The Reserve Bank of India’s digital rupee programme is now well past the trial phase and properly into the question of whether anyone actually uses it. The May 2026 picture is uneven, more interesting at the wholesale than the retail layer, and quietly different from the way it gets characterised in international coverage.

At the retail level, the digital rupee has not become the everyday payment instrument that initial framing suggested it might. UPI continues to dominate retail digital payment volume by an overwhelming margin, and the digital rupee retail token has not displaced it in any visible segment. Where retail adoption exists, it’s heavily concentrated in specific use cases — government disbursements to specific cohorts, certain merchant categories with regulatory advantages, and a small but real population of enthusiasts who prefer the instrument for specific reasons. The total retail volume remains a rounding error against UPI throughput.

This isn’t a failure exactly. It’s a clarification. The retail digital rupee was always going to compete with an exceptionally well-functioning instant payment rail, and the marginal benefit to a retail user of switching has been hard to demonstrate in practice. What the rollout has produced is the foundational infrastructure, the operational learning, and the institutional capacity. Whether retail adoption ever materialises at scale will depend on policy levers — interoperability requirements, programmability features that UPI doesn’t offer, specific use case mandates — rather than organic demand.

Wholesale CBDC is the more interesting story

At the wholesale level — bank-to-bank settlement, government securities markets, cross-border interbank flows — the digital rupee has produced more substantive activity. The integration with government securities settlement, in particular, has matured into routine usage that processes meaningful daily volume. The settlement finality benefits and the operational efficiency gains are real, even if they’re not headline-generating.

The cross-border angle has progressed less than the early signals suggested it might. Bilateral arrangements with central banks in trading partner economies have been announced and tested, but the move from pilot to routine operational use has been slower than the announcement timelines implied. The international plumbing for CBDC interoperability is harder than the domestic plumbing, and the political and regulatory complexity is substantial.

The wholesale settlement use case is where the next 18 to 24 months of meaningful expansion is most likely to come from. Whether that translates into a broader institutional shift, or remains a specialist tool for specific transaction types, is the open question.

What the banks are actually doing

Indian banks have invested significant capability into supporting the digital rupee, both at the wholesale and retail layers. The integration is technically functional — KYC layers, custody arrangements, transaction processing, regulatory reporting — at the major private and public sector banks. The bank-side capability isn’t the bottleneck.

What’s contested is how to commercialise the capability. The fee structures around digital rupee transactions are constrained by the same competitive dynamics that drove fees to zero in the UPI ecosystem. Banks have built infrastructure to support a payment method that doesn’t generate meaningful per-transaction revenue, betting on the broader value of being part of the rail infrastructure rather than the immediate economics. Whether that bet pays off depends on policy decisions that are still being worked through.

What’s quietly working

A few specific use cases have produced genuine traction:

Targeted government disbursements where programmability matters. The ability to issue digital rupee with constraints on use — for fertiliser purchases, for school meal programmes, for specific welfare categories — has been demonstrated in pilot and is being scaled in selected programmes. The administrative gains, the leakage reduction, and the targeting precision are all measurable. This is a use case where digital rupee genuinely beats UPI because UPI doesn’t natively support the programmability requirements.

Cross-border remittances at the wholesale layer. The corridor work between Indian banks and counterparties in specific markets has produced operational efficiency in trade settlement. The retail remittance corridor benefits remain mostly theoretical pending broader interoperability work.

Securities market settlement. The T+0 settlement for government securities using digital rupee has matured into routine practice and has produced measurable benefits in collateral velocity and operational risk. The use case is technical, not headline-grabbing, but it’s substantively important to the way the wholesale market functions.

What’s not happening

The dramatic adoption stories that would make the digital rupee a meaningful component of everyday Indian retail life have not materialised. The merchant category penetration has stayed shallow. The user adoption among UPI’s billion-plus users has stayed marginal. The expectation that programmable money would open up new commercial use cases has produced experiments but not at-scale adoption.

The honest reading is that retail CBDC, in a market with a highly successful instant payment infrastructure, is harder than the international CBDC discourse acknowledged. The Indian experience is genuinely useful data for other central banks considering similar projects in similar contexts.

What I’d watch

Three things over the next 12 months.

Policy moves on programmability mandates. If the central government or specific ministries mandate digital rupee usage for particular disbursement categories, the volume rises mechanically. Without policy push, organic retail adoption is unlikely to accelerate.

The cross-border interoperability work, particularly the BRICS-aligned corridor experiments. If these produce operational systems rather than pilot demonstrations, the wholesale story becomes substantially more interesting.

The interest-bearing question. The current digital rupee is non-interest-bearing, which limits its utility as a deposit substitute. If the policy framework changes to allow interest, the dynamics shift materially — both positively for adoption and negatively for the bank deposit base.

The honest summary: the RBI digital rupee in 2026 is a real and functioning system with meaningful but limited adoption. It’s neither the failure that some critics predicted nor the transformation that some advocates anticipated. It’s an infrastructure investment that may pay off over a longer horizon than the initial commentary assumed. The next two years will tell us which direction.