SBI Cards and Payment Services Limited (SBICARD) Earnings Call Transcript & Summary

Ladies and gentlemen, good day and welcome to the Q1 FY 2022 Earnings Conference Call of SBI Cards and Payment Services Limited. [Operator Instructions] Please note that this conference is being recorded. I'll now hand the conference over to Mr. Rama Mohan Rao Amara, Managing Director and Chief Executive Officer of SBI Cards. Thank you, and over to you, sir.

Thank you, Rasuja, and good evening, everyone. On behalf of SBI Cards, I extend a very warm welcome to you, and thank you for joining us today for the earnings call for the Q1 FY '22. My heartfelt gratitude to all of you for continued support and confidence in SBI Cards over all these years, and especially during the current pandemic. Your trust encourages all of us at SBI Cards to explore new horizons, drive the business forward and achieve the vision of our organization. I wish safety and best of health for all of you. I would also like to express my gratitude to my colleagues, who rose to the challenge and diligently followed their professional responsibilities. As you are aware, while FY '21 saw different stages of COVID-19, Q1 FY '22, too, got impacted going to the second wave. Among all ways, the silver lining is an intensified vaccination drive now in force in India with over 40 groups of vaccinations completed, well in line with the government's plan to vaccinate entire 18-plus population by the end of 2021. During COVID wave 2, there were state-specific lockdowns. However, the impact of the same was not as severe as was observed in the first quarter of last year. Also, another positive development is the revival witnessed in June 2021, which continues to be seen in 30 June 2021. As an [indiscernible] organization, we continue to closely monitor the situation. We are taking calibrated measures to build a robust, reliable and a resilient business. Despite the challenging environment, the SBI Cards managed to grow its parts and force and reached a milestone of 12 million cards. We continue to bolster our portfolio. And in a move to augment our premium portfolio, we announced the launch of a new co-brand relationship, Fabindia SBI Card during the quarter. As a measure of abundant precaution, we are continuously calibrating our credit filters for [indiscernible] liquidation. We continued with efforts to offer customers with easy payment options for a period -- over a period at lower interest levels, which have been received well. At the same time, we beefed up the collection efforts to minimize risks. Our employees are diligently pursuing various initiatives to fuel organizational growth. During the quarter, we took measures and initiatives to ensure their safety and well-being. Taking this further, during the quarter, we also organized vaccination drives in different cities for our employees and their family members. It's heartening to share that as a responsible organization, the SBI Cards is strongly focused on driving a positive change in the communities by ensuring their welfare and holistic development. During the period, besides regular initiatives, the SBI Cards also supported efforts to elevate the paying cost by COVID-19. The SBI Cards supported health infrastructure and capacity building at various government hospitals and medical colleges by providing PPE kits, COVID-testing kits, ventilators, setting up oxygen-generation plants, developing COVID care centers, et cetera. Other initiatives included providing dry rations to cook meal, among others. At this point, it is important to emphasize that, backed by our robust business model and learnings from the toughest phase we witnessed last year, we surely are much better equipped and prepared now. I'm confident that we will effectively navigate the business in FY '22 as well. Let me now take you through our financial performance for Q1 FY '22. We had strong business performance in the -- in Q1 FY '22 despite the quarter being impacted by a very strong COVID-19 second wave. Our resilient business model ensures that our key business volume indicators were significantly higher on a year-on-year basis. This reflects on our ability in managing business during COVID 2.0. Our new accounts for the quarter are 609,000, which is up 111% higher than corresponding figure on a year-on-year basis. And as mentioned earlier, our cards in force are now higher than 12 million. Our retail spend grew at 63% year-on-year to more than INR 27,000 crores. Corporate spend grew at 149% on a year-on-year basis to more than INR 6,100 crores. Receivables grew at 5% year-on-year to more than INR 24,400 crores. Our cards market share increased to 19.2% in May 2021 from 19.1% in March '21. Our spend market share has increased from 18.2% in Q4 FY '21 to 18.9% as of May 2021, as adjusted for disclosures by a major bank to stock exchanges. Profit after tax for the quarter is INR 305 crores, which is 74% higher versus previous quarter in Q4 FY '21, but 23% lower on a year-on-year basis. Total income grew at 12% year-on-year basis to INR 2,451 crores. Operating expenses were lower by 9% versus Q4 FY '21. Now moving to asset quality. The credit risk situation continues to be impacted by resurgence of second wave and overall macroeconomic variables surrounding us. To cover ourselves for future credit risk, we provided -- we continue with the management of overlay of INR 258 crores as on June '21. Our GNPA has come down to 3.91% as compared to 4.99% as of the end of March 2021. Net NPA for the period is at 0.88% as compared to [ 1.13% ] as of March '21. As of June '21, close to 60% of the total RBI RE book is less than 30 days delinquent. For the RBI RE book, which is between 30 days and 90 days, the company continues to provide a stage 3 ECL rate of 65.2%, which is equivalent to provision rate that is applicable for NPS. Further, for RBI RE book, which is greater than 90 days, the company has increased its ECL provision rate from earlier 80% to now 100%. ROAA, the return on average asset, for the quarter ended June '21 is at 4.5%, which is higher by 196 basis points as compared to 2.6% for March '21. And ROAE, return on average equity, stood at 18.8%, which is again higher by 761 basis points as compared to 11.2% for March '21. On liquidity and capital adequacy, our liquidity position continues to be strong during Q1 FY '22. Our capital adequacy ratio for period ended June '21 is at 26.1% as compared to 24.8% at March '21. Our Tier 1 ratio has moved from 22.6 -- I mean, moved to 22.6% from 20.9% as of March '21. We continue to enjoy strong credit ratings. Our credit ratings remain excellent with the A1+ and AAA ratings by CRISIL and ICRA for both short-term and long-term borrowings. These strong rate ratings by the rating agencies reflect our robust business and financial fundamentals. So with that, let's open the call for questions. Rasuja, you may please open the call for questions.

[Operator Instructions] The first question is from the line of Anuj Singla from Bank of America.

Sir, first question is with regards to your comment on credit filters, you mentioned that there are some changes you have done given the volatility in the market. Can you talk about that in a bit more quality, what kind of measures have been taken? And secondly, on the restructuring 2.0, what kind of response have we got to the -- in the -- in this -- until the month of May and June? And what kind of escalation we can see in the RBI RE book, let's say, in the next quarter? Or is the work done in terms of restructuring?

Yes. Anuj, coming to the first question, credit filters. This is a trend of continuous activity. I mean as part of portfolio management, we always change the filters. Wherever we see a kind of stress of delinquency higher than an acceptable level, we do make changes. Some broad level changes happened twice last year, one in May and one in November. That has ensured that the new acquisition is of acceptable quality. So apart from the big changes, I mean, some kind of micro-level changes will continue to happen, particularly -- I mean, we do source a sizable segment from Banca, particularly [ NTC ]. So I think recently, we have seen some small increase in delinquency rate. We have tightened the filters, introduced some additional documentation requirements so as to ensure that we are getting the right customer segment. So that way, the changes to credit filters is a continuous activity. Coming to your second question, RBI RE 2.0. Actually, RBI issued the circular in May, and we got the policy approved through both in June. So our dispensation was available only in the first week of June. So we -- by demand, I mean, the initial days, we saw a good demand of around INR 258 crores is the portfolio, I mean kind of RB RE 2.0 restructuring portfolio. We have -- based on the request we received from the customers. And after evaluating this portability and eligibility, we restructured around INR 258 crores, which is, in a way, if you compare our area portfolio was around INR 2,700 crores last year. So that way it is not even 10%. And what we have seen now in the July is like the requests have come down. It is not at the same level. And we don't see a kind of situation of having the same portfolio like last year, definitely. But it's too early to comment. I mean the flow has come down as compared to June.

Understood. And sir, one more, if I may. The cost of funds of 5.2% is -- I mean, it's a very impressive number. What -- how should we see this number panning out in the course of the, let's say, next 3 quarters?

I think to the take up of my finance team, Nalin and his treasury team, they have managed these cost of funds very well. Whatever opportunities were there in terms of kind of shedding the high-cost borrowings or high-cost facilities and then substituting with the low cost -- they, I mean, use it fully, all those opportunities are leveraged. But given the kind of current macroeconomic situation, where inflation is ruling high, I think perhaps the scope for further reduction, the cost of funds will be really limited, that is what our internal feeling is. While our [indiscernible] kind of [indiscernible] liquidity [ business that's ] prevalent in the system may continue to support, but if inflation continues to be high, then I think there may be a [indiscernible] way where policy rates may get revised. I mean this is our opinion. Then correspondingly, the borrowing rates will also increase. But as long as the rates are, I mean, liquidity is good and as long as the current economic align with -- current rate environment prevails, we will continue to look for opportunities. And I think [indiscernible] during the Investor Day, we have given the borrowing percentage versus the overall, long term versus the short term. Just to give you some comfort that we have increased the percentage of long-term borrowing from 28%...

34%.

34%. But while the cost of funds overall has come down, our long-term borrowing has actually, as a percentage, they increased from 28% to 34%. So we are constantly looking for opportunities, and that's the interval to keep the cost of funds to low -- as low as possible.

The next question is from the line of Amit Nanavati from Nomura.

So just a couple of questions. Firstly, just want to understand the linkage between your spend recovery and the revolver book, right? So even if I look at last year, in the second half, we had roughly around 71% cap increase in the overall book. But if I look at the revolver book per se, that is down. And even if I kind of add back the RBI RE book and add back the write-off there, it was still down like 3-odd percent in -- versus 2Q '21 or 4Q versus 2Q. So just want to understand, once you come out of wave 2 impact on spend recover, typically, just more theoretically, firstly, how long does it take to kind of come back in terms of optimum revolver mix? And secondly, more strategically, also we would have chosen to keep the revolver book lower by optimizing your customer mix. So in that sense, more like a practical level, how long does it take, the revolver book to kind of improve and the margins to kind of recover?

Yes. I will just add a couple of lines and Girish will supplement. If you look at our composition, our transactor balances, revolver and EMI kind of balances, revolvers were pretty high at 45% as of June 2020. That was a result more of a moratorium when the -- I mean, the hardship tools were not available at that time, only in others, only the restructuring tools were available. The moment the tools were available from other standards, part of those balances got converted into EMI. But then we have also given an RBI RE work where we would see a steady rundown in the portfolio. Some of it, of course, substantial portfolio got written off. But RBI RE portfolio, we were able to see some robust payments till March 2021. So that brought down slowly the percentage of EMI -- I mean the RBI, I'm talking about RBI RE percentage. It has -- it was, I'll say, at 9% at one point of time, maybe in December '20, came down to 8% as of March '21, and now it has come down to 6%. Despite RBI RE 2 portfolio, new portfolio of INR 258 crores. So as this portfolio weighs down, [indiscernible], as this portfolio pays off, to that extent, you can see improvement in the revolver. In fact, despite converting INR 258 crores of current balances into RBI RE, which is almost like a 1%, our revolvers improved in the quarter by 1%, 28% to 29%. So we always used to say, like a BAU kind of scenario will be 1/3, 1/3, 1/3. Transactors reached that level, now it is 33%. But as RBI RE portfolio comes down, we can see some kind of reversion to the mean in terms of revolver balances.

So just to add a couple of things there. From a quarter-on-quarter perspective, if you look at the movement, the asset, primarily on the revolver portfolio, is almost stable. There is an increase in the term asset balances, which is the EMI portfolio on the balance. There is also -- and this is despite the movement as our MD, sir, mentioned, that you have movement of earlier revolvers going into RBI RE and the pay down. So despite those movements, the balance revolver is almost stable. There is a decrease in the transactor because of the impact of COVID, which was not there in the end of March, but you could see that part of that until the month of June because June was also not fully COVID free. So that impact is also there. There is a decrease there. So that is why some mix things, you said looking at this. The other question which you asked, what is the kind of time frame that you should look -- we should look at from a conversion perspective? There are 2 kinds of conversion there. One is when we look at new spends, we try to convert a good quality of that spend into -- immediately into EMI kind of a portfolio, which is either through subvention or Flexipay products that we have. We see a good conversion rate there. So that is one way of conversion. The second is the revolving conversion, which happens on the asset, which builds up over a period of time. That takes normally anywhere between 12 to 18 months to get built up.

The next question is from the line of Subramanian Iyer from Morgan Stanley.

I had a question between [indiscernible] jump this quarter to 1Q [indiscernible].

Subramanian, your voice is not audible. It's breaking.

Is this better now?

Yes, slightly.

Okay. So my question was with respect to the loan yield. So there's been quite a good improvement in this quarter. And obviously, a lot of it has got to do with the favorable shift this quarter. So just wanted your thoughts on, basically, how do you think about the mix going forward? And also, the shift in mix this quarter towards the term loans was a bit unexpected because, obviously, in a quarter like this, we thought that term loans, which is probably [indiscernible] situation might [indiscernible] more towards [indiscernible]. So just wanted your thoughts on that. And also, were there any other factors which also resulted in the improvement in the loan yields this quarter?

So we -- going forward, as was mentioned, we will look at a 1/3, 1/3, 1/3 mix once RBI RE gets completely paid off from the portfolio. This was our -- broadly, this was a mix that -- with which we used to operate before the COVID period. And we had been operating with that mix for a fairly long time. So it was a consistent mix in our portfolio, but -- that we looked at that. We are seeing -- even now the movement is going towards that direction. So there are -- apart from, I think, some bit of revolving mix to be changed, but if you look at revolving RBI RE, which was -- actually, a part of revolver earlier that the mix would be approximately in the same 1/3, 1/3 range. So we look at that. On the term loan perspective that you talked about, there is a consumer demand towards -- whenever consumer -- large ticket site items or mobile phones or some of these items are being purchased online, there is a very strong demand to convert into EMI kind of a portfolio. And subvention is working very nicely there. So we see customers converting their balances into installment, both at the point-of-sale as well as later on. So that demand is very, very strong. Also, we have -- in this quarter, we have very strongly looked at some of our good customers and looked at giving additional loan to those customers so that asset can be built there also, which is a product which we sell, which we call as Encash, we build asset there also.

Okay. Just a follow-up question on the deduction in the RBI RE book. You can split that reduction of INR 789 crores into basically write-offs and actual repayments.

Yes. This -- out of INR 789 crores, I think bulk of it is RBI RE portfolio NPA, which was more than outstanding for more than 90 days, I think, around INR 600 crores, 640....

47.

INR 647 crores was outstanding as of March 2021. Normal situation, we would have recovered, at least resolved, at least a part of it. But given the kind of situation, second wave, I think most of it has actually were charged off. So that is the -- mostly, it is the RBI RE NPA portfolio. And that is also one of the reasons why, as a proactive measure, we increased the ECL provisions for more than 90 days from 80% to 100% based on the recent experience.

The next question is from the line of Karthik Chellappa from Buena Vista Fund Management.

We are not able to hear anything, Rasuja. Rasuja, we can't hear you. Can you hear us?

Sir, yes, I can hear you. Sir, please give me a moment. I'm facing some technical glitch. [Technical Difficulty] [Operator Instructions] We'll take our next question from the line of Bhavik Dave from Nippon India Mutual Fund.

I had 2 questions. One is on your incremental customer additions that you're doing. This quarter, we've had like significant additions via the nonbank or the open market channel. And if you look at the various breakups, we've either added more salaried, more category-type customers with a little younger age of like 30 and above -- sorry, 30 and below. And also, if you look at the geography mix, it's more Tier 1. So is there anything specific -- any new channel that we found, which is leading to getting back to maybe a different type of customer segment during the quarter. If you could just talk about that.

Yes. Mr. Bhavik, last year -- enter last year, our banker channel contributed more than 50% to our incremental sourcing. But this quarter, it so happened like a lot of restrictions were there in so many states, including states putting restrictions on the operating hours of branches, which came in the way of sourcing to this channel because even the customers were hesitant and even the branch team was also a function only for part of a day in some of the states, particularly in May and the end of April. That came in the way of sourcing by a Banca channel. So that was the reason -- open market, of course, never -- I mean, did not have the challenges to the same degree. So they continued with the acquisition, and that was the reason like open market accounts for maybe more than 65% and 35% roughly from Banca. And Banca earlier, because like we always work with the Shikhar model, and Tier 3, Tier 4, we were targeting through the Shikhar model, where we had the benefit of that [indiscernible]. So this is why Tier 3, Tier 4 had a significant contribution in the last year. But because the Banca sourcing itself has come down, now it will appear like open market is always active in Tier 1 and Tier 2 only. So that contribution will [indiscernible] got increased. So it is not like a specific strategy. It is more around our inability to operate Banca at the same level at which it was operating in the last year.

And sir, if you could just talk about -- we've seen that last 7, 8 quarters, you had a very good sourcing on the Banca channel. If you could just give -- qualitatively help us understand how is the customer behavior there being? How is the spend behavior being because those customers are also now getting online on the various apps or various e-commerce website. How has the graduation of these customers been over the last 7, 8 quarters? If you could just help us understand the balances or the spends have increased from these customers that we acquired via the Banca channel over the last 7, 8 quarters?

So the way that we have been acquiring Banca customer, as was mentioned, is through the Shikhar program. We run this Shikhar program in various batches. So we do it in program [Audio Gap] I think at this point in time, number 17 or 18. We take different kinds and sets of customers in each Shikhar program basis either a savings account or, let's say, a well set of wealth customers. So we look at those each Shikhar program like in that separate sense. What we have, however, seen across the portfolio is that people who become active and start using the card, their spend per card is broadly similar -- slightly lower than the normal open market customer in that city. So there is always a tiering level because Tier 1 spends are higher. Customers in Tier 1 spends are higher, Tier 3 are slightly lower. The low -- and even when we look at the breakup, the online spends are broadly similar. It is this issue at the point of sale because the number of point-of-sale terminals are not that much available in Tier 3. And the need is also usually not that much because you know the local grocer or the shop owner and those kind of things. So you -- but online spend is broadly similar. Where we were looking -- seeing some challenge as well in terms of getting the customer to start using the card and activating the card. So we've been doing a lot of programs. What we see is that it does take time for some of these customers to start using the card, but we have now figured out that on -- as the customer comes on board, doing an onboarding session, talking to the customer, explaining the product, does help. So we have put in those set of programs. The second thing, which I want to highlight is that the asset build up in some of these customers in Tier 3 is slower compared to Tier 1, Tier 2 or already carded customer. So the balance for a mature customer does almost become similar, but the maturity time frame is slightly longer for these customers.

Sir, second question is on the EMI part. When we are seeing more and more customers, both at the post terminal and also the balance in getting [indiscernible]. If you could just talk about what are the [indiscernible] rough [indiscernible] that we charge like maybe at the post terminal or when we convert the revolver or like the balances into EMIs. And how long does this book stick around? Like is the average duration 12 months or lower? Anything to read into that?

So we do not convert revolving balance into installment lending because that is not the right way. So key customers, if they are doing their fresh [Technical Difficulty] so if the customer is spending -- doing fresh spends on the card, then they'll be charges, 14%. And we also get the interchange on that transaction. So these are the 2 income streams that we make on those transactions. On the tenure part of it, the average tenure of people converting into installment lending at the point of sale is close to 8 months. So it's a very -- it's a short tenure thing. People prefer 6 months tenure or 12 months tenure primarily for conversion of new spends into installment balances.

We will take our next question from the line of Jaimin Shah from RWC Partners.

Quite commendable performance given the circumstances. I had 2 questions. The first question is on spend. Could you talk a bit on spends on how it's evolved from April, May and June and what you were seeing in July? And could you talk about the frequency as in -- in one of the slides you kind of show that spends on more of a daily necessity kind of stuff has gone up quarter-on-quarter. What has kind of suffered is essentially the discretionary and more importantly, the services, which was quite close. So I'm just trying to understand how you see the usage, also the data we see from RBIs on per transaction basis, the average ticket size has gone up significantly quarter-on-quarter or even in the data we have until May. Any color that would be great there.

So when you read RBI data, you should look at it that, that spend is both retail spend as well as corporate spend or B2B spend. It will include -- and the breakup is not given in the RBI data. So that's why you will not be able to get a mix on that. Typically, what happens is for the retail spends, the ticket sizes are lower. For B2B spends or SME spends, the ticket sizes are higher. So that's the insight while you are evaluating the RBI data. While on the first part of the question that you asked, you are right. Couple of trends that we see here is that discretionary spends have gone down. They continue to be down at this point of time because travel, lodging, hotels, restaurants are still not there. It is nondiscretionary items are actually -- are growing. What is also what we see is that online spend in Q1, in April, May, June, actually, on an absolute basis, were also higher than what we saw in Q4. So there is a trend line for -- as you can see on the slide, which we have presented, 55% of the spend is almost online. And so trend is basically that the customer is now buying nondiscretionary items online. And that is a kind -- and that is a consistent piece that we are able to see.

Right. Okay. No, that's helpful. That's helpful. And just on the asset quality side, your GNPA or whatever, your stage 3 includes all the RBI RE more than 90 days, right?

Yes, yes.

It's inclusive, right?

Yes, yes.

Okay. And I'm just trying to understand here, the INR 258 crore, you kind of gave that number, is similar on management overlay and the RBI RE 2.0. So essentially, is it fair to say that you're kind of covering 100% of RBI RE 2?

No. Jaimin, that's just a coincidence that the numbers end up being the same. That is not intended to cover the fresh RBI RE. Like we've mentioned earlier, the INR 258 crores that you see as management overlay is just the fact that we cover for any delinquent RBI RE at 65.2%. And any NPA RBI RE, we are, right now, covering it 100%. It's the sum total, of course, the fact that it is exactly equivalent to that is just a coincidence.

We'll take our next question from the line of Bharat Shah from ASK Investment Managers.

2 questions. First, you said that per se desired mix that you aim for is about 1/3 each transactor, revolver and EMI, but why would you desire 1/3 to the revolver of necessity? I mean does it have any impact on the way you assess the credit and the kind of -- you take a bit more liberal stance and issue cards accordingly? If the idea is to run the interest income, will it not structurally be resulting to an issue at some stage?

So when we said 1/3, 1/3, 1/3, it's primarily what we have seen from the consumer behavior over a period of time. And we are talking about this after demonetization, where we have seen that there are people, customers who are taught, and we are not talking about habitual revolvers or chronic revolvers or consistent revolvers. People, these days, if you have to -- let's say, if you purchase a thing, as I was mentioning earlier, there are -- rates are available as low as 14% to 18%, where the customer can convert into installment lending product. So customers who still revolve are either looking at it for a very short-term working capital requirements. So what they're looking at is that if they're able -- they will be able to pay the outstanding within 2 months, so that's the kind of thought process with which they revolve. Yes, you are right. Every customer who goes into finally a loss ultimately revolves before that. So that is a slow path. But from a consumer behavior, what we have seen for the last 4, 5 years consistently is that there are -- if the customer wants to pay back, and he has a planned thing, then there are installment lending products. That's why they have become 1/3. 6 years, 7 years back, this number used to be almost close to 0. So this is a trend in last 4, 5 years as it has happened. So revolving is where the customer need is getting fulfilled for a very short-term period. There are some sloppy payers also, people who forget to pay their outstanding at the due date. So they will fall into -- go into a revolving model and then come back and then pay back. So there are those customers also.

What I heard you saying that you prefer it this way. It's not an outcome of the consumer behavior, but probably you preferred it to be 1/3 each, which is what I was a bit curious about. Why would you structurally expect 1/3 people to revolve because that will then have an impact on the way you assess or judge the credit risk before issuing the card and all of that, I would have thought.

So let me clarify that further. We don't expect 1/3 of the customers to revolve. This 1/3 is of the asset. And that's -- the number of customers revolving is actually far lesser, much, much lesser. So this is now the trend line that we have seen over a period of time. And what it also does is, if the mix of assets is broadly in this 1/3, 1/3 range, [ the leads ], which you finally get are in the space where our business model is rightly positioned.

Okay. And secondly, when I structurally look at your business, where it is designed for a very superior ROE, which typically, most lenders can never hope to get because the substantial part of the income comes from the card spends. And typically, what we have seen is your noninterest income covers all your expenditure rather than the provisions. So therefore, structurally, that's a very superior business. You said, fundamentally takes care of [indiscernible] expenses at a practically very, very high ROE activity. And interest earning income portion also is a high interest component, ranging from EMI to your defaulting -- revolving customer all the way to [indiscernible]. Therefore, if you -- if your loan -- your credit losses are [ creating ] shape, even that part of the lending activity is superior ROE activity. If the 2 things we put together, can we kind of say that structurally, you should make more than 30% return on equity on a perpetuity, if not a higher number than that?

So thanks. Yes, this is the business model. You're right. I wouldn't like to comment on the perpetual EBIT at this point of time, definitely. Yes, right now, we are impacted just like everybody else with the COVID, and we have a stress book. So that is there. But yes, you're right. In terms of business model, this business can generate higher returns in most of the business because the income earning capacity per dollar of asset is definitely better than any other lending business or payment business that you can think of.

We'll take our next question that's from the line of Mahesh M.B. from Kotak Securities.

Sir, just 2 questions from my side. Sir, the question, if you start seeing more number of transactions on the online, and specifically, it starts getting concentrated on our 2 websites, how -- have you started seeing any change from a revenue perspective for you?

No, not as of now. But this is -- we believe that this is -- this concentration is there online between, let's say, the 2 large online players. But as the point of sale gets opened up, we believe that point of sale will get -- will increase. The second thing is, you should also note that, as mentioned, that there is a large set of categories are still not open. So the whole travel, the whole restaurants, the whole -- where the card is a -- used to be a natural medium of payment, those categories are not open. So they will also pick up. So it's quite very diversified from that perspective.

Can you just kind of [indiscernible] expanding the question. When there is sales which is happening on specific days around these platforms, do you have to kind of propose a specific incentive for them to be part of that? Or how does it work?

So the way that it typically works is that when there is -- if there is a cash back offer because that is typically the offer, which comes in with most of these large players, there is -- and that cash back is the customer value proposition with the customer directly gets that benefit. That benefit -- part of that benefit is shared between the merchant, the manufacturer and us. So there is a sharing arrangement, which happened. And between -- so that's from a fiduciary perspective. Otherwise, the advertising and other things are also looked at depending on who is able to put in what into the overall basket. So it's a commercial arrangement between 3 parties.

Okay. Yes, please, my question is a little bit different. If there are 3 very large credit card issuers, does the website provide a kind of or decide as to who does it partner for this program? Or all of them get an equal opportunity on how do you want to take this forward for that specific few days of offset?

See, end of the day, the program is being run by the website. So they can always individually do it or decide who amongst themselves. But we are a large player in this space. So almost all the websites, they connect with us, and we are in regular contact with them for various programs.

Sir, second question to Aparna. When you look at the spends today, Aparna, if you could just kind of qualitatively comment, have you started seeing spends coming across all the set of clients, which is a [indiscernible] of prime, subprime? How are you seeing these trends coming or shaping up in your targets?

So the spend, I don't think we should be looking at it as prime, subprime or any -- so one of the things we need to note is, like we mentioned earlier, there's a fair bit of dynamic portfolio management that we are doing. So if anybody is on the margin of -- or they have a very high probability of becoming delinquent, then the requisite amount of credit actions have already been taken on these lines. The lines have been brought down. And if they're trying to do any specific high-risk transactions, those will get declined. So that kind of activity is happening. Now -- so lines are available to spend only from the better customers. So we've not really seen any specific punching of spends coming only from the lower core segments or anything like that. We haven't seen any such grouping of this spend.

Can you actually reverse the argument and say the spend that you're seeing today is almost similar to what you saw in 2018, '19? Or you think you've not reached that point?

No. No, no. We -- it is thought because, as I told you, there are certain categories, which are completely shut at this point of time. They used to be mid-teens in the overall mix. So that is not there. The second thing also which you would note, which we have given in the presentation, is the categories like, for example, the whole nondiscretionary set, which is your utility base, telecom, okay? So they are -- these categories are typically -- they are used by everybody. So -- and the good part is that a large part of that was [ a clear ] going through cash or, for example, insurance payments through the insurance agents, where people used to give checks and make the payments now. All that is coming on online and is happening by the -- is being done by the customer regularly. So it's a very good sign actually from an overall digitization perspective.

We'll take our next question from the line of Suresh Ganapathy from Macquarie.

Just 2 questions. One is on the credit cost. Is there a way to really look at it? The last 2 quarters, it's been hovering around 10% to 11% of [indiscernible]. Do you think really the second half could see -- of course, notwithstanding a third wave, but can we see a substantial reduction in the credit cost numbers for the later part of the year? I'm not asking for a specific number, but can we really see this coming down?

I think last month in one of interactions, we said like we expect it going in downward trajectory for the credit card that was much before the second wave. So I think I can only say that a second wave has actually delayed a downward journey. But the pace at which it will decline is [indiscernible] macroeconomic environment, and we're gaining the complete collection efficiency like BAU kind of efficiency, we need to get back. Consumers also should be accessible, customers should also be accessible. But definitely, the trajectory will be downward. I cannot comment on the flow part. I mean how fast we will be able to reduce it, that we need to wait and watch. But over a period of time, as compared to May, the position in June was better. And again, as compared to June, July positioning is better. [indiscernible].

Okay. Okay. So that's clear. The other part is, of course, this question has been asked in a different way by various participants. So let me also try to attempt this. With the -- there is a big debate that the BNPL customers are going to be $40 billion, $50 billion in the next 5 years. Are you seeing some signs of aggressive pricing in the market, thereby, you're compelled to do? Or there is an issue of loss of market share in certain of these segments? I don't know, the industry data is actually not available, but how are you approaching this particular segment?

I think if we traditionally look at the kind of -- if it's our own company's internal experiences, I mean, typically, the limits will be minimum 30,000 to 40,000 kind of limit permit. Unit economics perspective, otherwise, it will not work out for us. But BNPL is in a totally different bracket, where it starts with a very small amount of even 5,000 and may have hover around 10,000, 12,000. But with regard to the potential, if you don't rule out the potential, definitely, the market will definitely increase. And mostly, it's very popular with the millennial and otherwise, the [ NPC ] kind of segment. But way forward, it will be -- perhaps, some convergence will be there where the BNPL customers can potentially become customers of credit card once they build a good credit history. Or otherwise, we will also be making journey towards it. So there will be a convergence point. But supposedly, it's like we have to use a right channel of liquidation, right value proposition has to be offered. So there will be some [ convergence ], definitely, for a period of time. There will be -- we may help to have different product offerings to cater to that. We will also help to articulate some more risk appetite going forward. But we are waiting and watching on when -- we will be calibrating our strategy based on the situation.

[Operator Instructions] Our next question is from Aakriti Kakkar from Goldman Sachs.

I have 2 questions. Starting with a quick one first. Do we utilize SBI YONO as a sourcing instrument? And if we do, could you please give us some idea around the proportion that is sourced from there?

Yes. The SBI YONO is definitely used both the app as well as there is an internal portal within the bank. So both are used because there's more convenient way of getting information in the specific customer who has given a mandate to the bank to share the information. So that is a preferred channel. That's definitely a preferred channel. And as of now, I think 20% of the sourcing happens through YONO. And...

20%?

Yes, 20% of sourcing happens through YONO. And it has potential for further increase.

All right. So sir, I'm looking at Slide 4 of the PPT. And in the total receivables, 32% is EMI. How much of it is converted at cost for the industry and for SBI Cards?

We have not declared that asset balance, out of the 32% how much is at the point of sale? But I can give you one indication that, close to almost 10% of the spend, new spends get converted into installment lending products.

Okay. All right. Okay. One last quick question. Sir, so 29% is revolver here in the total receives book, what is the duration of this revolver book?

It's about 3 to 4 months, okay? That's generally what we've seen in terms of how long the revolver [ sees ] the revolver. However, one point is that it's important to note that these are revolvers. So anybody paying between 5% and 100% -- 99%. So we do have customers who revolve, attain 40%, 50% as well.

We'll take our next question from the line of Ajit Kumar of AMBIT Capital.

So again on the sourcing, is this jump in the open market sourcing to 62% from normal 45% to 50% driven by co-branded partnership or normal open market sourcing? So could you please provide the breakup between co-branded and retail within the total markets that you used to provide this data earlier?

Okay. We -- see, we do not give the breakup among the channels in open market. But what I can tell you is that when we look at our total open market sourcing, our co-brand partnerships are a significant engine of growth for customer acquisition. So they do contribute a larger chunk. The other element is that the mix at any point of time also depends upon how the environment is. For example, in that quarter, when the malls get shut or the organized retail stores shut, then, for example, the fuel co-brand could sell more. But in the meantime, when the organized retail stores open back again, then the situation normalizes. So then mix changes in terms of how it is among the various constituents. So it's a large component, but we don't give the exact breakup among the different channels in open market.

Okay. So is there any difference in the sourcing cost per card between this co-brand and then retail brand within this open market?

So you're asking co-brand and retail. Co-brand is [indiscernible] are retail or open market card itself. So in terms of co-brand -- open market versus Banca, there's a difference in terms of the cost of acquisition per card. Cost of acquisition in open market tends to be on the higher side than Banca because in Banca, we get a lot more information from the bank, and we can run preapproved programs. Within open market, there are different channels in terms of tele sales, e-apply, co-brand and the normal point of sale. And those costs are staying in a certain -- there are -- these are 4 different cost elements within open market, within a small range of each other, but we don't specify that value. What I can tell you is that open market cost of acquisition is higher than Banca.

Our next question is from the line of Saurabh from JPMorgan.

Sir, just 2 questions. One is this stage 2 book -- one is what is your normalized level of stage 2 book? Shall we take that Q4 number as a normal number? And is this restructured asset over and above the stage 2 book, correct, the restructured non-NPA asset?

No, that's part of the stage 2.

Okay. Okay, okay. And what would be the normalized stage 2? Like, I'm just trying to compare what will be -- the interest 12-odd percent, what's the baseline number, let's say, you had experienced in fiscal '18 versus '17?

So pre-COVID, it was in the range of about 8-odd percent.

8%, okay. Okay. And the second is, so generally, on the interchange fees, so what you're seeing is an MDR pressure in the market. Have you experienced any interchange fee? Or do you think that, that thing could be under some pressure because the MDRs are coming down? Or you don't think that that's going to happen to the card companies?

So our interchange are broadly fixed and they have been consistent. There is -- if there is a change, the minor change happens on a month-to-month basis depending on the scenario. Because typically, for example, utilities will maybe carry a slightly lesser interchange compared to if somebody buys mobile from an online store. But if you buy a mobile from an online store or mobile from a POS, we get the same interest rates. There is no difference there. So the mix and depending on the monthly scenario and overall, same. It does make some bit of change, but we have seen it consistent for last 1.5, 2 years, 3 years.

Okay. So you've not seen any interchange issue? Okay.

MDR can be negotiated between the acquirer and the merchant. So that is a very different thing.

Our next question is from the line of Dhaval Gada from DSP.

2 questions. First is, what percentage of the EMI book gets converted at the time of purchase at POS? And the second one was, would we see a number outstanding in the [indiscernible] on that?

So on the first part, I just answered that sometime back. We have not given the breakup of that asset into what it is. But what I also, along with it, we have stated that of the new spend, which happens every month, we see close to 10% of them getting converted into installment lending products. On the EPP, Aparna?

Actually, I don't have that. How much is that?

300.

About INR 300-odd crores is EPP.

Our next question is from the line of Nitin Aggarwal from Motilal Oswal Securities.

A couple of questions. Like, firstly, if you look at the average ticket size of UPI has been increasing from 1,200 in FY '18 to now 1,900. So what is really driving the specific size increase for UPI payments? And how much of this, in your view, is driven by market expansion or substitution?

So UPI ticket size, we don't have that detail on that data as to where the transactions are happening because we are not on the UPI platform. However, the P2M that we hear -- that is getting declared, there is close to around 17%, 18%. So majority of it is P2P, still there. So 1,700, 1,800 is a good decent ticket size for small ticket transactions, which [indiscernible] people giving money to each other from that perspective. But that visibility, we don't have as to why the ticket price is moving up or down. But it's been quite stable at 1,800, 1,900 for some time now.

Okay. And secondly, on the presentation, Slide 10, like wherein we have given the spend categories, there is like a decline across all the categories, a double digit decline, category 2, 3 and 4, and then between 10% to 48%. And -- but overall retail spends are down 9%. So what is balancing this? So is category 1 like very big in size in terms of proportion of expense?

Yes, you are right. Yes.

So how much could this be like approximately in terms of overall spends?

So we've not given that breakup, but as you have rightfully figured out that overall spend, on a quarter-on-quarter basis are not down by that much. We are [indiscernible] INR 49,000 crores, we're around INR 47,000-something crores, okay? So it's in that range. And online actually has grown in the -- which we also said. We're not giving the exact breakup of these because these mixes -- departmental stores usually is a very large category. And this is after we have taken out fuel. If you look at the disclosure there, fuel and automotive, we have taken out because they are dependent more on what is the price of fuel at that point of time, and that is a growing category.

Okay. And lastly, just on the write-off number, if you can like mention it for the -- within the GNPAs, how much we have written off?

Sorry. The GNPA -- within the GNPA? So GNPA is a nonwritten of [indiscernible]...

The movement of -- yes. Movement of NPAs, basically.

So our NPA was 4.99% on a gross basis as of March, that has come down to 3.91%.

No, but the composition of it in terms of slippages, reductions and reductions by the [indiscernible], I guess.

Yes. I think unlike a bank, we don't do that type of disclosure, where we use the stages, then upgrade the -- I mean increase in the NPA balances. Typically, that is the way the banks do, but we have never given that kind of disclosure. But we give a kind of picture around what are the gross write-offs and what is the recovery, of course, the opening balance and closing balance will be there.

We'll take our next question from the line of Jai Mundhra from B&K Securities.

I have 2 questions. One is, while credit card is technically kind of a BNPL product only, but do you have a separate BNPL product? Or is there any restriction in your charter or maybe from a period that you can -- I mean, do you have a BNPL product as of now apart from this main offering?

No, we don't. As you rightly mentioned, credit card actually is the true original BNPL product because in this case, you can buy and then pay either in revolving terms on your own or paying 3 months installment or 6 months, 9, 12. The choice is with the customer, but we don't have a separate BNPL product as of now.

Right. And is there any restriction in your charter or maybe your understanding with parent?

What we can issue is a credit card. And so any BNPL type of product that -- if at all, will be in the lines of credit cards because that's what we can do.

Sure, sir. And the last question, sir, management overlay. Is this your over and above IRAC sort of provisions? Or this is just purely those provisions, which are -- which have not been netted off in computing PCR? Or -- yes. So is this like over and above, what is dictated by IRAC? Or this is purely those provisions, which have been set aside without -- which does not go into PCR?

So we provide as to be [ Ind AS ] [indiscernible] not IRAC. So the model shows up a number. Whatever we provide, in addition to the model, is what is the management result.

Which is already there in PCR calculation, right?

Yes, yes.

We'll take our next question from the line of Ravi Singh from Motilal Oswal Asset Management.

Yes. Sir, just one question on the retail spend slide. So the spend categories other than category 1, what are the trends you are witnessing in the second half of June and July so far after the unlocking started? And based upon this trend, what sort of normalization are you expecting? Do you think it will be a quick recovery in the spend level or it will be a longer recovery?

So on the spend side, what we have seen is that till the first week of June, it was quite soft. It started picking up actually as the rolling lockdown started to finish. So it is actually the second half of June, which was very strong, where this started to pick up and become better. We are seeing that July is coming out better than June, and it is stronger than June. So it is only after 15 June that things have started to pick up.

Right. But in the second half and July profile, trends are strong enough to suggest a recovery in the spends in category 2, 3, 4?

So see, as of now, it is quite early, but what we have seen in July is quite strong.

Ladies and gentlemen, that was the last question. I would now like to hand the conference over to Mr. Rao, MD and CEO of SBI Cards for closing comments. Over to you, sir.

Yes. Let me close the call by reiterating that our business fundamentals are very robust, and we follow all healthy financial and corporate governance principles, which form our core strength. We continue to closely monitor the situation, take all measures to minimize the risk and ensure a sustainable growth. The first quarter results I shared with you reflect our stability and growth potential. So lastly, the need of [indiscernible] is for all of us to remain safe and extremely cautious while pursuing our personal and professional endeavors. So thank you, and stay safe.

Thank you, members of the management. Ladies and gentlemen, on behalf of SBI Cards and Payment Services Limited, that concludes this conference. Thank you for joining us, and you may now disconnect your lines.