On the evening of 15 September 2026, a single announcement ended a six-year-old arrangement that most Indians had stopped thinking about. A Merchant Discount Rate of 0.4% would apply to UPI payments made to merchants above ₹2,000, effective 15 October. Within hours, screenshots began circulating on WhatsApp and Instagram stitching this together with two entirely separate stories — a US Congressional vote on tariffs, and a fresh round of H-1B restrictions — into one tidy narrative: India has caved to Washington, and your UPI is the price.
The question on a lot of minds: is that actually what happened?
It is a fair question, and it deserves better than either a viral thread or a government press release. What follows is an attempt to separate what is verifiably true, what is inference dressed up as fact, and what genuinely remains unknown. Some of the circulating claims hold up well. Some hold up only if you squint. And a few of the doubts being raised are legitimate enough that no amount of official clarification has actually answered them.
Let’s go claim by claim.
Claim 1: “The US House is voting on a Bill that introduces 100% tariffs on India. The Senate has already approved it.”
Partly true, but the framing is doing heavy lifting.
The legislation in question is real. It is the Lindsey O. Graham Sanctioning Russia and Iran Act of 2026, which cleared the US Senate 86–11 in August. On 15 September the House passed a procedural resolution 214–211 to take up the Senate amendments, setting up a final floor vote.
Here is where the viral framing diverges from the text. The Bill does not introduce a 100% tariff on India. It authorises the US President to impose secondary tariffs of up to 100% on countries that purchase Russian oil. Authorisation is not imposition — it hands discretion to the executive, it does not pull a trigger.
Second, India is not singled out. An amendment moved by Democratic Congressman Steny Hoyer names ten countries as eligible for these duties: China, India, Türkiye, Azerbaijan, Hungary, the Slovak Republic, the UAE, Singapore, Kazakhstan and the Kyrgyz Republic. India appears on a list, alongside two NATO members and a US security partner.
So: the Bill is real, the Senate vote is real, the 100% figure is real, and India is named. What is missing from the viral version is that this is a conditional authority over a group of countries, not a law that levies a tariff on India.
What remains genuinely uncertain: whether the executive chooses to use that authority, against whom, and at what rate. That is a political decision, not a legislative one, and nobody can currently answer it.
Claim 2: “The Trump administration is cracking down on Indian immigrants. H-1B costs are up and holders could be deported immediately on losing a job.”
Substantially true, with one important qualifier on tense.
The September 2025 proclamation imposing a $100,000 fee on new H-1B petitions filed from outside the US is real and was implemented by USCIS. It was subsequently struck down by a federal court, which held that the administration had exceeded its authority and encroached on Congress’s power to set immigration policy and levy taxes. The administration has appealed. In parallel, DHS has proposed a new fee of $103,265 on cap-subject H-1B petitions through the rulemaking route — which, unlike a proclamation, is the legally sturdier path.
On deportation: in September 2026, DHS proposed a rule that would eliminate the existing 60-day grace period an H-1B holder gets to find new employment after a job loss. If finalised as proposed, losing your job would make you removable far more quickly than today.
The operative word in both cases is proposed. Neither the new fee nor the grace period removal is currently in force. Proposed rules go through notice-and-comment, are frequently modified, and are very often litigated. The claim describes a real and consequential direction of travel; it just describes it in the present tense when the correct tense is future-conditional.
The detail about H-1B holders being predominantly IT professionals is accurate — roughly two-thirds of approved petitions are in computer-related occupations, and Indian nationals receive around 70% of H-1Bs. The pressure on this cohort is not imagined.
Claim 3: “The Modi government has given in to a US demand to get rid of zero MDR.”
This is where fact ends and inference begins — and the inference is not unreasonable, but it is not established either.
Two things here are independently, verifiably true.
One: the USTR’s National Trade Estimate Report on Foreign Trade Barriers, published in March 2026, does flag India’s payments regime. It raises concerns about zero transaction charges on UPI and RuPay, about government support for RuPay, about RuPay’s earlier access to credit-card rails on UPI, and about NPCI’s proposed 30% volume cap on third-party apps. This is not a rumour. It is in a public document from the office of the US Trade Representative.
Two: India has, six months later, introduced an MDR.
What connects them is where the argument lives. Correlation of timing is suggestive. It is not proof. The Finance Ministry issued a statement on 16 September explicitly rejecting the foreign-influence framing, saying India’s UPI policy decisions are made independently with the goal of a self-sustaining ecosystem.
A government denial is not dispositive either — governments deny things. But there are a few inconvenient details that the “capitulation” narrative tends to leave out, and they are worth putting on the table:
- The two largest beneficiaries of zero MDR are American. PhonePe is Walmart-owned. Google Pay is Alphabet-owned. Between them they handle over 80% of UPI volume. A framing in which zero MDR was an anti-American policy has to explain why its two biggest winners are US corporations.
- The 0.4% rate is not card parity. The claim that 0.4% was chosen to match debit-card MDR and let US networks compete does not survive a look at the RBI’s actual debit-card framework. Under the caps effective January 2018, small merchants (turnover up to ₹20 lakh) pay 0.40% on POS and 0.30% on QR; larger merchants pay up to 0.90% on POS and 0.80% on QR. So 0.4% sits at the bottom of the debit range, not at parity with it. If the objective were to level the field for Visa and Mastercard, 0.4% is a strange number to pick — it leaves UPI cheaper than cards for exactly the merchants who process the most volume.
- The domestic case for MDR predates the USTR report by years. The RBI’s 2022 discussion paper on charges in payment systems, industry bodies, and the Parliamentary Standing Committee on Finance have all been circling the zero-MDR sustainability problem well before any of this. The idea did not arrive in March 2026.
What remains genuinely uncertain: whether US trade pressure was a factor, the decisive factor, or no factor at all. Nobody outside the room knows. Anyone asserting certainty in either direction — “this is capitulation” or “this is entirely independent” — is telling you more about their priors than about the decision.
Claim 4: “Is this really about making UPI sustainable? Running it costs ₹20,000 crore a year, which is under 10% of what the RBI transfers to the Centre.”
Both numbers check out. The comparison between them is where it gets slippery.
The ₹20,000 crore figure is real and well sourced. Industry estimates put the annual cost of running the UPI ecosystem — infrastructure, servers, cybersecurity, fraud monitoring, NPCI and network costs, customer support — at roughly ₹15,000–20,000 crore. The Parliamentary Standing Committee on Finance cited an estimate of around ₹20,700 crore. This cost is distributed across banks, NPCI, TPAPs, PSPs and technology vendors.
The RBI transfer figures are also real, and larger than the post suggests: ₹2.69 lakh crore for FY25 and ₹2.87 lakh crore for FY26. So ₹20,000 crore is roughly 7–8% of a single year’s surplus transfer. The arithmetic holds.
Where the argument gets slippery is in the implied conclusion. The RBI surplus is not discretionary cash sitting idle. It is a statutory transfer of central bank income under the RBI Act, governed by the Economic Capital Framework, and it is already budgeted and already spent — it appears on the receipts side of the Union Budget and funds existing expenditure. “Just pay for UPI from the RBI dividend” is functionally identical to “fund UPI from general taxation,” which is a perfectly legitimate policy position. It is simply not the gotcha it is being presented as. The real question is one of budgetary priority — should ₹20,000 crore of general revenue go to subsidising payment rails versus everything else it could fund? — and reasonable people land in different places on that.
There is also a subsidy-fatigue angle worth noting. The Union Cabinet’s incentive scheme for low-value UPI transactions has been running at roughly ₹1,500 crore a year — nowhere near the ₹20,000 crore the system actually costs. The gap has been absorbed by banks and PSPs, most of whom have been loudly unhappy about it for years. That pressure is real and domestic.
What the Viral Version Leaves Out
Whatever one concludes about motive, the scope of the actual notification is narrower than the panic suggests, and this is worth stating plainly because it is factual rather than interpretive:
- P2P transfers remain free. Sending money to a friend, family member or landlord is unaffected, at any amount.
- The fee applies only to P2M transactions above ₹2,000, and is capped at ₹300 for a single transaction of ₹75,000 or more.
- Roughly 95% of merchant transactions are below ₹2,000 and are untouched.
- Small merchants earning up to ₹1 lakh a month via UPI QR continue at zero MDR.
- Essential categories — railways, fuel, education fees, government utility bills — carry a flat ₹5 cap above ₹2,000 rather than 0.4%.
- The merchant pays, not the consumer. No charge is levied on the person making the payment.
None of that makes the decision beyond criticism. It does mean the version circulating — that UPI is now “charged” — misdescribes what was announced.
The Doubts That Have Not Been Answered
Demystifying the viral claims is not the same as declaring the policy sound. Several questions remain genuinely open, and the official clarifications have not closed them:
Will merchants absorb the cost, or pass it on? MDR is legally the merchant’s cost. Economically, costs travel. The most likely outcome for above-₹2,000 purchases is a quiet price adjustment, a card-style surcharge, or a nudge toward cash. The Finance Ministry’s “consumers pay nothing” framing is technically accurate and economically incomplete.
Is ₹2,000 a permanent floor or an opening position? Nothing in the framework makes the threshold hard to revise. A threshold that can be lowered by notification is a threshold that can be lowered by notification. The GST-on-UPI rumours of 2025 were denied and did not materialise; the MDR rumours of 2025 were also denied, and here we are. That history is precisely why public scepticism of the current assurances is rational rather than paranoid.
Where does the 0.4% actually go? The split between acquiring bank, issuing bank, NPCI and the TPAP determines whether this genuinely fixes the sustainability problem or simply redistributes it. The public detail on this is thin.
Does this reopen the door for card networks? Not at 0.4% on its own — UPI remains materially cheaper. But the principle that UPI is a priced service rather than a free public utility has now been conceded. Principles, once conceded, are easier to extend than to establish.
Was there any public consultation? For a change affecting hundreds of millions of daily users and crores of merchants, the process was notably quiet. That is a procedural criticism entirely independent of whether the policy itself is right.
Reflections
The honest summary is uncomfortable for everyone. The viral narrative contains no outright fabrications — the Bill exists, the Senate vote happened, the USTR report says what it is quoted as saying, the H-1B measures are real, the ₹20,000 crore figure is sourced, and the RBI arithmetic is correct. What it does is stitch verified facts into an unverified causal chain, and present a conditional tariff authority as an imposed tariff, a proposed rule as an enacted one, and a timing correlation as an established motive.
The government’s position has the opposite problem. Every individual assurance it has given is accurate, and the overall picture it paints — a purely domestic, purely technical decision — asks the public to accept a coincidence of timing without offering anything to explain it. “Trust us” is not evidence either.
There is also a broader pattern worth keeping in view. The US imposed tariffs on Brazil in July 2026 following a Section 301 investigation that, among other complaints, cited Brazil’s Pix instant-payment system as disadvantaging American payment companies. Sovereign public payment infrastructure has become a live trade-policy issue globally. Whether or not it decided this particular notification, that pressure is real and is not going away.
Conclusion
Zero MDR was never free. It was a subsidy — carried by banks and payment providers, backstopped inconsistently by the exchequer, and invisible to the people benefiting from it. Ending an invisible subsidy always looks like an imposition, because nobody was ever shown the bill.
That does not settle whether 0.4% is the right number, whether ₹2,000 is the right threshold, whether the timing was coincidental, or whether the process was adequate. Those are open questions, and treating them as settled — in either direction — is the actual error here.
What we can say with confidence: the tariff Bill authorises rather than imposes. The H-1B rules are proposed rather than in force. The USTR objection is documented but the causal link to this notification is not. The ₹20,000 crore figure is sound but the RBI comparison is a budget-priority argument rather than a contradiction. And the fee itself is narrower in scope than the panic suggests, while raising legitimate questions the official clarifications have not addressed.
Scepticism of the official line is warranted. So is scepticism of the viral one. Both are asking to be believed on the strength of their framing rather than their evidence.
Cited Sources
- Lindsey O. Graham Sanctioning Russia Act of 2026 — S.5025, 119th Congress, full text
- House vote and the Hoyer amendment naming India — The Tribune, 16 September 2026
- USTR 2026 National Trade Estimate Report on Foreign Trade Barriers — Office of the United States Trade Representative, March 2026
- Finance Ministry rejection of foreign-influence claims — IANS, 16 September 2026
- UPI MDR framework and thresholds — Business Today, 15 September 2026
- RBI debit card MDR caps (effective January 2018) — Business Standard
- Cost economics of running UPI — Business Standard
- RBI surplus transfer to the Centre — Business Standard, May 2025
- H-1B $100,000 fee and its implementation — American Immigration Council
- Proposed removal of the 60-day grace period — TIME, 11 September 2026
Read more from us in our blogs
