Ladies and gentlemen, good day and welcome to Bajaj Finance Limited Q2 FY 2021 Earnings Conference Call hosted by JM Financial Securities Limited. 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 conference call, please signal an operator by pressing star then zero on your touchtone phone. Please note that this conference is being recorded. I now hand the conference over to Mr. Karan Singh from JM Financial. Thank you, and over to you, Sir.
Thank you. Good evening, everybody, and welcome to Bajaj Finance's earnings call to discuss the second quarter FY 2021 results. To discuss the results we have on the call Mr. Rajeev Jain, who's the Managing Director; Mr. Sandeep Jain, who's the Chief Financial Officer; Mr. Atul Jain, who's CEO, Bajaj Housing Finance; Mr. Anup Saha, who's Deputy Chief Executive Officer, Bajaj Finance; and Mr. Deepak Bagati, who's the Chief Risk Officer. May I request Rajeev Jain to take us through the financial highlights, subsequent to which we can open the floor for Q&A session. Over to you, Sir.
Thank you, Karan. Good evening to all of you. I'll be referring to the investor deck that we uploaded in the investor section of our website. Let me, without wasting time, jump to panel four. Q2, the quarter then went by and our assessment is all about gradual unlocking of the economy. Our focus as a company was all about restarting all our businesses, creating back to growth plans, given that COVID caused it all to go to zero between April and May. Continuous risk modeling to refine loss estimates, augmenting our collections capacity, and begin the implementation of our business transformation, which we talked about in Q1, knowing fully well that we will have a smaller aggregate economy over the next 12 to 18 months. That really, in summary, was what we were up to in Q2 as management. What that led to is AUM grew 1%, INR 137,000 crores.
OPEX to NII, which I'll provide some more color to over the next few slides, came in at 27.8%. PAT, after a lot of one-timers that we took, which I'll cover as well, came in at INR 965 crores, a contraction of 36%. ROE at this point in time is not really relevant, came in at 2.9%. Net NPA, given moratorium, given Supreme Court verdict, not really relevant, came in at 0.37%. I'll throw some more color on it in the next few slides. Let's give some color to the numbers. I'm on panel five. I'll cover one, two, three, seven, and eight points. While overall presentation is 49 pages, I intend to cover a few salient points for you to give you full texture on the quarter that went by.
Very quickly, point number one, restarted origination across all businesses except retail EMI card business, which used to contribute close to 450,000 accounts in a quarter a year ago, and wallet loans business, which used to contribute 250,000 accounts. Both these remain in pause and will remain in pause till January and March, respectively. Adjusted for that, we've now restarted all our businesses to virtually pre-COVID levels. Company overall booked 3.62 million loans against 6.47 million. You can see, half a million or 600,000 came in just from these two businesses. We had an option to restart both these businesses. From a risk standpoint, chose to push them to January and March. At this point in time, given that every month is important, we are fundamentally witnessing month-on-month improvement in volumes across all businesses at this point in time. In general, I would break the quarter into two.
We had a conservative stance on volumes till August, given extended moratorium and absence of updated bureau data. From September, which is from the month of September, the company has started to accelerate volumes across all businesses. We do expect even bureau data to go fully on stream by November, which should further add volume momentum. Please do realize that bureaus are "the radar" on which the consumer credit in any part of the world works. It is a headwind and from an information availability standpoint. Let's talk September. September, clearly, let's break this into three parts. There are a set of consumption businesses that we are in. The urban consumption business came in year-on-year at 72% of monthly volumes. Rural is starting to be back virtually close to 100%, came in at 91% of year-on-year volumes.
Credit card origination at 73%, e-com came in at 75%, auto finance was at 54%. If you look at loan businesses, whether it's personal loans, mortgages, rural personal loans, SME, were on an aggregate basis at 62% of last year's volume. In the current quarter, company acquired 1.2 million new customers. Overall franchise stood at 44.11 million customers. Point number seven, AUM moderated to 1%. We talked about it. It's INR 137,000 crore. We expect now the AUM growth to be at 6%-7%, versus 9%-11% forecast that we had actually given you in Q1. If the momentum by Q4 is stronger, which is possible, there may be a potential upside.
So far, when we look at the loan volumes that we have booked between June, some in June, but July, August, the reads for that has come at this point in time, and some part of September, the numbers are looking better than pre-COVID, but it's a little bit of oxymoron. Once business goes back to 100%, we can safely say risk metrics are same. At this point in time, based on our conservative risk stance of July, August, they ought to be significantly better, which they are at this point in time. Jumping to panel six, point number 10. The management plan at this point in time is that the last full pre-COVID month was February, is to go back to February 2020 volumes as a company on a month-on-month basis by March or April 2021. We do want to enter next year with forgetting about 2020, 2021.
Whether it's in terms of volume balance sheet standpoint or in terms of loan loss standpoint. We should have covered. We will front-load the loan losses, we will back-load the growth, so that net April 2021, we are back to where we were, where we would have been if COVID had not happened in April 2020. That's really the thought process. Back to February 2020 by March, April. Liquidity, clearly carrying a lot of liquidity, given how dark the days were between April, May, June. Company still had close to INR 27,000 crore of cash between SLR securities and liquid mutual fund investments. Total cost of liquidity, carrying cost of liquidity came in at INR 220 crore against INR 4,750 crore that we used to have in, we had in Q2 FY 2019-2020.
Given favorable market conditions, given the wall of liquidity that's sitting in the system, we will start to now dial down the liquidity buffer over the next six months, and expect to revert closer to pre-COVID liquidity by March, April 2021. We also believe that the cost of excess liquidity will normalize exiting Q4. Deposit book continues to grow, came in at INR 21,700 crore, a growth of 23%. The mix has also been shifted. We had provided guidance on that in Q1 as well. 75% is now retail, 25% is wholesale in that, versus 56/44. Point number 14, extremely important. We fundamentally had a choice to make in April when COVID happened, either cut costs significantly or grow volumes over the next six months. We, as you are aware, chose to cut costs. In Q2, despite higher fees and commission costs to the tune of INR 25 crore.
In a way, that line went up by INR 25 crores due to recovery commissions that are fundamentally sitting there. Overall, OPEX is, for Q2, is down INR 224 crores. As a result, NII grew 4%, but OPEX de-grew by 16%. I think it will go a long way in ensuring that we come out of this crisis stronger. Some of these cost cuts are structural, some of them will never come back, and some are transient. The transient cost will come back as we go back to growth, but the structural cost changes will be permanent in nature. If you took this number of INR 225 crores, I would say INR 120 odd crores is not going to come back, and INR 100 crores will come back. Just a ballpark level. Let's spend some time on credit costs, given the large provisions that we continue to take.
Point number 17 is extremely important. Sorry, let me just cover point number 16 as well. We go back to normalization of all our activities as a company, some of the operating expenses actions also we are rolling back. We have reinstated our quarterly incentive plan, some of the things that we'll ease as we move over the next six months, every month, as we get back to normalcy as a company. Credit cost, first line is an important line. Loan loss and provision estimates, as far as we are concerned, are on lifetime loss estimates. They are, in a way, truly ECL loan losses. That's really the way we see it as. This does mean that companies are accounting for additional loan losses, otherwise that may otherwise occur in FY 2022.
If you flow it the way it would flow, I can easily take you to first half of next year or maybe even a little longer. We are quite clear. We front-load the loan losses, absorb the shock, that the company is geared for growth as we get into FY 2022. It is possible that We do expect in the process, if we front-load the losses, that we go back to pre-COVID loan losses of 160 -180 basis points of average assets. If the recoveries are stronger, we fundamentally, based on ECL model undertaking higher, we may experience it is possible that FY 2022 may look lower net loan loss to average assets. It's possible, that's not really what we're banking on. We're saying we go back to normalcy. We took INR 1,370 crore. That's a number which is there.
Total provisioning coverage is now INR 5,099 crore. In general, we are seeing continued improvement in portfolio quality across all lines of businesses. If you draw a parallel, we started from 27% moratorium book, went to 15.7%. Moratorium converted into so-called 30 +, which is really what we stood at 8%. 15.7 in July has flown to 8%. That's really how you should look at the parallel, 27, 15.7, 8. Still a far distance from where we were a year ago. We are at 2.3%. 30 + was 2.3%, it's coming at 8%. But there is a continuous improvement, and if it was to persist, maybe overall loan losses can turn out to be lower. It's possible, but we will take that view as we experience every quarter. Last guidance we had provided was INR 6,000 crore-INR 6,300 crore, in addition to INR 1,150 crore that we've already taken in Q4 last year.
At this point in time, our loan loss models by, in general, by most businesses, if not all, are projecting an improvement to their estimate. We'll continue to roll forward these loss forecasts by each portfolio and hopefully I would say we are 80%-85% done. There is a 15%-20% residual frame on which we'll become fully clear. We are 80% clear. There's a 10%-20% clarity that will fully emerge by end of quarter three. It's not too far away. So far, we have taken in this year INR 3,400 crore. That's point number 22. Against the INR 6,000-INR 6,300 estimate. Residually, we have to take INR 2,600-INR 2,900 crore. In first two quarters, we have taken INR 1,700 crore on an average. If I took this number, we would take INR 1,450-INR 1,500 crore a quarter on the outer side as a company.
We have started to offer RBI's resolution plan. So far in the last seven, 10 days, we have offered resolution plan of INR 252 crores. Of that, INR 214 crores in mortgages and INR 38 crores in consumer. Largely it will be effective or applicable in mortgage business is really what our assessment at this point in time is. Point number 24, from a P&L standpoint, again relevant. We also reversed capitalized interest of INR 142 crores. So far in first half of the year, we've reversed capitalized interest to the tune of INR 361 crores. Gross NPA, Net NPA came in at 103 basis points and 37 basis points based on Supreme Court's interim order of not classifying NPA. Adjusted for that, it would have been 134 to 56 basis points. As I said earlier, these numbers are not really relevant at this point in time.
They'll become much more relevant in Q3 and will become fully relevant by Q4 only for entire financial system. Profitability, as I said, I'll provide some color. Contracted by 35%. PBT came in at INR 1,305 crores on account of massive one-timers. None of them are required to be taken. Let me make that point importantly. INR 1,370 crores. Our run rate as a company was INR 600 crores of loan losses. I can easily provide only INR 600 crores at this point in time. We took INR 1,700 crores. That's INR 1,100 crores of one-timer. INR 173 crores of additional liquidity, that's one-timer. Interest income reversal, we are not required to do, that's INR 142 crores. These numbers, if I add up to INR 1,400 crores. Technically, we could have announced a profit of, how much Sandeep? Add it up.
INR 2,700 crores.
INR 2,700 crores. That would not be demonstrating prudence. We want to front-load losses and back-load income rather than front-load income and back-load losses. That's the picture in general. Capital, we accreted. We ended at 26.6% as a company. Business transformation, that really 70% of the time is going for the management over the last four, five months. We want to become a moment of truth company across all our products and services. Reduce friction, generate velocity. We will deliver what we are saying by June, July 2021. Once done, what does that mean? It's not a technology project. It's a business project. It should mean significantly lower cost and at much higher velocity. That's really what the business transformation, once we deliver by June, July, would deliver. Let's just jump now to 35 very quickly.
We took away this panel in Q1, given the state that things were in. In essence, we said we must publish it, so we're publishing it. Again, with some changes. 44 million franchises. 54% is in general cross-sell franchise. It used to be, a year ago, it was 59%. As clients move into 30 +, you see the drop. This is despite a set of actions that we've taken on, we have various statistical scorecards, things like behavioral scorecards. People who paid during this period. In a way, our B-score models have also gone through change. Net-net, 59% of the franchise, which is so-called the best franchise, which was 56% in Q4, is looking like 54%. We still added 1.2 million new customers in Q2. Let's jump to an important slide, which is panel 46. This is an important slide.
The way you should read it is stage one, two, and three are defined below. Let me just cover that. Stage one is fundamentally defined as clients with no overdue on reporting date. Stage two is defined as clients with one or two installments overdue. Also includes our weak account, internal weak account framework. Weak account clients don't have an overdue, but are added to it, much more relevant for mortgages, and I'll cover that in a moment. Stage three, clients with three and above installments overdue. If you see stage one as of September 30th, INR 125,000 crores of assets are in stage one. Stage two, which is one or two installments overdue, is INR 10,913 crores, and stage three is INR 1,873 crores. If you exactly populated a year ago number, it's strangely coming close. It's INR 128,000 crores was the balance sheet.
Stage two versus INR 11,000 crore was at INR 3,000 crore. Stage three was actually higher. The reason Stage three was higher is because everything is being in a static mode. It's just a point to keep in mind. Against that, if you see stage one, the provisions are INR 1,571 crore. Stage two, against INR 10,900 crore, INR 3,500 crore. Stage three, it's INR 1,100 crore. Versus a 60-basis point standard asset provisioning in a way, we are at 130 basis points as of Q2. Versus stage two, 17.5% that we used to have, we are at 32%. Stage three, where we were 60%, we are at just a tad below 59%. This is really a summarized view. It changes by line. You see a lower stage two PCR because of being secured asset. You see the lowest in commercial lending at 4.5% on stage two.
You see 15.5% in mortgages, and so on and so forth. It's based on our past experience that this PCR has actually been arrived at. We just go back to where we were. Two changes that we have made. Fundamentally, the lifestyle business, we've talked about it in the past, it'll be there in the transcripts. Fundamentally, lifestyle is a small business. We don't want to make a change to it. It's less than INR 400 crore a year. We merged that. That's one change that we have made. The second change that we made is we've now made personal loan together. We were always publishing B2C business. Rural B2C was an aggregate output that we were publishing. Personal loan was broken into two. We have now aggregated it together. That's a second change.
Otherwise, there's no change in any of the panel. That's really the quarter that went by. I think I've covered all the important points. Happy to take questions between me, Sandeep, Atul, Anup.
Thank you very much. We will now begin the question-and-answer session. Anyone who wishes to ask a question, you may press star and one on your touchtone telephone. If you wish to remove yourself from the question queue, you may press star and two. Participants are requested to use handsets while asking the question. Ladies and gentlemen, we'll wait for a moment while the question queue assembles. Anyone who wishes to ask the question, you may press star and one. First question is from the line of Dhaval Gada from DSP Investment Managers. Please go ahead.
Yeah. Hi, Rajeev. Thanks for the opportunity. I had three questions. First was, could you share Bajaj's share in the OEM subvention pool? It seems that we've lost share and part of it might be a conscious strategy. I just want to understand how does this impact Bajaj's position in the market from a medium-term perspective, if there's any material impact to this strategy? That was the first, our share and the impact. The second was, in terms of, you've talked about this transformation journey. I just want to understand couple of important drivers to reach this 105 million customer and prospect customer, from where we are at about 44 million. What would be one or two major transformation changes that we'll be doing to achieve this target? The third one is related to sort of the originations being very low.
I mean, they were still down 40% at the end of September. Our fee income was quite resilient. We made a couple of important product introduction this year, which we talked about in 1Q. I just want to understand how much of the overall fee is sustainable. How does this impact overall business model? Those are the three questions.
Look, fundamentally, I go back to the PPT. It's not like we've been in a monopoly situation. We've been competing with various players in this business for the last many years, with some or two large players in the last three, four years. We used to have 70% of our OE share used to be us. Used to be, that number did not change. Based on COVID, I go back to point number two. In the last three years, if it didn't change, there was no reason for it to change. We took a stand that this is not the moment to grow. We want to grow as a company. We don't want to grow at any cost. That's an important management call that we took, that until moratorium exists, number one.
Number two, until bureaus fully normalize, until the darkest or we can see some light at the end of the tunnel in terms of reduction in bounce rates, number one, improvement in collection efficiencies, number two, we don't want to continue business as usual. These were the three drivers. As a result of which we decided to pull back till August. To give you a direct answer, the number used to be 70% of total OE share used to be with us. Number dropped to 60%-63%. Okay? It will go back in Q3. That's all. 70% of the sale, let me make a last important point, for us as a company comes from our existing customers. It's not like that customer has gone anywhere or the retailer has gone anywhere. Did we take a call?
Let me make a point that in the month of June, July, I would not do 40% of our business used to an EMI card being on 100% financing. For the last 10 years. We took a view, even for EMI card customers, we would do only 70/30. Maximum in some plans is 83/17. These were risk calls that were taken. We've released most of them, if not all of them. There're still some residual pieces left. I would say 95% we are done with it. There is 5%, 7%, that we'll further release as we see early November bounce data. That's question number one. Question number two, drivers of transformation. Look, the transformation frame was created as part of our strategic plan, which we roll every year in November. We were becoming quite clear that there is a degree of linearity in our lines.
If we had to create a non-linear way to grow, we needed something to change structurally. We created a frame called three in one, which fundamentally said in three clicks you should be able to do away, buy a financial service product across loan, insurance, mutual fund, cards for our existing customers. That's really where broking, that's really where what our focus was. We've significantly expanded the scope of that dramatically. It would have been a channel. If COVID had not happened, it would have been a channel for us. June, July, August, the company will work that way over the next three, four years. Let me make that point. On a continuum, it went from four to, I would say, eight, nine. That's really what has happened. It's not that we have thought through this in the last four months.
We significantly expanded the scope of it and the way we'll conduct business over the next few years. New product interest. As we deliver the second point which I made, it will dramatically enhance velocity of cross-sell. Okay? We reach customer when we want to. Customer still is not able to reach us. Do we have the orientation to create products? Do we have the product suite to deliver that? The answer is yes. You could buy a Pocket Insurance today. We have 120 products. Nobody else has it, just as an example. We have 33,000 customers every month in our franchise buy one of the Pocket Insurance products, just as an example. We have the product suite, we have the franchise. As you deliver three in one, we should see significant expansion in pull and generating fee play over the next few years.
Understood. Yeah. Okay. Thanks, [Arjun]. Thanks for this.
Thank you.
Thank you very much. Next question is from the line of Mahrukh Adajania from Elara Securities. Please go ahead.
Yeah. Hello.
Hello.
Yeah. Hi. My question is on flexi loans. What is the outstanding flexi loans at the end of September?
Total outstanding flexi loans is around INR 43,000 crore. I didn't cover that point-on-point number six, on panel number six. We did INR 8,600 crore in Q1. We did another INR 1,750 crore in Q2 as well in conversion. Largely, we will see this number to be INR 500 crores-INR 600 crores as we go ahead from here.
Got it. That is just conversion.
Total number is INR 43,000 crores. As I said, total number across mortgages to personal loans to each one of our portfolios, gold loans, LAS, so on and so forth, put together would be around INR 43,000 crores.
Got it. Just in terms of conversion, any particular segment where the conversion was the highest? Any color on segment-wise conversion?
In general, the take-up rate was between 45% and 50%. There's no specific variance.
Okay. Just one clarification again on the flexi loans, that, say, if a B2B customer takes a flexi loan, then the flexi portion will also be-
We don't offer it in B2B. It's only offered in personal loans, offered in mortgages, offered in gold. Gold loan by nature is a flexi loan. LAS by nature itself is a flexi loan. By nature, itself.
Okay.
So-
It's a consumer loan.
Yeah, exactly.
Okay. Perfect. Thank you.
Thank you.
Thank you very much. Next question is from the line of Ashish Sharma from Enam Asset Management. Please go ahead.
Thanks for the opportunity. Sir Rajeev, just a couple of questions on the asset quality and specifically the provisioning cost. The one clarification, which I think you mentioned that we are preempting little bit of a credit cost for FY 2022. Just wanted to get clarity that in no way that the way we think the credit cost will pan out, that there will be any slippage of FY 2021 credit cost to FY 2022. Basically, it's the reverse, that we are preempting little bit, which actually would have spilt to FY 2022, and we are preempting that.
Answer is, let me simplify the conversation, Ashish. Is it possible that just on a flow basis, as I said in the conversation earlier, can I extend this to September? The answer is yes. Okay. Can the credit cost fundamentally in the current year be lower by INR 1,200 crore-INR 1,500 crore on a flow basis? The answer is yes. Will we do it? The answer is no. That's one part of the conversation. We could easily do the number as, let's say, INR 1,200 crore. It could easily come in at INR 5,000 crore. In first quarter and second quarter, we'll be saying there's still more residual left. We are front-loading it. That's point one. Point two, which is that there'll be nothing left. Just to, for both of us be on the same page, we are not talking INR 100 crore, INR 200 crore.
If we're talking, the answer is fully, there's no residual left. We could always do a little higher in Q1 by INR 100 crores, INR 200 crores. That's what I meant.
Sure.
That also we won't leave it. Am I clear or did I leave?
No. I think that's about it. Second question would be on BHFL specifically. I think in terms of growth outlook there, I think in Q1, we had made a certain statement as to we will have to review the growth outlook there. How do you see the situation? The rates have come off even more from Q1 to Q2 in terms of outlook on growth and profitability for BHFL.
First point, like us, BHFL is also focused on going back to pre-COVID growth levels by March, April 2021. That's point number one. Let's just step back for a moment. Look, BHFL was created intent to create a standalone mortgage business. As we became very clear 2.5 years ago, that mortgage businesses are very different from consumer lending and MSME businesses. BHFL, however, being a two-year-old business, predominantly on the liability side relies on bank funding rather than money market funding. Because it is only two-year-old, it has only two years of financials. As our thought process was, as it completes the current year, it would have completed three years of financials. We will start to get a lot more access to money market funding. That of course, got pushed away by a year anyway, but inherently creates a cost disadvantage.
This business is about process and cost. You got to do business at a much lower cost, which is process, and you got to originate money cheap. The one part of the play, to be fair, it's got nothing to do with the management is, they do have, as management, a disadvantage that only time will fill the gap on. That's really what our thought process is. At this point in time, overall, if I look at them as a company, even for them, in the month of September, they have come back to 75% of pre-COVID volumes. Will they go back in March, April to pre-COVID? Answer is yes. Do they have an inherent disadvantage being a new kid off the block? The answer is yes. How will that get mitigated? Only time will kill it. There's no shortcut to building businesses.
Perfect. I think you've clarified on the liquidity buffer part. Going in the second half we will start-
Yeah.
Reducing the additional liquidity buffer.
Virtually now. Let me make that point. We are technically not going to borrow in both the companies. If we have to dial it down to where I have given guidance for, there'll be very little borrowing that we'll do. If we do any borrowing at all, that'll be in CP. CP for both the companies put together is down to less than INR 2,000 crores. Look at the kind of conservative stance that we took as a company, and I believe rightfully so, given how dark things were looking between April, May, June, July.
Yes, Sir. Perfect. Thank you, Rajeev and all the best for the next quarter. Thank you.
Thank you.
Thank you very much. Next question is from the line of Adarsh from CLSA. Please go ahead.
Sorry. Hi, Adarsh.
Hi, Rajeev. Rajeev, one question on asset quality panel on the loan against
We are losing you. At least I'm losing you.
Adarsh, say your voice is breaking.
Okay. Sorry. Is this better?
Yeah, this is better.
Better. Better now. In the asset quality panel, we see that most segments have seen a 5%-10% or say 6%-10% kind of SMA book. Loan Against Property has held up pretty well. It's broadly where you were in terms of SMA pre-COVID, right? It's quite surprising, or I would say positively surprising. If you can just reflect on why would this have happened? What's led to this being as resilient as this?
Look at the number. Panel 48. A year over, it was 60 basis points. Today, it is 236 basis points. What has happened is that cases have flown forward. You see, refer the stage two, 60 to 236, 51 to seven. Refer the number. Go to the next panel. home loans was 13 basis points, stage two. It's looking at 140 basis points. See, it's all relative from where they were, please understand, each of them are adjusted for margin. It's adjusted for margin conversation. We can't sustain a 236-basis point stage one in LAP business. We can't sustain 140 basis points stage two in home loan business. It's not sustainable. You run a home loan business, or you see in front of you, this is three-year data, two years data. Number runs at these levels.
Adjusted for risk, it's a pretty universal outcome in general so far it seems to us.
I was just saying that when I compared to business loans and everything, LAP held up well because this segment is still business owner, and it looks like it held up at 2.0%, 2.5%. It held up really well. The other way I like to ask this question is, and it relates to any SME kind of business. If you didn't have the flexi product, will the 8% number be materially higher for us, Rajeev, in terms of the SME book? How would you answer that question?
Look, I'll answer it a little differently. You can't model these numbers. We model numbers where 20% and 30% and 40% within. As the world went into lockdown, Fed's 12-year models failed in two weeks.
Okay. If I was to model, that thing will go up from having, we are publishing, let's say, consumer durable data for the last eight years. Number gravitated between 90 and 90 .5 . Eight years we are publishing. We are watching this data for 10. If I modeled it saying it will went to 92, what will happen? I would say let's not do business, because at 92, the business doesn't make money, so that we are all on the same page. We are risk managers. We take risks so that money can be made. You can't model these numbers. These models come out of once in a century crisis. Please understand this. We model for, okay, number went to 98 .5 , goes to 97.5. What will happen? We will say you cut the bottom 10%. At 92%, you have to cut 40% business.
Which you guys have a view on. Why are you not growing? You see, we can't talk from two sides of the conversation. At 92%, I have to cut 40% of the business if I have to ever model the business. Business is not run that way. This is a once in a lifetime crisis, produces this way. The important thing, however, Adarsh, is something else. Why have we not relaunched retail EMI and wallet loan? I have a view that we are in a business of cycles. In financial services businesses, cycles happen. In consumer lending, cycles happen. Some happen in 10, some happen in 12. You can stretch it to 15. Cycles will happen. There'll be one year of lower growth or a disproportionately lower ROE. It's a truth of our business. I've been in it for 27 years to know that.
What triggers it is not known. In that year when that happens, did the business lose three years of earning or a year of earning? That is an important conversation. retail EMI and wallet loans lost for me 2.5 , three years of earning. We have to repivot those businesses. These businesses, I don't have to repivot. That's the lens that we applied for what restarts and what does not restart. There's nothing else, Adarsh. You have to take these hits and move on and build back.
Last question, Rajeev, is you did speak about the cost. Part of it comes back and can you elaborate a little bit on part that doesn't come back, how that changes cost income maybe next year over a two-year period? You've obviously quantified some numbers on what the structural and transient cost numbers are, but how does that really change cost income say over a two-year period?
We are, Adarsh, one of the lowest, anyone OPEX to NII, at 31%. Today's number is anyway transient, so let's say it's at 28%. It'll slowly go back to, let's say, 31% over the next two, three quarters. Fundamentally, as we deliver three in one financial services for us as a company, do I expect the second half of the year could look sub 30%? The answer is yes. As we fully optimize that frame in 2020-2021, 2021, 2022, next years. In 2020-2023, can the number go to 29%? The answer is yes. We are not doing this as a hobby. It is a way we will conduct business. Cost of doing business will go down and the business that we do with customer will go up. Both will create a disproportionately lower OPEX to NII on a low base.
Got it. Which broadly comes back to the point that as you reiterated in the presentation, that COVID, because of changes on borrower level, either less leverage, more leverage, you're not seeing a reason to believe that income or your forecasted ROEs long-term change materially except for this one year of hit back.
Yeah. Answer is yes. Franchise is very large, distribution is very deep. We as management are surely but steadily going to go back to growth. Those are three important dimensions for us to just forget about 2021 as we step out of it.
Thanks. That was it.
Thank you.
Thank you.
Thank you very much. Next question is from the line of Kuntal Shah from Oaklane Capital. Please go ahead.
Hey, Rajeev. Good evening.
Good evening.
The questions I had was, what would be the amount of interest on interest under moratorium should the case be adverse against you? Can you throw some color on two-wheeler and three-wheeler business, which has seen some deteriorating metrics? We understand that the three-wheeler guys with business would be not paying back, but just your texture and view on it. Also, are you considering any additional credit card partnerships with any bank? Can you fund 3x the balance sheet size without banking license? Lastly, the question is, we noticed two things. You've been extremely cautious. Not only have you put brakes on the foot, but you have also engaged emergency brakes, and your numbers of provisions are same size even relative to the other lenders whose balance sheet size is probably 7x larger than yours.
Are you a canary in the coal mine or are you just overcautious? What's your view and what's your team view?
Kuntal, you asked five questions. One was about card, another was about bank, third was? Interest on interest.
Two-wheeler, three-wheeler also.
Two-wheeler, three-wheeler. five was cautious while growing. Yeah, understood. Yeah, emergency brake versus? Emergency brakes and accelerator. Okay. Look, interest on interest, government has given an affidavit to the Supreme Court. They have said that they would pay all financial institutions interest on interest. We are waiting for clarification. Is that part of affidavit that they would bear the cost? It seems the answer is yes. To whom? How? What? I think we have to just wait for clarity. The entire financial system is waiting for that clarity, so we are also waiting for clarity. That's point number one. Bank, at this point in time, we don't have a view on doing a bank. I think as the balance sheet grows; let it grow 2x to 3x. We are all here. We can always make a decision at that point in time. Question number three, card partnership.
The business has now started to grow again. As you said, it's back to 75% of last year's volume at this point in time. I said we have a strategic and important partnership with RBL. I've also said we want to be a dominant card issuer in this country among the top three, four. May require another partnership. At this point in time, given there's a significant dislocation, we are in a pause mode on that. Two-wheeler, three-wheeler, clearly, you have to break two-wheeler and three-wheeler. Clearly, that's the customer who's really at the bottom end of the pyramid across the ecosystem, across all lenders, so on and so forth. It used to be a book which in general would give much higher loan losses, price for risk, but that's really how it used to be. In that, clearly, three-wheeler has performed much worse.
If there are two parts of the book that may go into restructuring because of the nature of the business, one is lot more mortgage customers will go into restructuring, and two, three-wheeler book. I essentially see these two books or these customer segments for different reasons. Reasons not related to, one is a productive asset. As you know, mainly three-wheeler sells as a connecting point. If local trains in Bombay run, autos will run, and so on and so forth. They give you a point to point within a city. Until they restart, we have no option but to provide restructuring option to those customers. Mortgages because as tail risk emerges. If there is a tail risk that emerges, given it's a large part of clients or a customer's monthly outflow, may see restructuring requests. Emergency and acceleration, that's really what COVID has created.
We don't have an option. It's a crisis leadership test. Anybody who's working has to go through this at this point in time. There is no option. I think the management has rallied extremely well. I am quite proud of the way we have rallied. There is a third thing. We also press the clutch, doing transformation. Clutch is transformation, emergency brake is about credit cost and acceleration is about restarting. You're doing all three with two legs, you can imagine. That's really the way it is. I don't think it's set to change. Ajit Vanchal is smiling. I don't think it's set to change, at least for the next four or five months. We have to go through this. That's a response to your five questions. Quickly. Wish you all the best. Thank you.
Thank you.
Thank you very much. Next question is from the line of Mayank Bukrediwala from Franklin Templeton. Please go ahead.
Hey. Hi, Rajeev.
Hi, Mayank.
Thanks for taking my question. I wanted to check about the cross-sell portion of a business. You've indicated many times it's about 60%, 70%. What I wanted to specifically check is how much of that cross-sell happens onto a B2C, that is your personal loans platform, or to your mortgages platform. As in of the 23 million cross-sell customers, how many customers take a personal loan or a mortgage loan in a year from you? The second question is that you've declined your cross-sell franchise a bit. I missed the criteria on how you sort of declined the cross-sell franchise, and if we have close to 8% of our customers in moratorium in September, can that decline potentially be larger? The last question is on your online presence. Where are we in terms of online financing presence and how critical has it become to increase that?
Yeah. online cross-sell and decline be larger. Let me take the easy one. It's possible that the decline may increase. It's very highly likely. As bureau data becomes available, we will run a scrub on these clients. That's very much possible that we may see decline. Could it be 500,000 customers or a million customers? It's very much possible. It's difficult to put a finger on it at this point in time. We have to just wait for bureau normalization. Less of it, Mayank, will come from further flows at our end. More of it will come from bureau scrub because we have two measures. Has he paid me and has he paid the system? Has he paid me or not paid me, that's become largely clear. That I've largely cleansed. How many of them have not paid the system?
75% of our customers have bureau scores. It will become clear by November end. We may see some reduction there. It's very much possible. Cross-sell is a little more complicated answer. Differs by product. Differs by product, differs by risk, stance, and strategy at that point in time. The entire 23 million is available for cards business, mortgage business, personal loan business, insurance business, all businesses. There is an overlay that emerges as a result of risk and as a result of customer wallet. It's many aspects that go into it for determining it. Let me simplify the conversation. Let's just go to out of these 23 million, how many am I ready to give personal loan to? Let's say if you were to ask a question, 10 million customers I'm willing to give a personal loan to. Today, this number was INR 112 lakhs, INR 11.2 million.
We were ready to give personal loan to 11.2 million. Today, we are ready to give to 10 million customers across our urban and rural franchise. To five million people, we are willing to give a personal loan to. I'm just revalidating my number. To five million people, we are willing to give a credit card to. That's really how the stat plays. Right? Online, offline. Online is fundamentally a function of the size of the franchise. We have deep relationships both with Flipkart and Amazon. It's a function of franchise. Franchise converted into EMI card. That's really what it is. The franchise grows, that business would grow. As that business grows, its share with the e-com players would grow. That's why it's linked to the franchise rather than linked to growing e-com or de-growing e-com. If franchise grows, e-com would grow.
If franchise de-grows, whichever way I want to cut it, e-com would de-grow. You've to grow the franchise. We do foresee even in quarter two, where, as I said, July, August, the stance was "Let's just be very conservative or cautious." September was "Let's get into business." We still added 1.2 million customers even in a quarter like that. We will grow franchise, and as a result, e-com will grow.
Got it. Just one very small follow-up to this. You said you've got about INR 1 crore of customers that you want to give a personal loan to. My question is also that to how many of them are you actually able to give that loan to today, and can that number increase?
On a rolling basis annually, 10% land up taking . If I take a 10 million number, pre-COVID, we were doing one million personal loans. ± 100,000.
Sure. It's perfect. Would that be like the peak number at 10%?
No. That number will fundamentally multiply as three in one financial services come. We are not able to reach. 30% customers are DNC. We cannot even reach out to them. Just an example. Little more, 40% is DNC. We can't even reach out to them. There are various layers of this frame, but net, if I simplify it, 23, 10, one at a frame level.
Understood. Perfect. This is very helpful. Thank you so much.
Thank you.
Thank you very much. Next participant is Nishant Chavate from Kotak. Please go ahead.
line.
The line for the participant is on hold. Next question is from Sandeep Bapat. Please go ahead.
Sandeep, we are not able to hear you.
Sandeep, the line is open. Please go ahead.
Yeah. You hear me now?
Hello. Can you hear me now?
We were not able to hear you. Yeah, we can hear you now. Yes.
Just in terms of the initial bounce or zero DPD, any color on where it's starting and then it gets to sort of 8% 30 DPD? Any color on that would be helpful.
The answer is a little convoluted, but please bear with me. Are they coming down? Answer is yes. Are they at pre-COVID levels? They cannot be. That you understand my point, that the portfolio bounce is a function of those who are sitting in stage one, two, three . Until they go into charge off, they'll continue to be banked. Until you clean that balance sheet. That's one part of the conversation. Second is the new customers on top who are much lower risk until they get dropped into it. Two ways it'll go back to pre-COVID levels. The fresh customers who default much lesser in general. Depending on the product, the fresh customers default at much lower rate as they mature into the cycle depending on what type of loan their bounce rate increases. As we are aware, in April and May, nothing happened.
July, August, September, we are at 50% level. The top of the funnel is small. On top of that, there is a stock sitting there which has just exited moratorium and gone into 30 DPD. When both these happen are we seeing that they are at lower than pre-COVID? The answer is yes. Let me give you example. Will you see this 92% number that is sitting here in CD business on panel 47 by March go back to 98? The answer is yes. You will see it back at 97.5, 98%. These are eight-month loans. They would have washed. Top of the funnel would have been filled. In fact, before you see this, by February itself, we will see the current bucket bounce rate of this portfolio start to look similar to pre-COVID.
It's portfolio by portfolio, Sandeep, rather than a general point. Are they going down? Of course, they are going down by every measure. That had to naturally happen any which ways.
Understood. Just one follow-up on that. Pre-COVID used to be around 12%. Would we be north of 25 or around that zone or where would be roughly?
I go back again to the earlier two points. Until the stock sits there and look at this, until these 8% clients sit there, these 8% are anyway going to bounce.
Yeah.
If the new book is at, let's say 6% or 7% or 8%. Sandeep, I can give you a number, but it's not material, it's not relevant. That's all I'm just trying to articulate, nothing else.
Understood. Got it. Thanks.
Thank you very much. Next participant is Gaurav Kochar from Mirae Asset. Please go ahead.
Yeah. Thank you so much, Sir. A few questions from my side. First, on your flexi loans, what would be the average principal moratorium given to the borrower?
There are three kinds of loans we offer. As I said in Q1, you could take it. You could get a term loan, you could get a drop line flexi, or you could take a flexi. Flexi is nothing but like a line. You could use it by drawing down and paying down at a design level. Based on products, some of them come with a first-year interest-free payment only.
Okay.
Two-year interest-free payment. some come with six months. It's based on-
Right.
The feature that is created by the need of the product, nothing else.
Right. Sure. You mentioned around INR 43,000 crore worth of flexi loans outstanding as at September. Is it fair to assume that a large part of this was under principal moratorium in the month of September?
No. Let me make a different point. We blocked INR 5,500 crore, INR 6,000 crore of lines on some of these customers at the peak of COVID. It's a reverse point. We've released them. We have started our release that they were undrawn. We've restored that back to normalcy as we move slowly. This may go through some degree of change even as bureau data emerges. Some of them who may be today unblocked may get blocked. This is a quarterly refresh run that we run. That's not correct.
Okay. Sure. Any rough idea as to what percentage would be under principal moratorium, if you have that top of mind? Any number that you'd like to comment?
I don't have a number like that. Sandeep, do you have a number like that?
No. I think the recent loans in the last probably eight, nine months would be under somewhat of interest servicing.
Yeah, go ahead.
My sense is the number will be less than 25%, but I don't have a number readily available.
That's not a metric we track, so that's why.
Sure. Not a problem. Sir, my second question is on the stage two loans that you have disclosed. If I were to see in last six months between March and August, a lot of these customers were under moratorium. Now between August and September, it's just been one month.
Yes.
Is there a possibility, maybe in a worst-case scenario wherein customers would have enough cash or liquidity to pay off one EMI, but what is the comfort on the sustainability of this repayment trend? The 92% customers who have paid broadly, what is the sustainability of the subsequent EMIs in the month of October, November, December flowing in? Do you see a risk to this 8% going up is my question.
You're asking this from a company which is taking 3x the loan losses. Let me make that point first. I think it's an important point for me to make. The risk models that we have said that we roll do take into account how many times a client went into moratorium, one to all the way to six. How is he performing? How are we seeing their efficiencies and how long are we taking to collect from them, even if they default? All of that by line of business goes into the risk modeling frame. Tail risk. Is it possible in some of the businesses which are actually seen, let me make a point, who have returned to normalcy a lot more quickly, I see a tail risk there. Do I see a tail risk in mortgages?
Do I see a tail risk in some of our salaried personal loan businesses? Do I see a tail risk in professional loans? I see, because they've all largely reverted back to normalcy. It is clear that aggregate economy will be smaller. It is clear that the current year, things will contract anywhere between 10% and 12%. If I trade the tail risk between the hits that we will take and hopefully as the economy recovers from the shock, we see recoveries, net it should adequately cover.
Sure. Yes. That's good to know. My last question was regarding the growth. You mentioned in second half you'll be cautiously growing the book. Maybe you mentioned around 5%-6% of the AUM growth in this year, which implies roughly 5% sequential growth in the next two quarters. Any segments or any customers where you would target this growth? Is it the existing customers or you want to make fresh acquisitions? Any color around where would the growth come from?
It's data dependent frame. It differs product by product. It would also differ based on incoming data. In a way, it's a leap of faith to begin with that we've done business in July, August. We've become a lot more confident in September as we've seen July, August. It is going to be to the earlier respondent that I'm making, we'll continue to use clutch, brake, and accelerator all three at one point in time over the next five, six months to generate growth while ensuring there are no blind spots left.
Sure. All right, Sir. Message received. All the best and thank you so much.
Thank you.
Thank you very much. Next participant is Kuntal Shah from ICICI Securities. Please go ahead.
Yeah, thanks for taking my questions. Three questions. Firstly, in terms of underwriting standards, we would have seen that across the board. When we are seeing that in September we are back to almost 62%, had it been the normal case where it would have been and why would it be lower because of more centering of the norm? Second question is incrementally we are seeing banks tying up with payment solution providers for maybe the EMI on debit cards and all. Is there any risk which we see in any of the maybe the metro markets or so? Third, in terms of fee income, if you can give more granular details that quarter-on-quarter, if you look at it, despite the increase in the volume, it's still flat. No doubt year-on-year the fee income has been quite robust for us.
Maybe we are not seeing that uptick in the volume from sequential basis. What would have led to that, Sir?
Fee, OEM. That was the first question on?
Yeah.
On?
EMI debit card.
No, that is OEM.
No, overall underwriting standard, had it been like we are not [standard it], what level of business we would have been back to?
I understood your underwriting question a little differently. That, let's say, if COVID didn't happen, what would the number look like, right? Let's say, if you took a consumer durable, you're a EMI card customer, in general, your bounce rate would look like 4%. Okay? Is it looking like 2.5%? Answer is yes. As I said earlier, let me just caution. That was on the July, August banking because it's a conservative stance that we've taken. As we release some of this, we will see how it plays out. As I said to the earlier point, we will keep using clutch, brake, and acceleration based on the way we see data. I think, we've always run the company that way. It just has to be a lot more volatile at this point in time for the next three, four months.
OEM, payment solution providers, other lenders, it's a large market. It's a free market. There's place for everybody. They've all been there pre. They are there at the time of COVID, they'll be there post-COVID. We have to have a competitive offering, to continue to have a 70% of OE share, which we don't intend to dilute. Did we dilute in July, August? The answer is yes. Was it a management decision? The answer is yes. If that was a management decision, clearly, as we talk now, that is really where we have held over the last many, many years at 70%, 72% of OE's share. Fees, as I responded to few other people earlier, as we deliver three-in-one financial services, the share of fee as a result of product and customers should grow. It'll be a lot more bought or pull rather than push.
I would be really excited about cross-sell frame and products per customer as three in one financial services gets delivered. Kuntal?
Okay.
Yeah.
Yeah. Maybe in terms of the outlook, where should we see ideally fee to assets or maybe how should we look at it overall in terms of this friction once our three in one model also kicks in? How should we ideally look at it?
That's panel number 13. Are we clear that if you knocked off COVID, in the investor presentation on panel 13, that should we go back to 25%-27% AUM growth, 23%, 24% profit growth, a Gross NPA 140-170 basis points and a 3.5% ROA? The answer is yes. Will we go back to these six guardrails, from a long-term standpoint that we have fundamentally graded? The answer is yes. That's really why we are front-loading loan losses. We are back-loading growth. The objective is to go back to this long-term guidance that we have provided in general to public market investors. We will go back there.
Sure. Okay. Yeah. Thanks a lot, and all the best. Yeah.
Thank you very much. Ladies and gentlemen, that was the last question for today. I will now hand the conference over to Mr. Karan Singh for closing remarks.
Yeah, on behalf of JM Financial, I would like to thank the senior management team of Bajaj Finance for joining us on the call today. Thank you and goodbye.
Thank you. Thank you all for patiently. Thank you.
Thank you very much. On behalf