I recently read the following tweet:

I replied as follows:

Well, UPI is also part of Banking! (For the uninitiated, UPI is run by scheme operator National Payments Corporation of India, which is jointly owned by India’s leading banks.)

To which the OP replied back as follows:

Yeah, but much of UPI also involves banks losing the customer relationship in payments to third party apps. It’s unbundling of the payments relationship from the storage of money, by design. Which bank UPI service has been successful?

Well, that’s a feature, not a bug!

—–

Account-to-Account Real Time Payments like UPI eliminate float. Since float is a source of revenues for banks, you won’t find banks rushing headlong into an A2A RTP system anywhere in the world.

Nationwide A2A RTPs exist in 40+ countries in the world. Virtually all of them are result of regulatory diktat.

India is no exception. The Indian regulator was easily able to achieve forced distribution of UPI among banks since 70% of the banking industry is owned by the government.

In UK, where virtually all banks are in the private sector, it took two years for the regulator to “persuade” banks to join FPS, the local version of A2A RTP, in 2008.

American regulator(s) have still not succeeded in this endeavor despite making persistent attempts for nearly 10 years. (It may be a good time to mention that there are nearly 4500 banks in America.) That said, USA does have limited A2A RTPs like Venmo, Zelle, et al for several years.

Traditionally, Consumer Acquisition (acquiring of consumers who use payment cards and apps to make payments) is profitable for Banks but Merchant Acquisition (signing up of merchants to accept payment cards and apps) is a highly resource-intensive and low margin activity. This is true not only for payment systems like Debit Card with low-to-no Merchant Fees a/k/a Merchant Discount Rate (MDR) but also for ones like Credit Card that have 2-3% MDR.

By maintaining an optimal mix of Credit Card (quite profitable), Debit Card (not very profitable) and A2A RTPs (not at all profitable) businesses, Banks have managed to make reasonable profits in their payments business as a whole. Note that I’ve said “optimal” in the above sentence. For reasons that will become clear soon, it’s not “maximum”.

Enter #ZeroMDR in India. According to this new regulation, banks can’t charge any merchant fees for UPI payments (among others). They tried to compensate for that loss by levying nominal charges on consumers for using more than a certain number of UPI payments per month but, AFAIK, even that measure got shot down by the regulator (or the court of law).

As a result, Customer Acquisition and Merchant Acquisition have both become loss making. Even one quarter of loss raises eyebrows in the banking industry, which operates on the traditional PLBS business model. Ergo, UPI has stopped making much business sense for banks on the acquisition side.

OTOH, a high population, disproportionately large number of micro and nano merchants (70M for a ~$3T GDP of India versus 8M for a ~$20T GDP of USA) and the other factors highlighted in Don’t Go Global Without Cracking The Value Proposition For Foreign Markets provide hockey stick traction for UPI business. As a result, Fintechs engaged in UPI Customer Acquisition and Merchant Acquisition have become prime candidates for Venture Capital. VC-funded Fintechs work on modern valuation based business model, where losses are acceptable for a very long time (some would say forever). Ergo they have a seemingly-bottomless pit of money to fund cashbacks for customer acquisition and can afford to deploy feet-on-street for merchant acquisition.

Which banks can’t or won’t. Due to regulation, they can’t pull out of UPI business.

But they can minimize their exposure to a loss-making business by ceding Customer and Merchant Acquisition to Fintechs. Which is exactly what they have done. UX of banking websites and apps has nothing to do with it.

As long as fintechs get funding, they can remain aggressive in this space.

It’s been a win-win for all concerned – as the burgeoning volumes and values of payments processed by UPI stand testimony.

Except interoperability. More on that in a bit.

—–

Now let me come to the part in the OP that seems to insinuate that banks will become “dumb pipes” by losing the customer and merchant relationships in UPI to third party apps.

Well, the “dumb pipe” theory doesn’t work in banking, especially in a multi-corner network business like payments. Visa has no relationship with cardholders or merchants but is one of the largest financial services technology providers on the planet by revenues, profits and valuation. Ditto MasterCard.

Probably because they’re behind the scenes, the issuer and acquirer relationships owned by Visa (and MasterCard) are often underrated in the popular narrative but perhaps more than anything else, they’re responsible for creating the much-reviled retail payments duopoly.

Meanwhile, Banks in India are doubling down on their Visa and MasterCard credit card and debit card business. Since that’s not subject to #ZeroMDR regulation, it’s still profitable and sits well inside banks’ PLBS business model. See section titled “Wrest Control From Bureaucrats” in my blog post How RuPay Can Disrupt Visa And MasterCard for more details.

—–

One of the founding principles of UPI is “interoperability”, which stipulates that anyone with any UPI app should be able to send money to anyone else with any other UPI app by simply using the receiver’s VPA (Virtual Payment Address). For the uninitiated, there are 50+ UPI apps in India, out of which over 40 are from banks. For reasons advanced above, banks haven’t pushed their UPI apps. A vast majority of the market is owned by Google Pay, Phone Pe and PayTM – all nonbank fintechs.

Five years after UPI was launched, if you don’t know what’s VPA or what’s your VPA, I won’t blame you.

Like all startups, fintechs operate on a funding model that is premised on achieving the “Winner Takes All” outcome. Letting consumers use “any app, as long as it’s UPI” does not jell with that business model. Ergo, Fintechs must keep consumers on their own platforms and interoperability is thrown under the bus.

Not surprisingly, not a single leading UPI app makes it easy for their users to find out their VPA. Some even say that they go out of their way to make it hard.

I think @UPI_NPCI should ban and fine ALL UPI apps (that go out of their way) to hide the VPA from users – the basic spirit of interoperability has been destroyed.

I agree about the “hide VPA” part.

But “ban and fine” is silly. UPI is what it is today because of the massive distribution carried out by fintechs to acquire customers and merchants. If you ban and fine them, the whole UPI edifice will come crashing down.

TY. I disagree (not typo!). #ZeroMDR was sop for Merchant. UPI must balance it by letting PSPs soft pedal VPA & interoperability, drive market dominance via Vendor Dependence, whip up frothy valuations, & justify staying in a poor PLBS business.

Since VPA is not widely known, it’s hardly used to make UPI payments.

What people typically ask before initiating a UPI payment is “Are you on GPay?” and, if the answer is yes, they type the beneficiary’s mobile phone number to make the payment. If the answer is no, the next question is “Are you on PhonePe / PayTM?” – not what’s your VPA.

I don’t use UPI often – huge credit card fan! – but, on the rare occasions that I’ve used it to make a payment, I’ve asked the beneficiary for their VPA.  Not a single person has understood my question. As a receiver, I’ve not been asked by a single payer what my VPA is.

In effect, interoperability has been sacrificed and UPI has devolved into a few walled gardens / closed loop payment systems connected to bank accounts.

People are right in lamenting about that…

…UPI was meant to be an interoperable system and forcing people to use your own app to pay another user who is using your own app is just wrong.

But I tend to believe that UPI lost the interoperability plot the moment

it introduced a new thing like VPA instead of using the beneficiary’s mobile phone number

and failed to spec well-formed VPAs.

Doubt if a VPA like johndoe@okicici-ICIC0002390-000502054296-UPI will drive much interoperability, however prominently it’s displayed by a UPI app.

———-

I’m not for a moment arguing that that banking apps have great UX. In fact, I’ve pointed out exactly the opposite many times in the past. Cf. my op-ed entitled “Impact of Regulation on Financial Service Providers” in the Journal of Internet Banking and Commerce for an example of my writing on this topic.

But UX has nothing to do with the position of banks in the UPI consumer app and merchant app league tables. As we saw above, fading away from there is a feature, not a bug, of the business model of the banking indsutry!

Even if banks magically overcome the UX challenges I highlighted in Why Banks Will Never Catch Up With Fintechs, I’ll bet that they won’t bother to catch up with fintechs on UPI customer and merchant acquisition. (Unless, of course, #ZeroMDR reg is repealed.)

Interoperability, not banks, is the loser in UPI.

 

DISCLAIMER: This post is speculative and is the result of my connecting of the dots that I’ve seen in public. It does not purport to provide any inside track into the thought processes, business plans or financials of banks or fintechs around UPI, except if noted otherwise. This post is also not legal or investment advice.