Under collateralized settlement.
This week I thought I would talk about “undercollateralized settlement”. These may sound like the two most boring words in the English language — but actually, they may change the way we relate to each other as human beings.
Money in the zeitgeist
Tay Zonday. You know, the chocolate rain guy - made this video which has been trending recently.
I think it took off because it captures a zeitgeist of the current moment.
Mama, economy, make me understand all the numbers
why Daddy’s on a welfare plan — ‘turnin thirty, forty, fifty
gotta move in with my parents.
And the stocks go up but the jobs … disappear.
Anger with the bankers, anger with wall street, and anger with the corporate fat cats. Not all that new. It comes in waves but we’ve seen it before. What’s new is that people are starting to discuss and try to understand how money actually works. More lyrics from Tay:
It’s an I.O.U., remember dollars are a promise
When you borrow from a bank It’s not from other depositors
The money for your loan Gets created on the spot
Perhaps it’s because cryptocurrency is in the air. It’s actually quite exciting to me that people are starting to talk about how money is created!
But it doesn’t go far enough. We need to get beyond the anger and talk about how we can take that power back.
So let’s take a second to understand what’s going on with these bank ‘promises’.
Promises to pay
When you go to an ATM and check your balance, you see a number. Is that ‘real’ money? Not really, it’s a promise to pay that hasn’t yet settled.
If you withdraw the actual cash. That’s when the promise is settled.
As you probably know, the bank doesn’t have enough ‘cold hard cash’ in their vaults to settle all their promises at once.
As you may know, they don’t even have enough reserve bank deposits to settle all their promises at once.
That’s where the ‘ under collateralized’ bit comes in.
Sending money to phoenix
Why does it take a whole entire day (sometimes more) to send money to your sister-in-law in Phoenix? Even if you live in Phoenix, if she banks with a different bank from you, it can take forever for the money to get there.
This seems crazy. People often reasonably assume that banks just need to upgrade their technology, and there’s some truth to this — if we can send money halfway across the world in seconds using completely decentralized crypto then surely the banks should be able to get it together.
But the ‘hold up’ here is not just banks having a lot of old technology. Banks (most banks) are actually being actively stubborn about upgrading their technology because they definitely don’t want to ‘open up’ the process the are really using here. They like having a monopoly on this settlement process. It’s actually kind of fascinating what happens each night when the banks ‘settle up’ with each other.
Sending money to Phoenix takes three steps:
- The bank reduces its promise to you (reduces the numbers for your account in their computer)
- Your bank then makes a promise to the bank in phoenix
- The bank in phoenix then makes a promise to your sister in law (adds numbers to her bank account in their computer)
Step 2 is where it gets interesting. How do two banks ‘settle up’ on this promise? They certainly don’t just ‘leave their money’ in the other bank.
Of course, the sending bank could just put the cash in a truck and drive it over to Phoenix. But generally, they don’t. This is where the reserve bank deposits come in.
Reserve bank deposits are a special kind of bank account that your bank has with the reserve bank.
Each bank (or regulated depository institution) has ‘deposits’ on ‘reserve’ at the reserve bank. This creates a kind of hierarchy of banks making promises to other banks.
So, to transfer the money from your bank to Phoenix all they need do is ask the reserve bank to send the money from their account to the one belonging to the Phoenix bank. Since both accounts are on the same computer the reserve bank can just change some numbers in a computer and the story ends.
Often enough, when you read up on this story that’s where the story ends. But, there’s a really important step that’s been missed out here.
Because if you add up all the transfers over the course of the day, there’s a good chance that the transfers (or promises) from your bank to Phoenix will be ‘cancelled out’ by transfers that go in the other direction.
For instance, if all the promises added up together result in a promise of $120 million from your bank to Phoenix, and all the promises in the other direction add up to $90 million, then they only actually need to find $30 million dollars of actual reserve bank deposits to ‘settle’ the whole debt. The rest of it is simply ‘cancelled out’.
And banks prefer to collect up all the transactions that take place in a given day so they can net things out with other banks as much as possible, without having to move ‘actual’ money.
How to settle a promise
So to review, there are two ways to ‘settle’ an ‘under collateralized promise’.
Option 1. Get cash (or reserves) out of the bank and ‘pay the person’.
Option 2. Find matching promises and use these to ‘cancel out’ the payments.
Actually, you don’t even need an exact matched pair. Any ‘cycle’ (or loop) in the graph of promises can be used to cancel things out in this way.
In fact, the gap between narrow ‘fully collateralized money’ and the giant web of ‘under-collateralized promises’ is the gap on which most of the modern economy is built. At any given time about 93% of what we think of as ‘money’ — or about 90 trillion dollars — is in the form of bank ‘promises to pay’ and only a small fraction of that could actually be ‘collateralized’ at any given time. Every night when the banks ‘settle’ by cancelling out their debts, they are finding a way to ‘settle’ those debts that doesn’t involve having to actually collateralize those promises. (*Source: Worked for many many years at a financial research company on Wall Street and later at a Reserve Bank).
Banks as gatekeepers
So, maybe your take away from all this is - hey let's stop sending under collateralised promises and start sending ‘actual’ money. Maybe you think crypto is just better — because when you transfer bitcoin or crypto you are directly transferring the actual underlying asset. Surely that is better — none of this undercollateralized nonsense or promises-to-pay. Just send the actual token directly. No bankers. No third-parties. No interest, either (although that’s changing).
But there is more to this story. Because undercollateralized promises are actually kind of magic. When a bank comes into a community it creates new activity that didn’t previously exist — people can get loans to start new businesses based only on their idea and their ability to convince a bank manager that they can make that business real. And then, as Tay Zonday says:
When you borrow from a bank It’s not from other depositors
The money for your loan Gets created on the spot
It’s hard for some people to realise but it’s 100% true. And while it may seem bad it also has its upsides. This ability to ‘create money on the spot’ allows things to happen, to be ‘imagined into existence’ in a way that just wouldn’t happen if we stuck to only ‘narrow’ money.
Let’s say a young baker goes to the bank and borrows some money to start up a bakery. If the bank manager says yes, they enter numbers into the computer that are just an (as yet, unsettled) ‘promise’ in the form of digits in the baker's bank account. (The kid who walked in thinking of himself as ‘someone with an idea’ now thinks of themselves as a baker.)
What is, actually happening is they are swapping promises — the baker promises to pay the bank back in the future, and the bank makes a promise in the form of numbers in their computer that the bank will ‘settle’ for cash (or equivalent) on-demand.
If the baker pays someone whose bank account is already at the same bank, the bank just moves the ledger entries around, but if the baker pays someone at a different bank account, then the above dance of ‘overnight settlement’ happens.
So, why don’t banks just create infinite money?
At one level, this is because the ‘originating promise’ — the one that actually ‘created’ the money is the one from the kid who is now a baker. The banks can’t just ‘create’ money without those promises. But at a more practical level (and equivalently) it’s because the bank has to ensure their net flows in and out are roughly equal. The bank has money coming in from the baker, who is earning money selling bread, depositing it into their bank account and occasionally ‘paying off’ their loan. But they also have money going out from the bakers account whenever the baker withdraws cash or pays people at a different bank. Unlike for the reserve bank, if a commercial bank just ‘prints money’ — if they create too much money in the form of ‘bad loans’ — that will create a gradient of cash between them and other banks, such that on average the flow is mostly outwards. That is ‘promises to pay’ are mostly away from the bank and towards other banks or institutions without corresponding ‘promises to pay’ coming in. If they do that too often, or for too long then they will have to use their limited ‘reserve bank deposits’ to settle with other banks. And soon enough their ‘reserve deposits + cash’ will start looking like those of an Icelandic bank from late 2008. (This is called ‘going bust’).
For a bank, their reserve balance (which, we remember. is the money on deposit with the reserve bank ) is a way to keep score. If reserves go down sharply, the banks know they are making too many bad loans. Correspondingly, if they go up then the bank is likely to think about making more loans in order to make more profit.
Because none of this happens for free. Of course! And there is the rub. The banks want to maximize their profit by making as many loans as possible and earn as much interest as possible for their shareholders without damaging their reserve ratio (too much). That, essentially, is the job of a bank manager and/or risk department. Apart from the massive risk that they will periodically screw this up and throw the economy into recession there are two other huge downsides to this arrangement:
- The bank managers act as gatekeeper. It is the bank that gets to decide what business they believe is worthwhile and what is not. It doesn’t matter how good your finances are if you present yourself the wrong way. Or, more to the point if you have a skin colour, age or gender, or any other aspect of your person or your business idea that is outside of their comfort zone. If they don’t like the idea then, likely as not, that particular idea won’t happen. This excludes whole communities from economic prosperity and stifles creativity. Which sucks.
- And secondly, of course, they charge lots of interest.
Wouldn’t it be better if we could find a way to create this type of money for ourselves?
That is — create credit in a more distributed way without the gatekeepers.
To be clear I’m talking about the 93% ‘promises to pay’ not creating more of the ‘fully collateralized’ underlying assets. These are already being created at a great rate of knots on the blockchain. But this is not enough.
Because where and by who money is created is important. Crypto is amazing — as a new form of reserves. But it doesn't create money on the basis of a great idea. It can be used to invest in great ideas but the tokens themselves are created by miners or (with proof of stake) by those who already have tokens. In this way, we risk only further perpetuating ‘rich get richer’ dynamics (and ecological impact to boot).
The ability to create credit money — aka the 97% of money — is actually a social function. We as a society have collectively agreed that banks get to act as gatekeepers on this process. That rich shareholders and the people they appoint as bank managers and investment advisors should have this power. Credit Unions and locally owned banks are a step in the right direction. But we can do better. Much, much better.
Even more, if we track promises directly on and off-chain, we can start to see the real ‘cycles in the graph’ that are the basis for reciprocity in our economy. Data and visibility brings with it the ability to more consciously decide which cycles of reciprocity we want to support. New forms of reciprocity and collaboration can be possible if we do away with the banks that (not only act as gatekeepers, tax the poor and cause periodic recessions, but also..) create information loss by squeezing all our transactions through their narrow ledgers.
But I’m getting ahead of myself. Let’s get to all of that in future posts. For now, I hope we can all just take a second to appreciate that the settlement of under collateralized promises is a surprising and interesting thing!
__Miles, week 2