They go together.  Central-Bank Digital currencies (CBDC) look to be an inevitability.  Given that, can we at least use the UK government’s new digital identity trust framework to ensure some level of privacy.

Premise one – CBDCs are inevitable

The Bank for International settlements (BIS) is the club for central banks and it recently produced an important speech/ paper on CBDCs ( ).  Central bankers are herd animals; the paper indicates clearly that this herd is on the move.  Various discussion papers, proofs of concept and pilots going on in central banks all over the world are sending a mixture of “don’t lag behind” and “it’s okay” signals back and forth between central-bank decision makers, to the extent we at HBW now think that CBDCs are inevitable there is too much momentum for them to stop.

Premise two – It’s all about Central Bank control and their options

There are all sorts of possible justifications being touted by central banks as to why CBDCs are a good thing; some say consumers deserve access to a risk-free asset to replace the disappearance of physical cash.  Some say the parallel payment system will enhance payment systems resilience and financial system stability.  Others would argue that it allows for proper means of interest rates to go negative.  It’s also been argued as a potential way to improve financial inclusion.  And finally, many believe it will produce increased competition by nonbanks to take part in the world payment services.  HBW takes a more prosaic view; namely that it would give central bankers more control of more the financial activity and would give them another lever to pull for monetary policy for prudential regulation reasons.  If we are correct on this point then a number of architectural design points for a CBDC system follow.

Conclusion one – distributed ledgers are out

The basic point of permissionless distributed ledger technology (DLT), the sort used by most crypto currencies, is to allow people who do not know/ trust each other to collaborate and exchange.  This is not the context under which a Central Bank within a country operates; the Central Bank gets to choose who is trusted and who is not and everyone has to trust the central bank (whether they like it or not).

Permissioned DLT is the same technology but only a small number of players are allowed to take part; this might resemble the situation of the central bank and a number of money centre/ clearing banks.  It might give some system resilience but would reinforce the complex monopoly of money centre/ clearing banks.  Most of all it would seem to generate extra technical cost and risk when a central system would be coherent with the central bank wanting control.  We imagine CBDC having an architecture like the one below;

The central bank would hold accounts which carry the end users’ balances of CBDC in a central ledger.  The users would have an interface provided by a payment services provider (PSP) which might (or more likely might not) be a bank.  The PSP would be responsible for all the end user interaction such as registration, queries, user interfaces, account closures, fraud checks, etc.  We do not expect the Central Bank to want all the complexity of all this.  The central bank keeps a “thin” record of the account balance and transaction history and essentially only deals with PSPs.  The PSP holds no money so there is no risk to the user if it goes bust, hence no need for a deposit guarantee scheme.  If a PSP goes bust the user could access their account at the Central bank by a different PSP.  Their money would be preserved.  How the payments would work is illustrated in the diagram where the user A makes a couple of payment instructions and the transactions are passed on by their PSP to the central bank.  Theoretically the central bank could make all this work with DLT but it would be quicker, cheaper and lower risk without such technology.

Conclusion two – There will be no multi-country CBDC

One of the attractions of cryptocurrencies is that they would be useful for cross-country transfers and payments by not incurring FX and charges.  Since a key premise of the CBDCs will be about central banks of countries including it as part of their monetary armoury, there will be one CBDC for each country.  The central banks will not be offering multicountry CBDCs as that would be treading on each other’s toes and the total impact of all the national CBDCs will be to reduce demand for other potential stable coins such as Diem from Facebook.  Why would you want to hold a risky digital currency backed by Fiat currency when you can have one from the manufacturers of the Fiat currency directly?  We anticipate some form of international architecture based on collaborating central banks as illustrated below.

The essence of the idea is that there will be a network of Vostro/ Nostro accounts in the CBDC ledgers of the central banks to facilitate international payments.  There may well be a new Clearing Settlement Mechanism (CSM) to provide a parallel to SWIFT for the network of Central Banks.  The PSP and/ or FX provider will end up providing a service to the end user to allow foreign transfer using this network of central banks.

Recommendation – Use digital identity providers

The UK government has recently published a consultation on a digital identity trust framework.  CBDC seems an ideal situation to use it.  If the Central Bank is going to have an account for each and every member of the vast majority of the population (the Bank of England will want to have that degree of penetration to achieve its resilience and inclusion aims) then there is some concern about a central database with all the personal information needed to run an account.  Ideally one wants the Bank of England to hold just the ledger; all the connecting information to individuals such as name, address , date of birth, device IDs, biometrics etc. would be stored outside the ledger.  Many central banks are just arms of the government and putting are a lot of high quality data under their control may not be appealing.  One possibility would be to just trust the PSP’s (who are regulated) and leave the data with them, but a better solution would be to use a digital identity service provider as shown below;

This example is the registration; user A wants to set up CBDC account with the PSP and uses her preferred digital ID service provider to supply the necessary authentication to allow the account set up at the PSP and Central Bank.  The identity service provider could also be used for payment instruction authentication or confirmation of beneficiary requests.