#1 because the only way regular people can hold central bank money (as distinct from retail bank liabilities) is cash. And cash isn't convenient for most transa

#1 because the only way regular people can hold central bank money (as distinct from retail bank liabilities) is cash. And cash isn't convenient for most transactions. #2 Yes, most money is digital. But most of that money is NOT central bank money. Imagine you 'have' $500k in the form of digital money. How much of that will you still have if your bank goes bust? #3 Yes, this can be solved in a different way (as it has been in the UK). #4 Regular banking and online payments are built on legacy payment rails. Transactions between customers of different banks are slow or expensive or both. A CDBC does not need to use distributed ledger technology. Depending on your objectives, it could be 'just a database'. But that's the point: people are still trying to work out what those objectives are. Perhaps privacy and anonymity is an objective. Perhaps the opposite (all transactions can be tracked back to an individual) is an objective. Those objectives (and competitive forces) will determine how a CDBC is implemented. "the entity involved is already literally the arbitrator and source of truth of how much currency exists" Yes, but it might simultaneously want to increase transparency. How do you know the central bank is truthfully reporting M1? Could a distributed ledger help?