Tokenised Deposits. Which flavour will really help?

Not commercial bank money. That was last week’s conclusion. This week, a thought on what could work and on whether we even need a token. To help the debate, I introduced a definition.

A tokenised deposit could be either:

1.????????? Type A: issued by a commercial bank or organisation without a specific claim on specific assets.

2.????????? Type B: issued by a central bank, in other words CBDC.

3.????????? Type C: issued by a private organisation with a specific claim on a specific pool of assets.

So, Type A is of limited use. Specifically, it is great inside the four walls of the bank which issued it.

Next, let’s think about Type B: CBDC. This can only be issued by a central bank (CB). The clue is in the name. Let’s think about the theoretically possible. Allow that a CB issues some CBDC and allows it to sit in a reserve account at the CB or in its payment system aka an FMI.

Just from a technical perspective, this could have all the useful attributes identified in last week’s post:

1.????????? Singleness of money; a payment asset which settles in central bank money and is not an IoU, i.e. commercial bank money.

2.????????? Always on; a payments system which does not shut down. Albeit, even if 24 7 365 we will still need to have clear value dates in each currency.

3.????????? Programmable / interoperable; the balance in any payment system can interact with multiple external payment and settlement systems. This is one part of what is sometimes termed composability. So the payment asset P in any one currency can then serve for any payment (P), any FX settlement (PvP) or securities settlement (DvP) need. I call this a SPooL, a single pool of liquidity.

4.????????? Directly addressable: the balance in the system must belong to the beneficial owner.

Type C, a privately issued token, could achieve the same thing if the issuer was able to hold 100% of the deposits with the CB, and ideally add an extra layer of protection called “bankruptcy remoteness”. Fancy term, which means separate those client assets from the operating or issuing company. The reason for this is simple. In a bankruptcy event, depositors wait as unsecured creditors behind a queue of other claims; salaries, the VAT man, the tax man, social security etc. Bankruptcy remote would say that the holders of the token collectively have a claim on the reserves at the CB.

RLN, The Regulated Liability Network. This is an idea led by Citibank with support from many other banks. The basic idea is to create a universal ledger standard and network on which both commercial and central banks could operate a ledger. Although the thinking is not yet fully mature, I see it as generally sound. Aligning numerous Central Banks and their flock of commercial banks around a universal standard is both an admirable goal and somewhat the Holy Grail.

There is though one aspect of the design which troubles me, which is the mixing of commercial and central bank money. Let’s go back to the examples I cited above about my business with Credit Suisse and making payments. If my business were with Citi and I am paying another Citi client using RLN, that is a simple book-transfer. The technology is irrelevant, so too is central bank money.

If, however we are talking about creating a SPooL and a universal means of payment, then in the wholesale markets things are trickier. Imagine Bank A and Bank B are both Citibank clients. A has a balance of +20, B has a balance of -20. In a very abstract view, Citi has no money at the central bank. If Bank A wants to use the 20 to settle trades with parties outside Citi, anybody accepting that 20 for settlement is taking an IOU from Citi. There is no central bank money. So, balances in the RLN system are not a priori bankruptcy remote from the bank that is holding the balances.

The next plausible solution is Fnality[1]. This is a bank consortium led initiative to provide an on-chain payment solution for new digital asset related needs. Essentially it has all the ingredients of programmability and inter-operability we have talked about above. If it can interoperate with legacy systems which are not on a blockchain, then it is a very plausible SPooL. Fnality is a slow burner; it started as a research project in 2015 and is targeting a narrow range of currencies; GBP will be first, in 2023. with USD and EUR at some point thereafter. CAD and JPY should happen too. The design explicitly avoids the use of intra-day overdrafts; every penny in any currency is backed by central bank money. And the central bank money is bankruptcy remote from the operating company. I would offer the view that this is a very sound solution. However, the roll-out is slower than originally planned. The use of “blockchain technology” is something which has challenged the regulators.

The third plausible solution is RTGS.global[2]. This is a private effort, so it does not have the constraints of being run by a banking consortium. RTGS has developed the technology to operate ledgers and settlement processes in multiple currencies and to support all three flavours of P: DvP. P, and PvP. The tech has been built in cooperation with Microsoft and does not use blockchain. Like Fnality, balances would be backed by central bank money. RTGS would be a SPooL. With private backing, RTGS potentially has the wherewithal to scale quickly.

Both Fnality and RTGS.global would be FMIs. The technology is available to ensure that in each currency / jurisdiction, regulators can have as much real-time access to monitor proceedings as they want to. Regulatory approval means clearing a very high bar, and rightly so. Fnality has some approvals in GBP. As of early November 2023 Fnality had not commenced operations in GBP and RTGS’ license submission status was not public.?

For both types B & C, there is a but, and it is not insignificant. As a rule FMIs have “access rules” which dictate who is allowed to be a participant and have an account in the system. Typically, the requirement is for the participants to be entities which are domiciled and regulated in that jurisdiction. This results in something known as “tiering”. From the regulator’s perspective this is about oversight and monetary policy. If an entity has an account in a central bank-controlled payment system, balances are as low risk as you can get. A “risk-free asset” is not quite accurate, but the point is that in a crisis those with accounts would likely choose to hold a balance and not to use commercial banks and money market funds. This could quickly be de-stabilising.

When all of us who are trying to change the system and build tomorrow’s infrastructure rock up at a central bank to wax lyrical about our latest and greatest idea, faster than you can eat the biscuits and drink the tea, the CBs will ask: “Who has a claim against whom for what?” They are gauging the potential impact of your “new, new thing” on control over monetary policy. The starting point is they do not want to give any wider access than today. For the most part, that is a concern about end-of-day positions, although there is an issue to be managed around the impact of an intra-day issue.

Now, I will argue that all this new infrastructure is all about settling trades and making payments, so allowing wide access is important and I’d even stick my neck out and say that allowing wholesale market participants, at least regulated ones, to hold limited overnight balances, would be a very good and sensible thing. Just maybe some central bankers might agree with me. Then will come the “but” or, the “gotcha”, or the test of “if we give them an inch, will they take a mile?”. The conversation with say the Bank of England, would likely go like this:

“So, your system proposes that institutions we don’t regulate could hold a balance of GBP in the ledgers in your system during the day and overnight?”

“Yes”

“Let’s say one of those institutions is ZKB from Switzerland, whom we don’t regulate. If they have a balance of 100 GBP in your ledger at the end of today, what do they have a claim on?”

“They would have a claim for a 100 slice of the big cake of GBP in the omnibus account we would have as a payments system.”

“That would be outside the current scope of our access rules. Only those who are under the Sterling Monetary Framework are allowed to hold overnight balances.” ??

“But we only want to settle trades.”

“What would you propose?”

“How about we allow those ‘foreign’ regulated institutions to hold a limited amount overnight. Maybe one amount for G-SIBs, another for D-SIB[3]s or some other simple structure.”

“Just supposing we were to allow that, perhaps a cap of GBP 200 million for G-SIB and 50 million for a D-SIB. How would your system work if one fine day, ZKB decided to simply sell a fistful of CHF vs. GBP and sit long say GBP 1 billion overnight?”

“We assume market participants are rational actors and ZKB will only do that trade to settle trades today.”

“But, with what you propose, they could be irrational, could they not? What about if your system goes bankrupt during the day and stops operating before ZKB can use those GBP?”

These are complex issues. Of course we need new infrastructure. I remarked above about “you cannae change the laws of accounting”. Equally, we plumbers need to respect the “rules of monetary policy.”

A successful SPooL will require that we allow wide access. With that it must have a structure which fits within monetary policy norms. A structure which might allow unfettered access to the risk-free asset is not appropriate, nor even necessary. Just maybe, central banks will agree to some limited overnight holdings. That will require payment system operators to design a structure which ensures control. One possible combination is to require that foreign institutions have a domestic sponsor. If there is an intraday event, the rules would say the balances belong to the sponsor, who in turn has a liability to the foreign institution. At day end, there would need to be a process to “sweep” any balance, or excess balance, to the sponsor. An automated money-market loan. The result is the same as the situation today if an institution uses a correspondent bank. The institution can proactively manage its end-of-day balance. If it doesn’t it has a credit balance, which in accounting terms is an unsecured receivable vs. its Nostro.

I would offer the view that trying to completely loosen monetary policy controls is a hill that us plumbers should have not the ambition to climb & conquer. We will die a long slow death as we try to get what we want. The participant / sponsor construct, plus or minus some cap, would get us what we need. Payment Systems or SPooLs would be forced to have an end-of-day. As Mssrs. Jagger & Richards put it: “You can’t always get what you want, but if you try sometimes, you might just get what you need.”

A last thought is on the word “token”. My good friend John Whelan , with some help from BNY’s Larry Miller , recently sparked a useful debate by saying that we might be better served by calling the thing we are aiming to create a “controllable electronic record”. More on this next week.

Thanks for reading.

If you would like to talk more about things liquidity, please just type: “Can we talk?” into the comments box.

I am a long-time and long-in-the-tooth hawk on matters liquidity. I work closely with a great team of people at Planixs where I am the Liquidity Futurologist. Planixs is in the business of making sure FS firms can avoid what happens when liquidity dries up. I hope that together we can make liquidity sexy. ?

Want to know how?banks?really work? Maybe you are in a?FinTech, a?bank?or a?consulting?firm. Maybe trying to do something new.?Understanding how to make things work properly is a good investment of your time.

The Bankers’ Plumber’s Handbook is about how investment banking works, how to do operations?properly and keep control.?

The book tries to make it easy for you and includes a collection of real-life, true stories from 30+ years of adventures in banking around the world. True tales of Goldman Sachs and collecting money from the mob, losing $2m of the partners’ money yet still keeping my job and keeping an eye on traders with evil intentions.

So, you might like the tool kit, you might like the stories, or you might only like the glossary, which one of my friends kindly said was worth the price of the book on its own.?Or you might like all of it.

Go ahead, get your copy!

Hard Copy via Create Space:?https://lnkd.in/eSDz8tCu

Soft copy via Download:?https://lnkd.in/eac5zWBD?- Cheapest option

Please feel free to get in contact: https://www.dhirubhai.net/in/bankersplumber/ and [email protected].

?

#Liquidity #LiquidityManagement #LiquidityMatters #FinancialServices?#RegulatoryConfidence #Planixs #Realiti

?

[1] Disclosure: I have previously worked for Fnality.

[2] Disclosure: I was an advisor to RTGS.global from November 2022 to August 2023.

[3] D-SIB: Domestically Systemically Important Bank. One level down from G-SIBs.?