Heard this in one of Deepak Shenoy's podcasts (CEO of Capitalmind PMS) that UPI should be free because it enables banks to earn more float income. This additional income is more than enough to offset the costs to run UPI infra. Your take?
Appreciate you reading! Also, do share the link of the podcast: I don't think I've heard it!
I think this works from a bank perspective, but this money then also has to be distributed to your UPI Apps (TPAPs), PAs, and other infra layers that are actually driving innovation and usage of UPI. If we just focus on banks, then this goodness may not flow to the rest of the ecosystem, which is the problem, and these players are the one who will innovate on things such as alternate auth (biometric etc), customer experience etc.
So if banks can earn this income, but there is some structure where this money that is earned flows back to these other players, then it makes sense. Right now there is an incentive structure outlined in the Budget, but this has been gradually reducing over the years: From 2k Cr in FY24 to now ~437 Cr in FY25.
Somehow the fintechs have to earn money through whatever model that comes in: it can be subscription to UPI services that some folks pitch. It can be the float income, that is distribute it. The current incentive structure, or through MDR.
My view is: why overly complicate it? There is an established structure of MDR that cards, and global instant payment providers such as Pix use. Just use that, and reduce the MDR cost, its easier for stakeholders to understand, and therefore implement. The converse here is that then this may flow to the end customer, who ends up bearing the cost, which may not be what NPCI wants just yet, but that is something that can be figured IMO.
How would monetizing UPI transactions work for Groww and Cred though, since they're not really processing payments to merchants but instead (mostly) payments done by customers to these fintechs' UPI accounts for transferring the $ for share market purchases/credit card bills (Groww and Cred respectively)?
My POV: Credit card bills / share purchases ideally should not be charged, since the customer is transferring money from their own account: its a P2P sort of transaction, except here the "P" is the same individual.
In the case of share purchases, a case could be made, that since something is being bought, the platform that is facilitating this (a zerodha, HDFC securities etc) becomes the merchant, and some % MDR should be paid by them.
I see, thanks for your response. Another thing - in the calculations for MDR slide in your article, the network is who exactly? NPCI? Also, the acquiring and issuing bank don't need to be compensated to the level ordained by credit card networks, right? Simply because there isn't as much risk in UPI.
Finally, who's maintaining the tech stack that you refer to in the article? The 40p costs that go towards maintaining the network - is that paid to NPCI currently?
The network is UPI, think of it equivalent to card rails such as mastercard, VISA etc. NPCI is the org that oversees UPI. In terms of compensation: pricing of CC vs saving accounts payments is not based on risk, its just the fee structure that has been set up for a payment rail. There isn't risk in payment routing for credit cards: the risk is borne and locked in by the issuing bank at the time of underwriting the CC for the customer. Why UPI and other RTP networks are charging less, is so that there is less burden on the merchant, and its used as a real time payment method over cards. It increases acceptance.
NPCI covers its cost of maintaining this infra through these sources, but since banks / fintechs/ TPAPs cannot charge for this, that operational cost of maintaining their own infra, hitting the APIs etc is something that eats into their bottomline
Agree. End user chargeability will always be kept out of it.
But some pricing will have to come in for bigger merchants, otherwise this will never make money, and innovation will never really come in, especially not that top UPI Apps have reached critical mass. Hence my view: P2M transactions, for AoV > 2k, which is already coming in for PPI on UPI
Definitely agree with your take. If banks are making money, it should flow to fintechs who make the UPI infra possible.
Here's the link of the episode: https://open.spotify.com/episode/2gqJNgxEa6nLMt5ZPCjXI7?si=7khjXDrdQl6-OskOL5WT1w
Heard this in one of Deepak Shenoy's podcasts (CEO of Capitalmind PMS) that UPI should be free because it enables banks to earn more float income. This additional income is more than enough to offset the costs to run UPI infra. Your take?
Great article, as always!
Appreciate you reading! Also, do share the link of the podcast: I don't think I've heard it!
I think this works from a bank perspective, but this money then also has to be distributed to your UPI Apps (TPAPs), PAs, and other infra layers that are actually driving innovation and usage of UPI. If we just focus on banks, then this goodness may not flow to the rest of the ecosystem, which is the problem, and these players are the one who will innovate on things such as alternate auth (biometric etc), customer experience etc.
So if banks can earn this income, but there is some structure where this money that is earned flows back to these other players, then it makes sense. Right now there is an incentive structure outlined in the Budget, but this has been gradually reducing over the years: From 2k Cr in FY24 to now ~437 Cr in FY25.
Somehow the fintechs have to earn money through whatever model that comes in: it can be subscription to UPI services that some folks pitch. It can be the float income, that is distribute it. The current incentive structure, or through MDR.
My view is: why overly complicate it? There is an established structure of MDR that cards, and global instant payment providers such as Pix use. Just use that, and reduce the MDR cost, its easier for stakeholders to understand, and therefore implement. The converse here is that then this may flow to the end customer, who ends up bearing the cost, which may not be what NPCI wants just yet, but that is something that can be figured IMO.
Same logic for B2B payments? Transaction limits getting liberalized over time? UPI for B2B is still up for grabs (subject to transaction limits)?
Supreme value provided btw!! Thank you!
How would monetizing UPI transactions work for Groww and Cred though, since they're not really processing payments to merchants but instead (mostly) payments done by customers to these fintechs' UPI accounts for transferring the $ for share market purchases/credit card bills (Groww and Cred respectively)?
My POV: Credit card bills / share purchases ideally should not be charged, since the customer is transferring money from their own account: its a P2P sort of transaction, except here the "P" is the same individual.
In the case of share purchases, a case could be made, that since something is being bought, the platform that is facilitating this (a zerodha, HDFC securities etc) becomes the merchant, and some % MDR should be paid by them.
I see, thanks for your response. Another thing - in the calculations for MDR slide in your article, the network is who exactly? NPCI? Also, the acquiring and issuing bank don't need to be compensated to the level ordained by credit card networks, right? Simply because there isn't as much risk in UPI.
Finally, who's maintaining the tech stack that you refer to in the article? The 40p costs that go towards maintaining the network - is that paid to NPCI currently?
The network is UPI, think of it equivalent to card rails such as mastercard, VISA etc. NPCI is the org that oversees UPI. In terms of compensation: pricing of CC vs saving accounts payments is not based on risk, its just the fee structure that has been set up for a payment rail. There isn't risk in payment routing for credit cards: the risk is borne and locked in by the issuing bank at the time of underwriting the CC for the customer. Why UPI and other RTP networks are charging less, is so that there is less burden on the merchant, and its used as a real time payment method over cards. It increases acceptance.
In terms of maintaining the tech stack: NPCI maintains the UPI tech stack, that banks and other stakeholders such as TPAPs etc use. The 10p - 40p is a directional cost of 1 UPI transaction. NPCI manages this by taking some interchange fees, and some sort of membership fees from banks and fintechs is what I understand (you can check out this article here: https://economictimes.indiatimes.com/tech/technology/npci-posts-37-jump-in-fy24-net-profit-to-rs-1134-crore-revenue-up-42/articleshow/113291676.cms?from=mdr)
NPCI covers its cost of maintaining this infra through these sources, but since banks / fintechs/ TPAPs cannot charge for this, that operational cost of maintaining their own infra, hitting the APIs etc is something that eats into their bottomline
Habituated free digital payments are very hard to be charged now.
Agree. End user chargeability will always be kept out of it.
But some pricing will have to come in for bigger merchants, otherwise this will never make money, and innovation will never really come in, especially not that top UPI Apps have reached critical mass. Hence my view: P2M transactions, for AoV > 2k, which is already coming in for PPI on UPI