One97 Communications Limited (PAYTM) Earnings Call Transcript & Summary

Thank you for joining, and welcome to Paytm's earnings call to discuss our financial result for the quarter ending on September 30, 2024. We will start our call with Q&A after introduction to the management. [Operator Instructions] From Paytm's management, we have with us Mr. Vijay Shekhar Sharma, Founder and CEO; Mr. Madhur Deora, President and Group CFO; and Mr. Anuj Mittal, SVP, Investor Relations. A few standard announcements before we begin. The information to be presented and discussed here should not be recorded, reproduced or distributed in any manner. Some statements made today may be forward-looking in nature, actual events may differ materially from those anticipated in such forward-looking statements. Finally, this earnings call is scheduled for 60 minutes. A replay of this earnings call and transcript will be made available on the company's website subsequently.

We'll start our Q&A now. [Operator Instructions] First question is from Alok Srivastava from UBS.

I think first question is if Madhur or Vijay, if you could explain this DLG model with an example in terms of how revenue will accrue, how cost will be there, I think it will be helpful for everyone.

Yes. Thank you, Alok. I can begin to answer that. So as you know, in our previous model, we -- without DLG, which is a model that we continue with certain partners, we have a sourcing fee, which we have described earlier, 3.5% to 4%. So that continues as is. We upfront give a DLG to our partners, which obviously is well within the regulatory guidelines. So we give that DLG. We have -- as mentioned in the earnings release, we expensed that entirely currently in the quarter in which it was given. So in the last quarter, we have expensed all the DLG that we have given. And as a result of this, we get higher collection revenue during the life of the loan. The overall take rate net of DLG still works out -- and here, we're talking about merchant loans because we have given DLGs only for merchant loans. The overall take rate during the life of the loan, net of DLG costs still works out to be north of 5%, which was -- which is -- and as you know, earlier we did not have a concept of net of DLG costs. So we had an overall take rate. We think that the overall take rate will still be north of 5% net of DLG cost. So we do expect a significant amount of collection revenue over the life of the loan under the DLG model.

Also, I would go ahead and tell that -- this is Vijay. Thank you all for joining. Also, I'll go ahead and tell that we were reviewing our all regulated businesses, and we were seeing what is a market practice and regulatory sort of guidance on those businesses. And as you are aware, the more or less industry had matured or materialized, had reached towards a DLG-based structure while our commercial model never had any challenge from any audits, et cetera, with our lenders, we'll look at it but we were more about that, how can we make it very much aligned to like what everybody else does the business. So there, the commercial model and viability, we waited 'til the month where we were able to see that we have enough income. So we have reported in the quarterly earnings that we have actually taken the all DLG amount that we paid as a cost in the provisioned cost in the quarter. And I'm very happy to say this, that it does not change our profit guideline. We are going to make larger money, and this is going to be helping us net the DLG cost, if at all any quarter we are assigning in that. And expectedly, every quarter, there will be DLG cost. And I also want to extend next further that DLG costs, as you are aware, are going to be paid back to us, so higher revenue. So in a way, the business does not require additional equity capital or investment because this is in a rotation -- money is in a rotation here.

Okay. Sure. Just a follow-up on this, Madhur. This 5% net that you'll be making, is it over and above the up-front fees that you will make on the loan?

It's the total money that we make. So we make sourcing fee upfront. We have DLG cost upfront, and then we have collection revenue over time. The total of all of that, net of the up-front DLG cost that we talked about, will be north of 5% or is expanding north of 5 percentage.

Sure. And typically, are we talking about a tenure of 12 months or so?

Yes. So average tenure is about 12 months, 12 to -- I mean, most of our loans are 12 to 18 months. There can be some which are slightly longer, some which are slightly shorter. But yes, that is in the ballpark of what we do.

Sure, sure, Madhur. That is helpful. Secondly, on net payment margin, I think my calculation is suggesting some 3.5 basis point NPM. That's roughly a basis point improvement over last quarter. So how has this happened? Has there been some change in mix or something which has driven this improvement?

I won't comment on exactly what the improvement was. But yes, it has gone up, as we have indicated in our earnings release. That has been due to largely better monetization of merchants as well as control over our payment gateway costs. So it has been on both sides.

Okay. And should this be the run rate that we should be looking at in coming quarters?

Yes, we do think that we should be able to manage -- we should be able to maintain these sorts of levels going forward. We do have some quarterly volatility maybe because of festive being slightly better, slightly worse, and so on. And obviously, I'd just like to point out here that all of these numbers are excluding UPI incentive. So including UPI incentive for the year, we expect our payment processing margin to be significantly higher.

Next question is from Aditya Bagdia of Buoyant Capital.

Congratulations on good set of numbers. So just one question. Like you have mentioned in the presentation about increased monetization of the devices which were not in use, so if you could just highlight your strategy on that? And then what would be the rough take rate on those devices right now?

Rough take rate on the devices?

So just to be clear, yes, we are focusing on increased monetization of -- from our devices. The core piece of that, of course, remains subscription revenues. And those subscription revenues has been inching slightly upwards as we get greater active rates and so on. We have talked in the release about reactivations, but also pick up refurbishment and redeployments. So we're quite focused on that. So the core of it is subscription revenue. I mean, we've given a couple of examples of other places where active devices can also give us increased revenue. And there have been a couple of examples of that. But just to be clear, that relates to active devices, not to inactive devices.

Okay. And what would be the subscription revenues for the quarter? Like is there a change if you redeploy an inactive device?

Aditya, interesting thing is that once we pick up a device, that merchant does not remain a merchant anymore with us because we picked up the device. So one number gets reduced there. But if we refab and redeploy the device, we did not incur a CapEx of the size of typical CapEx. We did incur some refab cost which, in turn, signed up a new merchant. So what we've done and we've written in the detail there is that we will continue to pick up or recoup the devices from the market if the merchant is no more using them, refurbish them and redeploy it. So that is how we are trying to say we will more monetize the device. I hope I'm detailing it, and you can ask which part I should detail further please.

Next question will be from Siddharth Gupta followed by Pranav Gundlapalle of Bernstein. Siddharth Gupta, Voyager Capital.

I hope I'm audible. Great set of numbers. I have a couple of questions. Firstly, on the DLG model, if you could elaborate between the difference between the take rate we have on non-DLG loans and on the DLG loans. Secondly, do we plan on extending this DLG model to our other lending partners as well? And if we have an internal gap that we wish to have on these kind of models that we have set around INR 225 cores for this particular lender. And third, I wanted to understand if we have any plans on moving back towards the Wallet business with a tie-up with a potential bank in the future?

Maybe I'll take the first two questions and Vijay can address the Wallet question. So our net take rate for the DLG business, like I mentioned, should be a little bit north of 5% over the life of the loan. So when I'm saying that, that is net of DLG cost. That is not dissimilar to loans which are without DLG. I think the key thing that matters in a loan really is, as you can imagine, interest rates and so on, but also what is the credit loss. And that is where, as we have mentioned, we are seeing improving asset quality ever since we restarted doing the merchant loan business in March and April. So we're seeing improving asset quality. That's very encouraging. Our collection performance is also improving. So that is what affects the net take rate. But compared to the 2 models, which is where your question was, Siddharth, they are quite similar. Of course, in the case of the DLG model, the DLG cost comes upfront. So as this is -- once this is ramped up, then it becomes business as usual. But in the first couple of quarters, you'll see a bit of a drag, right? And obviously, we are having numbers that we have had despite the meaningful amount of DLG that we have given last quarter. On your second question, we are open to doing DLGs with more partners. I don't think we have a strong preference one way or the other. Like Vijay mentioned, this is the emerging market practice, and we are perfectly happy to be in line with what the market practice is. For various reasons, some vendors may not want a DLG, and that is okay as well.

Yes. So it's really interesting to know that not every lender has asked for it, but we also wanted to be particularly looking at the particular type of portfolios, et cetera, et cetera. So it is not a wider practice that we are going to go with. And that is why we've not set a limit or a number here either. Meaning, in other words, we are not saying that it will be on every loan or nor it is going to be with the cap. So that's what we are trying to say.

But then it leaves me open to the question that apart from, say, signaling to the lending partner that we are sort of backing up on the quality of the asset that we are bringing to them, what is our incentive to kind of go ahead, push forward with this model?

I told you a little bit, a while back, this has become an industry practice and regulatory best practice. So our interest was that let's remain in that within the model.

Yes. And the indication that we have also had from partners is that they are -- their appetite for doing -- for amount of business they do under DLG model is somewhat higher. So to your point that, that signal is not just a sort of qualitative thing, it does actually have an impact on how much business we do from their standpoint. And it's also possible that some new lending partners may have a preference for DLG model versus maybe less of a preference of non-DLG model. So our focus is -- listen, we're seeing enormous amount of demand at very good asset qualities, and this is very profitable business for us. So we do want to try to get into win-win partnerships with our lenders. So if existing partners can scale more and new partners, some potential new partners are finding it easier to start with this model and continue with this model, then we're open to it.

Okay. Got it. And maybe on the Wallet bit?

Yes, could you ask the question again, please? Can you ask the question again that -- are you asking us, you want to bring the Wallet back again?

Yes, are we envisaging bringing the Wallet back with another banking partner? Or is it a product that we have put on the shelf for now or then in your foreseeable future?

Because if you know, the Paytm Wallet was operated by Paytm Payments Bank, which is under regulatory supervision right now. We would wait for a clear direction on that side first.

Next question will be from Pranav Gundlapalle of Bernstein followed by Rahul Jain of Dolat Capital.

I'm just going to go back to the question on DLG. While the average take rate would be the same versus the earlier arrangement, would the sensitivity of the revenue to asset quality change with this model versus the prior model?

I think over the last 4 or 5 years, since we have started doing much in cash advance or merchant loan business, we have seen asset quality in a relatively tight range, right? And obviously, we have disclosed that every quarter as you're aware. I think within those ranges, and I would say, even give or take a couple of percentage points, as much as a couple of percentage points on expected credit loss, we don't expect the net take rate to be different or to be different than what I just mentioned earlier, right? So the answer is that in all reasonable scenarios, we don't expect that to be different. In theory, it can be different because it is a little bit more dependent on collection revenue. But as long as our expected credit losses are in that range or even somewhat higher, we don't expect -- the way this models out is that you will get the collection revenue, enough and more collection revenues so that the net take rate will be, like I mentioned, north of 5%.

Understood. So yes, I think in the normal circumstances, it probably will remain the same. I'm just trying to understand, let's say there's 0 credit losses for our portfolio, would we end up with much higher revenue? Similarly, if there is doubling, would we end up with much lower revenue?

So if you had 0 credit losses, we would expect we would have much higher revenues, much higher net take rates. If you had even meaningfully higher ECLs, we would still end up with, through the life of the loan, roughly the same net take rate. Of course, in the theoretical scenario where we breach, like you mentioned 2x, then yes, our collection revenue would be lower. But like I mentioned and as mentioned in the earnings release as well, the reason why we are doing this is to release more capital. Our lenders are very confident of the business. We are very confident of the business. We're seeing improving asset quality, and lenders are committing more capital on the back of it. And the timing -- in addition to the regulatory and the market practice, the timing has been when we feel comfortable that we are not going to have such scenarios.

Understood. No, I'm just trying to understand the benefit for the partners. So that was the basis of the question. The second question is on your cost base. You've seen again a material reduction this quarter versus last. Would it be fair to say this is a new base? Or is there room for further adoption in the near term?

At the moment, we're guiding to this being the new base although we continue to look at our cost base, particularly on people, software and other indirect expenses, which is the largest chunk, about 85% of our other indirect expenses to continue to find optimization opportunities. I am optimistic that we will find such opportunities. Like you may remember last quarter, we had guided to 5% to 7% for the reduction on the back of full quarter impact and so on. We exceeded that on employee costs. So we continue to remain very disciplined on trying to find as many areas as possible. And yes, so I think that is an ongoing exercise to find more and more opportunities, including AI-driven opportunities to find efficiencies.

Pranav, this is Vijay. I'd like to add a couple of points on both line items. Important to note that if you notice our contribution margins, they've gone back to near 55% without any UPI incentive, which is the guidance that we had done with the UPI incentive. And in my belief, this is going to be the new norm. We are very hopeful and sure that it should not cross or go much below than these numbers. And UPI incentive sort of what comes out will become on top of it, number one. Number two, there is one large cost, Pranav. Because I love the way you write those notes and I love to read them, so I want to tell you that there is a cloud cost, pretty much large amount of cloud cost. And we've seen one of our industry peers to make a cloud in a CapEx model instead of an OpEx model. And that's another way to look at the cost structure of that. So I personally would operate that as AI comes further, further, further, it will become -- that we are far more profitable on a point-to-point basis, but we are far more capital efficient also in using the capital the way that it's meant to be. So I personally remain committed that, technology-wise, for a per transaction and for per bp of revenue, we should be lower cost of people, lower cost of machines. And as you are aware, we are talking AI, which is a little bit of cost, but we would remain even further cost efficient on that. The models that we are deploying and the cost that we are incurring them are phenomenally low. I'm very happy to tell you we, in literally 10 months reduced 60% of our manpower cost on support. And that was so good and this is all triggered by [ Klarna's ] blog post that you would have seen out there. And we internally have created AI IVR. You can talk to a machine like a machine talking in a human. So interactive IVR response that used to be front-ending the merchants, now there is a single model that is answering them on a text. And if you want to continue the call on a call, it will continue there, and the agent will also know this. So these are even opportunities to fork out, to be a full-blown independent technology and software businesses. I'm very hopeful, like Madhur is saying, that this is a new norm that we will be further finding it out ways of our cost but not necessarily in people, but on a operating costs, which are beyond people costs. The point is that. And I also believe that our business with this new DLG model that we've done will bring us more capital towards credit disbursement to the merchants, and this is more of a future-forward underwriting help that our lenders have learned over the period and the quality that they have seen. Remember, the quality that they saw in last quarters gave us more revenue is the reason that we earn more credit revenue and more financial services revenue that we've written in the earnings call. So you're talking about higher-margin business, getting more excess of growth, opportunity of growth, while the machine-based leverage is growing this. So I'm personally hopeful. Let's see what goes up.

Next question will be from Rahul Jain of Dolat Capital, followed by Jayant Kharote of Jefferies.

Firstly, on the personal loan distribution business, when you see things turning better and the way we would like to grow in this space? And on the revenue recognition point of view, do you see the takeaway to ideally go up by 50% to 100%?

Sorry. Can you just repeat the second question, Rahul, explain that before I answer both questions?

Yes. I'm saying, in general, for the total business, do we see the take rate going up by 50% to 100% in the lending business.

In the lending business, okay. So on your first question on personal loan distribution, we have done a fairly good job of disbursing about INR 1,600 crores, INR 1,700 crores last quarter. But we do recognize, as we have mentioned in the release, that we have more work to do. And I think we have done a decent job this quarter of adding some new lending partners, and we do hope to add more lending partners in the next quarter as well. I think that is critical for us. That's scaling the lending partners that we have added last quarter. All those lending partners are relatively small on the platform right now as well as adding and scaling new partners. That really is the way we want to grow this business. And I think we've been talking about that for the last 2 quarters. And I think once we are able to do that and hopefully in the next -- when we're talking a quarter from now, we will have more proof points to show to you. We'll be very confident of scaling this business because the market opportunity is massive. It's just that you have to bring in as much supply as you can and as quickly as you can. So that is the key in terms of how does this business scale. There are other factors. There are other tailwinds that could also exist. We are seeing, in some sense, enough demand, right? So the number of financial products that we did last quarter was about 6 lakhs. Personal loan is obviously a subset of that. So there is enough and more demand. I think we -- it is fair to say that in the early days, us and some of our partners have been quite cautious as well. So there is an increasing penetration opportunity. And finally, I would say, and it is important and it is what it is, that we are going through a very cautious phase of the cycle as it should be, both for us and our lending partners. So maybe 6 or 12 or 18 months from now, that sort of overall market backdrop is more supportive. With respect to your second question, we do expect our lending -- or sorry, financial services revenues to continue to scale up. On a Q-on-Q basis, we did have a very good quarter despite -- on the revenue side despite the fact that the volume dispersed and number of FS products did not go up very fast. But we did get better revenues. And this does create a base for us to continue to scale it. I think as we get more volume through our platform on both ML and PL, we should be able to achieve that over the next couple of quarters.

And just lastly, if I could ask one more. On the add initiative of the Soundbox, do we see this could be a meaningful contributor in the near-term basis? And what should be the idea on that?

Rahul, I don't think it will be a meaningful contributor. The intention was to say that we really continue to innovate on the technology and the various aspects of merchant businesses. This is something that FMCG companies asked us and then Meesho asked us, so we thought that we will talk about it. Meaningful is a big number. So I'm going to say that these are one of those 20% experiment that our team continues to do it.

I would add that this channel of advertising is a bit of a differentiator. We would hope that this does scale, but these things -- specifically, this particular channel, any new channel does take a little bit of time to scale. And we also hope that we see knock-on impact of this on our overall advertising business, that we are able to offer a differentiated effectively publishing property to our advertisers. And as a result, they advertise -- they do this, but they also advertise more on our platform.

And we are trying to defend our product actually by creating more ways to earn revenue per product.

Next question will be from Jayant Kharote of Jefferies, followed by Anand Dama of Emkay.

This one is for Madhur. Madhur, I just wanted to understand the accounting for this DLG model. I get it that the net take rate should remain on MLs north of 5%. But the structure and through the P&L, should it be like a gross take rate of 10% minus the FLDG cost and the net take rate? And if that is the case, how many quarters do we take to reach that steady state sort of accounting gross and net numbers?

Yes. So the sourcing fee is taken upfront. The DLG cost is basis an ECL model. Given this is a new product we have taken on basis the ECL model, the entire cost of DLG has been taken upfront. It is in other direct expenses, so above contribution margin, and the collection revenues will come in financial services revenue over time. The time period of that, like we mentioned, that the average tenor of these loans are about 12 to 18 months. So the vast majority of that revenue will come in less than 18 months. Actually, the largest chunk comes in 12 months, but let's say vast majority comes in about 18 months and then there is some tail revenue because you have loans, which may have gone GCL, but you continue to recover those loans even past the full tenure of the loan or even past the last repayment date. So that's sort of a profile. These are not very long dated. But at least for the next couple of quarters, they create a bit of a drag compared to if we had to do these loans without DLG.

Just to clarify, 12 months from now, our gross take rate, if the ECL is holding up, goes to 8% to 10%? And net of FLDG, we will be at 5%? Of course, it can be better if the recoveries are stronger. Is that a correct understanding, 12 months from now?

That's broadly correct.

Great. And lastly, on the costs, again, phenomenal job, guys, I think, in 2 quarters to come down to these levels. How do we go from here? Is this -- now should we expect quarter-on-quarter growth on this especially the non-sales employee cost base? Or is there more headroom?

I think, Jayant, primarily, you should think of if we were to expand marketing costs. I mean that too when UPI, new customer, onboarding and market share cap, et cetera, show up, which I believe that for concentration risk will be done. So in my opinion, once that is taken care of, we would spend money on marketing, and that is the time that you should see this growing. As far as people is concerned, not really. But I do believe that UPI consumer growth offers an incredibly large opportunity for us. As you are aware, this business was earlier not being done by OCL but by our associate PPBL. And thanks to RBI and NPCI, they allowed us to become a TPAP player as OCL, and now we are awaiting new customer addition. And once that is allowed, we would spend on marketing. But albeit only large when we will take care of market share cap also. So you're expecting the increase only at a materially changing market condition, not otherwise.

Great. And if I could just squeeze one last one, is the active sales employees. I see that the base stabilizing around 30,000, 31,000 number. Given that you are doing a lot of redeployment of devices, should we expect this number to grow moderately rather than going back to that 35,000, 40,000 number?

You should expect to -- basically, we are also enhancing the productivity per employee actually behind the scene, let me also say. The AI that we keep singing a song of is also going to help us on per employee productivity in sales field. So while the number would increase slowly, but there will be further productivity enhancement.

Great. And congrats guys for a great set of numbers.

Thank you so much. Really appreciate. Thanks.

Next question from Anand Dama of Emkay, followed by Suresh Ganapathy.

Congrats for a great set of numbers. Is it possible for you to quantify what's the collection revenue that's included in the financial services?

Anand, we don't really break that out historically, and I would be reluctant to sort of put out another data point. But we have mentioned earlier that 3.5%, 4% historically was the sourcing revenue and roughly 1% and maybe in some quarters, slightly more. Let's say, we have said in the past, 1% is the collection revenue. So that's broadly the split on obviously, the non-FLDG model and nondistribution model. So it is not the PL distribution only, which obviously doesn't have the collection revenue. In the last quarter, it was slightly higher in terms of that split because we did a really good job of repayments and collections. And like we have mentioned, just merchants are just using mobile payments more. So that helps the overall performance of the merchant lending business. So it was slightly higher. So that's about as much direction as I could give you.

Sure. And this is primarily on the merchant loan and PL business both, right, put together?

Well, largely, it's merchant loan. Because merchant loans, just to remind you, we are doing on the same model we have been doing over the last 4, 5 years, obviously, now in part with FLDG also. On the personal loan business, vast majority of our business over the last 2, 3 quarters has been distribution only model. So there, there is really no collection arrangement. Our lending partners do their own collection.

And any update on the new loan products that you're talking about earlier on like home loans, mortgages, et cetera?

I think there are a few experiments and integrations going on with a few secured products. I don't think that at this point -- in the financial services business, that is our #1 priority because we see a big penetration opportunity in both merchant lending and personal loan. Like we mentioned earlier, in merchant loan, there's a huge amount of demand, and we just want to unlock more capital including through these DLG arrangements. And on personal loan, the focus is on adding more and more partners. So that is really the core focus of financial services. While we have work going on -- but I think you should really look at that as maybe having some meaningful contribution after the next 2 or 3 quarters as opposed to anytime very soon.

And lastly, is there anything pending from our side to be done for the NPCI to approve the new customer onboarding?

This is Vijay. So as of now, nothing seems like. We are just in a wait state. I wish we can start sooner.

Next question will be from Suresh Ganapathy of Macquarie followed by Nitin Aggarwal from Motilal Oswal.

So Madhur, just a little bit more on this FLDG thing. So if you have about INR 1,600 crores of portfolio, which is under the FLDG, and let's assume you're giving 5%, so you are telling that 80% is given in the form of bank guarantee and there's a cost associated with the bank guarantee? And apart from that, certain ECL cost which has been expensed through the P&L. Is my understanding right?

No, sorry, let me just clarify, Suresh. So we have given -- INR 1,650 crores is the AUM as of September 30. The dispersed amount is slightly higher than that, as you would expect, because some installments have already started -- have been repaid. So the disbursed amount is slightly higher than that. On that amount, we are -- the arrangement is that we give a DLG. The DLG is not 5%. It is meaningfully lower than 5%. 5% is the cap as you obviously know from the regulator. So our DLG is significantly lower than that. The arrangement is that, that DLG, there's no bank guarantee here, although there is flexibility to give it as a bank guarantee under regulation. Our arrangement is that, in this case, we would most likely give this as an FD, which would be lien marked. And all of that cost, 100% of the DLG that we gave, the arrangement is that we'll most likely give that in the form of an FD and that FD would be lien marked. And that entire amount of DLG that we are giving in the form of lien marked FD has been expensed in the last quarter. So in other words, there is no DLG given which is not expensed in the P&L this quarter.

Okay. So just to understand the numbers. I assume -- okay, let's assume it is INR 1,500 crores and the disbursement is higher, 2% suppose which is what you have given. That means you are saying 30% has been -- INR 30 crores has been expensed through the P&L. I'm just telling ballpark, just rough calculation.

That's right. That's right.

Yes. Okay. So INR 30 crores has been fully expensed through the P&L, and it is coming in the other direct expenses, right?

Exactly.

Okay. Let's hypothetically assume that the number is 5%; is the actual experience of that client is 5%. So effectively, we are saying that instead of INR 30 crores, it's closer to say INR 60 crores, INR 70 crores, what happens with the remaining INR 30 crores? Who bears the hit? The company bears the hit or what happens to Paytm? Is there a claw back arrangement because you are only expensed 'til INR 30 crores, right? Just throw the numbers off the air, but just to understand the mechanics at all.

Yes. So let me just simplify the numbers because they're all illustrative anyway. Let's say we do INR 100 crores of disbursal and let's say, we have 3% DLG, okay? So that INR 3 crores will be given in the form of FD. That entire INR 3 crores will be expensed, right? Now let's say the GCL is higher than 3% or higher than INR 3 crores on that book of INR 100 crores. All of that hit is credit loss for the partner, right? So in this case, because we have publicly disclosed it, I can say in this case, where we have given DLG, we're talking about SMFG where we have given a separate disclosure, right? And that is a part of their P&L. Because obviously, we are making interest income as well and on the remaining 95% or 97% of loans, which are good. So they're making interest income on that. Even net of that additional 2% or 3%, whatever we are talking about here, they have very [indiscernible], right? So the partner is -- so it is a DLG, in this case, the first 3%, as we were using in our illustration, the first 3% gets offset against the FD. After that, all the hit is [ ours ]. And then on top of that, depending on the performance of the book, we also make what we expect to be significant amount of collection revenue. But that might be right for [indiscernible].

But if it crosses 3% outline, they give you a collection incentive. Because you have already crossed the FLDG limit, right, and the company is not happy with the number being crossed, 3%. You will still get a collection incentive if the GCLs are higher than 3%?

Yes. So let's extend that same example. So we have INR 100 crores of dispersal. We have 3% DLG. And let's say, the partner has -- let's say, the book eventually ends up at 5% NCL, right? The first INR 3 crores, so the partner now has INR 5 crores of credit cost. They get INR 3 crores from us. And on that performance level, over the life of the loan, Paytm would not only not have any additional cost to the partner, but will also make significant collection revenue from the partner.

Okay, okay, okay. So this is clear. But just one final sorry, because this is an important thing for all of us to understand. Now the RBI is very clearly telling as per the rules and regulations, both synthetic FLDG and nonsynthetic structures like collections will be taken as a part of the overall 5% limit. Is somewhere down the line means saying this collection is not a nonsynthetic FLDG? Because the RBI may view it as a nonsynthetic FLDG, right?

I cannot speculate on that, Suresh, honestly. This has gone through extreme rigor to ensure that this is an alignment with the regulations, alignment with market practices, of course, and to the satisfaction of compliance teams of both sides. So this is -- the collection revenue, the collection arrangement that we have is very much permitted, and it isn't what you may call a synthetic DLG. We are not in the business of giving synthetic DLGs.

Okay. Okay. This is pretty clear.

Please feel free to reach out if there are any other questions. Thank you, Suresh.

Next question is from Nitin Aggarwal, and this will be the last question for today.

So two questions. One is like what do you plan to do with the cash on the balance sheet now that has increased after the sale of significant business? And what is the optimal number you will likely carry forward [ in ] requirements in terms of cash flow?

I think, Nitin, we want to address and visit that question with our Board. I think one of the things that we have decided is we want to be consistently free cash flow positive, which is not that far away for us now before we take that decision to the Board and come up with a sort of a firm guidance -- or not -- I mean firm guidance, as in a directional framework for what do we think of as excess cash, how do we think about returning that cash and, of course, before that, what are the uses of that cash. So yes, we have talked about this. This does come up on nearly every quarterly call, and I do acknowledge that we have INR 10,000 crores of cash. And I don't see any scenario under which we will be able to use up any meaningful percentage of that cash. But I think we have decided that let's get to -- the most important thing is the business should throw off cash, right? And we feel like we're getting close to that. And we'll absolutely address that with our Board and hopefully have a better answer for you once we have crossed that milestone.

Sure, Madhur. And the other question is like if I look at in metros in key cities, there is already a good coverage of like digital payment provisions [indiscernible]. So what is the TAM that you really look at? Is the growth in merchants coming in more from subpopulations or say something like going one level down metros? And both in numbers, I mean, as well as in the value in terms of GMV. What is the TAM that you are really looking at?

Nitin, this is Vijay. And while we all see a significant amount of Paytm and Paytm Soundbox all around us in metro cities, I'm going to surprise you by saying that the penetration scope is nearly double of what it is penetrated as of now in metro cities. I'm talking metro cities only. So we are talking significant large amount of TAM in metro city themselves possible. And the reason that metro cities, I'm talking is because there is a consumer merchant network effect. Because more metro city consumers pay using smartphones, mobile payments, the more number of merchant need Soundbox, number one. Number two, what we are seeing in metro cities and a little bit ahead, let's say, top 10 cities or top 20 cities, is that merchant now does not wait for a larger percentage of QR-based payments, Paytm QR-based percent payments. They are starting to take Soundbox very early in the journey of being onboarded itself. So here, I mean, numbers can be very comfortably -- I mean NPCI gave about 60 million merchant data, although I do not know the uniqueness check, et cetera, on that data. But I can say that there is a market of tens of millions more Soundboxes out there.

And lastly, just as a data point, how many inactive devices do we have? Any rank-up for merchants? How much is that number?

We don't -- so nothing of sort of that we know that this is the inactive merchant. We are trying to go activate the merchant, and we say that device is not working or picking up because of that. So it's an ongoing journey. What I can only tell you one thing here, Nitin, is that we found out that these devices, which we are picking back for the market are more like battery outage or some plastic, et cetera, refurbishments, et cetera, we see, which is not very large. So the keyword is that there, you go with opportunity of picking them back and then installing a refab device back. So merchant becomes active, revenue starts coming back. So that is why you are seeing there is a larger amount of -- larger amount of call out there.

Next question we'll take from Sachin Salgaonkar.

Apologies I signed up late. So if any of these questions is answered, sorry about that in advance. First question, just wanted to understand how fast could we see the wealth insurance business growing? And what are some of the new areas which you guys are exploring in terms of secured lending? And any color you could provide on that would be helpful.

Sachin, first of all, secure lending or the loans which are of low margin, et cetera, we've learned that everything ends up becoming the book for lenders. And if we are assigning secure lending distribution also on the same size of disbursement allocation, we'd rather lose an opportunity of another high better product for the customers. So we are not doing a large amount of secure lending. And that is one of the big factors, by the way, that personal loan disbursements were sort of not in line with what somebody would have expected. So in other words, we are not aggressively going against secured loans, let's say that. And then I personally suggested that this is something that we will look at later, if at all, we need to look at. Then when you talk about wealth and insurance, good thing is that these things, especially wealth market, has headwind -- tailwind where there is a great amount of opportunity happening and opportunity coming in. But when you look at insurance, there is the distribution that we are working on, are they going to become materially important? That is exactly the reason. If you notice, we have started to give number of transacting users or number of financial services customers. Because overall, the business model will look like payment customers and cross-sell to them. And cross-sell per customer will be the revenue. Because disbursements are -- or GWP or whatever we help underwrite -- help distribute won't be the primary number, but the revenue per customer is what we are headed behind the same. So yes, we expect them. We don't expect them in 1 or 2 quarters but do expect them in due course, let's say, second quarter, third quarter onwards showing up as better than today's contributions. And that is why we are trying to say financial services at large matters, not just credit distribution.

Very clear. A quick follow-up out here. So from a loan perspective, is it only the merchant loan book which would materially scale up, let's say, in the foreseeable future?

I think we see opportunities in both merchant loans and personal loans, Sachin. I think the demand side of it, the ability to do very, very sensible business in both of those is enormous. I think the dynamics are slightly different, wherein merchant loans, I think the growth of our devices business, after some interruptions that we had earlier this year, is now creating that ramp-up. On personal loans, I think we did talk about this a little bit earlier, but I'm happy to repeat, that one of the key things that we had talked about over the last 2 quarters is adding new partners. So we're pleased to say that we did add some new partners this quarter, but quite frankly, they're not ramped up yet. And we do have other partners in pipeline as well. So we need to get the new partners as well as potential new partners integrated and ramped up, and that will create the opportunity. There's no dearth of demand. It is just that we have to create more supply on the platform.

Very clear. Second question, I just wanted to understand a bit more on what [ point ] in the steady state EBITDA margin should we expect from the business, particularly given the fact that business model has evolved a bit after some interruptions?

Yes. I don't think we're necessarily giving that guidance right now. We have talked about -- Vijay talked about last quarter about being profitable in -- by having one profitable quarter by the end of the year. And despite the DLG costs that we talked about earlier in the call, we expect to not only meet that guidance but exceed that guidance, so deliver some significant profitability this year. And then with all the tailwinds that we have, especially on merchant payments, merchant lending, PL and the huge improvements that we have had in our cost structure, we expect to get very meaningfully profitable. So we are seeing this as getting to very meaningful profitability relatively soon, becoming a business that is throwing off cash, but specifically EBITDA margin range, I'll not get into.

Okay. And my last question, Madhur, is just on any thoughts on potentially cash returned back to shareholders. To your comments about trying to get into a cash position very soon, there's a guidance of EBITDA breakeven by, let's say, 3Q, 4Q. And potentially, at some point, as and when the PayPay IPO happens, there could be incremental cash which you guys could end up getting. So if we add altogether...

So...

Yes, sorry, go ahead.

Sorry, please finish.

No, sorry. The question was, do we see some kind of a potential cash return to shareholders, maybe not immediately, but at some point in next 12 to 18 months?

Yes. So we -- I think your observation is right that we do have more cash than we need for any organic or inorganic opportunities. And we do expect to get back in a position where we are generating cash. This question was asked about 10 or 15 minutes ago, I believe, from Nitin from Motilal Oswal. What we have said is that we're going to do -- I think we want to get to a position where we are sustainably throwing off cash and then have a very good robust framework with the Board about how do we think about the cash balance, use of money as well as use of excess money. So hopefully, in a couple of quarters, I'll have a better answer to this.

I'll also add on behalf of -- overall our Board discussion that we had, that we need to be fairly long haul in generating free cash before we start to think about returning it to the shareholders. There should be a trigger or a reason for it. And right now, these cash amounts that we are able to get, we are rather going to be focused on operating business and higher margin and more revenue and more profit.

Next question will be from Vijit Jain of Citi. And this will be the last question for today.

Sorry, I joined late so if this is a repeat, my apologies. My first question is so within the merchant loans business going forward, what kind of share do you think DLG will take as a -- in terms of disbursements going forward? Are you looking at a mix in mind? That's my first question.

Vijit, honestly, that's not a metric. We are -- and we have gone through and we'll share the transcript with you, we've gone through the DLG economics in some detail earlier in the call, as you would imagine. But in summary, over the life of the loan, we see this business as very profitable. We see regulatory clarity on this. We see certain lenders having a preference for this. And of course, we see a huge amount of demand. And it's a very -- like I said, this is a very profitable business for us with or without DLGs, right? So when you look at it from that prism, we are perfectly okay for business to be done on DLG model going forward. And I don't think there is a metric that we're trying to manage to that X percent should be DLG and X percent should be non-DLG. And we're going to find out more in the next few quarters, whether it stabilizes to a certain number, in which case, we can start talking about, hey, what do we expect. But we're certainly not managing to a number.

And my next question is going to be, are there any major areas within payments or financial services that you think you could still venture into? I know I've asked this question before. But once you get the RBI payment gateway license, does that open up things like cross-border? Is that meaningful? Is that something that you would like to do? That's my first question. And if you can just comment on if there's a time line on when you get RBI license and also on the NPCI side.

Thank you, Vijit. This is Vijay and I'm taking this question. So number one, yes, that is one of the big reasons that I'm personally -- whenever this discussion of returning capital to the shareholders comes, I'm suggesting that there are tremendous large opportunities for Paytm to grow within the current frame of scope of work. The first and foremost, let's go back to the origination of the movement when Paytm Bank originated. We gave the Wallet to the bank. We transferred the bill payment business to the bank. And in due course, the UPI also was operated by bank. Being at arm's length and being at a different entity, the Payments Bank had its own business priorities. And instead of growing UPI market share, it probably focused on other business models. So consumer side UPI market share for Paytm Payments Bank was something that we all know was not in the range of what other apps were showing. While, when you look at Paytm-operated Wallet or Paytm-operated QR, Paytm-operated different-different services, we were the market leader there very clearly and are the market leader very clearly. Now the consumer base of UPI is an opportunity which is way bigger than what we anticipate in a conversation as of now. Remember, the UPI consumers for Paytm is an open and wide-open opportunity. The product technology, market ownership, customer ownership, everything belongs to Paytm. And I really thank RBI and NPCI for making this possible and partner banks that as Paytm, we are able to get these consumers. So Vijit, think about right now, RBI asked and Paytm bank transferred 130 million customers and plus 200 million handles to OCL. And now once NPCI will give us a confirmation, we will be able to grow. And my friend, I'm telling you, we are not in this to become a mediocre player or remain a fringe player. We are here to solve for the problem of concentration risk or a market share. And we remain that aggressively committing towards consumer payment, like you've seen us committed towards merchant payments. So the very fact that Paytm will have an opportunity to play in the UPI consumer market share will be an extraordinarily large opportunity. As you can guess, once we have the customer on our platform, ownership of our customer on our platform, we will be able to grow tons of cross-sell of financial services to this consumer. And then obviously, there are byproducts like marketing services and advertising, et cetera, that we are talking about. And remember, in UPI ecosystem, when RBI allowed us to become a TPAP player, it very clearly marked us a responsibility to Paytm, that we will be able to potentially solve for concentration risks that the system carries. As an Indian player, as a pioneer of the mobile payment system in India, we remain committed and we remain committed to invest, expand our offering. Remember, as an Indian player, as India's homegrown champion, we are here to invest in payments, and there is a tremendous large amount of opportunity in consumer payments. When you come about the merchant side, as PA license, you should know that online payment innovations are something that we were able to bring QR on a tech stop so that the customers' higher trust, less fraud-prone and superior payments account systems happen. Since 2 years, we have not acquired new consumers, but we built dramatically large amount of technologies including cross-border, multicurrency, including quick commerce requires many other payment products. We are waiting to onboard as many more merchants so that we can onboard newer quality of product and offering to those customers. And I'm very happy to tell you, Vijit, you could figure out yourself, this foreign transfers, including another service, that omnichannel payments with systems are due to be very impactfully showed up in the market. Right now, imagine Tanishq is a customer for us in off-line, let's say. But they are not in online for us. We are not able to build an omnichannel payment. And India has to lead the way in the payment. I do believe there is extraordinary amount of opportunity in payment for us to grow a larger TAM. What we are right now focused, as you could see, is monetization, and we are showing the market that we can very well optimize and monetize. And in due course, we will -- when we get an opportunity, we will expand the TAM of the consumer and merchant both together.

Best of luck with everything and hope you guys have a really good festive Diwali ahead.

Thank you, Vijit and thank you, everybody, for joining us. I wish you all a very happy Diwali festive season, and wish you all -- that the support means a lot to us. We've been here and see you again after Diwali.

With that, we come to an end of this call. A replay of this earnings call and a transcript will be made available on the company website subsequently. Thank you all for joining. You may now disconnect your lines.

Thank you. Bye.