From October 15, a 0.4% Merchant Discount Rate kicks in on UPI payments above ₹2,000. But if digitisation has already saved the exchequer and the banking system tens of thousands of crores, why isn’t the Reserve Bank picking up NPCI’s bill instead of merchants, and, potentially, consumers ?
Kolkata, September 17: From October 15, India’s Unified Payments Interface will operate under rules very different from the ones it has followed for the past six years. The National Payments Corporation of India (NPCI) has finalised a 0.4% Merchant Discount Rate (MDR) on Person-to-Merchant (P2M) UPI transactions above ₹2,000. The announcement has triggered a political storm, protests from sections of the trading community, and sharp questioning inside Parliament’s standing committee on finance. Yet the more fundamental question in this debate has received far less attention: who has actually profited the most from India’s shift away from cash, the ordinary user, or the banks and the government themselves ?
What Changes From October 15
Under the framework finalised by the UPI and Services Steering Committee this week, all Person-to-Person (P2P) transfers and P2M payments up to ₹2,000 remain completely free. Small merchants receiving up to ₹1 lakh a month via UPI QR continue to pay zero MDR. Select categories, railways, telecom, insurance and fuel, attract a flat ₹5 fee per transaction above ₹2,000 rather than the percentage-based rate, while capital-market transactions (mutual funds, securities, stockbroking) attract a nominal 0.02% MDR. Both the Finance Ministry and NPCI have stressed that the charge falls on merchants, not directly on the customer making the payment. NPCI’s own data shows that more than 95% of total UPI P2M transaction volume falls below the ₹2,000 mark, meaning, by NPCI’s own reckoning, only around 4-5% of transactions will be touched by the fee at all.
| Transaction Type | Amount | New MDR |
|---|---|---|
| Person-to-Person (P2P) | Any amount | 0% (free) |
| Person-to-Merchant (P2M) | Up to ₹2,000 | 0% (free) |
| Standard P2M transactions | Above ₹2,000 | 0.4%, capped at ₹300 |
| Railways / telecom / insurance / fuel | Above ₹2,000 | Flat ₹5 per transaction |
| Capital markets (mutual funds, securities) | Above ₹2,000 | 0.02%, capped at ₹300 |
| Small merchants (up to ₹1 lakh/month via UPI QR) | Any amount | 0% (free) |
Source: NPCI and Finance Ministry announcements, September 2026
The Older Story: How Cash-to-Digital Already Paid Off for Banks and the State
To understand the current debate, it helps to look back a decade. ATMs, and later UPI, have quietly saved the banking system and the government enormous sums that rarely enter the public conversation. Industry estimates put the cost of a teller-assisted, in-branch transaction at ₹40–55 or more once staff salaries, branch rent, security and paperwork are factored in. That cost falls to ₹15–23 at an ATM, and to less than ₹1 for a UPI or mobile-app transaction, which runs mostly on server bandwidth and security infrastructure.
| Transaction Channel | Estimated Bank Cost (per transaction) | Main Drivers |
|---|---|---|
| Branch teller (in-person) | ₹40 – ₹55+ | Staff salary, branch rent, security, paperwork |
| ATM | ₹15 – ₹23 | Machine depreciation, power, rent, cash replenishment |
| UPI / mobile app | Under ₹1 | Server bandwidth, database, security infrastructure |
Source: Banking-industry cost estimates
A landmark 2014 Tufts University study estimated that the RBI and India’s commercial banks together spent roughly ₹21,000 crore a year running the cash economy, when India’s GDP stood at around ₹109 lakh crore. GDP has since roughly tripled to more than ₹300 lakh crore; scaling the 2014 baseline linearly would put today’s cash-handling cost at ₹60,000 – 70,000 crore. But UPI processed 24,162 crore transactions worth ₹314 lakh crore in FY 2025-26 alone, a volume physical cash could not have absorbed without a much larger ATM network, a bigger cash-in-transit fleet, and far higher note-printing and destruction costs. Factoring in that infrastructure, analysts estimate the system would have cost India ₹80,000 – 100,000 crore a year without UPI. Against that counterfactual, NPCI says it costs approximately ₹20,000 crore a year to run the entire UPI ecosystem, server bandwidth, fraud prevention and bank technical support included. The digital rail is processing exponentially more transactions at a fraction of the cost the old system would have demanded, and the bulk of that saving has accrued to banks and government, not to the individual user scanning a QR code.
Who Gets the New Fee ?
The same institutions that gained the most from the old cost structure are also positioned to gain the most from the new one. Under NPCI’s revenue-sharing formula for the fresh MDR pool:
| Stakeholder | Revenue Share |
|---|---|
| Issuer bank (holds the customer’s account) | 40% |
| Merchant acquirer (merchant’s bank/institution) | 30% |
| UPI app (payment service provider) | 20% |
| App’s partner bank | 10% |
Source: NPCI’s announced revenue-sharing framework
Industry estimates suggest the new framework could generate ₹15,000 – 21,000 crore a year in fee revenue, roughly comparable to NPCI’s own ₹20,000-crore running cost. Yet that money does not flow to NPCI’s balance sheet or the public exchequer; it is captured almost entirely by commercial banks and fintech platforms, with issuer banks alone taking the single largest share.
The Case for Making This the Reserve Bank’s Bill
This is where the argument for a different funding model gets its force. UPI today is not a peripheral convenience; it is the backbone of India’s domestic payment system and, in every practical sense, a digital extension of the country’s currency. The RBI already absorbs the full cost of printing, transporting, guarding and eventually destroying physical currency, a recurring, multi-thousand-crore expense, as a basic function of managing the currency system, without billing individual note-holders for it.
If UPI is functionally replacing that same flow of physical cash, the argument runs, there is a reasonable case for treating NPCI’s roughly ₹20,000-crore annual running cost the same way: as a cost of maintaining the currency and settlement system, absorbed by the RBI or the exchequer, rather than recovered through a new merchant fee. NPCI itself is a not-for-profit company promoted by the RBI and the Indian Banks’ Association, even though its equity is held by a consortium of roughly 67 member banks and payment entities. Given that the state and the banking system have already captured most of the financial upside of the cash-to-digital shift, avoided branch expansion, better tax compliance, cleaner direct-benefit transfers, critics argue it is reasonable to ask why the same beneficiaries should now also collect a new fee to cover a cost that is, relatively speaking, modest.
Critics further point out that because the new fee flows mainly into private bank and fintech revenue rather than back to NPCI or the public treasury, describing it simply as “the cost of keeping the system running” understates what is actually happening: it is also creating a fresh profit pool, captured largely by the same banks that already gained the most from the earlier cost shift.
UPI Is Not a Credit Card, and Shouldn’t Be Priced Like One
A recurring argument in the MDR debate compares UPI to credit cards, which have always carried an MDR. But the comparison undersells what card MDR actually buys the customer. Credit card MDR typically runs from 1.5% to 3% several times UPI’s new rate, but that fee funds cash-back, reward points, up to 45 interest-free days, and purchase protection. Debit card MDR, capped at 0.5-0.9%, comes with thinner rewards. UPI, historically, charged nothing and offered nothing in return, a fair trade. The new 0.4% fee changes the cost side of that equation without adding anything to the benefit side for the customer.
| Method | Typical MDR | What the Customer Gets |
|---|---|---|
| Credit card | 1.5% – 3% | Cashback, reward points, 45 interest-free days, purchase protection |
| Debit card | 0.5% – 0.9% | Limited rewards |
| UPI (before October 15) | 0% | No rewards, but no cost either |
| UPI (from October 15, above ₹2,000) | 0.4% (on merchant) | No rewards; a new cost enters the system |
Source: Published industry data
Credit card MDR is part of a self-contained business model in which the fee buys the customer something back. UPI’s new fee carries no such trade-off, which is why treating the two as comparable is misleading.
Small in Number, Large in Value – and Disproportionately Tax-Paying
NPCI’s statistic that fewer than 5% of transactions will be affected is technically accurate, but it understates the stakes. Because MDR applies only above ₹2,000, the affected transactions are, almost by definition, the higher-value end of the ledger – big-ticket purchases, business payments, bill settlements – which make up a disproportionate share of total transaction value even though they are a small share of transaction count. That segment also overlaps heavily with formal, tax-compliant economic activity: the same consumers and businesses already paying income tax, GST, fuel excise duty and assorted transaction cesses. If merchants eventually pass the cost down the chain, critics argue, it lands hardest on precisely the group already carrying the largest share of India’s tax burden – layering a new charge on top of existing formal-economy contributions. Government sources also indicate the MDR itself may attract 18% GST, though registered businesses would be able to claim input tax credit on it.
Consumers Are Already Nervous – the Survey Data
An August 2026 LocalCircles survey of roughly 45,000 respondents across 322 districts found that if merchants pass the MDR on to customers, 53% of UPI users could shift away from the platform for transactions above ₹3,000 – 27% to credit cards, 14% to debit cards, and 12% to cash or bank transfers. Only 12% said they would stick with UPI even if they had to bear the fee themselves. The government’s insistence that pass-through is illegal offers a legal safeguard, but it doesn’t fully answer the market’s underlying anxiety about where costs quietly end up once merchants start pricing them in.
The Politics of a Payment Fee
The decision has become a genuine political flashpoint. Congress leader and Leader of the Opposition Rahul Gandhi has accused the government of capitulating to American pressure, calling the charge a “UPI tax” that will benefit US card companies at the expense of ordinary Indians. Congress general secretary Randeep Singh Surjewala went further, describing the move as a “betrayal.” Opposition MPs raised the temperature further at a meeting of Parliament’s standing committee on finance this week. What is more striking is that criticism has also come from within the ruling camp’s own ideological orbit: Ashwani Mahajan, national co-convenor of the RSS-affiliated Swadeshi Jagran Manch, called the decision “most unfortunate,” arguing that UPI’s zero-MDR model had already forced American card networks to cede ground in India and saved the country significant foreign-exchange outflow, an achievement he says the new fee now undercuts. The government has pushed back firmly: Union minister Jyotiraditya Scindia said explicitly that passing the fee on to customers would be illegal and would amount to “a criminal offence,” insisting digital payments will remain free for ordinary citizens. A senior government official separately told the Press Trust of India (PTI) that the decision had already been taken and there was no question of reversing it.
The Other Side of the Ledger
The government’s rationale deserves a fair hearing too. To keep MDR at zero for small merchants, the Centre has run an annual incentive scheme whose cost has climbed steadily, from ₹1,389 crore in FY2021-22 to ₹3,631 crore in FY2023-24 and would only keep growing alongside transaction volumes. Officials argue that a modest, narrowly targeted fee on a thin slice of high-value transactions is a more sustainable model than an open-ended budgetary subsidy. Traders themselves are not unanimous: the Confederation of All India Traders (CAIT) has struck a relatively conciliatory note, pointing out that roughly 80% of retail transactions fall below the ₹2,000 threshold and will be untouched. Other trader groups are more wary, warning that merchants may split bills or route payments through personal UPI IDs to dodge the charge, complicating bookkeeping and potentially pushing some transactions back toward cash.
Expert Analysis
Put the numbers side by side and a pattern emerges. The shift from cash to digital has already saved the government and the banking system tens of thousands of crores a year, avoided a costly expansion of branch and ATM networks, and improved tax compliance. The fresh fee needed to keep that same digital system running, a comparatively modest ₹20,000 crore a year, is now being recovered through a mechanism whose single largest share (40%) flows straight back to the issuer banks that were already the principal beneficiaries of the earlier savings. That pattern supports the case that the RBI, much as it absorbs the cost of printing currency, could reasonably absorb NPCI’s running costs directly, or fund them through the exchequer, rather than through a merchant fee.
The question is not whether someone who can spend ₹75,000 is unable to pay ₹300. The question is about the underlying mentality. Firstly, no one can guarantee that this ₹300 charge will not increase in the future. Secondly, using this exact same logic repeatedly, multiple taxes and cesses are ultimately dumped on honest taxpayers, which amounts to a massive sum in total. The question is about the recovery of black money and how those recovered funds are being spent. The question is about wastage—despite implementing digital systems across all government departments, the old-fashioned use of paper continues to run in parallel. Today, a citizen has four to five types of IDs (Aadhaar, PAN, Voter ID, Ration Card, Passport, etc.), and the sheer volume of paper used in various government offices merely for KYC updates has not stopped even after digitization. The most pathetic situation belongs to those who trade in the stock market; they are forced to pay certain taxes even on their losses.
The Bottom Line
As of now, there is no indication that the government will reconsider its decision before the new rules take effect on October 15. However, statistics make it clear that the biggest financial beneficiaries of the transition from a cash-dependent economy to a digital economy have been the government and the banking system, while the common people have primarily just gained relief from unbearable hassle. The recent controversy over these new charges has therefore brought a major question to the forefront: when deciding who will bear the infrastructure costs of Digital India, are the systemic savings being evaluated with equal importance ?
Sourced from announcements and statements by NPCI, the Union Finance Ministry, PTI, ANI, and a LocalCircles survey.