How Payments Ought to Work
From First Principles
I discovered Dee Hock earlier this month. His books - Autobiography of a Restless Mind & One From Many - Visa and the Rise of the Chaordic Organization have become quick favorites of mine.
Dee Hock was a simple man who felt more at home in the Pacific northwest than in the center of the world in California. I can relate to Dee’s desire for simplicity. I hail from a small midwestern town in Northern Illinois, and have always enjoyed smaller towns and open space just like Dee did.
I love that Dee followed his own intuition for much of his career. He had a hard time submitting to corporate culture, back when there were few other ways to succeed outside of it. He spent his entire life wrestling with a few key questions:
Why are organizations everywhere, whether political, commercial, or social, increasingly unable to manage their affairs?
Why are people, everywhere, increasingly in conflict with and alienated from the organizations of which they are part?
Why are society and the biosphere increasingly in disarray?
Living in 21st century America, these questions are even more relevant today than they were in the 1960s when Dee started Visa. I feel as though I’ve seen enough of the inner workings of modern companies, universities, & government to know that something feels wrong. There is some sort of endemic disfunction that seems to be plaguing the vast majority of Western institutions. Some, such as Balaji Srinivasan, believe that the internet is the cause of our disfunction. I’ve heard him compare the internet to a kind of universal acid, and that hardly any institutions that were founded prior to the creation of the internet will survive it.
While information businesses have thrived in the 21st century, it feels as though the organizations that maintain our largest information networks are under a great deal of stress. The American financial system feels more confusing and complicated than ever. Massive money printing, an equity market that is up while real economic growth appears to be stagnant, and inflation that is hurting citizens around the world all give the impression that something might break. Social media companies, when they’re not fighting tooth and nail for your attention, seem to be under the constant pressure of governments to censor and surveil citizens the world over.
I can’t help but feel that Dee was right in One From Many when he discusses the need for new forms of organization to deal with our rapidly changing world.
The area that Dee chose to focus was on payments. And in One From Many he said that part of Visa’s founding was a result of a certain type of thinking. Specifically, Dee reflected on what our systems of payment had been, what they were currently, what they might look like, and what they ought to be.
It’s worth using this framework to think through payments from first principles for today, from the vantage point of a different world: one that is more connected and filled with far more advanced information technology than it was in the mid 1960s at Visa’s inception.
A Brief History of Payments
If we go back to ancient times, prior to the widespread use of currency, ‘payments’ happened via barter and what I would call ‘favors’ (which were really just a form of credit).
Where barter was used, trade was highly inefficient, because it relied on the ‘coincidence of wants’ (I just-so-happen to have what you want, and you just-so-happen to have what I want). ‘Credit’ in these days seems to have taken the form of favors enforced by physical violence. I’ll give you the berries we picked if you agree to give me a portion of the meat you hunt this year. If you don’t pay up, we’ll invade your tribe and take it ourselves. Deal?
Currency made all of this much easier because it removed the reliance on the coincidence of wants. If you’re interested in the forms that these early currencies took, Nick Szabo does a brilliant job of walking you through them in this post.
But once we had currency, payments still required physical transportation of this currency (usually metal coins) to provide payment. This limited the size of markets and also the size of transactions because there was a transportation cost associated with lugging around pieces of metal. And as Adam Smith cited in The Wealth of Nations, division of labor is limited by the extent of the market. In smaller markets, you get less division of labor, and with less division of labor, you tend to get smaller overall economic output.
Eventually, governments got smarter and began to issue paper money. The printing press made it easy to issue, and the lightness of paper reduced transaction costs & helped to expand the size of the market. As banking systems got more sophisticated, the check was introduced (see this study for a history of the check). The check would remain one of the dominant forms of payment all the way into the early 21st century. Checks made it ever more convenient to pay for goods and services across long distances because, instead of having to manage pieces of paper, you could carry around templated checks which could be issued in any amount at any time.
Credit was used all the while for payment, it just got much more sophisticated as governments became more stable and banking industries emerged. Colonial America relied heavily on credit, much of which originated in Europe. Many people came to the Americas and agreed to provide years of free labor in exchange for passage, food & shelter.
All of this worked as well as it could have, but it had huge costs to maintain. You have to cut down trees for paper checks & money. And you have to physically move checks around for them to be processed or to buy anything far away (i.e. via the Sears catalog). By the late 20th century, you had overnight flights and long distance truck hauls that consisted of nothing but paper checks.
Eventually, computers were invented and got really, really good. People like Dee Hock started to realize that money was really just information. And if it were only information, maybe we could invent better means of payment.
So Visa, Mastercard, and American Express pioneered the credit card and debit card industries, enabling the fastest and most convenient means of payment ever created. It’s done so well that card payments are how trillions of dollars worth of payments are still done to this day.
Payments Today
While card payments are still one of today’s most common forms of value transfer, there are still people who prefer to use cash, coin and check around the world instead. Businesses make use of other forms of payment as well, including ACH or Fedwire if in the USA, SEPA in Europe, MPESA in Africa, and SWIFT if making cross border payments. Many of these payment systems are plugged directly into central banks around the world, and at the time of their conception (in the mid 20th century for most), they were a cutting edge innovation.
But given the state of the information technology, most of them are incredibly inferior to what they could be. I won’t discuss our modern forms of credit (this would likely take a separate post), but instead focus on the digital payment rails of the present day.
Payments Are Stuck In the 1970s
In the US, the ACH network processes trillions of dollars worth of transactions per year. According to a study from the US Federal Reserve, the ACH network processed 69% of all non cash payments in 2020. But let’s think about how the ACH system works for a minute.
The ACH network processes transactions in batches 4 times per day during weekdays, and is closed on holidays and weekends. You can think of the ACH network a layer of abstraction atop the US Federal Reserve, which runs on COBOL servers (COBOL is effectively a dead programming language). This system essentially keeps a bankers hours. When looked at from a first principles point of view, the ACH network is just crediting and debiting accounts. It’s passing bits around, and it’s arguably simpler than what something like email does. But could you imagine if Gmail worked the same hours as the postal service?
You can send money faster with card, but the transaction typically does not settle until hours after your payment (which can admittedly be a feature in the case of fraud), and card companies often enforce transaction limits and charge relatively high fees. You could technically send money even faster with a wire transfer, but these transfers often cost between $15 and $40, which certainly feels exorbitant for what is just a simple transfer of bits. The reality is that our modern systems of payment are not even close to using the best of modern information technology.
The fintech industry has boomed over the past 2 decades, and has improved the UX of using money around the world by quite a lot. Visa, Paypal, Venmo, Cashapp, Square, and co have all made using money easier. While fintech companies are great at product design, they’re unfortunately forced to build on top of payment rails which are outdated by decades.
Cross Border Payments Suck
Sending cross border payments today sucks far more than most domestic forms of payment. Here’s a site I found which gives advice to e-commerce sellers on how to pay factories overseas to produce a product for you. It details everything from the $40+ wire transfer fees, multi-day wait times, and - look out, if you make a spelling mistake on the paper forms that banks make you fill out, it could take you up to 2 months to resolve the problem.
Are complicated payments processes like this good for international trade? Absolutely not. If anything, they are restricting the size of the market for businesses the world over. Especially given the amount of remote work happening throughout the globe, there is going to be an increasing level of demand for high quality cross border payment systems, and they currently leave much to be desired.
Who Controls Our Payment Systems?
These systems are centralized and monitored by banks and governments. Cash transfers, on the other hand are effectively untraceable. Some say that this makes cash the tool of criminals. This is true, but it’s also a tool for individuals living under oppressive regimes.
One thing that the framers of the US constitution seemed to deeply understand was that it wasn’t enough to hope that democratic voting processes would elect the right leaders who would protect individual freedoms. They recognized that human nature often led governments astray. The desire for power & control is a human tendency, and eventually, you’ll find yourself with people in positions of power who will seek to impose tyranny. So what did the framers do? They introduced many checks on the power of any individual branch of government. They acknowledged that ‘don’t be evil’ is inferior to ‘can’t be evil.’ By attempting to restrict the power of each branch of the US government, they reduced reliance on good governance and goodwill, and mitigated against the dark side of power and human nature.
Now that money is digital, and controlled via digital processes & centralized banking systems. These very rails of payment can be used as a means of control by governments. The CCP surveils every digital transaction of its citizens, and financial history is used as a parameter within their social credit scoring system. Your score isn’t high enough? Sorry, you’re blocked from purchasing certain types of services.
The powers afforded by centralized payment systems aren’t only being used by regimes we recognize as oppressive and authoritarian. In Canada (of all places), protesters in 2021 had their bank accounts seized under the guise of ‘public health and safety.’
Regardless of one’s political orientation, these measures of arbitrarily seizing bank accounts and regulating what one can do with their finances by government are troubling. I fear that these financials controls will be too convenient to use, and that governments around the world will normalize the use of financial controls to attain ‘compliance.’ Society seemed to get along just fine before governments were able to surveil the vast majority of transactions of its citizens. Should we really give a small group of people the ability to censor our ability to use money?
Where Payments Could Go
CBDCs
Central banks and large financial institutions are interested in more efficient means of payment. There is growing interest from central banks about CBDCs - central bank digital currencies. A central bank digital currency is usually considered to be similar to blockchains, but its quite unlike what we typically think of as a blockchain because it’s 100% centralized.
A blockchain is really just a giant, append only, open shared database. In a decentralized blockchain like Ethereum, Bitcoin, and the like, no single entity is capable of censoring transactions. But for a CBDC, the central banks maintaining them would have complete control to censor. China, the US, and the EU have all unveiled research on CBDCs, and it’s possible that there will be at least one CBDC live within the next 5-10 years.
In a positive CBDC future, payment systems no longer take a bankers hours and we have lightning fast, somewhat programmable money. Nation states probably use it to occasionally censor people they deem to be dangerous, and there are occasionally some false positives (i.e. people who are censored who shouldn’t be), but by and large the general population sees a net positive impact due to reduced transaction costs, increased transaction throughput, and programmability.
But a negative CBDC future is Orwellian. There is no financial privacy. Businesses are arbitrarily blocked from the payment system if the government sees their work as unfavorable. Dissidents are prevented from using the financial system entirely. Cash and decentralized blockchains are outlawed, and the financial system becomes an instrument used to enable tyranny.
The problem with CBDCs is that we’d have to trust governments to not use their newfound power for evil, and I for one am not comfortable with that.
Stagnation
There’s also an alternative reality in which we remain in a state of regulatory capture. The EU for one seems to take the position of emphasizing regulation over innovation, so maybe the European Union will continue this trend within the realm of finance.
The banking system within the US seems to benefit in many ways as a result of the lack of technical sophistication. For example, NACHA, the organization that maintains the ACH system in the US, seems to be very self congratulatory. The headline on the NACHA site is “ACH Network Prospers Through Economic Uncertainty,” and they seem to be generating quite a lot of revenue from selling training programs, certifications, and memberships that have price tags in the tens of thousands of dollars.
Banks benefit from overdraft fees and late payments which inevitably result from the payment system being closed on weekends and holidays. And they also benefit from the revenue generated by high wire transfer fees. In an episode of NPR from 2013 titled The Invisible Plumbing of Our Economy, the hosts discussed an upcoming vote that financial institutions were set to have on whether or not NACHA should shift ACH payments to enable same day processing. Eventually, this was implemented into ACH, but at the time in 2013, the group voted no. What was one of the most commonly cited reasons as to why same day was voted down? It was “financial institution opposition due to cannibalization of existing revenue.”
Maybe the current system is just too incentivized to keep things the same, and will fight to keep them so for a long time.
How Payments Ought to Be
Let’s break this down how payments should be on a few axes:
speed
programmability
user experience
access
control
privacy
Speed
Money should move as close to the speed of light as possible.
Programmability
We should be able to automate as many financial tasks as possible. Money should be moved by robots. We should also be able to define new ways for money to move. We should be able to stream money like we can stream any other form of data (an aside - this is possible today via the Superfluid, the protocol I work on in my day job).
It should be as easy to design new financial products and forms of money as it is to create any other software program.
User Experience
Using money should be like using any other tech product. Fintech companies are mostly doing a good job here already.
Access
Everyone in the world should be able to access the best of the world’s payment systems, regardless of where they live.
Control
Local governments should be able to impose some regulations at centralized points in the system (i.e. on exchanges, banks, or businesses), but they should not be able to own or control the payment rails on which money operates. No government owns TCP/IP (the internet protocol you’re using to read this), but they can regulate the businesses and individuals who make use of the protocol. Financial protocols should work the same way.
Privacy
People should have the right to financial privacy if they want it. Some believe that the private exchange of money enables crime. They would be correct.
But here’s a list of some other things that enable crime: cash transfers, private conversations, and pointy objects.
Freedom is a double edge sword. It limits the power of centralized institutions to stop bad actors before they’re able to commit crimes and acts of terror. But at the same time, it prevents governments from imposing tyranny on people in the long run. We should be very careful not to trade long term freedom for short term protection. If anything, embracing crypto could be the most pro-freedom, pro American thing we could do:
Open Blockchains + Fintech = The Way Forward
I’m biased, but I think that some combination of modern fintech & decentralized, public blockchains show a lot of promise. Used together, they have the potential to satisfy each of the criteria I just laid out for how payments ought to work. They already enable a system which is open, programmable, and outside of the control of centralized institutions or governments.
However, crypto has some ground to cover on user experience, true privacy, and speed. But there are many developments which are addressing all of these. Account abstraction & developer tooling are improving user experience. Zero knowledge Layer 1 blockchains and ZK rollups on Ethereum are enabling total privacy. Layer 2s, new Layer 1s, and new scaling solutions are addressing the problems of speed.
In my day to day work at Superfluid, I’m helping to build a world in which payments work in the way they ought to. But I don’t always get the opportunity to think as much about the long term (this is how startups can be). The crypto bear market of 2022 has been quite painful (though in many ways, deserved), and we’re focused on the adoption of our technology in the short to medium term.
But if you zoom out and think in decades, it feels like these improvements are inevitable as long as crypto continues to build. The world is increasingly online and connected, and as today’s institutions continue to show cracks, it will become ever more important to build a financial system which is made for the 21st century and beyond. One that’s the way it ought to be. Like Dee did it.