Your bank balance isn’t in the bank, and other alchemy

Patrick reads his classic Bits About Money essay on why your bank deposit is not what you think it is. He explains the capital stack that makes deposits appear riskless while funding genuinely risky businesses, and why the "no questions asked" property of money took the United States roughly a hundred years to engineer.

As you'll see below, Patrick updates the essay with commentary on SVB's collapse, the Voyager collapse and emergency injunctions about the finer points of ACH plumbing, and the GENIUS Act's stablecoin interest ban. He argues that crypto keeps rediscovering the same hard truth: things that behave like deposits without being deposits eventually break. When they break, they will break other structures they have wormed into, and they will tend to have wormed into a lot, because deposits are extremely useful and are perceived to never break.

Presenting Sponsors: Mercury, Meter & Granola

Complex Systems is presented by Mercury—radically better banking for founders. Mercury offers the best wire experience anywhere: fast, reliable, and free for domestic U.S. wires, so you can stay focused on growing your business. Apply online in minutes at mercury.com.

Networking infrastructure has a way of accumulating technical debt faster than almost anything else in IT. Meter handles the full stack (wired, wireless, and cellular) as a single integrated solution: designed, deployed, and managed end-to-end so there's only one vendor to call when something goes wrong. Visit meter.com/complexsystems to book a demo.

If meetings consistently leave you with hazy action items and lost context, Granola handles the transcription so you can actually participate and gives you searchable notes afterward. Try it free at granola.ai/complexsystems with code COMPLEXSYSTEMS

Timestamps:

(00:00) Intro
(00:20) Why revisit this essay now
(02:03) Deposits are money
(06:53) Heavily engineered structured products pretending to be simple
(09:11) Credit card charge-offs as an underappreciated welfare program
(10:16) Deposits as pink slime
(13:08) Silicon Valley Bank and information sensitivity in the real world
(19:06) Many things are quasi-deposits
(20:00) Sponsors: Mercury | Meter
(23:13) Many things are quasi-deposits (cont’d)
(25:10) Voyager bankruptcy
(32:29) How the FDIC resolves bank failures over weekends
(34:49) Making the magic happen
(35:13) The GENIUS Act and the stablecoin interest debate
(40:31) Sponsor: Granola
(47:45) Wrap

Transcript

Hey everybody. My name is Patrick McKenzie, better known as patio11 on the internet.

Why revisit this essay now?

I sometimes look back on things that I've previously written and wonder whether they've aged well or not. I have a piece from 2022 about the alchemy of deposits, the humble bank deposit, this thing that undergirds so much financial infrastructure, which talks about some of the downside risk that happens.

And if you've been following along in the intervening years between 2022 and now, there have been uncertain times for banking infrastructure and the stability of the deposit regime in various places. We also increasingly have the upstarts in crypto land attempting to use stablecoins as a product, which is quite similar to deposits. This is interesting because a few years ago they also had some products which were quite similar to deposits, and those didn't exactly work out well.

[Patrick notes: I am largely referring to the exchange Earn products, which were sold as deposits but actually represented effectively equity-adjacent exposures to various trading strategies. The exchanges which offered these blamed everyone but themselves for their failures. For example, Gemini placed customer funds with Genesis, which loaned them (30% of the loan book exposure!) to Three Arrows Capital, on the basis of a one sentence attestation of assets under management. Genesis subsequently filed for bankruptcy and Gemini Earn customers were locked out of approximately $900 million in funds. They had been told that Earn was as liquid as a checking account. This was a lie.]

And while I don't think stablecoins are rerunning those products exactly, I think some historical perspective might be useful. So we'll do what I usually do for this sort of episode: read the essay that was originally published back on Bits About Money, and then give some live commentary about updates and what's happened since then,and what I would say if you were just talking about this topic to me at dinner.

And so, "The alchemy of deposits," originally published on Bits About Money on July 7th, 2022.

The alchemy of deposits

The most basic product offered by banks is also one of the least understood. You've depended on its orderly operation your entire life. It runs society at microeconomic and macroeconomic levels. It required hundreds of years of trial and error, plus an edifice some would describe as the world's largest conspiracy theory that is actually true, to make it safe enough to bet society on.

Behold the terrible majesty of the humble bank deposit.

Deposits are money

The model we are taught as children is, like most models, useful but inaccurate. “You take this $20 bill to a bank and deposit it. They will keep it safe for you, and then give you $20 back in the future, plus a little extra for having the use of it in the meanwhile. We call that interest.”

Let’s start dissecting this transaction. You don’t deposit a $20 bill. You purchase a $20 deposit, coincidentally using a piece of paper with the same number on it. The deposit is a liability (a debt) of the bank to you. The bill which you gave the bank in return for the deposit is now theirs, the same as if you had bought a cup of coffee from Starbucks. On their balance sheet, it is now an asset.

The core action which one takes with a deposit is not actually withdrawing it (plus a bit of extra interest). Most deposits will not be withdrawn by the person who originally put them in the bank.

The actual core feature of deposits is that you can transfer them to other people to effect payments. Big deal, you might think. You can also transfer cows, seashells, Bitcoin, an IOU from a friend, or bonds issued by Google to effect payments.

But deposits are treated as money by just about everyone who matters in the economy, including (pointedly) the state. Economists can wax lyrical about what "treated as money" means, but the non-specialist gloss is probably just as useful: anything is money if substantially everyone looking at the money both agrees that it is money and agrees on the exchange rate for it. This is sometimes referred to as the "no questions asked" property. Money is the Schelling point for value transfers that all parties to a transaction are already at.

This is so fundamental a feature of deposits that, in developed nations, we don't remember that it isn't automatic. There isn't an ongoing exchange rate between Chase dollars and Bank of America dollars. Even attempting to imagine that invokes images of chaos. You'd have to specify a bank while doing a salary negotiation. Your rent might swing up or down based on your landlord's judgment of your bank's credit position. Newspapers might print daily charts, for the benefit of the business community, of the exchange rates of deposits for every bank in town, so that they knew how much extra to ask for when you wrote a check to settle a $20 purchase.

All these and more used to happen.

As an aside, I'm referring largely to the 1820s to 1860s, the so-called "free banking era" in the United States, when state-chartered banks were operating with relatively little scrutiny compared to modern regulatory regimes. This was also sometimes called the "wildcat banking era."

And it might be apocryphal, but one of the reasons people describe these banks as wildcats is that the banks were intentionally located so physically remote from where their depositors might be that only wildcats could find them. The reason to put the bank remote from the market, as opposed to putting a branch directly in the market where you would typically want to site one, was that the bank notes the bank issued were private money, mostly redeemable in specie, meaning gold and silver coins issued by the federal government. The purpose of the private bank note being issued by a bank remote to transactors was: it will be difficult for them to actually come out and try to take our coins from us. And therefore, if we have a temporary liquidity problem, or maybe some ill-considered loans, well, that won't cause a run on the bank, because running on the bank requires trudging through miles and miles of wilderness, so only the wildcats could spot us.

We've mostly forgotten this history, and we've mostly forgotten that it took about a hundred years of sustained effort from the United States of America, between about 1830 and about 1933, to successfully establish a set of controls, including deposit insurance, including national banking regulators, et cetera, such that there is automatic equivalence in deposits at any bank in the country.

These also happen today in places without well-developed banking systems. Crypto is a good example, to avoid stigmatizing developing nations. There is an exchange rate, constantly changing, between the stablecoins USDC and USDT, between both of those coins independently and the dollars they theoretically represent. Different rates prevail in different places and different transaction sizes. This makes stablecoin-settled commerce very rare relative to money-settled commerce.

As an aside, it's still very rare relative to money-settled commerce, but it's less rare than it was in 2022. The growth of stablecoins used for actual payment activity has been quite meteoric over the course of the last couple of years. The stablecoin.fyi paper from Artemis (et al) has some fairly decent data on this subject. So: extremely small relative to payments, including relative to payment methods that are relatively nascent, like Pix down in Brazil, but they are far and away the most used thing outside of speculation that the crypto community has developed yet. [Patrick notes: I discussed the stablecoin paper, which is methodologically quite interesting, a bit in the episode recapping some remarks I made at the Bank of England.]

Heavily engineered structured products pretending to be simple

From the consumer's perspective, deposits are "my money," functionally riskless. This rounds to correct.

From the bank's perspective, deposits are part of the capital stack of the bank, allowing it to engage in a variety of risky businesses. This rounds to correct.

The reconciliation between this polymorphism is a feat of financial and social engineering. A bank packages up its various risky businesses—chiefly making loans, but many banks have other functions in addition to the risks associated with any operating business—puts them in a blender, reduces them to a homogenous mix, and then pours that risk mix over a defined waterfall.

The simplest model for that waterfall is, in order of increasing risk: deposits, bonds, preferred equity, and common equity. Holders of all of these complex financial products have committed capital, under various terms, to the operation of the bank's business. The more risky the product, the higher in general the return and the more risk of loss.

Careful balancing of the amounts of capital in the various categories, and of the portion of risk absorbed by equity in particular, is supposed to make deposits so safe as to never get any of that risk sludge splattered on them at all.

This happy just-so story has been told by every bank in history, yet depositors actually periodically lost lots of money, and the banking system had large systemic crises until it was backstopped by deposit insurance. That's a deep enough rabbit hole that I'll cover it in a later essay.

And indeed, I wrote an essay about deposit insurance. 

Why fund the risks of a bank with deposits, as opposed to funding them entirely with bonds and equity (and of course, revenue), like almost all businesses do? From the bank's perspective, this is simple: deposits are a very inexpensive funding source, and the capability to raise them is one of the main structural advantages banks have vis-a-vis all other firms in the economy. This allows banks to price loans much cheaper than non-bank lenders can afford to, sell vastly more loans (due to more demand for cheap loans) sell vastly more loans for any given level of capital, increase their return to equity holders, and similar.

From society's perspective, the wide availability of cheap credit is generally considered a good thing, as it allows for productive investment, consumption smoothing over consumers' lifetimes, and a form of risk-pooling not entirely dissimilar to public support or insurance programs. (It is underappreciated that consumer credit is, effectively, one of the largest welfare programs in the United States. Charge-offs of, for example, credit card debt effectively transfer a private benefit to the defaulting consumer in return for a diffuse cost to the rest of the public, mediated by the financial industry. The net amount of them is almost as much as food stamps.)

As an aside, to pull more current numbers: in the current state of the economy, which is not particularly dislocated, we have a relatively typical level of credit card charge-offs, about 4.5% on $1.28 trillion of outstanding balances,which means the charge-offs are about $55 to $60 billion a year. SNAP benefits (food stamps) are about $95 billion a year as of fiscal year 2024. So it's the same order of magnitude, but they're not tracking to exactly the same number. It's about 60% as large as food stamps, which is pretty large. It never hits the public purse, and it's almost never discussed when people discuss the various support options that are available within the US consumer ecosystem. This is a very important one. 

Deposits as pink slime

Pink slime gets a bad rap. It's delicious, economical, and about as healthy as meat generally is. (Cultural note for non-American listeners: pink slime is a weighted and generally disparaging name for what the food industry generally calls "processed meat product"; much of American processed protein is made of it. Chicken McNuggets, for example, really are chicken meat, but they've been blended into a slurry then formed into the classic nugget shape, rather than being contiguous cuts from any individual chicken.)

This actually extends farther up the ladder of socioeconomic status than people generally expect it to. People broadly know that a McDonald's hamburger is a thing in the world, and that Chicken McNuggets are a thing, and that there's probably not an individual chicken who is mortally committed to the production of any individual nugget. What people broadly don't know is that if you go to an Applebee's or a Chili's or similar and get chicken in that classic fillet shape with the little grill stripes on it, that is also generally a piece of engineered processed meat product rather than an individual cut from an individual chicken.

Deposits, sitting on the liabilities side of a bank's balance sheet, balance out a bank's assets, including ones which are risky or illiquid, such as the loan book. If your deposit at the bank corresponded one-to-one with an identifiable asset, you'd have to care about that asset quite a bit. If your deposit was not merely spiritually funding my mortgage but actually backed by my mortgage, you'd watch my financial situation as if it were your own, because it would be. You'd be tempted to sell out of my mortgage and into, well, anything else, if you thought I suddenly developed a taste for mixing expensive alcohol with more expensive poker tables.

And then the people doing business with you would be inconvenienced as well, because while your payments would be notionally denominated in dollars, they'd actually be backed by something you understand a lot better than the person receiving the payment. And so they'd be less likely to transact without trust in you, might demand a risk-based discount, could delay transactions to do their own due diligence on me, and similar.

Specialists refer to this property as information sensitivity. (And if you want a great academic read on this, see Holmstrom, "Understanding the Role of Debt in the Financial System," BIS Working Paper 479, out in 2015. We'll discuss a worked example together in a moment.) The value of my mortgage is very tied to facts about my situation that an interested party could, through expenditure of effort, have an advantageous viewpoint on. It is information sensitive. The value of a deposit for 100 yen at the bank issuing my mortgage is precisely 100 yen. Every child knows it, and no amount of malfeasance will allow you to harm someone by engaging in an otherwise fair transaction for 100 yen and satisfying it with that deposit. Deposits are information-insensitive debt.

Another benefit of deposits, quite controversial, is that they allow money creation via a public-private partnership between the government and commercial banks. We'll have to address that another day.

Silicon Valley Bank and information sensitivity in the real world

Now let's talk about how information sensitivity plays out in the real world. If you recall back to 2023, Silicon Valley Bank found itself in a bit of a pickle.

The proximate reason for Silicon Valley Bank’s problem was that it had a large amount of deposits concentrated in basically the tech and biotech communities, and they used those deposits to fund a business. They had a business issuing loans, which was largely not their problem. And then they had a lot of their assets in agency-sponsored mortgage-backed securities, which are essentially obligations of the federal government. Those agency MBS have duration associated with them. They're medium-duration assets, the average duration of an MBS is about seven years.

And so there's a mismatch. Deposits are redeemable on demand from the user, and an agency MBS has the duration of, say, seven years. Now, when the Fed funds rate moves by 1%, every fixed income asset moves down in value by about 1% for each year of duration that it has. And so those seven-year duration MBS that were abundantly on the balance sheet of SVB were impaired by about 5% for each of those years, thus losing about a third of their value, because of the rapid ratchet in the Fed funds rate. That ratchet had been instituted to tame inflation.

The FDIC had some excellent letters to their friends at the Fed saying, "Hey guys, just so you know, if you rapidly ratchet the Fed funds rate, that's going to break the banking industry." That isn't an exact quote, they were somewhat more diplomatic about it. But the FDIC quarterly reports up until the mini crisis in 2023 make for excellent reading, particularly retrospectively. [Patrick notes: See my essay Banking in Very Uncertain Times, which was written early in the 2023 crisis and discusses an FDIC report from a few weeks earlier.

The very moderate language of that report was:

Unrealized losses on available–for–sale and held–to–maturity securities totaled $620 billion in the fourth quarter, down $69.5 billion from the prior quarter, due in part to lower mortgage rates. The combination of a high level of longer–term asset maturities and a moderate decline in total deposits underscores the risk that these unrealized losses could become actual losses should banks need to sell securities to meet liquidity needs.

]

And so that is essentially what happened. SVB failed to have a diversified base of depositors, and they failed to have control over their duration risk. And then they got caught really badly holding good assets that they had a liquidity mismatch for. And then we rescued other banks in that situation by doing a variety of backstopping schemes to prevent that crisis from going more systemic than it already was.

Now, what does that have to do with information sensitivity? Well, deposits, once they exist, get into everything, because the world has a vast variety of structures that have the need for a basically risky thing somewhere in the structure. And so one structure you can have in the world is a stablecoin. Let's say, for the purpose of this exercise, I take no particular point of view on whether stablecoins as a class, or any individual stablecoin, will have a stable value quoted by the market over extended periods of time. Crypto very definitely has a point of view on that question, but I am true neutral for the moment.

Let me observe that during the SVB crisis, Circle, the issuer of USDC, had about $3.3 billion [Patrick notes: misstated in audio as $4.3 billion] on deposit at Silicon Valley Bank, and that was at the time information-insensitive. Circle doesn't have to care about the health of Silicon Valley Bank, like every business in the economy generally doesn't have to care about the health of their bank, and they certainly don't have to look through the bank's balance sheet at the assets or at duration mismatch. 

This was brought up by detractors of the tech industry during the crisis, and honestly, I don't understand it. Nobody underwrites their bank in this fashion. Even professional managers of treasury functions at large financial services companies largely don't underwrite the banks. They just diversify on the assumption that, well, occasionally there will be something that our underwriting couldn't possibly catch with regards to a particular bank. And so we will diversify our sources and avoid having our exposure to any bank above particular thresholds.

So Circle had depended on SVB money being money. SVB money was suddenly, for about the period of a weekend, information-sensitive. You had to have a point of view on whether SVB was likely to make it through till Monday or not, or on SVB surviving through to Monday in some fashion, perhaps as a successor entity. To what degree will that successor entity get extraordinary federal support? To what degree will it be able to pay out uninsured balances?

And so the market ran an auction very, very quickly, pricing the portion of the USDC reserves that were at SVB, and the market's estimation of whether Circle or other entities within crypto would backfill for those reserves if those SVB deposits were not money-good. The conclusion the market came to for a few hours over the weekend was that those deposits were incredibly impaired, and therefore USDC itself was impaired. And so there were transactions that printed as low as the mid-eighties for USDC, which, if you had asked me that weekend, I would have said "yeah, USDC is probably money-good." Granted, I did not have money on any exchanges at the time, or I would have made some money by taking the risk that USDC was not money-good. But, for well-known reasons, people generally don't expect me to be a reliable oracle with respect to crypto asset prices. And indeed, I do not have a material amount of money in the crypto ecosystem at the moment either.

We'll return to some crypto stories in a moment, because while they're certainly not the only banking-adjacent issues that have happened in the last couple of years, I think they're a great window into the world of how deposits work and sometimes fail to work.

Many things are quasi-deposits

Banking is, when executed well, a very lucrative business to be in, and many firms and financial products replicate parts of it. These are sometimes described as "shadow banking," which is a word that might conceal as much as it illuminates.

Be that as it may, unsophisticated customers perceive many things where a database shows an entry in dollars as deposits, and (sometimes worrisomely) those are actually sold as being equivalent to deposits. If they're not money, they are not deposits.

Starbucks balances are a useful thing to have if you drink coffee, and are denominated in dollars (or here in Japan, in yen, etc.). Starbucks balances are not deposits because nobody takes them except for Starbucks. (If someone asks for money, maybe you should give them money. If someone other than Starbucks asks for Starbucks balance, hang up, because there is a 100% likelihood you are being defrauded.)

As an aside, the AARP (an American advocacy organization for older adults) did in the interim a public interest campaign of announcements telling people exactly that: don't ever pay for things with gift cards. Unfortunately, paying for things with gift cards is the core design goal of gift cards. I've discussed that in another essay.

Many products are not as clear-cut as Starbucks balances. The fintech industry has not covered itself in glory here. Sometimes firms misclaim a product to be a deposit where it is not. Sometimes they actually institutionally misunderstand the nature of the product they have created.

One would hope that that never happens, but just like the most expensive mistake in the history of programming being the assumption that programmers can count, smart people are doomed to continue discovering that just because a deposit is a complex structured product involving a bank which has a stable dollar value, not every complex structured product involving a bank which appears to have a stable dollar value is actually a deposit.

Voyager, a publicly traded company, marketed a deposit-adjacent product to users, paying a generous interest rate. Then a cascading series of events in crypto, outside the scope of this essay, blew up a series of firms, including one which had taken out a loan of hundreds of millions of dollars from Voyager. Suddenly, the information-insensitivity of Voyager's not-deposits was pierced, the pink slime appears both undermixed and undercooked, and customers now need to follow a bankruptcy proceeding closely.

I will give you an update on the bankruptcy proceeding in a moment.

(An interesting tangent is that, since all deposits are interchangeable by design,they are money, after all,that invariant is relied upon for the orderly operation of financial infrastructure. When something which was believed to be a deposit is discovered to not actually be a deposit, infrastructure around it breaks catastrophically. Matt Levine has an excellent extended discussion about how Voyager discovered that attaching the ACH payment rail to their deposit-adjacent product became a huge risk once they went under.)

Voyager bankruptcy

A brief history of time on the Voyager controversy. Voyager went under in July of 2022. Voyager was, prior to this, a relatively well-regarded player in the crypto space. They were publicly traded, well, in Canada. They were assumed to have a functioning risk function, although their risk function was not operating with the level of probity that they messaged to important stakeholders, to put it mildly. They had about 33% concentration in single-name exposure at one point, which is the sort of thing you would tell a bank and they would assume was a bad joke.

So they lost all the money. Well, what are you going to do? Originally there was a bit of a bidding process, and a well-capitalized firm in the crypto industry came in with essentially a bailout for Voyager: acquire all the assets, make customers whole on the liabilities, and just fold them into their commercial empire. Unfortunately, Sam Bankman-Fried discovered in November of 2022 that he didn't exactly have money to write checks to acquire anybody's businesses. And so that fell through. Then Binance took a whack at it and then walked away in April 2023 when they found out that the United States federal government considers them to be Bond villains.

And so the liquidation proceeded, and customers ended up recovering around 36%, about a year after Voyager failed. There was some ancillary recovery after that, after Voyager successfully pressed offsetting claims against the FTX estate and some other actors in crypto. But take 36 cents as a reasonable mark for the moment.

So we mentioned that things which are not deposits failed catastrophically when attached to ACH payment rails. Here's the nature of that failure. As I've discussed in other places, you have the right to reverse ACH payments in the United States under a variety of predicates, including in the case of fraud.

Fraud is a complex thing to adjudicate in white collar crimes, but the three elements of the offense are very simple. One: someone said something which was untrue. Two: you reasonably relied on that representation. Three: you lost money as a result of relying on that representation. So the thing Voyager said was: "If you give us dollars, we will give you Bitcoin." And the thing that many customers observed in the days immediately after the Voyager bankruptcy was: "I gave Voyager dollars. I don't have Bitcoin. I feel defrauded." And so those customers went to their own bank, Bank of America, or Chase, or wherever, and said essentially: "I paid money to an Internet merchant. They didn't deliver. Give me my money back."

And a bank customer service representative, who is not a US federal court judge, and who does not really have a mandate or training for reading through complex bankruptcy cases and explaining to an unsatisfied consumer what having an unsecured claim in a bankruptcy proceeding means, heard: "Oh yeah, internet merchant didn't give you the thing? Yeah, it happens all the time." And hit a button on their interface in the bank, and that mechanically reversed the ACH payment.

Now, Voyager was in bankruptcy protection at the time, and so Voyager could not pay out money except as approved by a bankruptcy trustee. And the bankruptcy trustee had not approved these payments because in bankruptcy land they are somewhat improper. You are not supposed to be able to get priority over other people that have a claim on a bankruptcy estate just by being able to call Bank of America up and say that you have been wronged by the bankrupt entity, but in actual fact, you can do that.

The money gets pulled back by ACH either way. Your bank does not make an exception to the ACH rules here. 

So whose balance sheet did those dollars that were restored to aggrieved customers come from? They came from Metropolitan Commercial’s balance sheet. Who's Metropolitan Commercial? Metropolitan Commercial is a bank. They extended banking services to Voyager and they thought they were very vanilla, low-risk cash management services, sort of a jumped up version of a basic business account.

 It turned out to be other than low risk because they got hit by a wave of these reversals and they were pulling money, again, not from Voyager, from Metropolitan. Whose money at Metropolitan, specifically? Our ordinary understanding is losses first hit equity, the shareholder’s money, which earns a return precisely because it is designed to absorb risks to Metropolitan's businesses.

Metropolitan went to the bankruptcy judge and asked for an emergency injunction and for essentially permission to change the physics of ACH in the United States. They argued that Voyager can’t dispute these attempts at reversals at the moment, both due to the rules of the ACH network and also because Voyager is somewhat busy at the moment. Because of this infelicity and how ACHs are working vis-a-vis the bankruptcy estate right now, this is putting a US bank in a bit of pressure, potentially. And, though this was not couched in exactly these words, they intimated it might actually cause us to fail and that would be a very bad thing.

 And so Metropolitan Commercial received permission from the judge to essentially start blanket rejecting these ACH reversal requests where that is not a thing that any bank is generally speaking, allowed to do. But bankruptcy judges have rather extraordinary power with regards to managing the equities under bankruptcy law.

 I mentioned this incident in Debanking and Debunking at the end of 2024, because the crypto industry has often said that banks just don't get it. They don't understand that we are a low-risk business of legitimate entrepreneurs selling legitimate products in the legitimate economy. We just want a checking account.

 And banks do in fact get it, because activities which you might think are low risk, are in fact not low risk. And banks need to know this. And when they get caught out in not doing their underwriting and giving someone something which rounds to just a business checking account for a publicly traded company that has responsible risk officers, what's the worst that can happen?

 The worst that can happen is Metropolitan Commercial. Their entire crypto practice went under over one client making essentially one mistake. One client made one bad loan, and that brought down a bank's entire line of business. Metropolitan wasn't ordered out of the business of crypto banking by regulators, which is something that has been alleged with regards to regulators a number of times. Metropolitan made the independent commercial decision partially in response to the regulatory headwinds, and also partially because they made a bet that crypto banking was low risk business to be in. Then they decisively lost that bet.

How the FDIC resolves bank failures over weekends

I'd also like to make another riff here. We've covered a case where attaching a deposit to a product which isn't a deposit can cause absolute chaos to break out when this fact becomes widely known, because the physics of deposits and the physics of non-deposit products interact very poorly with each other. And when everyone knows that, that's fine, but when you don't know it and you suddenly discover it in a crisis, that's a very bad situation to be in.

One of the reasons the FDIC resolves bank failures over the weekend as opposed to on any given Tuesday is that the much-maligned feature of banks, where banks don't do money movement over the weekends in the same fashion that they do it on weekdays, and don't work over banking holidays, creates something of a buffering effect on the banking system.

And so your FDIC resolution team has 48 hours, potentially a little more, to find another bank to take over a failed bank, or to create a successor vehicle, et cetera, to make sure as many of the deposits as possible, and certainly all the insured ones, are money-good come Monday.

Why do they do this? Well, one of the reasons is that when SVB fails, or when First Republic has a number of months of "will they, won't they" and then ultimately fails, or when Signature Bank fails over the course of a weekend, you don't want customers making the same calculation that people made about Voyager, where: "I sent my money to this bank in the course of ordinary banking business via an ACH payment. I don't know if it arrived there in any sense that matters. Maybe I should just reverse the ACH payment." That would create systemic risk in a variety of fashions, because this chains through other places in the economy.

The big worry in 2023 was that this wasn't a few banks individually failing for idiosyncratic reasons, that this was going to metastasize among this section of large regional banks in the United States, potentially even wider. And indeed we saw about two dozen or so firms undergoing 5 sigma plus deposit outflows over the period of a week or two, as other actors were wondering what other banks were exposed to similar underlying issues and the fallout of these existing bank failures.

[Patrick notes: Researchers at the Fed use their somewhat oracular view of national payments to calculate that precisely 22 institutions saw outflows greater than five standard deviations from their historical norms. How anomalous is that? Well, if you assume that outflows normally bounce around in Brownian motion, you expect to see roughly one bank with a five standard deviation outflow. Per decade.]

So resolving it over a weekend both preserves the equities of the people involved in the particular bank and doesn't create a miniature run dynamic conducted not through actual transfers out of the bank, but by buttons that customer service representatives are pushing at other banks to do things like claw back ACH transfers, dispute credit card transactions, and similar.

Making the magic happen

Capital adequacy ratios, reserves, and other not-so-secret sauce in banking are worth many books. Deposit insurance, a key ingredient in making deposits as reliable as they are perceived to be, is actually explainable in a reasonably compact form. Tune in for that next time.

And indeed I've written "Deposit Insurance Under the Coverage," which I recommend you read if you're interested in that sort of thing.

The GENIUS Act and the stablecoin interest debate

I have one other riff that I'd like to make while I'm in the studio, on the current discussions with regards to stablecoin. If you haven't followed stablecoin regulation in detail, there's been a GENIUS Act, because everyone in Washington, D.C. loves their acronyms. I don't remember what it stands for off the top of my head. [Patrick notes: Guiding and Establishing National Innovation for U.S. Stablecoins.]

One of the provisions of the GENIUS Act is that stablecoins aren't allowed to pay interest. And this is basically at the behest of the banking lobby. The banking lobby doesn't want stablecoins competing for market share with its core deposit business. In particular, they're worried about whether that would affect the deposit business at midsize banks.

We've been in an interesting rates environment over the course of the last couple of years. When I wrote the "Alchemy of Deposits" piece, we were at about 1.5% to 1.75% Fed funds rate. Currently it's about 3.5% to 3.75%, and in July 2023 it peaked at about 5.33%. You'll notice those are the Fed funds rate, and basically none of your deposits at a bank pay anything close to the Fed funds rate. Your checking account probably pays nothing, or something that is indistinguishable from nothing.

Mine is a solid one basis point. Yummy. Thank you.

But the reason for that is something called the convenience yield. The convenience yield basically means that the other features of checking accounts,aside from the interest, the ability to pay people, to call the bank at 8:00 PM and say "my card was stolen, please get me a new one," et cetera, these compensate you for giving the bank temporary use of your money to run their entire banking business. And so you don't need to be compensated as much with direct interest on the bank account.

This is a calculation that all consumers in an economy and all businesses are constantly making. Even relatively sophisticated businesses often don't do much on deposit pricing.

And so they get a lower-than-one deposit beta. Deposit beta means: when the Fed changes the amount of money the Fed pays to banks for depositing their money overnight at the Fed, banks don't pass 100% of that immediately to customers. They pass some number which is lower than 100%, expressed as a decimal. This has, broadly speaking, been trending down in recent years, essentially, experimentation at banks to see how sensitive their customers are to deposit beta really. And the answer is: not very sensitive. Most people don't make decisions about where to bank based on who offers the best rate on deposits.

The people who do make those decisions largely do so through intermediaries called deposit brokers. And so that's an important part of the funding mix for banks such as Signature, which relied on brokered deposits for a substantial portion of their funding. And that's one reason why the percussive run on Signature was as fast as it was, because it wasn't individuals making decisions to withdraw their family's checking account from Signature over the weekend. It was deposit brokers who were realizing on Thursday that they were probably in a pretty bad way, and hoping to get a wire out on Friday before any craziness happened over the weekend. And that weekend had a lot of craziness.

So the pitch from the stablecoin industry is: to date, people have been willing to use stablecoins in hundreds-of-billions-of-dollar sizes for essentially a convenience yield. There's some reason to hold Tether, we're not going to talk about it, but people are willing to extend their use of hundreds of billions of dollars of real money in return for getting something they can use to make money in the crypto economy, or for some other purposes which we don't have to discuss in detail here.

And the stablecoin industry has said variants of: we think stablecoins are convenient to use, and we don't think you should be prohibited from paying interest on them.

So far, the banking lobby has won that discussion. And then in the interim, stablecoins came up with: "Okay, we're not allowed to pay interest, but you banks, we hear you love your credit card reward schemes. Can we pay rewards? Rewards are kind of like interest, except we don't call it interest. And we'll have a complex multi-party system which will result in rewards getting remitted to the end customer without the stablecoin issuer actually issuing the rewards like they would issue interest to people with deposits. This fig leaf should assuage your concerns."

So, clearly that worked, right? And the banking industry and federal government looked at them like: "Yeah, no, that doesn't really work for us."

This is an ongoing discussion. The arguments for both sides are relatively understandable based on their individual view on the merits and not entirely cynical. It's not just people talking their book. Banks say: "Look, we pay interest on deposits because we have, remember, that hundred years of experience that the United States went through, from the time where deposits were basically vanishing acts to the time where they could be widely relied on as societal infrastructure. We did all that work. We do all that work on every Tuesday. The stablecoin issuers do none of it."

And the reason we aren't having constant problems in banks is because of all that work. As one tiny example: the capital adequacy ratios describing the complex capital stack that banks are customarily backed by. Tether's capital cushion is what, 0.3% of deposits? That's risky, even if you believe what Tether says about the composition of their reserves, which would also be risky if it were a bank. But Tether is being tethered off in the corner, and the banking lobby or other bank people would look at the more regulated forms of stablecoins, like USDC and similar, and say: "Okay, you guys are doing better than Tether, but that's a very low bar."

And structurally, even if you have high-quality assets, you can't make this product 100% safe. How many times do we have to reexperience this lesson? SVB had good assets. The agency-sponsored mortgage-backed securities weren't going anywhere. They just had duration risk associated with them.

You say you're going to have less duration risk because you're going to be solely in 30-day T-bills. That's wonderful. But if you're executing at the speed of the internet, when we know that a percussive run can result in outflows of tens of billions of dollars in a day, because we recently went through that, and the executives who were in charge of it will no longer work in the banking industry. What if we had tens of billions or hundreds of billions in outflows on one of your products over a weekend? That would cascade into all sorts of other businesses that we care about. And this is not a theoretical risk.

If you remember back to 2008, a different generation than the Zoomers had invented their own thing that was not quite a deposit, called the money market fund. And you guys have just made money market funds tradable instantaneously. That doesn't necessarily make them less risky.

And indeed, that nearly blew up the economy back in 2008 when one money market fund, the Reserve Fund, broke the buck briefly. One of the less well-understood things about the 2008 financial crisis is that there are many things that are money-like until they aren't, including, for example, the repo market built on top of treasuries. Short-duration treasuries are definitionally the riskless asset, but infrastructure built on top of them is not necessarily riskless during periods of stress. And so we don't want to smuggle systemic risk into society by simply paying people interest payments.

The crypto industry says their opponents are just talking their book. I believe the bank lobby is being relatively sincere with respect to: "Look, we are the people that society trusts to avoid banking crises. And the reason we get paid the big bucks is not because there's zero risk of a banking crisis."

And from the stablecoin side of things, I think they believe that generally speaking, the capability of databases and constantly-on market-making is a buffer against some of what they perceive as crises happening in the banking community. And I think they're not entirely right, and not entirely wrong, about that.

Broadly speaking, the "let's reinvent all of financial technology from first principles" that the crypto community has been engaged in results in them having relatively more working software for various things than the rest of the economy does. The market makers in crypto are sort of fair weather friends where they are very happy  to intermediate a wide variety of transactions, almost all the time.

[Patrick notes: See October 2025 for the difference between “almost all” and “all” of the time. During a brief but intense market dislocation, several major crypto market makers simultaneously withdrew liquidity, causing spreads on major stablecoins to blow out.

This particularly affected some products which are not stablecoins but which are pervasively missold to users as stablecoins, such as Ethena’s USDE product. To their credit, their CEO avoids characterizing it as a stablecoin during interviews, but as of this writing the main headline of their homepage is “Digital Dollars for the Internet Economy”, and crypto has a market structure where much information passes through intermediaries who are compensated to sell products using language which is even further out on the risk curve.]

When they are not, the crypto economy experiences some dislocation. This is interesting if you hold a lot of your net worth in Ethereum, but which doesn't affect people's ability to put food on the table, typically, because they're not getting paid their paychecks via stablecoins or similar. And in a future world where people are getting paid their paychecks via stablecoins, we would have some level of worry if they were not up to the standards the rest of the deposit system was built on.

So I don't imagine that's going to be definitively resolved in the next couple of months, although there might be some movement with respect to “clarity” on whether stablecoins can issue interest payments or not, or rewards, or some other mechanism on top of that.

What I most suspect will happen is that the banking industry is going to win the fight narrowly against stablecoins paying rewards. Then crypto is going to do some of their interesting application development and market infrastructure to come up with something which is not formally a stablecoin paying rewards. It will simply feel an awful lot to a customer as if they deposited a stablecoin and are suddenly earning rewards. For example, one could imagine a product manager for a wallet, perhaps a wallet attached to a custodial exchange, using some sort of on-chain interest market. One could imagine that experience being extremely accessible for the user, such that the user doesn’t ever perceive themselves as interacting with a smart contract.

Should this come to pass in a way which users enjoy but which doesn’t scale to trillion-dollar sizes, that might be a "compromise" that works for everyone. For at least the next couple of years. Until we get another set of headlines.

And so that's my brief riffing about the current state of stablecoins and the debate with respect to interest on them. We'll check back in on how that works out.

But thank you very much for listening to Complex Systems this week, and I'll see you next week.