In this article, Patrick McKenzie explores why community banks, often overlooked and dismissed as small-town relics, are actually financial dark matter with significant market influence. He offers insights into the functions and risks faced by community banks, and how they differ from larger, algorithm-driven institutions.

One of the best things about the Internet is that it both provides infrastructure for society but also demystifies that infrastructure. I’ve spent the last few years going deep on financial infrastructure while working at Stripe, and thought it might be useful to geek out about finance with software people and software with finance people.

Obligatory disclaimer: these are my own views, not those of anyone I am associated with.

Community Banking: The Dark Matter of the U.S. Financial System
The United States has many more banks than peer nations. There are more than 5,000 banks in the U.S. but only about 400 in each of the U.K. and Japan, for example. One reason for this is that the U.S. is dependent on community banks throughout much of the nation.

A community bank is a locally-oriented financial institution, generally much smaller than regional or national banks, focused largely on the “traditional business of banking” (taking deposits and lending) versus the capital markets functions that the “money center” banks also engage in.

Community banks are often assumed to run on small town values, glad handing, and sleepy scenes reminiscent of It’s A Wonderful Life. This causes them to be overlooked by technologists and the wider financial industry, to our mutual discredit. Community banks are actually financial dark matter; their market impact and the policy regime supporting them influence all Americans’ access to banking services and many fintech product offerings.

Sketch of a community bank

Let me sketch out a small software-enabled company with approximately 40 employees and a cap table that can fit around a dinner table:

The 50th percentile [community bank] … held nearly $169 million in total assets. … At a 4 percent margin (the difference between the yield on loans and the interest paid for deposits), a $169 million bank generates roughly $6.5 million in annual revenue. … [C]ommunity banking is essentially a highly regulated small business enterprise. [0]
Banks in the U.S. can be chartered by the federal government or the state; community banks (and their close cousins, credit unions) are dominantly chartered by states. They form the largest portion of the banking industry in exurbs and rural America; 20% of counties have no bank branch except through community banking [PDF]. While they’re smaller as a percentage of banking activity than they have been historically, they still represent about 10% of all loans and deposits kept in the country.

The business of a community bank is taking deposits cheaply and lending at a higher rate. Both of these functions are historically local in nature, but that is changing. Community banks sell themselves on convenience and having a human-centric relationship with customers, as opposed to the driven-by-algorithms approach of the money center banks.

The local nature of community banks is not just a source of warm fuzzies. It is also an important risk factor. If you are the bank in a factory town, and the factory closes, your deposits will be spent down (and so your cost of funding goes up) at the same time your loan customers suffer financial stress (and your credit quality deteriorates). This dynamic kills banks without diversified customer bases, particularly during financial crises. The global financial crisis is often remembered as a cautionary tale about big banks and the global financial system, but the friendly bank on Main Street suffered the worst; over 400 banks failed between 2008 and 2011. More than 85% were prototypical community banks; small, local institutions with less than $1 billion in deposits.

The financial system is a policy arm

The U.S.’s financial system is best conceived of as a public/private partnership. Each financial institution is in constant competition with all others for customer relationships, deposits, loans, and ancillary product revenue. However, the government does put a thumb on the scale, particularly for politically powerful groups.

Community banks are surprisingly powerful in the U.S., largely because they are very popular with their customers relative to Big Finance, they are indispensable for politically powerful local groups like landlords, real estate developers, and farmers through their commercial loan books, and they have earned an appealing narrative about financial access. It is difficult to overstate how dependent commercial real estate is on community banks in much of the country, and it is a symbiotic relationship; the investors who organize them are usually local business owners with heavy real estate interests.

Branches are not cheap to create or operate, and the large banks concentrate them in densely populated areas with relatively wealthy customers and businesses nearby. Community banks are willing to take worse economics to have branches in places where the large banks don’t; this keeps those places tied to the national financial system.

This is a policy aim of the government, both due to the economic impacts and because most residents being banked is core to orderly provision of benefits, taxation, and other government functions. A bank branch is a retail point-of-presence for the SSA getting funds to a pensioner, for the IRS collecting payroll taxes, and for the DEA interdicting fentanyl, staffed and funded by the private sector.

Impact of community banking outside the system

Because the continued existence and thriving of community banking is a policy goal, regulations or competitive measures which would disfavor them versus the large banks and other financial players are actively undermined. Fintech frequently feels the ripples of this.

Many technologists ask why ACH payments were so slow for so long, and come to the conclusion that banks are technically incompetent. Close but no cigar. The large money center banks which have buildings upon buildings of programmers shaving microseconds off their trade execution times are not that intimidated by running batch processes twice a day. They could even negotiate bilateral real-time APIs to do so, among the fraternity of banks that have programmers on staff, and indeed in some cases they have.

Community banks mostly don’t have programmers on staff, and are reliant on the so-called “core processors” like Fiserv, Jack Henry & Associates and Fidelity National Information Services. These companies specialize in extremely expensive SaaS that their customers literally can't operate without. They are responsible for thousands of customers using related but heavily customized systems. Those customers often operate with minimal technical sophistication, no margin for error, disconcertingly few testing environments, and several dozen separate, toothy, mutually incompatible regulatory regimes they’re responsible to.

This is the largest reason why in-place upgrades to the U.S. financial system are slow. Coordinating the Faster ACH rollout took years, and the community bank lobby was loudly in favor of delaying it, to avoid disadvantaging themselves competitively versus banks with more capability to write software (and otherwise adapt operationally to the challenges same-day ACH posed).

Durbin Exempt Interchange

In the wake of the global financial crisis, a broad swathe of reforms was passed to clip the wings of the financial industry. One amendment to the 2010 Dodd-Frank bill, named for its sponsor Senator Durbin, imposed a limitation on the amount of interchange that could be charged on debit card transactions. The amendment had a carve-out for smaller financial institutions with less than $10 billion in assets.

A full recounting of the effects of the Durbin amendment would have to cover free checking accounts, insufficient funds fees, and the competitive dynamics of interchange within a multi-sided credit card ecosystem. The part most relevant to technologists, however, is that the Durbin exemption for small financial institutions makes debit cards very lucrative if and only if they are issued by a small financial institution.

Think of a fintech company with a debit card. Square’s Cash App. Chime. Robinhood. Stripe Issuing, and users like Ramp and Bench. All of them, virtually without exemption, are brought to you courtesy of a small financial institution that you are unlikely to have heard of (unless you work in fintech or read Terms and Conditions for fun). Durbin exempt interchange is the revenue model which got a thousand pitch decks funded, because it is free to the ultimate consumer and protected by a moat guaranteed (for the moment) by Uncle Sam, contingent on the consumer choosing to use your app’s card.

This let a generation of those apps compete on user experience without being crushed by large banks' pricing power. Customer acquisition for fintechs would be difficult if you could walk into any of the top ten banks and walk out with a 1% rewards debit card (which previously existed but don't anymore at large institutions, due to the Durbin amendment capping their revenue substantially lower than that).

Fintech as the banks’ Finfriendemy

Fintech has a complicated relationship with community banking. Community banks had already suffered some of their business leaking to larger banks as banking moved increasingly online, but were not worried about large banks competing with them from branches in their core markets. The math never penciled out.

Online customer acquisition, however, has allowed well-funded fintechs to provide better services cheaper than the local small business financial institution. Some customers are voting with their wallets. This has become even more acute during the pandemic, when footfalls (the delightful bricks-and-mortar phrase for what technologists might call eyeballs) at branches declined precipitously. One of the key themes for the industry in the past year has been maintaining the sense of human connection with someone who never sees your staff anymore.

At the same time, partnering with fintechs has allowed some community banks to greatly strengthen their franchises. Cross River Bank is an interesting example; a major line of their business is originating loans for fintechs which are then largely sold to funding sources that the fintech has lined up (such as investors looking for yield in a low interest environment). This gives CRB an interesting cross section of credit risk across the U.S. footprint of their customers’ customers rather than just the credit risk local to the bank in New Jersey, and allows them to share in the risk with the fintech and associated lenders rather than taking it entirely themselves.

One commentator [1] waggishly described Affirm as an API between Peloton and CRB, due to Peloton being a very large account at the time of their IPO. This is probably unfair to all three parties; ~30% revenue concentration is certainly notable, but each company is in a very different business. None could engineer their to the desired user experience without both other partners.

It is an open question as to whether many community banks are positioned to create niches like this for themselves in the Internet economy. One model that reduces the sophistication required and level of strategic risk is platforms providing them capabilities which bolt onto the traditional business of banking.

Brokered deposits are one example of a function which is increasingly becoming a platform. Banks use deposits as a cheap source of funding for loans, but the growth of deposits in a local community is constrained by, among other things, the underlying economic growth of the community, and banks might max out their deposit base without being able to serve the demand for loans in their community. The traditional remedy is tapping deposit “brokers” outside the community, who largely represent large, sophisticated pools of money which will go anywhere with nationally competitive interest rates.

In recent years this is getting intermediated by software rather than by people serially negotiating over phones on a daily basis, to the mutual benefit of community banks, firms which have lots of end users with deposits, and those end users.

This model forms a major portion of the revenue mix for virtually every discount brokerage (with the notable exception of Robinhood). A brokerage might have billions of dollars of customer cash on hand but, if it doesn’t have an affiliated bank, very limited ability to turn that cash into net interest revenue.

A platform like StoneCastle intermediates between the brokerage and banks, solving both incredibly fun operational challenges of storing and moving money, and effectively bidding the deposits out daily to the banks most willing to pay to support their loan books. The brokerage's customers get FDIC insurance on their deposits and liquidity which might feel quite close to cash. Meanwhile, the net interest spread the broker collects for this subsidizes the other services of the brokerage. This allows them to offer e.g. commission-free trading and asset management "for free." (It is popularly believed that payment for order flow is the primary subsidy brokerages receive. This belief is straightforwardly wrong for every brokerage except Robinhood.)

Brokerages have gigantic staffs of engineers, customer support agents, and other professionals. All of them are paid for by, effectively, the imbalance between checking account balances and demand for commercial real estate loans at a particular small bank in Kansas.

The financial infrastructure the world rests atop of causes many underappreciated ripples in our lives. It is full of these delightful rabbit holes. I hope to explore many of them with you in future issues of Bits about Money.

Further reading
[0] This quote is from The Most Fun I Never Want to Have Again, a swashbuckling startup tale about a Georgia community bank. It is part personal career narrative, part fundraising journey, part paean to the public image of community banking, and part behind-the-curtain look at how sophisticated executives at them evaluate their own success. Highly recommended.

[1] Marc Rubinstein, the above cited waggish commentator, writes a monthly newsletter on banking which is compulsory reading for anyone interested in the space.

The community banking community itself publishes surprisingly many podcasts, particularly during the coronavirus era. They’re an interesting mix of marketing content, shooting the breeze with peer professionals, and (to mangle the Picasso observation) artists getting together and passionately talking about where to find cheap turpentine. I particularly enjoyed an episode about the business benefits of colocating a cafe (even when, as it turns out, regulations forbid a bank from actually selling coffee).

Community Banking and Fintech

Discover the surprising progress of financial innovation in this thought-provoking essay. Contrary to popular belief, the world is actually improving within our own lifetimes, and much of the progress has been in under-the-hood infrastructure that goes unnoticed by the general public. Find out about the emergence of free-as-in-beer banking, the transformation of the checking account landscape, and the role of competition in driving change. Don't miss this fascinating exploration of how financial innovation is shaping our lives and what we can expect in the future.

If you say the words “financial innovation” many people incorrectly believe it is largely negative (or at least risk-taking) or not happening. This is incorrect. The world is actually getting better, within our own lifetimes. Much of the improvement has been to under-the-hood infrastructure not paid much attention to by the general public, but since the financial system touches almost every part of our lives, it is worth knowing what has happened and on which axes we can expect further improvement.

Free-as-in-beer banking
Advocates for the un- and underbanked have long had one request of the financial system: can you just get us a free basic checking account, please. This was the white whale for decades. It was technically possible but would have been extremely burdensome for community banks, because it is (fundamentally) a call to subsidize some depositors at the expense of others. That subsidy would have been bearable for the largest, most broadly distributed financial institutions in the world, but would have covered almost the entire deposit base of many community institutions. The tough part about subsidies is that someone has to pay for them. As we discussed previously, community banks exercise quite a bit of influence in the U.S., and so proposals for "free banking" got quietly killed.

So it would probably surprise advocates, burned from this discussion in the 1970s, 1980s, 1990s, and 2000s, that in 2021 the U.S. has nationally distributed free banking offerings that users love, that are profitable for the industry, and that do not stem from a regulatory mandate.

The key driver of the change has been competition for the deposit business of regular households. It's a fascinating story.

Perhaps some readers may not remember this, but prior to the mid-1990s checking accounts had price tags. In the 1990s, “free checking” became a marketing phenomenon, partly aimed at growing the number of banked households but mainly a pricing war between a consolidating industry hunting for share in local markets.

Free checking wasn’t free. Instead of most depositors paying a predictable (and relatively small) fee for their checking account, a tiny portion of the depositor base was assessed many, many $25-$35 fees stochastically based on how frequently their incomings and outgoings were temporarily mismatched. This resulted in a perverse tax-the-poorest-harder situation, where more than $10 billion annually was collected, with the large majority of it coming from less than 10% of banked households [PDF]. Overdraft fees made the math for free checking work for most of the country, at ruinous cost for a portion of it.

But. Things. Are. Getting. Better.

Online and, especially, mobile banking have given consumers more up-to-date visibility into their funds flows and allowed them more control over fees. That alone wouldn’t have helped (and didn’t help) folks whose primary problem was that they didn’t have enough money for obligations at least some of the time.

What has helped there is competition by apps, which a) knew that users who experienced them virtually universally hated overdraft fees and b) successfully decreased servicing costs for marginal users to the point where they could be profitable on tiny amounts of per-account contribution margin.

This combination let them ostentatiously kill overdrafts and use that for marketing. Chime’s recently highlighted “Fee-free overdraft up to $200.” Cash App similarly leaned into this for years.

The financial industry thought, for many years, that the customers added or retained by these improvements would be disproportionately expensive to service and low-revenue. Cash App has experimentally disproven that. Their cohort graph is a thing of beauty (see page 9 here [PDF]); enterprise SaaS level net expansion revenue on cohorts of consumers recruited at a $5 cost of acquisition.

This did not go unnoticed in the financial industry, particularly in 2020 when “widely distributed branch network” became a tougher thing to sell on. Online-focused banks are aggressively competing on this axis. Ally Bank dropped overdraft fees entirely. Some in the industry expect at least one top-10 bank to shortly experiment with it for marketing purposes, which (given competitive dynamics, particularly vis regulators) will likely cause a lot of following.

We’re not done with improvements here, given approximately $10 billion in overdraft fees assessed a year, but it was considered impossible for decades to do the thing you can now get from multiple competing providers available on every smartphone.

Sidenote: Smartphones themselves are a beautiful example of technical and financial innovation making new capabilities almost universally available. When the iPhone debuted it was high-end consumer hardware. These days entry-level Androids are given away to attract mobile service contracts.

Fast transfers
If you’re an internationally-minded American in the financial industry, you get ribbed about two things constantly: checks (“what do you mean the country that invented high-frequency trading mails around pieces of paper to move money and that those pieces of paper have account security tokens literally printed on them in block letters to make them easier to read”) and transfer times.

“Moving money between bank accounts takes 3-5 business days” has been the rule in the United States for most of my life. This quietly changed and many people have not really noticed. It happened via a combination of sustaining innovation, greenfield transfer networks, and heuristic credit decisions made at superhuman rates by computers.

The Automated Clearing House (ACH) moves about $60 trillion (!) a year. It was started as a backoffice improvement to solve the problem that banks processing checks would have to mail or courier checks between each other, and gradually expanded into electronic payments over time.

ACH payments historically operated on a daily or twice daily batch processing, and they’re (importantly) “negative confirmed.” This means that you do not get immediate reliable feedback that an ACH transfer or withdrawal has succeeded. The association’s rules require financial institutions to report errors (via even more batch processing) within 5 business days, and given that errors are expected in the ordinary course (most commonly, insufficient funds, since funds availability is not and cannot be guaranteed at transaction time), businesses and banks have to make decisions on what degree of risk they’re willing to tolerate.

ACH moved to same-day settlement (rather than next-day settlement) in 2018 for most transactions. This has accelerated funds availability, but it did not achieve real-time or near-real-time guarantee about funds sufficiency, so giving the user money before the 5 day window expires still runs some credit risks.

Banks have gotten much better at evaluating this credit risk and accelerating payments for low-risk transactions, such as federal government transfers, established direct deposit relationships, and others. The “others” is the most interesting category, because it’s a problem which would have been intractable prior to computers but is now fairly pedestrian data science, and using it is uniformly pro-consumer (giving people access to their own money faster).

Depending on the institution, some of the factors that go into heuristic acceleration might include age of account (more established accounts experience less errors, of virtually all types, than newer accounts), frequency of the transferer/transferee relationship, experience across accounts regarding the transferer (“Shouldn’t we treat Google transfers as money-good? They literally never bounce.”), desirability of the customer relationship (“Shouldn’t we be willing to take a tiiiiny bit of credit risk to keep e.g. a dentist’s custom? They’re very lucrative.”), etc.

In a pattern we often see in finance, this innovation has occurred in parallel with new user behavior and new networks, most obviously the Zelle consortium in the U.S. and app-driven closed payment networks like Venmo and Cash App.

Zelle was, in large degree, a response by banks to avoid being disintermediated by the payment apps, and now processes more than $300 billion of transactions annually. These are both free to consumers (banks subsidize them to keep their depositors from going elsewhere) and almost instant, including outside of banking hours. Venmo and Cash App both have transaction run rates around $250 billion per year.

Perhaps counterintuitively after quoting twelve digit sums, the impact on consumers of these innovations is larger than it looks. Single-day or single-weekend liquidity mismatches are responsible for large financial and social consequences, particularly for less well-off users, ranging from multiple $35 overdraft fees to getting one’s power or water service turned off. Being able to tap friends and family for funds instantly, or move money between one’s own accounts instantly, helps to avoid this.

We see this in B2B services, too. Many small businesses operate very close to the margin essentially all of the time, and the number one request from businesses of almost all sizes at Stripe is accelerating payout timing. It turns out there are many ways to do this at an acceptable cost for acceptable levels of marginal risk if one is creative, and we continue to be very, very creative. That helps to turn Stripe into an overlay network on the global financial system. We can do internal settlement at Internet speed even between businesses that have a very complicated least-cost route between them through existing financial institutions.

Remittance costs are cratering
A phrase that no one has ever said: “I really enjoy making international wires.”

Nonetheless, this has gotten better over the last few years, especially over heavily-used corridors. (In remittances, “corridor” is a term of art meaning an origination location, a destination location, and a currency or currency pair. For example, “US to Mexico while converting dollars to pesos” is one of the busiest remittance corridors in the world; the opposite direction is a different ballgame for structural and demand/supply reasons. As a result, professionals think of it differently.)

Anecdotally, as someone who has for almost 20 years had to move money between the U.S. and Japan, all-in costs have declined by 90% for small transfers and double digit percent for large ones. Japan to the U.S. is a relatively high-cost corridor (mostly due to user willingness to pay), but Wise (nee Transferwise) has gotten the pricing on a $1,000 transfer down to about $10. The best provider in the market used to charge $40 plus a (conveniently underdisclosed) ~2% spread on the currency conversion.

Interestingly, there are now multiple Japanese banks that actively court foreign depositors (which was almost unthinkable 20 years ago), and they finally do for retail users what sophisticated counterparties have long had available: constantly publish reliably available prices for currencies with tight spreads. One of these banks currently quotes a spread of .3 JPY per dollar, which works out to about 0.25%.

More active corridors like the aforementioned U.S. to Mexico have similarly benefitted from vibrant competition by the financial and non-financial sectors. Walmart is a major player here (through partnerships, a frequent feature of non-financial firms offering financial services), and has fairly competitive pricing, at about 3% at the typical remittance (about $300).

One of the major themes in driving remittance pricing down has been focusing on operational costs, which for remittances are dominated by retail presences at both endpoints of the transaction. Transitioning remittance senders to apps (and encouraging them to do their cash management in either the traditional banking sector or in adjacent ecosystems like pre-paid cards) reduces the labor and retail footprint required on one leg of the transaction. In cases where both legs use apps (or websites, or phone centers), the transaction can asymptotically approach free.

Since remittances are a high-salience user need for some user groups, financial institutions (and other businesses) on both sides of the border can choose to sacrifice revenue on remittances to win more business from those customers. We see this pattern again and again: because financial transactions are high-salience and often high-frequency, non-financial firms would prefer to commoditize their complement and drive the cost of transacting to nearly zero. This was historically difficult for a host of reasons, but technological and business innovations are making it so achievable that many don’t realize how quickly it is happening.

Common themes
Innovation in finances often starts at the “edges” of the traditional financial ecosystem but moves in, sometimes gradually and sometimes rapidly. It is not the sole province of either tech startups or banking incumbents; partnerships between the two are responsible for most interesting product development in the space. The competitive interplay between startups (with attractive new user experiences) and incumbents (with operational scale and distribution advantages) is a core driver of improvements becoming ubiquitous. Another is regulatory or quasi-regulatory action; finance is networks at every scale, and sometimes fiating the existence of a new equilibrium is more efficient than hoping that everyone simultaneously decides to move there together.

Stripe believes that financial services for businesses are also getting better, through measured and responsible innovation happening at financial institutions, software companies, and elsewhere. We partner with leading financial institutions to make capabilities that either didn’t exist or were limited to only the most sophisticated businesses more widely available. This helps put startups on a level playing field with incumbents, increase the pace of innovation for all industries, and improve the lives of everyone who transacts on the Internet.

That's why I work there. If building better financial infrastructure for the Internet to enable the next few decades of financial innovation sounds like an interesting challenge, we're hiring aggressively.



Financial innovation is actually happening

Patrick Mckenzie explores the complex world of credit cards and how they make money. From the intricacies of bundling and unbundling to the various revenue levers, the author breaks down the financial dynamics behind credit cards. Get ready for a deep dive into the world of payments and discover the hidden complexities that shape our everyday transactions. Whether you're a casual credit card user or a finance enthusiast, this essay will challenge your understanding of the financial services industry and leave you with a newfound appreciation for the humble plastic in your wallet.

Payments are deceptively complicated—everyone has used them and thinks they have good intuitions for how they work. However, payments require coordination of a dance between different parties who have extremely different incentives, both on a transaction-by-transaction basis and what they get out of participating at all.

Consider the humble credit card. Swipe it. Tap It. Dip it. HTTP GET it. You have probably used one, mostly oblivious to how it is a complicated bundle of services with a pricing structure strictly more complicated than a venture capital fund’s.

Here is how your bank thinks about it. (A useful jargon word to know: if you have plastic with your bank’s name on it, that makes the bank the issuer of that card. This helps to distinguish it from the other banks involved in a credit card transaction.)

Bundling and unbundling

It’s a truism that there are two ways to make money in financial services: bundling and unbundling. Credit cards aren’t just a bundle, they’re a Mandelbrot set of bundles. The bundles contain bundles. It is bundles all the way down… and all the way up.

One way to think of bundling is as cross-subsidization: you can charge users (or other parties, but we’re getting ahead of ourselves) more for X to give them Y for less than they expect, or even free (or negative!) Credit cards are cross-subsidization engines, both within a particular card as used, within a portfolio of customers using a particular card, and across a financial institution’s customers.

This last bit is very important: sometimes credit cards make money by losing money on the card itself. This is fairly rare, and mostly limited to a small number of issuers at the higher end of their product lines and customer archetypes. You can lose money on e.g. a particular entrepreneur’s personal rewards card to capture their deposit banking business, mortgage, and business banking, and while most financial institutions don’t set out to do this, individual accounts being single-product losers within large portfolios of users is unexceptional and very, very planned.

But let’s talk about the vastly more common case cards are designed around, where a good-fit, non-fraudulent user is expected to pull their own weight revenue-wise.

Revenue levers for credit cards

Net interest. Interchange. Fees. Marketing contributions. There, that’s (just about) every way your card can make money. Taking them in turn:

Net interest

Credit cards facilitate high-frequency minimal-human-involvement extensions of relatively tiny consumer loans bundled (ba dum bum) into an ongoing relationship with parameters negotiated very infrequently relative to individual transactions.

It’s often forgotten, but prior to credit cards, many Main Street retailers like e.g. pharmacies maintained hundreds or thousands of credit accounts for customers individually, necessitating their own back offices, accounting, and collections headache. This was in the ultimate service of getting customers to choose them over competitors, transact in larger sizes, and come back more frequently, the “only three aims of marketing.”

Credit cards represented banks saying: “You know, if you had a specialist doing that for you, it would be much more efficient. They’d have computers doing the math, not bookkeepers. They’d have departments doing collections, not clothing salespeople worried about offending customers who they’d need again at Christmas. They’d have access to cheap deposits to fund loans, rather than expensive working capital. They’d be adequately capitalized against losses, rather than having tiny margins backed by almost no equity, like most retailers. They’d diversify against regional and sectoral risk, rather than being all-in on the plant down the road still being open.”

So credit cards generate loans. A lot of loans. The traditional business of banking makes money on loans by funding them with a mix of cheap deposits and more expensive equity, charging adequately, collecting a spread, and using a portion of that spread to pay for operational costs and defaults.

Credit cards made the business of making loans work much better, by encouraging loans to be automated (rather than bespoke), by encouraging them to be more frequent, and by making them be iterative games rather than one-shots. This, and credit cards’ other revenue streams, let banks relax the “credit box” for loans originated by cards versus comparable unsecured signature loans.

The credit box, a metaphor from consumer underwriting, is conceptually a matrix which maps customer quality, loan amount, and similar to the prices you’d need to justify the underwriting decision or an outright denial, representing “This loan is negative expected value at all conceivable prices and we shouldn’t make it, at all.” Relaxing the box means approving more customers, approving smaller (less lucrative) and larger (riskier) loans, and giving borrowers better pricing.

Interestingly, credit cards make originating loans a distinct business from holding loans. Depending on a bank’s appetite for consumer credit risk and how much in deposits and equity they can back their loan portfolio with, they might not be able to satisfy all customer demand with their own resources. Some banks will package “pools” of those originated loans and sell them on to investors, earning a fee for the ongoing servicing of the loan (they are still the ones receiving payments and on the other end of the telephone, after all) and generally a premium to the pool’s face value (because investors are willing to pay more than $100 for the promise to repay $100 and 12% annual interest over time).

But as mammoth of a business as lending is, lending is not the reason credit cards took over payments in much of the world. Interchange is.

Interchange

When you swipe your card at the local cafe, multiple sales have necessarily happened. One is selling you a coffee. One was selling you a credit card. And one was selling the cafe on the desirability of accepting your credit card brand.

The sale to the cafe went something like this: You’d sell a lot more coffee if you accepted our credit cards. The best coffee drinkers carry our plastic, and they will drink coffee where they can use our plastic. You should pay us a bit for bringing you these desirable coffee drinkers, just like you’d pay for an ad in the paper that brought you desirable coffee drinkers.

That fee is called interchange. (Technically speaking, the industry dices up the fee into a few different parts and has different names for them, but let’s agree to call it interchange for the moment.)

The lion’s share of interchange goes to the card issuer… at least for the moment (oh don’t worry, we’ll get to that). Issuers have this argument for why they should get most of the fee: they do most of the work in the ecosystem. They signed you up for a card. They take the credit risk if you drink your coffee but don’t pay your bills. They will answer a phone call at 3 AM in the morning on the second ring if you have a problem with the card.

A much smaller portion of interchange goes to the credit card processor, to the acquiring bank, and to the credit card network. (Stripe makes a very large portion of our revenue by taking a small portion of the cost of interchange which we charge our business users.)

Interchange makes cards so valuable you’re paid to use them

Interchange ended up with different regional equilibria as credit cards ate parts of the payments pie worldwide.

In the United States, issuers quickly discovered some customer archetypes which absolutely printed money via interchange. A major one was business travelers, who were largely not customers of the credit but only needed the cards for money movement. And they moved a lot of money, mostly between their employers and airlines/hotels.

Competition for the business of business travelers caused one of the most important innovations in both consumer banking and the travel industries ever: cross-subsidization of credit card customer acquisition with travel company loyalty points. This economic engine became so massive that it is now worth strictly more than the airlines themselves, and it sparked a change of practice across U.S. cards: competing aggressively for customers by rebating interchange in the form of either rewards (such as airline loyalty points) or cash back (a post-transaction discount).

This had fascinating implications for the microeconomics of credit cards. For one, competition for desirable credit card users is so intense that profit margins for banks decline midway up the credit score ladder (and for some segments are actually persistently negative), before recovering for the most desirable users (who spend so much that their interchange finally outruns the rewards expense). (See Table 3 in this PDF.)

For another, this ended up being an almost peculiarly American experience. In Europe, regulators were worried about the cost of interchange to businesses (rather than consumers) and capped it. Since issuers didn’t have the margin to compete on rewards paid for by interchange, they instead leaned into branding and convenience, and credit cards became a smaller portion of the payment mix (about 47% of electronic payments, compared to almost 70% in the U.S.).

Curiously, in Japan, interchange is uncapped (and, according to the government, maddeningly opaque) but financial institutions held the line on rewards at 1%. This makes card issuance an extremely profitable business to be in in Japan, so much so that it subsidizes the rest of consumer banking in Japan’s persistent low interest rate environment (which depresses the net interest margin that consumer deposit accounts generally generate most revenue from, historically and in much of the world).

Fees

Fees have gone out of favor in much of consumer-facing finance, but credit cards have historically been rich sources of them. You can broadly categorize them into account fees and usage-based fees, with the former applying to most holders of a particular card (though issuers frequently waive them for marketing/etc purposes) and the latter based on user behavior that the issuer wants to discourage and/or price.

The two dominant types of credit card fees are over-the-limit fees and late payment fees. Both have declined as a percentage of the revenue mix over the last several years, due to consumers having better visibility into their usage (largely via mobile apps and IVR systems) and due to regulatory pressure to compress fee levels. The CARD Act alone probably returned over $10 billion a year to consumers.

Marketing contributions

One insight the industry had is that there is a limit to business’ desire to pay for payment acceptance but a much higher willingness to pay for customer acquisition. As technology has allowed credit card companies tight loops to their customers, they are increasingly attempting to nudge their purchase behavior in provable ways then invoice businesses for a portion of the marginal revenue driven.

A very customer-visible example of this is on Square’s Cash App, which periodically offers “Boosts” which rebate much more than interchange rates for particular purchases. These are sometimes paid for by the issuer as a marketing expense, but more frequently they’re the marketing spend of the boosted business. For example, Cash App (as of this writing) offers 10% off on one use at any grocery store and 5% off up to ten online purchases at Adidas. Without any internal knowledge, the first of these is very likely to be Square subsidizing the customer to motivate future behavior; the second is likely Adidas paying Square to send them more sneakers-hungry consumers (and send less to Nike). This sort of thing makes credit cards into a “channel” where advertisers compete with money, just like they compete for placement on retailer’s shelves or in the weekly insert in a newspaper.

If you are familiar with Bank of America or Chase mobile apps, you may have noticed partner rewards programs that periodically offer you, e.g., 5% cash back for a transaction at Starbucks. These are administered by a publicly traded company called Cardlytics, which charges the likes of Starbucks to drive them business, pays the consumer incentive out of the marketing spend, and also pays the bank for lending them the customer relationship. It is real money; they paid banks more than $100 million in 2020.

It is believed by many that banks make lots of money selling "your data." This is not a significant contributor to the economics of credit cards, for reasons which are slightly too complicated to get into in this piece. The short version: much like Google and Facebook, issuers can demonstrate to the most sophisticated organizations on the planet that they can deterministically influence actual purchasing behavior. That's easier to sell than a CSV file and worth more to more businesses.

That is substantially every way to make money with credit cards. Balancing these against each other is a fascinating exercise in marketing and product design, which will revisit later in this series.

Debit cards: a horse of a different color
Debit cards are a very similar product with enough under-the-hood differences that they deserve their own moment in the sun. In particular, due to a quirk of U.S. interchange regulation, they basically fund most of the fintech industry.

See you next time for it. Later in the series, we’ll discuss how the microeconomics of these products have helped turn payments infrastructure into a platform and ecosystem.

How Credit Cards Make Money

Patrick McKenzie delves into the hidden financial infrastructure of debit cards. While credit cards often steal the spotlight, McKenzie argues that debit cards are a critical piece of the financial puzzle and have unintentionally transformed retail banking. By examining the economics of retail banking and the impact of debit card usage, he reveals how these seemingly simple cards have driven fintech innovation and changed the way we manage our money. 

Credit cards are a bundle of a payments instrument and loan origination. You might think debit cards must be strictly less interesting, given that they lack the loan origination. They’re actually a core piece of financial infrastructure, accidentally upgraded checking accounts, and drive (and fund) much fintech innovation.

Debit cards made retail banking a better business
Much of financial services concerns stocks and flows; money sitting mostly idle or money moving between places. Financial service firms can charge for either of these or both, but typically one predominates.

Prior to the introduction of debit cards, core deposit accounts for retail users were a boring business to be in. (Banks historically consider consumers and small businesses to have strikingly similar banking needs and, largely for organizational reasons, group them together. We’ll call them “retail” users for simplicity; different banks call them “core banking”, “community banking”, and similar internally or in their financial statements.)

In the United States and much of the world, payments are free or near-free for retail bank customers. The defining feature of U.S. deposit accounts, check processing, both lends them their name (“checking accounts”), contributes billions of dollars of operational cost and credit risk, and is assumed to be free for both sides of the check transaction. This was historically subsidized by net interest earned on one’s balances and by using the core deposit relationship, and the frequent stops by the bank branch it used to entail, to sell loan products which had a healthier margin profile.

An interesting wrinkle about net interest is that most customers don’t contribute much. In much of the U.S., median checking account balances at account creation are only around $3,000. On a portfolio level, they increase slowly over time as customers' incomes rise slowly over the course of their career and as they save money.

At a 4% net interest margin, this account would only contribute about $120 in margin during its first year. In the U.S., the industry generally estimates approximately $350 a year in costs to maintain a checking account, of which approximately $120 is direct marginal cost (as opposed to e.g. a pro-rated portion of the cost for the branch footprint).

So most checking accounts (and particularly, most checking accounts where the customer had no other products from the bank) were negative contribution margin. That sounds better in an annual report than "We lose money on most customers even when everyone does everything right."

They were subsidized by banks and, indirectly, by customers of banks with larger balances, who were mostly unwittingly paying a higher spread between their own deposits and their loan pricing. This was a considered policy choice, because (as previously covered) the government prefers almost all members of society to be banked, but it was not an easy policy to implement around the edges. In particular, neighborhoods which had low concentrations of retirees and small business owners were difficult to service profitably on the branch model. Regulators and legislators attempted to put a thumb on this scale for decades, with mixed success.

Enter debit cards, which like many technologies were available (1960s) before being common (1980s) and then becoming ubiquitous (1990s). Debit cards  fundamentally transformed the economics of retail banking.

Consider again the median bank user, who might have a pre-tax income of around $60,000, post-tax post-transfers cashflow of $3.5k a month, and rent of approximately $1,000. (These numbers likely sound low to many readers; remember that the median American is not a professional employee in a coastal city.)

This user might very well have somewhere on the order of $2,000 of debit card transactions a month. This implies about $300 a year in interchange revenue, substantially more than their contribution from net interest.

Debit cards were a major structural factor in the retail banking boom of the 1990s, because neighborhoods which were once marginal prospects for new branches were suddenly both a) directly margin contributive immediately while b) having several embedded options on neighborhood economic growth, aging of the customer base, and cross-sells within the customer base. Neighborhoods that had handily merited a bank branch for the same reason now saw increased competition, from multiple banks first and eventually from multiple branches per bank, as each bank worried that a competitor could steal their debit cards by being one block more convenient.

This isn’t just an American story. Here in Japan, the persistently low interest rate environment for the last 30 years has made retail banking a very bad business to be in. However, over the last 15 or so years, the Japanese financial industry and government have made a concerted push to move the nation from being cash-based to using electronic payments, which for most typical consumer purchases means credit cards. The government’s current target is 40% of the payments mix, about double the current amount, which would put it on parity with the present state of play in the U.S. (and behind Asian peer nations, at around 50%).

Debit cards as an infrastructure upgrade
In addition to revolutionizing the economics of retail banking, in the U.S., debit cards quietly (and mostly accidentally) introduced instant business-to-consumer payments.

Many European readers are amazed that I keep harping on this, but “money moves much slower than data” has been a core feature of the U.S. experience for my entire life. The reasons behind that could fill volumes, but for now please just accept that we run the largest economy in the world on top of financial infrastructure that truly does lack a core primitive you’d expect it to have.

Anyhow, a bundled feature of credit cards and debit cards alike is that businesses can refund transactions. This is broadly considered a boring implementation deal. And they can even over-refund transactions in some circumstances. This was largely planned, to the extent it was planned, as an operational convenience. A customer might buy $150 of dresses in two transactions, return them to the store, and receive a single $150 refund rather than having the clerk hunt down both original transaction IDs to refund.

Eventually, someone realized that over-refunds plus debit cards accidentally create a novel payments network. You could just capture a debit card and then refund money never actually spent on it to transfer the customer money. That money would arrive in their bank account substantially instantly, rather than in the 3-5 business days that was more typical in the U.S. until recently. This required no negotiation with their bank, travelled over nearly universally accepted existing infrastructure, and involved no mutli-party stakeholder negotiation over whether a migration would disadvantage any economically significant part of the financial ecosystem.

This feature quietly lurked in the financial industry for many years but only became widely adopted in the last decade, first by the gig economy and then by wallet apps.

Typically, the apps offer their users a choice: payouts at the speed of banking, for free, or payouts at the speed of the Internet, for a small convenience fee. Cash App charges 1.5% with a minimum of a quarter. Lyft charges a flat fee of 50 cents.

This is an extremely high margin service and it is very, very, very popular with users. Many report that it transforms the nature of their interaction with the underlying application; delivery driving for casual drivers becomes something that you can burst up to fill an immediate cash need or mentally allocate against a particular desired expenditure (e.g. “Drive for 3 hours to afford a night out starting immediately after you log off.”)

The actual physical implementation of this once used niche financial players but these days is offered directly from Visa and Mastercard. Stripe offers it as a product called Instant Payouts.

Interestingly, pricing instant payouts serves an important packaging goal for fintech applications: the actual thing that the user wants isn’t money in their bank account faster. It is to be able to meet an obligation at a known time in the immediate future. Charging a convenience fee for instant payouts allows fintechs, and businesses with embedded financial infrastructure like gig economy platforms, to position their own debit cards as a free alternative with the same instantaneous funds availability.

This allows the fintech to shift the cost of instant payments from their own users to the businesses their users transact with. In particular, in the U.S. at present, they’ll almost certainly do this by having a debit card issued by a Durbin exempt financial institution.

Durbin-exempt interchange
We covered this briefly in the issue on community banking, but debit card interchange in the U.S. is presently capped to 21 cents plus 0.05% of the transaction. This is much, much lower than credit card interchange.

This was passed as part of the sweeping Dodd-Frank legislation in the wake of the financial crisis. If you imagine society as being in a perpetual dialogue with the financial sector, you can conceptualize this as a demand: in return for partially paying for your bailouts, commercial users require you to not charge us nearly as much for payment services. Find another way to subsidize your retail bank users; it’s not our problem.

Community banks had this to say about that: we didn’t cause the global financial crisis. We weren’t there on Wall Street creating exotic derivatives on asset backed securities. (Community banks might, ahem, hope that anyone who understands the mortgage supply chain stayed had too many things to worry about to protest this point.) We remade our retail franchises around the expectation of debit card interchange. If you cap the price on it, you’ll kill our institutions, cause millions of Americans to lose access to branch banking, and further roil the commercial real estate market, which we fund in most of the nation.

Senator Dick Durbin heard arguments like this, substantially agreed with them, and proposed an amendment to Dodd-Frank exempting banks with less than $10 billion in deposits from the interchange cap. It passed, and the fintech industry (which barely existed at the time) accidentally inherited a business model.

The fintech industry calls institutions under that cap Durbin-exempt issuers. Most of them continue issuing debit cards to their local communities, as they’ve done for decades. A handful of them partner with other firms to provide debit cards (and other banking services) to those firms’ customers.

This was an unintended structural alliance, but is extremely favorable for businesses engaging in it. Fintech companies offer relatively deep integrations with regulated financial services without themselves having to become banks, by piggy-backing on underlying banks’ capabilities to offer those services. The banks handle the compliance, custody, risk management, and a portion of the customer service, as they always have, but the fintechs provide a scaled national go-to-market strategy that no community bank could independently develop.

The most visible beneficiary of this has been the neobanks, which in the U.S. at least are almost invariably software companies that have a mobile app which integrates tightly with a debit card provided by a partner bank. That is (by far!) the easiest pathway to launch a banking product in the United States; it takes literally years, tens of millions of dollars, and substantial execution risk off of go-to-market. Most neobanks will attempt to figure out a scalable approach to attracting and retaining customers then try to increase contribution margin through other products. (It is an interesting open question whether neobanks in the U.S. will ever outgrow their banking partners, either for operational reasons or because society demands it of them.)

Many infrastructure providers, Stripe included, further abstract financial services (like debit card issuing) into suites which they offer to platforms building up on them. This “embedded fintech” lets companies whose core competence is software development for a particular vertical industry deeply integrate financial services into their core platform without actually providing it themselves.

Innovation is happening apace in this space, but as of today, the main monetization engine for this sort of relationship is the Durbin-exempt interchange on debit cards. The software platform, fintech platform, and bank split on the order of a hundred and change basis points (1.X%) of each transaction of their customers.

The value proposition for customers is an interesting one. Its a faster way to get access to one’s money. But the money is also… better? Because it is enhanced by the software provided by the platform, which can mirror a ledger of it (like any financial institution could) but use intimate knowledge of the customer’s business to make their software offering categorically better given that it is aware of transaction-level data about how money flows.

This lets platforms do things like e.g. automated tax reporting, bookkeeping, business analytics, etc, on top of their core services, without needing to directly charge their own users for this. The payments revenue is (from the perspective of the platform) extremely high margin.

SaaS companies offering "money, but better" thus act as an acquisition channel for relatively small banks which would otherwise have their own software offerings comfortably crushed by the largest banks in the world, which in the past few years have actually started to introduce very good mobile applications. Those banks have not, however, started innovating with respect to particular business or customer verticals; their retail offerings are (sometimes painfully) one-size-fits-all.

These services tend to make sticky products even stickier. Businesses can churn off a SaaS product, but once the SaaS product is intimately integrated into the flow of operations of their business, churn rates decrease markedly. It is difficult to get more intimately integrated with a business than to be directly in their cash flow. Since SaaS companies' enterprise values essentially move inversely with churn rates this is extremely powerful for them, even on top of the direct revenue contribution.

And thus a measure intended to rap Wall Street’s knuckles for the financial crisis has inadvertently caused rapid innovation in the main financial surfaces that many consumers and, particularly, small businesses interact with.

Infrastructure: it’s a fascinating world.


Further reading
Many readers have asked where I’d suggest starting to get a better handle on these topics. Payment Systems in the U.S. is my go-to recommendation for a quick survey, although I don’t love the book. The American Banker magazine has many interesting factoids about e.g. the economics of checking accounts in its back archives.

The Federal Reserve system (and its counterpart in other nations) publish very detailed reports on these subjects, and while their discoverability can frequently be poor the comprehensiveness is wonderful.

Finally, if you have dozens of hours to invest in the topic, getting familiar with the quarterly and annual reports of a small number of banks will, over time, really help to develop intuitions about both banking and the payments industry.

If you’ve not previously worked professionally on this, try First Republic’s reports; they’re large enough to have good production value on their reports but they have a very simple business to understand relative to e.g. the largest banks in the world. Wintrust and Silvergate are also good for similar reasons. After you feel like you largely understand what a typical bank’s reports look like, try Green Dot Bank or Evolve Bank and Trust to see how interaction with fintechs changes the equation.

If you find this sort of thing interesting, Stripe is hiring aggressively. As always, my views on these subjects are my own, but I like to think that I've learned a bit on this industry in my five years here.

Debit cards are hidden financial infrastructure

Patrick McKenzie takes us on a journey through the changing landscape of payments in Japan. Exploring the differences in credit card usage, McKenzie highlights the unique culture surrounding money and the obstacles faced by Japanese consumers when it comes to accessing credit. With his personal and professional insights, he gives us a glimpse into the future of payments in the country and hints at the potential impact these changes may have globally. Whether you're a payment geek or simply curious about the evolving world of finance, this article offers a fascinating perspective on the Japanese payments space.

Payments in Japan

Patrick McKenzie delves into the world of bank transfers as a payment method, shedding light on why they haven't become the norm for most transactions. He examines the core features of payment methods, including total cost, customer user experience, certainty, settlement time, and reversibility. McKenzie explores the substantial innovation happening in bank transfers, particularly outside the U.S., that will likely lead to them becoming a larger portion of the payment mix in the coming years. This thought-provoking article uncovers the intricacies behind payment systems and offers insight into the future of banking and transactions.

Bank transfers as a payment method

In a fascinating exploration, Patrick McKenzie unveils the complex and surprisingly organized world of credit card fraud. Drawing on his professional experience in the field, McKenzie sheds light on the diverse ecosystem created by fraudsters, complete with HR departments, certifications, and even headhunters. Delving into the realm of payments fraud, he uncovers the various methods employed by professionals to misappropriate money from the legitimate economy. This eye-opening article provides a unique perspective on an industry that operates with the precision and sophistication often associated with legitimate businesses. Prepare to have your conception of crime and fraud turned on its head.

I recently wrote a tweetstorm that touched on the supply chain for credit card fraud, and many people said they were surprised by it, so I thought I’d delve into that topic in a little more depth.

The business of fraud is almost as diverse as the business of... business. And it really is a business. Organized fraud has many of the accoutrements of professionalized systems with differentiated labor, like finance itself has. There are HR departments, professional fora, certifications, headhunters, and the like.

This is a surprising fact about the world, and not at all obvious. The mental model most have for crime is that it is largely committed by impulsive, not-terribly-well-organized individuals or small groups. That happens in finance, too, but the bulk of the work is done by professionals.

Different classifications of payments fraud
One could write several books about fraud without scratching the surface of the topic, and unfortunately the margins of this newsletter contain too little space for them. (The best I have ever read, by far, is Lying about Money by Dan Davies.) So I’ll take the liberty of focusing on payments fraud, which is near and dear to my heart.

Obligatory disclaimer: I work professionally adjacent to this for the last few years at Stripe, and got my start in fraud at a previous employer doing anti-spam research (where the adversary has a truly surprising amount of overlap with payments fraud). The following represents only my own views.

Payments fraud misappropriates money from someone in the legitimate economy by convincing an actor in the legitimate economy to transfer it to someone it does not belong to. There are a million varieties of it; it happens on every conceivable payments rail. Almost anyone in the supply chain of legitimate payments could be targeted by it.

In the industry, we sometimes find it clarifying to bucket fraud by a matrix of intent on the part of the current holder of payment credentials and the intent of the business those credentials are being used at. The ill-intent/ill-intent quadrant is the one most people think of when they think of e.g. credit card fraud, but fraud occurs in the other quadrants ill-intent quadrants, too. Still, this is the easiest to think about: someone steals e.g. a credit card number and wants to turn it into money. How does that happen?

Stealing credentials wholesale
Some credit card numbers are stolen by individuals. One could pickpocket a wallet, memorize the number of a patron at the restaurant one works at, or sweet talk someone over the telephone into giving it to you by misrepresenting your identity.

But this is not the dominant way cards (or other payment credentials) are stolen.

Amateurs talk strategy or tactics. Professionals talk logistics. Most misappropriation of payment credentials happens at an industrial scale. The industry is often referred to as “carding”, though every method to transfer value is vulnerable to something in this vein, not just credit cards.

Sometimes this happens via security research against firms that see lots of credentials legitimately. A famous example was the Target hack, where approximately 70 million credentials were compromised by malicious software running on legitimate point-of-sale hardware that routinely interacted with physical cards. The software was injected by hacking the network using credentials misappropriated from an HVAC vendor, which had itself ultimately been compromised because someone clicked on a link in email that installed malware on their machine.

There are multiple overlapping ecosystems of evil implicated here, and often they form supply chains, where one firm’s outputs are another’s inputs. For example, there is specialization of skill implied in sending malicious email at scale, in compromising the machines of a wide variety of small businesses, in exploiting those machines to e.g. become cloud infrastructure for bad actors or purloin credentials from them, in conducting offensive network security research, in developing malicious point-of-sale software, and in operating said software to steal the cards.

That’s all theoretically within the reach of a single dedicated individual, in the same way that building a software company is within the reach of a single dedicated individual, but most software is not written by individuals acting alone and in isolation. Trust this multi-time solo founder: it’s a heck of a lot of work for one person.

There are other ways to steal cards. You could, for example, send spam email to hundreds of millions of people offering desirable goods at unbelievable (or believable) prices, send them to fake e-commerce sites you control hosted on cloud infrastructure, and get their credit card numbers that way.

Notice that this re-uses a few intermediate outputs from the above example. It’s inefficient for every firm to implement their toolchain from scratch every time they try to do something, which is fundamentally why the economy has firms specialize and buy tooling and services from each other.

Lists of known email accounts, themselves often put together by specialist market research firms (but evil) consuming the work of data scientists (but evil) consuming the work of web application pentesters (but evil), are definitely a product that can be bought. “Ratware” to automate the sending of email such that it ends up in inboxes, like a mail services provider (but evil), is a thing you can buy off-the-shelf or commission from boutique consulting shops (but evil).

This economic activity is coordinated in ways not dissimilar to legitimate commerce. Much of it happens over industry forums (but evil), but there are also marketplace sites (but evil) which have customer service teams (but evil) to administer reputation systems (but evil) so that unsuspecting customers (but evil) don’t get frauded by fraudsters while frauding the fraud they fraud professionally.

And that’s just to get the cards!

Turning purloined credentials into money
Suppose you have a small amount of text which doesn’t belong to you. You can’t eat text, pay your employees in text, live in text, drive your children to school in text, etc. You want to turn that text into money.

You will turn to the services of a “casher”, or in the alternative you will put your valuable text (payment credentials) for sale such that cashers can purchase it, for example at the above-mentioned fraud marketplaces.

The term Dark Web gets thrown around quite a bit at this point, because it is very evocative, but a lot of this actually happens on the same networks that everything else happens on rather than e.g. on sites which are accessible only over Tor. (Tor is open source security technology developed by the U.S. government originally. Not everything in the supply chain of Evil Inc. is intrinsically bad! Most of it is browsers, programming languages, operating systems, etc etc etc, that are the same as the ones you use.)

Why have the operational split between carders and cashers? The reasons are similar for specialization within firms and professionals engaged in carding, with an additional wrinkle, which is risk. Just like the legitimate finance industry has extremely developed systems to price risk and transfer it to people who want certain slices of the exposure in exchange for certain return profiles, Evil Inc. has relatively safer professional specializations and relatively riskier professional specializations, and having them occur by different people, perhaps in different firms, perhaps in different countries both decreases the total amount of risk in the system and helps allocate it efficiently.

In traditional finance, most risk is denominated in money. In fraud, risk is sometimes denominated in risk of law enforcement action against you and sometimes denominated in adversarial action by fellow criminals. And sure, money, too.

There are as many ways to be a casher as there are to be a carder, and (much like one could be a carder just by stealing a wallet or memorizing a number) one could get in on the ground floor of cashing by acquiring a single credential. You could, perhaps, take that one credential to any e-commerce store you like, buy a thing, have it shipped to you, and then re-sell it for money or enjoy it for your personal use. Casual fraud is very much a thing, and (speaking generally) the targets of it are often the same sort of businesses that struggle with casual shoplifting. (Shoplifting is also sometimes an organized business, as many American cities are currently realizing to their discontent, but that is another newsletter entirely.)

When the legitimate holder of the card finds out that this happened, they’re likely going to call their bank to complain. The bank will reverse the transaction, and the e-commerce store will have to return the funds. They will also pay a fee to the financial system for mediating the dispute, which is designed to encourage companies to contribute their time and talent to zealously protecting the ecosystem.

But back to the bad guys. Notice that taking physical receipt of goods involves some amount of risk, in that it leaves records in the legitimate economy about a physical location which may be tied closely to you personally. If you do this once, you’re abundantly unlikely to come to the attention of law enforcement. But if you make a habit of it, you’re… still, unfortunately, unlikely to come to the attention of law enforcement, but quite likely to come to the attention of fraud departments, which will attempt to shut you down.

So cashers will do other things, both to achieve operational scale and to reduce their risk profile. One example is triangular e-commerce fraud, which (like much fraud) is so brilliant that if it weren’t absurdly evil you’d almost have to admire the folks who thought it up.

The basic mechanic is: open an account on a legitimate e-commerce platform. You operate like a real business, selling valuable things to people for money. You acquire customers by running ads or sending email or competing on SEO or pricing aggressively or what have you.

Then, when a customer orders something from you, you fulfill it at need by going to another e-commerce company (ideally, not on the platform that you transact on), signing up as a new customer, and ordering the thing your customer bought with delivery to your customer. You pay with a stolen credit card.

This is really hard for the e-commerce platform to detect, because you look an awful lot like a legitimate business! You will have many satisfied customers who legitimately paid you money and legitimately received exactly the product they expected to receive! Very few will notice who sent it, and if they do, they are probably the sort of weirdos who care about e-commerce business models and know that dropshipping is a thing that legitimate firms also do.

A funny aside: I first learned the word dropshipping in high school, at my first job, working for a publicly traded U.S. office supplies retailer in their order entry department. During the interview, the rep I was speaking started to explain how to input a dropshipped order in their system, realized she had used jargon, and started to explain it.

I said “I understand the term; that’s when you buy something after the customer orders it and ship it directly from your supplier rather than from your own warehouse. Your free shipping offer doesn’t apply and the customer needs to have an elevator up to the floor the delivery happens on or it will be an additional $75.”

She asked how I, a high school student, knew that.

I said “It’s in your catalog, in the fine print after the order form.” She asked why I had a copy of an office supplies catalog, I replied “I made it my business to know your business.”, and that, reader, is the smartest thing that I will ever do in a job interview.

Other variants on cashing involve money laundering in the legitimate financial ecosystem, often using purloined credentials to buy things-that-aren’t-quite-money then reselling those things for actual money. For example, gift cards are very money-adjacent, and there are thriving businesses that buy and sell them online. Much use of them is perfectly legitimate. Carders could either use purloined credentials to purchase gift cards from an original issuer then resell them for value or use purloined credentials to purchase gift cards from a gift card marketplace. (Entrepreneurs who run gift card marketplaces either go out of business or become the payment industry’s most ruthlessly effective fraud busters.)

Turning deposits into value close to you
Suppose you have an account at a legitimate e-commerce platform, gift card provider, or similar. You want to accept payouts from it, but may not physically live where they support users or you may just not want to connect your (extremely surveilled) banking information with the nexus of your criminal activity.

You might hire the services of a so-called “money mule.” That person will use an identity, very often their own or that of a person in their family, to receive your funds. They will then move them around in the financial ecosystem the same way they generally move money in the financial ecosystem, send most to you, and keep something for their efforts.

Mules are the mugs of Evil Inc. Many of them don’t even realize they’re participating in a fraud. They’re recruited by advertisements pitching work-from-home opportunities in e.g. “being an accounts receivable clerk.” The basic job description is “We periodically email you descriptions of incoming payments to your bank account. You will forward those to us via a mechanism we specify, for example taking out cash and then sending it via a remittance service to our office abroad. Keep 10% as a commission.”

Money mules are very, very likely to come to the attention of fraud departments and (frequently) the authorities, but this is a very, very appealing pitch, and people fall for it (with varying levels of understanding that it is not legitimate) all the time. They’re difficult to exclude at scale from the financial ecosystem because they’re legitimate people with legitimate e.g. checking accounts. They pass KYC screens with ease; they legitimately are who they say they are and have the government-issued documentation to prove it.

An especially vexatious variant of money mule, from the perspective of the financial industry, is businesses operating as mules. Businesses have legitimate need for moving money around, sometimes in large quantities, at high velocity, in ways which look a little weird if you don't work in them but are probably legitimate. They also get extremely vexed if banks routinely ask them stupid questions about their internal operations. This tends to create a controls environment where business accounts are less surveilled than consumer accounts, though quantifying that is difficult.

My wife once remarked to me, as I was wrapping up one of my startups in Japan, that she had seen an ad for people buying closing companies to “recycle” them and that I could get ~$1,000. “Honey, that’s a crime and the person buying the business is a criminal.” “What, no. They just don’t want to go through all the rigamarole that you had to to open a company and then get a bank account for it.” “No no no, trust me on this, anyone with this pitch is overwhelmingly likely to be a criminal and when the police come looking for the stolen money the person whose name will be on the documents would be me.”

This doesn’t necessarily even require the business to fail and be sold. Sometimes the businesses are in on the scam, sometimes even simultaneously with using the opened accounts for their legitimate business. Sometimes they’re using the accounts simultaneously for legitimate purposes and don’t understand why the thing proposed by their new business partner / client / investor is in any way illegitimate.

Fraud is an fascinating deep rabbit hole
If I hadn’t fallen backwards into entrepreneurship I think it is highly likely I would have ended up working in fraud full time (on the side of the good guys, to be clear). It is fractally interesting and involves a cat-and-mouse game between very, very talented people.

The financial industry invests a staggering amount of effort into defeating fraud, but that isn’t the only priority. The right amount of fraud is not zero. We as a society care, for example, that consumers, even low-sophistication consumers, should be able to walk into any financial institution and walk out with a bank account. A lot of interventions against e.g. money mules would trade off with that societal goal.

If you’re interested in this subject, particularly as a technologist, Krebs on Security has a lot of great reads.

The fraud supply chain

Patrick McKenzie dives into the intriguing world of gift cards, analyzing their role in the payments industry, the different types available, and the often misunderstood psychology behind their use. Discover how gift cards go beyond just a simple substitute for cash, offering personalization and storytelling that enhance the gift-giving experience. Uncover the complexities of these seemingly straightforward financial instruments and explore their potential in facilitating transfers and functioning as a valuable form of currency. Prepare to have your perceptions challenged and your understanding of gift cards expanded in this captivating essay.

There are few things comedians and personal finance writers agree on, but one comes up every holiday season: “Gift cards. For when you want to give someone money, except worse.” Like many topics in financial infrastructure, they’re a fascinating Gordian knot of user needs, business incentives, government regulation, and infrastructural weirdness. Let’s start unraveling it.

How the payments industry thinks of gift cards
The industry taxonomy for gift cards (and gift certificates, and pre-paid cards, and similar instruments) primarily splits them into open-loop and closed-loop systems.

Closed-loop gift cards are the classic retailer gift card, where someone buys a known value of future goods/services, typically paying 1:1 in cash, and those goods/services only come from the business which issued the card.

Open-loop gift cards are issued under the aegis of a payments network, like Visa or American Express, and are approximately as widely accepted as other cards carrying the same brand, not just with the company which directly sold them to the customer. (As always, there’s a fascinating level of detail hiding behind the word “approximately.”)

There is more product differentiation in open-loop gift cards than in closed-loop cards. Some are sold primarily for the traditional gift card use case, to enable non-cash transfers of funds in a convenient physical or electronic format between people who know each other. Others (like e.g. Amex Serve) are sold as financial facilitators directly to the end user; they effectively function as simplified bank accounts.

There are also some much quirkier options, including one which deserves an issue for itself as some point: “Medicare pays doctors using pre-paid gift cards” is certainly the most surprising sentence I’ve read recently, and being unsurprised by wildly off-label uses of payments infrastructure is my job.

The job to be done, or, users are less irrational than believed
Clayton Christensen, a Harvard Business School professor sadly no longer with us, popularized a framework for thinking about products called “jobs to be done”, which helps to conceptualize rational user behavior that looks surprising.

One example he frequently used was that McDonalds sells most of its milkshakes not to children as a dessert but to adults early in the morning. The job to be done of the milkshake was not, as it turned out, delivering the taste of chocolate. Milkshakes were an engaging and enjoyable meal substitute for people who had not had breakfast and needed to consume calories while driving to work.

Personal financial columnists and comedians do not understand the jobs that gift cards do. They are not cash-but-worse-in-every-way. Closed loop cards allow a giver to personalize a gift to the recipient. A Barnes and Noble gift card to a favorite niece tells that niece that you remember she is a reader, unlike those uncreative family members who would just give her cash.

If anyone’s psychology is being manipulated, it is likely the recipient, on a theory very similar to the one that causes many readers’ employers to offer a gym stipend instead of the equivalent amount of salary. You appreciate the money more, not less, because of the story about your relationship it comes with.

In a running theme for Bits about Money, gift cards are misunderstood because they are a financial product slightly outside of the financial mainstream. Most middle class users treat them as primarily a payments instrument, but they are also used to effect transfers.

In this regard it is not merely important that they look attractive in a birthday card but also that they’re available for cash everywhere, require no identification or ongoing banking relationship to purchase, do not charge a fee like e.g. Western Union, and can be conveyed over a text message or phone call. They're not worse cash, they're better Tide in the informal economy.

This makes them a fraud risk (and we’ll get to that later). Ransomware operators and fraudsters were using gift cards to move victim’s money from them through the financial system while evading controls long before cryptocurrency existed. But it also makes them extremely useful to some consumers at the economic margins.

My bank gives me free coffee and a smile to get me to come in more often; some consumers’ coffee shop gives them free banking and a smile for the same purpose. I can get coffee in many places; many of them cannot get banking elsewhere.

Open-loop cards are even more obviously designed, marketed, and described to regulators as low cost (... not really) bank account substitutes for people we’d prefer to be allowed to engage in commerce but who, for a variety of reasons, do not have bank accounts.

How closed-loop cards work under the hood
Back in the day, many retailers ran their own gift certificate programs, since the certificates were bearer instruments, relatively simple to print for a retailer, and relatively operationally easy to bookkeep, both at the corporate and POS levels.

Gift certificates got substantially more complicated as consumer-grade printers became capable of producing effectively indistinguishable forgeries, forcing retailers to verify them at the POS level, which implies an increasingly complicated supply chain to track each certificate to a database entry and verify it at will.

Gift cards exploded during roughly the same period; in addition to solving this issue by having the database more naturally integrated, they offered a superior user experience (you could re-use them) and a superior experience for the business (you never had to make change with actual currency or take the customer experience hit implied by refusing to do that).

But if retailers had issues running paper certificate supply chains, running card networks was even more obviously a non-starter, and almost all of them ended up outsourcing the manufacture, bookkeeping, payments stack, POS integrations, fraud monitoring, and compliance issues to dedicated gift card processors. In the U.S., one of the major ones is run by Fiserv, an absolutely mammoth infrastructure company that few people have heard of. (It is a miniscule portion of their totalbusiness.) There are many others and, unlike the card payment networks, they are largely not consolidated into regional or worldwide entities.

Closed loop card providers are in the business of moving bits, not in the business of holding money. Generally speaking, for a closed loop card which is sold directly from its own stores, the business hopes the money never goes out the door again. Things are slightly more complicated for e.g. buying a Chipotle gift card from the rack at Walmart, but those funds quickly move from Walmart to Chipotle rather than resting at Chipotle’scard provider.

Why businesses love gift cards
Geeks in every field love describing the world through the lens of their own field, and it’s very tempting for a payments geek to wax laconic about the financial underpinnings of gift cards, but retailers will describe, at substantial length, why they have gift card programs, and most of the reasons are not the financial stack.

They’re good at achieving the only three goals of marketing: they bring in new users, they bring users back, and they increase basket sizes. That would, by itself, justify the huge investment in gift card programs. The following factors are more of icing on the cake, for all but the largest programs.

“Float”

Many people are fans of Warren Buffet and describe any arrangement where a user pays for services in advance of receiving them as generating “float” which the business can invest, hopefully profitably.

This is not a terrible mental model for gift card accounting, but it breaks down in the details, and while Warren Buffet spins a very appealing yarn about insurance industry accounting, trying to understand gift cards through that lens is a bit like learning long division in reference to floating point numbers.

When a company sells a gift card, they book a liability to the user (they’ve made a promise to provide something of value in the future) and an asset (cash). This is revenue neutral. When a transaction happens in the future, the liability goes down (the promise having been fulfilled) and revenue is recognized.

In between the present and the future, the company has cash. In insurance, by regulation and practice this would be invested in a portfolio of largely low-risk highly-liquid assets, and the insurer would keep the returns generated by that portfolio, allowing it to be profitable potentially even if it had eventual claims higher than the premiums it took in. This is the core engine of (part of!) the insurance industry.

Most retailers do not have sufficient gift card liabilities and do not have sufficiently sophisticated treasury teams to make a meaningful amount of income off an investment portfolio. Also, they have a much better option. Unlike insurers, who must maintain reserves by regulation, most businesses issuing closed loop gift cards are not required to maintain reserves. They can simply use the cash as a source of working capital for the business, similar to other sources of working capital, except substantially free. Even in a capital-drenched economy, it is tough to beat free.

(Though, who knows, perhaps negative interest rates will get retailers to start issuing reverse gift cards. “Here’s some money if you promise to buy a coffee later! Please, please, take it off our books!”)

Breakage

What happens if gift cards aren’t redeemed, or are only partially redeemed? The remainder is deemed as “breakage” after a few years. The business gets to recognize it as revenue and keep it.

Historically, there was a heuristic that businesses were allowed to use to determine that the possibility of redemption was remote, but these days computers are better and accounting standards require that retailers or their partners use historical norms of actual redemption behavior to age card balances and recognize breakage.

Breakage is free revenue… except in those jurisdictions where it isn’t.

Many jurisdictions have complex rules governing the treatment of lost property. Most gift cards worldwide are issued in the United States, and many American states consider an unused gift card a special case of unused property. They require the card issuer to escheat (remit to the state) funds which have been deemed abandoned, so that the state can hold them in safekeeping for the original owner or their heirs.

That’s the theory. In practice, because gift cards are often sold anonymously for cash, very few gift cards escheated will ever be returned to their original owners or the heirs. The rules for accessing escheated assets are also complicated and have a high frictional cost, often requiring e.g. multiple documents to be notarized, which tends to make them non-viable at typical gift card sizes.

(Surprising no one who knows me, I was so fascinated by escheatment when I learned about it that I had to try. I found the great state of Missouri had been helpfully sitting on $12 overpayment of a utility bill for almost 20 years after university, and spent about 6 months and $75 in notary fees and certified mailings prior to getting the $12 back. Thanks, Missouri.)

The broader practice of escheatment is less odious than it is for gift cards specifically; most usage by value is for actual financial accounts whose owners may have e.g. passed away without informing their heirs about the accounts. There is actually some societal value in having a department of people tasked with maintaining records for generations (in many states, they’re charged with doing it forever), answering letters, and disbursing funds.

Breakage is much lower than non-professionals believe. In the U.S., the industry estimates that approximately 2-4% of the ~$100 billion in annual sold cards will break. This is not a dirty secret or cynical PR play; the numbers are maintained extremely accurately by computer systems. You can read them in the annual report of your favorite retailer, where they’re almost always far below the level of materiality.

One exception, because of its mammoth size: Starbucks. Japanese Internet companies often buy banks intentionally. Starbucks made a mid-sized bank mostly by accident. In FY2020 they had $1.5 billion in “deposits” and $145 million in breakage revenue. This is higher breakage than typical due to small average card sizes (less incentive for users to work to retain partially spent cards), small ticket sizes, and relatively anomalously successful use of the cards as a corporate gift method in e.g. swag bags, employee traveling, and similar.

The joys of gift card compliance
Escheatment is the primary compliance headache for closed loop gift cards, and neither your local pizzeria nor Target want anything to do with it, so it gets outsourced to the infrastructure providers (and then to the government).

Gift cards are also a money laundering risk, and they’re relatively under-regulated given the degree of it. They’re cash equivalents. The features which make them attractive to legitimate users at the fringes of the financial system also make them attractive to bad actors at the fringes of the financial system.

There was a fascinating article which described one Eric as a modern day currency broker, swapping Bitcoins for gift cards and back again, earning a spread on transactions. Currency brokers are paid for offering liquidity and taking inventory risk. Eric was paid for neither. He was part of the fraud supply chain.

The Bitcoin part is almost entirely a red herring. In one extremely literal example of regulatory arbitrage, some states require closed loop gift cards to be redeemable for cash. The supply chain constantly recalculates ways to turn misappropriated value into physical gift cards located in these states, then sends a casher out to the store to collect clean cash. There was once a Reddit comment from an extremely frustrated retail clerk wondering why people inthe neighborhood made him spend so much time counting out refunds of $500 gift cards.

Much more on this later. Until then, enjoy the winding down of the year, and hopefully financial infrastructure brings a tiny bit of joy this holiday season.

More than you want to know about gift cards

Patrick McKenzie delves into the fascinating world of secondary markets for gift cards. He explores the nature of gift cards as privately issued money and the similarities to the free banking era. McKenzie uncovers the inner workings of gift card marketplaces, where you can buy or sell gift cards at a discount, and explains the unique features of closed loop gift cards that enable this business model. Discover the intricacies of this largely unknown economy and how it professionalizes the demand and supply of gift cards.

Gift cards (covered previously in this space) are a form of privately issued money. The U.S. has prior experience with widespread circulation of privately issued money during the free banking era, where “wildcat banks” issued bank notes backed by deposits of precious metals and the general creditworthiness of the bank.

These bank notes were not equal to each other; they traded at discounts to their face value which floated with sentiment about the issuing bank. Newspapers published daily reports of what going market rates were, so businesses could accept them with less risk, and there was a specialized economy of moneychangers in areas where many competing bank notes circulated.

My colleague Sebastian Bensusan at Stripe had a chat with me about gift card marketplaces, a field in which he has substantial professional experience. I was previously aware they existed, but had no idea how deep the rabbit hole went.

Gift cards are, gloriously, privately issued bank notes backed by the full faith and credit of Chipotle.

A sketch of a gift card marketplace
If you’ve never encountered one of these businesses, there are many of them a Google away from you. (I won’t name any in this piece, for social reasons.) They will sell you gift cards at a discount to face value, or buy gift cards from you at a larger discount to face value. They make money on the spread between these two prices.

While they are described as marketplaces, in traditional finance terms they aren’t exchanges but rather are market makers. In the large majority of cases, you are not trading with another person facilitated by the marketplace. You are trading with the marketplace itself, which will trade with another person days before or after trading with you.

There are two really important features of closed loop gift cards which, especially in the United States, uniquely enable this business model.

The first is that gift cards, by regulation, have extremely long expiry periods, cannot be expired, or must be exchangeable for cash at face value, depending on the state. This was instituted to protect consumers from losing money to breakage. This tends to make the cards hold value at least as long as paper currency does (or as long as their issuer is liquid).

The second is that, for user experience reasons, almost all issuers have a website which will, without requiring any sort of account or authentication, allow you to check the balance on a gift card with no information other than the serial number printed on it. This is to allow users to plan their shopping, but it also allows gift card marketplaces (and other participants in the supply chain) to know, with certainty, the value ofthe thing they are buying and selling, and to update that knowledge in near-real time. (Technologists will correctly predict that there is a fun cat-and-mouse game going on here regarding scraping and that retailers usually do not have sufficient programmers to win it.)

The marketing message of the sites is “Sometimes you get a gift card and it’s just not quite right, perhaps at the wrong store or a bit more than you need. You don’t want to be ungrateful and ask the giver to change it for you. We’ll pay you cash for it or allow you to exchange it for another gift card, for a small fee.”

That has probably happened at least one time in history, but that marketing message could not support an economy in gift card secondary sales, and yet that economy manifestly exists. Sebastian clued me into why.

Successful markets professionalize at least one of demand or supply
A common theme among marketplace startups is that they match two parties, the demand side and supply side, and that often this monetizes “casual” undermonetized assets in the world. Take Airbnb for example: you just happen to have an extra airbed lying around, someone just happens to be in town that week, wouldn’t you both be happier transacting than not if Airbnb could make it easy for you.

A common theme marketplace startups learn as they grow is that casual demand and casual supply are far, far less common than they need to be to support technology companies at typical marketplace take rates (the fees earned on transactions). Kevin Kwok has written about this in far more detail.

So who are the professionals in gift card reselling?

One potential answer is fraudsters, but that is obviously unsatisfactory. A marketplace which doesn’t stay ahead of fraudsters will find itself turned into a money pump, as it pays real money for gift cards, gets paid with stolen credit cards by fraudsters, and then has those purchases reversed by the banks of the defrauded individuals. It’s obvious that a marketplace which doesn’t solve this problem will close doors in a matter of weeks, and indeed fraud management is oneof the largest technical and operational concerns of gift card marketplaces.

It turns out that the broader economy contains many individuals and firms which, either through unexpected circumstances or through their own actions, are “naturally long” or “naturally short” gift cards. The naturally long users get far more gift cards than they could ever use personally. The naturally short ones require far more gift cards than they will receive from Santa, even if they’re very, very nice this year.

Where gift cards collect in the economy
Sock drawers, obviously.

No, there are particular individuals and firms who receive a vastly disproportionate number of gift cards.

How do we know? Because, since gift card exchanges need to defang fraud or they will die in weeks, they spend better-than-bank levels of effort on Know Your Customer. This is not because they are specifically required to by regulation but because buying gift cards from an unscrupulous source can see those cards drained after you’ve paid for them.

Cryptocurrency enthusiasts would tell you that simply transferring the number of a gift card doesn’t actually transfer the scarce asset, because the sending party still has the number. The structure of closed loop gift cards requires trust, and that need for trust causes marketplaces to know the business of their suppliers very intimately, and to actually investigate the representations those suppliers make.

Some people receive gift cards as de facto compensation.

For example, an ancient tradition in America is that teachers receive apples and other produce from students on the first day of class. This doesn’t actually happen anymore, but it remains in our cultural memory, so much so that teachers are often associated with the apple (you’ll see a lot of them in e.g. iconography on school supplies).

In the present day, many parents want to “tip” their children’s teachers, but they generally do not do so with cash. It is absolutely verboten in the professional managerial class to accept cash tips; that is the province of cabbies, hair stylists, and restaurant servers. Parents know this, and instead of giving a cash tip might send their child with a holiday card containing a gift card. Sometimes it is explicitly for the use ofthe teacher; sometimes it is explained as offsetting the cost of classroom supplies. (Many teachers in the United States furnish classroom supplies themselves, at an average cost of around $400 a year. Trust me, in a previous life I sold a lot of bingo card activities to them.)

This can result in teachers getting hundreds or low thousands of dollars of gift cards around the holidays, from a potpourri of stores which do not match their personal consumption needs. So every year, they come in droves to gift card marketplaces to liquidate them.

This is abundantly verifiable by the gift card marketplaces. Mrs. Smith, who represents herself as a teacher from Kansas, may have a Facebook account talking about being a teacher in Kansas, be listed on the website of a school in Kansas, have received a distinguished teacher award from the state of Kansas, and be willing to take a selfie with her last paystub. As any teacher would tell you, there are no stupid questions, noteven "Is Mrs. Smith factually a teacher in Kansas?"

A similar example is doormen. In Chicago, where I grew up, it is common practice for residents of buildings with attending staff to give each of those staff a tip at the end of the year. This is apparently also common in New York City. My parents recently lived in a building with one attendant and 90 doors. The typical tip is about $100. You can do the math from there. The doorman earns a substantial portion of his yearly income in the form of tips.

Some tenants make those tips in gift cards, and receiving a substantial portion of one’s annual compensation in a set of unmatched gift cards is inconvenient, so they also turn up to the marketplace in droves.

But individuals are not the primary gift card suppliers. Businesses are.

Businesses which intentionally expose themselves to gift cards they did not issue.

Many people exist in the cash-based economy and at the economic margins of society. Some businesses serve clientele who are composed primarily from that socioeconomic stratum, and they develop strategies to serve them better while making a profit.

Consider the corner bodega, for example. If you go there for a routine purchase, ring up $16.29, and then discover that you only have $15, that might mean you have to return some items. But some bodega owners will ask you “OK, if you don’t have money, what do you have?” And if you have e.g. a $20 Fandango movie theatre gift card, the bodega owner might say “Good enough! See you later.”  (They can,in principle, check the balance in real time using the instructions on the back of the card, or they could just trust someone they’ve known from the neighborhood for years.)

These gift cards go into a drawer for a while, after which the bodega owner sells the “portfolio” in a bulk transaction to a gift card marketplace (or a reseller who will do the same). They have been taking a substantial convenience fee from their customers for this service and will then realize it in cash.

Similarly, pawn shops and currency exchanges are both very prevalent in the economic lives of people in the margins, and a gift card is good for cash at many of them. They also bundle and sell the cards, because they cannot match supply and demand locally in a convenient fashion.

It is worth noting that economic precarity in the United States is generally not characterized by having no access to money, it is characterized by not having consistent access to money. Poor people get and give gifts, too. Gift cards also form part of the economic infrastructure of their neighborhoods. It becomes known that you can text an unbanked relative money by walking into any retailer and buying a gift card in cash, and that they will be able to convert that back into cash in minutes without needing to e.g. show ID that they may not have.

(It is probably also the case that not all of these gift cards are acquired legitimately, but the supermajority of economic activity is legitimate, everywhere.)

Who buys thousands of dollars of gift cards
A few patterns:

Businesses which resell fungible durable goods

Let me sketch a different retailer: a local cell phone kiosk. One important way they make their rent is stocking fungible items where there is perpetually high demand by people who are in a hurry, such as e.g. USB cables or phone chargers.

Where did that item come from? I had always assumed “The store owner has an account with AliExpress and bought a large box of them at $5 each then resold for $20.” It turns out that this is not always the case.

Some stores source their inventory from other retailers. They might drive a few hours from downtown New York City to upstate New York where Best Buy has a collection of perfectly good USB cables available for slightly lower than the market price in New York City. Say, 5% lower. This is classic arbitrage.

They then buy several thousand dollars of USB cables, chargers, etc etc, and drive back down.

This seems like a rough way to make a living, but retail entrepreneurs are often scrappy individuals who will put up with subjectively punishing lifestyles to build lives for their families in America. And they are savvy.

What if, instead of buying from Best Buy for cash, they purchased with a gift card that they got at a 6% discount? That more than doubles their margins. And how are they going to buy that gift card from a marketplace? With the best cash back credit card they can find, of course. Their 5% margins are now 13%, courtesy of other retailers and the payments industry.

And that justifies eight hours in the car followed by weeks of moving USB cables one at a time to stressed businessmen and tourists in Manhattan.

Businesses where inputs are sourced at retail

Home Depot is an interesting case. Many builders and contractors in the U.S. will purchase inputs to their trade, such as e.g. lumber or plumbing fixtures or similar, from Home Depot at retail prices. They might need thousands of dollars of material to do a job, and they might have an agreement with their customer (the client or the general contractor) that they invoice for “time and materials”, with inputs being reimbursed at cost orcost plus a small margin. They may need to show receipts.

Home Depot accepts Home Depot gift cards at par, and will print this on receipts. If you buy $1,000 worth of lumber, you get a receipt saying that you bought $1,000 worth of lumber. If you actually paid $940 for the gift cards that you made the purchase with, you conjure $60 of margin (worth perhaps an hour of your labor) for a few minutes of playing the gift card game.

And if you ever buy $1,000 of lumber… you probably buy an awful lot of lumber over the course of a year.

A lot of Home Depot gift cards are issued to people who e.g. want Dad to have something for his hobby projects, but they end up getting redeemed by business owners in the trades.

Consumers who are stacking discounts

There exists a calendar in clothing, based on seasons, fashion trends, and inventory management needs of stores. Some savvy shoppers, who prioritize price rather than trendiness, are acutely aware of this calendar, and plan for it in advance.

Most discounts in retailing do not stack. Gift cards, however, very much do stack with retailer-provided gift cards. $100 of H&M’s private money is worth $100 of trendy shirts at sticker price or $200+ if e.g. cashed in during the Labor Day sale.

Because clothes are extremely difficult to resell (they are not fungible due to sizes, there are real and perceived cleanliness issues, etc), professionals will not generally source their own clothing inventory at retail. Because clothes are produced at extremely attractive margins (East Asian manufacturing complex go brrrrr), retailers can offer extremely generous discounts at some points during their annual calendar.

This combination causes gift cards of specialty clothing shops to generally sell at large discounts, but be worth a multiple of that discount if used savvily. And so savvy hobbyist shoppers might buy $1,000 face value of gift cards for $600 and turn that into $3,000 of clothes (at sticker price).

Note that this introduces some interesting inventory risk, for the cards not the clothes. Cards enter the system at gift giving holidays like e.g. Christmas, but inventory liquidations are not tied to the gift giving calendar, they’re tied to the retailer operational calendar. So the system accumulates gift cards at Christmas and then discharges them (for a profit) around e.g. Labor Day. (Labor Day marks the “end of summer” in the United States and willtypically see retailers liquidate their summer inventory in favor of fall/winter inventory.)

The markets are extremely efficient
If you’re wondering why gift card marketplaces show different discounts for different cards, the above factors are why. Natural supply/demand mismatch is very different depending on retailer-specific factors, and that will influence which cards trade at tight discounts and which trade at wide discounts.

The marketplaces also exist in frothy competition with each other. There is no pretense of customer loyalty, on either buyer or supplier sides. They’re in it for the best offer they can get, and (because both buyers and sellers are largely professionals) they comparison shop. This turns the market into a ruthlessly efficient outpost of finance, like e.g. routing stock trades, instead of something like e.g. bank deposits where many consumers are happy receiving less interest in order to have a branch closer to their home, a teller they have a rapport with, or a color scheme they find attractive.

So why aren’t these markets even larger and more important to the overall economy? Because they are acutely supply constrained.

There will always be a buyer for a dollar sold at 95 cents.

Issuing businesses, though, have very limited incentive to encourage their cards to end up on marketplaces. (“Your opportunity is my margin!”, they might think, misquoting Bezos to their own purposes.)

There are a few closed loop card programs which find it rational to sell large blocks of gift cards at a very modest discount to the marketplaces, but it is a Faustian bargain for the marketplaces. If they’d make that deal with you, they’d make it with anyone, and so your competitive advantage disappears on half of each trade. You’re left with trying to win business on the retail side, where you exist in perfect,fratricidal competition with your peers.

There is also, fascinatingly, a metamarket, where aggregator sites attempt to step higher up in the purchasing funnel (by out-SEOing the marketplaces for e.g. “cheapest Starbucks gift card”), then list the marketplaces. They will, of course, want a cut of the action, and that causes marketplaces to bid down their margins to basis points.

Success as a marketplace involves attempting to attract as many natural accumulators of cards as possible, while avoiding fraudsters, and there just aren’t that many cards escaping into the secondary market as there are people who’d prefer to buy a dollar at a discount.

But there is still enough of an opportunity to support a thriving ecosystem, piggybacking on the marketing needs of retailers, the liquidity needs of consumers at the economic margins, the social anxieties of the American professional managerial class, and the relative difficulty of sending money to a friend via text message.

The secondary market in gift cards


Patrick McKenzie delves into the world of Buy Now Pay Later (BNPL) services and sheds light on their fascinating but often misunderstood nature. He explores how BNPLs work, their core customer base, and their crucial role in consumer transactions. McKenzie examines the mechanics of BNPLs, payment timing, and the credit offering they provide. He also discusses the implications of late fees and the underwriting process. Whether you're a curious consumer or interested in the business perspective, this article offers valuable insights into the evolving landscape of payment innovation.

Payments innovation has accelerated in recent years, including multiple fundamentally new payment modes. Most nations did not historically see a new mode in a typical decade. We've had at least five since iPhones hit pockets. One capturing a lot of attention and payment volume recently is Buy Now Pay Later (hereafter, BNPLs). BNPLs are fascinating and unfortunately not understood very well, partly because (in an ongoing theme for Bits about Money) their core customer is not in the social class which spends most effort in understanding the financial system, and partly because they combine old mechanisms in a way which is very novel. Let’s dig in.I work at Stripe, which builds infrastructure for the global Internet economy. We work extensively with BNPLs and with the businesses that use them, just like we aspire to work extensively with every way a customer and business would like to transact. I will note for hygiene purposes that there are three large global brands (Affirm, Afterpay, and Klarna) which are each publicly traded. My opinions are my own and are informed by my general professional expertise. Factual representations below are on the public record (or are my errors).How BNPL presents to a userThe overwhelming use case for BNPL is in consumer transactions for discretionary medium-duration non-durables or medium-amount durables, for reasons which will be very apparent later. Think makeup, clothes, and exercise bikes, but not (generally) food or cars.A new user for a BNPL is recruited either at the point-of-sale (POS) in an offline transaction or during the checkout flow during e-commerce. They’re presented with an endorsed third-party offer of credit: sign up for the named BNPL provider and, upon successful underwriting, get offered terms go here.The bread and butter offering in the BNPL space is “pay in four installments”, though there are others (pay in 3, revolving credit, etc). To make this discussion succinct and comprehensible for non-specialists, we’ll talk mostly about the mechanics of “pay in 4.”The timing of the installments is crucial to the offering: the first one is drawn immediately (typically from a debit card collected from the user during the instant underwriting flow), with the other ones being drawn automatically from the same card 2 weeks, 4 weeks, and 6 weeks after the transaction.This is a credit offering; the user is getting use of someone’s funds for a few weeks. Another crucial point is that the credit is free to the user. The installments are equal and include no interest charge if paid as agreed.What happens if the user doesn’t pay as agreed? That’s complicated, and highly sensitive to where the user is located and which BNPL they’re doing business with. It is dictated both by desired user experience, by how willing the BNPL firm (or their capital partners) is to subsidize credit losses, and by local regulation of consumer lending. Some BNPLs have no late fees; some have ones which are capped at low dollar amounts. Late fees are far less crucial to the BNPL economic model than they are for consumer lending generally, particularly subprime consumer lending, which BNPLs are sometimes (unjustly) grouped with.BNPLs underwrite each user (and each transaction) individually. This is entirely done by a computer and happens in milliseconds.Subsequent BNPL transactions follow the same model, except the underwriting process is faster since the user already has an account and doesn’t have to repeat data entry.How BNPL presents to an accepting businessAs we covered for credit cards, the payments industry charges partially to facilitate the orderly movement of money and data and partially for marketing. Businesses have three goals of marketing: increasing customer propensity to use them, increasing frequency at which they come back, and increasing amount spent (“basket size”, in retail lingo).BNPLs pitch themselves to businesses as more marketing efforts and less simple payments rails. They’re more expensive than cards by about 300 bps. (Payments nerds always quote prices in basis points, hundredths of a percent, rather than in percentage terms. In addition to this being a shibboleth to get oneself mocked on Twitter, it reduces ambiguity. If I had said “3% more expensive than cards” you might not know whether I meant “so about 3.1% plus some fixed fee” or “about 6% plus some fixed fee.” It is, of course, the latter.)Clearly BNPLs need something to justify costing twice as much (or more) as the default option. They claim they radically increase conversion rates (by X0%), basket sizes (by X0%), and repurchase frequency, without substantially cannibalizing card sales. (You can read the specific claims on their sites.) The math follows very quickly from there, particularly for retailers in segments like fashion and makeup who have high margins.An important subtlety: installment financing is not a new thing in the world. Unlike retailer-provided financing, the transaction is over immediately from the retailer’s perspective; they get paid very quickly (settlement times are broadly in line with credit cards) and do not have to book a deferred revenue liability. This is extremely important for retailers, and is something they pay for both with cards and with BNPLs. Moving the credit risk to the financial sector insulates them from it. Moving the deferred revenue liability makes their businesses structurally more valuable.Retailers depend on generating a high volume of transactions on an ongoing basis, and having a large percentage of them be financed by the retailer would tend to make the retailer look very highly leveraged, which stacks with other structural needs of retailers to leverage themselves (to e.g. finance inventory) and would tend to decrease their enterprise valuation. (For similar reasons, many large retailers that offer installments do so via arrangement with a financial partner who takes the risk in return for fees and, potentially, interest revenue.)Who owns the risk?Here’s where the financial engineering magic happens.The BNPL is a software company brokering a very complex credit transaction in a fashion which makes it free to the user, by producing ex nihilo high interest debt, manufactured out of the interchange fee the business paid.Most BNPL loans start with a bank, which is probably surprising to non-specialists. The dominant reason is that consumer lending is, in much of the world, a monopoly granted to the regulated banking sector. It was granted in return for that sector’s guarantee that it will not engage in the long and sordid list of anti-consumer practices possible in debtor/creditor relations.The bank originates the loan but does not desire to economically own it. They retain a small piece (on the order of a few percent) and sell the rest to someone who has agreed in advance to purchase it.In principle, that could be the BNPL provider itself. In practice, BNPL providers don’t want to own their originated loans either.The dominant reason is capital efficiency: BNPL providers are software companies which code up consumer loan factories, but they do not themselves want to be in the business of borrowing and lending at a spread. They do not, unlike the large banks which dominate consumer lending, have gigantic franchises which attract deposits at extremely low interest rates that they could use to lend out. Instead, they’d have to lend from their own capital, and capital for software companies is extremely expensive relative to bank deposits.So the BNPL provider looks for capital partners. There is a substantial amount of variation and nuance here, but broadly the counterparty is a large pot of money which seeks attractive yields at low but non-zero risk. That counterparty will offer the BNPL one of two things: a flow forward arrangement or warehouse financing.Finance loves making simple things sound complicated because that ensures the continued employment of a lot of finance professionals at high wages. Here’s the intuition you need for both of these instruments: “If you came to me tomorrow with a loan for $100 at 25% APR, since I trust you to underwrite good loans, I’d buy that loan from you. I have very few opportunities to buy low-risk 25% APR debt in the current interest rate environment. I don’t need to trust that loan, per se, I need to trust you, because I intend us both to be repeat players on this transaction and we both understand how statistics work. I am not buying the performance of that loan, I am buying the performance of random samples from your loan-creating machine.”Multiply that intuition by a lot, and that’s BNPL financing in a nutshell. (We'll elide that, as large software companies, the software / marketing / administrative side of the operation also needs to be financed, could issue debt and equity, etc etc.)The main difference between forward flow arrangements and warehouse financing is that there are two transactions in buying a loan and they don’t necessarily have to be coupled. One part of the transaction is providing capital. One part of the transaction is absorbing risk. In warehouse financing, the BNPL and the capital partner split the non-payment (or late payment) risk between themselves. In a forward flow arrangement, the capital partner agrees to bear substantially all of the risk. (From this follows some largely correct intuitions about which is more expensive, which banks are likely to engage in versus which is largely the province of hedge funds and similar, and the like.)Where did that 25% come from? To save you from having to prove this to yourself in Excel, here’s the intuition. A $100 pay in four transaction needs $75 of capital backing it for six weeks (because the user immediately pays a quarter upfront). The average amount of capital required over the term of the loan is half of that, due to repayment. If the BNPL was willing to pay 2% of the original transaction (sliced from their fee) to the capital provider in lieu of interest, the capital provider would receive $77 in repayments for their committed capital for 6 weeks, representing about 2.67% yield on $37.50 over 6 weeks. This works out to 25% annualized, give or take.Bam. You have just invented a relatively high-interest consumer loan out of nothing, without charging the customer a penny more than their purchase naturally costs.Of course, it is not out of nothing; the business is paying for it, as a marketing expense.Sephora wants to sell more Jelly Mint Glossy Lip Tint. They have a variety of options to incentivize this purchase, from giving you a discount to offering a free tote bag to underwriting use of credit (provided by Klarna). They are largely ambivalent which specific inducement it takes, as long as people buy more makeup, more frequently, from them.BNPLs sometimes re-use existing rails“Rails” are payment jargon for the technological and financial links between various entities which allow money movement. Visa provides rails, debit card networks provide rails, etc etc.One interesting thing BNPLs are beginning to do is to pivot the model. Historically they worked only with customers sourced by partner businesses on transactions at those businesses. This is strategically speaking a rough place to be. Increasingly, they want to own the customer relationship and re-use it at other businesses, the same way the bank backing a store card can convince you to use that card everywhere else on the same network.Card networks are the best network effects businesses ever, and to avoid having to re-build the merchant side of those networks, the BNPLs largely build this offering by riding on the existing card rails. They arrange for the issuance of a debit card (yes, of course, Durbin exempt in the U.S.) which you can use almost anywhere.The debit card typically comes with two options. One is less interesting as a product: it works virtually the same as any other debit card, making a withdrawal against your bank account via whatever local mechanism there is for that sometime after purchase. In this case, the BNPL only earns interchange revenue.The more interesting one is that they make it trivial to turn those debit transactions into credit transactions, generally after they’ve been made, within the linked app. As usual, these transactions are interest-free.Who pays for the credit here? The BNPL firm themselves, partially out of the interchange revenue (using much the same logic as before), and partially out of their hope that this causes you to use their core offering more frequently at merchants they have a first-party relationship with.As with all consumer lending products, controlling cost of customer acquisition dominates long-term profitability of the business. The best way to control it is to amortize it over long customer lifetimes. BNPLs don't want to acquire one transaction; they want to acquire a continuing stream of transactions from a customer, ideally changing their purchasing behavior for certain sectors and charging their partners for the change.Finance as social commentaryThe finance industry has not covered itself in glory over the centuries, and there is a reflexive distrust in many quarters that financial innovations could actually help consumers.One argument made is that BNPLs, by extension of credit to people who don’t have traditional credit available, allow customers to overextend themselves. (This is, for social reasons, very rarely articulated as “I am better at math than women who shop at Sephora, think they will not make good choices, and accordingly think they should have less choices presented to them.” But it is that argument.)One response to this argument is that BNPL users are less concentrated on the socioeconomic margins than they are believed to be. When credit cards were introduced, much sniffing happened over allowing people to borrow money to eat dinner, but credit and debit cards are used by the entire socioeconomic spectrum. Some BNPL users find it the only source of mainstream credit available. Some BNPL users are financially sophisticated, could trivially get (or already have) credit cards, and simply enjoy accrual accounting for capital purchases like e.g. clothes.But to ease the concerns of benign paternalists, BNPLs have a structural advantage over traditional credit arrangements: their duration is known in advance and known to be very short. A credit card is a commitment which implicitly requires you to have some sense of your earning potential, career stability and growth, and outgoings over a multiyear time horizon. BNPL asks you to look six weeks into the future. Many people, even people less fortunate than steadily employed professionals, have fairly tight understandings of what their finances will look like six weeks out. “Pretty much like today” is a good heuristic.BNPLs, due to their structurally high cost, tend to be used by businesses with structurally high margins. This means they’re less likely than credit cards to be used to purchase e.g. groceries. Stacking food bills for weeks might easily get someone at the margins to a point of overextension, but they’re far less likely to need to stack clothing or makeup purchases. If overextended, they can simply choose to stop or delay new discretionary purchases. Even defaulting on the BNPL is likely less impactful than defaulting on other debts; the chief consequence is not being able to use the same BNPL provider until one makes good on prior commitments. This is an impact, but it blocks the thing you use on discretionary purchases, not the thing which keeps the lights on.Moving the cost of credit from the end user to businesses dealing with them also insulates consumers from self-digging interest / late fee holes.A separate argument is that BNPLs increase risk to society itself, because society ultimately backstops consumer debt. How quickly we forget that the the global financial crisis happened, etc etc.Because new sneakers are less critical to the capital stock of developed nations than housing is, the total risk engaged in by all capital providers to BNPLs is several orders of magnitude less than mortgages and always will be. There will never be socialized losses for BNPL capital providers; they have a profitable and contained niche. The parts of it within the regulated banking system are structurally low risk and in the regulated banking system.The argument that BNPL and capital providers have to themselves, though, is even stronger than that sketch.We have decades of financial and ethnographic experience of what happens when modern consumer debts go bad. At a micro level, the narrative is typically overextension followed by an earnings shock (sudden unemployment, loss of earnings, medical issues, divorce, etc). At a macro level, debts go from largely uncorrelated across borrowers to extremely correlated across borrowers based on the business cycle. Broad trends in the economy hit many people at once over a period of months or years, leading to a rise in defaults, which plays out over several quarters in a row.BNPL firms and capital providers say “If the economy goes south, we have multiple countermeasures we can undertake in days or weeks, not in quarters. Credit cards can’t retroactively unmake the loans they were happy to have made a year ago but less happy about now. We structurally don’t have any loans older than about 6 weeks and those will age off our books in, hmm, 6 weeks or less. We also underwrite new purchases per transaction rather than per credit line, and therefore get our pick of where we spend our risk budget. We can adjust underwriting based on geography, business sector, basket size, even the individual items purchased, to stay open for business while temporarily slowing our riskiest transactions. Or we could go to capital partners or businesses and find a price for that risk that makes everyone happy.”These arguments have not been tested in the unhappy half of a business cycle, but they do not sound as implausible as “makeup and mortgages are both systemic risks to the financial system.”Broadening New Payment LanesNew double sided marketplaces (which effectively all payment rails are) are extremely difficult to incubate in the world, which is one reason why platforms are so powerful: snap into them and you can save a lot of the effort on one or both sides of the market. Much of the world is, like my home of Japan, moving towards an equilibrium where there are many accepted payment methods and more coming all the time.One thing that Stripe can hopefully offer to upstart payment methods is broad reach on the business side. To businesses, we can offer decreased integration cost and technical churn as new methods start to win consumers’ hearts and wallets.This is good for both businesses and consumers. Credit card networks were and are a stupendous achievement for the world, but a bit of competition helps to keep them, like anyone, on their toes. This keeps prices (net of rewards) down for the rest of the economy and spurs ongoing product innovation happening.Sometimes those product innovations are genuinely powerful for various parts of society. Sometimes they help people buy sneakers.Sometimes they do both.

BNPLs: Businesses Needing Provided Legibility

Bits About Money

Patrick McKenzie delves into the current state of banking in uncertain times, examining the recent failures of multiple U.S. banks and the potential for a widespread crisis. McKenzie explains the underlying factors contributing to these failures, including the impact of rising interest rates on asset prices and the consequences for financial institutions and individuals alike. With a straightforward and sober-minded approach, this thought-provoking essay provides valuable insights for anyone seeking to understand the fragility of the banking system and its implications for the broader economy.

Banking in very uncertain times

In this thought-provoking essay, Patrick McKenzie dives into the world of banking and explores the concept of deposit franchises as natural hedges. As the banking crisis continues to unfold, McKenzie examines the structural issue of rising interest rates and its impact on asset prices and bank solvency. With a sharp wit and financial expertise, McKenzie challenges conventional wisdom and provides a unique perspective on the risks taken by banks and the perplexity surrounding interest rate risk. Don't miss this illuminating exploration of the banking sector and the lessons to be learned from its current turmoil.

Deposit franchises as natural hedges

In this eye-opening essay, Patrick McKenzie delves into the world of credit card debt collection, shedding light on an often overlooked aspect of the finance industry. He explores the lifecycle of defaulted debt and provides insights into the different archetypes of credit card users. With personal experiences in dealing with debt collectors, McKenzie offers a unique perspective on the brokenness of the system. If you've ever been curious about the inner workings of debt collection and its impact on individuals, this thought-provoking piece is a must-read.

Credit card debt collection

Discover how credit cards are undergoing a transformation as Patrick McKenzie delves into the flaws of the current system and the solutions being developed. In this enlightening article, McKenzie shares insights from his experience at Stripe, explaining how the industry is addressing issues like transaction failures and the security risks associated with card numbers. Gain a deeper understanding of the challenges faced by the credit card ecosystem and the engineering work being done to improve efficiency and security, ultimately leading to a better user experience. Buckle up and prepare to have your perspective on credit cards changed.

Improving how credit cards work under the covers

Patrick McKenzie delves into the complex world of deposit insurance maximization as a service and the challenges it poses in analyzing systemic risks in the banking system. In this article, McKenzie explores how financial service providers navigate the limits of deposit insurance to offer treasury management solutions to companies. Gain insights into the intricacies of deposit accounts and the pricing of banking services, shedding light on a lesser-known aspect of the financial industry. Discover the complexities behind why startups and businesses choose to keep large sums of money in bank accounts and uncover the underlying dynamics of this practice.

Deposit insurance maximization as a service

Patrick McKenzie delves into the complex world of Know Your Customer (KYC) and anti-money laundering (AML) programs in the financial industry. While these acronyms may sound familiar, McKenzie goes beyond surface-level understanding to reveal the nuanced and often misunderstood nature of these regulatory practices. In this thought-provoking essay, he explores how these programs aim to stochastically manage crime and terrorism by requiring financial institutions to implement their own risk management strategies. McKenzie challenges preconceived notions about the effectiveness of KYC and AML, providing a fresh perspective on their role in mitigating risks in the industry.

KYC and AML: beyond the acronyms

In his candid review, Patrick McKenzie dives into the captivating world of cryptocurrencies, dissecting the book "Number Go Up" by Zeke Faux. McKenzie, a long-time skeptic of cryptocurrencies, commends Faux's unbridled contempt for the industry's scam promoters. He also explores the book's exploration of investigative journalism, the allure of numbers in finance, and Faux's engaging writing style. If you've ever been curious about the ins and outs of the crypto world or enjoy a thought-provoking discussion on the intersection of finance and journalism, this review is a must-read.

A review of Number Go Up, on crypto shenanigans

In this engaging essay, Patrick McKenzie delves into the world of fantasy miniature painting and explores how disruptive innovation has transformed the industry. He traces the supply chain backward, uncovering how advancements in plastic manufacturing and the power of the internet have enabled a global network of small firms to produce highly customizable miniatures. McKenzie discusses the challenges faced by traditional manufacturing industries and highlights the abundance of options available to hobbyists today. Whether you're a geek or simply curious about the intersection of hobbies and professional interests, this article is a fascinating read.

Plastic (and payments) in the fantasy supply chain

In this eye-opening essay, Patrick McKenzie delves into the world of credit cards as a legacy system. He highlights how the industry's reluctance to adapt to the changing times has created vulnerabilities, particularly in the face of modern challenges such as online transactions and card fraud. McKenzie's analysis sheds light on the complexities of the credit card system and raises questions about its long-term viability in the ever-evolving digital landscape. Prepare to challenge your assumptions and gain a deeper understanding of the infrastructure behind this ubiquitous financial tool.

Read Bits About Money

Choose Your Reading Path

Full Backlog

Best Hits

You'll start with:

Community Banking and Fintech

Financial innovation is actually happening

How Credit Cards Make Money

About Bits About Money