BIS research head says distributed ledgers cannot replace settlement liquidity

A constraint beyond the ledger

Hyun Song Shin, economic adviser and head of research at the Bank for International Settlements, presented an economic model on November 9, 2018 that separated two problems often combined in blockchain discussions: agreeing on a transaction history and supplying enough liquidity for institutions to settle large payments.

The presentation, delivered at the University of Chicago’s Cryptocurrencies and Blockchains conference, accepted that distributed-ledger technology could maintain reconciled copies of payment history among participating institutions. It also described trial systems in which banks transferred digital tokens redeemable at a central bank. In those trials, Shin said, decentralized consensus typically required validation by a supermajority of approximately 75% to 80% of participants.

His central conclusion was narrower and more consequential: technical agreement on a ledger does not itself finance payments. Even a network whose nodes agree perfectly about balances and transfers can produce less settlement activity when participants face meaningful credit costs.

What the model claimed

Shin said the value of payments processed through a conventional real-time gross settlement system can exceed a participant’s central-bank deposit balance by more than 100 times. Incoming funds are therefore recycled into outgoing payments, while central-bank credit and private credit from outside the payment system help sustain the flow.

Those figures were parameters and institutional observations used to motivate the presentation, not the result of a newly disclosed transaction-level dataset. The surviving BIS record does not identify a particular payment system or measurement period for the “over 100 times” comparison, limiting independent replication.

In a simplified two-bank example, the model found that communication could not produce a cooperative payment outcome once credit costs crossed a threshold. A broader model treated the problem as a public-good contribution game among an unspecified number of banks. A clean, reconciled ledger formed one public good; the provision of credit needed to support outgoing payments formed another.

The model’s interpretation was that decentralized coordination remained possible, but only across a narrow range of underlying conditions. It often failed to reproduce the high payment volumes associated with central-bank balance-sheet support. That was an analytical finding, not proof that every distributed ledger would fail or that existing payment architecture could not be improved.

Market and protocol context

Crypto markets were comparatively subdued on November 9, 2018. Kraken’s venue-specific daily report recorded $77.1 million of trading across its markets and displayed bitcoin at $6,342, down 1.31%; ether at $208, down 2.46%; and bitcoin cash at $556.50, down 5.53%. The surviving report does not clearly define the precise reference timestamp, conversion methodology or daily-candle boundary, so those values should not be treated as universal market closes.

The same date also illustrated the operational difference between open cryptocurrency networks and supervised wholesale settlement. CEX.IO announced that it would halt bitcoin-cash deposits and withdrawals three hours before the network upgrade scheduled for November 15, 2018 at approximately 16:40 UTC. It also said customers would receive a matching amount of a new coin if the disputed upgrade produced a chain split. That was an exchange custody policy, not a guarantee about which chain would retain the Bitcoin Cash identity or market value.

Together, the records showed why institutional adoption involved more than copying blockchain mechanics. Operators still had to decide who supplied liquidity, which assets were recognized, how disputes were resolved and which balance sheet ultimately absorbed settlement risk.

Scope and later context

Shin expressly stated that the views were his own and not necessarily those of the BIS or its member central banks. The November 9 presentation created no regulation, payment standard or central-bank commitment.

The BIS later noted that an updated version was presented at the University of Cambridge on January 22, 2019. That later presentation confirms continued development of the analysis but does not change what the November 9, 2018 record established.