The Unified Payments Interface (UPI) is one of India’s most successful pieces of economic infrastructure. In July, it handled nearly 24 billion transactions worth almost ₹30 trillion. That scale and ubiquity make UPI more than a convenient payments product. It is an essential part of India’s digital public infrastructure (DPI).
Hence, the legislative change opening the door to a merchant discount rate, or MDR, deserves scrutiny. The amendment does not itself impose an MDR. But it removes the protection that had kept MDR at zero for UPI and RuPay payments, leaving future charges to executive decisions.
The finance minister has assured the country that any MDR would apply only to merchants and not consumers. It was needed to fund investments in infrastructure, innovation and security.
Banks, payment service providers and the National Payments Corporation of India (NPCI), which operates UPI, spend money on servers, cybersecurity, fraud detection, dispute resolution and settlement. That is estimated to cost the industry about ₹20,000 crore annually. Government reimbursement covers only a fraction.
The costs of UPI cannot be wished away and banks and payment companies should not be expected to subsidize UPI indefinitely. But that does not settle the question of who should pay.
There is an important distinction between cost and price. Many types of public infrastructure are costly, but we don’t collect a toll at every point of use. That’s because the benefits spill over to society at large. UPI has strong public-good characteristics, large positive externalities and powerful network effects. Its value rises as more people and merchants use it.
Zero MDR was an important reason for UPI’s rapid spread. It made digital payment resemble cash. More than 85% of person-to-merchant transactions are small ticket. A payment of ₹20 for tea or ₹150 to an auto driver is made without either party calculating a transaction cost. Such small merchants operate on thin margins and consumers make several such payments every day.
That MDR will be paid by the merchant and will not affect consumers is incorrect logic. It can either be absorbed or passed on through higher prices. The economic incidence of any levy does not necessarily fall on the person on whom it is formally levied.
India has some experience with tiny transaction levies. Recall the 0.1% Banking Cash Transaction Tax introduced in 2005 and withdrawn in April 2009. The analogy is not exact, but the lesson is useful. A tiny fractional cost may appear negligible but influences behaviour when repeatedly imposed. UPI was kept free to nudge movement away from cash. Let’s keep it that way.
UPI’s benefits go far beyond mere convenience. In semi-urban and rural areas, growing dependence on digital payments creates pressure to improve telecom connectivity. A weak mobile signal that prevents a shopkeeper from receiving payment is an economic bottleneck. Its increasing usage, therefore, becomes a pressure point for improving the country’s telecom network coverage itself.
It similarly promotes digital literacy. Someone who learns to scan a QR code, verify the recipient, authenticate a payment and check whether money has arrived gets more comfortable with the digital economy. That familiarity can extend to banking, e-commerce and accessing government services.
Compare UPI with Aadhaar, another foundational digital infrastructure. Authentication by Aadhaar is easy, widely accessible and free. We don’t impose a toll on every Aadhaar authentication merely because this process has a cost. UPI should be approached with a similar philosophy. The social benefit of both come from ubiquity and low friction.
Currency, an alternative to UPI, has to be printed, transported, sorted, stored, guarded and replenished. Banks have to maintain ATMs, branches and staff. Digital payments additionally encourage business formalization and improve transaction trails for tax compliance. Hence, the net social cost of UPI versus the alternatives should factor this in.
There exist ways of paying UPI’s ₹20,000 crore annual cost.
The Reserve Bank of India’s (RBI) dividend to the Union government was ₹2.86 trillion in 2025-26. The cost of UPI is just 7% of that amount. A transparent reimbursement mechanism, linked to audited costs and efficiency standards, can compensate banks and payment providers without imposing charges on merchants.
When there are huge social benefits from digital infrastructure, the cost should not be recovered at every point of usage.
Is UPI being made unfree under international pressure? The US Trade Representative’s 2026 report complained that India’s policies favour UPI and RuPay over foreign payment suppliers. They certainly have disrupted the merchant-fee model of global card networks. Yet, Google Pay and Walmart-controlled PhonePe together process more than 80% of UPI transactions. Foreign participation is hardly absent.
UPI is one of India’s most important contributions to global DPI. Countries are studying the India Stack because India demonstrated that identity, banking and interoperable payments can work at population scale. Its success came from reducing transaction friction.
The costs are real, but the gains are greater—from financial inclusion, formalization, digital literacy, better connectivity and economic efficiency. Let’s keep UPI as public infrastructure rather than just another commercial payments service.
The author is senior fellow with Pune International Centre.