© 2024 Alpha Spread Limited. All Rights Reserved.
www.alphaspread.com is not operated by a broker, a dealer, or a registered investment adviser. Under no circumstances does any information posted on www.alphaspread.com represent a recommendation to buy or sell a security. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, investment advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. In no event shall Alpha Spread Limited be liable to any member, guest or third party for any damages of any kind arising out of the use of any content or other material published or available on www.alphaspread.com, or relating to the use of, or inability to use, www.alphaspread.com or any content, including, without limitation, any investment losses, lost profits, lost opportunity, special, incidental, indirect, consequential or punitive damages. Past performance is a poor indicator of future performance.
You don't have any saved screeners yet
Intrinsic Value is all-important and is the only logical way to evaluate the relative attractiveness of investments and businesses.
Ladies and gentlemen, good day, and welcome to FY '24 earnings conference call of SBI Cards and Payments Services Limited. [Operator Instructions] Please note that this conference is being recorded.I now hand the conference over to Mr. Rama Mohan Rao Amara, MD and CEO. Thank you and over to you, sir.
Thank you, Yeshashvi. Good evening, everyone. I'm pleased to welcome you to the Q1 FY '24 earnings call along with my senior management team. I also take this opportunity to introduce Mr. Shantanu Srivastava, our Chief Risk Officer.This fiscal is special to us as it marks a key milestone in our organization's history. As you are aware, SBI Cards celebrated its Silver Jubilee on 15th May this year. Through these years, the company has exhibited business resilience and strength to stand the test of times. Our customer's patronage and trust have been instrumental in our success, and we remain committed to delivering value and growth to all our stakeholders.As we meet today, the global economy continues to be impacted by the prolonged geopolitical tension and volatility in global financial systems. However, India is well poised to sustain its growth momentum. In its annual review report, Ministry of Finance has said that Indian economy appears to grow more globally than before.India's GDP has grown by 7.2% in FY '23 as against an earlier estimate of 7%, owing to a strong Q4 performance. It is reassuring to know that India continues to be the fastest-growing major economy and should remain so in the next few years.Digital Payments continue to grow at a fast pace with transaction volume growth of 56% year-on-year in FY '23, which is expected to grow 4x by FY '27. Indian credit card industry continues to grow at a steady rate. As per RBI June 2023 data, there are around 89 million cards outstanding in the country. In June 2023, credit cards' spends have been at INR 1.37 lakh crore, marginally lower than record high INR 1.4 lakh crores issued in May '23.Coupled with highly under-penetrated nature of Indian credit cards market, this underlines the continued relevance, attractiveness and long-term sustainability of the credit cards business. Many progressive regulations have also reinforced the growth prospects of the industry. These have balanced the healthy competition in the industry with customer-friendly measures.RBI Payments Vision 2025 outlines that card acceptance infrastructure will increase to 250 lakh touch points, which will further boost transactions in India. Being the largest pure play credit card issuer in India, the SBI Card is at the forefront of industry's growth by driving up the penetration of credit cards in the country. As an agile organization, the SBI Cards continues to asses and explore opportunities that can support the growth momentum. With improving consumer sentiments and stable macroeconomic conditions, our aim has been to leverage this window to test and expand the variety in customer base while being prudent.During Q1 FY '24, we have introduced Paytm SBI cards on the RuPay network, thus offering more choices to our valued customers. We have started roll-out of SBI Card issued RuPay credit cards on UPI platform on the limited number of pre-packs and this will be expanding into other assets. Over a period, we expect this to be a significant contributor to our transaction volumes.We continue our investments in technology. Our app platform available both on Play Store and App Store, was revamped this quarter by enhancing the UAE, and U.S. to provide ease of navigation with access to several service and features just like digital KYC and cross-sell features. Our app has been downloaded by 13 million gross users with 7 million unique monthly login by customers. It has been rated 4.5 out of 5 on Play Store and 4.6 out of 5 in App Store. The app has been revamped in a phased manner, several more features to be rolled out throughout the year.Let's now look at SBI Cards business overview in Q1 FY '24. During this quarter, the SBI Cards pursued and achieved strong growth across most key business metrics, once again demonstrating resilience and sustainability of our business model. We crossed 17 million cards in source milestone.Our clip stands at 1.73 crores with a growth of 21% year-on-year. This market share is at 19.6% in Q1 FY '24. Our new account in Q1 FY '24 stands at 10.97 lakhs with a healthy 22% year-on-year growth. That current quarter saw a moderation in growth in line with our experience during Q1 for any fiscal. Our share of new account sourcing from bank on open market channel in Q1 FY '24 stands at 54% and 45% respectively. We continue to leverage the networks at a customer base that SBI has.We endeavor to keep our market share at net card addition at around 20%. Our card spend growth has been strong, registering a growth of 24% year-on-year at INR 73,913 crores. In fact, spends in Q1 FY '24 are the highest ever quarterly expense for SBI Card, superseding the performance we achieved in the previous quarter. Retail spends have contributed INR 58,347 crores with 28% year-on-year growth, thus, making it the best ever quarter for us in retail spends.Corporate spends have been stable and have contributed INR 15,565 crores with a 10% year-on-year growth, in line with our measured approach for the segment. Our spends per average card has grown from INR 1,70,000 in Q1 FY '23 to INR 1,73,000 in Q1 FY '24. Specifically, we have seen a healthy growth in spend per average card per our retail spends which have grown from INR 1,30,000 in Q1 FY '23 to INR 1,37,000 in Q1 FY '24. During Q1 FY '24, spends growth have been driven by growth in categories such as travel, dining, and entertainment, apparel, education and utilities.Online spend continues to be robust with a share of around 55% in retail spends. Cost witnessed a faster growth than online in Q1 FY '24. Our market share in card spends stands at 17.8% in Q1 FY '24. With robust cards spends, we have also seen a healthy 30% year-on-year growth in receivables too. Our receivables have grown to INR 43,271 crores as of June end.I would like to highlight that our share of interest earning receivables have grown to 62% in this quarter versus 61% in Q4 FY '23. Our revolver stayed stable at 24% and EMI receivables were at 38% of the portfolio. Most importantly, our receivables per card have grown to INR 24,949 for the Q1 FY '24 versus INR 23,202 in Q1 FY '23.Coming to the financial performance in Q1 FY '24, our total revenue in the quarter is at INR 4,046 crores registering a growth of 24% year-on-year. In Q1 FY '24, our revenue from operations is INR 3,912 crores with 46% year-on-year growth. Interest income saw significant share of our revenue from operations, which has been growing steadily. In this quarter, interest income contributed 46% share and has seen a growth of 30% year-on-year.Our PAT for the quarter that is Q1 FY '24, stood at INR 593 crores. Our cost of funds during the quarter increased to the 37 basis points. We expected the short-term rates to be lower and eased to normalize, however, higher than expected short term rates and opportunistically borrowing more long-term debt as long-term rates eased off, resulted in our cost efforts being higher by 37 basis points quarter-on-quarter. The long-term borrowings, as a result, increased from 35% to 37%.Given the predominantly short-term nature of our assets, we would be comfortable with long-term borrowings at the current level. During the quarter, we were also able to mostly transmit the increase in cost of funds to our asset book, resulting in increase in the yield on growth by 20 basis points, helping us keep our NIM almost stable at Q4 FY '23. We expect the cost of funds in Q2 to be marginally higher by 5 to 10 basis points from Q1. NIM is expected to be stable next quarter.Our asset quality. As of Q1 FY '24, GNPA marginally increased to 2.41% from 2.35% as of last quarter. Our gross credit cost increased to 6.8% in Q1 FY '24 from 6.3% in the last quarter. As we shared with you in the last quarter's earnings call, our credit cost was elevated on account of stress emanating mainly [Technical Difficulty]. We have noticed that the delinquency behavior for the 2019-2020 has been relatively worse, considering its weightage in the portfolio. The lifetime billing with the behavior service vintage deviates from the expected lifetime behavior. We have identified the swift segments and have taken suitable portfolio actions. We have also intensified our collection efforts for this cohort.While the overall new sourcing qualities within the desired delinquency band, we have proactively discontinued sourcing of less profitable low limit cards, terminated a few pilots which are not really significant, increased documentation and verification requirements for certain low-income segment, and reduced the sourcing from certain geographies.We also took portfolio actions on potential high-income segments by reducing their credit limits and tightened eligibility criteria for cross-sell to these customers. The collection efficiency for the industry, particularly the letter buckets have not come back to pre-COVID level, which could be partly be attributed to the change in customer behavior.We reviewed our collection strategy and simplified field efforts including enhancing the field infra in geographies with highest risk portfolio. Our collection teams continue to leverage digital tools available.We are certain that the gross credit cost have increased in Q1 FY '24, thus improved flow rates in the month of June, that is June 2023, resulted in lower credit cost as compared to April and May months. Initial trends in July also indicate continued improvement and we expect the credit cost for the coming quarters to be on a downward trend.In addition, the delinquency for our recent sourcing vintages is better and in line with expectations. As the weight of these recent better quality vintages increases in the portfolio, the overall portfolio quality will also improve. As you would have noticed, we have increased the sourcing from self-employed in the Tier 3 and Tier 4 segments in the last few years. The delinquency behavior for these segments have shown an improved trend over the last few years, thus increasing prosperities and therefore the increased consumer demand from Tier 3 and Tier 4 cities in India, will continue to drive growth for the financial sector.Government initiatives, growing consumption and driving SME enterprises in these cities encourage us to focus on these cities. Access to SBI network in these cities gives us an edge over other players. We are closely monitoring the performance of this loan segment and we'll calibrate our approach as and when appropriate.With the steps taken in Q1 and encouraged by the results, we expect credit costs for the second half of FY '24 to trend back into our target range of 5.8% to 6.2%. Our cost to income for Q1 FY '24 is at 56.4%. We find improvement in cost to income ratio to 56.4% versus 58.1% in Q4 FY '23. Our profitability ratios continue to be robust. In Q1 FY '24, despite an elevated credit cost, our return on average assets were at 5.1% versus 5.4% in Q4 FY '23. With steps taken to reduce the credit cost and increase share of digital sourcing over the next few quarters, we expect to maintain a healthy ROA going forward.In summary, or in conclusion, India's growth story remains intact, and the domestic consumption is encouraging. At SBI Card, we have always maintained an agile approach and have taken well calibrated measures to ensure sustainable and profitable growth. We embarked on this fiscal year on a positive note as we gauge from our strong performance across most key business parameters. Amidst this ever evolving macro environment, we stay focused to take advantage of the immense growth opportunities that the biggest credit card market continues to offer.Yeshashvi, we are now open for questions.
[Operator Instructions] We have a first question from the line of Mahrukh Adajania from Nuvama.
Sir, my question is on credit cost. So the 2019 customers, what would be their rough contribution to current outstanding? Any rough idea?
Yes. This is Shantanu, here. The current composition of our portfolio is about 16%. So 2019, so think about 16% of our [ revenue ].
Okay. And what would be the range of credit cost in the near term? So obviously, you've taken a lot of portfolio actions, so say for the next 2 to 3 quarters for the rest of FY '24, what is the range of credit cost we can expect over the next 3 quarters?
As I said in my speech, I think we have -- we started seeing a positive trend down in the sense like we started the -- seeing decline. Month of June, we had a little bit of -- looking at the entire month of June, because better than the previous month like April and May. Essentially it means like it has peaked in the first quarter up to me, the credit cost, monthly we are monitoring now very closely. So it started declining. We are confident with the kind of measures what we have taken and initial trends what we are seeing, after September quarter, we will be back to our target range of 5.8% to 6.2%. So you can actually estimate like, where we will be. We are at 6.8% in this quarter. But after the quarter, we will be in the target range. So you can estimate like how it will trend.
Okay, got it, sir. Got it. And sir, my last question is on margins. So see, rates are cut at some point in time, not immediately, but RBI cuts rates, then how will your margins behave, not only from a cost of funds perspective but from yield. So how soon do you pass on rate cuts to your customers in terms of yield? How will the yield side of things move?
See, if you look at our past record, whenever the rates were declining or whenever the rate environment is very benign credit of environment, we have always been the beneficiary in terms of improving yield. That has been the case because of the lag effect, like the way, what you are seeing now. With the rates are increasing, our liability is getting repriced faster but it takes some time for us transmit the rates fully to the asset side, but exactly it will work in our favor when the rates are declining. So any declining, while I'm not asserting any guess here, so when RBI will declare, I mean, reducing the rate, but that will be very positive for our margin.
We have our next question from the line of Anuj Singla from Bank of America.
Sir, first question is on, again, the credit cost. So we talked about, we have identified this 2019 cohorts, where we are seeing stress. So can you talk a bit about what is the profile of these customers? What went wrong there? And what confidence do we have that the incremental sourcing we have done, maybe in the previous years or the years after that, we are not going to witness the same or similar stress in those cohorts?
So this is something that we've been very focused on in the last 3 to 6 months. Incidentally, I joined in April, so just about 3 months ago. And ever since that we have been running diagnostics of the problem and collectively thinking about the actions should be taken, some of them are already underway well before I came. It takes some time to execute, and then result also take some time to show up, but that's been happening and we are seeing green shoots already. But to answer specifically what you were seeing in 2019 sourcing was, that the behavior of this cohort was somewhat different from what we noticed in the prior or subsequent cohort, in terms of both peaks that this particular cohort witnessed which is around 30 month par. The level of that peak as the speed at which that peak was crossed and the delinquency behavior cabled out. That took a some sort of different trajectory compared to the other cohorts. This is not very different from what we expected from the model, but it took some time for the discrepancy or analogy to show up because it was coming just off the pandemic. So that point of time to distinguish the portfolio from the others was not -- or distinguishment of the portfolio from the others wasn't very easy. But as and when we did it, we've started taking those actions in terms of light increases, ad hoc voice certification of collection efforts.In terms of the other subsets of this segment, some of the charts are available in the investor deck. You can see that alongside these portfolios, there is a component of our portfolio that come from self-employed. In this duration of the last 3 or 4 years, the contribution of self-employed segment has gone up by about 4 percentage in the 3-year period -- in half year period. Similarly, at the same time, the delinquency behavior of this segment has actually improved and there are 4 other similar sub-segments of this portfolio that -- we call SE, whether it's Cat A, whether it's Cat B and salary statement over this 350 portfolio. All of those have moved towards -- we move towards the higher component as you try to generate revenue. But the riskiness of each of these segments has improved in the last 3 or 4 years. So that's a hardening thread and as the portfolio matures, and we get a higher and higher component from our more recent sourcing that we find is more suitable and giving us much better delinquency performance in terms of behavior, we are encouraged that the overall delinquency of the portfolio will improve and it's on that basis that we are giving you the comfort that and we just spoke about 2 minutes ago.
But anything specific here? Is it self-employed -- was it open market or bank outsourcing with a cohort came from? Any color you can share there, or it's a mix?
It's actually a mixed bag. We can't really point out towards only one component, but there were sub-segment we're seeing whatever cohort we identified and the look a likes of those cohorts that we're seeing through our portfolio, those are the ones we're taking actions on from our portfolio actions point of view or from our marketing actions point of view or from our relation action point of view.
Got it, okay. The second question is with regards to the growth trajectory. So Mr. Rao talked about these intervention as well. You've also put in Slide 6 to focus on profitable segments, cutting off credit limits, and also sourcing maybe some of the pin codes, they're not sourcing for. How do we see the 1 million net card addition target per quarter in that context? So should we be expecting a lower number in the coming quarters?
I think we are still mindful of clocking, that's the 300,000 per month and about a period of a quarter 1 million kind of thing, only as rightly said, like some of the steps, the recalibration, et cetera what we have done. I think it's the seasonality of what we've seen in Q1 combined with that you can see, there is result but there is no change in our aspiration. But only thing is, it will take a month or 2 before we hit the kind of mark. It is the initial aspiration to reach 300,000 in a couple of months, and up to gradually -- increase the run rate, but we'll will try to be more careful and we will try to balance it with how the credit cost is also behaving. So we'll be mindful of both.
We have our next question from the line of Piran Engineer from CLSA.
Yes. I just wanted to understand your delinquency chart on slide 15. So how exactly do I read it? So all of them are up. They reached their peak in 24 months on book and then go down, but you're also saying this is ever 90 plus. So shouldn't that just always be increasing?
So, this is the standard industry slide and you will notice the shape of the curve is very similar toward they might have been elsewhere --
No, no -- I meant -- no sir, I just meant that, say the 30-month on book number should always be higher than the 24-month on book, which should be higher than 12 months on book, because this is ever 90 plus. So I'm not sure if I'm understanding this chart properly.
Yes. So here, this is an incremental number. You are right. If you actually do cumulatively, it will continue to grow. But the point here that we wanted to show us that it -- the maximum peak after a buildup of assets happens at around 24 months, and after that the incremental number of people coming into 90 plus keeps on going down and you just manage that portfolio. So that's the chart that's representing that term, and also index. So it's not the absolute value. It's the index value.
Got it, got it. Okay. So it's an incremental number of 90 plus, okay. But in that case, the dotted line which is CY '19, now it's been like 4 years, right, since it's CY '19 and that number has come down anyway. So I don't see why it should impact us in 2023. Am I making sense?
It's about 16% -- it's about 16% of our assets and about cycle or 20% of our NPAs. So slightly it's postured downward. The incremental amount of credit cost that we're carrying because of this particular port is about 14 basis points.
No, no, sir, I don't think I've communicated my question well. I'm just saying that, so if I compare the CY '18 line and the CY '19 line, they are probably pretty similar. But last year we did not see any sort of hit due to the CY '18 line, but this year we are seeing the hit from the CY '19 line. So I'm not really understanding why?
Yes, it's the composition of the cohort in the overall asset. So CY '18 ran off because and it is not contributing significantly to the overall rate cost, because it's much smaller in value. In '19, it's 15% in our value.
This chart, see on the slide, Piran, is -- for the -- at the portfolio level doesn't show the as Santhanu pointed out, the percentage contribution in the portfolio. Just behaving a portfolio behavior. It's the behavior of putting ever 90 -- 90 plus in percentage. When you put this in context to the percentage of this vintage in the portfolio, that's where it starts to show up in a higher rate.
Makes sense.
Also the other point of putting this chart was to also show you that the vintages which are belonging to 2020, '21, '22, which are the latest vintages which we acquired in last 3 years, are almost on a index level at the peak of around 0.59 or 0.6 compared to what we were acquiring earlier. So as these new vintages build up asset, automatically the weighted average will start to come down. So the point is that, the subsequent cohorts are speaking earlier and tapering faster.
Okay, got it, got it. And the second one just for Rashmi, now if RBI does not hike anymore cost of funds should be stable or do we see that increasing and in the last one year, how much have we increased our EMI yield by?
Our EMI yield has gone up by around 20 plus basis points. The yield on our EMI book would have gone up by about 120 to 150 basis points. Even if the RBI doesn't increase rates any further as Mr. Rao called out, we do expect the cost of funds to go up this quarter as some of the other liabilities that come up for repricing, both on the short-term and as well as on the long term as well. Remember our long-term book which you would have borrowed 3 year back is at a much lower price. So factoring all of that, we called about 5 to 10 basis points increase in cost of funds at Q2. I'm hoping post that, the rates kind of become normal for the second half of the year.
We have our next question from the line of Abhishek M. from HSBC.
Yes. So I had a follow-up on this yield on EMI. How much of the 120 to 150 bps would have translated already and how much would be get repriced as we go along?
The numbers that I called out is the translation of the cost of funds going some -- the increase in cost of funds being passed on the EMI booked.
No, I meant, the increase in yield on the EMI book, 120 to 150 basis points, that would be applicable only the book sourced through over the last, let's say 2 to 3 quarters, right? So there would be a part of the EMI which is not --
That's right.
Okay. So anything....
What you would like to know is....
So what I would like to know is that what part of that EMI book is yet to reprice?
So we can't reprice our existing book at all. The EMI loans are a fixed rate loan.
Okay. See, Abhishek, what happens is, you know the nature of the business is, we extend only fixed rate kind of loans, so they will be fixed for the period or tenure of the loan, but every month we make certain disbursements, spends converted into loans or otherwise encash kind of loans being given are click-to-pay, or balance transfer, et cetera that you can opportunity to transmit the rate increase. So it takes a few months for the entire portfolio, because considering the life of the portfolio, which is around may be 11 months or less than a year, it takes that kind of time for the entire portfolio to get repriced.
Got it. So 11 months to a year is the --
On the FlexiPay and Subvention, more than 80% of the book would have already got repriced because it is as -- average it's tenure is 9 months. So it's only some of the 18-month book will be left and cash is where the repricing is still -- continues to happen because the average ticket size -- average tenure is close to 33 months.
Exactly. Okay. Okay. So how much of that encash book would have been at new rates. Let's put it that way. Anything in the last 2 quarters, basically.
Actually, less than 12 months is fairly less because personal loan ticket size is primarily in 24 months and above. So the new addition that is happened into that book or incremental book is at a higher rate, the replacement cycle will continue to happen over a period of time.
Got it. And what part of that EMI is on encash versus the other 2, like FlexiPay and Subvention?
We have not declared the breakup of the all 3.
Got it, got it. Okay, no problem. So that was on yields. The second one was on this collection teams, which were enhanced in certain geographies, et cetera. Do you think this will have a sort of follow-on impact on costs on an overall basis on OpEx?
So I think, this is our [Indiscernible] And what we do there is a -- the full model is based on variable cost. We only kind of save the money comes in. So this addition is actually on the free channels, which is basically a [Indiscernible] channel. So we are not --
While the absolute collection cost will go up, but there are obviously benefits coming in as well as better collections and better recoveries as well. So, we are mindful of that numbers.
And also better collection efforts that lead to lower LGDs as well that helps with our credit cost.
We have our next question from the line of Bhaskar Basu from Jefferies.
Yes. I just had a follow-up question on this 2019 cohort. I just wanted to understand, I mean, before these accounts slipped. And I assume that these are spends made by those customers now. They would have ideally slipped into a revolver book and then subsequently become an NPL. So why haven't really been seen any impact on revolvers because of this whole slippage and why do they continue to be there in the system? I mean they should ideally probably rejected out after they become an NPL? That's all from my side.
So I think Bhaskar, the kind of behavior what we have seen with regard to this cohort is, while they have been bearing some record -- payment record, the moment we become stressed, particularly with the moment we cross 30 plus, we have seen behavior weigh, it is trade flow practically. That means our ability to normalize from this segment was limited. That is what we have noted.So that means you will not see -- the rate is not like a regular revolver rate, they continue to pay maybe more than 5%, 10%, 20%. But if it's just like an inability to pay, which will be more like a wash in the revolver. So that way you won't see a significant increase in the percentage of revolver because asset is also growing on the other hand, -- other side. So it won't be a kind of a very big increase on account of it. But it is only the segmental behavior when you track only, then only we'll come to know that actually there is some -- it's not a regular revolver, it will more like a deliberate revolver.
And the fact that we continue to see that stress, does it mean more and more borrowers from that pool continue to default? Is that how it works or is it the same borrower making some payments, continued to make higher spends. Is that the reason why this number continues to be elevated?
So for the last 2 quarters, as we had identified these accounts, we did see some increase, but if you look at this slide which talks about the credit actions that we've taken, obviously we've taken steps to now make sure that their utilization of the credit card limit, and therefore the addition to the stress reduces. And that's where we are hoping that as that stress reduces and the collection efforts help us in recovering the already lent out money, is where it will help us to reduce the credit cost.
So we are very careful about 2019 segment. So, and there is a deeper cautiousness, as -- if a customer has to revolve, we go after that customer at the initial stages itself. As I said, they are going through collection efforts are more -- are higher, try to identify look alike, do credit declines, very limited declines in some of those segments. So there are multiple actioning which is happening. We are restricting ourselves from cross-selling products to these customers at this stage. And that's how the whole asset percentage has come down to mid teens for this segment. And it will continue to go down in weighted as we grow the good assets. which is coming, as you can already see from that slide 15 which is coming from 2020 onwards vintages. Okay. So that is the way that we're looking at it. Those new vintages as you see, they are peaking at a lesser rate, but that also along where it goes that revolve rates are also lower which we are fine with, because the idea is to keep the revolve rates if we have the revolve rates at 24% and the get credit cost within control, we can continue to grow from that by working on the term lending portfolio.
We have our next question from the line of Swetha from Elara.
Sir, any sale to ARC this quarter or anything anticipated in near future?
So we keep looking at these opportunities as and when we think it makes more sense to fill up our portfolio to the ARC, we will do that.
So nothing as on today?
So if your question is, did we do anything this quarter, the answer is no.
Okay. And my second question is somewhat pertaining to the previous question. So look, if I look at systemic delinquencies on the credit card portfolio, they have been clearly rising at least for the banks. So in such a scenario. I mean we did highlight the fact that incrementally the new portfolio has been showing a decrease in delinquencies. But then in such a scenario generally it tends to be like that, in a scenario where delinquencies tend to raise, your revolver book tends to pick up. So can you just give us color on how the revolvers look like from your own?
Your question is that the delinquencies are increasing and therefore the revolver should go up.
Yes.
No. So the way that you look at it is, if it is -- we only look at -- so what is the revolver? Revolver is a customer who pays up to 5% or more on the outstanding balances. Okay? These are the people who are current also and they pay interest. So they are the people who are good revolvers and that is what we have been stating in earlier calls also that these days even that revolve behavior has changed. Earlier people used to be -- they will be revolving, 6x, 8x, 9x in a year. Now people revolve, maybe 2x or 3x as and when the short some fund requirement is there. Okay?So people who keep moving from transactor to revolver, revolver to transactor, so that happens. If the customer becomes delinquent, this is a payment. In any case after a point of time, whether it is a 45th day or 60th day, the card gets blocked. And we will not allow the customer to spend further onto the card if the earlier payments have not come. And in any case, after 90 days, it becomes Stage 2, Stage 3. So that is a very different value compared to the overall amount that you look on revolver. So revolver is a very different ballgame all together.
Okay. Then how is the color on the revolver book going forward? We are currently at 24%. we have been around that percentage for a while now. So how do you see this going forward?
So, good point. So on the revolver book, what we have seen is that even when we look on a month-on-month basis, it is stable at around -- there is not much variation. It is hovering at around 24% or so, plus minus 20 bps, 30 bps in that scenario. We have seen -- as the new portfolio also maturing, because as you'll see it was 2020, 2021 good customers maturing. This is a trend, which was shown in slide 15 of 90 plus, but we are seeing revolving behavior in those segments also. And this time the evolving behavior is that the customer would revolve for one or 2 months and then again become a transactor. So -- which is a very good sign. So it is stable at 24% -- at this point 24%. We don't foresee any change at least going further in next 3 to 6 months. And then we will keep you people posted at if we see changes in that behavior on the positive or other side.
We have next question from the line of Prashant Kothari from Pictet.
Yes. Just wanted to understand the lifecycle of the customer. I mean why does it takes so long time to even kind of understand that this was the kind of the bucket of customers, which is going bad like the portfolio, which kind of started in 2019. It's 4 years old and now we are understanding, recognizing, realizing that this was a bit of a weak bucket. Why feedback mechanism is so kind of slow. And secondly if this the case, that will take about 3, 4 years --
Sorry, we couldn't get the last part of your first question.
Yes. So the question is, why is the feedback mechanism so slow? Because what I -- as your feedback would be very fast in the credit card business that you will know which customers are bad or good. And therefore you can take corrective actions much sooner. But why is it so slow? So that's the first part of your question. And second, related to that is, if the feedback mechanism is slow really, then how do we kind of keep the incentives aligned in the business? Because people are getting rewarded maybe on, I don't know maybe on monthly, quarterly. yearly, performance, but if the problem takes whether 3, 4, years, actually to emerge properly, then how is the incentive system lined in internally?
I'll take the first part of the question if I understood correctly. Your question was, why does it take so much time to identify the problem areas and for the actions to be taken, and the results to be seen? If that's the question, that's especially true for the 2019 vintage from the chart that we were just looking at. And I extend -- some of that earlier I will elaborate further. See, until the -- that portfolio is peaking, 2019 between the particular 30th month period, the aberration wasn't there. The aberration was, the way that curve has then started or come down and that will be aberration and that we should have picked up some time in '22, but because we are coming out of the pandemic, remember '21 was still the pandemic year, that is a year in which the second wave happened.So the only part of '22 is then we have potentially found the problem and then acted on it. we didn't -- wasn't so kind at that point of time because the entire portfolio comes to at that point of time. This aberrant behavior between who is apparent, bigger in the year and that when the actual [Technical Difficulty]. So that slight delays on account of coming from the pandemic period to non-pandemic or normal situation. I hope that answers some part of your question.
Yes. And how do you keep the incentives aligned? I mean if it is taking 2, 3 years for us to understand, where the underwriting is right or wrong. How do you keep incentive aligned?
You are asking about the incentive for the salespeople?
Yes, incentive for the salespeople and incentives for the underwriting team that they are taking the right decisions, when they are getting new customers?
Okay. So underwriting teams are not on variable incentives or anything.
So on incentive, the salespeople are on incentive and there are strict metric that we measure their performance on and their payouts are made only after 3 MOB -- 4 MOB performance of the accounts that has been posed by them. The underwriting team is not on incentive, but there is no individual decision that's happening in the underwriting team. There are scorecards and models that are being used for underwriting our customer.
[Operator Instructions] We have our next question from the line of Gao Zhixuan from Schonfeld.
Yes. So just some data keeping questions on this 2019 cohort. Just want to understand there has been more. So can you provide us with number one, what's the revolve rate of that cohort as of this quarter. And number 2, what's the credit cost of that 16% book for this quarter, please?
The second question is you want to know what is the credit cost for this 16% 2019. We didn't get the first question. You want to know as to what is the --
What's the revolver rate. So book number?
Revolving rates in 2019, credit cost in 2019. We do not usually give out revolvers by vintage. I think that's something that we won't be able to share with you.
The second part of the question is, I think you answered earlier as well. The 2019 vintage contributes about 15% of our assets and over 20% of our NPA. As is -- adding about 14 basis points to our credit costs as we speak in the current quarter [Indiscernible] would have been 14%.
Okay. So on the revolve rate on a cohort. Appreciate it. You can share that number. But are there any -- is it higher or lower than the overall book level?
Give us time. We will look at the data and tell you. Just give us time.
Somebody from the Investor Relations team will reach out and share the data and questions with you.
We have our next question from the line of Rahul Jha from Bay Capital.
Yes. My question is, the guidance that you've given on credit cost, is it on the gross basis or net basis?
We are talking about gross. gross basis.
We have our next question from the line of Saurabh from JPMorgan.
Sir, just wanted to know what is the approval rate right now for new credit cards sanctions in the open market book? And what was at let's say, pre-COVID? That's the only question.
So we usually don't give segment wise or open market versus Banca approval rate. Yes, we can tell you that the open market approval rates are higher. They run anywhere between on an average 55% to 65%. They will continue to go in that range. And they are broadly similar to what we had pre-COVID numbers. Okay. Now, you have to also look at this in light of that the sourcing pattern and the way the customer acquisition has completely changed now. Okay. So at the initial stage it fell, after taking some for 4 or 5 filled, from the customer, we are able to either give a soft approval, soft decline or a referred status to the customer. Now after that, once the application is fully filled, then you look at what is the status of the customers. So it is not about -- it becomes a 2 step thing to be able to start digitally to reduce the cost. So approval rate, that can be looked into very different line all together.
Okay. And during 2020 and 2021, fair to say that this would have been maybe lower 30%, 40%?
No, nothing of that sort. It is actually -- these days the filtration, we try to do upfront itself. The modeling status and then we try to do after 4 or 5 filled, if a customer is spread this state or he has a let's say, a stable score already, whole lot of models go upfront itself to be able to see whether we want to go ahead with that customer or not rather than waste our time. There we don't want to bother.
Okay. So your approval rates haven't changed through 2020-2021 even today?
The point I'm trying to make there is that they have changed, how the context and the situation has completely changed. If you look at completely only online, the approval rate would be in the range of let's 10% to 15%.
Other thing to just to supplement. It's Girish. In addition to the bureau scores. I think we also started looking at some alternate data as well, wherever it is available. Maybe because of the partnership or otherwise. Now, at least we started looking at even account aggregators. So that's where I think the percentage is that much, but underlying it has changed for a period of time as compared to pre-COVID period, kind of extra due diligence which we do for some, maybe segment having low score, et cetera, we do that extra due diligence in terms of validating the information or being sure about the profile
We have our next question from the line of Dhaval from DSP.
Yes. Just one clarification. So, you said about 14 basis point is due to the 2019 cohort. Is that correct? So about INR 60 crores.
That's right.
And the other point was just the next quarter, you expect somewhere between 0 to 14 basis points, as the impact and the following quarters, it should be back to BAU, so about 6.2% to 6.3% kind of gross rate cost?
Yes. So, this quarter -- after this quarter, we expect to come back to our target range. That's right. Current quarter. September quarter.
Okay. Right. So my point is that it won't -- 14 basis point is the maximum impact that we can see in the September quarter, is that's the point I'm trying to get.
I think the trajectory will be -- credit cost will only be coming down. In order to reach even the target range, it has to come down in this quarter itself, which we are shared like in month of June, we have seen good rate and again July also it is holding good. So this gives us the confidence to say like a September quarter overall gross rate cost will be lower than June quarter, but reaching the target range will happen only by way OND. So we believe it has peaked in the month of May and it just started to come down.
We have our next question from the line of Krishnan ASV from HDFC Securities.
Yes. My question is partly related to what Prashant asked earlier which was about the feedback mechanism being too long compared to what conventionally we would think given the kind of signs and the kind of analytics that is now available on credit cards. I thought this feedback loop would be shorter. Having said that, if you say that this was indeed not something that you will see. I mean, to one of the earlier questions by Anuj, you said there wasn't too much different about the 2019 cohort in terms of origination. It was a mixed bag, both open market and Banca. So the cohorts that we originated thereafter, did we get them right just by pure accident or I mean, it seems that if there was nothing different about the 2019 cohort compared to the others, and what is it that we got right and how accidental or was that?
2020, if you recall, was the COVID year. And therefore, like every other organizations, we also had tightened up credit norms. And therefore, you will see that '18 and '19 behave almost similarly. '18 peaked a little higher than '19 and is now obviously came up earlier and the percentage contribution was lower as that. But the benefit that we got that the COVID made us all tightened our credit norms. And as a result, you see '20 performed better, learning from '20, fine tune -- and we also fine-tuned our models post the COVID once the market opened up and we also started sourcing from the market. We fined tune the models, taking in the inputs from the COVID period and prior to that as well and that helped us in the quality of the portfolio post that being set up.
So it's all right, Rashmi. I think my limited point was. I mean, given that much of the market has matured towards early warning signals, right? Four years past is kind of far too long for a portfolio to be, I mean, bleeding and the -- and as not knowing about it in terms of why is 2019 cohort behaving very differently. What is it that we did and where is it that you tweaked it tighter? That was the limited point I was trying to get to, given, I mean all those....
It's not that we did not see the EWS. Obviously, we have a strong EWS mechanism and we do work on that but as Shantanu pointed out that the early years on this particular portfolio, it was also colored by the COVID behavior. Remember that we were required to give restructuring to our customers, et cetera. And as a result of that, that behavior in a way -- because of the restructuring and the COVID, it camouflaged weird the behavior of this portfolio. And last year as well, and we called that out in the early -- in the previous earning calls as well, that end of last year -- calendar last year is when we realized that the signals of EWS for this particular portfolio is hitting us, telling us that the behavior is not as per where it should be statistics MOD and that's where we are partly taking actions on it. So through the cycle of the customer, we do keep monitoring the EWS but as if that it got camouflaged under the RBI RE and the COVID period.
We'll move on to the next question from the line of Shubhranshu Mishra from PhillipCapital.
So when we speak of the Tier 3 and other cities, how many total cities are here and do we plan to change these Visa or Mastercard cards with RuPay credit cards which is acceptable on UPI, given the lack of peers infrastructure in these cities? That's question one. Second is, in the open source customers, how many customers already have a credit card? And if we have a similar number for SBI sourced customers? Thanks.
On second one, on the -- so the sourcing strategy of open market is to leverage the existing carded customers and -- so today, primarily we target customers who have been carded by the industry. We do not really disclose or we do not really publish the data around the percentage of carded customers in either of the segments. But primarily, the strategy is to A) use the digital channels to, use the co-brand sourcing channels to provide these customers with an alternate value proposition, which is compelling enough for them to have an alternate card in their wallets. Similarly for the Banca channel, the strategy is to also reach out to some of the non-carded segment because of the relationship with the bank, which makes it compelling for the Bank to have another relationship apart from CAPA or any other relationship that they have. Shantanu?
Yes. On the Tier 3, Tier 4, we are already giving cards there which are RuPay cards. So customers now with the new draft guidelines of RBI, the customer preference will also get taken into consideration and as MD sir mentioned in the opening remarks, that we are already going live with UPI on RuPay cards. We are testing it. We have not opened it for the last ESPs, but for RuPay portfolio, the testing is taking place. Once that happens, we believe that it will get more transactions in Tier 3, Tier 4 towns from these customers.
How many Tier 3 and other cities are we catering to? That question is still unanswered. How many cities are there?
So we grew more than close to around 200-250 cities in overall scenario, okay? So these days, the definition of Tier 1 and Tier 2 is that Tier 1 is a top 10 and Tier 2 is what we look at the state capital. And beyond that we say Tier 3 and Tier 4.
Also a graded approach, so it's not that you fully operational in every city. And we keep revisiting that strategy every now and then, depending on the penetration and the portfolio quality.
We have our next question from the line of Abhishek M and from HSBC.
Yes. Just on this business development incentive income, like last quarter, also a lot of conversation happened and what I want to understand is, when does the measurement start? So, when -- does this start at the beginning of the year? And then, your partners, see how much volume you are generating for them and then they pay you an incentive. So from when does this measurement start?
So on the business development incentives with these partners, some of these deals are long-term deals which are 4, 5 years in period. There are certain set of milestone -- achievement on those milestones, there is a set of income streams, which follow. That is not typically worth. We have usually constructed those deals so that we are able to achieve at least one milestone in a year or so. So that's how it gets constructed. But there are some deals which -- where we are able to get business development incentives for some specific activity to be done, let's say, in 3 months or 6 months or 9 months.
Okay. So the reason for this kind of fluctuation would be that you've got a big milestone payment maybe in the fourth quarter and then you are now getting the usual regular payments, is that correct? And then as you progress through the year, it should take a higher trajectory or it could go lower as well.
So we can't give you estimate around that. Usually what starts to happen is that there is some amount of evening up which happens, because as per India, if you look at a whole period and look at income streams in that way, but there are some opportunistic arrangements for business development that you can get for a period of time, that we also -- we keep on look out for.
We'll move onto the next question from the line of Pankaj Agarwal from Ambit Capital.
So these cohorts originated post CY '20, are they behaving differently in terms of spends and revolve rate as well?
So you are right. The spends that we are getting from these cohorts is higher. And that is what, how you are able to see our average spend per customer, continuing to go up. The revolve rates are lower, but they are still happening. So we are now stabilized at 24% on the weighted average portion. So this is how the behavior is. Higher spend slightly, lower revolve. But we are getting good EMI traction in these cohorts.
And second, this competition in the sector, is it leading to --
Mr. Agarwal, I request you to strictly ask one question.Participants are requested to stick to one question, please. We have our next question from the line of MB Mahesh from Kotak Securities.
Yes. Just one question. This chart on exhibit 15, you do it by number of customers or is it by outstanding loans or by spends? How do you do this?
Which chart are you referring to?
Which chart are you talking about, Mahesh.
This delinquency is ever 90% plus data that we all have been asking, how do you make this chart?
I think this is value-based chart.
It's the value.
It's the value.
How do you do it in value, because the customer has been -- could be a revolver, he could be EMI customer, he could have been -- generally been a normal customer who is paying on time. How exactly have you created this chart?
So Mahesh, that way to look at this chart is, this is ever 90 plus. Okay? So this is -- so a customer, when he goes into 90 -- initially when the customer comes on the books, the way to read it, in the first 6 months, very few customers will go into 90 plus. Okay. for that court. So you will have -- people will come. They spend on the card. They will pay, who will 90 plus. Broadly people who will be wrong selection might go into that area. As the portfolio builds up, people start spending on the cards. Some people will become transactors. Some people will become, as you are saying revolvers. Now, people who are revolvers, they are typically been more than 5% of the outstanding balance. Some people when they don't pay anything on below -- below of the outstanding balance, you hit pause in revolver because they are interest income but they go into 30 plus, 60 plus, 90 plus. When they're going to 90 plus, this is where their job starts.
Okay. Sir, just one question, is there any possibility that we could have also, what is the kind of customer base that was acquired during this period, in a sense that during that period, was the activity levels of origination more towards the below prime segment, which has cost this problem? Or you think that they transferred into below prime eventually after COVID hit them?
So Mahesh, I think as Shanthanu has been saying earlier also in the call, these customers are behaving as -- but if you look at the chart, 2018-2019 broadly look similar. Okay? It looked similar till the time for sales 30 plus and that time COVID was also happening. So even though there were some -- there warning signals, but whether it was because of RBI RE or because of the customer behavior, wherein I think that was straightened out. And it is 6 months back, we were very clear that this is the segment, which is causing us issues. The last call we declared that and the actioning has already been all on these customers and as Sir was saying, I think it is, we believe that it has peaked out in May and it is now on a downward trend. So we should see that in the next quarter. The automatically, the overall thing should start coming down.
We have our next question from the line of Ajit Kumar from Nomura.
So if you can quantify what percentage of your card base are let's say, less than INR 30,000 and whether that proportion has increased or decreased in the last few quarters? That's it.
Can you repeat the question?
What percentage of your card base are having limit, let's say, less than INR 30,000 and whether that proportion has increased or decreased in last few quarters?
So we have not declared the credit limit breakup of our portfolio. I can only tell you that the average credit limit of the customer is more than INR 90,000 for the full portfolio. So that's where the average is. Medians are slightly lower, but that's where they are.
Thank you. Ladies and gentlemen, that was the last question for today. I now hand the conference over to Mr. Rao for closing comments. Over to you, sir.
Thank you, Yeshashvi. Let me thank all the shareholders, investors and business partners for their continued trust and support. I would also like to thank my colleagues at SBI Cards for their unwavering commitment to ensure the company's success. SBI Cards has and will continue to steadily move on the part of the significant profitable growth. While we cannot control the external factors, but we do believe that our strong business model, our agility as a business, and adaptive approach equips us well to keep fueling our growth in the future. Thank you, all.
Thank you. On behalf of SBI Cards and Payment Services Limited, that concludes the conference. Thank you for joining us. And you may now disconnect your lines.