UPSC Darpan

EconomyGS318 September 2026

UPI Merchant Fee: The Final Framework, and Who Actually Pays and Gains

Open in the app — quiz, notes, Mistake Vault

The news

The National Payments Corporation of India released its circular on charges to be levied on certain UPI payments from October 15. Per The Hindu's explainer, the Merchant Discount Rate (MDR) is a fee for using UPI paid by merchants to payment processors and banks; consumers will not directly pay it, and the Ministry of Finance has emphasised that banks have been advised to ensure merchants do not pass MDR charges on to customers, and that UPI application providers are expressly prohibited from imposing platform fees or hidden charges on users. MDR will be paid by mid to large-sized merchants receiving UPI payments in excess of ₹2,000 per transaction, at 0.4%; for transactions of ₹75,000 and above the MDR is capped at ₹300. All Person-to-Person transactions remain free regardless of amount, and P2P makes up about 37% of total UPI transaction volume. Transactions of ₹2,000 or more in essential and thin-margin sectors such as railways, telecommunications, insurance, fuel and agricultural inputs attract a flat MDR of ₹5. Capital market transactions such as payments to mutual funds, stockbrokers, dealers and equities attract an MDR of 0.02%, capped at ₹300. Small merchants receiving up to ₹1 lakh per month through UPI QR codes under the Person-to-Person-Merchant category are exempt. Person-to-merchant transactions above ₹2,000 make up just 2.5% of all UPI transactions by volume; payments to merchants up to ₹2,000 make up another 60.5%, so in total 97.5% of UPI transactions remain free. While ₹29.8 lakh crore was transacted over UPI in August 2026, P2M transactions above ₹2,000 were ₹5.99 lakh crore, meaning the maximum banks and payment processors can earn from MDR is about ₹2,400 crore a month or ₹28,000 crore a year. Of the MDR collected, about 40% goes to the payer's bank, 30% to the merchant's bank, 20% to the UPI app or Third-Party Application Provider, and 10% to the Payment Service Provider bank; 5% of total MDR collections will go to a dedicated fund to promote UPI adoption among small merchants. Yes Bank is the payer bank in more than 50% of all UPI transactions and the payee in about 55%, followed by ICICI at 18.3% and Axis at about 19% respectively; PhonePe accounts for about 46% of UPI transactions by volume and Google Pay another 32%. The Economic Times reported that the government will monitor the rollout to ensure merchants do not pass charges to consumers, that GST on MDR can be taken up by the GST Council with input tax credit available to merchants, and that RuPay debit cards are kept free of MDR. Retailers' bodies including AIMRA, representing about 150,000 neighbourhood stores, and the South Indian Organised Retailers Association have petitioned the finance ministry to scrap the charge; new-age brokers Groww, Zerodha, Upstox and Angel One, who together account for about 62.5% of NSE's active client base, want the MDR capped at ₹5 and the exempt threshold raised to ₹20,000. SEBI Chairman Tuhin Kanta Pandey said the regulator will examine broker and asset management company concerns.

The chain in one line: MDR of 0.4% applies only to P2M above ₹2,000 → that is 2.5% of UPI volume but ₹5.99 lakh crore of August's ₹29.8 lakh crore value → maximum ₹2,400 crore a month for the ecosystem → split 40-30-20-10 across payer bank, payee bank, app and PSP → because Yes Bank is payer in over 50% and payee in about 55% of transactions, most of the 70% bank share concentrates in one bank.

Static syllabus linkage

  1. What MDR is and who pays it. The Merchant Discount Rate is the fee a merchant pays to the banks and payment processors that carry a digital transaction. It is deducted from what the merchant receives, so the customer never sees it as a line item — which is why the entire policy debate turns on whether merchants will quietly raise prices instead.
  2. The zero-MDR regime and its legal basis. Zero MDR on UPI and RuPay debit transactions has been in force since January 2020, implemented through Section 10A of the Payment and Settlement Systems Act, 2007 and Section 269SU of the Income-tax Act, 1961. Reintroducing a charge therefore required an explicit policy reversal, not merely a pricing decision.
  3. Who runs the rail. UPI is operated by the National Payments Corporation of India, a not-for-profit company set up under Section 8 of the Companies Act by the RBI and the Indian Banks' Association. It is not a statutory regulator; the RBI regulates payment systems under the Payment and Settlement Systems Act, 2007.
  4. The four-party split, explained. A UPI payment involves the payer's bank, the payee's bank, the app the customer uses (a Third-Party Application Provider), and the Payment Service Provider bank that connects the app to the network. The 40-30-20-10 split of MDR simply pays each of those four for their role.
  5. The 30% market-share cap that was never enforced. NPCI announced in November 2020 that no third-party application provider should exceed 30% of UPI transaction volume. PhonePe at about 46% and Google Pay at about 32% are both above it. The cap was repeatedly deferred because enforcement would mean turning away customers from the largest apps.

Why UPSC loves this

  1. Digital Public Infrastructure is a flagship GS3 theme. UPI, Aadhaar and India Stack are standard answer material. What is new here is the financing question — who pays for a public digital utility once it reaches scale — which is a much sharper question than 'what are UPI's benefits'.
  2. It is a clean case of political economy. The Opposition frames it as capitulation to US card companies, an RSS-affiliated body objects on small-merchant grounds, retailers want it scrapped and brokers want it re-shaped. Multiple actors opposing the same policy for different reasons is exactly the structure examiners use to test whether a candidate can separate interests from arguments.
  3. The data allows a technocratic critique the politics misses. The transaction-level numbers show a narrow behavioural reach but a highly concentrated revenue gain. That is a competition-policy point, and it is the analytical gap most answers will leave open.

Prelims nuggets

  • UPI is operated by the National Payments Corporation of India (NPCI), an umbrella organisation incorporated as a Section 8 not-for-profit company, promoted by the RBI and the Indian Banks' Association.
  • Payment systems in India are regulated by the RBI under the Payment and Settlement Systems Act, 2007; zero MDR on UPI and RuPay debit card transactions was implemented via Section 10A of that Act and Section 269SU of the Income-tax Act, 1961.
  • Under the framework effective from October 15, 2026: MDR of 0.4% applies to person-to-merchant transactions above ₹2,000, capped at ₹300 for transactions of ₹75,000 and above; a flat ₹5 applies in specified essential sectors; 0.02% capped at ₹300 applies to capital market transactions; all person-to-person transactions remain free.
  • Small merchants receiving up to ₹1 lakh per month through UPI QR codes, classified under Person-to-Person-Merchant (P2PM), are exempt from MDR.
  • NPCI's November 2020 directive capping any third-party application provider's share of UPI volume at 30% has not been enforced.
  • NPCI's other products include RuPay, IMPS, NACH, AePS, BHIM, NETC FASTag and Bharat BillPay.

Analysis

  1. The fee's reach is narrow but its revenue is not. Only 2.5% of UPI transactions by volume attract the 0.4% charge, and 97.5% remain free. That makes the government's 'most users are unaffected' claim factually correct. But those few transactions carried ₹5.99 lakh crore of the ₹29.8 lakh crore transacted in August, so the revenue is roughly ₹2,400 crore a month, about ₹28,000 crore a year. Low behavioural reach and high revenue concentration can both be true, and an answer that reports only one of them is incomplete.
  2. The distribution of that revenue is the real competition story. Seventy per cent of MDR goes to banks — 40% to the payer's bank and 30% to the payee's. Yes Bank is the payer bank in over half of all UPI transactions and the payee bank in about 55%. A fee designed to make the payments ecosystem self-sustaining will therefore route a disproportionate share to a single institution, not because it processes transactions better but because of how sponsor-bank arrangements evolved. That is a structural artefact, and it deserves regulatory attention independent of whether the fee should exist.
  3. The design is unusually careful, and that is its own problem. There are at least five regimes running at once: 0.4% above ₹2,000, a ₹300 cap above ₹75,000, a flat ₹5 for essential sectors, 0.02% for capital markets, and full exemption for small merchants below ₹1 lakh a month. Each carve-out protects a genuine interest. Together they create exactly the compliance complexity that GST's early years showed falls hardest on small businesses, who must now determine which of five rules applies to a given receipt.
  4. The pass-through prohibition is the weakest link. The Ministry has advised banks to ensure merchants do not pass the charge to customers, and apps are barred from levying platform fees. But a merchant does not need a surcharge line to recover 0.4% — the price simply goes up. FMCG retailers operating on 8-12% gross margins and mobile retailers on 0.75-1.5% net margins will not absorb it silently. A prohibition that can be complied with in form and defeated in substance is not a real safeguard, and the honest framing is that the cost will partly reach consumers through prices.
  5. The capital-market objection is different in kind, not just in degree. Brokers argue that transferring money into one's own trading account is not paying a merchant, and therefore should not attract a merchant fee at all. That is a definitional objection rather than a pricing complaint: a fund transfer to oneself creates no merchant-acquiring service for anyone to be paid for. The four brokers raising it account for about 62.5% of NSE's active clients, so the practical effect is a friction on retail market participation — the opposite of a stated policy goal. SEBI's decision to examine it is the right forum, since this is a securities-market access question wearing a payments costume.
  6. The unenforced 30% cap undermines the sustainability argument. The official case for MDR is that the ecosystem must become financially self-sustaining. But the same ecosystem has left NPCI's 30% market-share cap unenforced while PhonePe reached about 46% and Google Pay about 32%. Charging merchants to sustain a system whose concentration rules are not applied invites the fair question of whether the sustainability problem is a revenue problem or a structural one.
  7. The 5% small-merchant fund is the one piece worth watching. Five per cent of MDR collections — roughly ₹1,400 crore a year at current volumes — is to be routed to a dedicated fund promoting UPI adoption among small merchants. If it is transparently administered, this is a redistributive feature that partly answers the equity objection. If it is not, it becomes a levy on merchants that funds an unaudited pool, which is why the framework's silence on who administers it and how it is reported matters more than its size.

Possible Mains question

"The reintroduction of a Merchant Discount Rate on high-value UPI transactions raises a question India has avoided since 2020: who should pay for a digital public utility once it becomes indispensable?" Critically examine the design of the new framework and its distributional consequences.

Model approach

  1. Introduction — frame it as a financing question. Open by noting that zero MDR since January 2020, implemented through Section 10A of the PSS Act and Section 269SU of the Income-tax Act, made the state and the banks bear the cost of UPI's growth, and that the new framework revisits that choice.
  2. Body 1 — set out the design precisely. Cover the 0.4% above ₹2,000, the ₹300 cap at ₹75,000 and above, the flat ₹5 for essential sectors, 0.02% for capital markets, the P2PM exemption and the continued zero charge on P2P.
  3. Body 2 — establish scale with the data. Use 2.5% of volume against ₹5.99 lakh crore of value, and the ₹2,400 crore monthly ceiling, to show that narrow reach and significant revenue coexist.
  4. Body 3 — make the competition-policy argument. Show that 70% of MDR flows to banks and that one bank sits on both sides of the majority of transactions, so the fee's benefit is structurally concentrated; connect this to the unenforced 30% cap on app market share.
  5. Body 4 — assess the consumer-protection safeguard honestly. Argue that a prohibition on pass-through cannot survive thin merchant margins, and that the realistic policy question is how much reaches consumers through prices rather than whether any does.
  6. Body 5 — separate the capital-market objection. Explain why a transfer into one's own trading account is conceptually not a merchant payment, and why treating it as one imposes a friction on retail market participation.
  7. Conclusion — propose the conditions for legitimacy. Conclude that a fee on a public digital utility is defensible only if its revenue-sharing formula tracks cost and risk rather than legacy arrangements, if the small-merchant fund is transparently administered, and if the concentration rules the ecosystem already has are actually enforced.

Administrator's brainstorm

As a banking regulator, how would you respond to the finding that MDR revenue concentrates disproportionately in one or two banks?

First establish whether the concentration reflects genuine differences in infrastructure investment, uptime and risk borne, which would be a legitimate competitive outcome, or whether it is an artefact of how sponsor-bank and PSP arrangements were allocated when UPI was scaling and nobody was paying attention to who sat on which side of a transaction. If it is the latter, the revenue-sharing formula should be revised so that each participant's share tracks the transaction-processing cost and risk it actually bears, which is the only defensible basis for a regulated fee. Publish the concentration data quarterly, because a sharing formula whose outcomes are invisible cannot be held to account. And treat the unenforced 30% application cap as part of the same problem rather than a separate one: a system cannot credibly claim it needs revenue to be sustainable while leaving its own structural limits unapplied.

You are in the Department of Financial Services. How would you make the anti-pass-through rule enforceable rather than aspirational?

Accept first that you cannot police a price rise, only a visible surcharge, and design accordingly. Prohibit and penalise explicit surcharging at the point of sale, which is detectable and provable, and build a public reporting channel where a customer can upload a bill showing a UPI-linked charge. For the invisible pass-through, use measurement rather than prohibition: track price movements in the specific thin-margin categories most exposed to the fee against a control set, publish the comparison, and commit to revisiting the threshold if the data shows sustained divergence. Pair this with the publication of the small-merchant fund's receipts and disbursements, since the fund is the policy's own equity instrument and its credibility depends on being auditable. And keep the review date fixed in advance, because a fee introduced with a monitoring promise but no scheduled reassessment tends to become permanent by default.