Nithin Kamath warns UPI MDR could hit brokers

Nithin Kamath, Zerodha founder and CEO, has criticised the proposed UPI Merchant Discount Rate (MDR) structure for investing and broking payments. His core argument is that a broker can end up paying a charge on every UPI fund transfer even when the customer does not place a trade. The comments have been widely shared on X and discussed across investing communities, with attention on how payment charges can alter brokerage pricing models. Kamath said MDR is “probably inevitable” given UPI’s scale, but he wants the structure to reflect how broker accounts are funded. He also warned that free equity delivery trades may not stay free if brokers have to absorb a new cost on client transfers. The debate matters because it connects payments policy directly to retail investor costs.

What the proposed UPI MDR framework says

The new framework discussed on social media focuses on UPI transactions above Rs 2,000 made to merchants. Under NPCI’s announced structure, capital-market transactions including mutual funds, securities, stockbrokers and dealers attract an MDR of 0.02 percent. That MDR comes with a cap of Rs 300 per transaction, and Kamath has called that cap steep for broking use cases. He has highlighted that some other categories are set to pay a flat fee per transaction. The discussion has also referenced that these charges would apply regardless of whether the payment leads to a revenue-generating action for the merchant. For brokers, the merchant in the payment flow is typically the brokerage, even when the payment is simply a wallet-like transfer into a trading account. The framework has been described as starting from October 15 for capital-market transactions.

Why broking differs from normal merchant payments

Kamath’s central point is that a UPI transfer into a broking account is not the same as paying for a product or service. In many merchant categories, a payment and a sale are closely linked in time and intent. In broking, a client can add funds for later use, or move money in and out based on opportunity, volatility, or personal cash flow. A broker cannot force a customer to trade after transferring money, Kamath said. That means the broker can incur costs without receiving brokerage or any related revenue. This mismatch becomes more visible when a charge is applied to every transfer, not only to executed transactions. Kamath has argued that a single MDR design across all merchant types does not reflect these differences.

The cost problem when customers do not trade

Kamath has warned that the MDR could create significant costs even when customers do not execute trades. He said brokers may struggle to absorb such costs indefinitely, especially when the charge is applied repeatedly to funding transactions. The critique is not that MDR should never exist, but that it should not be structured in a way that makes non-trading behaviour expensive for intermediaries. He also noted that if brokers cannot pass on UPI charges to customers, then there is effectively no limit to the cost a customer can impose on the broker without generating business. This is a strong statement because it frames the issue as an uncapped liability in practice, even if each payment has a cap. Retail clients often use UPI for convenience and speed, so high-frequency transfers can be common even among low-activity traders. The broader implication is that payment charges can reshape how brokers design funding flows and client experience.

Kamath’s hypothetical example that sparked debate

To illustrate the economics, Kamath shared a hypothetical scenario that circulated widely in market forums. He described 10,000 customers making 50 UPI transfers of Rs 2 lakh in a month without executing a single trade. At the applicable MDR, he said such activity could potentially cost a broker about Rs 2 crore without generating any business. The example is not presented as an actual Zerodha data point, but as a stress-test of how the pricing model behaves. It highlights a key fear: costs scale with transfers, not with trades. The scenario also underscores why brokers are focused on the cap, because a high cap can still allow meaningful charges on large transfers. Online discussions around the example largely focus on whether brokers would change pricing, limit UPI use, or pass charges on to users. Kamath’s argument is that the structure should prevent these distortions rather than create them.

Why free equity delivery brokerage is being mentioned

Kamath explicitly linked the proposed MDR to the possibility of charging for equity delivery trades. He said Zerodha currently does not charge brokerage on equity delivery trades because the economics allow it. However, he added that if every UPI transfer starts carrying an additional cost irrespective of whether the customer actually trades, he does not see how brokers can absorb it indefinitely. This is being read by many retail investors as a signal that business models built on low or zero delivery brokerage could face pressure. The point is not that delivery brokerage will definitely change, but that a new recurring cost can force firms to revisit pricing. For the broader broking industry, any shift in delivery pricing would be visible because delivery investing is a core retail use case. The discussion also touches on quarterly settlement rules, which Kamath referenced as another factor that can complicate cash movement and associated costs. Overall, the message is that payment policy can flow through to end-investor charges.

The structure Kamath supports for broking MDR

Kamath has said he supports the introduction of MDR on UPI, but wants a different structure for broking. He suggested an MDR of around 0.02 percent with a cap of Rs 5 or Rs 10 per transaction for broking transfers. His focus is on reducing the cap sharply from the proposed Rs 300 for capital-market related UPI payments. He also acknowledged that even a lower MDR does not fully solve the problem of customers transferring money without transacting. Still, he described the smaller cap as “much more reasonable” for the industry’s cash-management reality. The suggestion implies that for broking, the cap matters as much as the rate because transfer sizes can be large. It also implicitly aims to keep UPI as a low-friction funding method for retail investors. The broader ask is a category-specific approach rather than a one-size-fits-all MDR.

MDR rates and caps being discussed

Social media posts about the issue have repeatedly contrasted capital-market MDR terms with other sectors. The key details being shared are summarised below, based on the framework and the proposal Kamath put forward.

The contrast has fuelled debate about whether broking should be treated more like a utility flow than a discretionary merchant purchase. Kamath’s view is that the proposed cap is misaligned with the realities of repeated client funding transfers. The table also helps explain why some investors are focused on the cap more than the percentage. Large-value transfers can hit the cap quickly, so the cap becomes the effective price point in practice. This is why the Rs 300 cap has drawn criticism even alongside a low percentage rate.

What investors are watching next

The immediate market relevance is not limited to Zerodha, because the underlying payment rail is widely used across brokers. Investors and traders are watching whether brokers introduce convenience fees, change free delivery policies, or redesign how clients add funds. They are also watching for any refinements that create a separate MDR framework for broking payments, as Kamath has requested. Another watch point is how the MDR applies in common funding patterns, where clients may top up multiple times and trade only occasionally. The debate also reflects a broader policy trade-off: funding a payments network while keeping it affordable for high-frequency, low-margin use cases. For retail investors, the practical question is whether the cost stays invisible inside brokerage economics or becomes an explicit fee. For the industry, the question is whether the structure recognises that a fund transfer is not the same as a completed transaction. Kamath’s posts have helped frame the issue in those simple terms, which is why it has stayed in the spotlight.