Bordo and Levin favor an account-based CBDC system (managed or supervised by the central bank) rather than central bank issued tokens in the blockchain.
They emphasize the Friedman rule and the fact that interest paying CBDC affords the possibility to satisfy the rule:
These … goals – … a stable unit of account and an efficient medium of exchange – seemed to be irreconcilable due to the impracticalities of paying interest on paper currency, and hence Friedman advocated a steady deflation rather than price stability. But the achievement of both goals has now become feasible using a well-designed CBDC.
Interest paying CBDC would imply—payments to account holders. Bordo and Levin do not discuss the political economy implications. They are also silent about the transition from the current system with deposits to a new system with interest bearing CBDC in which demand for deposits would drastically fall.
Bordo and Levin favor abolishing cash to render monetary policy most powerful. Eliminating the option to withdraw cash would also eliminate the lower bound on nominal interest rates and would render unnecessary any “inflation buffer” of 2 percent or so. Monetary policy thus could move from positive inflation targets to a price level target.
Their paper contains a long list of useful references.