We could have the commercial banks, and others if necessary, run nodes. They could verify payments on “accounts”, or addresses, (if we want to keep the old meaning of bank account, that’s fine, then we could instead call these addresses) which are held by the commercial banks. These addresses could be non-anonymous such that requirement for know-your-customer still reside with the commercial banks (the central banks really don’t want to have to get anyway near KYC requirements). Each such “account”/address could be linked to addresses of currency at the CB, which at that point was anonymous.
The CB could then have full control of the issuance of currency, control interest directly on the currency itself, everything being anonymous to the CB.
If the currency is going to be centrally issued then what reason is there to use a DLT given all of the drawbacks it would introduce?
DLTs aren’t a solution to a technical problem, they’re a technical solution to a trust problem that comes with some pretty hefty implementation baggage and performance constraints.
If you have to trust a single issuer anyway (as would be the case with a CBDC) then the core reason for using a DLT doesn’t exist.
Since each miner is linked to a publicly known institution, attacks on the blockchain could be punished with jail time and revocation of the authorization.
Proof of work is just a very advanced anti spam protection system. In decentralized systems people can create as many accounts as they want. You need to make sure that they can't.
I'm just saying this is how it could work. It's not obvious whether its a good or bad idea.
Clearly neither of those two rationales make any sense whatsoever - and they might even seem like a downright ridiculous suggestion to some readers of this response. If that’s you, then welcome to wild and whacky world of public spending.