Earlier this week the Bank of Thailand (BOT) announced the results of its proof of concept (PoC) for a wholesale Central Bank Digital Currency (CBDC). Wholesale CBDCs are targeted only at financial institutions, and eight took part. R3, the enterprise blockchain company, was the technology partner alongside Wipro. The overall conclusion was positive, though some areas were identified as needing further work.
The eight banks that participated were HSBC, Standard Chartered, Bangkok Bank, Krung Thai Bank, Bank of Ayudhya, Kasikornbank, Siam Commercial Bank and Thanachart.
“The outcome of Project Inthanon has demonstrated that DLT has not only increased efficiency but also reduced risks and overall costs in the financial ecosystem,” said Pimolpa Suntichok, Senior EVP Commercial Banking Solutions, Siam Commercial Bank.
At the beginning of January, the Bank for International Settlements (BIS) published a survey and report on CBDCs. Sixty-three central banks accounting for 90% of the world’s economic output responded, and 70% of them are either working on or about to start working on CBDCs. However, the report emphasized that most of the work is purely exploratory. Very few central banks are considering a general-purpose CBDC in the next 1-3 years. And the time frame for the first wholesale CBDC seems even longer according to the survey.
Project Inthanon, named after Thailand’s highest mountain, had three aims. First as a tool for collaborative learning about distributed ledger technology (DLT) by Thai financial institutions. Secondly, to develop a PoC for the payment infrastructure or Real Time Gross Settlement system (RTGS) using a tokenized version of the Thai currency, Baht. And finally to evaluate the impact of a tokenized RTGS.
In this case, apart from tokenizing cash, the bank tested liquidity saving mechanisms (LSM). Quite often multiple payment transactions can be resolved by netting payments. The BOT said their design found a “compromise between the optimised liquidity used and priority of transactions.”
Additionally, the solution tokenized bonds. The BOT was able to provide liquidity automatically by enabling bond tokens to be used as collateral. Bond tokens could be exchanged for cash tokens, ie. Delivery versus Payment (DvP), without an operator as the intermediary.
By doing away with the intermediary, banks can settle with each other around the clock. In turn, this reduces credit risk by avoiding liabilities building up outside of BAHTNET operating hours.
Plans and scope for improvement
Two further phases are planned with the second phase aiming to expand on the Bond tokens. It will target coupon payments, interbank trading, bond redemption, and interbank repos (short term collateralized lending between banks). Plus it will explore regulatory compliance particularly for non-Thai residents, as well as fraud detection. In the third phase the BOT plans to explore interoperability with legacy systems and other platforms, including cross-border transactions.
The results were very positive with resiliency as one of the main areas that need to be explored. Part of the issue was the decision to use a single notary even though a multiple notary setup could have been tried. In R3’s Corda the notary prevents double spends. By only using a single notary it introduces a single point of failure.
However, there were also other resiliency issues. For example, if the node for one bank goes down during “gridlock resolution” it causes a failure in the payment netting process.
As many central banks experiment with blockchain, by sharing the results, other banks can make even more significant progress.