Happy 2024! I had been hoping to have a year-in-review for 2023 ready by now, but a combination of illness and travel bushwacked me in January. So instead of that, we’ll go with regular programming for today and return to that later.

Brief housekeeping notes: one, if you like BAM, you’ll enjoy an interview I did with Tyler Cowen. Two, I’m sketching out the year both for BAM and other professional commitments, and am shooting for the same 20-26 issues that I was shooting for in 2023. This year I do not plan on my family again immigrating to America, which crushed my productivity in summer and lead to not hitting that forecast in 2023. As always, if you have comments or concerns, the inbox is open.

Check cashing

One of the reasons I covered the checks as a payment method recently was to lay the groundwork for talking about some of the fascinating alternative financial world around them. In the main, this helps people at the socioeconomic margins turn payroll and other checks into cash (or otherwise immediately spendable value) in return for a fee. This is not how you, reader, probably deal with checks, and the existence of this industry / product have been controversial for many years.

Hopefully you’ll soon understand why, and why it persists in existing and its customers persist in using it.

A note at the top: check cashing is distinct from payday loans, despite overlapping customer bases, overlapping retail presences, and the centrality of checks to both enterprises.

An oversimplified explanation of check presentment

In an ideal world I would have one essay already covering the nitty-gritty of check presentment but I don’t and so we will here do the handwavy version.

Pretend you are a banked individual, which in the U.S. context means that you have a checking account in your own name trivially available to you. You are handed a physical check. You would prefer to have money. What do you do?

You deposit it with your bank, naturally. This could be at the teller window, through the ATM, or (most likely if you’re reading an essay about financial infrastructure) through the “remote deposit capture” feature of your bank’s mobile app. You might not know it by that specialized name, and instead think “I use my iPhone to take a picture of the check then money shows up.”

Regardless of how you send the check to the bank, the next process is the same: the bank will make an electronic copy of the check, which is exactly the same legally as the check. This was made possible about twenty years ago by the Check 21 Act. Your bank will then electronically communicate that copy to the bank the check is drawn on (i.e. the bank whose routing number is printed on the bottom of the check).

That bank will, generally speaking, pay your bank money on the next business day. I will be intentionally handwavy here as to how, since the miracle that is net settlement is not the focus of this essay. Just assume that money magically arrives tomorrow.

Now, if you’ve paid attention when cashing checks over your lifetime, you may notice some confusion here. It is highly likely that you’ve waited for longer than one business day to receive funds from a check before. It is also likely that many of you recall at times not waitingat all; you received either full or partial credit for the check the same day you deposited it. What is up with this discrepancy?

A full discussion of Regulation CC, the Expedited Funds Availability Act, and banks’ sliding scale of willingness to be accommodating will have to wait for another BAM. Instead, let’s focus on the most fundamental issue.

Depositing a check requires an extension of credit

As we’ve previously discussed, all a payment ever has been is a message about the status of a debt with some level of certainty associated with it. A check is a specially formatted message. The “movement” of money tomorrow is another message. But neither of those messages encode absolute certainty that the payer is certainly discharging their debt to the payee.

If something goes wrong in this process—the overwhelmingly most likely one is that the payer doesn’t have the funds to cover the check (NSF, or “insufficient funds”), but the check being fraudulent or unauthorized is also possible—that wrongness may not be discovered before money “moves” to your bank. And so that payment can be recalled from your bank to the bank the check is drawn on. This will likely result in the bank attempting to recall the money from your account.

And so by presenting your check, which you think is substantially terminating a transaction, you are actually creating a new credit extension with your bank. They are extremely aware that you just asked them to advance you money, even if you are not aware that you did that. They already partially underwrote this extension of credit; that is why you were not shooed out of the building when you originally asked for a checking account.

And note that this credit relationship has two sources of risk. One is with the check writer: is that check going to “bounce” (be returned as NSF) after presentment? One is with the check payee: if the bank attempts to recover a bounced check from their customer, will that customer make the bank whole?

If you are banked in the United States, that means that a profit-maximizing institution looked at you and said “In expectation, almost all checks this person presents will be good, drawn legitimately on the accounts of individuals or firms who do not make a habit of bouncing checks. Of the tiny, tiny number of checks that this person deposits that will bounce, and honestly it will probably be zero over the lifetime of their account, we have high confidence that they will make us whole. They have credit here.”

There exist a variety of ways to be unbanked in the United States. One, which is socially awkward for advocates to take explicit notice of, is that a bank could look at a seeker of credit and say “Actually, in expectation, checks you present to us have a much-higher-than-baseline risk of bouncing. When they bounce, and our best estimate is you’ll bounce multiple times per year, our desired outcome is you make us whole and pay a modest penalty under our agreement. We assess that you present a material risk of not doing this. You might have gotten an A for 92% in school but if your checks are 92% likely to be good money that’s a hard no from us. We decline to extend you credit under this product. We may extend you credit in other fashions, if you ask. You’ll find that in those products credit costs a lot more than the cheap credit embedded in checking accounts, reflecting the elevated risk of doing business with you.”

Advocates at this point often call the banks racist, classist, and stupid.

Let’s pretend you asked decisionmakers in deposit franchises for their point of view here. They’d be happy to tell you. It would rhyme with this:

You know, we have been in this game for decades, and like to think we are pretty good at it. We did not make up that estimate of creditworthiness. We most definitely did not engage in illegal discriminatory practices, like inferring creditworthiness from zip code. We know that would be extremely probative data if we were allowed to use it, because we’re in the data and math business and good at our jobs, but egads. Can you imagine the fines we’d pay? The headlines? Compliance keeps a file to scare young analysts with. No, we paid a few dollars to get a report from ChexSystems, which said that the literal same person who wants us to extend credit ripped off the bank down the street for $450 a few months ago. If you were in consumer banking, and you’re not because we are having this discussion, you would recognize that as multiple years of the contribution margin of a checking account relationship. Do you want to extend them credit? Then bonne chance. We decline.

That is certainly not the only pathway to being non-banked, but it is an extremely common one. Often, consumers self-select out because they try banking for a while and then repeatedly get assessed high fees which they do not feel are legitimate, such as fees for overdrafting their accounts. To bang a very old drum, the decision to move from everyone-pays-a-Netflix-subscription-for-banking to banking-is-free-except-we-assess-high-fees-if-you-screw-up created winners and losers. We called that one "free checking." Descriptively it subsidizes the middle class by using fees assessed stochastically to people in persistent economic precarity. In particular, young members of the middle class (college students and recent graduates in their least-well-off years) benefitted a lot.

So let's look at something that recent college grads very rarely encounter.

How cashing a check works if you’re not banked

You begin with a physical check. You generally physically walk it into a local business, which in Chicago are most commonly “currency exchanges” but which can in principle be done in many businesses known by many names (much like e.g. Western Union transfers). You ask to cash it. You speak to a clerk with a high school education, who likely remembers you from previous interactions. The clerk asks you to endorse the check, a ritual we’ll return to in a moment. She swiftly pays you cash from a drawer and retains the physical check. She offers you a receipt, and you do with it exactly what you do with a receipt from McDonalds. 

The cash does not match the value printed on the check. The fee you were charged is prominently disclosed on the receipt, on the wall, and in printed material you are passively offered but do not take, for the same reason that you’ve never walked home with a McDonalds menu.

That pricing grid looks something like this:

A brief aside about endorsement

Apologies to non-Americans who are wondering whether they’ve wandered into Westeros, with clearly medieval payment methods plus a likelihood of being beset by monsters, but let me explain a traditional ritual of our people.

The reverse of a check contains a small area where you can “endorse” the check. This means signing it and optionally leaving the bank an instruction as to what to do with the check. This is established by a combination of law and ancient practice.

One common endorsement historically is that you can endorse a check to another person, i.e. instruct the bank to make payment to the person you nominate, not to you. This enabled many use cases back in the day. A family might have two people earning wages but only one with title to a bank account; the second could endorse their wages to the first. A business could endorse a payment made by a customer to the business’ owner or to an employee or to a supplier. A common endorsement was simply “pay to cash.” (Banks hate that one and mostly don’t offer it any more. You can probably predict why.)

The practice of endorsing is why check cashing can exist as a business, because you can endorse your check to the check casher. As of that moment, by the magic that is the U.S. legal system, your payer no longer owes you money; they now owe the check casher money. The now-endorsed check instructs the check casher’s bank and the payer’s bank to cooperate to make this side agreement happen.

As long as we are on this tangent, have you ever written “For mobile deposit only at Your Bank Goes Here”? Wondered why?

This is to prevent an annoying fraud vector where someone takes a legitimate check and then deposits it roughly simultaneously at multiple banks. Each bank will come to the conclusion it is a legitimate check (because it is) paid to someone with an account with them (true as it goes). Then they will begin the process of crediting their customer. Only at some later point will it be discovered that the check was presented multiple times. If the actor is good at being bad, they can use this to extract money from the victim banks. Asking you to endorse the check in the above fashion spoils it for future fraudulent use at other banks. And there, now you know why you are subjected to a minor annoyance.

This is also why the app/bank can’t do this for you. They need you to physically spoil the specimen that only you have and prove you did so (with a photo of the endorsement).

Many people hate check cashing and everything about it

If you are reading this, you probably are relatively wealthy, are almost certainly of high socioeconomic status, and quite plausibly have never paid one red cent for check cashing in your entire life.

But that check cashing business exists, basically entirely, to siphon a small amount of dollars off of thousands of relatively poor people. It can’t make the math work any other way. Every time you see a currency exchange when driving around town, you can be safe in the knowledge that there were thousands of poor people paying the vig last month and the owner expects most of them to pay the vig this month, too.

Most of the customers of that check business are not in a position where they can afford to be indifferent to $9. And yet the check cashing business will rake the first $9 out of their monthly public benefits payments. Or it will take their wages for the first hour of every pay period at McDonalds. Or it will insist that, when the homeowner whose lawn was mowed by an immigrant offers a $100 Christmas bonus, that Scrooge must get his $3 first.

Advocates hate this business model with a passion unmatched by ten thousand burning suns. They like to quote statistics like “[t]he average unbanked worker in Illinois spends $574 a year to cash their payroll checks.” (I am not relaying this quote for its truth value.)

There exists a deep academic literature about the unbanked and underbanked, much of it written by scholars who are self-consciously advocates. They have the same degree of neutrality on the merits that historians of the Civil Rights movement have on Jim Crow. That is to say, their disdain oozes from the page.

Personally, I would not attempt to dissuade anyone from their aesthetic feelings with respect to this business model. That would not be a good use of anyone’s time.

Do I hate this business and everything about it? Eh, I hate poverty, certainly. Almost everything poverty touches will have terrible elements about it, because the definition of poverty is terribleness caused by scarcity. If you can’t find the terrible, you’ve either left poverty behind or you're not looking very hard.

And, not to put too fine a point on it, it’s really hard to expressively, passionately hate this business and not hate the young lady working in it, the decisions and life challenges of its customers, and similar. And if you hate one particular building in a poor neighborhood, and then start rigorously thinking about the liquor store, or the police station, or the supermarket, or the church, or the public school, or the home in the poor neighborhood, I think your mind will start going to some pretty dark places.

So maybe let’s move past the aesthetic revulsion, which you’re entirely welcome to, and just look at what is going on here.

The internal logic behind that pricing grid

There exist two sources of credit risk in cashing a check, as we covered earlier. The pricing grid directly prices credit risk for the payer of the check, via bucketing them. It is not the most discerning risk analysis ever conducted in the financial industry, because you need to be able to explain it to extremely unsophisticated customers and only barely more sophisticated staff.

The world’s financial system is predicated on the U.S. government being definitionally zero credit risk when denominated in dollars. Every other kind of debt in the world is defined in reference to a Treasury.

A portion of the price of every cell in that pricing grid is credit risk. Just like the spread between a bond and a Treasury of the same duration is a reflection of marginal riskiness, the spread between personal checks and government checks is a reflection of “the credit risk of the types of people who most commonly write checks to poor people.”

As you can see, by simple subtraction, this spread is non-zero but low.

Now what causes the price to be so hard high for the reference risk? Well, Treasuries certainly pay out to someone, but not everyone in the world who says they own $1 million in Treasuries actually does. Government checks don't have credit risk directly, they have operational risk which becomes a credit risk. In the case where either the person cashing them isn't the person named on the check, or where the government later comes to the conclusion that it didn't really want to pay them, that money could (at some risk) be clawed back from the bank, and therefore from the check casher.

It is also useful to understand that the consumer of this service is not solely paying for credit risk. A consumer of hamburgers at McDonalds is not solely paying for processed meat product. The physical storefront the transaction is conducted in pays commercial rent (in, given that this is a check cashing business, highly likely a low-rent area). The high school graduate who remembers you from last month, and who has many challenges of her own, wants a day’s wages for spending a day talking to poor people about money. Somewhere in the enterprise there exists a much more expensive professional who spent weeks of work writing up compliance procedures for a money services business, likely secured a license for the same, and then convinced a bank that it should accept the custom of one of the highest-risk legal businesses.

That bank does not cash the daily envelope of checks for free, either. It had a consequential commercial negotiation which took notice of the business’ risk profile, high operational costs associated with their custom, and the near certainty that they would be very annoying to work with. The bank ultimately quoted a price per check and, very likely, a minimum amount which would be assessed monthly.

How much? Eh, prices are prices. How much does a pound of potatoes cost? I don’t know. It depends on what kind of potato. It depends on where you live. It depends who you are buying the potato from. It depends an awful lot on whether you buy your potatoes by the pound or by the truckload or by the megaton. (Not an exaggeration if you are McDonald’s logistics system, right? Some check cashing businesses are owned by multi-state chains.)

But, as someone who has more professional experience with checks than with potatoes, if you twisted my arm I’d say “Indicatively, the bank charges 5 to 25 cents per check.”

Persistent identities as a KYC possibility

So we mentioned that there are two types of credit risk here, and how the payer credit risk is explicitly priced. How is the payee credit risk priced?

It isn’t. The check casher assumes that, as a first approximation, if a check bounces, they are going to lose money. Any other result is a windfall. Absorbing these credit losses is about 20% of the total costs of their check cashing operation.

That probably strikes you as pretty surprising, but if you earn $6 gross on a $200 transaction, and one out of every 250 checks bounces, well there you go. 40 bps default rate on 230 bps of revenue of which ~200 bps goes right back out the door.

The main procedural control for this that check cashers have is persistent identities. Your bank relies on a version of this, too, except in banking it is spelled KYC and is by regulation and practice excruciatingly formally documented. Many check cashing places will allow you to cash a check without having a government-issued ID, because many customers are from socioeconomic strata which routinely do not have a government-issued ID. Many check-cashing places will allow you to cash a check with a name which you are known by to the community but perhaps not what your teacher called out at roll call.

Some might take a picture of you the first time you do business with them. Some might simply rely on the social network and recollection of the young lady manning the cash register. Some will ask for and retain a copy of a government-issued ID or substitute paperwork, much like the DMV will. That could be a lease or utility bill or similar. Regardless, the purpose is the same: if you bounce a check, and you don’t pay it back, staff will be instructed that they don’t do business with this #(#%#($#)*$ anymore.

There are a spectrum of words that the staff and management could call you there, and none of them are phrased “defaulting customer” or appropriate for inclusion in this essay. All of them are, in fact, used at at least some establishments. As we’ve established, this is not a building in the part of town where people are ruthlessly socialized to not say really bad words about poor people. Compassion gets burned out of the owner and burned out of the clerk in basically the same ways that it gets burned out of everyone else in the neighborhood.

A brief discussion about class distinctions in America

One very real reason this type of business exists in the world is to be a firewall between social classes and the businesses that serve them. Check cashing establishments insulate banks, which are indispensable for cashing checks, from needing to talk to certain people.

A check cashing business is “alternative finance." It is alternative to the banking world of smartly dressed middle class employees, free coffee, and firm handshakes.

A check cashing clerk and a bank teller look to many to be similar jobs done by similar people and crucially they are not. Bank tellers do not make much money but know they must present as middle class. They work in a built environment where surveillance is absolutely ubiquitous and where deviant behavior (like using certain prescribed words) will have one referred to an alternative court system for swift and certain punishment.

That is to say: bank tellers work for an American corporation with an HR department. And bank tellers, in their hearts and in their actions, internalize the class that they must, must, must present as. There are classes of people that the bank does not want to do business with. (Banks are, as we have frequently covered, not allowed to say this in as many words.) The tellers do not want to speak to them, either, and this disdain radiates from them as palpable waves.

The clerk at a check cashing business is not a bank teller. She does not disdain talking to poor people; being able to do that in such a way that most poor people end up liking her is her job. Don’t take my word for it; take the customers’. We have studied this industry extensively. We ran surveys. The customers keep saying things like “I like my local check cashing place because the girl behind the counter is kind and doesn’t judge me like those #%*(#%( at the bank.” You can present as being kind to almost all of your customers and be obviously unemployable as a bank teller.

You will deal with thousands of customers. If you use “kind girl behind the counter” language about the 0.01% most aggravating customer once, you will not be a bank teller tomorrow. So bank tellers basically never use those words, and instead can inflect “Can I help you, sir?” in a way which leaves absolutely no doubt as to how welcome the new arrival to the branch is.

Then the economists running the survey typically scratch their heads and try to squeeze that feedback into homo economicus’ model of the world.

Readers might be thinking I am unnecessarily besmirching the good name of the field of economics here here, and so I will recount a representative sentence from a very good journal article verbatim:

To more effectively bring these unbanked individuals into the financial mainstream, it is essential for policymakers to recognize that these consumers have made these interdependent decisions in accordance with their marginal-cost-marginal-benefits calculations.

That is a lens of looking at reality, sure, and if you share that lens, you are welcome at my poker table any time. My marginal-cost-marginal-benefits calculations suggest your confidence in inferring ground truth from partial data is misplaced.

(Now you shouldn’t expect from me the level of academic rigor associated with scholars at e.g. Harvard; on the Internet, we cite our sources. That quote was originally published by the Chicago Fed and then the Review of Economics and Statistics. Full cite relegated to JSTOR for those interested. And again, as the literature in this field goes, that is a really strong entry. A particularly interesting finding is the relationship between being married and using check cashing businesses, which surprised me.)

Classes aren’t monoliths, of course. Many people assume you’d clearly never visit a currency exchange if you have a checking account. In time, the literature actually learned to ask that question and found that, decisively, no, some banked people do happily pay to cash checks. This observation caused some confusion. I am confused why this fact is confusing; some people who are capable of cooking have also been known to go to McDonalds, even though McDonalds isn’t free and doesn’t taste like a burger at home. That is, of course, one of the points of going to McDonalds.

Anyhow, if you want to dive deeper into that topic, look how various groups/papers/etc delineate between unbanked and underbanked. It will frequently come down to “Underbanked means you’re not unbanked but you still eat at McDonalds consume alternative financial services.” See this representative example.)

Check cashing on phones

So if we’ve identified the costs associated with retail establishments as one reason why check cashing costs money, and that many unbanked or underbanked customers do not have happy human interactions with financial service providers, you might wonder if technology can successfully cash checks while being cheaper and less judgemental about one’s presentation of class.

Yes, it can.

As always, I’m not endorsing any particular provider here, but let me point you towards a couple of models.

In one, a pure-play company like Ingo Money has both a direct-to-consumer offering and some ability to whitelabel it. (A whitelabel arrangement means that some other firm does the work of attracting the customers and then uses Ingo’s technology stack, financial rails, licenses, or any subset of the above to provide the check cashing service. To the customer, this feels a lot like they're "just" using the partner the whole time.) Ingo lets you essentially do remote deposit capture but instead of the deposit being to a bank account and there being a hold period, deposit is to an alternative financial product (like e.g. a prepaid card) and the funds are released basically instantly.

Take a look at Ingo’s fees and compare them to the representative fees for brick-and-mortar cash checking. For a $500 paycheck, it’s $5 versus $11.65 in favor of the app. This is the fintech dream; the delta is basically entirely “your costs are our margin.” Because the fintech nightmare is “our costs of customer acquisition explain why people still pay rent for commercial locations in high-traffic areas”, there is a loyalty program which gives steep discounts for repeated use.

And then there’s one decision which I just love aesthetically: if you’re willing to wait ten days, Ingo will discount your fee straight to zero. Why ten days? It is past the window where fraud discovery will result in the funds being clawed back plus (ahem) a bit of annoyance tacked on as a product decision.

I expect very few of their customers take them up on that. Most McDonalds burgers are not consumed by people who can cook a burger at home. They lack that capability; understanding this is an important part of understanding their life and the role of McDonalds in it. But free check cashing being available now where it wasn’t before is a straight-up win for the customers and the world.

Another model is embedding the check cashing into a larger suite of services. Cash App is a notable standout here (and, IMHO, probably the most interesting financial product the tech industry has created for the un-/underbanked.) On the backend, Cash App has simply convinced a bank to have a wider risk envelope as part of the price for working with a client who represents a very large portfolio of customers. The most useful thing Cash App can convince those users to do, which it spends substantial effort on doing, is signing up for direct deposit to their Cash App. (Really, it is to the partner bank, but from the user’s perspective eh it shows up in Cash App and all their friends use Cash App and you could turn it into actual paper money so does it really matter what some bank in Nebraska thinks? Nobody there will ever talk to you or talk down to you, so no.)

This is not, by itself, a full solution to the headaches implicit in banking people who are above average risk. You can read elsewhere about some of the implications.

There exist other models here, too, and other factors which are reducing need for this form of financial service. The continued march of direct deposit and earned wage access products keeps more of the wage pie in the pockets of workers, even those in diminished circumstances.

One useful thing the advocates (and others) accomplished over the last twenty years was convincing public benefits distributors to not solely use checks to distribute benefits. For example, "electronic benefit transfer" (EBT) replace the monthly check with a specialized variant of a pre-paid card which can be refilled. This sometimes results in the cost of distribution falling less on beneficiaries. Sometimes.

What do you want me to write about next?

The recent focus of checks has been a bit on the classical side of the financial industry and a bit depressing (see above). I’m eagerly accepting nominations for topics in 2024. Drop me an email or hit me up on Twitter (@patio11). As always, the writing calendar is driven by some combination of what people ask me about, what I find myself puzzling about on any given day, and what fills obvious holes in the Internet for the sort of people who’d read Bits about Money.

The title references a famous rap song: "Money, power, and respect / First you get the money / Then you get the #$(#%in', power / After you get the #$#(%in' power / Mutha*#(%#%s will respect you."

We recently covered the long shadow that checks cast on modern financial practice. The most important genre of check historically was the paycheck, settling an employer’s liability to their employee for services rendered. Most paychecks (and direct deposits, and earned wage access vehicles, and similar) are not sent directly by employers, but instead go through a payroll provider. Payroll providers are the original financial technology company, and were some of the first scaled adopters of computers generally.

You have almost certainly depended on payroll at some point in your life and probably never thought much of it. And it is fascinating! Let’s discuss a bit of history, a bit of politics, a bit of financial technology, and a heavy, heavy dash of extremely boring schlep in service of larger societal goals.

A bit of a disclaimer off the top: I’m indebted to a discussion with Ian Zapolsky at Check about some of the history and current practice of payroll processing companies. Check received an investment from Stripe, where I am an advisor (and past employee). Neither Check nor Stripe necessarily endorses what I write in my own spaces. Any errors in fact are, as always, my own. Credit for taste in financially-oriented rap music goes to ChatGPT.

Why does payroll exist, anyway?

Payroll, as an institution, exists to support the deputization of (mostly private) employers as tax collectors by the state.

In the United States, prior to the Civil War, almost all federal spending was funded by tariffs on alcohol or tobacco. The war was extremely intense and expensive, and the winners mostly chose to fund it by instituting a temporary income tax, at a flat rate of 3% of income above a relatively generous threshold. This might or might not have been legal, strictly speaking, much like the contemporaneous suspension of habeas corpus, but needs must when the future of the nation is on the line. (A note to non-American readers: you can Google this if you want to get into fascinating minutiae of U.S. history, but take note that a disturbing percentage of authors who are extremely invested in Abraham Lincoln’s supposedly tyrannical behavior on taxation are using that to distract the reader from the institution of slavery.)

The actual collection of this tax was from taxpayers directly. The combination of the tax being new, tax incidence being a relatively small portion of the population, and taxpayers being mostly on their own recognizance lead to a fairly large “tax gap”, which is the wedge between taxes theoretically owed to the government (per the law) and those actually reported by taxpayers.

America re-instituted the income tax in the early 1900s (via a constitutional amendment, this time) as a result of that most cherished of American traditions: widespread populist revolt over tax rates. Specifically, tariffs on imported goods hit the pocketbook of the emerging middle class the worst. The middle class exercised (and exercises) substantial political heft in America, and directed its representatives to move more of the tax burden to the wealthy. The term of art for this in tax policy is “progressive” taxation (where one’s effective tax rate goes up with income and/or wealth); any tax on consumption (including tariffs) will almost definitionally be regressive (the opposite), because the wealthy consume less of their income and/or wealth than other social classes.

The first implementation of it attempted to do lump sum collection, but taxpayers found this surprising and unpleasant. In particular, a large portion of the economy was still agrarian, and the farmer’s harvesting calendar is impossible to reconcile with a single fixed national tax deadline. (As both farmers and people who play Stardew Valley know in their bones, not all crops arrive on the same schedule.)

Enter tax withholding

Actors in government had a bright idea: instead of requiring a single lump sum payment of income tax annually, they could spread the payment over the course of the year. It would have been too operationally intensive for government to arrange this with every wage owner directly. Taxpayers would have revolted about the marginal paperwork burden, to say nothing of the constant reminder that they were paying taxes. So the state deputized their employer as a tax collector to accomplish it.

The mechanism would be withholding: the employer would, using a formula, estimate the taxes you would owe for your next payment, divide them ratably over the year, and withhold an estimated tax payment from each payment to you. They would “relatively swiftly” convey this payment, on your behalf, to the government over the course of the year. You would file a return once a tax year, reconciling your reported income and withholdings made by your employer, resulting in either you paying the difference to the government or, more commonly, receiving a refund from the government for tax overpayment.

This solves many operational issues for the government. It makes its own income much smoother and more predictable than under an annual payment regime. (This returned government to income schedules which more closely matched those of excise taxes. Although business owners might understandably feel otherwise, it is not in fact the case that all alcohol is consumed on tax day.)

Withholding also reduces fiscally consequential tax non-compliance. Most individuals who forget to file a return have already paid all taxes that are due, and may be forgoing a refund against their own interest.

It also has one side effect in the political economy, whose consequences we deal with a century later: withholding is quiet and sending in tax payments is loud. Most taxpayers, including relatively financially sophisticated taxpayers, do not carefully scrutinize their payroll withholding statements. Most do not understand how their payroll tax is calculated; it is simply a law of nature, handled by dutiful clerks somewhere. The only time they really think about tax withholding is once a year when the government pays them. In that blissful moment, their thought is not “Wow, I am paying so much in taxes!”

Tax withholding is one of a very few ways in which you can licitly bribe someone with their own money.

An aside about tax preparation software

It is widely believed in the tech industry that the reason the United States requires taxpayers to calculate their own tax returns, which is not required in many peer nations, is because Intuit (who make Turbotax, the most popular software for doing one’s taxes) spends money lobbying policymakers to oppose the IRS creating a competing product. People who believe this have a poorly calibrated understanding about the political economy of taxation in the American context. 

I will have to take notice about uncontroversial but politically inflected facts about the world we live in to describe why you must use the software you use. If you’d prefer to not get politics mixed in with your finances and software, mea maxima culpa. That said, a democratically accountable government which deputizes the private sector to achieve state aims is invariably subject to the political process, and this is on net a good thing. To the extent one has a complaint about the outcome, one's complaint is not with some unaccountable or corrupt actor in a smoky backroom somewhere. It is with one's countrymen.

In particular, the tech industry zeitgeist that blames Intuit for us needing tax preparation software fails to understand the preferences of Congressional representatives of the Republican Party. Any fairminded observer of U.S. politics understands the Republicans to be institutionally extremely interested in tax policy (and tax rates in particular), in the sense that doctors are interested in heart attacks. Their most recent platform includes the quote “Republicans consider the establishment of a pro-growth tax code a moral imperative. More than any other public policy, the way the government raises revenue—how much, at what rates, under what circumstances, from whom, and for whom—has the greatest impact on our economy’s performance.” This is far from the only flag proudly planted by the elected representatives who enjoy the enthusiastic support of about half of Americans for their views on tax administration.

The specific policy implications of those shared values are frequently outsourced, in a fashion extremely common in Washington and critical to your understanding of U.S. politics. Washington has an unofficial ecosystem of organizations and public intellectuals who, by longstanding practice, have substantial influence on policy. When a Republican candidate promises to voters that they are anti-tax, as their voters (particularly in primaries) demand they must be, the thing they will offer in support of that is “Grover Norquist gave me a passing grade.”

Norquist runs Americans for Tax Reform, a non-profit political advocacy group which opposes all tax increases. ATR is institutionally skeptical of withholding, because they believe that withholding allows one to increase taxes by stealth. I don’t think it is excessively partisan to say that, if one phrases that claim a bit more neutrally as “withholding increases tax compliance by decoupling public sentiment and policy changes,” the people who designed the withholding system would say “I’m glad the National Archives makes our design documents so accessible. We wrote them to be read!”

And, relevant to the question of whether Intuit controls U.S. tax policy: it can’t, because that would imply they have wrested control from Norquist. Norquist considers a public filing option a tax increase by stealth and opposes it automatically. (I offer in substantiation ATR’s take on a specific policy, which was bolded for emphasis in the original: “Americans for Tax Reform rejects the use of unauthorized taxpayer dollars being used to expand the IRS into the tax preparation business and urges states to reject participation in the program.” You can find much more in the same vein.)

Anyhow, back on the topic of withholding taxes: they still have political consequences, including consequences directly relevant to how individual Americans interface with their government, more than a hundred years later.

Withholding taxes were an operational disaster in early implementations

Calculating, collecting, and remitting withholding taxes was extremely labor intensive in the early years. Businesses used to have entire departments of payroll clerks, who needed to do per-employee calculations manually or on primitive mechanical calculators.

Fun fact: this was one of private industry’s first scaled uses of women in the workforce, and the job title was “calculator.” AI took our jobs and thank goodness it did. The typical workload of a (human) calculator persists in the present day only in Japanese megacorps which, and this is in no way an exaggeration, use it as a form of a psychological torture to coerce employees to quit “voluntarily.” OK, I guess that is less of a fun fact.

Withholding was designed without substantial input from the business community. It was so hated that it almost triggered a "tax revolt"; refusal to pay taxes as an act of protest. After World War One was over, the War Revenue Act (which had traded phasing out withholding in return for phasing in more onerous income reporting requirements) lapsed, and planners got a few years to debrief what had gone wrong. They had a better game plan ready for 1935’s rollout of the Social Security Act, which assigned Americans unique(-ish) identification numbers for the first time. This time, user research included America’s relatively nascent large national employers like e.g. Ford and GE, who exactly dictated what they’d need from the government to make withholding happen on a scale of hundreds of thousands of employees. (Think how many rooms of clerks that implied! The imagination boggles!)

In addition to the clerks, Big Business demanded the government accommodate modern practices like using specialized hardware from the Computing-Tabulating-Recording Company. It had recently completed a corporate rebrand to reflect their cutting-edge technological supremacy: International Business Machines. To a very real degree, the spec for U.S. withholding taxes became “What can be quickly calculated on a machine that IBM can deliver production units of this year? Write that law and we will send a check to Washington, every month, like clockwork. Or have it your way. Write another law. Then, you can come to Detroit and explain to the boys they need to give you what's coming to you.”

So what happens in payroll, anyway?

In broad strokes payroll processing hasn’t changed in a hundred years:

A business owner, or their payroll department, first calculate the amount of money they owe to employees in a pay period. The length of that pay period is established partly by custom (two weeks is most common in the United States, monthly is more common in Japan) and partly by operational complexity. Weekly was considered too exhausting for payroll clerks to contemplate decades ago and most businesses haven’t moved to it due to the persistence of tradition. (They have made some changes to get workers their pay faster, which is a topic we’ll return to.)

During the process of running a payroll, the payroll department or payroll provider will apply current laws to calculate deductions from payroll. Almost all of these will be taxes or tax-adjacent, like contributions to tax-advantaged employee retirement plans. There are some important exceptions which you absolutely must support, like e.g. garnishing the wages of parents who are delinquent on child support obligations (or, for that matter, certain delinquent debts owed to the nation).

In parallel with this calculation, money has to fan out. Is the service provided by payroll providers the calculation, the recordkeeping, or the money movement? Yes.

In U.S. practice, payday is almost inevitably on a Friday. The payroll provider will make an ACH debit against the employer’s bank account within a few hours of them running payroll, ideally (from its perspective) on Monday. This moves the total cash outlay for payroll in a single transaction to a settlement account owned by the payroll provider. Then, on Thursday night or early Friday, it will schedule payment via ACH push or other mechanisms to each employee. Typically, that will have that money be spendable in their accounts at Friday. (ACH direct deposit has almost entirely replaced physical paychecks.)

These are not the only payments made by payroll providers, and one might argue they are not the payments made which actually justify the existence of payroll providers, but they are the ones most people think of when they think of payroll. We will return to the other payments in a moment.

Why did businesses largely choose to outsource payroll processing? Because jurisdictions with taxing authority saw the federal government succeed with withholding taxes and copied the effort. There are many thousands of jurisdictions in the U.S. with this authority. Even at very small scales of business, decisions which seem very minor about who you employ suddenly pull in huge amounts of the collected tax code of jurisdictions which might feel Very Far Away to you. Keeping up with what laws and rulings you're subject to and the changes to them very quickly becomes a full-time job. In addition to being basically the first fintech, payroll processing is the first regulatory compliance layer as a service. (You'd predict from this, mostly accurately, that payroll processors are far less economically dominant in e.g. Japan, where taxing authority and collection is far more centralized than in the U.S.)

“Where is the risk transfer?”

That was my first question for Ian, and we both laughed ruefully at it. Almost every movement of data and money in finance also includes an (often underanalyzed!) risk transfer as part of the price of the services.

The biggest risk in payroll is that the business’ bank account doesn’t have money to cover the payroll debit in full, in which case the debit will be “returned” by the bank, resulting in the payroll provider not getting the money in their settlement account. Now, counting days on your fingers like all good financial technologists, you might think this is unlikely to be a problem. Monday plus two business days for an ACH return window is Wednesday. The money can’t leave before Thursday. No overlap in timeline means no risk.

So here’s the rub: that Friday deadline is, by law and custom, very hard. If you, as a business owner, fail to have money delivered to employees by Friday, you have “missed payroll.” This is one of the most legally consequential screwups a business owner can do. How developed is the law in this area? Missing payroll is listed in the Bible (James 5:4, etc) as a “sin crying out to heaven for vengeance.”

The Monday deadline, on the other hand? Eh, business owners sometimes get busy. I probably averaged Wednesday for payroll submission at the last company I ran.  “Thou shalt not submit thy CSV files on Wednesday” is not written in the Bible.

And so here is the product decision confronting every payroll provider: do we strictly impose the Monday deadline, or do we let favored customers slip it until, say, Wednesday? OK, maybe gulp Thursday afternoon? Or do we cede the custom of many, many employers to a provider capable of playing ball here?

The payroll industry overwhelmingly allows businesses to transform “late” payrolls into “on-time from the perspective of the employee” payrolls by taking balance sheet risk. They pay the employees before they know they will collect from the employer in a durable fashion. This is a considerable risk.

In the event a business misses payroll, the likelihood that it is not simply an operational stumble but actually insolvency is, over the universe of all firms, very, very high. Individual runs of payroll are very large relative to the fees earned by payroll providers; at my last company, two weeks of salaries/etc for engineers/etc was ~$200,000 against a payroll service charge of a few hundred dollars a month. Clearly, taking credit losses even at a very low incidence rate does not pencil out well. (A payroll provider will typically reserve the right to recover against employees if there is a return on a payroll run. Nobody likes that resolution. One reason: employees of a defaulting employer will frequently not have the money a day later, either, not least because their employer might have told them “Uh you probably don’t want to keep your last paycheck in the bank over the weekend, call it a hunch.”)

And so payroll providers, unlike most software companies in the we-run-a-complicated-spreadsheet business, will have to underwrite their customers as they onboard them. Most businesses in the economy are extremely thinly capitalized. Underwriting will often focus on reviewing books, demonstrable (past) cash flows versus anticipated payrolls, etc. Doing this well is make-or-break for payroll providers.

(Other risks payroll providers have to be cognizant of include account takeovers and fraud. “Payroll is a money hose”, as Ian mentioned, and by design will cause the payroll provider to spray money into accounts it has no history with and has minimal documentation on until the first transfer of four figures per account. A legitimate business which gets their payroll account taken over is in for a very bad day. A payroll provider which is convinced to run payroll on behalf of a business which either doesn’t exist or only exists to enable a later crime via a “long fraud” will also have a very bad day. There exist many people in finance and tech companies who think of little other than “How can we identify an account takeover before we or our client lose money in an irreversible fashion?”)

What about those other payments?

A lot of people have read about Warren Buffett’s investing genius over the years. One key to it was using the “float” earned by his insurance companies. Float is, in the insurance context, money which customers have paid you (premiums) which covers risks which you will (as an actuarial matter) owe them back someday. In the present day, you can invest it (subject to some constraints) and keep the returns for yourself.

Many people see float absolutely everywhere. “I bet a payment processor just rakes it in from float!” Well, not so much, because float being interesting requires a large amount of money multiplied by an extended timeframe multiplied by relatively high interest rates. Non-specialists tend to forget that “timeframe” bit and overestimate float by a factor of ten or more. (Payment processors are keenly aware that their customers want their money faster at virtually every margin and are keen on improvements to make this happen, even if they notionally cost a minor revenue opportunity.)

Which I mention to excuse the following: payroll firms actually do earn material amounts of float, in a fashion which is rare for financial technology businesses. Why? Precisely because payroll runs are large relative to fees for payroll services and holding times can be quite considerable.

We are not talking about the float between Wednesday and Friday primarily here. We’re talking about the float earned on “funds held for clients”, which mostly constitute money that they’ve withheld on behalf of particular employees but not yet remitted to the tax agency (or similar).

You can read the disclosures for your payroll firm of choice for the boring mechanics here. I recommend ADP’s, since it is a giant in the industry (~17% share nationwide) and publicly traded. ADP keeps client funds at a trust account at a trust bank that they established. (Was that decision voluntary? One might speculate that if you happen to have plural percentage points of all wages earned in the nation on your books on any particular Tuesday, almost all of it owed directly to the government, some people in Washington would really prefer you had it in a maximally regulator-legible wrapper versus another structure. Anyhow, OCC charter secured, congrats all around.)

So in economic substance, when your Chicago pizzeria runs payroll on Wednesday for payday on Friday, ADP constructively receives the money Thursday-ish and pays out the majority of it Friday-ish.

But the withholding taxes that the pizzeria owes to the IRS and to the Illinois Department of Revenue? Those ADP custodies until the next day they are required to pay it out, which will (in expectation) be about two weeks later, not about one day later. (Various payout schedules prevail for various taxes and various taxpayers, etc, but this suffices to show the general pattern.)

The money is custodied on behalf of the pizzeria—it is still their money—but the interest earned on it goes to ADP. It is a lot of dough: ADP earned more than $800 million on client funds in fiscal year 2023, representing about 18% of EBIDA. Payroll providers aren’t as dependent on interest income as discount brokerages but it is a very important part of the model. (The main revenue line, for ADP and other payroll providers, is the fee they charge businesses for services.)

An interesting detail here: payroll companies are a conveyor belt for money. That conveyor belt looks pretty short if you just count from the payroll being run to the final transfer of withholding tax. It starts to look much, much longer if you consider “Well, if a software company hires an engineer in 2023, and has attrition of 10% annually, in expectation haven’t they basically committed to paying something like 85-90% of current salary in 2025 once you count the impact of annual raises/etc?”

When you multiply that intuition over a large and diversified book of business, and subtract out the negligible churn rates of payroll providers, you (as a payroll provider) could get comfortable conceiving of your conveyor belt as being much, much longer than it looks at first glance. So should money on your conveyor belt be earning demand deposit interest rates? Oh clearly not, short-term Treasury rates will of course be higher and have even less risk. How about mid-term Treasury rates?

And if mid-term Treasury or agency MBS rates are higher than your corporate short-term borrowing rates, you could borrow short and lend long. How do you know you’ll have the cash flow to support the borrowing? Dude. Look at the conveyor belt of money which charges you nothing for borrowing from it.

Welcome to the wonderful world that is being a corporate treasurer at one of the fairly few companies where one’s fiscal asset allocation strategy is both really interesting and also really matters to results.

A detail of a detail here: the flipside of this being an excellent reason to be in the payroll business is, indirectly, one reason why payroll providers are so terrible. To do this licitly, you need to get a money transmitter license everywhere your customers want to pay as many as one employee, which (very quickly) means all 50 states and then some.

This is not hard relative to creating a fusion reactor, but is hard relative to creating a functioning Rails app. This means that the minimum cycle time to create a new functioning payroll provider is “a few years” unless you’re using an underlying provider via some sort of white label arrangement. Most ways to do that just end up replicating their infelicities, so really innovating requires all the licenses first, which implies years of boring scutwork prior to doing the valuable, fun, and lucrative bits of innovating on payroll software.

Where is the frontier in payroll?

It was a revelation a few years ago when one startup payroll provider (Gusto, about which I will say “I have used them as an employer and yet I have employer-side payroll stories that take longer to tell than ‘I ran payroll successfully’”) debuted a web application for employees which wasn’t terrible. Pedestrian applications of technology create enormous value in financial services, film at 11.

In terms of actually new capabilities, the hot new thing in payroll recently is called “earned wage access.” This is mostly for employees at the lower end of the socioeconomic ladder. We covered that the payroll cycle in the U.S. is generally two weeks, for legacy reasons owing to the workload capacity of payroll clerks. In principle, computers could trivially do it daily, but business owners largely have not decided to do that.

Employees low on the socioeconomic ladder are therefore, economically, extending their employers a loan of (on average) one week’s worth of wages at zero percent interest. They are themselves are often in debt, frequently paying high or unconscionable rates on that debt.

EWA attempts to make it incentive compatible to fix this. In it, the payroll provider convinces the employer to allow employees to opt-in to receiving their wages not as a paycheck or an ACH deposit into their bank account but as a deposit onto a prepaid card provisioned by the payroll provider. That deposit is made daily not biweekly. Critically, these wages are being advanced by the payroll provider and not actually coming from the employer, which is probably surprising. (This is free to the employer; no cost aside from the wages/etc you’ve agreed to pay.)

Why is this incentive compatible for the payroll provider, when it could be advancing wages to e.g. a pizzeria chef who might depart employment two days into that 14 day pay period? Because people who are relatively low on the socioeconomic ladder tend to spend most of their paycheck relatively soon after they receive it, and much of the spend will be by presenting that prepaid card to pay for e.g. groceries. Every business accepting the card will pay interchange to do so, in the neighborhood of 2-3%.

And so you can napkin math this out something like this: an employee earns 250 days of wages in the year. If they spend half of their wages via using the prepaid card (with the other half being tax withholding, rent, and similar things which will not result in the provider using interchange), the provider will earn something like 2.5 days of wages for doing financial engineering. They will also absorb, on average, 7 days on average of risk of separation from employment. And therefore, factoring in estimated annual risk of separation and anticipated recovery in event of separation, both of which can be measured empirically as you roll out the offering, there is some solution to the equation where the payroll provider earns a day’s wages without working a day.

Then, it does that for tens of thousands of employees in parallel. This is lucrative relative to earning the interest alone on a small percentage of a few weeks' wages, or on the rates businesses will tolerate to run payroll for one pizzeria employee at the margin.

Other fun innovations include giving people money in ways they find actually valuable. A major one for Check has been enabling instant EWA-style to employees’ Cash App accounts. (Cryptocurrency advocates will, inevitably, mention that this would all be so much simpler as less expensive if everyone just used stablecoins. Convincing U.S. workers to actually desire stablecoins as opposed to e.g. Bank of America dollars or Cash App dollars has been rather more evitable.)

Coming up next in Bits about Money

Merry Christmas, somewhat in advance! I’m planning on a recap issue for the year, likely on the Friday before Christmas.

If you’re looking for late Christmas gifts, wondering how to spend your continuing education budget while nervously glancing at the calendar, or perhaps want to book an expense in 2023 and save on taxes, Bits about Money offers paid memberships.

After the year in review, we’ll return to the usual potpourri of topics. One relatively close to the top of the list is the check-adjacent alternative finance products that are reinventing the paycheck cycle for people lower on the socioeconomic spectrum. (In contrast to EWA, above, some of them don’t require integration into the payroll provider and/or facilitation of the arrangement by one's employer.)

Speaking of payroll software, a thing I’d love to write at some point is why government payroll system modernizations are the most doomed and cursed of any software projects. Perhaps next year!

A recurring theme for Bits about Money is that financial infrastructure is heavily path dependent. We’re (oft unknowingly) standing atop decades and centuries of work which came before. Sometimes, the specific contours of that work, and of decisions made decades ago, is a roadmap to what continues to work today and what doesn’t.

Let’s talk checks.

A brief programming note: We're fast approaching the end of the year, and I still owe you all a few issues of Bits about Money, so you might be getting three issues or so prior to Christmas rather than the more typical publishing pace around here.

A brief digression for people who use functioning payment forms

Checks are the oldest form of widely deployed non-cash payments instruments. As we’ve covered previously, all a “payment” is in the payments industry is a message about the status of a debt. While almost all checks are pre-printed these days, the earliest ones were simply a handwritten letter to one’s banker, instructing him to pay money the bank owed to you to the party you named.

Checks became central to payment systems through a property called “negotiability.” The earliest checks look like a three-way transaction: buyer of a good or service (who wrote the check), seller (who turned the check into money), and the banker (who facilitated the transaction). It fairly quickly developed that checks were much more useful than this transaction appears on the surface: the person accepting the check could sell it, because the notion of a debt owed to you has value. This would allow them faster access to their cash and allow someone else an opportunity to profit from transactional friction, by reducing it.

There is a fascinating history of the negotiability of checks, of the deposits at different banks trading at varying (and constantly changing) discounts to par, and of checks being the primary form of paper money for more than a century in the United States. We’ll elide most of it here. Let’s focus on the most boring possible implication of negotiability: your bank should be willing to accept that other bank’s check.

That is an interesting usage of the word “should”, right? Should they, because it would be convenient to you? Or should they, because it is good business? Or should they, because other stakeholders in the financial system, like say the government, mandate it? Yes, in all these senses of the word “should.”

Check settlement in the pre-computer era

One of the most important standardizations that you’ve probably never thought of is the Uniform Commercial Code, which dates to the late 1800s. It was an early and extremely successful implementation of “model language” for statutes: each of the several United States was convinced to change local state law to match a change being made effectively simultaneously in every other state, without an explicit federal coordination mechanism.

The most lastingly important thing in the UCC is that it standardized checks. Instead of them being creatures of state contract law, dragging decades of precedent and complex bespoke negotiations behind every specimen, they became almost exactly describable by recounting a short description of the face of the check. We (very intentionally!) made checks “dumb” to allow the system around them to be much smarter.

The UCC facilitated banks clearing each others’ checks. (“Clearing” is a magic finance word. Clearing a check refers to completing the process which the check agrees to: the writer sees money leave their account and the person depositing the check sees it enter theirs. This is much more complicated than it sounds in this quick gloss.)

Note that “facilitated” does not imply “this then became very easy.” One infelicity which became increasingly stressed as improved transportation and communication technologies caused more commerce in the U.S. to be other-than-local: you had to physically move checks between banks to clear them. If there are two banks, then you need two messengers, each carrying a bag of checks once a day. If there are four banks, you need twelve messengers, because Bank A needs one each for B, C, and D, and each other bank needs the same. If there are thousands of banks, then each bank needs to… oh dear.

More than a century before we invented the computer science to describe what would have happened, we invented the solution to it: centralized clearinghouses. New York’s first was in the 1850s; London had already pioneered the mechanism a century earlier. Instead of your bank sending the check to the originating bank to clear it, it would send it to the central clearinghouse. Each bank thus only needs one messenger a day, not thousands. (As you can imagine, in the real world, there were actually multiple messengers carrying multiple bags. Still, the stylized truth of this is important.)

And thus we come to an important fact about the U.S. payments ecosystem: every way of moving money between banks was designed in relation to the capabilities necessary to facilitate nationwide clearing of paper checks. The most common method of interbank payments in the U.S. wears this history in its name: ACH stands for “Automated Clearing House.”

The historical timeline for interbank payment settlement in the U.S. comes to us, in a direct and at-times maddening fashion, from “How long did it take to move data between New York and San Francisco when the best available technology for this was rail networks, the transport protocol was handwritten paper, and the cloud computing was floors full of clerks doing repetitive mathematics using mechanical adding machines and AI (artisanal intelligence)?”

Some funny consequences of checks underpinning everything

Credit underlies much more than people generally believe it to. Checks are an unusually direct example of this. The capability to write checks requires a credit extension. The capability to accept checks requires a credit extension. Checks, by existing, promiscuously distribute credit. Promiscuously distributing credit has some knock-on consequences.

Consider the case where you pay for a basket of groceries with a check. (This probably strikes many in the audience as anachronistic. Work with me here because it’s actually terrifyingly relevant to modern payments systems.) At the point which you check out, you owe the grocery something, and promise to make good upon this debt. Retailers historically did a direct extension of credit at that point; users would settle up weeks or months later.

Checks move the extension of credit (in part!), from the store to the banking system. At the point which you write the check, the grocery has to make a decision whether to release your chicken, with extremely imperfect information. Your account may or may not have good funds in it presently; the store cannot know. More importantly, the store cannot know whether your account will have sufficient funds when the check is presented a few days in the future.

Why did we do this?! Why accept mere promises, promises conditional on unobservable future events? Because systematically taking relatively small amounts of risk created immense value. Groceries prefer selling more groceries to selling less. Extending consumers credit tends to increase the number of chicken breasts they consume at the margin. Consumers prefer to eat chicken versus going hungry. Chicken do not get a vote.

Note that the extension of credit is recursive and systemic here. The grocery store will issue paychecks; those paychecks each embody credit risk. (We will return to the fascinating topic of credit risk in payroll systems in a later issue.) That credit risk in part comes from the uncertainty as to whether your check for the chicken was good. The credit risk there comes from factors like your general capitalization, degree of conscientiousness regarding financial management, and whether your most recent paycheck is good. It is credit risk all the way down.

One control which we made for checks to reduce systemic risk continues to have consequences more than a century later. Most disagreements between you and a grocery store are beneath the notice of the law. If you and your grocery store have a disagreement about a check specifically, you can go to jail. The crime is sometimes called “uttering”, for charming historical reasons. 

It is well within the norms of checking accounts for them to become overdrawn as e.g. users miscalculate the amount of money they have on deposit, timing issues with the posting of checks break in ways that are not perfectly predictable in advance, and users make mistakes. This is something that product managers at banks design into the unit economics of checking account. NSF (insufficient funds) fees were for at least two decades or so an important part of the revenue mix.

We arrested and jailed, and continue to arrest and jail, at least some people for something which looks very, very similar to what earns other people a $30 NSF fee. (And, of course, if you’re a desirable banking customer, they’ll waive that fee because the waiver is simply good business.) In theory, uttering is distinguished from simple overdrafting by there being an element of intent to defraud. In practice, uttering is when you’re a poor person using a checking account and a string of people reviewing your behavior view it unsympathetically. If a grocery store loss prevention employee, a police officer, and a prosecutor each refuse to stop the process, NSF means you’re Now Somebody's Felon.

Sometimes your behavior will look unsympathetic because you are a poor person. Sometimes your behavior will look unsympathetic because you definitely were trying to pull one over on the grocery store. Sometimes your behavior will look unsympathetic because you have done this many, many times before. Sometimes these are all the same picture.

Phrased that way, it sounds almost fantastically unjust. And… it’s complicated. Many people, of all social classes, consumed many chickens obtained on credit because certain specially-formatted pieces of paper were believed to be money. That belief relied, in material part, on a state guarantee that they were money. That guarantee was backed by the state expressing substantial displeasure about the abuse of certain specially-formatted pieces of paper.

And from this follows an underappreciated consequence of modern financial infrastructure: credit cards and debit transactions make the economy more efficient. One way you can measure that efficiency gain is that their users rarely go to jail. Partly that is because you can do a near real-time prediction of whether the transaction has good funds behind it at the point-of-sale. Partly it is because we more fully move credit risk from the grocery store to the bank. 

Society frequently considers the sophistication of different participants in determining their level of legal recourse. We consider grocery stores presumptively worthy of a relatively high degree of protection; we will jail someone over $20 of chicken. We consider banks usually capable of doing their own risk analyses, and if they make an underwriting decision that costs them $2,000, well, society assumes banks have people who are good at math. Part of that math is “credit card issuance is fantastically lucrative, in part because you can charge the grocery store when it sells chicken to someone using a credit card.” And so part of the cost of accepting credit cards is the cash rewards, part is for computers, and part pays for “not jailing poor people.”

Money in transit

It’s useful to understand the historical physical reality of check clearing because we loved this model so much we re-used it for most later forms of payments. Banks which routinely cleared checks from particular other banks, due to geographic proximity or because they were large institutions in e.g. New York, would open correspondent account relationships with each other. A correspondent relationship is simply a bank having a bank account inside another bank. Bankers historically use Latin to make this sound more complicated than it is. (For shibboleth value: nostro is “our account at your bank”, vostro is “your account at our bank.”)

This made collecting payment more efficient: instead of frequently having money move between banks, you could total up all of the incoming payments (checks presented by your customers drawn against the other bank today), total up all of the previous period’s outgoing payments (checks presented by the other bank’s customers drawn against you a while ago), “net” those against each other, and then make a single internal accounting entry against your correspondent’s vostro. Your two banks would then only periodically rebalance where they held their money, which in the old days involved sometimes literally sending a stagecoach over with gold or silver and in more recent days would typically take the form of a wire transfer.

This relationship decreases the amount of money in transit at any given time, which has important operational efficiencies and decreases credit risk. We have made many, many more improvements over the years.

One intellectually interesting one: the Check 21 Act was designed to bring check clearing into the 21st century. The most important part of it was bulldozing a coordination problem: banks were split on whether the paper part of paper checks actually mattered or not. Check 21 said definitively “Your opinion as a bank is irrelevant: an electronic reproduction of a check is the same as a check. If you absolutely need physical paper, your counterparties are allowed to write that electronic reproduction onto completely new physical paper then send it to you.”

Why did we do this? Because it greatly speeds up presenting checks for payment. They get scanned in at one institution then transmitted electronically and not physically to the clearinghouse. The clearinghouse then will generally electronically send it to the institution the check is drawn on. This can save literally days. It is another step in making checks dumber to make the surrounding system smarter: the check is almost entirely vestigial at this point.

So how could anyone have been against this? In a recurring theme for this column, many small banks didn’t believe that they had the prompt technical capacity to receive their checks reliably electronically. It was 2003, a surprising portion of small banks still had immaterial usage of the Internet in their operations. Most of them did not own their technical destiny but rather are at the mercy of various vendors. (Plus ça change.)

Check 21 said “OK, the rest of the industry hears that explanation, and we actually sympathize a bit, but we won’t let you block an industry-wide improvement. We will instead give you a carve-out: you and you alone will still get the daily delivery of a lot of paper. It will just be new paper, with substitute checks printed on it, from a printer we have arranged to locate very close to your check processing address. We will legally compel you to treat that paper exactly like the special magically formatted check paper. You are very good at processing that sort of paper, because you’ve done so for decades; even though you are small, you are still a bank and still operationally competent.”

More interestingly to most readers of this column, Check 21 also paved the way for remote check deposit as a product. Your bank’s mobile app very likely allows you to take a picture of a check to deposit it. This desirable user experience would be much less possible if the mobile app had to, after you had taken a photo, read the routing number and tell you “Actually, sorry about this: that bank needs the check presented physically to pay it. We understand you’re angry at us, because we’re your bank, but this is out of our hands.”

That sounds crazy, right? But you’re dealing with that exact argument today for why you mostly can’t do instantaneous payments between U.S. banks. The system which would allow it (FedNow) is opt-in for both sending institutions and receiving institutions. There are legitimate and heady reasons why some financial institutions think operational and risk issues would make supporting FedNow very difficult for them to tackle. With those institutions as holdouts, the network graph of “Who could you send money to from your current bank account?” is currently very sparse. This makes it less likely for your bank to both a) decide to join as a sender and b) actually expose that sending capability in the app, because they know it will generate annoyed customers and little else. Instead, they’ll continue pushing you to (slower) ACH payments and (broadly worse) Zelle payments. (An in-depth description of why Zelle is not a particularly good product qua bank payments will probably have to wait for another day.)

Deposit accounts and their discontents

A final funny wrinkle for the moment: the “standard” bank account in the United States is a checking account. (This is not true internationally; checking accounts are sufficiently weird in e.g. Japan that if you have one you’re assumed to be much closer to Toyota than to a natural human. Having a checking account is considered weird because Japan has its own quite path-dependent history in the coevolution of bank accounts and business practices. Back home, checks are used almost exclusively for large commercial transactions and routine person-to-business or business-to-person payments happen via e.g. instantaneous bank transfers.)

Since the standard U.S. bank account is a checking account, even if it cannot write checks, it is necessarily a credit product. Banks must manage credit risk. This makes opening up standard bank accounts far higher ceremony than it is in many peer nations. You can (easily) be declined services, for example because your previous use of bank accounts landed you in ChexSystems, an industry-wide niche credit reporting agency which basically exists solely to blacklist people from accessing checking accounts in the future.

This is not simply because banks hate poor people and want them to suffer. Routine use of a bank account in the United States will not infrequently cause a credit loss. Margins on small bank accounts are very thin; credit losses can easily be larger than several years of them. Some people will walk away from that credit loss, in which case the bank (basically) eats it. One choice which keeps bank accounts broadly available to people, including people who do not have demonstrable assets or credit history, is defaulting to not handing out bank accounts to people who demonstrably cause credit losses and then walk away from them.

And so people, including in the financial industry, who want to provide financial services to people in economic precarity run headlong into this. "Creditworthy" sounds like a value judgement but is, simultaneously, just a prediction about the weather. Some places see more rain; some customers see more credit losses. Frequently advocates who want to bank the un- or underbanked are also simultaneously furious at the banking industry for improvidently extending credit.

How have we dealt with this? We have seen a fascinating boom in products which exist along a spectrum from “definitely not a bank account but has some characteristics you’d associate with one” to “a full, traditional checking account.” We’ll cover that another day.

James Bond films have a certain formula to them. It is more interesting when seen from the perspective of the villain.

He has long been adjacent to money and power, but craves more. Several years ago, he successfully escaped his low-on-the-ladder job at an existing institution. He built a base of power that is independent of institutions. From it, he successfully puppets any organization he needs to. He and his organization are from elsewhere, everywhere, all at once. They have no passport and fly no flags; these concepts are thoroughly beneath them. They move around frequently and are always where the plot requires them to be, exactly when it requires them to be there. No law constrains them. Governments scarcely exist in their universe. To the limited extent they come to any government’s attention, no effective action is taken. The villain rises to the heights of influence and power.

This continues for years.

Then we suddenly hear E minor major 9. We begin the film, telling the end of the story, mostly from Bond’s perspective. The villain is just another weirdo who dies at the climax in act three.

Life imitates art

For years I have used the phrase “Bond villain compliance strategy” to describe a common practice in the cryptocurrency industry. 

In it, your operation is carefully based Far Away From Here. You are, critically, not like standard offshore finance, with a particular address in a particular country which just happens to be on the high-risk jurisdiction list. You are nowhere because you want to be everywhere. You tell any lie required to any party—government, bank, whatever—to get access to the banking rails and desirable counterparties located in rich countries with functioning governments. You abandon or evolve the lie a few years later after finally being caught in it.

Your users and counterparties understand it to be a lie the entire time, of course. You bragged about it on your site and explained it to adoring fans at conferences. You created guides to have your CS staff instruct users on how to use a VPN to evade your geofencing. The more clueful among your counterparties, who have competent lawyers and aspirations to continue making money in desirable jurisdictions, will come to describe your behavior as an “open secret” in the industry. They will claim you’ve turned over a new leaf given that the most current version of the lie only merely rhymes with the previous version of the lie.

And then we begin the third act.

So anyhow, Binance and its CEO Changpeng Zhao (known nearly universally as CZ) have recently pleaded guilty to operating the world’s largest criminal conspiracy to launder money, paying more than $4 billion in fines. This settles a long-running investigation involving the DOJ, CFTC, FinCEN, and assorted other parts of the U.S. regulatory state. Importantly, it does not resolve the SEC’s parallel action.

How’d we get here?

A brief history of Binance

Binance is, for the moment, the world’s largest crypto exchange. Its scale is gobsmacking and places it approximately the 100th largest financial institution in the world by revenue. The primary way it makes money is exacting a rake on cryptocurrency gambling, in particular, leveraged bets using cryptocurrency futures. To maintain its ability to do this, it runs a worldwide money laundering operation with the ongoing, knowing, active participation of many other players in the crypto industry, including Bitfinex/Tether, the Justin Sun empire, and (until recent changes in management) FTX/Alameda.

In his twenties CZ worked in Japan (waves) and New York for contractors to the Tokyo Stock Exchange then Bloomberg. In about 2013 he got interested in crypto and then joined a few projects, including becoming CTO of OKCoin, another Bond villain exchange. Being a henchman is an odd job, so he decided to promote himself to full-fledged villain. In 2017 he did an unregistered securities offering (then commonly spelled “ICO”) for Binance. 

Binance rose meteorically from then until recently, essentially gaining share at the expense of waning Bond villains. To oversimplify greatly, it carved up the less-regulated side of the crypto market with FTX, with Binance mostly taking customers in geopolitical adversaries of the U.S. (most notably greater China) and FTX mostly taking them in geopolitical allies (most notably, South Korea, Singapore, and the U.S.). But the cartels did not partition the globe in a way which fully insulated them from each other.

These operations were intertwined and coordinated. How intertwined? Binance was a part-owner of FTX until SBF decided successfully capturing U.S. regulators was a lot more likely if his cap table named more Californian trees and fewer Bond villains. How coordinated? The name of the Signal chat was Exchange Coordination.

This eventually led to grief as CZ (mostly accurately) perceived SBF was using the U.S. government as a weapon against Binance. He retaliated by strategic leaking, leading to a collapse in the value of FTX's exchange token, a run on the bank, and FTX's bankruptcy.

Where was Binance in all of this?

Binance did a heck of a lot of business in Japan in the early years. This officially ended in March 2018 when the Financial Services Agency, Japan’s major financial regulator, made it extremely clear that Binance was operating unlawfully in serving Japanese customers without registering in the then-relatively-new framework for virtual currency exchange businesses.

As an only-sometimes-following-crypto skeptic, this was the thing which brought Binance to my attention. Binance was piqued, saying that they had engaged the FSA in respectful conversations and then learned they were being kicked out of the country from a news report. Having spent roughly twenty years getting good at understanding how Japanese bureaucratic procedures typically work, I surmised  “...that is a very plausible outcome if you start your getting-to-know-you chat with ‘Basically I am a James Bond villain.’” I think that was the first time the metaphor came to me.

The order expelling them listed their place of business as Hong Kong, with a dryly worded asterisk stating that this was taken from their statements on the Internet and “...there exists the possibility that [this information] is not accurate as of the present moment in time.”

That, Internets, is how a salaryman phrases “I am absolutely aware that you maintain a team and infrastructure in Japan.”

Did Binance exit Japan? Well, that depends. Did CZ personally return to Japan? Probably not. Does Binance continue to serve Japanese customers? Yes, though (Bond villain!) it pretends not to. Where does Binance’s exchange run as a software artifact? As a statement of engineering fact: in an AWS data center in Tokyo. ap-northeast-1, if you want to get technical.

 (Someone needs to write an East Asian studies paper on how Tokyo became Switzerland for Asian crypto enthusiasts due to a combination of governance, network connectivity, latency, and geopolitical risk. I nominate anyone other than me.)

Binance also maintained an office in Shanghai, with many executives working there. It was raided by the Chinese police. Binance denied that the office existed. The spokesman’s quote was pure Bond villain: “The Binance team is a global movement consisting of people working in a decentralized manner wherever they are in the world. Binance has no fixed offices in Shanghai or China, so it makes no sense that police raided on any offices and shut them down.”

This was a lie wrapped around a tiny truth. Internet-distributed workforces containing many mobile professionals do not exactly resemble a single building with all your staff and your nameplate on the door.

Of course, on the actual substantive matter, it was a lie.

We know it was a lie, because (among many other reasons) we have the chat logs where the parts of their criminal conspiracy that operated in the U.S. complain that the parts of their criminal conspiracy that operated in Shanghai kept information from them that they needed to do their part to keep the crime operating smoothly. Coworkers, man.

Binance’s Chief Compliance Officer, one Samuel Lim, apparently is not a fan of The Wire and never encountered Stringer Bell’s dictum on the wisdom of keeping notes on a criminal conspiracy. He writes great copy, most memorably “[We are] operating as a fking unlicensed security exchange in the U.S. bro.” He and many other Binance employees have helpfully documented for posterity that their financial operations teams were, for most of corporate history, working from Shanghai.

Binance also operated in the state of Heisenbergian uncertainty, sometimes known as Malta. Malta has a substantial financial services industry, which welcomed Binance with open arms in 2018 and then pretended not to know him in 2020. This continues Malta’s proud tradition of strategic ambiguity as to whether it is an EU country or rentable skin suit for money launderers. ¿Por qué no los dos? Despite this, Binance would continue claiming to customers and other regulators for a while that it was fully authorized to do business by Malta.

Binance operates in Russia, to enable its twin businesses of cryptocurrency speculation and facilitating money laundering. In 2023 it pretended to sell its Russian operations.

Binance operated in many jurisdictions. The U.K.: kicked out. France: under investigation. Germany, the Netherlands, etc, etc, they required non-teams of non-employees at non-headquarters to keep track of all the places they weren’t registered doing their non-crimes.

A core cadre of the Binance executive team is currently in the United Arab Emirates, where CZ hopes to return. He professes that he will await sentencing there, and pinkie swears that he will totally get back on a plane to the U.S. to show up to it. For reasons which are understandable by anyone with more IQ than a plate of jello, the U.S. is skeptical he will make good on this promise, and is currently, as of Thanksgiving 2023, attempting to keep him in the U.S. He is physically present to sign what Binance advocates believe is the grand compromise to put all his legal worries behind them.

A defining characteristic of Bond villains is that they think they are very smart and everyone else is very stupid. To be fair, when you play back the movie of the last few years of their life, they keep winning and their adversaries look like nincompoops.

Then, they get extremely confident and begin to make poor life choices.

How did this work for so freaking long?

Much like the optimal amount of fraud is not zero, the global financial system institutionally tolerates (and actively enables) some shenanigans at the margin. You can think of Binance, Tether, FTX, and all the rest as talented amateurs capable of engaging the services of professionals. They followed advice and grew like a slime mold into the places where shenanigans are wink-and-a-nudge tolerated.

Why tolerate shenanigans? Some shenanigans are necessary to keep the world spinning.

China has grown into an economic superpower via capitalism while also at times officially having private property ownership be illegal. That circle cannot be squared. We, the global we, want Chinese people to not live in grinding poverty. That requires economic growth. Economic growth required making things the world wanted. Selling those things required integration into the global economic order. That required a willingness to ignore things the Communist dictatorship said were crimes, while simultaneously saying “Oh, bankers definitely, definitely shouldn’t facilitate billionaires committing crimes.”

As I’ve remarked previously, we similarly have complicated preferences with regards to Russian oligarchs. In some years, money laundering for them is, how might a gifted speaker phrase this, “[b]ringing our former adversaries, Russia and China, into the international system as open, prosperous and stable nations.” In other years, money laundering for them is described as funding Russia’s war machine.

Finance is messy because the world is messy.

Some of the shenanigans aren’t strictly necessary or planned, but society considers an expenditure of effort required to curtail them to be wasteful or to compromise our other goals. We had all the technology required to CC regulators on every banking transaction years before slow database enthusiasts decided all transactions would eventually be publicly readable and persisted forever. We simply chose not to implement it. It would have been quite expensive and infringed on the privacy of many ordinary people and firms.

But Binance, and others, forgot the critical step, to the annoyance of their engaged professionals: you have to eventually stop growing and keep to a low profile. You have to simply be content with being fantastically rich. If you do, you can continue showing up to the nicest parties in New York, owning expensive real estate in London, and commuting to a comfortable office in Hong Kong or the Bahamas or many other places.

But crypto kept growing until the control systems could not ignore them any longer. And the control systems cannot continue to avoid knowledge of the crimes.

So, so many crimes. Many of them are what crypto advocates consider as utterly inconsequential, like serially lying on paperwork. And also Binance gleefully and knowingly banked terrorists and child pornographers. That’s not an allegation; that has been confessed to. There is no line a Bond villain will not cross. They will cross them performatively.

And, surprising even me, some crypto characters consciously adopt the aesthetic of Bond villains. Le Chiffre, the villain in Casino Royale, owns a fictional house. That house exists in the physical world, where a location scout said “This certainly looks like the sort of place a Bond villain would live.” Jean Chalopin owned that literal, physical house. (c.f. Zeke Faux’s Number go Up, Kindle location 1175.) As previously discussed, Chalopin is a professional bagman, and his largest client was previously Tether.

What happens to Binance now?

Some believe that Binance admitting to being a criminal conspiracy is actually good news, not merely in the memetic “good news for Bitcoin” sense, but because this upper-bounds Binance’s exposure somewhere below “The United States forcibly dismantles the most important crypto exchange and much of the infrastructure it touches.”

The immediate consequences are about $4 billion in fines. Despite being one of the world’s largest hodlers, the U.S. will not accept payment in Bitcoin, and Binance has agreed to pay in installments over the next two years. CZ and Binance will be sentenced in February.

Some people think the grand bargain was to avoid him getting imprisoned. The actual text of the agreement says that Binance gets to walk away from some parts of it if he is sentenced to more than 18 months. (Senior officials told the NYT they are contemplating asking for more than that.)

Probably more consequentially, the settlements are going to force Binance to install so-called monitors internally. Those monitors are effectively external compliance consultants, working at the expense of Binance in a contractual relationship with them, but whose true customer is the United States. The monitors have pages upon pages of instructions as to exactly how they are to reform Binance’s culture by implementing recommendations to bring onboarding, KYC, and AML processes into compliance with the law everywhere Binance does business, and sure, that is part of the job.

But the other part of the job is that they’re an internal gateway to any information Binance has ever had, or will ever have. This can be queried essentially at will by law enforcement, with Binance waiving substantially all rights to not cooperate.

You might reasonably ask “Hey, doesn’t the U.S. typically require a warrant to go nosing about in the business of people who haven’t been accused of a crime?” And, to oversimplify half a century of jurisprudence, one loses one’s presumption of privacy if one brings a business into one’s private affairs. All of Binance’s customers and counterparties gave up their privacy to Binance by transacting with it. The U.S. has Binance’s permission to examine all of Binance’s historical, current, and future records, at will, for at least the next three years. It also secured a promise that Binance would assist in any investigation.

And so, if one were hypothetically not yet indicted by the U.S., but one had hypothetically done business with one’s now-confessed money launderer, one’s own Fourth Amendment protections do not protect the U.S. from hoovering up every conversation and transaction with Binance.

All of this is certainly good news and we can put this messy chapter behind us, say crypto advocates.

How are Bond villains actually regulated?

Was the Bond villain strategy ever going to work? Did Binance have a reasonable likelihood of prevailing on jurisdictional arguments, like telling the U.S. that the Binance mothership had no U.S. presence and so it should not be subject to U.S. law? No. Crikey, no. The system has to be robust to people lying or acting from less-salubrious jurisdictions, at least to the extent it cares about being effective, and at least some of the time it does actually care about being effective.

The U.S.’s point of view on the matter, elucidated at length in any indictment for financial crimes, is that if you have ever touched an electronic dollar, that dollar passed through New York, and therefore you’ve consented to the jurisdiction of the United States. Dollarization is very intentionally wielded like a club to accomplish the U.S.’s goals.

There exist some not-very-sympathetic people one could point to who ran afoul of this over the years who are still much more sympathetic than Binance. Binance intentionally used the U.S. market and infrastructure to make money. The U.S. was essential to their enterprise. Many peer nations can, and will, make a similar argument.

Binance had tens of percent of their book of business in the U.S. They were absolutely aware of this, knew that some of those users were their largest VIPs or otherwise important, and took steps to maximize for U.S. usage while denying they served Americans.

Their engineers didn’t accidentally copy the exchange onto AWS or deploy it to Tokyo by misclicking repeatedly. The crypto industry playbook for doing sales and marketing looks like everyone else’s playbook for doing sales and marketing. They get on planes, present at events, send mail, hire employees (or otherwise compensate agents), open offices, etc etc.

If having an email address meant you didn’t exist in physical reality anymore, the world would be almost empty.

CZ personally signed for bank accounts for some of his money laundering subsidiaries at U.S. banks, like Merit Peak and Sigma Chain. The SEC traced more than $500 million through one of those accounts.

One major rationale for KYC legislation, as discussed previously, is that it makes prosecuting Bond villains easier. Even if compliance departments at banks are utterly incompetent at detecting Bond villains at signup, having extracted the Bond villain’s signature on account opening documents is very useful to prosecutors a few years down the road. Why have to do hard work quantifying exactly how many engineers work on which days at Binance’s offices in San Francisco when you can do the easy thing and say “Hey, fax me the single piece of paper where the Bond villain signed up for responsibility for all the crimes, please.”

Why do Bond villains sign for bank accounts in highly regulated jurisdictions? Partly it is because of beneficial ownership KYC requirements to open bank accounts. Partly it is because finding loyal, trustworthy subordinates is very hard if you’re a Bond villain, and Bond villains (sensibly!) worry that if the only name on the paperwork is a henchman, eventually that henchman might say “You know, actually, I would like to withdraw the $500 million I have on deposit with you.”

(This is why Bond villains frequently have e.g. the mother of their children sign for bank accounts. Bond villains, again, think everyone else is stupid, and that no one will cotton onto this.)

A subgenre of challenges in people management for Bond villains: you have to hire experienced executives in the United States to run the U.S. fig leaf for your global criminal empire. The people you hire will, by nature, be experienced financial industry veterans who are extremely sophisticated and have access to good lawyers. This combination of attributes is the recipe for being the best in the world at filing whistleblower claims. I expect a few previously executives at Binance U.S. are eventually going to take home the most generous pay packages in the entire financial industry for a few years between 2018 and 2022.

To make this palatable to the American public, those whistleblower rewards are not courtesy of the taxpayer; they’re courtesy of money seized from previous Bond villains. A portion of Binance’s settlement(s) will go to pay the whistleblowers at the next Bond villain. It’s a circle of life.

News that will break shortly

Different regulators have differing ability to prosecute complex cases, but they basically all have the ability to read simple legal documents. That is one of the things they are best at doing.

Binance will suffer a wave of tag-along enforcement actions, in the U.S. and globally. Partly this will be for face saving; global peers of the U.S., which Binance has transacted billions of dollars in, will largely not want to signal “Oh we’re totes OK with money laundering for terrorists and child pornographers”, and so they’re going to essentially copy/paste the U.S. enforcement actions. They will then play pick-a-number with Binance’s new management team, who will immediately cave.

The earliest version of this is probably only weeks away, but Binance will deal with it for years.

More interestingly, and likely more expensively, the SEC is going to hit Binance like a ton of bricks. They were one of the few regulators which opted out of consolidating with the DOJ’s deal. They think they have Binance dead to rights (they do), and tactically speaking, the deal makes their life even easier. Binance has waived ability to contest some things the SEC will argue. The SEC can now proxy requests for evidence to Binance’s monitors through other federal agencies.

Binance has had the enthusiastic cooperation of many people who walk in light in addition to their co-conspirators who walk in shadow. Those people, lamentably inclusive of some in the tech industry who I feel a great deal of fellow-feeling for, are going to start cutting off access to Binance. Compliance departments at their corporate overlords, which were either entirely in the dark or willing to be persuaded that a new innovative industry required some amount of flexibility with regards to controls, are (today) having strongly worded conversations which direct people to lose Binance’s number.

Binance has pre-committed to helping with the efforts to cauterize them from the financial system. They also pre-committed to assisting in, specifically, the investigation of their sale of the Russia business. That investigation will conclude that the sale was a sham (a Bond villain lied?) designed to avoid sanctions enforcement. Ask your friends in national security Washington how well that is going to go over.

Binance is going to be slowly ground into a very fine paste.

Many crypto advocates believe the U.S. institutionally wants to see Binance reform into a compliant financial institution. They are delusional. The U.S. is already practicing their lines for the next press conference. This course of action allows them to deflate Binance gradually while minimizing collateral damage, which responsible regulators and law enforcement officials actually do care quite a bit about.

The U.S. is aware that many high-status institutions and individuals, which are within the U.S.’s circle of trust, actively collaborated with Binance. Most of them will escape serious censure.

A few examples will be made, especially in cases where it is easy to make an example, because the firm is no longer operating financial infrastructure. This will take ages to happen and be public but relatively quiet, insofar as senior U.S. regulators will not get on TV to make international headlines announcing it. It will be one of the stories quietly dribbled out on a Friday to the notice mostly of people who draft Powerpoint decks for Compliance presentations. If you want a flavor for these, join any financial firm and pay attention during the annual training; you’re stuck going to it, anyway.

You seem a little smug, Patrick

I’m not breaking out the Strategic Popcorn Reserves yet, but I will admit to a certain amount of schadenfreude here. The world was grossly disordered for many years. It has corrected a relatively small amount.

We are a nation of laws. I’d support reforming some of them; a lot of the AML/KYC regulatory apparatus harms individuals who have done no wrong. Much is not well-calibrated in terms of societal costs versus occasionally facilitating a Bond villain’s self-immolation.

However, in the interim, one cannot simply gleefully ignore the laws because the opportunity to do so allows you to become wealthy beyond the dreams of avarice. Even staunch crypto advocates looking at Binance’s conduct see some things they are not happy to be associated with. Not all of the crimes were victimless crimes.

There exists the possibility that there is some salvageable licit business in crypto. People enjoy gambling. But if you factor out the crime, the largest casino in the world is not that interesting a business relative to the one that Binance et al ran the last few years.

I do not know if we’ll ever have a world with this scale of crypto businesses without the crime. The crime was the product. An opportunity to transform global financial infrastructure was greatly overstated and has not come to pass. I do not expect this to change.

The New York Times recently ran a piece on a purported sudden spate of banks closing customer accounts. Little of it is surprising if you have read previous issues of Bits about Money. The reported anecdotal user experiences have a common theme to them. Banks frequently present to their users as notably disorganized, discombobulated institutions. This is an alarming and surprising fact for the parts of society that are supposed to accurately keep track of all of the money.

Why does this happen? Why does it happen across issues as diverse as bank-initiated account closures, credit card or Zelle fraud, debit card reissuance, and mortgage foreclosures? Why does it happen in such a similar fashion across many institutions, of all sizes, who exist in vicious competition with each other and who know their customers hate this?

Banks are extremely good at tracking one kind of truth, ledgers. They are extremely bad at tracking certain other forms of truth, for structural reasons. In pathological cases, which are extremely uncommon relative to all banking activity but which nonetheless happen every day and which will impact some people extremely disproportionately, the bank will appear to lack object permanence. Every interaction of the user with it feels like being Bill Murray in Groundhog Day: the people you’re talking to remember literally nothing of what they’ve promised before, what you’ve told them, and the months or years of history that lead to this moment.

How did we end up here?

Recordkeeping systems

Like every bureaucratic system, banks run on a formal system of recordkeeping which requires an unrecognized, illegible shadow system to actually function. The interactions between those systems, and what they are optimized for tracking and not optimized for, cause a lot of the pathologies that people see. The seminal text on this, focused on government bureaucracies, is Seeing like a State.

Because banks are filled with extremely creative people, we call the primary system banking is conducted on a “core.” The largest banks in the world have complicated bespoke subsystems for this, but most banks are not in the software development business, and instead license a system from a so-called core processor like Jack Henry or Fiserv.

One could fill a book with architecture diagrams for a mid-sized financial institution. The key thing that non-specialists need to understand is a) the “core” does a lot of what you think of banking as, b) the core interfaces with many other systems which make up a bank, c) in particular, the core interfaces with the ledgers of the bank, and d) all of these systems together cannot represent reality nearly as well as you’d hope.

They typically grow over the years by accretion, caused by the normal processes of software development, regulatory changes, and competitive pressures. No system will ever be able to answer all interesting questions about a user; that is formally undecidable in computer science. Banks are extremely, painfully aware that the ordinary operation of the business of the bank will occasionally drop things on the floor. They have long-since automated the fat head of customer issues, and the long tail is kicked over to operational and customer support teams.

Every time responsibility moves between subsystems, be they different organizations, different computer systems, or different groups within the bank, some percentage of cases will simply break. The boundaries of systems are responsible for a huge percentage of all operational issues at banks. (They’re also where most security vulnerabilities live: systems A and B usually agree on reality, but a bad actor can sometimes intentionally get them to disagree, in ways which cause the bad actor to gain value before A and B reconcile their view of reality.)

A major technological advance over the course of the last few decades has been ticketing systems, which strikes many technologists as being crazy, because they’re almost the simplest software that you can describe as software. All a ticketing system does is enforce an invariant: if there is a problem with a case number assigned to it, and it goes between Group A and Group B, Group A needs to know it no longer is responsible and Group B needs to know it is now responsible. Then you can do can’t-believe-they-pay-us-for-this computing and observe things like “Group B is now working on 10,342 cases”, “There are 76 cases which Group B has not acted on within the last month”, and “Ginger seems to be anomalously unproductive at closing out cases relative to her nearest coworkers.”

So why didn’t ticketing systems solve this problem? Part of it is that the problem is self-referential: the ticketing system is not the core. The ticketing system is not the subsystem that is directly responsible for anything of interest to you. The ticketing system is an entirely new system, which requires integration with other subsystems and which will frequently need to do handovers to them. This interface is frightening, unexplored territory where new classes of issues that you’ve never seen before can spring up.

Bank systems are an interesting combination of designed and accidental. They accrete like sedimentary layers. A particular force which affects banks more than most institutions is that the banking industry has undergone decades of consolidation. When banks merge, one bank doesn’t simply eat the other and digest its balance sheet and people. They end up running their systems in parallel for years while working out an integration plan. That plan will, almost inevitably, cause one of the systems to mostly “win” and the other system to mostly “lose”, but for business reasons, something of the loser will be retained indefinitely. It now has to be grafted onto the winner, despite frequently being itself decades out of date, having its own collection of grafted acquirees partially attached to it, and needing expert input from people who are no longer with the firm.

Users can watch this play out in real time. For example, I banked at First Republic and also bank at Chase, which now owns First Republic. In something which sounds unimpressive and would blow the mind of bank CTOs from as recently as ten years ago, both sides of the bank understand that the same person has an account on the other side. (You wouldn't think testing Social Security Numbers for equality requires any high-tech wizardry, and you'd be right. The thing which was actually hard was building a process to allow complex ad hoc bidirectional synching of systems that were not built in tandem with each other.)

But because that integration is ongoing and will take years to resolve, neither part of the bank knows consequential things the other part knows about me, even where it strikes most people as obvious that they should. Chase eagerly communicates timelines for transitioning the home loan that First Republic very definitely never wrote. It is utterly clueless about the Line of Credit that they factually did extend. And it will require a lot of midnight oil from hundreds or thousands of people for most of another year before I can walk into a Chase branch and ask what the balance is on an account serviced by First Republic's core.

Human accountability and its malcontents

So let’s talk about how banks spackle over the infelicities in their systems.

First, the bank builds many subsystems which interface with its core processing systems and ledgers. These systems are built so internal bank staff can see what a customer has done in their accounts and, perhaps, act upon those accounts on their behalf.

For those keeping score: yep, this interface boundary is another place which can cause the bank to fail to agree with reality. Relatively simple programming issues can cause the staff-exposed view of an account to fail to agree with reality known to the bank.

For example, they not infrequently fail to show some staff transactions which are “pending.” In many cases, “pending” has consequences which are extremely similar to being finalized from the perspective of the user, but a particular system might simply not show them. You’d think that is a confusing choice to make and often underrate the possibility that no one ever made this choice, not really. Sure, it exists (inarguably in this case) in code, and that code might be described in a requirements analysis document that someone handwaved together 18 years ago, but nobody ever said “Nah, exclude pending transactions”. This was a simple oversight, projected into the future indefinitely, to the enduring annoyance of old hands among staff and the continued surprise of non-specialist users. You might assume that senior members of operational staff have the ability to write a memo to engineering or procurement to tell them that the software that makes up the bank is broken. That is a thing which exists at surprisingly few firms.

(A repeated experience of my time at Stripe was watching engineers embed with Ops teams for a day, then run back to their laptops while saying “I’m so sorry! I can’t believe we did that to you! I will drop everything I am doing and fix it immediately!” In many cases, those bugs had existed for months or years. I watched senior engineering leadership ask senior Ops leadership why they had never been asked to fix them. Ops replied that their long experience in the financial industry had taught them that Ops never gets to use software which isn’t broken and that complaining about this is like complaining about gravity.)

Banks aggressively partition staff based on job duties and levels within those duties. The most relevant silo for retail consumers is actually a series of parallel silos which staff front-line phone support for the bank. Often, each line of business gets its own silo, which accounts for much of the Your Princess Is In Another Castle that happens when you call a bank with a seemingly straightforward question and then get passed between various departments.

Most products offered to retail consumers and small businesses are relatively low margin, in absolute dollar terms. To be able to offer these, banks use various methods to cram down their support costs. Offshoring is often the face of these initiatives, but stratification by skill levels and powers granted is probably more important to understand.

A fairly typical setup for a financial institution will have the retail bank support teams stratified into Tier One, Tier Two, and Tier Three. Each has management located with them, who may or may not be shared across tiers.

You truly haven’t lived until you’ve tried paying for your college education by being a Tier One customer service representative, like your humble correspondent did. (Not at a bank, thank goodness; I might have stayed in a field with that many interesting problems presented.) The reason Tier One exists is that the median problem, so-to-speak, from a retail user is not actually a problem. That retail user is extremely unsophisticated about the bank account, finance in general, and frequently many other things in life. Tier One exists to handhold this individual in getting something very straightforward done, or to pass the call off to Tier Two.

Many people in our social class want to be extremely compassionate in explaining challenges that some people endure. Sometimes this compassion extends to believing that people with substantial challenges don’t exist or don’t exist in any large numbers. It is extremely important to understand that those challenges exist and that they will dominate your frequent fliers for support. Some people have only emerging competence in English but will want services from a bank which does business in English. Some people, frequently with large account balances and long successful histories with you, are experiencing age-related decline in their faculties and need to be protected, frequently without that being a capital-F Fact within the system yet. Some people are crooks. Some people have a very interesting relationship with the truth, and say many things to banks which probably felt true to them in the moment. (Is that fraud? Eh, it’s complicated.)

But to zoom into one particular way people can differ from each other: Some people are not as intelligent as you are. That is uncouth to say. In the United States, almost every large organization will institutionally tamp down on any explicit discussion of it. They all must structure their affairs to deal with the reality of it, though.

Think of the person from your grade school classes who had the most difficulty at everything. The U.S. expects banks to service people much, much less intelligent than them. Some customers do not understand why a $45 charge and a $32 charge would overdraw an account with $70 in it. The bank will not be more effective at educating them on this than the public school system was given a budget of $100,000 and 12 years to try. This customer calls the bank much more frequently than you do. You can understand why, right? From their perspective, they were just going about their life, doing nothing wrong, and then for some bullshit reason the bank charged them $35.

The reason you have to “jump through hoops” to “simply talk to someone” (a professional, with meaningful decisionmaking authority) is because the system is set up to a) try to dissuade that guy from speaking to someone whose time is expensive and b) believes, on the basis of voluminous evidence, that you are likely that guy until proven otherwise.

And so every Tier One rep will talk to dozens of folks a day. Many of those calls are… fairly aggravating, from the perspective of the agent. Tier One has limited ability to do anything useful; this depends on the firm and the silo within the firm, but they are largely read-only interfaces to money. They have a few pre-programmed buttons to push which get 90%+ of people they talk to to not call again. They execute scripts and flowcharts, written by people better paid than them, which gate your access to Tier Two.

In the best operated systems in the world, Tier One gets about one tweet worth of context to pass over to Tier Two when doing a handoff to them. (“Cust didn’t rec new debit card to Japan plz next day air + waive fee.”) Most financial institutions are not the best operated systems in the world. The bank “forgets” about your issue as soon as you’re off the line with Tier One, and needs to be told it entirely de novo when you speak to Tier Two.

Tier Two typically spent a few years in Tier One and has begun to specialize in a subfiefdom of banking. They have emerging competence into the nitty gritty of operations at their institution, at least with regards to that subfiefdom. They’re paid more, though not by much. They’re typically given more ability to do what my shop called “accommodations”, which means self-authorizing a resolution for a customer which costs money. Tier Two might be able to, for example, credit an account a small amount of money for an arbitrary reason and have the bank charge it off as an operations loss.

Your humble correspondent had a soft limit of approximately $200, below which no number was worth trifling my management chain or a specialist about. An interesting observation about the physics of money is that Tier Two could conceivably cost the bank more by authorizing accommodations than they earn in salary. A line manager apprised of this probably will not investigate it for more than five minutes before deciding that the bank is satisfied.

Then you have Tier Three, which at some firms sits in Customer Service and at some firms sits in Operations. There exist some ambiguity and spectral ranges here, but at some point the job changes in character from “low-wage peon reciting a script” to “professional who has a career doing this and is no longer managed on a tickets-closed-per-hour basis.”

Tier Three, let’s call them for simplicity, engages in constant firefighting, because at the volume of transactions (and other sources of cases) in all-but-the-tiniest financial firms, something is always on fire. Sometimes you’re covering for hiccups in the technical systems of your institution or counterparties. Sometimes someone has found themselves in an odd edge case or been passed around for ages between departments. Sometimes you’ve received an escalation, about which more later.

A user of the banking system will often have to redundantly explain themselves when they hit Tier Three, for the same reason as they did when they hit Tier Two. However, because they’re no longer operating in time-starved tickets-per-hour crunch mode, Tier Three has richer access to systems at the bank and more ability to forensically reconstruct procedural history, including history that is not ledgered. They will frequently do this both to do their jobs and to do the legwork for other professionals at the bank who might have decisionmaking authority in some cases but do not have the access or acumen to pull together a view of a case from disparate systems.

The gaps in experience between getting passed around tiers are replicated for being passed around departments. Say, for example, that the bank owes you a check and you do not receive it in the mail. The vast majority of checks sent through the mail arrive without issue, but the bank knows that it will have to reissue some of them. There is a process for doing this. Unfortunately, because this is a relatively infrequent issue, Tier 2 does not have a Reissue Check button available to them. Instead, their interface to this process is likely “Raise a ticket with Ops and tell the customer someone will call them.” There is no system available to Tier 2 which can verify that that call was actually made. The agent has no basis in their training or experience to know whether the bank routinely makes that call. It is quite possible that success rates on that call being placed are quite low, even if you ask for it three times consecutively, and that the bank is entirely institutionally unaware of this.

And so the customer will feel frustrated and they have been lied to, Tier 2 certainly doesn’t feel like they’ve lied to anyone (they read the script, it’s a Tuesday), Ops feels like the world is on fire because the world is always on fire, and senior bank management cannot detect this problem because no metric available to them is capable of disaggregating it from the complex monster that is the financial system.

Two embedded surprises about bank staffing

Many traditionally-minded users of banks assume that someone at their branch can likely help them with issues. Due to the deskilling of the bank branch, the people at a bank branch, including the branch manager in many firms, can only offer solutions to relatively straightforward problems. For the other ones, they also have to call into a support phone tree. Sometimes the bank will have ability to e.g. share context between their screen and the Tier 2 rep; sometimes they’re literally incapable of proving to the bank that they work there. (You might think I’m joking. To beat a drum: the level of technical sophistication across the spectrum of U.S. financial institutions varies wildly.)

The other surprise is that substantially every financial institution has a parallel way to reach decisionmakers in every area it operates in, which skips most or all of the tiering system and the technical and organizational scar tissue that it carries. This goes by different names in different places but “escalations” is a fairly common one.

Much like the United States has decided, in its infinite wisdom, that caseworkers for immigration and passport services should be staffed in every Congressman’s office and not at the agency that actually handles immigration or passport issuance, there very likely exist people at the bank whose job is working the bank more than it is working for the bank. A number of functions which are not ordinarily customer-facing are given the contact information for that group, with the instruction “In case of emergency, skip Tier Everything and talk immediately to the highly-placed troubleshooting team.”

If you are a reporter and call a bank for comment about a widow on the cusp of being improperly foreclosed upon, you will (fairly reliably) find your words forwarded to the troubleshooting team within a few minutes. If you’re a regulator and intervene on behalf of an individual, same result. You can absolutely achieve this as a civilian, too; a paper letter to the VP of Retail Banking, Office of the President, or Investor Relations will often cause the bank to swing into motion in the same way. (I wrote a few hundred letters like that as a hobby back in the day.)

These folks are professionals who are capable of keeping paper notes and having day-to-day recollection of things they have done in complex cases. They are managed and incentivized in a way which allows them to have agency. The formal customer support organization is very, very bad at this, at every tier. It is very difficult to do in a model where you’re constantly bouncing cases around individual reps and between departments.

Is this because banks are malicious? Are they willing to grind retail users to bonemeal in the pursuit of another cent of earnings per share? The truth is a bit more mundane: supporting people with can-do-anything-you-throw-at-them professionals is ridiculously expensive and getting moreso over time. The per-case cost for the troubleshooting team can be more than 100X that of the tiering system.

Retail customers have relationships where they pay highly-educated high-agency jacks-of-all-trades to provide professional services in arbitrarily complex situations. They hate the experience of those relationships. They hate their medical bills. They are incredulous that lawyers bill hundreds of dollars per hour (in six minute increments).

You can get something approaching this level of service out of a bank, too, and the private bank generating $150,000+ in annual revenue per client would be happy to make your acquaintance. But if you want phone calls to the bank to be both answered at 2 AM and absolutely free, you want the tiering system. Society wants the tiering system. It is why a high school student with a paper route can open a checking account, get a debit card, and start buying things on Amazon. It is why bank branches can be operated in working class neighborhoods.

As a sophisticated user of the banking system, a useful skill to have is understanding whether the ultimate solution to an issue facing you is probably available to Tier Two or probably only available to a professional earning six figures a year. You can then route your queries to the bank to get in front of the appropriate person with the minimal amount of effort expended on making this happen.

You might think bank would hate this, and aggressively direct people who discover side channels to Use The 1-800 Number That Is What It Is For. For better or worse, the side channels are not an accident. They are extremely intentionally designed. Accessing them often requires performance of being a professional-managerial class member or otherwise knowing some financial industry shibboleths. This is not accidental; that greatly cuts down on “misuse” of the side channels by that guy.

It is also much more institutionally palatable to the bank and other stakeholders like e.g. regulators. No financial institution can say “We offer differential service levels to our community based on their education level, perceived social class, and perceived capability to bring power to bear on their behalf.” Every financial institution factually does that. The successful way to phrase it is “We offer contextually appropriate services to the entire range of customers, who come from all walks of life, and also we respond with alacrity to any issues impacting our important stakeholders via a variety of programs.” 

Society has goals which conflict with banks being good at banking

I hate sounding like a conspiracy theorist about banks, which for whatever reason seem to attract a disproportionate amount of attention from people who believe the Illuminati and lizardmen are conspiring to corrupt the free peoples of the world. And so ordinarily I do not want to say crazy things like “Sometimes banks suck because we want them to suck.”

Sometimes banks suck because we want them to suck.

In the specific case of “Why did the bank close my account, seemingly for no reason? Why will no one tell me anything about this? Why will no one take responsibility?”, the answer is frequently that the bank is following the law. As we’ve discussed previously, banks will frequently make the “independent” “commercial decision” to “exit the relationship” with a particular customer after that customer has had multiple Suspicious Activity Reports filed. SARs can (and sometimes must!) be filed for innocuous reasons and do not necessarily imply any sort of wrongdoing.

SARs are secret, by regulation. See 12 CFR § 21.11(k)(1) from the Office of Comptroller of the Currency:

No national bank, and no director, officer, employee, or agent of a national bank, shall disclose a SAR or any information that would reveal the existence of a SAR. Any national bank, and any director, officer, employee, or agent of any national bank that is subpoenaed or otherwise requested to disclose a SAR, or any information that would reveal the existence of a SAR, shall decline to produce the SAR or such information, citing this section and 31 U.S.C. 5318(g)(2)(A)(i)...

If the United States brings its subpoena power to bear against a bank teller and asks them about a SAR, they’re supposed to say nothing. That is the law! (Regulation, well, if one wants to be technical.) It is designed to be enforced against the interests of the United States of America! Customers have far less access than the U.S. awards to the U.S.! So does the teller, incidentally: to avoid constantly violating this, Compliance at most functioning institutions has long-since decided that SARs will live in their own walled garden of a subsystem, seen only by the people responsible for drafting them and sending them to FinCEN.

That subsystems’ interactions with every other system are, of course, a site for extremely painful hilarity to happen. If, for example, a SAR is misfiled because that subsystem doesn’t share the same view of account ownership as another part of the overall system, investigating that problem might require telling the customer that they were investigated, which you cannot do. And because this is insufficiently Kafkaesque, at some financial institutions, you can get a SAR filed for knowing what a SAR is, because “advanced knowledge of anti-moneylaundering procedure” is a characteristic only of financial professionals and terrorists. Compliance training can tell e.g. personal bankers to please look at the Know Your Customer questionnaire and see if “Professional background: I work in finance” is bubbled in and then draw the appropriate inference.

You might think I am joking. I am utterly not joking. Most of the times infelicities in the world have a logical explanation to them, a structural cause where each individual link in the chain sounded good at the time and the result just happens to be suboptimal. And sometimes the world is absolutely batshit insane.

So what can be done about this?

Like many structural problems, banks lacking object permanence didn’t happen overnight and can’t be fixed overnight.

A lot of the fix is technical. In the not-too-distant past, there were zero—zero—financial institutions which were competent at software. There are now a handful of them, after the expenditure of many tens of billions of dollars. That was the price of getting without-loss-of-generality Chase to the point where in-house engineers can cause the retail web app to react to me having an account at First Republic less than a year after their purchase of that bank.

Although it certainly doesn’t feel like it to people who hit edge cases, the tiered support model is a technology which took us decades to popularize and which made the world much better. It brought down the cost of financial services and supported product innovation which would have been impossible under the mid-century bank staffing model. We could not have credit cards or discount brokerages without the tiered support model. The biography of Charles Schwab makes this point persuasively at considerable length: competent telephone operations were instrumental to bringing equity ownership to the middle class. You should prefer a world with credit cards and discount brokerages to one which doesn’t have them, even as you listen to hold music occasionally.

It will similarly take decades to roll out the best-functioning refinements on customer service at scale to the entirety of the financial system. Partly this will happen through continued consolidation; the more banks Chase ends up owning, the higher the average operational competence in the U.S. financial system is. (And that’s… saying something.)

Partly this will happen as banks increasingly tap external providers for technology where the right things are recorded automatically and actioned appropriately. Partly, I continue to expect Operations to come further into its own as a high-status discipline, and to rewrite the internal structure of banks just as engineering has done over the last two decades.

Partly this will happen as banks increasingly partner with firms that impose a tech-inflected view of the customer experience. Google is, for example, a legendarily hostile organization to attempt to get customer support from. Google is also beloved by users because of overwhelming competence in shipping products that work almost all of the time. If you think that talking to a compassionate human is a core part of the banking experience, there are many banks in Iowa who will sell that service to you. If you simply want to access your money on your phone and have that almost always work, Cash App will happily operate as a front end over Lincoln Savings Bank to make that happen. (There are many layers to that onion. No particular equilibrium is necessarily the “right” one!)

And partly, we as a society have to make some tradeoffs. We want something from 12 CFR § 21.11(k)(1) . It was not written by accident or because the drafters were stupid. Every future with 12 CFR § 21.11(k)(1) in it will include many Americans whose bank accounts are closed for no reason that can be disclosed to them. Many of them will have done nothing wrong.

Plausibly, we should decide to stop doing the thing that no one wants us to do. And, as a particular thing which could help unlock that: if one cares a lot about the experience of people at the socioeconomic margins, one should perhaps spend less time fulminating about greedy capitalists and spend more time reading Requests For Public Comment by relatively obscure parts of the administrative state.

A book review is a curious artifact. By convention, it purports to stand in judgment of the work of an expert who has labored for months to create something of enduring value in the world. Few book reviewers, on the other hand, have ever done anything meaningful, and none has ever spent a book-equivalent effort on a book review. Still, hope springs eternal among would-be reviewers that they alone will write a review worth reading. And so I throw my hat into this accursed ring.

Number Go Up, by Bloomberg’s Zeke Faux, is the best book yet written about cryptocurrency. It is even better if you haven’t closely followed the rogues gallery committing a Shrekian onion of scams.

And yet I found it at times exasperating, perhaps because of the narcissism of small differences. I am also a long-time cryptocurrency skeptic, and perhaps if one squints a great deal a professional in financial writing, and it is not exactly the book I would have written. Of course, I did not spend two years of my life diligently talking to ne'er-do-wells and writing down what they said.

Most writing about cryptocurrency by professional journalists suffers from taking the representations of scam promoters at face value. This book does not have that problem: its tone is unbridled contempt for its subjects.

Some books might choose to unspool this point of view quietly over time, before gutting the subject with a single arc phrase and an implicitly raised eyebrow.

The byline for the Prologue announces that, as of February 17th, 2022, that “Total Value of All Cryptocurrencies: $2 Trillion (Yes, Trillion with a ‘T’)”

We, somewhat surprisingly, do not return to this device. It is drive-by contempt, thoroughly earned by the subjects but as yet unearned by the text.

Finance has a conceit that the numbers are important, indeed, that the world is ruled by important numbers. Financial journalism adopts that conceit when it is convenient, but (mostly accurately) assumes that the reader couldn’t read a balance sheet if their life depended on it. So in place of numbers, financial journalism is vibes, recounted conversations, and (at its best) snippets of documents contrasted with the conversations to cast doubt on the vibes. The numbers only matter insofar as you can vibe to them.

Which, to be fair, is a metacommentary on the crypto industry, which also pretends to be about crystallized math and its implications for humanity but cares about numbers to the precise extent that Number Go Up. Crypto and journalism love each other; where else in finance can one find such rich vibes (and such an engaged audience clicking merrily) unencumbered by the dreary necessity of ferreting numbers out of PDFs and doing fourth grade math on them.

What does Number Go Up do well? It sends our protagonist Zeke all over the world in search of lies. Bloomberg clearly has a very generous travel budget and if you missed the subtle hints dropped by inclusion of multiple chapter-length travelogs of Switzerland and the Bahamas, this subtext is eventually promoted to text. Zeke is, very definitely, a character in his own book. He’s engaging, he’s sometimes self-aware, and by God, does he write well.

Zeke finds more liars than lies, more incredible characters than an incredible plot, more gonzo journalism than a world gone absolutely mad.

A throughline of the book is what investigative journalism is believed to do by civilians, what it actually succeeds at, what it fails at totally, and what it institutionally believes is out-of-scope. The book is not very reflective about this, or perhaps it was carefully designed to examine this throughline while seeming at a surface level oblivious to it, in which case this salaryman stands awestruck at the accomplishment.

For example, we’re treated to a subplot about how hilariously and consciously Bond villain some of the relevant Bond villains are. I have, for years, used “Bond villain” to describe how the rogues central to crypto seem to flit effortlessly between jurisdictions, wherever the plot demands, without seeming to ever be hindered by regulations or government action. Zeke handily one-ups me: the central characters own houses previously filmed as Bond villain lairs and seem to adopt the aesthetic of Bond characters intentionally.

Jean Chalopin, the chairman and CEO of Deltec, the bank which for an extended period was Tether’s proxy in moving money around the global financial system, really did have a hand in bringing Inspector Gadget into the world.

One could imagine him stroking a white cat as he talks with Zeke about the reclusive Tether CEO Jean-Louis van der Welde. Oh no, that sounds absurd. See, they actually meet van der Welde at a casino, Chalopin’s introduction being a (smiling) “If you screw this up, I’ll kill you” and van der Welde ironically observing that he is “[t]he man that does not exist.” Great stuff!

Zeke gets absolutely nothing useful out of van der Welde. He is conscious of this and announces it to the reader. But he gets great narrative texture to cover chapters recounting months of patiently following the Tether saga.

And it must be said, Zeke is far-and-away the best mainstream reporter on the Tether beat, which is institutionally treated as a boring Internet backwater and not the largest financial fraud in history. Anyone Seen Tether’s Billions is the first and last useful expose written about the company in the mainstream press, and coined the best bon mot about the conspiracy: “[Tether] is quilted out of red flags.” I am still jealous that I did not write that line first.

Zeke fingers CFO Devasini as the primary motive force behind Tether. He manages to find an offhand comment on an Italian politics article linking him to a pseudonymous blog where Devasini praises Madoff and fantasizes about following in his footsteps. Holy #*(#%#*(! You get little flashes of this throughout the book: patient, occasionally inspired journalism by the best reporter on the beat, who sets up to nail Tether to the wall and ends up… not doing so.

He meets Bitfinexed, who deserves a Pulitzer and will not get one. Bitfinexed is a pseudonymous Twitter troll who is neither a financial industry expert nor a buttoned-up professional reporter. He has been more right than wrong about Tether in what some people believe is an annoyingly loud fashion since 2017.

Bitfinexed hopes Zeke has sufficient goods to put Tether away. Same, same, muses Zeke. They part disappointed in each other.

I have been informed by financial journalists before that they are not law enforcement and cannot be reasonably expected to ferret out multi-billion dollar frauds before law enforcement does. In my childlike naïveté, I always thought that Lois was Clark’s equal partner because she was damn good at being a reporter. It turns out that reporters are actually just the vibe police, but will not disabuse a gullible civilian who believes them to be heroic in stature. (Law enforcement? Yeah, also mostly incompetent at detecting multi-billion dollar frauds before they are exposed, and will say (quietly) that that isn’t really their job, either. So I guess, by process of elimination, it is Bitfinexed’s job.)

Zeke is handed a detailed trove of information on Tether’s holdings in August 2021, by a source who requests anonymity. They include short-term bonds, which Tether had told the world to expect, small positions in a grab-bag of commodities futures, a few hedge fund holdings because why the heck not and, oh yeah, billions of dollars of Chinese commercial paper. Zeke finds no angle with these documents, though a short seller makes what seems like a sincere offer to buy them for $1 million. Zeke is constrained from accepting due to ethical concerns.

(For those keeping score at home, the Chinese commercial paper point means you are reading in 2023 about a journalist who got explosively newsworthy documents in 2021 confirming what Bitfinexed had specifically and loudly claimed between, oh, 2017 through 2019 and thereafter. I realize there are serious institutional constraints that prevent the Pulitzer from being awarded to a loudmouthed frequently off-base pseudonymous troll, which is a pity, because they ensure it will go each year to a journalist less right and less early about a less important story.)

Zeke avoids a lot of the smoke and mirrors deployed by Tether and crypto generally. Unfortunately, the smoke and mirrors detector is not 100% well-calibrated.

Take Chalopin. He’s depicted as charming and affable, with a roguish sense of humor. In a podcast promoting the book, Zeke takes pains to mention that Chalopin has always treated him honorably.

A brief digression

Please excuse a brief interruption from this review of Number Go Up while I discuss extratextural knowledge about one of the characters.

Chalopin is a bagman. Chalopin is a bagman. Chalopin is a bagman. There are other interesting facts about Chalopin, but they have to be filtered through remembering he is a bagman. The roguish charm and personal honor are important professionally significant attributes about why he is so damned good at being a bagman. He makes one forget he is complicit in the conspiracy during conversations about the other co-conspirators.

Chalopin established a worldwide web of correspondent accounts in the name of Deltec Bank & Trust, Ltd. to allow Tether and their co-conspirators to launder funds. This lets one send a wire from a bank which has a compliance department that is alive, awake, and uncorrupted by Tether, because if Deltec is willing to be complicit one doesn’t need to name Tether on the wire.

Compliance will only see a payment to a reasonably well-respected local bank to ultimately pay a bank in the Caribbean. This chain of activities might sound exotic to you but is routine and non-objectionable in offshore finance. One could, for example, be paying a bank to buy a Treasury bill from them. There is nothing wrong with that, unless one is in fact doing business with a banked Bond villain rather than with the bank itself. The wires have a lie on them to allay suspicion (nothing new there) and tell Deltec which pending transaction to tell Tether to clear.

Chalopin is the very image of a colorful-but-ultimately-inoffensive bank CEO, which is how he convinced state and federal regulators to approve of him buying a U.S. bank. He did this to assist Tether and FTX in committing a control fraud to secure unfettered access to FedWire.

The New York Times outed Chalopin as controlling Moonstone Bank. Financial journalism sometimes does actually do the thing everything believes it does: ferreting out wrongdoing before the feds indict everyone involved!

Oh, to have been a fly on the wall in that investigation. How could it possibly have happened? Let us speculate.

Perhaps a source emailed the New York Times a packet of documents. Perhaps that source followed crypto skeptics Twitter. Perhaps a participant in it DMed him a link to a consequential press release. Perhaps that source understood that Maryland Business Entity Search exists and that D20033544 is more interesting than almost everything in it. Perhaps that source knew that D20033544 was significant because new bank holding companies are announced in the Federal Register. Perhaps that source was absolutely dumbstruck when he saw Chalopin’s name at the bottom of the Maryland document, because he knew Chalopin to be a bagman. And, because a genre convention of Bond films is that expository scenes happen in locations the audience finds exotic, perhaps that call was placed from Tokyo.

Sadly, the world will never know, because journalists don’t reveal their sources.

Chalopin buying a U.S. bank is a very significant fact, because Chalopin is a bagman, and bagmen should not gain control of regulated U.S. financial institutions. However, some complex mix of lacking-object-permanence and believing-in-innocent-until-proven-guilty apparently makes it difficult for both regulators and financial journalists to take consequential real-time actions in light of Chalopin being a bagman.

But they do, eventually, and so bully for regulators and financial journalists.

Chalopin apparently pinky swore, when acquiring a U.S. bank, that he was not going to immediately turn it into a puppet of Tether. OK, technically speaking it was probably only a promise to not change the business model, but renting yourself out as a skinsuit is not in fact a traditionally accepted form of community banking. After FTX blew up, the relevant regulators were reminded, somehow, that Chalopin is a bagman, and issued a very stern cease and desist.

This killed the bank the bagman and the Bankman had conspired to puppet.

Reading the tea leaves, Chalopin will likely get to be roguish and charming to a very different audience than Zeke, and Chalopin’s travel expenditures for that gonzo adventure will be covered by the United States.

We return you to your regularly scheduled book review.

The tenor of most crypto coverage

Coverage of the crypto industry tends towards fawning and, when Number Go Down, becomes immediately disinterested in who lost money and the coverage’s complicity in that loss. To his credit, Zeke flies out to the Philippines to talk to Axie Infinity “scholars” (victims). Axie Infinity was a pyramid scheme covered largely incuriously in many high status places, including (it must be said) in Bloomberg, and which received promotion, enthusiasm, and money from people in tech who should have known better, including (it must be said) investors at A16Z.

Zeke hires a Tagalog interpreter to talk to people far more intimately affected by Axie than most of the crypto community was. That act is a valuable public service. They say things which are very predictable to any competent person who thought about Axie for more than a minute. But they are real people who exist in the real world, in circumstances well-paid Americans would not wish on their worst enemy, who we built a machine to exploit because Number Go Up. But we speedrun Axie's rise and collapse. Crypto has a new pay-to-earn to pitch (Zeke covers a few pivots) and Zeke himself is off to Cambodia, for the book’s (by far) most disturbing chapter.

Slavery is an institution which continues to exist. We react to that with the appropriate amount of moral horror, and nowhere near the appropriate amount of physical action, for understandable reasons. Zeke flies to Cambodia and pulls out his I’m An American So You (Probably) Can’t Kill Me card to get far physically closer to slaves than most Americans ever will.

Some of the slaves are being used to staff cryptocurrency scams operated by Chinese gangsters. Those scams take payment over Tether; this is probably a more incidental fact of their slavery than the text tries to leave you with. One gains sympathy for Sam Bankman-Fried, and isn’t that a sentence, when he tells Zeke that the situation is fucked and he has no idea what to do about it.

In a famous scene in The Wire, a killer of men flippantly tells a criminal defense attorney “I’ve got a shotgun, you’ve got the briefcase. It’s all in the game though, right?” The point is that both high status and low status actors are, in the language of the scene, parasites on the suffering implicit in the drug trade. The criminal defense attorney is a criminal defense attorney, though, which is why I want to carefully say that I remembered this scene while reading the Cambodia chapter but do not think it is analogous. I remembered it at the point where Zeke expresses moral distaste for a Vietnamese YouTuber interviewing former slaves, and in some cases directly causing them to be “former”, in the service of wringing money from Google. “It seemed distasteful to turn human suffering into YouTube content.”, Zeke writes, in the middle of a book published by Crown Currency whose Kindle edition has an MSRP of $13.99.

But Zeke is much more self-aware than this discordant note would suggest, which is why I mention it. For example, he later muses that he might have zeroed in on the wrong part of the FTX story when focusing on whether Sam Bankman-Fried was actually an effective altruist. I can relate; I was so sure SBF's job and source of economic advantage was being Tether's #1 bagman (never developed a strong feeling on whether he was 100% aware of this) that I overdiscounted the possibility he was also e.g. abominably bad at some of the things his sideline in running a cryptocurrency exchange and hedge fund required of him. Why discount that? I believed the day job was a sufficient explanation for having lots of money to throw around and he struck (and strikes) me as intelligent, which I assumed implied competence.

Many people have in the wake of the fiasco reexamined the evidence and updated towards him being more evil and more competent; I have yet to see anything which causes me to update away from him being sincere, charismatic, and utterly incompetent. Finance is aware this is a risk in senior leaders and that is why it has balance sheets and accounting controls. I modeled someone with a fantastically lucrative career in making crime palatable to high status institutions as being someone who probably had really good accountants (he did) and probably didn't run the business out of Google Sheets and Post-it notes (sadly, also true).

Memo to self: more math, less vibes in future.

A book not yet written

Every non-fiction book is paired with a notional anti-book, the part of the story which through some alchemy of authorial intent and happenstance was not told. The anti-book to Number Go Up would have covered, in appropriate detail, how crypto (and not just Sam Bankman-Fried) is making a concerted effort at regulatory capture and claims to have cleaned up its act. The anti-book would quote, at length, the adults who are newly in the room, and the consequences of their actions.

In the anti-book, the Canadian pension fund CDPQ which invested several hundred million dollars into Celsius would have been interviewed at length about their decision to buy equity in an insolvent financial entity at a non-zero valuation. Neither financial journalists nor their readers are particularly skilled at decoding balance sheets, but investment committees at pension funds are. The anti-book would print the balance sheet that the pension fund was swindled by, then quote an Accounting 101 professor explaining how even a C student could see it might as well be written in crayon, then quote the pension fund’s investment memo. Then the anti-book would triumphantly say “And then they lost all the money, because of course they did.”, and be very smug about it.

Celsius, a hive of scum and villainy even by crypto standards, whose CFO ended up remote working from an Israeli prison before the firm even got into trouble, gets a lot of airtime in Number Go Up. This is because their CEO Mashinsky is colorful character and fits in well with irascible incompetents.

But that’s the point, the anti-book would make great hay out of. This is the institutional, buttoned-up, the adults-have-arrived crypto. This is what the pension funds invest in. This is what pension fund CEO Charles Emond claimed was subject to extensive due diligence “... with many consultants and experts involved.”

Number Goes Up has a chemically-drenched chapter covering Jimmy Fallon during his career arc as a promoter for NFTs. NFTs, for those struggling to keep acronyms straight, were a briefly popular way to run unregistered securities offerings, after crypto realized that the last word in Initial Coin Offering (ICO) would eventually wake up a securities regulator.

Who cares about any of that? I want a book about what Charles Emond is smoking. Jimmy Fallon’s job is convincing the world he is capable of witty banter. Charles Emond’s job is making sure retired Canadian teachers can continue to eat. Hundreds of hours of evidence dispute Jimmy Fallon’s competence but on the plus side no abuse of the public trust he is capable of would result in people dying.

I would gladly read hundreds of pages of carefully recorded interviews and written documents by the adults in the room. Let the world behold the probity of our class and the exacting standards of our institutions.

We gloss over a publicly traded company (Voyager) which blew up alongside Celsius. It had the same risk controls (the empty set is equivalent to itself) and the same capture by a major client. That is a story worth telling, even if it is not nearly as cinematic due to lacking a serial fraudster CEO who seems seized by a compulsion to leave video evidence.

BlockFi barely gets mentioned either. They were, if you missed them, another adults-in-the-room crypto lending play. BlockFi’s CEO Zac Prince had an excellent conversation with his investors on a public podcast on June 24th, 2022. I think you could fairly characterize that conversation as bullish and attempting to distinguish them from the amateurs that had just blown up because they didn’t understand what e.g. asset/liability mismatch was. BlockFi filed for bankruptcy on November 28th. BlockFi apparently learned something about the shape of the world in about a hundred days which was surprising to them; perhaps the anti-book could explore what they learned.

At what point did Digital Currency Group realize that their subsidiary Genesis’s insolvency, in the same fracas, meant that DCG was cooked due to extensive self-dealing they had engaged in to paper over earlier, mostly distinct, stupid mistakes and malfeasance? You could write a book just about the GBTC arbitrage trade that blew up DCG.

The stakes are higher than they are believed to be

Number Go Up would be one of the most important works of investigative journalism ever if it showed the spread of the rot. Crypto marched through our most important institutions like cordyceps with a gain-of-function upgrade. It was a malevolent memeplex which corrupted some through ideology, some with crass lucre, and some by merely showing them a good time.

The corrupted were not merely no-account jokers, drug-addled former childhood actors, or members of hithertofore obscure Internet subcultures. They include every level (every level!) of our most significant national industries, regulators, and politicians, past and still serving.

FTX bought the time of Bill Clinton and Tony Blaire. Sam Bankman-Fried was believed by those in the halls of power to be the largest source of future funding to the Democratic Party. Meanwhile, his cutouts were busy sewing up the other side of the aisle. The conspiracy documented their strategy to remove crypto-skeptical legislators in writing.

A senior FTX advisor who, as a Stanford Law professor and major Democratic bundler, likely understood the law here, described that it would be unfortunate if the cutouts were understood to be “fronts”, in writing. Para 113. Financial journalism detects that the story there is that she is Sam Bankman-Fried’s mother. Homer (the Greek one not the yellow one) would be jealous of achieving that level of vibes, but we mostly don’t expect Homer to inform our views on governance.

Tether eschewed the quiet influence business and tried to buy a sovereign nation. (Never wanting to fail to hedge their bets, FTX very likely tried to buy a nation, too. They put a different variant of that plan in writing.)

Where are those interviews? Or one with Arthur Hayes, who bragged that he could suborn a nation for the price of, and this is a quote, "a coconut."

Heck, if you pull on that string a little longer... well, as The Wire says, "[Y]ou start to follow the money, and you don't know where the fuck it's gonna take you." Plausibly it takes you to somewhere broader, deeper, and darker than crypto.

Looking forward to the new revised edition when it is published

Zeke expresses some level of forward-looking regret that he might never get handed as juicy a story as crypto was in 2021 and 2022. There still exists juice, Zeke! Go get the juice! You still have a de facto worldwide monopoly on juicing! The rest of financial journalism cannot find an apple in an orchard at harvest time!

One imagines there are intelligence services that look at crypto and sigh, perhaps aspiring to some day match their operational cadence and breadth of activity. Perhaps one could interview senior ex-officials of them, and get a lot of quotes which would rhyme with “Holy #*(%*#(%*#( #*$(#(%*#% they were ambitious but an important part of the job is not getting caught. A good thing they did get caught, because otherwise the most Dangerous Professionals in the world would work for them now.”

I wish institutions, across the spectrum, would take crypto more seriously. Many in crypto have wanted to be heard sympathetically; seriousness and sympathy are not the same.

As it is, the book is the most important book about crypto, with the rest of the story very undertold. Zeke apparently (per his appearance on a podcast) has reserved the option with his publisher to add a postscript about Tether once they get what is coming to them.

I predict it will not be 5.075% on a 2 year T-Bill.

One interesting lens for understanding how industries work is looking at their waste streams. Every industry will by nature have both a stock and a flow of byproducts from their core processes. This waste has to be dealt with (or it will, figuratively or literally, clog the pipes of the industry) and frequently has substantial residual value.

Most industries develop ecosystems in miniature to collect, sift through, recycle, and dispose of their waste. These are often cobbled together from lower-scale businesses than the industry themselves, involve a lot of dirty work, and are considered low status. Few people grow up wanting to specialize in e.g. sales of used manufacturing equipment.

One core waste stream of the finance industry is charged-off consumer debt. Debt collection is a fascinating (and frequently depressing) underbelly of finance. It shines a bit of light on credit card issuance itself, and richly earns the wading-through-a-river-of-effluvia metaphor.

A disclaimer: I have had substantially more at-bats with debt collectors than most people, as a result of an old hobby of writing letters on behalf of debtors to their lenders and non-affiliated debt collectors. I did this over the Internet, on my own volition, because it seemed pro-social and I was extremely underused by my actual job at the time. This experience leaves me with strong opinions on the debt collection industry; a frequent archetypical person in need of a letter was a Kansan grandmother in diminished financial circumstances who had been harassed for months. I’m going to try to keep these views to a dull roar here, in the spirit of spending more mental energy on discussing why the system presents as so broken.

The lifecycle of a defaulted debt

The majority of debt which defaults in the United States is revolving credit card debt. (Without addressing the politics of it, the impression among many informed people that most defaulted debt is medically-related is the result of successful advocacy work rather than being substantially based in reality. Trustworthy numbers are not hard to come by, due to a combination of regulatory supervision and the U.S.’s almost unique degree of population-wide debt surveillance via credit reporting agencies. We’ll return to credit reporting some other day.) Assume I’m talking about credit card debt below, though the mechanics for medical debt bear overwhelming similarity. (I’d address it specifically except that healthcare economics is a fractally complex topic once you bring e.g. insurers, public programs, and the like into the mix.)

Credit card issuers bucket users into various archetypes, personas, and predicted lifecycles, because behavior is extremely heterogeneous and this is important from both a marketing and risk management perspective. Most users of credit cards, probably including readers of this column, believe they are the typical user of credit cards; no bucket is typical. If I were to make some informed guesses, relative to the population-wide distribution, you, reader, use credit cards in preference to debit cards as a payment instrument, do not routinely revolve balances, and hold some combination of student loan, auto, and mortgage debt which dwarfs your credit card debt. And so it is critical to understand that most defaulting credit card debt is not held by people who act like you.

Most credit card debt which defaults non-fraudulently was incurred in the relatively distant past. Credit cards are revolving accounts; one’s balance can increase (even as one successfully makes payments) for years as one pays off a portion of prior balances but continues purchasing with the card. This is normal and expected use of the card, planned for thoroughly, much like never carrying a balance is also normal and expected use of the card, planned for thoroughly.

A small percentage of borrowers, carefully tracked and generally oscillating between 2.5% and 5% depending on the overall health of the economy, will go delinquent on credit card debt. (Some issuers specialize in certain parts of the credit spectrum and, as a result, will have sharply lower or sharply higher delinquency rates. American Express, for example, specializes at the high end and typically has delinquency close to 1%. Capital One made its name in so-called subprime credit cards, though it has diversified since, and typically tends towards the high end among banks whose names you know. There is a largely hidden ecosystem of banks you don’t know that issue very expensive products to poor people; you can accurately predict their default rates exceed anything mentioned above.)

Default typically begins by missing (or underpaying) a scheduled payment. That in itself is a theoretical breach of contract but not very outside the ordinary for a card issuer; they will generally automatically assess a fee but take very little action. Most borrowers will recover before they are 30 days late, which is the point at which most issuers start to treat an account as being a credit risk rather than a minor operational issue.

After 30 days, issuers will typically work the account internally, using a combination of communication methods to nudge the user into payment, until one of a few things happens. The happiest is the customer gets current on their account. An outright refusal to pay is substantially less likely, and can result in an issuer moving up timelines. But the most common is that the borrower ghosts the issuer for a few months.

Consumer debt issuance is generally, by law, an exclusive privilege of regulated financial institutions. (The other big one is taking deposits.) Society wants many things from regulated financial institutions; one of those things is having accurate books, because stealth losses cause financial institutions to fail and frequently leave society holding the bag. As a result, the Federal Reserve has a Uniform Retail Credit Classification and Account Management Policy.

This tail wags the dog. Many decisions about account servicing, which a naive conception of debt might assume are between borrower and lender, are done with the goal of aligning servicing to accounting standards. In this way, the books impose their will on reality, and where the books and reality differ, reality frequently adjusts itself to accommodate the books. (As my buddy Kevin frequently muses, states are gonna see. Organizations which are tightly tied to the state, like regulated financial institutions, will develop a vocabulary and processes for seeing like the state sees, and that edifice tends to capture non-state methods of seeing that they run in parallel.)

In particular, credit classification requires that financial institutions “charge off” delinquent debt. Mechanically, they have previously accumulated an allowance for delinquency on their books as a liability; they move a bit of that allowance to a bad debt expense. In principle, nothing has to happen to the actual debt. In practice, financial regulators encourage institutions to seek certainty and finality around this.

Financial institutions achieve certainty and finality by packaging portfolios of bad debt together and selling them to non-financial institutions. This durably moves them off of the books and realizes a very, very small residual value.

The debt collection industry

There are essentially two halves of the debt collection industry. A portion of it works on an agency model: a lender can have e.g. a law firm or similar work a debt on its behalf during the several month period where the delinquent debt lingers on their books, in return for a performance fee (often 15-30% of face value) should the borrower make good on the debt.

But debts, much like people, only age in one direction. Unlike people, debts universally decline in value as they age. We’ll return to that topic in a moment. Most debt which is not being serviced successfully by the first party lenders will not long be worked by their agents. It is instead sold to debt buyers, who will then attempt to collect the contracted amount (plus fees provided for in the contract, which are legally legitimate, and not infrequently fees which were not provided in the contract, which are frequently extremely questionable). Subtract the original cost of buying the portfolio and their substantial operational costs and the remainder is profit.

Most defaults are small. This fact drives everything about debt collection; it has to be done scalably, by the cheapest labor available, with a minimum of customization or thoughtful weighing of competing interests. The average defaulted credit card debt is on the order of $2,000, the median is between $500 and $1,000. These are processed like McDonalds burgers, not like grant proposals.

Debts are sold as part of a portfolio, where (typically) thousands of relatively similarly situated debts in a cohort are sold as a packet. The value of portfolios is a huge discount to the face value of the debts; at the point where a lender has only worked it themselves and the debt is a few months delinquent, portfolios generally fetch about 5 cents on the dollar. That value will continue to decay over time. There is an entire ecosystem of brokers supporting contractual infrastructure to convey these debts to buyers and insulate the issuing financial institutions from the actions of the debt buyers.

Partly, that is due to regulatory risk. (Particularly in recent years, regulators have begun using prudential regulation of financial institutions to strongly suggest that financial institutions adopt their social goals. There is a tortured argument by which a debt collector being unsavory in the process of debt collection would damage the reputation of the bank, which could damage an item on its balance sheet, which could damage its financial stability, which fact a regulator actually has jurisdiction over, and therefore regulators can discourage debt sales.)

Probably more surprisingly, financial institutions care acutely about brand damage, particularly to their own partners that can contractually obligate that. Many credit cards are co-branded; you, as an unsophisticated consumer, might think you have a credit card from Amazon, Best Buy, or Apple. You don’t, but those three firms are acutely aware of your miscomprehension, and they want to continue earning your business for the next 40 years. And so they might have hypothetically contractually bound Chase, Citibank, and Goldman Sachs (respectively) to never mention their co-branding partner in connection with debt collection and to bind subsequent debt buyers to the same.

This is a frequent cause of debtors mistakenly believing that debt collectors are confabulating debts against them. “But I haven’t ever done business with Citibank!” I’m slightly-more-than-unsympathetic to that, because of the relative sophistication of borrowers versus Best Buy’s marketing department. On the topic of true confabulation, oh yes, that happens minimally hundreds of thousands of times a year; debt collectors adopt business processes which simply make debts up because accuracy requires competence and costs money.

The debt collection industry is, and I say this as someone who is capitalist as the day is long and attempts to be non-political in public, among the most odious hives of scum and villainy as exists in the United States. The business is sordid and virtually immune to reform, despite decades of trying. The Fair Debt Collection Practices Act was passed in 1978! It is older than me! I learned to cite it in 2004 against the same abuses it was designed to prevent! The situation did not markedly improve in the last 20 years!

This is not because of a lack of virtue or a lack of laws; the structure of the industry colliding with the socioeconomic reality of defaulting debtors basically ensures that it will be a miserable place populated by miserable people who will project leveraged amounts of misery into the outside world in the hopes of collecting a tiny sliver of defaulted debts.

Debts are conveyed to the debt buyers as large CSV files with minimal supporting documentation. The legal reality of a credit card debt begins with a contractualized promise to pay. You might assume that the owner of the debt necessarily has read that contract. Reader, if the debt has been sold, they have not merely not read that contract, they have likely not received a copy of that contract. They might have the contractual right to ask the seller of the portfolio to ask the entity it bought the portfolio from to follow a few more links in the chain to eventually ask the financial institution for a copy of the contract. In principle, financial institutions always have the contract… somewhere. In practice, they will frequently not organize themselves to actually locate it; this business is off their books and Operations has better things to do than hunting in the archives for a paper copy of a low-value contract signed several years ago.

Again, we’re talking about promises frequently denominated in the hundreds of dollars. It is a consensual social fiction that there is actually a legal process operating here. That consensual social fiction has real consequences, and sometimes bubbles up uncomfortably in actual courts of law, which we will return to in a moment.

The rights of debtors are observed by both primary lenders and eventual debt buyers mostly in the breach. One of those rights is to a written “debt verification”, with specified information in it, and (surprisingly, if you haven’t worked in this field) despite that being the law many debts are sold in such a fashion that the buyer couldn’t produce a responsive verification even if they wanted to. That isn’t even a political claim; it’s just the engineering reality of which columns are in their CSV file.

The former advocate in me will observe that the single most effective method for resolving debts is carefully sending a series of letters invoking one’s rights under the FDCPA (and other legislation) to a debt collector who is operationally incapable of respecting those rights, then threatening them with legal or regulatory action when they inevitably infringe upon them in writing, leading to them abandoning further attempts at collection.

This effectively makes paying consumer debts basically optional in the United States, contingent on one being sufficiently organized and informed. That is likely a surprising result to many people. Is the financial industry unaware of this? Oh no. Issuing consumer debt is an enormously profitable business. The vast majority of consumers, including those with the socioeconomic wherewithal to walk away from their debts, feel themselves morally bound and pay as agreed.

Why are debt collectors so bad at debt collection? Partially it is because credit card issuers are large national institutions with large, automated processes sitting atop a legacy of corporate acquisitions, IT migrations, and similar that makes availability of non-critical information extremely fragmentary. They then want to dump that complexity through a very small pipe (CSV files) onto the debt collection industry.

That industry is largely not characterized by large, nationally scaled, hypercompetent operators who happen to have decades of institutional inertia. Instead, it is heavily fragmented into strata of mid-market and smallest-of-small-business firms, governed by a patchwork of regulations on the national, state, and local levels, and runs (in large part) on faxes and post-it notes.

The operations of a debt collection firm

Roughly three quarters of debt is bought by ten large firms and one quarter is bought by so-called mom-and-pops, but this is complicated by the resale of portfolios which were worked by the majors. Mom-and-pops then buy them at a deep discount with the goal of getting the residual value which the majors did not successfully capture.

Suppose a debt collection firm has bought a portfolio. They will first "scrub" it, which means using automated or semi-automated processes to enhance the fragmentary data they have and prioritize the portfolio for collection efforts.

For example, one stage of the scrub will be associating a credit profile with as many debts as possible. Credit scores are extremely good predictors of who pays their debts; that is what they are designed to measure. You will get sharply better results on a per-call basis calling people with a 750 FICO versus a 450 FICO, accordingly, you should call all the 750s first and more frequently.

A second scrub will typically remove dead debtors, because they infrequently answer their phones. (Debts are not inherited in the United States, a fact which the debt collection industry frequently demonstrates strategic ignorance of.)

While reading government reports, I chanced across a mention that a novel form of scrub uses third party databases of so-called litigious debtors. This was surprising to me (as far as I know, it was Not A Thing when I was doing advocacy), but makes a lot of operational sense. You can trivially Google providers with [litigious debtor scrub] as a keyword.

The FDCPA and state legislation provides for automatic damages for illegal behavior from collectors, the incidence of illegal behavior is extremely high, and a debt collector with a high school education and three months of experience will frequently commit three federal torts in a few minutes of talking to a debtor then follow up with a confirmation of the same in writing. (You think I am exaggerating. Reader, I am not. “If you don’t pay me I will sue you and then Immigration will take notice of that and yank your green card” contains three separate causes of action: (frequently) a false threat to file a suit where that is not actually a business practice of the firm, a false alleged affiliation with a government agency, and a false alleged consequence for debt nonpayment not provided for in law.)

As a result, private companies compiled databases of (public in the U.S.) court filings and organized them by Social Security number, address, and similar to allow debt collectors to identify which debtors are aware of their legal rights. In principle, a debt collector could do anything they wanted with that fact, like being extra careful to follow the law in contacting them. But the economics of debt collection do not counsel careful, individualized consideration of credit card debt.

I will bet you that, in practice, they simply avoid collecting against anyone who demonstrates ability and financial resources to enforce their rights. This is one for the history books of borked equilibriums. We devoted substantial efforts to pro-consumer legislation to address abuse of (mostly) poor people. We gated redress behind labor that is abundantly available in the professional managerial class and scarce outside of it, like writing letters and counting to 30 days. (People telling me they were incapable of doing these two things is why I started ghostwriting letters for debtors.) We now have literal computer programs exempting heuristically identified professional managerial class members from debt collection, inclusive of their legitimate debts, so that debt collectors can more profitably conserve their time to do abusive and frequently illegal shakedowns of the people the legislation was meant to benefit.

After scrubbing their lists, debt collectors will attempt to locate contact information for the debtors. You would think this would be fairly straightforward, given that they were given that information in the CSV file they purchased. But information is fragmentary and outdated, and people likely to default on debts also have complicated and frequently changing relationships with their addresses and phone numbers. So there exists an ecosystem of “skip tracing” providers, delightfully (ugh) borrowing the jargon from the business of bounty hunters, to match up fragmentary information with current contact information.

Operational and IT issues at this stage create a lot of the harm in the debt collection industry. For example, to use an example ripped from my own experience (which was part of over $100,000 of confabulated debt that caused me to fall down the credit advocacy rabbit hole), consider a fragmentary piece of information identifies a P.J. MacKenzie at an unspecified town in Illinois as owning a debt in the few hundred dollar range. That name is a fuzzy match to one Patrick Johnathan McKenzie of Chicago, Illinois. Here you are welcome to your moral intuitions as to whether simply asserting that I owe you a few hundred dollars is legitimate. The debt collection industry frequently assumes, as a matter of business practice, “I mean, probably you do factually owe me, and if you don’t, oh well, then you won’t be one of the ~8% of people who we successfully collect from.” No individualized application of human labor is required to reach this conclusion. No one is accountable for the mistake; no one made it, not in the usual sense of people making mistakes. People architected a system which generates this outcome by default.

And then begin the calls. To you, if the above processes have successfully found you. To your friends and relatives, if they have not. (The FDCPA provides that debt collectors are only allowed to contact a debtor’s family to find current contact information. They do this with gusto, including when they have current contact information. It is used coercively, because many debtors will make sacrifices to find $250 so their mother stops getting calls.)

Debt collectors typically work in call center environments, assisted by technological improvements to greatly increase the number of calls per hour each collector can do. These include so-called predictive dialers, which (as a sketch) dial several debtors in parallel for each collector, since most calls will not be answered or will go to voicemail, then connect only those that pick up to a debt collector waiting in the queue. Maximizing the efficiency of callers is key to the economics of debt collection; labor frequently costs more than than underlying debt does, even when one is paying four cents on the dollar and hiring collectors straight out of high school. An unhappy consequence of this is that debt collection firms build extremely scaled systems to project externalities into the phone network and the people attached to it, including very many people who never owed them a dime. But for the (dubious) cover of law, this is what technologists would call a volumetric denial of service attack: a relatively small expenditure of resources allows a collector to disable a relatively large number of phones.

If there is one thing that unites debtors, it is the cacophony. The phones ring and ring and ring. They ring at all times legally permissible and many times not. They ring from the same collector calling you twice a day. They ring from the same debt being collected by multiple different agencies, several of which sold the debt on but neglected to update their internal systems to stop collecting on the debt they no longer own. They ring from dozens of different firms, because the circumstances which caused you to go delinquent on one debt caused you to go delinquent on several debts. Each of those is collected in parallel by a process which considers every other debt collector a competitor for your scarce dollars and must outcompete them by ringing you harder, faster, and more persistently.

What does a debt collector hope to get from a call?

The goal of calls is to get a verbal promise to pay and payment credentials. The most useful payment credential, from the perspective of a debt collector, is a checking account number, but in a pinch they will also usually take debit cards or credit cards. (Many people who have defaulted on debts still have access to credit, because in deteriorating financial circumstances one could default on some but not all credit lines. There is also an extremely efficient financial industry capable of extending credit profitably at a wide variety of predicted repayment likelihoods.)

Many promises to pay are not promises to pay in full, but rather “a plan” to pay the face value of the debt plus accumulated (and accumulating) interest over time. Most debtors who agree to payment plans will not successfully complete those plans. This is, from the perspective of a debt collector, not a problem. If they paid five cents on the dollar for your loan, and successfully convince you to pay twenty cents today and the balance over monthly payments over a year, you know what they’re left with in month three when you default? Ability to collect from you again, or to sell your debt for more than they paid for it, because you are a better credit risk than most people in the portfolio. You proved that, by paying.

Many less sophisticated customers assume that they have substantial control over sending payments via check. All checks have the account number, printed on the face of them, and collectors will push aggressively to get that number over the phone versus waiting for a check to arrive. The goal of having the number is to present the bank holding the checking account with an electronic ACH debit or a “demand draft”, in both cases representing that the account holder has pre-authorized the debit. How many debits did a debtor agreeing to the above sketched plan pre-authorize? Not less than thirteen.

They will frequently be very surprised to learn that, and they will be frequently surprised when their financial institution backs the debt collector, who often has a sudden fit of competence and manages to record this part of the conversation.

The amount of gamesmanship that debt collectors get up to regarding payments could use its own column. The verbal authorization to debit your account (“payment plan”) was probably silent on timing, so they will use their substantial experience in dealing with people in poor economic circumstances to target the dates and times of day when benefits payments or paychecks have posted to your account. They’ll surge collection attempts at 3 AM on the 1st of the month, learn the differences between various banks as to when Social Security payments post, etc. They’ll also do things like e.g. splitting the agreed upon payment into smaller payments, with the goal of getting at least one rather than having the payment declined for insufficient funds (NSF, in financial industry parlance). This has frequently magnified fees assessed by banks to debtors, because NSF fees are frequently assessed per-occurrence.

Why does this continue being so broken?

Incentives rule everything around us. The structure of the industry, technical issues, and similar alluded to above cause most of this, and they have continued causing it despite decades of effort to correct.

There are also human capital issues at play. Discussing them is a bit difficult, because the people who physically make phone calls for debt collectors share many of the life challenges of debtors as a class. They are poorly educated, poorly paid, poorly managed, and churned through viciously. The work is soul crushing; turnover at the large firms is 75-100% annually. Call centers are already generally a high-stress, terrible working environment even when one is doing routine customer service or order entry work; a debt collector is the type of call center worker that no one wants to talk to. Successful debt collectors need to both harden their hearts against tales of woe (and personal abuse) from debtors and also successfully execute on some combination of moral suasion and threats to eke out a few percentage points at the margin in collections. Success in the field is the difference between e.g. 7% and 8% collection.

Few competent people remain in this field over an extended period of time; an effective, morally upright debt collector would be a better telephone sales rep at a tenth the stress and two to five times the total compensation. As a result, most calls being made today are not by effective, morally upright debt collectors. Illegal threats may be more effective than diligent suasion and management frequently, through incompetence or careful refusal to watch the sausage being made, ignores that collectors are making threats but tracks per-collector productivity very carefully. Solve for the equilibrium.

As a former advocate, I’d report that it is never in a debtor’s interest to verbally speak to a debt collector under any circumstances. One’s likelihood of being abused or lied to, including in financially consequential fashions, is high, and one’s ability to counter that is minimal. Instead, force them to do all correspondence on paper, where lies are self-documenting, illegal threats are immediately admissible to regulatory processes or court, etc.

The incentives and hiring pool available to debt collection firms cause them to put most of their limited competence into scaling phone teams and very little competence into individualized written correspondence. That would require them to successfully keep a folder about a debtor for a month and route communications to the desk with that folder; this is an edge case for them and uneconomical to do at the hundreds of dollar range. This means they will, through a combination of incompetence and strategic decisions by management, frequently simply abandon collection activity where that would require them to read and write.

Suing people with robots

Most debts are not litigated, but various debt collection firms have various strategies here. Almost no consumer debts are worth serious amounts of professional attention, but above about $1,000 or so (depending heavily on jurisdiction), they can be worth doing high-frequency lawyering on.

Basically, the debt collector will either itself or, through use of a law firm, file hundreds or thousands of lawsuits in a single court against its roster of uncollected debtors. Those lawsuits will be entirely templated. A frequent detail that emerges in news reporting is that they’ll often be physically robosigned, so that the relevant lawyers don’t give themselves repetitive stress injuries by needing to affix their name to all of the things they are swearing to under penalty of perjury.

You might have a view of the judicial system where two highly educated advocates joust in front of a neutral finder of fact like a judge or jury. This ain’t that.

The goal of this business process is generating so-called default judgements. They’re both judgment that a default happened and also the judgment that happens by default; the overwhelming case for issuance is when one party is summoned to court and does not show up. Many debtors will ignore correspondence about court cases believing that ignoring them will make the problem go away; this is the opposite of the truth. Many will lack transportation, not be able to take off work, etc. In any event, filling 100 court cases means that, in expectation, your lawyer will collect 60-95 default judgements or more. Not a bad day’s work.

Default judgements turn a debt that hasn’t been collected voluntarily into a financial asset with value. They can be used to garnish the contents of bank accounts or wages paid to debtors, in many jurisdictions. For debtors who own property, which is a fact pattern that occurs much more commonly than you probably would guess, a default judgment can frequently be turned into a lien against the property. This swaps an unsecured consumer debt for a senior claim against the debtor’s home equity. That claim is itself salable into an ecosystem of lien investors, and the debt collector will frequently sell it for (say) 80 cents on the dollar, realizing a profit over their low basis in the debt and collection costs.

Should a debtor actually show up to the courthouse, the lawyer will almost invariably offer an on-the-spot settlement. Fifty cents on the dollar seems to be popular, anecdotally. Avoid the expense and stress of a legal proceeding, etc etc, this is obviously incentive compatible.

When debtors actually contest their debts, or have competent legal representation, the debt collectors frequently get beaten like drums. Many judges take an unsurprisingly dim view of lawyers who profess to have personal knowledge of the facts of the case then can’t locate the contract they are suing over. (Again: they do not and never did possess a copy of the contract. The lawyer does not actually know any facts of the debt other than that the firm purchased a name, address, and perhaps a Social Security number in a CSV file. They don’t have the contents of any previous correspondence about the debt, owing to some combination of incompetence and unwillingness to present evidence of crimes in court.) Then comes some gamesmanship, where the debt collector will attempt to ask for an extension to get their paperwork in order and the debtor’s attorney will push for dismissal, claims under the FDCPA and similar, and costs.

This makes for good human interest pieces occasionally, but the economics make the lawsuit machine very profitable to run even accounting for losses, and accordingly it continues robosigning cases by the tens and hundreds of thousands.

What can be done about this?

Many people have suggestions for obvious improvements here, which might rhyme with the Federal Trade Commission’s suggestions from 2010 or the Consumer Financial Protection Bureau’s spate of rulemaking from 2021 through 2023 or the FDCPA from 1978.

I’ll leave rulemaking to the rulemakers, as this essay has gone on long enough. The state of reality is very bad, not because any single person woke up this morning attempting to maximize evil, but because a complex interaction of structural factors over multiple different parts of society got us to where we are.

Untangling that Gordian knot is hard and has empirically eluded us.

The slow-motion banking crisis we’ve covered for the past several issues of Bits about Money is slightly awkward for me to talk about, for social reasons. I decided to lean into the awkwardness a bit this time.

(Hopefully we’ll be back next time on non-banking topics. To the extent the U.S. banking system is solvent and boring there should be more interesting things available to write about! It is only when large portions are suddenly insolvent does it become irresistibly interesting.)

There are two general sources of awkwardness when writing about banks: one, the industry guards its reputation very zealously. Surprisingly small mentions of individual firms can bounce around enough to reach people professionally relevant to me. (As always, while I am a past employee and current advisor at Stripe, it does not necessarily endorse what I write in my personal spaces.)

Two, I am inclined to scrupulosity in disclosing relationships when writing about firms. No one anywhere seriously expects a conflict-of-interest disclosure over a checking account, but once one is talking about e.g. mortgage-sized sums, you have to at least think about it.

And so I have a relationship which I’d like to disclose, which is actually a jumping off point into a wider discussion of things misunderstood about the recent (and ongoing) crisis.

First Republic’s Personal Line of Credit

One of the best products offered in the U.S. financial industry the last few years, and broadly underappreciated, was First Republic’s personal line of credit. (The division of Chase currently operating as First Republic has a product named the same thing as of this writing and it is not materially the same product for reasons we will soon discuss. The Information reported recently that First Republic is quietly sunsetting this business.)

Their publicly generally available offer was an unsecured signature loan for up to (generally) $100,000. It had a 2 year draw period, during which one would pay interest only, then amortization over the remainder of the pre-chosen lifetime for a total of 7, 10, and 15 years. There was no penalty for early repayment, not that repaying early would have been a good idea.

That’s already a combination of attributes you would not find many banks offering. Now here’s the kicker: it was, until the recent failure of First Republic, written in quantity at a sweetheart interest rate.

How sweetheart? Well, as of early 2021 (for an example we’ll return to), you could get a 7 year fixed rate of 2.25%. (This was net of discounts, which we’ll return to.)

This was ludicrously favorable and that was an intentional strategy. We’ll talk about the strategy more in a moment.

In which I unexpectedly found myself in San Francisco

First Republic would only write loans within their branch footprint, which (to simplify slightly) is “the parts of the United States most densely populated with high net worth people.” I have lived almost all of my adult life in Japan, and so could not take advantage of this. (Japanese banks, despite the prevailing interest environment, are markedly less likely to write unsecured consumer loans at low interest rates, for a variety of reasons.)

Then mid-pandemic I suddenly found myself running a non-profit (best known as VaccinateCA) which ran the United State’s covid vaccine location information infrastructure. It is a somewhat wild story, and I’ve told it before elsewhere.

In the very early days of that effort, knowing that I couldn’t survive continuing to work from midnight to midafternoon and that official California would eventually react with extreme displeasure to being puppeted by a geek in Tokyo, I bought a one-way ticket to San Francisco. I told my wife and kids I’d be back either in a few weeks in the case of failure or much later in the case of success. It ended up being about 5 months.

One of my main jobs for VaccinateCA was stumping for money. Concurrently with raising money for the charity, I also needed it for us, because maintaining a second household in San Francisco is not a cheap proposition and I was temporarily on leave from my employer. I didn’t want to complicate fundraising discussions with my own situation, so VaccinateCA paid me $1 and I absorbed personal costs by shattering the (metaphorical!) piggy bank.

Here I’ll acknowledge some residual middle class guilt for talking straightforwardly about finances, but I think it’s useful to understand both for this story and for the larger issue of understanding the banking system.

(The banking system must, of course, frequently process transactions which are larger than various comfort levels. The class norm of not talking about money silos the information about those transactions, to the detriment of both class members and also the wider public. This has been replete in discussions of the banking crisis. Many commentators react in mock horror to discussions of accounts with more than the FDIC insurance limit in them, as if that didn't include almost every company with more than 10 employees in the country.)

So, the awkwardness: the piggy bank had about $100,000 in it. This seemed like a dicey amount for an open-ended commitment with functionally no income while supporting two households in high-cost cities.

Now I could have gotten creative in financing personal expenses, but I preferred spending almost all of my time on VaccinateCA. So I had a brief negotiation with First Republic, where I asked for (and got) a $100,000 credit line “for cash management purposes.” My recollection is that this took less than two hours total, inclusive of time to write the loan application.

I was not raised to be enthusiastic of debt, but inking that credit link was an enormous relief for me. It meant that I could almost ignore my family’s personal financial situation for the duration of VaccinateCA.

I eventually drew all of it. (To make a long story short: our charity raised from a variety of tech industry funders, frequently with a substantial lag between verbal commitment and receipt of the wire. We were operating at a cadence much faster than most funders. When we received a commitment to funding, I sometimes advanced money to the charity with the intent of recouping it after the donation had actually arrived. This was to accelerate shots into arms, our sole goal. After we had operated for a few months, the funding environment changed in a fashion that made not all promised grants actually arrive. In lieu of causing the charity to shutter early, I recharacterized my loan to it as a donation, and the marginal cash saved paid salaries and expenses in our final weeks. This ended up being $100k out of the $1.2 million we raised.)

The fundamental purpose of bank loans is to enable measured private risk-taking by leveraging a small amount of bank equity (from risk-taking investors) with a larger amount of risk-adverse deposits. Sometimes the risks are opening a restaurant or buying an apartment building in an up-and-coming neighborhood; here the risk was a crash project to build charitable medical infrastructure during a crisis.

Risk is not a four-letter word. Society wants restaurants, apartment buildings, and crash projects to build charitable medical infrastructure. The banking system enables a higher rate of creation of these goods than would prevail in an environment where only risk capital was available to fund them. This is its main social purpose; the checking accounts and payments infrastructure and tastefully decorated branches and bonus checks are all consequences of it.

Society should be thrilled it has banks, like it should be thrilled it has power plants. The alternative is a far worse world.

So you’re a bank underwriter

Let’s play the world’s most boring game of Dungeons and Dragons: pretend you are sitting on First Republic’s credit committee. What do you need to see in a loan application packet to underwrite this loan?

Well, you need KYC information, clearly. That’s straightforward; you had a U.S. passport passed over a counter at a branch. It matches an existing U.S. credit profile, which both solidifies your KYC story and also answers most of your worries about credit risk.

Underwriters would traditionally ask about capacity to repay, and while this product was offered on sweetheart terms, it was underwritten reasonably rigorously. Two discounts offered to the rate were contingent on depositing 10%-20% of the line of credit amount in a First Republic checking account. This both directly decreases risk via acting similar to collateral and indirectly decreases risk because most people who are poor credit risks can’t come up with $20,000 in cash.

Now you turn to the income story. Here is where most bank underwriters would have noped the heck out: my documented prior income was “weird” by the standards of U.S. banks. It wasn’t W-2 and was denominated in yen. (W-2 is the U.S. tax form issued by employers to document wage income, and one’s “W-2 income” is the most legible form of income for the U.S. financial system. All other forms of income, of which there are many, are harder to underwrite to.)

One thing which First Republic historically did very well was parsing certain varieties of “weird.” I was pleasantly surprised to see that the loan application anticipated partially international transactions; there were pages of the workflow dedicated to that. This was not their first rodeo.

An aside: You’d be surprised how many U.S. banks, of all sizes, are completely incapable of dealing with this as a matter of procedure. As a consequence, they have large lines of business utterly incapable of touching anyone who needs to present non-U.S. dealings in their file. Banks with incompetence regarding mobile people include, strikingly, many which have substantial international operations in capital markets and commercial banking. The biggest banks in the U.S. brag se habla español and then are utterly befuddled that immigrants exist, leaving the Spanish-speaking ones to Seis (a small angel investment of mine). Anyhow, back to the more functional bits of banking.

Underwriters aren’t concerned with past income, per se. They want it as a proxy for future income, and therefore future ability to repay the loan. And here, First Republic was simply willing to stretch a little for a desirable customer. Sure, my immediate future looked upside-down financially, but they believed my mid-to-long-term career prospects were fairly good, and were willing to go along for the ride. (One might sensibly wonder “Did your balance sheet make any difference?” and the answer was a resounding “Nope.” They were utterly uninterested in e.g. private tech equity, on a “Don’t even show us the docs we will not count it in your favor for this product” level.)

Why did First Republic stretch here? Was it due to a one-off exception? Not having been at the credit committee’s meeting, I can only speculate, but I speculate that this was rubber stamped as being clearly within the parameters of this product. I had a somewhat-higher-than-typical degree of weirdness in my application but the product was designed to attract the business of people who’d routinely have weirdness like e.g. working at a startup they founded, earning most of their income via carried interest and not on a W-2, etc.

Successful Millennial Generation Strategies

Sometimes companies do other-than-straightforward things for strategic reasons. This is often the cause of a lot of external speculation, sometimes verging on conspiracy theorizing.

It is underappreciated that publicly traded companies will frequently write down their strategies, explicitly and at substantial length. And so we don’t have to speculate why First Republic offered sweetheart deals on lines of credit.

Per their 2021 annual report (emphasis added):

Our next-generation client strategy continues to be highly successful. Drawn by our Personal Line of Credit, Professional Loan and affiliate programs, younger client households grew 14% during 2021. We’re attracting younger urban professional households even earlier in their careers. This strategic initiative is intended to engage younger clients with specific products to spur trial and subsequently build deep, lasting relationships. It’s been transformational at First Republic. At year-end, millennial households represented over 40% of First Republic’s total consumer borrowing households, compared to only 12% in 2015.

First Republic talked this product up to investors, regulators, and other stakeholders for years. (The heading for this section is stolen from their quarterly reports; they recycled it frequently.)

They had a structural problem common in the banking industry: a commanding share of their deposits were held by retirees. (It is broadly underappreciated how much wealth in the U.S. is held by seniors, almost entirely due to lifecycle factors.) The typical behavior of older households is to spend down their savings. The bank would be inconvenienced if it saw large deposit outflows (oh howdy was it aware of that risk), and so it made a bet intended to pay off in decades: get young millennial professionals early, in their pre-rich years, and then hug them tightly for life.

Interestingly, due to banking regulations, you’re not actually allowed to say This Loan Is For Young Professionals because age is a protected class. You’d have to go very far into the archives to find this loan explicitly referred to as being for young professionals in the marketing material. Compliance eventually read it and (one presumes) gave Marketing a non-recorded phone call whose contents are very predictable. So strategically, the loan program was designed to generate additional business from young professionals, even if the bank underwrote individual loan applicants without regard to age.

A brief digression about permissible and impermissible discrimination in lending: I have no reason to believe the bank actually discriminated on age, at least outside the socially-accepted way of discriminating against FICO scores that incorporate time-credit-profile-has-existed, which reliably disadvantage the young by proxy. To the extent we say age is a protected class, anyway, in U.S. practice this broadly means the old are protected against the young but not vice versa. (Surprised? See page six.)

This line of credit grew out of an earlier program, where First Republic offered it explicitly for debt refinance. The basic sketch of that was “You’ve spent a lot of money on education and now made it into a high-earning job; refinance your education with us at a sweetheart rate and we capture your deposit business in return.” By 2020 that was so successful it expanded into lines of credit, which didn’t merely capture backwards-looking debt but could potentially fund forward-looking expenditures.

Building the deposit franchise

The goal of this loan program was to function as a loss-leader for the deposit franchise. First Republic’s business model was to charge richly for providing good banking services to well-off customers.

(For a combination of culture-of-banking and regulatory reasons, it was open to the general public. Almost anyone could have opened an account there with $25. But the core of their business, to a much larger degree than most banks, was high net worth households and the relatively small organizations they run. Their typical business client only had a few hundred thousand dollars on deposit; it was someone's local charity or small business, not a large enterprise.)

They paid their customers almost nothing on their deposits (0.40% in 2022, up rapidly from 0.07% in 2021), lent them back out to the same customers for more-than-nothing, and ran a very profitable business for many years.

They didn’t expect to lose money, per se, on lines of credit. The normal usage of “loss leader” is to describe supermarkets’ practice of pricing certain high salience items (e.g. milk) at much tighter unit margins than they price most of the store at. This doesn’t mean that they necessarily lose money on each gallon of milk they sell, though negative unit margins occasionally happen in the world.

First Republic expected nearly zero credit losses in this program and they were right about this expectation. They used to quote a total number of individuals who were presently late in their quarterly presentations and you could count to it on one hand. If you wanted the feeling of rigor, you could go to the crunchy part of the credit quality analysis and see that on $3 billion in outstanding unsecured loans (of which this program was a subset) they had $1 million in loans which were 30-60 days late and no defaults whatsoever.

First Republic additionally bragged that the households attracted by their Successful Millennial Generation Strategies™ were largely self-funding. The strategy was designed to pay off over a period of decades but was on track to outperform plan. In a period of mere years it had already gone from (effectively) their older, wealthier depositors subsidizing acquisition of younger, less wealthy customers to being internally self-sustaining. Accounts of slightly older millennials were already bringing in enough deposits to almost cover the loans of slightly newer vintage households.

How did that happen?

For one thing, these were lines of credit, rather than loans. A line of credit can be attractive in option value terms without actually being drawn upon. If you incentivize AppAmaGooBookSoft employees to move their core checking accounts to you by offering them sweetheart rates if they ever need it, many will take you up on that without ever needing it.

For another, the terms of the lines themselves incentivized partial coverage of loans by the borrowers themselves. Interestingly, this was more on a handshake than a contractual basis. My paperwork quoted a 0.50% interest reduction for maintaining at least a 10% of the line ($10,000) deposit average and 0.75% total for 20% ($20,000).

For, as best I can determine, operational or software reasons, First Republic wasn’t actually capable of dynamically altering the loan rate every month in response to one’s actual savings behavior. They just assumed you’d keep to the handshake, wrote the final interest rate into the contract and loan servicing database, and charged you that regardless of your balance every month.

And for a third, this product was designed to (and successfully did) attract high-earners during a period of their lifecycle where they would, in expectation, go from having almost no assets to having substantial liquid wealth and a large paycheck arriving every two weeks. The modal newly hired tech employee in San Francisco who owns no mattress frame and needs an account for their first paycheck will not have zero dollars in that account in five years.

I recall a funny conversation during my account opening, which is an almost-too-good-to-be-true window into the socioeconomic weirdness that is being a young professional in tech. The banker assisting me asked how much I wanted to open the checking account with.

Me: “I think a hundred.” (As we had already been discussing the line of credit offering and my rationale for seeking it, I assumed this was unambiguous.)

Him: “For clarity, do you mean a hundred… dollars? Or…”

Me: “Oh, no, sorry. A hundred thousand dollars.”

Him: “I’m glad I asked. We’re running a promotion for new checking accounts, and…”

Me: “Ah yeah, it’s a funny thing in this town that it could have been a hundred or a hundred thousand.”

Him: “Oh, you wouldn’t be the first to mean the other hundred.”

All of those are plausible for a lanky San Franciscan who shows up to open a checking account while wearing a track jacket. Some portion of those new relationships will prosper, and the bank prospers with them.

First Republic took substantial losses on these (and other) loans

During the initial phase of the banking panic, people concentrated on large losses in banks’ portfolios of marketable securities, partially Treasuries but mostly mortgage backed securities. Tsk tsk, bank risk managers, why are you speculating on interest rates and not performing the traditional function of banking, making solid loans backed by strong credit?

First Republic is no longer with us not because of losses on their available-for-sale or held-to-maturity securities but rather because of large losses on their loan books. The overwhelming majority of them were on fixed rate mortgages secured by primary residences in places like New York and San Francisco. That was many tens of billions; these loans were only a handful of billions.

Here it is useful to point out that bond math applies to loans in addition to bonds: a 1% increase in prevailing interest rates decreases the value of the loan by approximately 1% per year of duration.

As a worked example, the line of credit I signed from early 2021 has approximately 2.5 years of duration still on it. (I will be paying it for five more years, the average amount for those 5 years is half of the current total, etc.) Interest rates rose by about 5%. And so First Republic took more than a $10,000 paper loss on my business.

I’m still paying! I still bank there! My financial situation has improved markedly since I opened my account! Everything is going exactly according to plan! And yet, when replicated across their loan book, that evaporated many tens of billions of dollars of equity.

This would have been survivable had they not suffered $100 billion in deposit flight, in the early stages of the banking crisis.

When Chase bought First Republic, they effectively received an inducement from the FDIC to cover the “pain” of buying loans (new assets of Chase) which had decreased in value at the same time as they absorbed deposits (new liabilities of Chase) which had not decreased in value. The total cost of those inducements was estimated by the FDIC at $13 billion. With respect to me specifically, I think they got about $10,000. Not bad work if you can get it.

(Here I’ll make the obligatory disclaimer that I’ve been a Chase customer for a very long time. Without going into inappropriate levels of detail, let's say that they have richly earned me performing exactly to contract.)

The broader picture

Every time a firm goes out of business unexpectedly, a tiny bit of light goes out of the world. I mourn a bit for First Republic, much like I mourn the local barbecue joint that couldn't make Tokyo rents during the pandemic.

It is likely that the customer service attitude and risk taking culture that was distinct to First Republic will eventually be fully subsumed into the Chase borg.  Should we, as a society, be happy about that? It is probably the least worst option we had in 2023, contingent on a fiscal response to the pandemic which broke large portions of the banking sector.

It has been quite popular for various parties to point the fingers at bank management teams, saying that impressively impecunious operation of the core business of banking is the proximate cause for the crisis. The crisis is caused by the pace of change in interest rates. Everything else is commentary.

There are many more good banks out there, which were and are in what I’ve called the sweat and smiles business. They made good loans to good borrowers, taking limited risk in the service of encouraging private risk-taking.

Those banks are dead as a result. Academics estimate that there are thousands of them. Almost all of them are still shambling around, like well-dressed extras on the Walking Dead.

The current social consensus is that we presently expect most to limp their way out of the crisis. I think the consensus underrates the need to recapitalize the banking sector, to the tune of several hundred billion dollars.