UPI Now, Pay Later

The next paradigm in consumer credit and a critique of ‘Neo Banks’

I hold one fintech belief very closely: the scale and innovation lie in Payments, and the deep profit pools lie in credit. This does not mean that payment businesses are any worse off — value creation in these businesses works entirely differently. What is important to note, though, is that the aggregate value creation opportunity in credit is several times larger than in payments. There is a corollary to this as well — the aggregate value destruction opportunity in badly-run credit businesses is potentially as large as the value creation opportunity, but we will leave the obituary writing for another day.

The central tenet of this blog is that there is an emerging opportunity — probably for the first time at scale in India — at the cross-section of payments and credit. This will have a lasting impact on the world of consumer payments and consumer credit as we know it. I break down this thesis into 7 episodes, starting with some context setting.

Context and Perspective on Consumer Payments

There is one well-established truth about Payments in India: Unified Payment Interface (UPI) is a behemoth. It has revolutionized digital payments in India, and as a modern money movement network, UPI is way ahead of its global peers, it achieved a phenomenal milestone of 10 billion transactions in August 2023. The greatest success of UPI lies in its design — it’s a free public good. I am a good old capitalist, but I believe utilities and public goods need to be regulated and need strong price controls, as they tend to end up becoming monopolistic. In the modern world, there is no more important public good than the money movement network, and the rest of the world seems primitive without a network like UPI. However, what do I know — Visa is almost a US$ 500B payment network and all I can do is write blogs!

Being a public utility allows UPI to have a unique business model. It’s free, and it’s transparently free because it’s instantaneous. I believe this business model ensures universal acceptance among merchants, which has been the single most important point of success for UPI.

Here is a sidebar: Indian merchants don’t like to pay a discount rate for accepting payments. It’s not that they cannot afford an merchant discount rate (MDR); they can, there is a deep-seated irritability with parting one’s money as MDR.

I believe UPI would become less ubiquitous if there were an MDR associated with UPI transactions.

Now with this preface, the first thing I attempt in this blog is to envision a world where Credit on UPI becomes ubiquitous. You know, UPI Now, Pay Later. What would this world look like? Is there room for innovation and value creation? Is this the next big wave in consumer credit or is it just a fad?

Boom and Bust Cycles in Consumer Credit

In the world of consumer credit, there have been several waves. I want to start with what many consider the first Short Term Personal Loan (STPL) boom, all the way back around 2006. The reason I have vivid memories, and this is a personal favorite topic of sorts, is that this was a topic of discussion during my business school interview, and I have always tried to keep pace with the ups and downs in the consumer credit world ever since. Back to the story — the boom that started in 2006, picked pace in 2007 through 08 and came to a screeching halt by early 2010. Part of it was linked to the GFC and the bubble that preceded it, and part to the absolute lack of any Know-Your-Customer (KYC) or underwriting norm being followed by the then very aggressive non-banking financial companies (NBFC). The key takeaway of this four-year cycle, looking back, was that there were maybe 3–4 NBFCs that came out of it bruised, but stronger, and were able to invest and have now created behemoths in the Financial Services world.

This phenomenon has applied to various forms of consumer credit again and again over the past 17 years since my story began. This has happened in Microfinance, Gold Loans, Affordable Housing Finance, Consumer LAP, and so on. The last iteration of this phenomenon started with the wave of Buy-Now-Pay-Later (BNPL) and has now settled in the shape of digital ultra-short-term and short-term personal loans, with maybe three important companies emerging as strong contenders today. We are privileged to have seen the journey at very close quarters as investors in Kissht. As I said, the value destruction opportunity in fintech is tremendous, and there have been several spectacular failures over these years, but the less said about those, the better it is.

Transaction Credit Opportunity

I believe that we are at the cusp of a new wave in consumer credit and that this will take the shape of what I call transaction credit.

I think UPI acceptance becoming ubiquitous is going to play a crucial role in transaction credit becoming the next wave in consumer credit. A lot of money has been spent on making UPI the ubiquitous platform it is today, Credit on UPI is where the monetization happens.

The converse also holds — in the absence of UPI, transaction credit would likely see a much slower uptick, we will just see credit card issuers issue more cards to a narrow set of customers, and utilization levels (spend per user) will not ramp up from current levels. What this effectively means is that transaction credit players in the Indian market will need to adopt a very different model as compared to credit card players, the most significant difference coming from the extremely low (no?) MDR on UPI transactions. More on this in E06.

The why now of transaction credit needs to be mentioned briefly. Most of this relates to incremental growth in per capita disposable income which means that we now have a critical mass of aspirational users who are saving less and spending more. At a larger scale, this catches up, and with tomorrow looking rosier, credit starts to embed itself in the equation. Transaction credit fits extremely well in this equation — instead of taking a line of credit and splitting it into smaller chunks, the user spends based on their convenience and can revolve the aggregate into equated monthly installments at risk-governed rates.

User Segmentation and Behavior

I believe that the transaction credit on UPI opportunity will coincide with the credit card opportunity, and there is a need to understand this from a user segmentation standpoint.

The chart below (roughly) breaks up the user base into four segments based on two key characteristics I believe are the most relevant to defining user behavior. The underlying story is a lot more complex, but we are writing a blog here and not building a business.

First a few definitions/descriptors:

• The transactor almost always pays the (equivalent of her) Credit Card bill on time, whereas the Revolver builds balance over time, intending to pay it all back at a later point in time (when hopefully things improve a lot).
• The bargain hunter will move from instrument to instrument and platform to platform to find the best deal, whereas the loyalist — while looking for the best deal — would do less of this. As a result, Bargain Hunters will want to have access to as many credit instruments as possible.

Before we talk about the user segments, I think it’s important to get some ground realities out of the way; here is the most relevant data:

• India has about 93 million Cards In Force (CIF). Based on estimates, these cards have been issued to ~45–50 million users. This means the average card user has 2 cards.
• The CIF base is growing at ~1.8–2.0 million cards per month. This is the fastest it has ever grown, and the pace is expected to increase.
• Public Sector banks (except for SBI Card) add a total of ~330k cards per month, which means their already tiny market share is shrinking.
• The Big 5 card players: SBI Card, HDFC Bank, ICICI Bank, Axis Bank, and Kotak now have a lion’s share of users, and strong momentum — they add between 150–350k cards per month at a very large scale. A clutch of smaller players — particularly IDFC First, RBL, Canara, Bank of Baroda, IndusInd, and Yes are starting to ramp up card programs, but their commitment to the long term remains to be seen (more on this subsequently)

I believe all these four segments will — over time — get access to transaction credit in the form of a card, or via a UPI-only instrument. Over a longer period of time, all these users will have access to multiple credit instruments, and at that time, the spend per instrument will become a key success factor.

I do think there is only one segment where the Credit Card, in its current form, wins over Credit on UPI. In my view, going forward, the Credit on UPI opportunity is significantly larger than the Credit Card opportunity. There is an obvious convergence, RuPay Credit cards will offer UPI out of the box. It is the MDR that matters more than the convenience of using UPI, and I argue this more in E06.

The easiest determining factor for driving spending on an instrument has been rewards, but as we will see in a subsequent section, rewards will not be an easy option on UPI-led transaction credit platforms.

A Critique of Neo Banks

Neo Banks, or startups ‘aiming to disrupt banking for millions of Indians’, have been a sort of holy grail investment in the Fintech VC world. We are talking consumer Neo Banks, and at my last count, there are more than 20 of them that have raised nearly a billion dollars over the past five years. That’s a lot of money, but then startups in India raised a lot of money in general over this period. The critique here, though, is that there really hasn’t been a business model other than credit that seems to be sticking.

With that said, I have somewhat of a traditionalist view of credit. It’s not a growth business, but rather a risk business. I also think that millions of Indians do not need a disruptive bank, they just need a functional one. This has several cascading implications:

• Disruption typically means feature-rich. While many customers are keen to experience these features, they are not willing to pay for them.
• Disruption as a value prop is an expensive one when it comes to acquiring customers. Two things create customer pull in the world of financial services — (a) being there at the time of a need and (b) brand. Disruption as a brand value prop takes years to establish, and a much shorter period to become a commodity. And being there at the time of a need means running a ton of ads and suffering a low click-through rate. In a nutshell, in Financial Services, Disruption means high CAC.
• Unlike in the case of other startup categories, I want to bank with my mother’s bank. Banking is a serious business, and parents often have a significant impact on how young millennials bank. And Neo Banks often don’t get to play a role in this conversation.

As a result, the NuBank of India will take a few more years (decades?) to build. In the meantime, the competition in the consumer credit space will continue to increase as well-funded neo-banks start to issue transaction credit instruments as Credit Cards or as Credit on UPI instruments.

Core Tenets of Transaction Credit

As you read through this section, note that there is a central assumption I make. It is that for Credit on UPI to succeed, most transactions need to be MDR-free. I will revisit this assumption and write a follow-up in few months, but this is my strongly held point of view for now.

A: Zero or near-zero MDR

I think the Credit on UPI revolution lives and dies by this tenet, and I will try to describe it.

If a small merchant — think the small electrical shop behind my dad’s house — receives a payment on UPI, he expects to receive 100% of the money in his account. If the issuer starts levying a 1.1% MDR on this transaction (assuming it settles using a credit on UPI loop), the merchant starts turning off credit transactions in a heartbeat. I actually think the RBI will start asking merchant acquirers to turn off credit on UPI acceptance on all merchant QR codes by default. And if that happens, Credit on UPI is dead on arrival.

So how does Credit on UPI work? I think for Credit on UPI to succeed, the interest rate needs to be high enough so that issuers can offer an MDR-free transaction. This is not simple — an MDR-free transaction means a significant negative carry for the issuer. In the case of a credit card, the issuer is paid by the merchant for the credit-free period offered to the user in the form of an MDR.

There is a second-order effect — in the case of Credit on UPI, the revolvers must pay for the credit-free period enjoyed by the transactors. Therefore, assuming a current split of transactors and revolvers (see SBI Card, the ratio is ~35–65), the interest rate on revolve needs to go up by as much as 18–24% to get to credit card equivalent economics, which means an APR of 50% + for the revolving user.

B: Rewards need to be re-thought

This brings me to the user segmentation. I believe Credit on UPI needs to be offered only to revolving customers. Adding transacting customers to bump up card spends and earn MDR does not work in this world.

Which means rewards need to be rethought. In the world of credit cards, the ground reality is that revolvers pay enough interest for some of it to be farmed back into rewards that are mostly used by the transactors. In the world of revolver-only instruments, similar rewards don’t really work. The user segment changes and rewards need to be rethought ground-up.

This has a significant impact on user retention, and on spends prioritization for customers.

C: Customer engagement takes place on payment platforms

In the world of Credit on UPI, user acquisition happens at two levels. First, the fintechs acquire users and issue them a credit instrument. Second, the payment platforms acquire users and enable them to pay at the point of purchase.

In my view, users will not use the issuer platform as their payment platform. To illustrate, I have never seen anyone open their HDFC bank app to pay at the neighborhood store. We (predominantly) use Google Pay, PhonePe, or Paytm. In the world of Credit on UPI, the user will simply add her credit instrument to her payment app of choice, and just continue her merry ways.

Now you may say that the fintechs issuing these Credit on UPI instruments are no HDFC bank, but I submit that they are no PhonePe either, and I bet that winning a user from PhonePe is just not going to be easy. I contend that the battle is lost, and it’s best not to invest in trying to ensure that the user uses your app as a payment platform. Gamification and rewards tend to be very effective, but I wouldn’t count on the user opening the issuer app a few times a day.

D: CAC is crucial

Yes. Fintechs that cannot solve for CAC don’t make it. We saw this in BNPL, we will see this kill many Credit on UPI platforms over the next few years.

E: Collection muscle unclear.

Yes. Fintechs that are built with collections as an afterthought don’t make it. We saw this with a host of personal loan platforms, we will see this kill many Credit on UPI platforms over the next few years.

A Word of Caution: Debt Trap

Household financial assets in India are at their lowest ever. The gross savings rate (Savings to GDP ratio) has declined at a rapid clip — from ~37% in 2011 to ~30% in 2022. Now sure, part of this is linked to an increase in the value of real assets and investments, but it’s easy to see that spending has increased at a rapid clip, as has credit. This shows up in household credit, which nearly doubled last year. Again, some of this goes into real estate, but I would contend that a fairly significant proportion goes towards other personal expenses, a sign of increasing aspiration for a large spending consumer segment.

Now this rapid increase in indebtedness is not unique to India, there is an increase in leverage across the world, most notably in the US. To be fair, an increase in indebtedness is in some ways necessary to improve the standard of living in India. India has always been a consumption juggernaut, and credit allows consumption to grow even faster. Simplistically, if tomorrow is going to be better, we can always grow into paying off the debt, and that remains the most likely scenario.

However, contagion across markets works in funny ways. The Indian consumption economy has been one of the most resilient across the world — not been dramatically impacted by global macroeconomic fluctuations for over a decade now. Importantly, the consumption economy looks at its best — people are spending more all around us. However, there is a note of caution that I need to point out — a lot of this consumption is credit-fueled. How domestic credit behaves in the wake of a global credit meltdown — which looks more of a likelihood every passing day — is anybody’s guess, and we could see a temporary cycle in small-value personal loans and transaction credit.

However, the long-term story is clear as day. Retail credit penetration in India will continue to go up, and there remains a tremendous opportunity to create new, disruptive models in transaction credit.

PS: if you got to this point, I appreciate the effort, you must really want to understand fintech. I will say though, I am unapologetic about the title. It needed to be this to get the conversation going!