The debate over UPI MDR has moved beyond whether merchants should pay for large-value transactions. A report by IIT-Bombay professor Ashish Das and former RBI official Pragya Das argues that the payment system’s next funding model should be built around the economics of banking, public digital infrastructure and merchant scale rather than a broad revenue share for banks.
The paper, titled “UPI at a Crossroads: Reintroducing the MDR”, proposes reducing the merchant discount rate on specified merchant UPI transactions from 0.40% to 0.08%. It recommends a 0.06% share for third-party application providers such as PhonePe and Google Pay, while retaining a 0.02% share for payment service providers, predominantly banks. Its more consequential proposal is to remove the existing 70% revenue share received collectively by issuing and acquiring banks.
That argument places UPI in an unusual institutional category. It is operated by the National Payments Corporation of India but functions as shared digital infrastructure used by banks, payment applications, merchants and consumers. The report’s central question is therefore not simply how much a merchant should pay. It is who should finance a system that has been designed to make digital transactions widely available and, for most users, free.
The report argues that banks already derive substantial value from current and savings account deposits, known as CASA deposits. According to the study, banks recorded a net interest margin of more than Rs 4.85 lakh crore from CASA deposits in FY26. The authors describe this as large and disproportionate compensation for providing a limited range of services at no charge.
“Banks earn substantial interest margins from the CASA depositors. They should use the same towards running and improving the UPI ecosystem and not rely on MDR revenue sharing from offline transactions,” the report said, according to PTI as quoted in the Times of India report.
This is the report’s most important institutional intervention. Rather than treating every UPI transaction as a separate commercial service that must generate a fee, it treats the payment network as an ecosystem supported by the wider banking relationship. Under this view, banks’ low-cost deposits are not unrelated to UPI. The accounts that hold customer money also help banks fund their operations, while UPI gives those accounts a convenient transaction layer.
The paper proposes that banks could set aside around 3% of the funds that CASA depositors forgo through lower interest earnings. It estimates that this could amount to about Rs 15,000 crore for operating and upgrading the UPI ecosystem. The proposal would shift the funding burden away from a transaction-level revenue share and towards the broader balance-sheet benefits banks receive from deposit accounts.
The report also asks the Reserve Bank of India and the government to direct banks to treat UPI as an essential service that must be offered to remain in the banking business. It compares this approach with the adoption of Core Banking Solutions, which eventually became a basic requirement for banks. This comparison is significant because it frames UPI not as an optional add-on supplied by individual institutions, but as a foundational banking capability.
The immediate policy backdrop is an NPCI circular issued on September 15, according to the report. The circular introduced a 0.4% MDR for specified person-to-merchant UPI transactions above Rs 2,000. Merchants, rather than consumers, would bear the fee. For transactions of Rs 75,000 or more, the charge would be capped at Rs 300.
The arrangement does not apply uniformly to all UPI payments. Transfers between individuals and most routine merchant payments would remain free. Transactions above Rs 2,000 for essential services, including railways, telecommunications, fuel and insurance, would attract a flat Rs 5 charge. Capital market transactions, including mutual fund investments and stockbroking, would carry a lower 0.02% MDR capped at Rs 300.
The report’s objection is partly about the structure of these charges and partly about their relationship to the cost of providing the service. It questions whether a charge of Rs 300 on a peer-to-merchant transaction worth Rs 75,000 or more is reasonable. It also cites RBI regulations requiring service charges to remain broadly in line with the average cost of providing the service.
This matters because an ad valorem fee, calculated as a percentage of transaction value, can increase the charge even when the underlying payment process does not become proportionately more expensive. The report therefore recommends that if such a model is adopted for UPI, the maximum charge should be limited to Rs 40 per transaction.
The proposal also tries to separate merchant capacity from merchant vulnerability. It recommends making UPI mandatory for merchants with annual turnover of Rs 50 crore or more, while limiting charges to merchants above that threshold. The report equates the threshold to roughly Rs 4 crore a month and says such a structure could account for approximately 90% of the MDR proposed by NPCI.
For smaller merchants, the distinction could preserve the practical advantage of UPI: a low-friction payment option without a charge that materially affects thin margins. For larger online merchants, the report suggests a different funding route. It recommends introducing a “digital payment fee” on online merchant transactions involving goods or services purchased from e-commerce companies.
That recommendation recognises that the payment ecosystem is not used in the same way across all merchant categories. A neighbourhood shop, an online platform and a large service provider may all accept UPI, but they have different transaction volumes, bargaining power and ability to absorb fees. A uniform MDR can therefore produce different consequences depending on the merchant’s scale and business model.
The report also addresses the relationship between UPI and credit cards. It argues that credit card users should not place an additional burden on person-to-merchant UPI users and recommends lowering credit card MDR. It questions the rationale for applying a credit-card MDR structure to UPI, saying that the two payment systems do not necessarily have the same economic model.
This distinction is central to the public infrastructure question. UPI transactions generally move through bank accounts, while credit-card payments involve a separate system of credit, risk and settlement. The report’s position is that the existence of a familiar fee model in one payment network does not establish that the same model is appropriate for another.
The study also recommends reconsidering generous fee-free ATM withdrawal allowances, arguing that such allowances are harder to justify when consumers have access to convenient digital payment options. It supports continuing the Payments Infrastructure Development Fund, saying an extension would help payment aggregators bring more merchants onto digital payment platforms.
These recommendations show that the paper is not arguing for the complete absence of payment-system revenue. Instead, it proposes a different allocation of costs. Large merchants and online commerce could contribute directly; banks could fund the system through the value they derive from CASA deposits; payment applications and service providers could receive a narrower compensation share; and public support could continue through the infrastructure fund.
The policy challenge is that UPI’s success has created both scale and dependence. The platform is described in the report as public digital infrastructure that has made payments more convenient and improved economic productivity. At the same time, the absence of a widely accepted revenue model has left policymakers looking for ways to support its operation and expansion without weakening user adoption.
The proposed 0.4% MDR framework is intended to establish a sustainable revenue model for the digital payments ecosystem, according to the NPCI circular described in the report. The IIT-Bombay paper warns that the framework, if introduced in its current form, could weaken public confidence in UPI and make the payment system a source of dissatisfaction.
That concern is less about a single transaction charge than about how costs become visible at the point of use. UPI’s public value has been built partly on its simplicity: users can make payments without confronting a separate fee in most routine cases. Charges imposed on merchants can eventually affect prices, acceptance decisions or the willingness of businesses to offer particular payment options, even when consumers are not charged directly.
The report’s alternative is to preserve the user-facing simplicity of UPI while making the funding arrangements more explicit and differentiated. Its proposals would require decisions by NPCI, the RBI, the government, banks, payment applications and merchants. They would also require clarity on which transactions are treated as essential, how charges are calculated and how funds for system maintenance are collected and governed.
What the evidence supplied in the report establishes is that India’s UPI debate has entered a new phase. The issue is no longer only whether digital payments should remain free for consumers. It is whether the infrastructure supporting that convenience should be financed through transaction charges, bank deposit economics, large-merchant contributions, public funds or a combination of all four.
The next stage will depend on how NPCI, the RBI and the government respond to the competing proposals. For Urban India, the outcome will determine whether UPI remains a broadly accessible public utility, becomes a more differentiated commercial service, or adopts a hybrid model in which the largest participants carry most of the cost.
