Brilliance of UPI & its Next Chapter (#60)
UPI's evolution to dominate retail volumes, and its big steps towards international, credit cards, and offline transactions
Welcome to the 60th issue of Unit Economics. In this article, I revisit all that makes UPI tick and discuss some critical developments within the space. Dive in!
The ideal payment experience is when you almost don’t notice it. And the slower it feels, the more you notice. Which is why cash keeps doing well. Keep aside the anonymity, and we will find the speed & finality of cash payment as its strongest suit.
Digital and bank-paper based methods have long attempted to replicate this experience. But with banks and digital interfaces, the distance between the payer and the payee stretches longer with each middleman. And this can get tedious. A credit card payment, usually simplified into a four-party model, can have fifteen or more hops to cross from initiating the authorization to completing the settlement.
This is not by chance. Financial institutions choose to prioritize (A) quick authorization, which allows for a handshake at the point of sale, and (B) trust in clearing and settlement, which may then happen with a delay.
As with all things money, the institutional focus on maintaining a high success rate (SR) is primary across all three legs: authorization, clearing, and settlement. In our dream for a cash-like digital experience, the criticality of a high SR as a trust factor acts as a limiter. And the result is an ecosystem restricted to a T-plus-some-days mode of payment for decades.
Fortunately, in India, the Reserve Bank of India (RBI) developed a strong muscle to mend that early in the 21st century.
The central bank launched Real Time Gross Settlement (RTGS) in 2004, allowing large-ticket inter-bank transfers to be settled in real-time and on a one-to-one basis, with the processed transactions deemed final and irrevocable in nature.
A year later, similar convenience was imagined for retail and corporate customers who dealt with lower ticket sizes through the National Electronic Funds Transfer (NEFT) system. NEFT launched in Nov 2005 and, instead, operated on a batch settlement cycle that ran at hourly intervals, appearing close to real-time.
The centralised and almost real-time settlement payment systems were a big leap toward the desired experience. Eighteen years later, RTGS (73.7%) and NEFT (16.5%) combine to process over 90% of all digital payment transaction value in the country, making up roughly the entire bucket of corporate transfers and a good share of the large ticket retail transactions. In this while, RTGS evolved to operate around the clock (24x365) and is a leader in large value payment systems across the world.
On paper, these fund transfer systems took us closer to the dream. But explain that to an average customer. At the point of sale, RTGS or NEFT seem inconsiderate. They require time, correctness, and a lot of patience.
Think of a NEFT transaction process for example.
Naturally then, cash dominated at the point of sale. And cards, despite higher settlement times, were preferred next due to the convenience of swipe and the comfort of quick authorization they offered.
To visualise the state better, imagine a payments landscape where you would distribute your savings between cash and a bank account. Now, if you are thinking about making a large payment or are in a hurry to remit it somewhere, you would have to rely on the centralised funds transfer systems.
If instead, you were making a point-of-sale purchase from your everyday merchant, you would have to either carry your wallets filled with enough cash or inconvenience the merchant with a longer settlement time through cards, assuming the merchant would even accept card payments.
This need-basis payment method switch was the fragmented reality until the mid-2010s.
The holy grail, however, lay in a cash-like payment experience that could (A) be easy for customers to perform, (B) settle money quickly, and (C) be replicated at the smallest of merchants. All at the same time. This stayed aspirational.
The Unified Payment Interface (UPI) changed that. UPI was not the first easy-to-use digital payment method. There were digital wallets and cards. It was neither the only one that allowed money to be settled quickly. The funds transfer systems of RTGS, NEFT, and IMPS had been doing that for over a decade.
But it was the first to do both. Additionally, the QR form of payment & the MDR regulations helped achieve the third goal of large-scale merchant acceptance. And in UPI, we then had a payment method that came close to the cash-like digital experience that we had imagined. All with the added benefit of never requiring the money to be withdrawn from your bank account.
In mid-2016, UPI launched with 21 live member banks. It was only rational to expect that it would work. Yet, no one could imagine how remarkable the speed and scale would get.
14 months post-launch, UPI had already hit monthly transaction volumes of 100 Mn. Add 24 more months, and by Oct-19, monthly volumes had increased 10X to reach 1,000 Mn. The domino continued: 2,000 Mn in Oct-20, 4,000 Mn in Oct-21, and 8,000 Mn, now, in Jan-23. Unstoppable.
Today, UPI powers monthly transactions worth ~₹1300 Bn and constitutes more than 64% of all digital payment volumes in the country. Interestingly, 2/3rd of all UPI value continues to be relatively large-value P2P transactions (more than ₹2,000) as shown below.
You would think that after seven years, the numbers would appear simpler. But even in Jan-23, UPI showed 74% YoY growth in volumes and 56% YoY growth in value!
The ingenuity of the payment service aside, a good part of the credit must go to the NPCI for constantly building and innovating for important use cases, including making QRs interoperable, allowing bill payments through BBPS, recurring payments via Autopay, access to low network transactions through UPI Lite, USSD transfers using *99#, among many other initiatives.
In short, the NPCI always found ways to keep the UPI ecosystem busy. And it is no different today. With so much going around, we may have taken UPI for granted and I thought it useful to take a few steps back and attempt to understand what is going on in the space. I will summarise some highlights that feel most critical.
Volume cap for TPAPs at crossroads with UPI goals
The NPCI had sought an aggressive 30 percent volume cap for all Third-Party App Providers (TPAPs) in Nov-20, given the dominance of PhonePe and Gpay. The initial plan was to enforce the guideline by Jan-21, but it has seen been extended multiple times with the last extension putting the timeline at 31-Dec, 2024.
Why has it been extended again?
In the circular, the NPCI wrote that “in view of the significant potential of digital payments and the need for multi-fold penetration…imperative that other existing and new players (banks and non-banks) shall scale-up their consumer outreach for the growth of UPI….,”, suggesting that the volume cap stands against the more critical growth objectives of UPI. Further, the circules pushes the onus on non-leading TPAPs to increase their market share in driving the market cap goals, offering some flexibility to the market leaders.
It appears that with the UPI volume share for PhonePe at ~50% and GPay at 33% - the guidelines are unlikely to be met even by Dec 2024. But as long as the ecosystem continues to innovate, I presume that the NPCI would want to prioritise the growth of UPI over the volume cap guideline. Which may mean that the caps would either see further extensions or relaxations.
Crossing the borders with UPI
With PhonePe introducing UPI International this week in a few countries, the UPI is slowly bringing its global ambitions to reality. Over the last couple of years, NPCI had extended partnerships with central banks and groups across Bhutan, Singapore, South-East Asia, Nepal, UAE, and Europe, among other geographies, with the vision to allow Indians the ease of using UPI for international travel and for making foreign remittances less costly and more efficient.
This is a big development that has been announced rather silently. I imagine that the roadmap for the year will want to focus on three parts, (A) stability of existing infrastructure for countries where international UPI payments are live, (B) adoption rate in Singapore, UAE, and other important travel corridors, and (C) integrating international payments with other groups/countries.
I believe that this is too early to draw conclusions on the initiative, but if UPI is able to deliver the same experience on international payments as it does for domestic, then given the current hassle of currency conversion, managing cash, etc. for foreign transactions and the affinity towards UPI of the travelling population – UPI can potentially gobble up more than a chunk of the international travel market in quick time.
Offline Merchant Acquisition & Transactions
Some teams, incl. CRED, are jumping on the UPI opportunity, with a focus on scaling their offline acquisition of merchants. Offline acquisition can be costly, but it a good way to acquire fast and build brand perception.
For others more evolved such as Paytm & PhonePe, they see acquiring opportunity in adopting solutions that help them reach users in no/low network areas.
The latter initiative is in line with the RBI’s push for Facilitating Small Value Digital Payments in Offline Mode, and may turn out to be a slow burner for a few years. On a P&L level, offline or low network payments are likely to be a cost-center but may appear a necessary evil, by helping UPI apps maintain feature parity with competitors & by buying some brownie points of the regulators, if nothing else.
Credit Cards on UPI may be a win-win
After some noise, credit cards on UPI functionality is live on certain platforms, with the cards on the RuPay network being the first to be allowed to be linked to UPI IDs. Unlike the bank accounts linked to UPI IDs, the linked credit cards will only be allowed for transactions on merchant VPAs.
For customers, the money will continue to be charged to their credit card accounts and the merchants may discourage accepting UPI payments via credit cards due to the MDR expected to be charged on such transactions.
Another new area to draw conclusions on, but I will be keeping an eye out for a couple of developments.
With the battle for market share and volumes still hot on UPI, this functionality presents a new battleground to fight over. I believe that the UPI apps may want to offer incentives to users to link credit cards on their platforms first, and it will be interesting to analyse what they come up with.
RuPay continues to be a small fish in the credit card market, with some reports (quite surprisingly) putting the number as high as 20%. With rumors of Visa and Mastercard joining the band, the plastic cards may be threatened for the wallet share, and the UPI apps may become instruments of distribution & engagement for CC issuers.
Final few thoughts…
UPI is going from strength to strength and making big dents wherever it goes. A lot of it is down to the proactive nature of the NPCI and the central bank, who have fast supported each of the new initiatives that we have discussed above.
I feel particularly excited by the gaps that credit cards and international payments on UPI may fill, and how issuers and customers may react to them.
We will surely revisit this topic sometime again this year, but I hope this has helped paint a healthy & exciting picture of the thriving UPI ecosystem. Until next time!
If you have any views or feedback to share on the topic, feel free to add a response below or to share your thoughts with me over Linkedin. In case you feel your friends or family would be interested in reading about payments, feel free to share the blog with them as well. See you in a few weeks!
What do you make of the fact that the cash-circulation to GDP ratio is now higher than it's long-term average?