Banking & Capital Markets Systems · Chapter 10 of 13
The digital rupee, and what its own numbers say
India's central bank digital currency is a live pilot with published figures — which makes it a rare chance to see adoption without the marketing.
These three chapters are the banking-side record. If the mechanics are unfamiliar — what a hash chain is, what consensus costs, how a smart contract works and how the money actually gets stolen — the companion book Blockchain, Honestly covers all of it from first principles, and its four-question test is what these chapters apply.
A central bank digital currency (CBDC) is money issued directly by the central bank in digital form. That last part is the whole point, so it is worth being exact about what is different.
What is actually different from the money in your UPI app
The balance in your bank account is a claim on your bank. If the bank fails, you are a creditor, and deposit insurance covers you up to a limit.
A note in your pocket is a direct liability of the Reserve Bank. No intermediary can fail underneath it.
The digital rupee (e₹) is the second kind, in digital form. The bank distributes it and runs the wallet, but it is not a deposit and it does not sit on the bank's balance sheet.
UPI is not competing money — it is a fast way to move the first kind. That distinction is the single most common confusion about CBDC, including inside banks.
India runs two pilots. Wholesale (e₹-W) launched 1 November 2022 with nine banks, settling secondary trades in government securities, and was later extended to the call money market. Retail (e₹-R) launched 1 December 2022 with four banks in four cities, reaching seventeen banks and more than twenty-six cities by March 2025.
The milestone worth understanding properly.
On 27 December 2023 the retail pilot crossed one million transactions in a day, which was widely reported as the moment adoption arrived.
RBI's own Deputy Governor later confirmed it was reached substantially because participating banks began paying part of employee benefits directly into staff CBDC wallets. Once that push eased, daily transactions fell about 90%, to roughly 100,000 by June 2024.
The milestone measured the incentive, not the demand. This is a pattern worth recognising far beyond CBDC — whenever a number is hit shortly after someone was given a reason to hit it, ask what happens when the reason stops.
The most recent dated primary figure: about 120 million cumulative transactions and over ₹28,000 crore, as of 5 December 2025.
RBI's response to the stall is itself informative: it changed distribution, not technology. In January 2025 MobiKwik and CRED became the first non-bank fintechs to offer e₹ wallets, sponsored by Yes Bank, with limits of ₹10,000 per transaction and ₹50,000 a day. The diagnosis was that people were not reaching it, not that the ledger was wrong.
Is the e₹ even a blockchain?
Less than you would assume, and RBI has been reasonably candid about it. The wholesale pilot is token-based; the retail pilot is described as having blockchain components, but it is a permissioned ledger with a small RBI-controlled participant set and banks as the only nodes. There is no mining, no public validation and no open participation.
Apply the four-question test from the blockchain book and it does not obviously pass: there is an accepted trusted operator, and it is the central bank. The Bank for International Settlements has said as much — a centralised ledger is likely superior for money except where rule of law or contract enforcement is weak. Which is not a criticism of the e₹. It is a reason to look for the capability that does not exist on the old rails, because that is where the case actually lives.
Programmability — the one genuinely new capability
Money on a programmable ledger can carry conditions in the money itself, rather than in a rule enforced afterwards by whoever holds the account.
The Ratnagiri example, which is worth carrying in your head
IndusInd Bank disbursed programmable e₹ to farmers in Ratnagiri, Maharashtra, restricted so it can only be spent on agricultural inputs.
The consequence is not the restriction. It is that the bank could extend credit without full collateral, because the money physically cannot be diverted — so the main reason for demanding collateral is gone.
About 50 farmers, aiming at roughly 1,000. Small. But this is a capability an ordinary payment rail does not have, and it changes a lending decision rather than a settlement time.
By 2026 there are roughly ten welfare-linked pilots of this kind. Be careful with the ‘$80 billion’ figure that circulates — that is India's total welfare budget, not an amount moved through CBDC. The source says so itself.
Cross-border, and why India sits out one of the projects
- Project Nexus is not a CBDC project, despite constant confusion. It is a Bank for International Settlements effort to interlink existing fast payment systems — UPI and its Southeast Asian equivalents — targeting transfers in about sixty seconds. RBI became a founding member on 30 June 2024, targeting 2026. No go-live has been announced.
- mBridge is genuine central-bank-to-central-bank CBDC settlement (China, Hong Kong, Thailand, the UAE, Saudi Arabia). India is an observer, not a participant.
- There is a real, non-political reason for that, and it is a good example of policy constraining architecture: the rupee has limited capital-account convertibility. A CBDC bridge that is fine for trade settlement could also open a capital-flow channel RBI is not ready to allow. The blocker is monetary policy, not technology.
Check yourself