RBI's Digital Rupee Moves Closer to Reality (#58)
Reserve bank begins CBDC pilots, and debates Digital Rupee's (e₹) design and use cases
Welcome to the 58th issue of Unit Economics. I am back writing after a long work and travel break, and for today’s write up - we will go deeper into the fascinating beginnings of Digital Rupee. Dive in!
How did we get here?
Money has a long history. It took centuries to collectively move from a barter to a physical form. Initially, physical money represented value through commodities that were deemed to have intrinsic value. Slowly, we graduated to forms of money that were more representative in nature, from commodities to metallic coinage. Later, the monetary policy function strengthened with the development of sophisticated banking systems, which led to the popularization of banknotes.
Paper money and coinage gave nation-states control of the money supply, and established centrally controlled monetary systems. This meant that the ruling governments and central banks took upon themselves the sovereign function of currency issuance. And with banks, the forms of money evolved further towards electronic payment systems of cards, wallets, and bank transfers.
Today, cash is estimated to drive ~18% of all transaction value – going stronger at point-of-sale transactions. India, particularly, has been exemplary in the development of an electronic payment ecosystem through real-time bank transfers, cards, and prepaid instruments. Yet, nearly half of all transactions continue to be in cash in the country.
But physical cash is both operationally and socially costly, hinders innovation, and lacks efficiency. And these limitations are known. So why is cash still popular? Some dialogue and analysis suggest that absolute anonymity, wider accessibility, and the assurance of finality of settlement at the point of purchase justify the strength of cash for consumers.
This justifies and leads us to the potential counterpart of cash: CBDCs.
I had written on CBDCs on a couple of occasions (here and here) and had captured how widespread the experiments on digital cash had gotten. This time, the fervour has reached India.
Until three years back, the RBI had hardly given thought to the idea of digital central bank money. But the developing optimism around the CBDC experiments, as rigorously covered by the Bank of International Settlements (BIS), likely forced the central bank to look more keenly at the prospect. Keenly enough that the RBI launched the first pilot of one type of Digital Rupee (e₹-W) on 1st Nov, and is at the final stages of the rollout for the other (e₹-R) already.
The RBI had given us some insights into how it was thinking about the Digital Rupee last month in the concept note it published. We will take the rest of the space to mull over it here.
But to start with,
What is Digital Rupee (e₹)?
The Reserve Bank defines CBDC “as the legal tender issued by a central bank in a digital form. It is the same as a sovereign currency and is exchangeable one-to-one at par (1:1) with the fiat currency”. That is, it is central bank money, only digital.
The Digital Rupee (e₹) can be issued in two forms:
Wholesale Digital Rupee (e₹-W): designed for settlement of interbank transfers and related wholesale transactions by select financial institutions
General Purpose or Retail Digital Rupee (e₹-R): an electronic form of cash meant primarily for retail transactions by the public, to be made interoperable with other payment platforms and stored in an e-wallet on-demand
The RBI conducted its first CBDC pilot for e₹-W on 1st Nov and saw the participation of nine banks, who cumulatively tested 48 transactions worth ₹275 crores. At its core, the e₹-W is expected to lessen the settlement risk in the inter-bank market and improve its efficiency “by pre-empting the need for settlement guarantee infrastructure or for collateral”. The adoption of e₹-W would largely depend on (A) the development of technical architecture, including e₹-W’s integration with RTGS, and (B) the savings it brings to the settlement costs, including for liquidity, margin funding, etc.
The RBI suggests that the pilots, in the following months, may explore the use-case of cross-border payments, where today - central bank acts as a central counterparty, and of providing access of G-Secs, CPs/CDs to retail participants by bypassing the bank-account route, among other potential use-cases.
On the other hand, the pilot for e₹-R is in the final stages of the rollout – with eight banks participating in the trial. The expectation, it appears, is that each bank would do a closed user group testing with 10,000-50,000 users, who would perform P2M transactions in select locations. The testing, it is believed, will help give insights into the many design considerations for digital currency. What would these be?
Design Considerations of the Digital Rupee (e₹)
Any design choice for the Digital Rupee (e₹) must exist without interfering with the features of the currency. The features suggest that the Digital Rupee
Will be a sovereign currency issued by the central bank
Will appear as a liability on the RBI’s balance sheet, as opposed to existing forms of digital money that appear as liabilities with commercial banks
Must be accepted as a medium of payment, legal tender, and a safe store of value by all citizens, enterprises, and government agencies
Will be freely convertible against commercial bank money and cash,
Will be fungible legal tender, not requiring the holders to have a bank account, and
Should help lower the cost of issuance of money and transactions
Since we have already understood the two types of CBDCs, we are left with five critical design considerations for the currency.
1. Who administers the CBDC?
The question seeks to understand the respective roles of the central bank and the private sector in facilitating access and use of e₹. There are three models that the RBI considered:
Direct model (Single-Tier): this would be a direct CBDC system wherein the central bank would be responsible for all functions, including KYC, issuance, account-keeping, transaction verification, etc.
Indirect model (Two-Tier): consumers hold their e₹ in an account/wallet with a bank or another regulated intermediary. The obligation for issuance of e₹ to end-consumers would fall on intermediaries, while the RBI would be tasked with tracking wholesale balances and matching them against retail balances with consumers.
Hybrid model (Two-Tier): in this model, commercial intermediaries (payment service providers) are held responsible for all customer-associated activities, while the central bank maintains a ledger of retail transactions.
The indirect model is, expectedly, deemed the most receptive of the three, due to its parallels to the existing money supply model – causing the least disruption to the financial sector. The authorised intermediaries, the RBI suggests, will be termed as Token Service Providers (TSPs) and will be responsible for (A) the issuance of e₹ to end customers, (B) account-keeping services, (C) customer verification via AML/CFT checks, and (D) transaction verification.
2. Should CBDCs bear interest?
There are two possibilities here: e₹ bears interest (remunerated), or it does not (non-remunerated). If e₹ were to be remunerated with a positive interest rate, there would be
More e₹-R adoption due to the higher attractiveness of the Digital Rupee as a store of value,
Better pass-through of the repo rate changes from RBI to the economy, strengthening the effectiveness of monetary policy
But the other side can equally turn ugly rather quickly.
A deposit-like e₹ would move away from cash and cause disintermediation in financial systems as consumers migrate their deposits to the digital central bank money – causing ripples in the credit creation ability of banks, eventually putting pressure on their net interest incomes
Further, banks would be required to maintain additional liquidity buffers to support e₹ demand, mandating higher upkeep of collaterals
This is a tricky decision. While the non-remunerated e₹ would be (A) closer to cash and (B) least disruptive to the existing financial system, the remunerated e₹ offers an opportunity for more effective monetary policy – allowing the monetary policy to respond more powerfully to the economic conditions. To counter disintermediation and associated risks of interest-bearing e₹, the interest rate may be kept lower than bank-deposit rates, and/or conversion limits may be set on an account-level.
To start with, it may make sense to launch with a non-remunerated e₹ to limit the disruption to existing systems, while migrating to lower-than-bank interest rates in the longer term.
3. Should the e₹ be account-based or token-based?
A token-based e₹ would act similarly to a banknote (bearer-instrument), implying that whoever ‘holds’ the token at a given point would be presumed to own them. The system would emulate a paper currency transaction, wherein merchants at a retail store need only validate the paper currency handed to them – not the authenticity of the payer.
An account-based e₹ instead would require verification of the individual or entity’s identity prior to the creation of an account, and the keeping of a record of balances and transactions of all holders of the CBDC to indicate ownership of balances.
Transactions over account-based e₹ would be rapid and secure once an account is successfully verified. In contrast, in a token-based e₹, ownership of each token must be stored in a secure ledger and new transactions must be collected into blocks and verified before their addition to the ledger – likely increasing the settlement latency.
e₹-W supports the account-based model due to its (A) need for instant settlement, and (B) legal status of wholesale bank accounts. For e₹-R, the RBI suggests, a token-based e₹ would offer more relatability to physical cash as it would be withdrawable from banks as we withdraw physical cash.
4. What degree of privacy should apply to CBDCs?
The anonymity of e₹-R is crucial for it to resemble physical money. However, while complete anonymity would enable maximum adoption of the e₹, its potential for illegal transactions may undermine KYC measures.
This suggests that it may perhaps be best to offer consumers intermediate levels of privacy instead of complete anonymity. A suggestion of Managed Autonomy argues for anonymity in small-ticket size transactions and traceability of higher-value transactions, which may be possible through the issuance of anonymity vouchers for small-values.
5. Fixed or minimum value denominations?
The RBI, in its efforts to emulate physical cash, goes further to suggest e₹ be offered in fixed denominations of ₹50, ₹100, ₹200, etc.
Issuance of physical cash in fixed denominations makes sense operationally. However, restricting e₹ to the same would (A) restrict the flexibility of consumers that are used to making non-fixed quantity digital payments via UPI, and (B) complicate P2M transactions that require settlement of an amount different from those available in fixed denominations – since the concept of giving back change in a digital transaction would imply an additional and a rather unnecessary transaction.
While the decision is awaited (and keeping aside the consideration of technical performance issues), my vote here goes for a min-value-based denomination system for the flexibility it would provide.
The Way Forward
The RBI is transparently working towards a phased implementation strategy, testing the idea of an account-based e₹-W and a token-based e₹-R through various stages of pilots. The operationally controlled environment will emulate multiple test cases and help decide the currency design and deployment plan, post which the pilots will be deployed at a larger scale – before finalizing the design and taking the call on the full public launch. The onus would continue to remain on causing minimal to no disruption of the financial system.
Early thoughts on the design suggest a Retail Digital Rupee (e₹-R) that would be issued indirectly via a two-tier model and would sit along the existing payment systems by being non-remunerative, token-based, moderately anonymous, and interoperable. On the other hand, the Wholesale Digital Rupee (e₹-W) would likely differ by being issued directly, adopting an account-based and non-anonymous form.
The functionalities and operations of the Digital Rupee (e₹) will extend beyond to include multiple other considerations, including:
The distribution of responsibility and the architecture for customer due diligence at the point of onboarding, and for transaction verification.
The ownership and creation of CBDCs, on whether the RBI would require a separate technical subsidiary to handle the responsibility
The offline capability of the e₹, given that more than 40% of the population continues to lack access to the internet and a higher percentage struggle with acceptable network speed. The VISA protocol for Offline Payment System (OPS) provides direction for this use case.
Each of these is a structural decision meant to redesign a form of money that has remained fundamental to the nation’s psyche: physical cash. There is little evidence or motive to suggest that e₹ would substitute physical cash, and it would go a long way if e₹ could emulate the same level of trust, safety, liquidity, and settlement finality. This, however, appears a well-thought-out beginning to the idea, and we will keep revisiting the topic with more notifications, pilots, and circulars.
If you have any views or feedback to share on the topic, feel free to add a response below or to share your thoughts with me over Linkedin. In case you feel your friends or family would be interested in reading about payments, feel free to share the blog with them as well. See you in a couple of weeks!