Ladies and gentlemen, good day and welcome to the SBI Cards and Payment Services Limited Q3 FY25 earnings conference call. As a reminder, all participant lines will be in the listen-only mode, and there will be an opportunity for you to ask questions after the presentation concludes. Should you need assistance during the call, please signal an operator by pressing star, then zero on your touch-tone phone. And now I'll hand the conference over to Mr. Abhijit Chakravorty, MD and CEO of SBI Cards. Thank you, and over to you, sir.
Thank you, Rao [guess]. Good evening, everyone. I'm pleased to welcome you to the earnings call along with my senior management team. Projections indicate that digital economy will account for 20% of GDP by 2026 and exceed $1 trillion by 2028. UPI transactions, for example, saw an 8% month-on-month increase in volume to 16.73 billion in December 2024, the highest volume for the digital system since it became operational in April 2016. As per the RBI December 2024 data, credit card spends have grown to INR 1.88 lakh crore in value terms, with transaction values exceeding INR 1.14 trillion on e-commerce platforms. This trend further highlights the increasing preference on credit for online purchases. At SBI Cards, we understand and remain committed to focusing on Indian credit card industry's growth potential and increasing digital payment adoption.
As India's largest pure-play credit card player, with an 18.7% market share of cards in force, we are well-positioned to drive the sector's strong growth. A significant milestone for us this quarter has been surpassing the INR 2 crore cards in force mark in December 2024, reflecting SBI Cards' expanding customer base and market presence. Let us now look at SBI Cards' business overview in Q3 FY25. We have sustained growth trajectory with strong business performance while continuing to focus on customer experience and operational efficiency. As part of key initiatives, we rolled out many customer-centric initiatives and campaigns, for instance, the Khushiyon Ka Utsav promotional campaign during the festive period of October-November 2024. 100-plus festive offers were introduced across varied categories on an array of brands and segments. Collaboration with e-commerce platforms Amazon and Flipkart were launched to drive high-spend customer engagement and penetration.
During the quarter, we have launched our hyper-personalization platform and have commenced running campaigns. The platform now gives us the capability to run customized and targeted campaigns. The key objective of the hyper-personalization platform is to enhance customer lifetime value through personalized customer engagement. This will be done through scaling up and augmenting existing capabilities to more than 10 times and achieve higher levels of personalization and one-to-one communication through the SBI Card Mobile App. We continue to grow end-to-end digital acquisition capabilities. For instance, owing to the integration of the SBI Cards print platform with SBI YONO and SBI Internet Banking, the majority of banker new account acquisition is now being achieved digitally. We have leveraged AI to automate dedupe stage in application processing journey, bringing in efficiency, cost benefits, and error proofing on application decisioning. We continue to focus on varied ESG initiatives.
Our commitment to sustainability continues to deliver results, as evidenced by our consistent A rating from MSCI, registered with SEBI to provide ESG rating companies this year. Our ESG rating of 16.5 from Sustainalytics places us in the low-risk category. These ratings reflect our dedication to integrating ESG principles into our business operations. Coming to financial performance, the results for the company for the quarter are in line with the seasonality of the business.
The key financial highlights are: On new accounts, I am pleased to share that during this quarter, we have achieved a significant milestone. Our cards in force have crossed the 2 crore mark, witnessing a healthy 10% year-on-year growth. During Q3 FY25, our new account acquisition was at 11.75 lakh, with a 7% year-on-year growth. This is significant considering our continued focus on selective acquisition during the past few quarters and to grow in a calibrated manner.
SBI Cards' CIF market share in Q3 FY25 is 18.7%. Share of new accounts sourcing from banker and open market channels stands at 55% and 45% respectively. Regarding spends, retail spends remain strong at INR 80.792 crore, with a 10% year-on-year increase. Corporate spends are stable at INR 5,301 crore. SBI Cards' spends market share is at 15.6%. Retail spends per average card have remained steady at INR 1.62 lakh, similar to Q3 FY24, despite the festival season being spread over September and Q3 this year. Categories comprising consumer durables, furnishings, and hardware, apparel, and jewelry witnessed 36% year-on-year growth for the nine-month period this fiscal. Online spends are robust, with a share of around 58.5% of total retail spends for the nine-month period. RuPay spends at UPI terminals continue to grow and have shown a growth of more than 45% over the previous quarter.
Department stores and grocery utilities, fuels, apparel, and restaurants continue to be among the top five categories for UPI spends. Tier 2+ customers are utilizing this facility more, as it increases the number of acceptance outlets for RuPay Card. Total revenue has grown to INR 4,767 crore, almost similar to last year. It is higher by 5% quarter-on-quarter, contributed by both interest and the fee income. Profit after tax for the quarter is at INR 383 crore, 30% lower year-on-year. Receivables as of 31st December 2024 are at INR 54,773 crore, registering a 12% year-on-year growth. Receivables per card is INR 27,052 versus INR 26,438 in Q3 FY24. Interest-earning asset and revolver rate is stable at 60% and 24% respectively. With newer and quality vintages reflecting a lower revolver rate, we expect the receivables to grow at a lower rate. Cost of funds has remained stable at 7.4%.
We expect the cost of funds to remain at around current levels going forward till we see any rate cut action. Our cost-to-income for Q3 FY25 is at 53.5%, almost similar quarter-on-quarter, owing to festive offers during the quarter. Regarding the asset quality for the company, the refreshed data from the credit bureaus suggests that the credit card industry defaults have remained elevated due to the continued stress in unsecured portfolios. We have been taking several steps to strengthen our new acquisition, underwriting, and portfolio management framework. As a result, we have been seeing a continued reduction in our flows into delinquency and an improvement in the delinquencies of the new acquisitions. The asset quality composition is also improving and is moving towards higher bureau scores, as indicated by the increased proportion of prime and above-prime segments in the newer acquisitions, as well as overall portfolio.
This is evidenced in the following. Our GNPA for the quarter has remained stable at 3.24% as compared to 3.27% in the previous quarter. We expect the GNPA ratio to start improving now. The gross credit cost has increased by 40 basis points to 9.4% from 9% in the previous quarter. We believe we are at an inflection point in our credit cycle. As we further intensify efforts on our tightening underwriting standards, portfolio management, and collections, we expect the credit costs to start moderating. However, the rate of moderation will depend on the changes in the unsecured lending ecosystem and macroeconomic factors. Our liquidity position continues to be strong. Capital adequacy ratio for the period ended 31st December 2024 is at a healthy 22.9% as compared to around 18.4% in the same period last fiscal. This will support our growth aspirations.
Liquidity coverage ratio for Q3 FY25 is 114%, as against statutory requirement of 100%. ROA for quarter is 2.4%, and for the nine-month period, it is 3%, so the credit card industry continues to offer growth opportunities. As per reports, the number of credit cards in use in India is projected to continuously increase, as per projections up to 2029. As India's largest pure-play credit card player, we realize the potential and are well geared to take advantage of these opportunities. Thank you, and now we are open to questions.
Thank you very much. We will now begin the question and answer session. Anyone who wishes to ask questions may press the star and one on the touch-tone telephone. If you wish to remove yourself from the question queue, you may press the star and two. Participants are requested to use handsets while asking questions. Ladies and gentlemen, we will wait for a moment while the question queue assembles. The first question is from the line of Mahrukh Adajania from Nuvama. Please go ahead.
Good evening, sir. So, sir, I have a few questions that if you see your write-offs, they continue to increase and other provisions are declining. So, what would give you the confidence that write-offs would taper off from fourth quarter? Because that is essentially what will bring down credit costs, right? Or is it that asset quality will start improving and then credit costs will improve with the lag? I mean, how do we view this? When do we see credit costs improving and what will drive that? Write-offs, they've been continuously rising and sharply. So, will they taper off now? Is this the peak? That's my first question.
So, thank you. Thank you for the question. Then, what we have seen in the financials also, the lesser provision is indicating towards a better asset mix. Somewhere, the Stage 2 and Stage 3 composition has reduced, and we are seeing it to continue during the quarter. And as we stated just now, we are also seeing reduction in the into 90s, that is, into delinquency. All these taken together now gives us an indication that things are going to improve from here onwards. And we see that from next quarter, the improvements should be visible.
Now, to what extent that will depend? That will depend on how the larger portion of our Stage 3 will behave. If you strictly speak about write-offs, you will be knowing that we have an NPA pool. We continue to work upon them. The NPA pool has reduced. But then, how we continue to work upon them and extract most out of it will depend to what extent the write-offs will be reducing. But we do see an inflection point, whether it is gross NPA and the credit cost, both concerned.
Okay. And so, basically, from Q4, it should look better. We don't know how much better.
Exactly. Okay. And then, and let me go back to my, a couple of quarters earlier call, where we had indicated that what we are seeing over the quarters is that by year-end, we were expecting the improvements to come up. And today, what we can see is that probably we will be adhering to what we expected.
Got it, sir. And so, what kind of receivables loan growth can we build for the next year? What do you think?
So, Mahrukh, as you are aware, that for the last almost six to seven quarters, we have been very selective in our new customer acquisition, and that asset is building up. And as you can see, the receivable growth has not slowed down. So, what we are looking at at this point of time is that on a year-end basis, for this quarter, you've seen a 12%. But next year, what we are looking at is in the range of somewhere around 12%-15% as a receivable growth.
Got it. Okay. Thanks a lot. Thank you.
Thank you. Before we take the next question, we'd like to request participants to please limit their questions to two per participant. Should you have a follow-up question, we request you to rejoin the queue. The next question is from Krishnan ASV from HDFC Securities. Please go ahead.
Yeah. Hi. Maybank speaking here. As Girish [guess] just mentioned, it's been nearly two years since our origination has been tweaked, improved just to kind of get better at this quality of sourcing. Despite that, the write-offs or the overall credit costs seem to be hitting a record high almost every quarter. Also, it's fairly evident that the collection environment has been fairly tight, has been fairly difficult, both for, I mean, banks and NBFCs. I just wanted to understand what is it that SBI Cards is able to control other than sourcing? Because you do a hard job in sourcing now. But other than sourcing, is there anything else in your control? Because that's what the franchise gets value for. What is it that you're able to control in the rest of the DBD?
So, basically, if you look at our portfolio, the identified customers who are likely to go through some stress. So we've seen, and even the market reports suggest that because of the rapid build-up of the retail lending going through a boom, especially personal loans, the leverage has built up in the portfolio. So what we have done is, of course, to understand customers who have that increase in leverage or who will display a tendency to increase leverage based on their behavior, based on their transaction pattern, and take prompt action so that we can at least reduce our exposure to these customers. So we are able to do that, and we have talked about limit decrease in the past as well, and we continue to enhance our early warning framework so that we continue to identify and take action before these customers go into delinquency.
Because once they go into delinquency, as you rightly said, the environment is tough, and if there are too many so basically, we have to cut our losses before if we identify the right set of customers. We've managed to do that successfully over the last few quarters.
And the success of these actions is being felt in the performance metrics that we track, so for example, for the first time in several quarters, the 30-day delinquency and the 90-day delinquency have come off appreciably by 20 basis points, and also we see similar trends in the industry, and also the performance metrics around our flows into delinquency, as MD sir mentioned, they have been improving. They've been improving consistently over the last one year, and this takes time to show up in results, and we're seeing the beginnings of that right now. At the same time, the other half of the equation, which is the flows from delinquent pool into write-off pool, that also for the first time in several quarters has started improving.
Okay. Okay. Understood. Just one other question. This is majorly to do with what proportion of your customers have actually been defaulting? I mean, I just so one is obviously the value that you tend to report every quarter. But what proportion of these customers are actually reflecting in that credit cost of 9% last quarter, 9.2% this quarter? How many of these customers are still there in the pool who can spring a surprise on you?
We don't disclose these numbers, Krishnan, first of all. And two, in case of a product like a credit card, which is revolving where customers can come and pay you any time of the day, any time of the month, it's very difficult to track and at a point in time measure as to how many have defaulted and how many have not defaulted.
No, what I mean what I mean to say is, are these
if the average asset is 27,000, you can calculate that when that customer is defaulting on an overall basis, it's over close to a lakh or so. Okay? So you can estimate what is the number of customers.
Yeah, that's correct. Girish, I was just wondering whether these are customers who have leverage even more than the average balance. That's what I'm trying to get at. Is that the reason why they are defaulting?
Yeah. Obviously, they have leverage over them. They have high leverage, yes. But they do not go above balance.
Okay. Fine. Thank you. Thanks, ma'am. Thank you. Thank you. All the best.
Thank you. Next question is from Piran Engineer from CLSA. Please go ahead.
Yeah. Hi. Thanks for taking my question. Firstly, just a clarification on the previous thing. Did you mean that the average and.
Average? Sorry, we didn't get the question, Piran?
The ticket size is one. Is that what you meant in the answer to the previous question?
Your voice is breaking, Piran. You have to repeat that.
One second. Am I audible now? Is it better?
Yeah.
So just a clarification on what you said as an answer to the previous question. What was INR 1 lakh? Is it the average ticket size of an NPA account?
No, no, no, no, no, no. This is close to so one lakh is you have to take it. It's not an exact number. It's in that range. This is the average, what we would say, balance of a defaulting customer. When the customer gets into a write-off or becomes a defaulter, that is the kind of balance that it is. Otherwise, the average asset is around close to 27,000, which you can see from the statement.
Yeah. So correct. So the MPLCA [guess] average is one lakh. The average NPA customer has one lakh with you.
No, it is not that number. It is brought around in that range.
Okay. Fair enough. So my first question really is on you have spoken about how the new customers have lower delinquency trends, the cards that you sourced in the last one or two years. How are the revolver trends versus earlier vintages that were sourced?
Typically, the revolver trends are we are able to see after three to four quarters. We see the revolver trends as of for the, let's say, vintages more than six to nine months, which we have sourced acquired as of now, are slightly lower than what is the average of the portfolio. We would continue to view them over a period of time and see how they develop further. But they are slightly at the lower end of the spectrum.
Okay. Fair enough. And then secondly, in terms of what happens when an NPL customer pays you back? Do you reinstate the same limit, or how does it work? Or is this credit card just cut off?
So if the customer pays back his entire due and becomes current for some set of customers who we feel to be now creditworthy, we will reinstate. But we may not reinstate at the same limit.
Okay. Okay. Fair enough. That answers my question. Wish you all the best. I'll get back in queue.
Thank you. The next question is from Shweta from Elara. Please go ahead.
Thank you, sir, for the opportunity. So I recall last quarter you were mentioning the strong repayment history led customers also were facing slight difficulties in repayment. And second, the stress was also apparent for customers with higher credit card limits. So how is the scenario now?
The scenario is what we see in our NPA pool continues to be almost the same. We see the mix continuing. So when we see our write-offs have gone up, the composition remains varied. As we see, as we saw earlier, even the mix will be an older vintage also, a customer doing well and being a transactor or turning to revolver, again turning to transactor, having a good history, then defaulting. That trend we have seen. Since our NPA pool continues to have this kind of a mix, we will continue to watch them and having a close look at them after having taken whatever actions, portfolio actions we could have taken on them.
As I said, that, based on whatever portfolio actions we have taken so far, the delinquency inflows are lower. Overall, the mix continues to be on a better side. Probably people are able to manage with whatever limits they have. They are transacting within what their limit they give. It will depend on what the NPA pool behaves over the next quarter.
Sure. Fair point. The second revolver portfolio, although marginal, has seen marginal spike. And given the current scenario, shall we read negatively into this revolver and potentially this could be an NPA customer? Or how are the indications for the revolver pool customer behavior?
No, Shweta. At the end of September quarter, we had very clearly stated that the numbers there, because of festival season having started somewhere around 18th, 19th of September, there was a lot of transactor NEA which got built up. Hence, the ratios were looking slightly different. And we had stated in September quarter itself that this would normalize by end of December. So this revolver has been stable as of now for the last four, five quarters.
Sure. Thank you. I'll come back in the queue.
Thank you. Before we take the next question, a reminder to participants that you may press star and one to join the question queue. The next question is from Nishant Shah from Millennium. Please go ahead.
Hi, sir. Thank you for the opportunity. So these are a couple of questions from me. Just a follow-up question first to Marukh's question. This quarter, we've seen a sharp dip, right? And you're saying categorically that this is the peak. But how should we think about modeling for credit costs going ahead? You answered this a bit qualitatively that it depends on the macro environment. But any initial kind of guidance or even a range that you can give that is this something which is like a peak and a plateau, or is it a peak and a reversal? Will the normalization be much sharper given the kind of sourcing we've done?
How should we think about the downward normalization of these credit costs? Let's say for FY26 or 27, more longer term, if you could give some guidance about how we should think about where these credit costs should settle at? That's question one. I'll ask my next question later.
Yeah. So very difficult. I will repeat. Very difficult to give a firm number of credit costs considering what uncertainties we have seen over a period of time. As I said, as I said, that the customer behavior has been changing very sharply over a period of time. And there is a segment of customer which I have stated it earlier also. There is a segment of customer which was flowing into write-off straight away without paying me a single penny.
Now, with all that taken together and on one side and on the other side, whatever portfolio actions we have taken, whatever collection efforts that we continue to make on early into the buckets, all that taken together has already given us. This has brought us to a stage where we are seeing the stage one, stage two composition significantly better. Even the stage three composition has come down. Now, this should not be related to the present credit cost. When we are saying if there is an improvement in the asset quality and the stages, definitely it will translate to the credit cost. How much, to what extent, it will depend, then we will like to see it over a period of time ourselves and very difficult to give a guidance for entire FY26 at this stage.
Okay. Fair enough. And the second question also on the other page. So around before COVID, we used to operate at a much higher level of revolver. And with the new sourcing that we've done, it's come understandably at a different revolver mix. How should we think about as the delinquencies start to come off, hopefully, should the revolver mix start to inch up from here? Or is this the new level of revolvers that we should kind of model in that 25% or thereabouts is what the new normal for the industry or for you guys is?
Any thoughts around that would be helpful. Because initially, if I recall, a lot of the reduction in revolvers was also attributed to fintech players doing a lot of STPL and small loans. That market seems to have slowed down significantly. So is there something to think about that revolver mix should start slowly kind of inching up from here?
So Nishant, as I was stating earlier, in fact, some of the new vintages that we are getting are as of now showing a slightly lower revolving behavior than the average spectrum that we have. We don't expect the revolver rates to inch up from here at all. If they remain stable or a percentage point down, that would be a good place to be in. We are, however, trying to work more at the point of sale, EMI conversion, and the other product which we call as FlexiPay, which is where the customer can convert the spends before the payment due date into installments. So that is the portion on which we are working at, and more activity is being done there. We are not trying to induce the customers to revolve at this stage at all.
Perfect. Sure. I have more questions, but I'll come back in the queue. Thank you so much.
Thank you. Next question is from Pranav Gundlapalle from Bernstein. Please go ahead.
Hey, good evening. Thanks for taking my question. You said one question on your sourcing shift that you've done. If you look back pre-COVID, you had a certain model for sourcing, which hasn't helped on the underwriting front, but it didn't allow you to grow or maintain a certain market share. Now that you have tightened your sourcing strategy, do you think the growth will be a lot lower versus the industry? Or is it more yet now throttled down because of the environment? But even with this new approach, will we be able to grow at the same pace as you did pre-COVID, let's say?
No, I'm only to ask if whether we will continue to keep growing our market share.
If you have sourcing better customers, can you still grow at the same pace as what you used to do in the past?
So growth rate at this point of time will be slightly because in the near future, in the next three to four quarters, what we are looking at is that the growth rate will slightly come down. So it will not be maybe at a rate of 30% or so, which was happening earlier. But it will tend toward we are looking at around 18%-20%. So this is how we are looking at the growth rates as of now in the near future. But we will continue to add new customers and maintain market share or grow market share.
Understood. Thank you.
Thank you. The next question is from Roshan Chutkey from ICICI Prudential AMC. Please go ahead.
Yeah. I wanted to understand how the minimum-due-paying customers, just about minimum-due-paying customers, not necessarily the entire revolver pool, but just the minimum-due-paying customers, how have they been trending for the past few quarters?
What is minimum-due-paying customers? So that data that's been improving. We've not even we don't share that data.
We don't share that data out.
That data will not be available.
We don't share that data.
No, I don't want the data around it. Just a trend of it, if you can talk a little bit about it.
Yeah, that is improving. So we can just share that the number is improving.
Okay.
I guess, sir, the combination, you see, you have to look at the combination. It's the minimum due being paid or entire dues being paid. All that together has contributed to the improvement that we are seeing now. It's not a standalone, it's not a standalone activity result that where only minimum dues being paid out. It's a combination of both the entities, minimum dues as well as the total dues being paid. All that taken together has contributed to the improvements, what we are seeing.
Okay. Next question was essentially because if the customers, if the pool of customers are just about paying minimum due, that means probably they are stressed, right? That is an indication of stress in some sense. Right? So that is the reason why I've asked this question. But if you are saying that these are increasing, then it's good to hear that.
Not necessarily. Not necessarily. They also get converted over FlexiPay and the EMIs also.
Right. Understood. Understood. Okay. And the other question I had is, I guess Sir mentioned that the return of the pool of incoming pool into the return of bucket is improving. But when I look at the data, 22% increase sequentially, right? What was 90% plus for us at the end of Q2 has probably been returned off the end of this quarter now, right? If that is an increase of 22%, how are we saying that the write-off pool is also improved?
Yeah. So the metric we spoke about earlier, there were four metrics that we wanted to allude to. One was the flow into delinquency. The second was delinquency to write-off. So the first has been coming down sequentially month on month almost for the past 12 months.
But the second part of the equation, which was the flow from delinquency into write-off, that flow had not been coming up. It was actually increasing for the better part of the last 12 months. In the last three months, however, it has been coming down. So that reversal or that change in trend has happened about three months ago. And that is what is contributing to the outlook that we are giving to you right now. It's also measured by two other metrics, which are 30% plus and 90% plus. Both have been coming down for us and also for the industry over the last quarter.
Thank you so much. That's all from me.
Thank you. The next question is from Dhaval from DSP. Please go ahead.
Yeah. Hi, sir. Thanks. A couple of questions. First is on the credit cost. So if you look at the sort of normalized level of credit cost for the business for about 67%, and if you look at the nine-month number, it's about 8% net credit cost. Do you think based on the current metrics that we are seeing, next year credit cost should be around the nine-month level? Or do you think there is hope that it can fall even further from where we are?
Qualitatively, if you can just think about when should we normalize, basically, that's the eventual question. So that's the first one. And the second is relating to cost to income. Any thoughts that you have on where the business is headed with this change in the underwriting that you've done? What's the cost to income sort of steady state that one should think about for the current underwriting that we are doing? Yeah.
The first one, I already stated that it's very difficult to give a number and predict a number. We are seeing I mean, we will see how it goes. But then, as I said, we are seeing a gradient, an inflection point reached, and then it should improve here onwards. But then being able to, it's very difficult to give a credit cost number. The aspiration can be anything. But then having seen so much happening in the market and in our portfolio, impacting our portfolio over the last almost four-to-five quarters, it will be unfair to give a number without really seeing how the gradient takes shape.
Also look at in terms of confidence, if I may ask, as an external investor, what should one look at to get confident that we'll be closing to 8% for next year or 67%? What is that one or two metrics that you think we should look at so that you can give comfort to investors? It's a sort of very important question. When we say peak, it's understandable, but it can be a hundred basis point reduction and thereafter stabilization like Nishant asked. And it can be a sharp fall. So that's what we're trying to get through.
No, but when you have to go back to a couple of quarters early into the year, when we had clearly stated that it will not be a sharp fall, we are on record on that. Because it can never happen that anything can improve overnight. Then if something improves overnight, then something was wrong in the entire system. So no. What we are saying is we'll wait to see how it happens, and then we will together see how it improves. But then we are forecasting an improvement. That's what we will continue to say as of now.
Got it. Got it. And on cost to income, you said?
From Dhaval, this quarter, obviously, the numbers—the last two quarters, the number has been around at 53.5%. So expect somewhere—in fact, these two were the festive quarters, so the number was higher. Quarter four number on the cost to income will be lower. So on an average, as we said earlier as well, the number should be around that 52, 50, 50, maybe not 50, 55 in a very broad range, but around a 52 kind of a number for the year.
Perfect. Yeah. Thank you for all the best.
Thank you. Before we take the next question, a request to participants to please limit your questions to one per participant so that the management is able to address questions from all participants. We take the next question from Manan Agarwal from ICICI Securities. Go ahead.
Yeah. I want to ask about the selection efficiency for the last one of the summer.
So we don't disclose a number, Manan.
Okay.
But the trend is improving, as I think a lot of data points otherwise we have been sharing between MD Sir and Shantanu and Nandini, they've shared a lot of other data points which indicate that the collection trends are improving.
And I attended all the con calls of micro loans, microfinance companies. So it is estimated that Q4 will peak out for the microfinance industry for credit cards and all. So what's your take on FY26 in particular about the credit cost and ROA?
Credit cost and ROA both?
Yeah.
So I think, Manan, as we've been saying, and I'll repeat whatever has been said by the speakers earlier, that we are saying there's an inflection point. And I think we'll have to wait for the quarter four results to come out to be able to give you any direction for FY26. The indication from whatever metrics that Shantanu and Nandini have shared, they are showing a positive trend, but the quantum, etc., is something which we still will have to wait for the full quarter results to come out.
Okay. Fine.
Thank you. The next question is from the line of Bhavik Dave from Nippon India Mutual Fund. Please go ahead.
Yeah. Hi. Good evening, sir. Just a question again. I'm circling back to the credit cost or the write-off comment that you're trying to address here. So if I do a very simple math where I take your gross write-offs and if I divide it by three, I get a number of INR 450-odd crores that we may be doing. Obviously, we do not look exact, but that would be the number that we'll be clocking monthly. And we were doing INR 3,350-odd crores previously in the previous quarter is what I see if I divide the gross write-off number by three, right? Very, very basic math. Are we saying that this INR 450-odd crores monthly that we were maybe writing off, that number has significantly come off? Because without data points, just qualitatively talking about this number going down makes little sense, right?
Because if you give us a trend line that this number is coming back to that 300-odd crore levels, then that's the way to think about it. It would be great if you could just help us with this. But where are we going in terms of monthly write-offs? Because unfortunately, we're not seeing any good trends in terms of slippages. When I see a slippage number, that has been on the higher side. So just wanted to get a sense on this one here.
So again, I will repeat. See, it's too early to give any numbers, too early to predict what will be the reduction. We have to see a scenario where there is a segment of customers who are sitting in an NPA pool and proceeding towards write-off unless we are able to collect from them and/or hold on to them wherever they are.
It's a combination of all these actions taken together which define how much will be, how many of them will get written off and what will be done on. It varies quarter over quarter. But as we said, that since we have seen a lesser inflow into stage three altogether, somewhere it will result into a lower write-off. Also on the performance metrics, slippages are actually improved in this quarter, and so has the PCR rate and the ECL rate and also the GNPA rate, although marginally.
Thank you. Next question is from Nitin Aggarwal from Motilal Oswal. Please go ahead.
Yeah. Hi. Thanks for the opportunity. Shantanu, sir, firstly, if I look at the proportion of salaried customers this quarter in terms of new sourcing has come down. While you mentioned that we are pretty much close to the peak in terms of credit cost and things should likely improve, how do you really compare between the self-employed and salaried in terms of their repayment behavior? That's one. And secondly, wherein you said that revolver rate, we are not expecting any improvement from here. I just wanted to check on as to how the revolver rate is different between UPI and non-UPI? Because UPI is something which, while it is small today, but it is growing fairly fast. How should one look at it in the medium term?
In terms of salaried versus self-employed, in self-employed, definitely the focus more is to look at the cash flows. And whatever sourcing we do, we do look at their bank account information, whether it is from the account aggregator channel or if the sourcing is from the banker channel. And the sourcing from the banker channel has increased in this quarter. So basically, it's an underwriting which understands the profile, understands the credits and debits in the account. And accordingly, we give a card and give a limit.
Right. But sorry. Your voice has gone off.
Sorry. Now audible?
Yeah. Yeah.
Yeah. So I was saying that because this number has trended down, but still, things are likely to come on credit card and it should get better from there. So this is the reason why I'll ask this question because the general notion is that salaried segment performs better than self-employed. So now with this mix coming down, should this not continue to cause worries as such in terms of incremental delinquencies?
No. Because you see, when you're underwriting, you're looking at the customer profile in general as well as, like I mentioned, his cash flow information. So this should not result in an increase in delinquency. And through the account aggregator channel, we are able to see the transactions in the bank account and assess the cash flow of the customer. So I don't think that is a worrying signal. A lot of these customers come with their bank statement entry.
Thank you. Before we take the next question, we request participants to please limit their questions to one per participant. Next question is from Jignesh Shial from InCred Capital. Please go ahead.
Yeah. Hi. Thanks for the opportunity. Basically, I have one data-giving question quickly. We are seeing a surge in investment sequentially. Anything to add into it?
This is in line with the regulatory requirements. There was a graded increase in the requirements for keeping HQLA, which is highly liquid assets, for LCR purposes. And we've been adding on to that number based on the requirements.
Okay. And much basic question.
So the liability book is also growing. So there are two reasons to that. It's a function of how your liability book is growing. And the second one was the regulatory requirement. So our liability book is growing as well as the regulatory requirement required us to keep adding about 10% on a regular basis, on a phased manner. And that's the reason.
Understood. And just the trend on fee income side, I think it's more or less flat if you see sequentially. Overall trend has been consistently declining on a YoY basis. So how do you see more co-branded cards coming up together? Along with it, obviously, online spend coming up, which basically means we generally take a little lesser cut on it. So how do you see the fee income trend going forward? Do you see that improvement can come up, or what's the view?
The fee income actually is made up of three parts. There's a composition that we put out on the investor deck around page 11, which talks about I think it's the spend-based, installment-based, and subscription-based is where you see the stickiness on the fee income side. There has been some bit of a moderation on the fee income, one because on account of the number of cards that we are doing, that obviously factor is an important factor as to what makes up the subscription-based fee income. The spend-based has a seasonality attached to it.
I would think that as we do more cards and as the amount of spend that we increase, this number should go up. Where we might see a bit of a moderation would be in the installment-based. That could see some bit of a moderation, but the other two line items should see it growing with the business growth.
So as our acquisitions increase, so the income will also be increasing going forward. That's your fair assumption, right?
That's right. Yeah. And of course, the spend will also grow, so that part of the fee income will also increase.
Perfect. Sounds good. Thank you so much.
Thank you. The next question is from Rohan Mandora from Equirus Securities. Please go ahead.
Good evening. Thanks for the opportunity. I'm sorry to touch upon that flow rate question again. So if we look at your Stage 2 plus Stage 3 in 4Q FY24 as a percentage or absolute amount in 3Q, 3Q is still higher. And the commentary that we gave in 2Q was still that in the first half of current financial year, the flow rates have improved. So what you said in the call today also is that the flow rates into delinquency is lower, delinquency is to write-off is higher. But this is not reflecting in the Stage 2 numbers that we are seeing right now.
So if you can share what is the 30 plus, how has it moved in the last two, three quarters, and also what is our share of exposure in, say, very high risk and high risk category of bureau rating customers? Something around that would help to get some sense on how the improvement is happening.
Yeah. So the 30 plus and 90 plus numbers, as we mentioned earlier, they have come off significantly in the last quarter, quarter ending December versus quarter ending February.
So what if I give numbers that would help?
But we don't disclose those numbers in that level of precision.
Okay.
Sorry. Can I first, Rohan, ask you a clarification before we further answer your question? You just mentioned that if stage two and three put together, you're not seeing an improvement. And where have you taken the numbers from?
On the presentation data that comes out. So I'm comparing 4Q versus 3Q, 4Q 24 versus 3Q 25, right? Yeah. So see, if we look at the numbers was 5.7 plus 2.8.
So it's the 4Q number, sorry. See, if you look at the quarter two FY25 and quarter three FY25, and you calculate the percentages to the NEA number given at the top, and you add the Stage 2 and Stage 3, you will see a reduction in the stock of Stage 2 and Stage 3 for quarter three as compared to quarter two.
Yeah. There's a margin improvement from.
You said the spend-based is sorry?
There's a margin improvement on 2Q, even in absolute amount from 2Q to 3Q. But what I'm saying is, see, in the previous quarter, also the commentary was that flow rates into delinquencies have been coming down. So if I'm comparing, and this was the commentary for the first half as well of current financial year. So if I'm trying to look at from the 4Q FY24 numbers, wherein your stage two and stage three were 5.7 and 2.8 respectively, and if you look at stage two plus stage three combined, it was around INR 4,300 crores. That thing has now gone up to around INR 4,900 crores. So that flow rate coming down does not match with the data I shared comparing with the year-end number and the nine-month number for the current financial year.
So Rohan, I don't think, and we'll probably have to go back and check, but I don't recall that we've ever said that the flow rates have been improving in the last two quarters. We've just said if there was a mention made on the delinquency numbers, it was to do with the sourcing that has been done in the recent part. There we said the SPF [guess] is seeing a better performance compared to the older portfolio. But this time, we are calling out and saying that our 30 plus and the 90 plus numbers are looking better. As I mentioned, if you do a total of our stage two and stage three, even that number is lower.
And as you mentioned earlier as well, that what flows into write-off is nothing but the stage two and the stage three flowing into write-off. And this is a number which is improving. And the flow rate, in addition, for the earlier buckets are improving, which is where we've been able to make a comment saying that we think we are at an inflection point.
Just to clarify, last quarter, there was a specific question around this point. And the answer we gave was that while entry into delinquency is improving, the next leg, which is delinquency to write-off, that was worsening. And that is mentioned very clearly in the FAQ from the Q&A transcript from last time. And this time, there's been a reversal. So the first time, we're saying the second half of the equation also has behaved properly, which means improvement.
So while the entry into delinquency continues to improve, where we are now seeing is an improvement in the stage two to stage three, which is where we are now saying that the write-off number should be lower, and therefore, the credit cost should be moderating.
Got it. Got it. Thanks. And just to reconfirm, for FY26, any guidance that we're given on the revenue growth or spend growth?
We just gave it. We just checked it. We're looking at somewhere around 12%-15%.
Thank you. Next question is from Subhranshu Mishra from PhillipCapital. Please go ahead.
Hi. So the first question is around the managing position. So we've got quite a few people as directors from the parent SBI. Why do we need someone at the MD who may or may not have requisite professional experience in card business? Why don't we hire a professional MD CEO from the credit card industry to run the business, whereas we can have board member representation from the parent? The second is, through cycles, what has been the recovery from write-off accounts in year one, year two, and year three? And the third is the number of customers who have two cards, number of customers who have three cards, and number of customers who have four plus cards. Thanks.
The first one, I think we don't need to comment.
I want to comment on the second one. You asked for Subhranshu to gather the information. We don't disclose that at that kind of a level, where we kind of disclose as to how many customers have three cards, four cards, and five cards. I don't think anybody in the industry discloses that number, and we will not do so as well in the future. We haven't done that, and we will not do it as well. On the recovery, as you asked the question that how much are we recovering from the written-off pool, we did see a bit of a dip in the recovery from the written-off pool in the last one or two years. There is some, again, green shoots.
Again, we do not share the exact numbers, but the recovery numbers, as you can see, for this quarter were higher from the previous quarter. The recovery numbers did get impacted as per the overall environment that was there for the last 12 to 18 months, but we have seen a higher recovery number for this quarter compared to the previous quarter.
Now, what I'm asking is that through cycles, what percentage have we recovered from the written-off pools, not in the last 12 to 18 months? Through cycles, what is that number?
We don't share that number, Subhranshu.
Okay. What we do share is that we do get recoveries for a long time after write-off. That's been the case.
See, and we've said that as well in the past, and we will maintain that, that we do continue to get recoveries for a longer period from the written-off pool. But in full transparency, we have shared as to what the recovery number has been quarter on quarter.
Thank you. Before we take the next question, a reminder to participants to please limit your questions to one per participant. The next question is from Pranuj from J.P. Morgan. Please go ahead.
Hello. Thank you for taking my question. So one thing you mentioned was that you constantly monitor the leverage levels of the customers and take corrective actions if you see increase in stress. So just a clarification, is the corrective action that you can take only in terms of limiting the spend limit that the customer has or limiting the credit limit that you give to the customer? Second is on the frequency of scrub. I think previously you guys issued a monthly. Will you be shifting to fortnightly now that the regulations allow for that? Thank you.
Yeah. Basically, the actions, yes, are reducing the credit limit, reducing certain kinds of spend, doing early blocking. All these actions are taken. Secondly, with respect to the bureau pull, we get a bureau trigger product we have subscribed to, which basically means that whenever the information is updated in the bureau for a certain set of accounts, we get information on T plus one. If there is an increase in leverage or inquiries or there is payment, etc., happening, we get to know, and we accordingly plan our actions. With respect to the 15-day window, so while everybody has started the process, since we subscribed to this product, we will anyway get the information. But yes, we are looking at how we can reduce our bureau pull period with the new updates which are happening fortnightly.
Thank you. We'll be able to take one last question. We take the last question from Rohit Jain from Tara Capital Partners. Please go ahead.
Yeah. Hi. Can you hear me?
Yes, please.
Yeah. Go ahead.
Yeah. My question is that this guidance of loan growth for next year, 12%-15%, given that we are now incrementally sourcing better vintage customers, and a lot of poor-quality customers would automatically get weeded out because of the stress over the last two years, is it fair to say that this now seems like more sort of a medium-term trajectory, and this is how it would be for the next one or two years?
So this projection is for a period of next 12 months, in 9 to 12 months. Things can change. If we see that the credit card trajectory is very different, things are different on a macroeconomic perspective, things can change either way. It will depend on how we see the macro environment and our portfolio behaving over a period of next three to six months. This is a growth rate as of now, looking at the future environment in isolation.
I have just one more question. On the cost to income, you said for the next year, the broad range is 52%-55%. But in an environment where, let's say, growth picks up, would it be fair to assume that the cost to income would also go up?
Yes. Yes. It's a function of how many cards that we do. If growth picks up, absolutely. And by the way, that number that I gave you about 52%-50% was for this year, but you're absolutely right. The growth will determine the cost to income ratio.
Got it. Got it. Thank you. Thanks a lot.
Thank you very much. We'll take that as the last question. I would now like to hand the conference over to Mr. Chakravorty for closing comments.
Thank you, Rao. And thank you, everyone, for your continued trust and confidence in SBI Cards. And have a fabulous year ahead. Thank you.
Thank you very much. Very much. On behalf of SBI Cards and Payment Services Limited, that concludes this conference. Thank you for joining us, ladies and gentlemen. You may now disconnect your lines.