Fintech Weekly Deep Dive — UPI MDR’s Diwali Delay | Week of October 11–17, 2026

Executive Summary

The most consequential fintech development this week was not a new fee but a delay: on October 13, NPCI said the proposed merchant discount rate (MDR) on selected UPI merchant payments would begin “immediately post Diwali,” rather than October 15. No replacement date was announced. The delay followed merchant objections and arrived just as businesses were preparing for the festive-sales period. It is a reprieve, not a cancellation. The proposed framework still sets a 0.4% charge on eligible person-to-merchant (P2M) payments above ₹2,000, with special rates and exemptions; person-to-person transfers remain outside it.12

That distinction matters because the headline “96% of transactions remain free” describes transaction count, not how much value sits in the remaining 4%, which merchants qualify, or how much revenue the framework will produce. Nor does the official promise that customers will not be charged settle the economic question: the merchant is the formal payer, while prices, discounts and payment choices are where costs can travel. This week’s postponement, a pension regulator’s request for an NPS contribution exemption, and questions about surcharge enforcement expose a design problem, not simply a scheduling one. A sustainable UPI needs dependable funding; a defensible fee needs transparent cost evidence, clear classification and a remedy customers can actually use.

The Story in Depth

Context

MDR is the fee a merchant pays for accepting a digital payment, normally divided among parts of the payment chain. It is not the same as a fee debited from the customer’s UPI account. India’s policy history, however, made “free UPI” mean more than that technical distinction. Government-supported reimbursement helped promote low-value digital payments before a 2020 legal change prohibited charges on specified modes and small transactions. Earlier schemes explicitly compensated acquiring banks for eligible low-value transactions; in 2021–22 the Cabinet approved a ₹1,300 crore incentive scheme, followed by a ₹2,600 crore one-year scheme for FY2022–23.345

During Parliament’s 2026 Monsoon Session, the Taxation and Other Laws (Amendment) Bill passed both Houses; a government bulletin later said it received presidential assent and amended the Payment and Settlement Systems Act framework. The government presented the change as enabling a later MDR decision, not a consumer charge, with rates to be set by an NPCI-led UPI and Services Steering Committee.6 On September 15, the detailed schedule followed. The standard rate was set at 0.4% for specified P2M transactions above ₹2,000, capped at ₹300 for transactions of ₹75,000 or more. Certain essential or thin-margin categories—including railways, telecom, insurance, fuel and agricultural inputs—were assigned a flat ₹5; capital-market payments were assigned 0.02%, also capped at ₹300. P2P transfers, P2M payments up to ₹2,000 and eligible small merchants under the zero-MDR P2PM category were excluded.17

This is not a blanket charge on every UPI payment above ₹2,000. The payment’s purpose, merchant classification and category matter. For an eligible ₹3,000 sale at the standard rate, 0.4% would be ₹12; a ₹75,000 transaction reaches the ₹300 cap. A ₹100,000 transaction would still be capped at ₹300, an effective 0.3%. The exemptions are meaningful, but they also make classification decisions part of the price consumers and businesses experience.

The RBI has treated payment-system pricing as a policy question for years. Its 2022 discussion paper canvassed who should pay for digital payments and how costs might be shared; its Payments Vision 2025 called for evaluation of charges across payment systems while cautioning that affordability and user confidence matter.89 The policy question is therefore not whether payments have costs. It is whether this rate schedule corresponds to those costs, creates the intended investment, and allocates burdens without undermining acceptance.

What Happened This Week

At the start of the week, the NPCI FAQ still showed October 15 as the implementation date. Earlier reports described requests to defer the change, but on October 11 the published status remained unchanged. On October 13, NPCI’s UPI and Services Steering Committee postponed the start until “immediately post Diwali,” saying the revised date would be notified separately. Reporting by The Hindu and Business Standard confirmed that the planned October 15 rollout had been pushed back; the new date and any revised terms were not supplied.210

The timing is not incidental. Merchants had been told to prepare for a near-term change during a major retail season, while traders’ associations argued that a fee could compress already-thin margins or alter how customers pay. The Kerala Hotel and Restaurant Association said some members planned to stop accepting UPI payments above ₹2,000 from October 15. That threat was not evidence of a nationwide boycott, but it showed how an uncertain rule can affect checkout behavior before the first fee is collected.11

Two unresolved boundary questions also surfaced. First, PFRDA confirmed to Mint that it had asked the Department of Financial Services to treat UPI contributions to NPS through D-Remit as P2P transfers. A subscriber transferring money into their own pension account is not obviously buying a merchant’s goods or services, even if the technical payment route resembles a merchant transaction. PFRDA’s request was not an exemption, and the authorities had not published a final answer by the reporting date.12

Second, the fee’s formal incidence does not guarantee that customers never encounter its cost. The Finance Ministry says merchants must not pass MDR on to customers, and NPCI’s September 18 consolidated circular says payee PSPs must ensure no surcharge is levied for UPI transactions; any convenience fee must be disclosed before the user selects the payment mode.13 Lawyers interviewed by Mint nevertheless raised questions about how a customer would identify a prohibited UPI surcharge and obtain a refund. A ban without a clear complaint owner, service deadline and enforcement reporting is harder to rely on than a posted rate.14

Why It Matters

The 96% figure is a protection claim, not a complete impact analysis. The government’s stated estimate is that approximately 96% of P2M transactions will be unaffected because they are below the threshold or covered by the small-merchant exemption.1 That is a useful measure of transaction frequency. It is not a measure of the share of rupee value affected, the number of merchants that will pay, the proportion of each merchant’s turnover received through UPI, or the fee revenue that will be raised. An occasional high-value payment can be a small share of counts but a large share of a merchant’s payment value. Those separate denominators should be published before the policy is described as having only a 4% impact.

Scale makes that distinction consequential. NPCI’s September statistics record 24.07 billion UPI transactions worth ₹29.37 trillion—about 802 million transactions a day.15 Even a small eligible fraction sits on an enormous network. But it would be misleading to multiply all transaction value by 0.4% and call that prospective revenue: the rate applies only to selected P2M transactions, with exclusions, category-specific rates and a cap. The published 96% figure alone is not enough to estimate the base. NPCI should publish anonymised count and value bands by merchant category and exemption status, plus a revenue estimate that can be independently checked.

The fee cliff creates incentives as well as revenue. At the standard rate, the difference between a ₹2,000 payment and a just-over-₹2,000 eligible payment is not merely a fractional charge; it can create an abrupt cost difference. A merchant may absorb it, change a discount, promote another rail, or ask a customer to split a bill. The last option is not automatically fraud, but it is an obvious response to a sharp threshold. The ₹300 cap softens the percentage rate on the largest payments; it does not answer whether a transaction threshold is the best way to distinguish routine retail from high-value commerce. Evaluation should track split payments and substitution, not just total UPI volume.

Merchant classification is now a consumer-protection issue. P2M, P2P and P2PM sound like network labels, but they decide who bears a charge. The NPS D-Remit request demonstrates the weakness of classifying only by settlement plumbing: a person’s contribution to their own regulated retirement account may not have the economic character of a shop purchase. Similar questions may arise for loan repayments, insurance premiums and other transfers with a defined purpose. A public, searchable category register and a simple correction route for wrongly classified merchants are necessary if exceptions are to be trusted.

A sustainable rail needs more than a new revenue line. RBI’s policy papers acknowledge that digital payment services have real costs and that fees can be part of a viable system. NPCI’s case for charging is that participants need resources for technology, security, customer support and wider acceptance. That is plausible, particularly for the banks and providers that carry operational responsibilities. But the policy documents reviewed do not provide a public, independently audited cost base showing which costs the MDR is meant to recover, how much revenue the schedule is expected to produce, or how that revenue will change reliability, fraud response or acceptance. A merchant levy is not automatically an infrastructure budget.

Analysts disagree on the right balance. An IIT Bombay report reported by Business Standard urged rationalisation of the framework, warning that trust depends on the way charges are designed and communicated.16 Economists Atul Mehta and Charan Singh, writing in Mint, set out the funding gap and the limits of relying on public incentives alone; their comparison also highlights that industry estimates of operating costs are not the same as audited, participant-level accounts.17 The Reserve Bank Governor had earlier said a reasonable MDR was unlikely to materially affect UPI volumes, while NPCI chief Dilip Asbe argued that the charge could support ecosystem investment. Those are informed expectations, not post-implementation evidence. The postponement means the volume-impact test has not yet begun.18

A good design would publish its theory of change: which costs are being funded, who receives the proceeds, what security or acceptance outcomes will be measured, and when rates will be reviewed. The proposed revenue share should be transparent to merchants and users. If money is meant to support small-merchant acceptance, disclose how much is ring-fenced, who qualifies, and how many merchants are reached. “Sustainability” is a goal; it is not an auditable spending plan.

Data & Metrics

  • September 2026: 24.07 billion UPI transactions, worth ₹29.37 trillion; roughly 802 million transactions per day, according to NPCI data reported by Business Standard and available through NPCI’s statistics portal.15
  • Government estimate: approximately 96% of P2M transactions remain unaffected. This is a transaction-count estimate, not a value-weighted estimate or independent audit.1
  • Standard MDR: 0.4% on specified P2M payments above ₹2,000, capped at ₹300 for payments of ₹75,000 and above; selected categories have separate rates.17
  • Historical public support: Cabinet-approved incentive outlays included ₹1,300 crore for FY2021–22 and ₹2,600 crore for FY2022–23. These were schemes with defined periods and eligible modes, not a direct forecast of current UPI operating costs.45
  • No independent public estimate of the new framework’s total eligible transaction value or expected annual MDR collection was identified in the material reviewed as of October 18.

Expert Views

  • Government and NPCI: The stated objective is long-term ecosystem sustainability while keeping P2P, small-value payments and eligible small merchants free. The government says the charge belongs within the merchant-payment ecosystem, not the treasury; banks must prevent merchants from passing it to customers.1
  • RBI: Governor Sanjay Malhotra said a reasonable MDR was not expected to cause a major reduction in UPI volumes. That is a forecast made before the postponed rollout, not evidence of the actual response.18
  • NPCI leadership: CEO Dilip Asbe has argued that a modest MDR need not materially reduce use and could support investment in the payments ecosystem. The case depends on revenue being used for that purpose and on costs not reappearing as covert consumer surcharges.18
  • Research and economists: The IIT Bombay report calls for a more carefully calibrated design to protect trust; Mehta and Singh argue that public incentives alone may not be a durable funding model. Neither position removes the need for audited cost and outcome data.1617
  • Pension regulator and traders: PFRDA’s D-Remit request is a concrete example of classification uncertainty. Trader groups’ objections, including the Kerala association’s proposed refusal of high-value UPI, show the risk of payment substitution when implementation is unclear.1112

Consumer Impact

For customers, the direct policy promise is that UPI remains free to use: no MDR is debited from the payer’s account, P2P transfers remain free, and the framework prohibits a UPI-only surcharge. As of October 18, however, the October 15 start had been postponed and no new date had been announced. Consumers should not assume that a merchant can add a fee just because a payment exceeds ₹2,000; the proposed rates apply only to eligible categories, and the merchant is the formal payer.

The indirect risk is subtler. A merchant facing a new cost can change prices or discounts across the board, steer customers to cash or cards, or make acceptance less predictable. A broad price rise cannot automatically be labelled an MDR surcharge, so enforcement needs to focus on explicit UPI-only charges and clear disclosure. Keep an itemised bill and UPI reference if a separate fee appears, ask the merchant to identify its basis, and complain through the bank or payment app while retaining both records. The public rules should name the institution responsible for resolving such cases and require a time-bound response.

For small businesses, the exact category and monthly receipt criteria matter more than the headline threshold. The P2PM exemption needs accessible eligibility checks, written reasons when an account is reclassified, and a way to correct errors before deductions start. Pension savers and other users making non-retail transfers need an answer before the fee begins, not after a charge has been routed through a merchant code.

Looking Ahead

The next test is whether “post Diwali” becomes a transparent, technically prepared launch or an open-ended notice that leaves merchants guessing. Before setting a date, NPCI and the Finance Ministry should publish one definitive schedule; clarify D-Remit and comparable non-purchase flows; release transaction counts and value bands behind the 96% estimate; and set out complaint, refund and audit procedures. They should also publish the cost categories and expected use of MDR proceeds, then review the framework against uptime, fraud losses, merchant acceptance, customer complaints and payment substitution.

A short delay is sensible if it gives merchants and payment providers time to update systems. It is less defensible if the pause only postpones an unresolved argument. The decision to collect a fee can be revised after evidence; public trust, once merchants print new QR instructions or customers encounter unexplained checkout charges, is harder to restore. UPI’s next phase should be judged not by whether it is nominally free or nominally sustainable, but by whether the people paying into the system can see what the money buys—and whether ordinary users retain a genuinely usable, dependable payment choice.

Sources

Primary documents and data

Reporting, research and analysis

  1. Ministry of Finance / PIB: UPI MDR framework, exemptions and consumer protections ↩︎ ↩︎ ↩︎ ↩︎ ↩︎ ↩︎

  2. NPCI: October 13 announcement · Business Standard: coverage of the deferral ↩︎ ↩︎

  3. PIB: 2018 Cabinet approval to subsidise MDR on low-value digital transactions ↩︎

  4. PIB: 2021–22 incentive scheme for low-value BHIM-UPI P2M and RuPay ↩︎ ↩︎

  5. PIB: FY2022–23 incentive scheme and ₹2,600 crore outlay ↩︎ ↩︎

  6. PIB: government statement on the amendment’s purpose · PIB: legislative record of passage by both Houses · New India Samachar / PIB: report of presidential assent and PSS Act amendment ↩︎

  7. NPCI: FAQs on MDR for select UPI P2M transactions ↩︎ ↩︎

  8. RBI: Discussion Paper on Charges in Payment Systems (2022) ↩︎

  9. RBI: Payments Vision 2025 ↩︎

  10. The Hindu: NPCI postpones MDR implementation until after Diwali · NPCI’s official October 13 post ↩︎

  11. The Hindu: Kerala hotels association objects to high-value UPI MDR ↩︎ ↩︎

  12. Mint: PFRDA seeks an exemption for NPS D-Remit contributions ↩︎ ↩︎

  13. NPCI: UPI Consolidated Circular v1.0, September 18, 2026 ↩︎

  14. Mint: Consumer surcharge rules and enforcement concerns ↩︎

  15. NPCI: UPI ecosystem statistics · Business Standard: MDR deferral and September UPI data ↩︎ ↩︎

  16. Business Standard: IIT Bombay report calls for MDR rationalisation ↩︎ ↩︎

  17. Mint: Atul Mehta and Charan Singh on UPI cost and pricing ↩︎ ↩︎

  18. Business Standard: RBI Governor on possible MDR impact on UPI volumes · Mint: NPCI chief on MDR and funding ↩︎ ↩︎ ↩︎