Central Bank Digital Currency or CBDC today tops the agenda of every Central Bank. While the People’s Bank of China and archipelago economies like the Bahamas or the Eastern Caribbean States have actually launched CBDCs or pilots or are about to do so, others are actively evaluating the pros and cons of the architecture, technology, operating model and many other aspects of this new form of money. Yet others are waiting and watching with the intent to learn from experiences and insights.
To put it simply, CBDC is a digital version of the cash (fiat money) that we use in everyday life, and just like cash, the liability of CBDC lies with the Central Bank. This difference between CBDC and the money that we move electronically between bank accounts is that the latter is backed by these commercial banks, not by the Central bank. The move from physical currency to digital currency can be significant because apart from removing the operational costs and anonymity of cash that makes it so easy to use for illicit activities, CBDC provides another way to pay digitally even without a bank account, brings with it benefits such as financial inclusion and modernization of a country’s payment system, and triggers innovation through competition among payments players. Cryptocurrencies may be popular but they have received bad press due to their volatility and uncertain future. Stablecoins that are pegged to currencies like the dollar are only slightly ahead in reputation, but even they remain unregulated. Now that even the Fed is working on a CBDC framework, the Crypto-Stablecoin landscape is bound to undergo substantial changes.
CBDC can be Wholesale or Retail. Wholesale CBDC provides banks with a digital version of their respective Central bank reserves and is meant to be used by Real-Time Gross Settlement (RTGS) systems for settling domestic transactions between banks. Wholesale CBDC can also be used to settle cross-border transactions.
Retail CBDC (also called general-purpose CBDC) provides customers with a digital version of the bank notes and coins they use for their everyday transactions. Each time this digital money is issued to their bank accounts by the Central bank, customers can withdraw these units to fund their CBDC wallets and pay for transactions at stores or online. The amounts and the intervals at which these units may be issued by the Central bank depend on a variety of reasons.
CBDC can be account-based or token based. Account-based CBDC refers to digital currency balances held in Central bank accounts, ownership of which is linked to an identity. Successful transactions are reflected by debiting the payer and crediting the payee accounts, much like today’s digital payments systems. Token-based CBDC are digital tokens representing bank notes or coins of specific denomination and are linked to digital addresses (not to any person or legal entity). A transaction represents a change of ownership/address like when a banknote changes hands.
Central Banks are considering implementing Retail-only CBDC or Wholesale-only CBDC, or both. The choice varies depending on economic, political and geographical reasons. The archipelago economies for example have found CBDCs to be critical as the hundreds of small islands face unstable communication and power, and the risk of natural disasters that make it exceedingly difficult to operate a cash-based economy. Emerging economies on the other hand are favouring Retail CBDC to further their financial inclusion agenda as many in those countries remain underserved in terms of access to payment systems. The Digital Dollar Project on the other hand, is said to be triggered by China’s eCNY CBDC efforts that could potentially threaten the global position of the dollar.
Wholesale CBDC (wCBDC)
wCBDCs are accessible only to regulated financial institutions (FIs) that hold reserves in their accounts with the Central Bank to settle interbank transfers. The mode of operation of settlements across banks is similar to that of the traditional system – debiting the FI account with a net debit position and crediting the FI account with a net credit position.
The difference is that in the traditional model, these accounts are credited or debited without any actual transfer of values, but a wCBDC could use a digital token to actually transfer value from Sender FI to Receiver FI without any intermediary. Moreover wCBDC technology could (a) connect to other markets such as securities or foreign exchange, (b) allow for smart contracts (e.g., a payment settles only if an asset has been successfully delivered, or variable interest rates are applied depending on a set of criteria) (c) reduce fraud through real-time monitoring of fund movements, and (d) simplify cross-border payments by minimizing the number of intermediaries. This potentially makes wCBDCs faster, safer and less costly, and an extremely attractive proposition for most countries planning a launch.
Building a wCBDC however requires the involvement of private parties and even banks, since Central Banks cannot be expected to have all the knowledge and experience to build this technology.
As of August 2022, Saudi Arabia, UAE, Singapore and Malaysia are planning to launch wCBDC pilots, while Bahrain and Switzerland are in the wCBDC development stage.
Noteworthy wCBDC Pilots
wCBDC can have many benefits as we saw above, but even without wCBDC, FIs today have digital access to central bank money, so in a way, the wCBDC flavour already exists. When DLT and other technologies link multiple domestic and international markets to make wholesale and cross-border payments more innovative and efficient, wCBDC will have truly arrived!
Retail CBDC (rCBDC)
rCBDC is a different ball game altogether as it impacts all retail payments made by the general public. A successful implementation of an rCBDC can be transformational for an economy, and this is apparent from its growing coverage in whitepapers, blogs, essays, discussion notes and point-of-view articles from leading consulting companies, central banks, financial regulators, international FIs and thought leaders around the world. Despite the different opinions around rCBDC implementation, it is generally believed that rCBDCs are meant to be developed for the public good, and its design should enable an open payment platform that is conducive to innovation.
rCBDCs vs Faster Payment Systems
It is natural to ask how CBDCs can better today’s faster payment systems that ensure instant receipt of funds in the beneficiary account. The behind the scenes reality is often overlooked – that faster payments need wholesale settlement between the banks involved, and that the same may take time which is tantamount to a short-term loan and associated credit-risk situation between banks. CBDCs in contrast enable direct transfer between users of each transaction in real time and are thus settled directly in Central bank currency – a distinct benefit when it comes to cross-border payments.
Account-based vs Token-based rCBDC
rCBDCs can have two designs: Account-based or Token-based.
In an Account-based design, both payment initiators and beneficiaries would need to identify themselves. Such an rCBDC will need to be linked to an identity scheme with KYC mandates. The Central bank will have to set up individual KYC-based accounts for every individual to issue account-based rCBDC. The proposed CBDC in India is account-based.
A Token-based design comes with full anonymity, and transactions are authorised using public-private key pairs and digital signatures similar to passwords. In a sense these tokens are similar to cash. This model ensures privacy but is prone to be used for illegal activities like money-laundering. Moreover, users need to remember their private keys. It is also argued that token-based transactions take longer to settle on the blockchain than account-based instant payments. Thus it is generally considered that token-based systems cannot serve the public interest. South Africa, Senegal, Brazil, Cambodia and some others are favouring this option all the same, for reasons of their own.
A July 2021 report ‘Progress on Research and Development of E-CNY in China’, the first English language whitepaper on the topic published by the People’s Bank of China (PBOC), defines China’s retail e-CNY (Digital Yuan) as ‘the digital version of fiat currency issued by the PBOC and operated by authorized operators. It is a value-based, quasi-account-based and account-based hybrid payment instrument, with legal tender status and loosely coupled account linkage’ – implying that it is both an account-based and a token-based digital currency, or something midway. E-CNY allows immediate settlement in contrast to credit card or e-money payments. Although e-CNY is meant for low-value retail payments, the paper seems to indicate that e-CNY is interoperable with reserve accounts at the PBOC which might imply that e-CNY has the design elements to be used as a wCBDC in the future.
Back in India, the RBI is exploring the option of implementing an account-based wCBDC and a token-based rCBDC using a graded approach.
Three main CBDC models are discussed depending on how the Central bank works with commercial banks and intermediaries in the new ecosystem. In all cases, rCBDC are the liability of the Central bank.
(A) Direct rCBDC
The Central bank onboards individual/retail clients with KYC-compliance either directly or via intermediaries, settles retail payments directly in real-time, and keeps a record of all balances and transactions. This model is unlikely to be accepted as it involves huge responsibilities on the part of the Central bank which will have to perform all activities of Banks in addition to its own duties as the financial regulator of a country.
It would also stifle innovation by banks and other payments players in the long term. Iceland, Denmark and Senegal had started to focus on this architecture, but presently only Iceland is in the research stage.
(B) Indirect or Intermediated rCBDC
Intermediaries such as Banks or PSPs onboard individual/retail clients with KYC-compliance, and process retail payments. The Central bank holds the wholesale ledger balances of the Banks/PSPs, processes wholesale payments, and closely supervises the intermediaries to ensure that wholesale balances add up to the sum of all retail accounts. This model is similar to the existing retail payments system.
(C) Hybrid rCBDC
A blend of (A) and (B) above, in a hybrid design, intermediaries onboard individual/retail clients with KYC-compliance, and process retail payments. The Central bank keeps a copy of all rCBDC balances and periodically updates the same. An advantage of this model is that if an intermediary fails, the Central bank is always there to back any claims.
It is believed that two-tier models are better suited for a balanced financial system where each party does what they excel in. The hybrid model is favoured as it combines the advantages of the other two models by letting intermediaries play their part while letting the Central bank store all the necessary information that can be accessed in an emergency. China’s E-CNY follows a hybrid design.
An important aspect of the 2-tiered design is the competitive playing field for intermediaries that the Central bank must ensure to safeguard the financial health of the economy. To guarantee that CBDCs are not used as a store of value (especially during a crisis) and consumer bank balances are maximized (as it is today), CBDCs could offer zero interest or lower interest than bank holdings, or even negative interest to restrict the amount of CBDC holdings (just like Central banks aim to minimizing cash hoarding today). In account based CBDC systems, restrictions could also be imposed on the amount of CBDC that can be held in consumer accounts.
The technology infrastructure of CBDC will depend on the architecture selected. The direct rCBDC design places huge responsibility on the Central bank that will require massive technological capabilities. By contrast the intermediated design will be similar to the current scenario, with banks and PSPs taking over the load of retail payments. The Hybrid model introduces a level of complexity to the intermediated model, as it requires Central banks to maintain all retail balances as well.
The technology chosen can either be a conventional centralized database or a decentralized DLT. The two technologies differ in the way transactions are validated and updated. Centralized databases usually have different nodes that are controlled by the central or top node. The DLT system is most likely to be a permissioned-DLT where authentication is done only by designated node operators. This eliminates any illegal activities associated with permissionless DLT, like those used for Bitcoin and other cryptocurrencies. However, maintenance of a permissioned DLT can be costlier and more time-consuming, and hence less favoured than a conventional design. This would mean that DLT would not be optimal for the direct rCBDC design, especially in large countries like China that are in any case sceptical about DLT being able to handle huge loads such as ‘Single’s Day’ shopping surges. However, DLT could be used in an indirect rCBDC for wholesale payments, as these would be far less than the retail load – this already been successfully tested by existing blockchain platforms. DLT could also be chosen for smaller areas. Intermediaries too could choose to use either technology or both for their operations within the same CBDC jurisdiction.
CBDC technology must allow privacy while enabling implementation of financial regulations and standards.
For intermediated or hybrid architectures, the infrastructure can either be managed by the Central bank or be delegated to technology suppliers (e.g., NZIA in the Bahamas, Accenture in Sweden). However, some countries like China and Canada who have decided not to rely on technology vendors, are building these systems themselves to have full control over them.
Jamaica, Ecuador, and Uruguay are among the few countries that had been taking the conventional database approach (Jamaica has already launched). Brazil, Nigeria, Eastern Caribbean Islands, Sweden, S. Korea, Saudi Arabia and the UAE are taking the DLT approach (Nigeria and the Caribbeans have already launched). Many others like Russia, China, Turkey, Bahrain, Canada and the Bahamas have so far opted for both.
Central banks can either issue rCBDC directly or as is more likely, through designated banks (as in China’s pilots). The digital code for each CBDC unit has a unique identity that cannot be changed (just like a rupee note or coin) and can be held in a digital wallet and transferred to other people’s CBDC wallets or be used to scan in-store codes for purchases.
Fixed denominations for token-based CBDC (as we now have for physical notes and coins) are generally favoured over minimum-value based CBDC as they would be more trusted by the public and easier to process
Discussions on CDBC wallets revolve around two types: (i) a conventional hosted CBDC smartphone wallet app that can be designed by banks, FinTech or payments players, and/or (ii) a hardware wallet in the form of biometric smart cards or wearables that can be loaded with CBDC and used offline. The smartphone wallet app can be used to track expenditure, set spend limits, or link credit and debit cards – just as we do with our mobile wallets today. The smart card/other hardware can be issued by banks and others even to unbanked customers who cannot afford smartphones – furthering the financial inclusion agenda in emerging economies.
To offer varying degrees of anonymity and access, the wallets could be linked with different KYC levels – the strength of which would determine transaction limits.
With globalization comes an important need that Central banks must consider while planning a CBDC: cross-border payments. Legacy technology platforms with limited operating hours, non-standard message formats, complex compliance checks, and long transaction chains causing incessant delays plague cross-border payments today. CBDC solutions can minimize these challenges.
Linkages with other CBDCs could be arranged at a retail level or wholesale level. Retail linkages can enable consumers to hold multiple foreign CBDC balances by allowing them to buy foreign exchange directly instead of going thru intermediaries that charge fees – just like we do today with forex cash before travelling abroad. This will be easier to implement with a token-based design that foreigners also can use. However, account based systems would need to collaborate with their foreign counterparts to implement the same using national digital IDs that CBDC regimes must accept. The account-based cross-border design may not be easy to implement but it is a better choice to limit ‘currency substitution’ (such as ‘dollarization’ where the dollar has displaced the local currency in many countries).
Linkages between existing wholesale systems could also be arranged if a Central bank can hold local CBDC accounts of domestic and foreign banks that operate within its jurisdiction. A cross-border payment would then simply involve a transfer from one bank’s CBDC account to that of the beneficiary.
If two Central banks in two countries allowed commercial banks to hold CBDC accounts in both regimes, then a bank can send payments from its own domestic CBDC account to its foreign CBDC account, which would then make it easy to transfer the same to the beneficiary bank’s CBDC account in the other regime.
A Multi-CBDC arrangement that joins multiple CBDCs to operate across borders is an interesting possibility and can make cross-border payments even more efficient. The BIS Innovation Hub, together with the Central banks of Hong Kong, Thailand, China and the UAE are working together to build a prototype platform, called ‘mBridge’ a wCBDC project that explores the capabilities of DLT to support multi-currency cross-border payments. This phase of the project has recently been declared successful by the BIS with cross-border payments worth $22 having been transacted on the mBridge test platform.
Considering the extensive ongoing work on CBDCs around the world, it is expected that sooner rather than later, they will become a reality in most countries – although they may not completely replace cash anytime soon. The overarching goal will be to serve the public interest through fast, secure and convenient ways to pay.
While the CBDC architecture, technology and access types may differ depending on geopolitical and other factors, one hopes that more and more countries will work on linking their domestic CBDC systems to make multi-CBDC a reality and cross-border payments a breeze. CBDCs are here and will be the next big game changer in payments.
- Central Bank Digital Currency Tracker – Atlantic Council
- Rise of the central bank digital currencies: drivers, approaches and technologies – BIS Working Papers, No 880, August 2020
- EY article on CBDC – May 2022
- The technology of retail central bank digital currency – BIS Quarterly Review, March 2020
- CBDCs: an opportunity for the monetary system – BIS Annual Economic Report 2021
- Central Bank Digital Currency (CBDC): Wholesale CBDC Global Developments – Payments Canada, 2022
- Behind the Scenes of Central Bank Digital Currency – Emerging Trends, Insights, and Policy Lessons – IMF, February 2022
- Are Central Bank Digital Currencies (CBDCs) the money of tomorrow – Deloitte, 2020
- Central bank digital currencies for cross-border payments – Report to the G20 -BIS, July 2021
- Progress of Research & Development of E-CNY in China – People’s Bank of China, July 2021
- Concept Note on CBDC – RBI, October 2022