More Central Banks Going for CBDC, May Not Use Existing DLT – BIS
The Bank for International Settlement (BIS) has presented central banks seeking to issue central bank digital currencies (CBDCs) to the public with design stage choices even as it identifies 17 selected projects either researching, piloting, or, like China, have ongoing work in place at the moment.
The bank for central banks sketches out in its special feature Quarterly Review some key technological design considerations for a retail CBDC based on consumer needs and associated technical design choices that could help interested countries with their decision.
Unlike current electronic retail money which it says represents a claim on an intermediary rather than functioning as the digital equivalent of cash, the BIS notes that “CBDCs could potentially provide a cash-like certainty for peer-to-peer payments” and, at the same time, “should offer convenience, resilience, accessibility, privacy and ease of use in cross-border payments.”
They did not investigate the case for or against the issuance of a CBDC, its implications or management. Rather, they identified key attributes a CBDC should consider: whether to feature intermediaries, run on a conventional or distributed infrastructure, have account- or token-based access, and enable retail interlinkages across borders.
Based on a CBDC pyramid structure, the BIS stresses the importance of the operational architecture to interested central banks to ensure a cash-like safety and convenience for consumers like intermediaries confer on the existing payment system. Their decision to either base the infrastructure on a conventional centrally-controlled database or instead on distributed ledger technology (DLT) which differs for their degree of protection from single points of failure will help them meet the consumer’s need for cash-like payment safety against the insolvency or technical glitches of intermediaries and outages at the central bank, it adds.
The underlying trade-off between privacy, ease of access and ease of law enforcement depends on whether access to the CBDC is tied to an identity system (i.e. account-based technology) or via cryptographic schemes that do not require identification (i.e. digital tokens-based technology). Tied to whether the CBDC is account- or token-based is the enabling of cross-border payments which could be arranged via technical connections at the wholesale level built on existing systems or novel interlinkages.
On the costs and benefits of using DLT as the CBDC infrastructure, the BIS states that while DLT stores data multiple times and in physically separate locations like a conventional infrastructure, they differ on how data are updated. Conventional databases achieve resilience by its multiple physical nodes being controlled by one entity unlike by different entities with DLT in a decentralised manner and harmonised between the nodes through consensus mechanisms.
“The overhead needed to operate a consensus mechanism is the main reason why DLTs have lower transaction throughput than conventional architectures. Specifically, these limits imply that current DLT could not be used for the direct CBDC except in very small jurisdictions, given the probable volume of data throughput. However, DLT could be used for the indirect CBDC architecture, as the number of transactions in many wholesale payment systems is comparable with that handled by existing blockchain platforms, as also demonstrated in several wholesale CBDC experiments conducted by central banks.”
Their vulnerabilities differ too. While the top node of a conventional architecture could fail by a targeted hacking attack, for example, the consensus mechanism in a DLT could be put under pressure by a denial-of-service type of attack. That DLT essentially outsources the authority to adjust claims on the central bank balance sheet to external validators is also a concern. BIS says using DLT can only be advantageous if the network is trusted to operate more reliably than the central bank.
It adds that ongoing assessments of DLT-based proofs-of-concept tend to be negative particularly as it is not clear that scalable implementations can actually rely on the technology.