UPI MDR: Who Should Fund Digital Convenience?
Protecting everyday access while funding dependable payment acceptance. Our perspective connects precise merchant classification, transparent settlement and permitted revenue to measurable reliability and service.
Perspectives · The Banking Balancing Act
A success worth protecting
UPI has made a complex banking journey feel ordinary. A customer scans, confirms and expects the merchant to know the outcome almost immediately. That simplicity is valuable because it removes payment friction from everyday life.
It can also hide the work underneath. Banks, the National Payments Corporation of India (NPCI), payment service providers, third-party applications, acquirers and merchants support identity, routing, settlement, fraud controls, cybersecurity, reconciliation, complaint handling and recovery. “Free” is the customer experience. It is not a description of the operating cost.
The revised Merchant Discount Rate (MDR) framework announced on 15 September 2026 brings that distinction into public view. The Department of Financial Services says the finalised threshold structure takes effect on 15 October 2026. As of this article’s publication on 7 October, the announced rates are a readiness matter; activation is stated for 15 October. [1][2]
The constructive question is how India can fund dependable payment acceptance while protecting customer access, small-business participation and trust.
What the framework changes—and what it protects
The official announcement keeps person-to-person UPI transactions free. Person-to-merchant (P2M) payments up to ₹2,000 also remain free of MDR. Small merchants in the Person-to-Person-Merchant (P2PM) category, receiving up to ₹1 lakh per month through UPI QR codes, retain zero MDR on all transactions. Customers are not to pay MDR, merchants are not to pass it on, and UPI applications are prohibited from adding platform or hidden payment fees. [1][2]
For specified general P2M payments above ₹2,000, the announced rate is 0.4%, capped at ₹300 for transactions of ₹75,000 and above. Selected essential or thin-margin sectors use a flat ₹5 above ₹2,000. Capital-market transactions use 0.02%, capped at ₹300. The announcement also provides for an amount equivalent to 5% of MDR collections to support small-merchant adoption; the detailed fund framework was still to be finalised in consultation with the Reserve Bank of India (RBI) within three months of the September FAQ. [1][2]
These distinctions aim to protect everyday use and smaller merchants while asking a narrower group of commercial transactions to contribute. The PIB estimates that 96% of merchant transactions remain unaffected. That is a count of transactions; it does not establish the share of sales value or margin affected for an individual merchant. [1]
A broader view of payment economics
Merchants receive value through convenient acceptance, reduced cash handling, digital records and integration with commerce systems. Their contribution still needs to be assessed against their economics: a small percentage of sales can represent a meaningful share of gross margin.
A separate distinction is needed between gross MDR and bank profit. The official framework says MDR is distributed among ecosystem participants, including banks, payment service providers and application providers. Each participant carries different costs and incentives. An acquirer may fund merchant onboarding and support. A remitter bank authenticates and processes the payer side. The network coordinates routing and settlement. An application provider maintains the customer experience. The amount deducted from a merchant is therefore not automatically the amount retained by one bank. [1][3]
Recognising infrastructure costs is the beginning of a fairness assessment. It also requires examining who contributes, what they receive and how service obligations are funded. Transaction processing costs are not perfectly proportional to purchase value. A ₹50,000 instruction does not necessarily use twenty-five times the compute of a ₹2,000 instruction. Larger values may, however, bring different fraud, settlement, dispute and control exposures. Percentage pricing is partly a way to distribute funding, not a precise meter of server effort.
RBI’s 2022 discussion paper framed the enduring tension well: an efficient system needs affordable access and an appropriate return for operators. It also observed that payment providers incur costs to create secure systems, comply with requirements, acquire users and continue operating. The paper was consultative and did not itself establish the 2026 rates, but its economic questions remain relevant. [3]
A small rate can have different effects
Consider an illustrative ₹10,000 general-retail P2M sale after the stated effective date, with verified eligibility for the 0.4% rate. MDR would be ₹40. At a fictional gross margin of 2%, that is 20% of the ₹200 gross margin; at 20%, it is 2% of the ₹2,000 gross margin. This calculation excludes taxes and other charges, assumes no exemption, and uses invented margins rather than an industry benchmark. Gross margin is not net profit.
The same rate therefore creates different operating pressures. A merchant assessment should compare the full cost and value of each permitted payment route, including cash handling and reconciliation, without assuming those benefits always offset the deduction. The payment firm’s assessment needs a separate view of its retained share, shared costs and service commitments. Neither perspective can be inferred from the headline percentage alone.
The operating dependencies behind one rate
Classification is a financial control
The correct rate depends on what the transaction is, who receives it, the amount, the merchant category and whether a special programme applies. NPCI explains that the merchant category code is assigned by the acquiring bank during onboarding. Under the new framework, inaccurate or stale classification can now change the settlement amount. [2][4]
A neighbourhood merchant may cross the P2PM monthly threshold. A fuel station may be coded as general retail. A capital-market payment may travel through a route that obscures its purpose. These are not merely data-quality issues. They can produce incorrect deductions, refunds, complaints and unequal treatment.
A bank therefore needs governed category evidence, effective dates, review triggers and a controlled reclassification process. Merchant growth should not turn into a surprise charge because a back-office threshold was crossed without timely communication.
Settlement must explain itself
A correct calculation is insufficient if the merchant cannot reconcile it. The settlement record should show the gross transaction, MDR, tax treatment where applicable, net amount and an intelligible reason code. Reversals, refunds, partial refunds, duplicate deductions and delayed adjustments need defined treatment.
This is where strategy meets production. Pricing engines, merchant masters, payment switches, settlement systems, general ledgers, tax processes, statements, dispute tools and support scripts must share the same rule version. If one system changes on 15 October and another changes a day later, the policy can be right while the outcome is wrong.
Connect revenue to resilience
The Government’s stated objective is to support infrastructure resilience, cybersecurity, innovation and customer service. Revenue creates capacity to invest; it does not prove that the investment occurred or improved outcomes. [1][2]
Banks and payment firms should connect the new funding model to measurable service commitments: availability, uncertain-transaction resolution, settlement accuracy, fraud-loss control, complaint ageing, merchant onboarding quality and recovery performance. Otherwise MDR risks being seen as a new deduction rather than a clearer exchange of value.
The second-order effects to watch
Mode steering and price pass-through
Even when customers are not directly charged, merchants may respond by steering larger purchases toward cash or another instrument, changing discounts or spreading costs across prices. The official framework says merchants should not pass MDR to customers. Implementation still needs monitoring, clear complaint routes and practical merchant education. A rule can prohibit a surcharge more easily than it can eliminate every indirect response.
Transaction splitting and category arbitrage
A threshold can encourage attempts to split a purchase into smaller payments. Protected categories can create pressure to remain misclassified. Detection needs to distinguish unusual behaviour from legitimate purchasing patterns. Proportionate monitoring, evidence-led review and fair correction help preserve access while maintaining the intended categories.
Competition can improve—or concentrate
A predictable revenue model may help smaller providers invest and compete. But the distribution formula matters. If the participants doing expensive merchant support or fraud resolution receive too little, service gaps remain. If large players can bundle payment acceptance with unrelated services below cost, smaller providers may still struggle. The existence of MDR does not settle how value and responsibility are shared within the ecosystem.
Inclusion depends on more than a zero rate
Protecting P2PM merchants and small payments is important. Inclusion also depends on connectivity, devices, assisted onboarding, local-language support, fraud recovery and confidence that a completed sale will settle correctly. The proposed small-merchant fund can help, but its governance, eligibility, measurement and allocation will determine whether it reaches the places where acceptance remains difficult.
Four tests for a fair model
- Protect the customer promise. No UPI payment fee should appear through a renamed platform, handling or convenience charge. Customer communication should distinguish transaction limits used for risk control from MDR thresholds used for merchant pricing.
- Make merchant treatment explainable. Classification, rate, cap, exemption and net settlement should be visible and disputable. Optional services should be separately contracted and priced, not used to disguise a payment charge.
- Connect revenue to operating outcomes. Ecosystem participants should know what service obligations the funding supports and measure whether reliability, security, support and reach improve.
- Review behavioural evidence. Track adoption, mode steering, complaints, merchant attrition, misclassification, fraud and geographic coverage. If the design creates material exclusion or gaming, adjust the rules rather than defending the first configuration.
A production-readiness agenda for banks
Before activating the framework, a bank should be able to answer seven operational questions:
- Which dated rule version is in production, and can the bank prove that no new charge was applied before its effective time?
- Which source establishes P2M, P2PM, sector and capital-market classification, and who approves a change?
- How are the ₹2,000 threshold, ₹75,000 cap and special rates tested across ordinary payments, reversals, refunds and exceptions?
- Can a merchant reconcile each deduction to a transaction and reason code?
- Can customer and merchant support explain the same rule without suggesting that customers must pay?
- How are incorrect deductions detected, corrected and reported?
- Which service and inclusion outcomes will management review after implementation?
The bank also needs a cost-to-serve view. That view should separate fixed infrastructure, variable processing, fraud, dispute, merchant service and regulatory costs. It should not assume the new MDR share covers everything, nor treat any remaining gap as permission to weaken essential controls.
When the advice should change
The case for threshold-based MDR becomes weaker if small merchants are misclassified at scale, customers face disguised charges, or the ecosystem cannot show better resilience and service. The case becomes stronger if protected groups remain protected, deductions are transparent, competition broadens and measurable operating outcomes improve.
The appropriate model may also differ by use case. A high-volume public-utility collection has different margins and social importance from discretionary general retail. A micro-merchant has different bargaining power from a national platform. Good policy recognises these boundaries without creating so many categories that the system becomes impossible to administer.
Our perspective
Treat UPI MDR as a governed reinvestment compact: a clear contribution tied to a dependable service.
The customer promise should remain clear: UPI payments are free to the individual. Merchant contribution should be limited, correctly classified and explainable. Ecosystem participants should connect the revenue to dependable acceptance—security, resilience, accurate settlement, timely exception handling and wider inclusion.
This creates shared accountability. Merchants preserve the protected customer price; banks and payment firms explain deductions and demonstrate service outcomes. Regulators and the ecosystem can then review observed behaviour alongside policy intent.
UPI’s success came from making payment feel effortless. Its next phase should make the funding model equally trustworthy: visible enough to challenge, precise enough to administer and purposeful enough to improve the service it supports.
Key takeaway
A sustainable UPI funding model protects everyday access, treats merchants proportionately and turns permitted revenue into verifiable payment reliability.
Learn by deciding
Explore the companion Decision Labs:
- The Payment Feels Free. Who Funds the Rail? — classify merchant payments, explain settlement and connect funding to the customer promise.
- The payment timed out. Now what? — practise resolving uncertain payment outcomes without creating a duplicate business effect.
Sources & further reading
Primary sources rechecked on 7 October 2026. The policy framework is dated 15 September 2026 and the Department of Financial Services FAQ states an effective date of 15 October 2026. This article discusses the announced framework ahead of that date; institutions should use the final operational instructions for production activation.
- Government of India, Press Information Bureau, “UPI Continues to Remain Free for Peer to Peer Transactions and 96% of Merchant Transactions”, 15 September 2026.
- Government of India, Department of Financial Services, “Frequently Asked Questions: Merchant Discount Rate (MDR) on Select UPI (P2M) Transactions”, 15 September 2026.
- Reserve Bank of India, “Discussion Paper on Charges in Payment Systems”, 17 August 2022. This consultative paper presents economic questions and does not itself establish the 2026 rates.
- National Payments Corporation of India, “UPI — Frequently Asked Questions”, merchant onboarding, acquiring-bank and merchant-category-code sections.
The examples, operating framework and Our perspective conclusion are original editorial analysis. They do not constitute legal, regulatory, tax or implementation advice. No third-party passage, chart or image is reproduced.
Put this perspective into practice.
Explore the concepts, make a decision and test what changes when the situation changes.
Topics: Perspective, The Banking Balancing Act