The debate over UPI MDR has moved beyond whether merchants should pay a charge for accepting digital payments. A study by IIT Bombay proposes a new funding structure for the payment system, including a lower merchant discount rate for selected transactions, a revised share for payment companies and banks, and a larger contribution from high-turnover businesses. The recommendations place the financial sustainability of India’s most widely used digital payment network at the centre of a policy debate that will affect the architecture of urban commerce.
UPI has become part of the everyday operating system of Indian cities. A customer can pay a tea stall, a neighbourhood grocer, a shopping mall or an online retailer by scanning a QR code. That apparent simplicity depends on a network of banks, third-party app providers and other payment service providers. The central question raised by the IIT Bombay study is not whether UPI is convenient, but how the infrastructure behind that convenience should be paid for as transaction volumes and use cases expand.
The study, titled “UPI at a Crossroads: Reintroducing the MDR”, has been authored by IIT Bombay professor Ashish Das and former Reserve Bank of India official Pragya Das, according to the report. It recommends that the MDR on certain merchant transactions be reduced from the declared rate of 0.40 per cent to 0.08 per cent. The proposed rate would be divided between third-party app providers and payment service providers, which are primarily banks.
Under the proposed structure, third-party app providers such as PhonePe and Google Pay would receive 0.06 per cent, while payment service providers would receive 0.02 per cent for their services. The study also recommends ending the share of MDR currently proposed for banks. This is based on its argument that banks already earn from the deposits that support the banking system and should contribute to the cost of maintaining and improving UPI through that income stream.
The report cited by Aaj Tak Business says banks’ net interest margin from current account and savings account deposits was more than Rs 4.85 lakh crore in the financial year 2025-26. The study proposes using about 3 per cent of funds linked to CASA deposits—approximately Rs 15,000 crore—for the UPI system. It also recommends treating UPI as an essential service for banking businesses and assigning banks a clearer responsibility for maintaining and expanding digital-payment infrastructure.
This proposal shifts the funding conversation from a direct transaction charge to a broader institutional contribution. In the model described in the study, banks would not depend on their MDR share to recover the cost of participating in UPI. Instead, the system would be supported partly through the economic value banks derive from their deposit base. The recommendation also reflects the fact that UPI is not a standalone app service. It is connected to bank accounts, merchant acceptance, payment processing and the wider digital infrastructure through which money moves.
The study also places a greater funding responsibility on large businesses. It suggests that businesses with annual turnover of Rs 50 crore or more could be required to accept UPI payments and could be charged an appropriate digital-payment fee on online merchant transactions. This would create a distinction between everyday digital payments and the commercial transactions of larger enterprises that are better positioned to absorb or account for payment-processing costs.
That distinction matters because the existing UPI ecosystem serves very different users. A small street vendor, a local shop and a large online platform may all display a QR code, but their transaction volumes, margins and payment requirements are not comparable. A uniform approach to payment charges can therefore produce different effects across the merchant economy. The study’s proposed structure attempts to recognise this by directing more of the cost recovery towards large businesses while keeping ordinary person-to-person payments outside the proposed charge framework.
The report says the proposed MDR would be ad valorem, meaning that it would be linked to the value of the transaction. It recommends a maximum charge of Rs 40 per transaction. The study also calls for consideration of a lower MDR for credit cards so that the difference in charges between digital-payment methods is narrowed. Other recommendations include reconsidering the existing limit on free cash withdrawals from ATMs and continuing the Payment Infrastructure Development Fund to help bring more merchants into digital payments.
The policy context is already changing. According to the report, a government arrangement announced on September 15, 2026, set a 0.40 per cent MDR for selected merchant UPI transactions. The charge is to apply to merchants and not be passed on to customers. Person-to-person payments and specified exempt categories are outside the arrangement. The IIT Bombay study’s lower 0.08 per cent proposal therefore enters a policy environment in which the principle of charging for some merchant transactions has already become part of the discussion, even though the final structure remains unsettled.
The most important unresolved issue is classification. The payment industry has argued that UPI transactions cannot all be governed by one simple rule because the system supports several categories of use. Utility payments, loan repayments and capital-market transactions may have different commercial characteristics and regulatory requirements. Companies linked to the share market have questioned the logic of applying an MDR cost when a customer adds money to a brokerage account, arguing that the transaction may not generate direct income for the company.
These distinctions reveal the administrative challenge behind the debate. UPI is often experienced by users as a single payment method, but its institutional operation includes multiple categories of transactions. Some payments could previously have been made through net banking, IMPS, NEFT or RTGS, while others are linked to merchant purchases, financial services or account funding. A charge that is appropriate for one category may not be appropriate for another. Clear rules would therefore be necessary before any revised structure can be implemented across the network.
The timing of implementation has become another point of contention. Fintech companies and business groups are seeking an extension of the October 15 deadline to January 2027. Their stated concern is that the rules for different transaction types are not sufficiently clear and that banks, aggregators and merchants need more time to prepare. Industry representatives have also argued against introducing the system during the festive season, when transaction volumes and business activity are typically especially important to retailers.
The deadline question is not simply an operational matter. Payment rules have to be reflected in contracts, technology systems, merchant agreements and settlement processes. When different categories of UPI payments carry different treatment, participants need to know how each transaction will be identified, who will bear the cost and how the charge will be calculated. The demand for more time indicates that implementation depends as much on administrative clarity as on the headline rate itself.
For ordinary users, the study does not recommend a direct charge on routine UPI payments. The report says the aim is to create a sustainable way to fund the system without imposing charges on everyday payments. Under the government arrangement described in the report, the MDR applies to selected merchants and cannot be passed on to customers. That distinction is central to the public impact of the proposal: the immediate question for users is not whether UPI will remain free at the point of payment, but whether the cost will be distributed among banks, payment companies and businesses without weakening acceptance.
The study’s recommendations also raise a broader governance question. UPI has functioned as shared digital infrastructure, but its operating costs are distributed across institutions with different business models. Banks provide account and payment infrastructure, third-party app providers manage user-facing services, payment service providers support processing, and merchants depend on acceptance to reach customers. Deciding how each participant should contribute requires more than setting a percentage; it requires a clear allocation of institutional responsibility.
The evidence supplied in the report confirms that the debate is still at the proposal stage. The IIT Bombay study has offered a specific alternative: reduce MDR on selected merchant transactions to 0.08 per cent, allocate 0.06 per cent to third-party app providers and 0.02 per cent to payment service providers, remove the proposed bank share, and use part of banks’ CASA-linked earnings to support UPI. It has also proposed a larger role for high-turnover businesses and a clearer classification of transaction types.
What remains undecided is whether the government and relevant regulators will accept any of these recommendations, how the final categories will be defined and whether the implementation deadline will move to January 2027. Those decisions will determine whether UPI’s next phase is funded through merchant charges, institutional contributions, or a combination of both. For India’s cities and businesses, the outcome will shape the costs and responsibilities attached to the digital infrastructure that now supports everyday commerce.
