AU Small Finance Bank Ltd

I now hand the conference over to Mr. Prince Tiwari, Head of Investor Relations. Thank you, and over to you, sir.

Thank you, Sejal, and good evening, everyone. Welcome to AU Bank's earnings call for the second quarter of FY 25, and we thank you all for joining the call today. The format for today's call will be very similar to what we have had in the last few quarters, where we'll start with the opening remarks from the management for the first 10 to 15 minutes. And then we'll follow that up with a Q&A session with the participating analysts and investors for the next 40 to 45 minutes.

To start the call, we'll have our MD and CEO, Sri Sanjay Agarwal, share his thoughts on the overall performance of the bank for the quarter as well as the strategic outlook for the bank.

Besides Sanjay ji, we also have the senior management of the bank on the call today to answer your questions.

For the benefit of everyone, I'd like to request that each member kindly keep the number of questions to 2 so that we can take everyone's question and join back in the queue should we have more questions. You may reach out to the IR team at any time after the call for any data-related questions.

So with that, I'll now request Sanjay ji to kindly start today's call by sharing his thoughts.

Thanks, Prince. Good evening, everyone. I'm sorry my throat is bad, so excuse me for the any inconvenience. But thank you for joining the call today evening. I'll take this opportunity to talk about 4 things; operating environment, our growth and profitability, asset quality and our key building blocks for a sustainable bank.

Let me start with an update on the operating environment. As you will be aware, overall operating environment has remained challenging during the last 6 months with persistent high inflation, a long election cycle, both at the national and state level, intense heatwave in quarter 1 followed by unusual heavy rains in quarter 2. This has adversely affected continuity of policymaking and underlying business activity with the unorganized and informal segments impacted more in quarter 2 than quarter 1.

Around 80% of our book is in this informal self-employed customer segments where our customers are dependent on business earnings rather than the fixed monthly income. Any discontinuity directly impacts their earnings and cash flows. There are some early signs of pickup in economic activity over the last 3 to 4 weeks and we remain optimistic of an improved operating environment in later half year as consumer confidence, rural demand, private investment up again. H2 has always been a seasonally stronger half than H1.

Despite the challenging environment, our performance in quarter 2 has been strong across both growth and profitability. On the deposit side, we had been saying for some time that our franchise is strengthening, and it's giving us the desired results in terms of both growth and cost.

This quarter's strong performance was a reflection of this. I'm very happy to say you that our total deposits crossed the milestone of INR1 lakh crores to stand at ~INR110,000 crores, registering a quarter-to-quarter growth of 12.7%.

Our CASA deposits grew by 11% quarter-to-quarter. All key parameters remain on track with CASA ratio at 32% and CD ratio at 86%. Our cost of funds was stable at around 7.04% compared to 7.03% in quarter 1 FY '25. And we now expect the full year cost of funds to be in the range of 7.10% to 7.15% range rather than 7.20% to 7.25% initially expected.

Our team also has done a lot of work on improving our branch profitability. 43% of branches, which were in existence before December 2022-23 are now profitable in comparison with 25% in December '23. This improvement is driven by increased productivity and focus on overhead cost control.

On the asset side also, we had a robust quarter with our GLP portfolio growing around 5% quarter-to-quarter, 9% on a year to date basis. Our loan portfolio now stands at ~INR105,000 crores, and we remain on course to hit our growth target of 25% for the year despite slowdown in MFI and calibrated approach in credit cards.

All our secured asset verticals, be it wheels, MBL, HL and Commercial Banking are performing well, including gold loans. Disbursement was up by 18% quarter-to-quarter in MBL and 12% in Wheels. And more importantly, 81% of our disbursement was in high ROA assets.

Our yield on assets -- yield on gross advances was broadly stable at 14.4%. We are leveraging Fincare's distribution network and increasing our product coverage in each touch point and to do that, we are starting with wheels in around 250 touch point and MBL in 100-touch point by the end of the financial year, and it's more to coming in years.

In terms of profitability, we had a strong quarter with pre-provision operating profit increasing by 19% quarter-to-quarter and profit after tax increasing by 14% to close to INR571 crores despite headwind on credit costs in unsecured segments. This was driven by all key ROA component. We are able to maintain our yield and cost of money. We had another strong quarter of performance in other income, which increased to 32% of net interest income versus 27% in quarter 1.

We have started to monetize all our investments and our aim would be taking this higher in coming quarters. The strong performance during the quarter was broad-based, driven by all the key components. Example, we are growing our insurance distribution business, which has around 14 partners by increasing penetration in our customer base. Deposit and lending-related fees and charges are increasing with the business volumes.

Credit card interchange and other fees will build up as the business matures. All investment has started being monetized, including early signs of offtake in AD-I business and Wealth Management. Treasury income from interest rate movement, PSLC or capital markets are market

dependent and can impact other income by 5% to 10%. This quarter, strong capital market also aided our other income.

Further, we expect all the fee income component to pick further in quarter 3 and quarter 4. We were able to maintain our operating expense at quarter 1 levels despite higher business volumes.

This has been an area where we have focused a lot in the last 6 months, and we believe some benefits of scale will now gradually come into the play.

Our cost of income ratio has reduced from 61% in quarter 1 to 57% in quarter 2, and I believe this is sustainable. This was driven by higher efficiency moderation in overhead cost and calibration of credit card issuance numbers. We also did 11% higher disbursement in quarter 2 with broadly the same manpower cost. Operating expenses would be seasonally higher in H2, but I expect cost-to-income ratio to be around 60% for the full year.

And now regarding the asset quality. In our secured retail asset, which is 2/3 of our book and consists of Wheel, MBL, HL, and gold, we are comfortable with our asset quality. We remain well collateralized and ultimate losses in our portfolio has been low historically, even though our gross NPA may remain elevated in the interim. The uptick in our GNPA in this asset class during the quarter is driven by seasonality, and weather-induced challenges.

As we did communicated previously, we expect a 3% annualized credit cost through the cycle for MFI, and we are building our provisions accordingly. This cycle has come earlier than anticipated and credit loss would be higher in the short term. We have a strong and diversified MFI franchise led by an experienced team with conservative underwriting policies.

For example, we have one of the lowest average exposure per customer in industry with 26,000 outstanding per loan size. We are the sole lender in 41% of our loan book. Our book is well diversified across 61,000 villages with no district more than 3% and no state having more than 12%.

We have taken some additional corrective actions in terms of strengthening of underwriting norms and collection team too. There are some early signs of recovery in the leading asset quality indicators in November but we expect the credit cost for the full year to remain higher than our expectations.

In our credit card and unsecured lending business, which we are building digitally, which we have taken many learnings from the last 3 years of our data in unsecured lending, and we are implementing these learnings in our digital underwriting on a regular basis. We've also calibrated new card issuance and strengthened our collection teams.

Credit costs in this portfolio in quarter 2 are elevated as per the expected lines and headline numbers is further impacted by lower base effect. We expect credit costs to stay elevated in

second half also. Overall, our credit cost for the first half was around 1.28% on GLP, which is higher than our expectations. We'll see how things shape up in the second half, particularly in the unsecured segment.

As things stand today, we expect credit costs to remain elevated and finish full year around broadly the same level as Half 1 based on the current outlook. There could be a variance of 5, 10 bps, depending on the economic condition in the second half too.

To recap on profitability, we achieved an ROA of 1.7% in quarter 2 and H1. Our underperformance in asset quality was offset by sustained growth in other income and improved cost/income ratio. Our margins were broadly stable. For the full year, we aim to defend ROA of 1.6%. Credit cost -- credit costs in unsecured would be a key monitorable, but I expect us to sustain our momentum in other income and cost-to-income ratio.

As I also mentioned earlier that it takes about 10 years for a bank to develop fully. We continue to work on our foundation to create a forever bank. We have invested in building a full product suite for retail and business banking customers. We have a pan-India distribution network with over 2,400 touch points. We have invested in our tech outlook, both customer-facing and core technology. Our digital bank, AU 0101, continues to gain traction, and we are focused on driving efficiency through our tech investments.

Our leadership team has remained stable, and our governance pillar is working fine. And as you all know, we have recently filed an application with RBI for voluntary transition to Universal Bank and upgrade to Universal Bank license will strengthen our deposit franchise with a better brand acceptance and higher customer perception of safety and trust.

Well, thank you so much for the patience. I'll hand over to Prince for further proceedings. Thank you.

Thank you. Thank you, Sanjay ji. And Sejal, we can now open the call for question and answers.

The first question is from the line of Renish from ICICI Securities.

Congrats on a good set of numbers. Just 2 questions. First, on the margin. So earlier, we were guiding at around 5.7%, 5.8%, and now we are guiding at 6%. So how confident we are about sustaining yields and cost of funds at current level especially if there is no rate cut till March '24, I mean, sorry, March '25.

So Renish, thank you for your kind words. No, if you see us, because we were expecting our cost of money close to 7.20 - 7.25% in the beginning of the year, and we want to guide around now 7.10 - 7.15%, right? So there is a reduction of close to 10 bps or 15 bps in overall cost of money and we are also able to get a better yield on our asset side.

Of course, we are not sure about how we want to build our MFI book as of now. But still, there is good kind of traction in our wheels, MBL, housing, gold and all those kind of book, including commercial banking. So we expect that our NIM should get protected close to 6%. And it's very

early, honestly, if you ask me that because I'm not expecting too good a quarter in terms of GDP quarter 2.

So -- it may be a little more ask, but let's have finger crossed because there may be a rate cut even in December also, right? But I'm expecting a rate cut in maybe in February also. That's my last bet. Overall, there is enough room for us to protect NIM.

And secondly, sir, on this RBI circular on sort of revisiting foreclosure charges on micro and small customers. So what is our early assessment? And because seeing one of the slides, we did mention that we have increased sort of fees on lending and deposits. So if, let's say, this circular has to come with some, I don't know, modification or observations whatever -- what is the impact we should have? Or is there any impact at all?

So you are asking about the foreclosure charges on SME book. So Renish, I would want to wait for the clear guideline because Governor only spoke briefly in his call, -- so because it is on the fixed book or it is on the variable book still to be watched out. So we haven't done any calculation. But if foreclosure charges gets weighed on the variable book, then it won't be a much impact.

The next question is from the line of Kunal Shah from Citigroup.

So first question, when we look at it in terms of the provisioning, so it seems like incremental delta is not much flowing in from MFI, but particularly from credit cards plus PL/BL as well as the secured retail when we look at 1.15% to 1.28% and the awaited credit cost out there and you indicated that secure should come back. Is it like the catch-up in the MFI which will offset the improvement in the secured and commercial banking? And that's the reason we are still continuing with that guidance of 1.28% credit costs. And does it include contingency buffer?

Absolutely, Kunal, you are spot on because quarter 2 -- sorry, second half would always be better for secured assets. That's the trend we have seen over the years. So we are -- because of course, the environment is not that great, honestly, as of now to comment much on that. But I'm still hopeful that for a secured asset, the second half would be much better than the first half.

And that's -- but microfinance events happen dramatically in the last 1 quarter only. So that space is still to be watched out. October has given us little better hope than September or August. So we want to finger cross. So we don't want to overcommit ourselves that our credit cost can be in the range of 1.15 - 1.20% as we promised, right?

So we really want to become -- we want to really play safe there and hope that secured asset will pull back, and we might have extra costs on MFI or maybe credit cards. But still, we want to protect our ROA of 1.6%. The overall idea is to really protect our ROA 1.6%. some figures may not be in our control. So that's the way we want to build ourselves in H2.

And secondly, on LCR, now down to almost 112-odd percent, no doubt, it's indicated that you carry sufficient high-quality liquid asset and the non-SLR investments. But how should we look at it given the draft LCR circular also being there? And how would we tend to improve this LCR ratios going forward here? So what are the initiatives which we would be taking?

Kunal, again, I would say that circular is in WIP. Lot much data is coming from various sources that it may be deferred, it may be put in with a phased manner and all those things. But as of now, anything around 115, we are very comfortable. And if that circular comes in, we will see, we'll go to the board and we'll go to the Risk Committee and figure out that what are the best level for us in that given data, right?

Because circular will also become very stringent. It will cover all the kind of events or circumstances where a bank may be needing the money, right? So I think we haven't discussed much internally till now because, again, there is no clarity on specific about LCR circular. So we want to keep, again, that discussion open. And we'll comment once that circular comes.

The next question is from the line of Nitin Aggarwal from Motilal Oswal Financial Services.

A few questions. One is like around the same on the credit cost, we are guiding again to be a similar number as 1H on credit cost. So is it more out of conservatism or are we expecting 3Q to be like worse off than 2Q because 4Q typically is the strongest for us?

Nitin, quarter 2 was not so great and that remains historically also, right? My worry is it will be more on MFI, the whole challenge on MFI started coming from quarter 2 only, right? As I commented earlier that October has given us some hope because October is better than September, right?

But we really want to see 1 more quarter to really finally comment that where we are heading in MFI, right? And that's a 7% of our book. You know 41% of our customer doesn't take any loans from anybody else. We largely remain very conservative in our underwriting over the year in Fincare, right? Very experienced team.

We have built lot much around our collection efficiency because of learning of AU. So I think we are working as a team there, right? So -- but still, we need one more quarter to really assess the real impact of this whole slippages -- and so that is one thing.

And second, the credit card, where we have taken a very kind of call where we just want to understand more digital credit, right, because anything to be done around manual credit in the credit card or unsecured lending is not viable. So the bank has to learn digital credit, right? So we are putting a lot of controls, a lot of learning in present process.

So there may be a case where our book may not grow and our NPLs in terms of absolute amount can be there, right? So in the percentage terms, it might look more, but we want to be very sure about how we really want to build our credit card from here onwards. So that [two] book is very small, just around 11% to 12% of our overall. It won't affect us much.

But in that sense, and I strongly believe that retail assets will pull back, and of course, commercial banking is in line, right? So -- but I want to really play a conservative rather than more aggressive here because there are little unknowns. We need to see that economy really picking up back because our segment is very I would say, informal segment who do on a daily kind of working to really earn their livelihood, right? So that uptake has to be there in the market

to really give you that comfort that our credit cost might be lower than what we have done in H1.

So -- and the other question is, I think the slippages during the quarter has gone up sharply Q-o- Q. So what is the mix of this? Like if you look at between secured and unsecured assets some color around that.

Deepak Jain this side. So from the unsecured side, the slippages came around 33% out of the total slippage during this quarter. That is from the secured side. The unsecured is mainly constitute of credit card, MFI, and personal loans.

Nitin, Quarter 2 was worse than quarter 1. So you are asking in [725] slippages -- 33% is from unsecured and 67% is from secured -- so that's the way we want to say that the entire segment was effective, right? But that remains the case. It's not we have to overly read this data except the MFI.

Got it. And how will this number be in Q1’FY25, say, what this number be in the first quarter, 33% unsecured contribution now. This was around 25%. 25%. And lastly, like when you look at the exposure to the borrowers more than 5 lenders which you mentioned in the slide, how is that likely to move because the MFI disbursements are not happening and you would have like internally done workings around as to how the repayment trend will be over the year. So how do you expect this exposure to move and so as to assess by when we can expect this credit cycle to end? So some thoughts around that.

So Nitin, this is Rajeev Yadav. So as you stated, we have 8% of our borrowers who have taken loans from more than 5 lenders ( equal to or more than 5). And as you know, the industry guardrails are there on a number of lenders just 4. Fincare guardrails are more conservative than the industry guardrails. So I personally believe this number will gradually decline over the next few months in our portfolio.

In fact, Nitin, Prince, here, if you look at the data point, I think from June to September, it has gone down from 11% to about 8%. So it has been reducing as we speak as well.

The next question is from the line of Pritesh from DAM Capital.

Just a couple of questions. First on your CASA. So that has grown very strong this quarter and especially driven by CA -- so any bulkiness in that CA? Or is it like something positive happening on it?

So no, obviously, Pritesh, as we have been saying for quite some time, there's a lot of work that's going on in the banks behind the scenes in terms of trying to granulize the book, trying to grow much more. And I think from a CASA perspective, this quarter as well as last quarter also, if you saw the SA grew pretty well. Last quarter also, there was a 5% quarter-on-quarter growth.

This quarter, it was 8% quarter-on-quarter growth on SA book.

On CA book, I think we didn't see any growth last quarter because of the base effect from March number. But this quarter, there was a gradual buildup, plus also there are a few segments that we have been now trying to entrench predominantly around capital markets, whatever is happening in the IPO segment. So how do we capture a pie of that particular share? And some of those customers that we have onboarded who invest with us in IPOs and other things. That also is helping that CA business. But maybe I can ask Uttam ji to add more color.

Yes, Pritesh, I'm Uttam here. Just to supplement here, as Prince said, that CA overall growth contributed basis diversified segments. So we started focusing on that. So money has come from our government clients also and we have solution-based approach there. So we garnered from there also. Commercial banking also, we are working hard on business banking, agri banking, all those segments. Capital Markets clients also, we have acquired some. So IPO-linked money has also come.

So I think all the verticals towards CA, which we said earlier, I think the focus on CA that has come. It's not chunky money, as you said, but I think transitional money is there, and some money comes in month end also, that is there, but not that any optical money, can say that.

Sure. Just on another question on MFI. So we've given a good color in our slide, but just wanted to check in terms of geographic perspective, which kind of geographies for us, particularly is seeing some problem. Of course, we've heard the industry north and west of the India, but there has been other geographies as well, which have been causing problems. So anything on that? Any geographic color.

Yes. So the one thing that you would have noticed on the slide that you sort of shared was that we are one of the most diversified platforms, and this has been the norm for Fincare as we built this microfinance portfolio over the last decade and our top 3 states are 35%. And in a pretty much in a credit cycle like this, a very solid diversification across states and across districts is a core fundamental framework of risk management.

Besides, we are a rural finance player, which also helps us in terms of stability of customers and our ability to sort of work with them through any cycle. Now having said that, the more difficult states are somewhat very common which the industry has talked about. So in our portfolio and which is true for all I would assume, it's the Bihar state which is slightly more challenging.

The second more challenging state is Orissa. The good news is that our Bihar state is actually the eighth largest state at about 7.86% of our microfinance portfolio. And Orissa, which I named is way down. It's only INR150 crores, it will be about 2% of our portfolio. So fundamentally, the more difficult states, which have had greater flows in H1 are relatively smaller for us. There are obviously a few more states which are -- which are in the yellow zone. These are in redder zone, but those pretty much can be managed with less credit losses over a point of time.

The next question is from the line of Shailesh from Centrum Broking.

Sir, 2 questions from my side. First is on MFI portfolio. Have we considered or taken any CGFMU cover on that? And if yes, what percentage of portfolio would be covered? And second,

I was looking at the PCR number. So there has been some drop sequentially for last few quarters.

And currently here is 61%. So how should we read that number for the year-end?

So on CGFMU, the answer is still last financial year, we have not taken any cover. We have only started taking the CGFMU cover from this quarter. So the stand-alone Fincare bank, we have not taken any cover specially Yes. So I think on the PCR, if you see including write-offs, it's close to 82% (corrected – 84%) right? And if you go sector by sector, we haven't given you that data, but internally, we track because we are running 5 or 6 different asset classes, right? And largely, we are into secured.

And our net credit loss is not about 20 bps or 25 bps in my secured assets, be it vehicle, be it MBL, be it gold, be it commercial banking over the years. So in my opinion, we adequately covered in the secured assets. And if you go because we haven't shared the data internally, but our PCR on the unsecured segment is close to 90%, including MFI, MFI credit card, personal loans, business loans and all those things.

So in that sense, as we move forward, it will be in calculation done based on our provision policy, right? And so we can't comment that at what level it would be because if the delinquency is decent, then the PCR is less, right? So we are not targeting any number there, but we are very comfortable because overall, it's 82% (Correct – 84%). And as of now, it's 63% (excluding technical write off) on an overall asset basis. So I think it's good to go there, in my opinion.

So just a follow-up on the CGFMU, what percentage of our portfolio is covered under CGFMU? In MFI book or overall?

In MFI - We've just started in Q2, and the number is -- it's negligible, very negligible.

So we have just started, Shailesh. So we didn't use to take it. So you can assume it to be almost nil. We'll update you as we move forward.

So the reason I'm asking is that as I understand, as we increase the disbursement in the segment, the 1% premium would start kicking in, in our P&L, right? That is where it will work?

But we will be charging with the customer, right?

So I think we will do a fair degree of risk-based thinking on the portfolio. So we will not necessarily be thinking of doing it across the book. But again, these are early days, we'll come back on it.

We don't do that. Let's put it this way. We don't do that. We have just started off and that also for a very small segment of customers. This is our internal risk profiling. So honestly, there's nothing more to compare at this stage.

The next question is from the line of Pranuj from J.P. Morgan.

Couple of questions on my side. One is your incremental spreads continue to be better than your bank book spread despite your slowing down the disbursements in MFI and credit cards, it's 15.2% versus 14.4%. So what segments are driving this? And can this continue going forward also for the next couple of quarters?

So Pranuj, I think what we have said that basically, it's a function of 2 things. One is, obviously, after the Fincare acquisition, the MFI happened at a higher rate that resulted in both the things that resulted in a onetime upward movement in Q1 that you saw, which is 15.8%. And this quarter, it went down because the MFI again, disbursements were pretty muted, right?

But having said that, I think the more important aspect there is and what we have written in the presentation as well. that all our core assets, right? If you look at wheels, if you look at MBL business, if you look at housing, everywhere, we have enhanced disbursement yield. I mean we have been doing it since last year as well, but I think it has picked up real pace this particular year.

And in the first half, we have sustained enhancement in yields on wheels upwards of 30 basis points in MBL upwards of 40 basis points and in housing, almost 130 basis point plus, right? So I think all of this is helping to drive that entire disbursement yields.

So it should sustain at higher than at least on the bookings going forward despite...

For this financial year, definitely, that's the plan. And then we see how things unfold.

Understood. Very clear, Prince. And 1 last thing. I see that recently also, I think, again, you guys have rationalized your SA deposit rates. So what -- how would you be moving forward with this thing going ahead as compared to other mid-tier banks? Because I think it was 3% up to INR1 lakh and now is 3% up to INR3 lakhs. So what would be the thought process over here?

Yes. So we keep -- I mean, again, it's a function of market feedback that we get from the ground.

And to that extent, we keep optimizing it. The idea is over a period of time as the brand builds up as we have more customer base, we will narrow down the premium that we offer over the large private sector bank.

Currently, we do it more optically in certain buckets wherever we get a room. And this time, like we did that in the first quarter. And we saw -- we didn't really see too much outflow of deposits. In fact, deposits grew in that particular bucket. So to that extent, we took further steps this quarter and we have now rationalized in up to INR3 lakh as well, taking that number from INR1 lakh to INR3 lakh and you'll see the response. And if the response has been positive then we'll further look at how we want to calibrate it. But it's more a function of how the deposit build out is happening and how we are able to narrow down the premium over a period of time.

The next question is from the line of Nidhesh Jain from Investec. Due to no response from the current participant, the next participant -- the next question is from the line of Rohan M. from Equirus Securities.

And this is on opex. So like in your opening comments, you mentioned that there was 11% higher disbursement sequentially, but employee cost was similar. So just wanted to understand, is there a change in the payout structure or origination mechanism? And also when the cost is contained sequentially or the expectation of an improvement in this -- is there a reduction in spend on tech or branding or any of these kind of one-off nonbusiness linked spends. Some color around that?

And secondly, incremental disbursement spreads are down 80 bps point Q-o-Q. So how should one look at it going ahead into second half? After all this is on account of lower disbursements on MFI, but just your view here? And thirdly, if you can give MFI collection efficiency for the quarter or the overdue book in MFI?

So the collection efficiency for quarter 2 was 98.44%. Quarter 1 was 99.07%. So overall, in H1, we have delivered the current bucket collection efficiency of 98.75%, which we believe is better than pretty much what we are hearing from the industry. Over to you, Sanjay.

Yes. So this is Sanjay this side. So again, in terms of opex, we remain very focused in terms of the productive levels in terms of managing the overall manpower in system, right, and I think there was, I would say, a very unprecedented kind of focus in the last 6 months to really contain the cost. And we haven't stopped any kind of investment, be it tech or be it our digital framework or whatever is necessary for to build and franchise for future, right?

But you will appreciate that we've got around 140-odd branches from Fincare. So we haven't done any kind of branch expansion this year. That is helping us to be at lower cost. We have also have really done a lot many things around the entire hierarchy, where we were building lot much different SBUs. And there, we were building lot much manpower for future -- but we have gone back and restrategized ourselves that if something -- some role is not required. We have actually zeroed that also.

We are also getting our synergy benefit because of Fincare. That is also coming up. And it will come -- it will be more and more in maybe next 18 to 24 months. So that is why we have said that in spite of having around 58% H1 cost-to-income, we might have around 60% on a full year basis because there would be some payout and all those things, which will only come out in the second half -- so -- but I would say the trade has started.

The silver lining is that we are now coming from 64% to maybe 60% this year. And then the scale benefit will start kicking in. So -- but the roadmap to maybe to 55% in maybe next couple of years or maybe 3 to 4 years is now there, right? And so -- and the team has also get matured enough. The automation is playing their role. So the growth has been now really not to that extent that you need to invest more and more. So I think a lot many things is on now on the table and we are executing it to the core, right? So that is on the opex.

Your other question was around, of course, MFI was not that bigger book for us, it was maybe capped at 10%. We are around 7% as of now. And finger-crossed if things get improved, we can quickly build that book back -- but overall, I would say the 14.4% our yield is largely protected because our yield on wheel, yield on MBL, yield on gold, or yield on HL has gone up.

Commercial Banking also has able to price their assets more -- so that is there. And -- but I think more than yield, it is about cost also, right? So I think the overall NIM will be protected. That's the overall sense.

Because if I look at the Slide Number 16, our incremental disbursement deals are down and incremental cost of fund has gone up. 1 bps. No, sorry, you are saying... 20 bps increase.

Yes. So of course, Rohan. So as we said, that's a function of the mix, right? -- as the -- because each of the core businesses, as we have said, has been able to increase the disbursement yields.

Commercial Banking has been able to pass on. And the MFI and credit card, MFI predominantly has been lower in terms of disbursement share, and that's a pretty optical ratio mix, right?

Cost of funds have gone up because in the first quarter, when we spoke, if we recall, we didn't really raise a lot of deposits given that we already had quite sufficient liquidity. And in Q4, if you go, our incremental cost of funds used to be 7.7%. In fact, the incremental cost of fund now is at 7.3%. So we've been able to bring that down. In this quarter, you saw our deposit growth were pretty robust, right? So I think it should sustain here unless there is a change in the mix further in terms of disbursement.

And Rohan, one more thing, that the gross advances, which is 14.4%. So our incremental yield is also more than that. So it's a 15.2%, which is higher than our gross advances yield. So that 14.4% will not come down.

Sure. And on that opex lever that you have given, sir, just on the employee piece you didn't touch upon like disbursements were up. So is it that the incentives are deferred to 3Q? Or what has happened there?

No. So, there was no deferment of any kind of setup. There's no change in the policy, right?

Whatever was policy was being followed earlier, same policy still gets followed. I think there has been a lot of, let's say, a lot of with a very large magnifying glass, I think we are looking at every line item, and we're trying to figure out that where exactly efficiencies and productivity can be drawn upon, which are the costs we can do away with.

Like as Sanjay ji said, some of the replacement hiring that were to be done, probably we figured out that can be managed with -- maybe at a lesser cost, right? So all of those things are on an ongoing basis, we have been working on for the last 6 months. And that's where the cost to income decline that you are seeing is there, and that's why we are pretty confident that we'll be more sustainable from here on.

And you should see full year around 60% and not necessarily 63%, 64% that we saw last year.

The next question is from the line of Nidhesh Jain from Investec.

Firstly, on the credit card business, what is the strategy from here onward? I couldn't understand that we will go more digital -- is it sourcing digitally? Or how are you thinking about trade for card business from here onwards?

Nidhesh, so no, my sense was that we were doing credit card underwriting basis on digital footprints, right? And because if you want to build an unsecured book or credit card book, it's difficult to build a brick-and-motor franchise because of the cost enrollment. So we began with that only, and we want to be on that path.

We have learned hard way in the sense that how good digital credit is and what are the challenges we faced over the years. And -- so there was an elevated credit cost. So we have learned a lot and has put in lot many things now, this is understanding over the years. And now that is why our issuance is not to that extent. So we really want to build more cautiously and with more guardrails around as of now, right?

But there are not many things are coming back in terms of to understand customer like account aggregator can help you have more, I would say, data which is reliable, right? There can be other means which we are able to figure out. So I think there will be -- again, we want to go back to our previous number, but I don't want to put any day or any month or any year on it, right?

Because I would say very -- it's a very delicate product, right? So we need to handle it carefully.

And we haven't built so big that we can't go back and correct ourselves. So both the book, whether be it credit card or unsecured lending. And one thing is very sure, Nidhesh, we need to understand this because if we have a desire to become a big bank, then this portfolio needs to build there, right? And the team is committed, and we are all committed to really learn and do the right thing. So my sense is that you have to give me another maybe 2 quarters to really come out with an explicit plan how and how big we really want to build this book.

Second, on microfinance, in H1, our credit cost was around 3%, right? Yes.

And any guidance on the full year credit cost on microfinance?

A difficult one, Nidhesh. As I told you now, this dramatically happened last quarter only. I would only say you this that October, November is better than September, but very, very early days, just 23, 24 days of October. But my sense is this because Rajeev is also on call. So there will be a little elevation from here.

But I think we predicted 3% cost to be honest, before we acquire this book, right? So as of now, it's in that same bracket, but it may go up, but you know that microfinance gives you 5% ROA, right? It might not be giving us 5% ROA over the year, it might give us maybe 3% ROA, right?

But it is still a far better than any other book because it gives me SMF book also or agri book also, right. So the overall understanding to acquire this business still holds true.

And we are optimistic that because this is not an event risk. This is where lender had become more aggressive over the period, right, and that is getting in our state through all the means.

Rajeev is on the call that you can ask him also that it may become more smoother, faster than what we are thinking, right? So Rajeev, do you want to comment more. Okay. So is it fine, Nidhesh?

The next question is from the line of Abhishek M from HSBC.

So the first one is on cards. If I look at the yields, clearly, Q-o-Q, that has gone up from 13.8% to 14.3%. Since the book is stable, is it mostly because transactor balances, etc. are reducing and the mix is changing more towards revolver EMI. Is that how it is evolving? And since you will take about a couple of quarters to recover from -- or to get a plan in place basically, this book should just keep tightening from here on?

No, no. So, Abhishek, Prince here -- so obviously, we have gone slow, but I think the book is still growing, growing at a slower pace than where it should have been, right, or where it was in the last 2 or 3 years. I think there's a very clear effort that we have been putting in terms of growing our mix of earning customers, right?

You earn on a credit card book through your revolvers as well as through your EMI customers.

You don't want your revolver to go out of a certain range because then there could be an associated credit cost, right? So you want a very optimum mix of people who take EMIs and people who revolve vis-a-vis the people who only transact and don't earn you anything, right?

So I think this last -- if you go back in last December, this number for us, the people who used to make us money, right, which is basically revolvers plus the EMI. That percentage was about close to about 37- 38%, that has now gone to about 44%, right? So there is -- the EMI book has gone even from March, where it was 17- 18% it has gone to about 20% as on September, right?

So there's a very clear effort that's going on in terms of monetizing the customers. And it's also happening because the base is expanding. So we have much more acceptance at a lot more merchant points. So to that extent, we are also becoming relevant for the larger merchants, which probably earlier would not really involve us in their promotions and other schemes.

And one thing was in the last 2 months, 3 months, the industry has also moved towards some interchange charges, which is moving upward in direction, including your revolving book rate of interest is also on the upward direction. So everything is supporting to the increased margins in the credit card.

Right. And given that a large part of your book is greater than 760 bureau score and then nearly 90% is 740 plus, which is a reasonably good score. So where do you see most of the credit cost actually coming from? And since you said that you are learning from your previous underwriting using digital means, can you give some sense of where things went wrong and what you need to correct now?

So I'm Sanjay this side. To be very honest, we sourced around 10 lakh cards. And out of that 10 lakh cards close to 75% cardholders does not use any of the credit. They just use the card. They may not be using the card, but they pay on time, right? It's just 25% people take the credit. Maybe minimum credit days period or maybe some kind of loans and everything, right?

Out of that 25%, only 3% people default. So in that metric sense, I don't think we have done wrong thing, but that 3% people actually uses their limit to the full extent. So the credit cost is around 7%, right? So I think we might have gone over in terms of credit lines. I think that's the first learning we got it that we shouldn't have gone aggressive on the credit line because -- in terms of acquisition, you sometimes want to take a market by giving more or a larger credit line than what the customer deserves. That was the first learning.

So we have started reducing the credit line now based on our data analytics. Second thing, generally, in a digital credit card underwriting model, we don't meet customer right? And customer tends to move from each place, right? So I mean, you need a very strong collection mechanism to really recover money from the customer. So that was missing. So we have built that again. We are now building to that extent that we become a lot, much stronger on the ground.

Third, it may be like the kind of income estimation program which we should run because it's difficult to judge customer income estimation based on as of now. So there was no account aggregator 2 years back, right? So income estimation was done based on credit bureau -- basis on other attributes. So I think there, we also have gone back and have changed the entire data point so that we start assuming customer income estimation to the extent possible, right?

I mean these are 3, 4 things which we have corrected over course. And I believe that the new credit card which we're issuing from the last 6 months have shown us a different data point, right? So that's why this business can give you 5% ROA on the longer run. So -- and I think if India is a data kind of country where data is available, then we should learn this business, right?

So that's the overall from our side, right?

Thank you so much for the granularity, very much appreciated.

The next question is from the line of Dhaval from DSP Asset Managers.

Just 2 questions. First is on the credit cost guidance for the full year. I just wanted to get a clarity around what surprised us from where we started the year. So based on the data that you've shared, it seems like your gross -- so at the start of the year, we are assuming the MFI portfolio to be around 3% at the upper limit, including the contingency provisions. And right now, first half, we are tracking about 3.5%. So that's a 50 bps delta on 7% approximately of the loan book, so 20 bps overall credit cost.

And similarly, on credit card, we expected 6.5%, which is now tracking close to 8.5%. So 4% of the book doing 200 basis points higher, so let's say about 8, 10 basis point delta there. And then the residual book is slightly higher than what you were expecting at the start of the year. Is that broadly how the math shapes up from that 120 credit cost to 150, 160 kind of credit cost for the full year, like the 20 bps delta from from MFI and 10 bps from credit cards and small bits from the residual book. Is that correct ballpark?

So Dhaval, I'm Prince here. So there are 2 things, right, which is what we have given on Slide Number 8. Of course, we securitize some books. So 150, 160 are looking on gross advances. I think we should look at the loan portfolio level, and that's where we had guided that there will

be a certain mix and basis which, as you rightly said, probably we were expecting microfinance to be at 8% of the book and 3% of the credit cost.

Now so one, obviously, there's been a slight shift in the mix as well because the disbursements on some of the businesses have been lagging, right? And that's also the environment. So we want to be more calibrated in growing unsecured segment at this point in time, given that the industry itself is going through a challenge. So that impacts the mix. But more importantly, I think, as Sanjay said on the opening remarks as well and to earlier question, that second quarter has been a bit surprising, both on secured as well as unsecured side, more on unsecured, less unsecured, but definitely some bit on secured as well, right?

Like say, for example, what we have shown in that slide is if we expected a 80 basis points on a secured book, then we got actually hit of 90 basis points in the first half. And part of it is because of the rains and other things that happened during that particular period, right? Second half seasonally remains strong. So what we are saying is there should be a pullback in that business, right?

We are not really sure how the MFI is going to play out. We need some more time, but we said October obviously has been somewhat positive and optimistic for us. We'll have to see how those trends sustain. So on an overall basis, what we are saying is that if you look at the first half, second half credit cost at this point in time, we do believe that we should be able to contain at similar levels.

But you'll have to give us leeway of 10 basis points here and there, subject to changing the mix or any kind of further deterioration in the unsecured portfolio.

So Dhaval. Sanjay this side. So let's take one by one, right? When we met in quarter 1, right, we -- before quarter 1, we expected that our cost of money will be very high, right? But during the 6 months, we figured out that it won't be so high, and we recalibrated our cost guidance there, and it was now 7.25%, and now we are saying 7.10%, right? So that remains strong for us.

In terms of asset growth, asset yield, what we said, we delivered it, right? I think the other income really has made us very strong in terms of our traction, be it insurance, be it third-party distribution well, maybe AD-I, all those things, right? It has remained very strong. It's very sustainable, right? I think if you go back and if you go ahead and see the opex commentary, we were more I would say, we're not expecting this kind of reduction in cost because we have done a lot many things, but this remains beyond our expectation in a positive way.

But when we spoke in quarter 1, we haven't able to figure out that market would dramatically change like this. Quarter 2, to be very honest, was very disturbing in terms of overall economic sentiment in the country. There was an unusual rains. We are coming out from heat in quarter 1.

There was a side effect of election results. The confidence was down. And so in there, the slippages was more than what we anticipated.

But it hasn't gone so bad, honestly, right? So it is maybe what we promised was around 110, 115 is now reading around 130, right? So I believe that -- I still believe that based on the -- of course,

the subject to the whole economic activity coming back in H2, I expect that our retail asset will again perform at the same level with what we have promised, maybe a 5 bps here and there.

But I think the finger crossed is for the microfinance because it just started quarter 2 so we need to see 1 more quarter to really assess the real impact, whether it is going beyond 4% and whether it's going beyond 5%, difficult to comment as of now. But credit card, as you know that, we have on the course of correction, and we are doing it. So there may be a little upside based on the lower book. If it would have been the same kind of run then the cost would be in the same range, so 6.5% right?

So I think this is overall on the credit cost. But I would say that in overall scheme of things, I still believe that ROE will be protected. There is something which has been done so well for us.

There is something which is beyond our control. So we might not be doing so well there. But overall, I expect that 1.6% is still defendable, and we are working for that numbers, right?

Understood. Just one small follow-up. The contingency that we planned for the year, will we still have some more contingency in the rest of the year? Or that may get pushed out to next year? Just a clarification.

Dhaval, we said that, if I go back in my earlier commentary that if the credit cost is less than 3%, we want to buffer it up to 3% because we believe that on a 10-year basis, the MFI will give you a 3% credit cost every year because that was the data shown us by CRISIL at their research.

But if the credit cost in first year itself is 4% or 5%. I know that it may be around 3% next year, and then it will be 2% a year after, right? So we would only build a 3% buffer,if it is below 3%, right? It won't be -- we would not want to build a buffer if it's above 3% because then it's a very normalized credit cost.

Understood. So there is a chance that we utilize the contingency that we've created...

That is very small amount, Dhaval, INR17 crores is a very small amount. We haven't think much about it right. It's good to have, to be honest.

Just one last thing on the opex side. So sir, last year, we had positive surprises. I mean, in the sense that opex was rising and we were continuing to invest this year. On the other hand, that is a line which is sort of coming down more than what generally the expectation was. I mean if you could just give 1 or 2 big examples of what's changed in terms of like you gave -- I mean I think Prince talked about having more focus on each of these. But just 1 or 2 big items, if you could just call out -- and so just to get some comfort around sustainability of this number being around 60% and then lower...

There are 2 things, I would say 2 things: one, productivity of people. We are really worked hard on our people's productivity. And so that is why in spite of -- we have grown ourselves by 11%, but there is no growth in our people's number, right? And second, I would say the synergy because of Fincare. We are not opening any branches. There are a lot of leadership, which has come up. The size is there, there is a commonality in our many back-end functions, right? So there are no people required at Fincare side for many back-end functions. So there are some

synergy which is coming up. And we still -- more and more will come up as we move more and more deep in our journey. These are 2 things.

Thank you. Ladies and gentlemen, you may press star and 1 to ask a question.

Thank you so much. Very happy Diwali from our side.

Thank you. Ladies and gentlemen, we will take that as the last question. I would now like to hand the conference over to Mr. Prince Tiwari, Head of Investor Relations, for closing comments.

Yes. Thank you. Thank you, Sejal, and thank you, everyone, for joining the call today. If you have anything residual, you can all reach out to the IR team, and have a good day and Happy Diwali to all of you from our side and see you soon. Thank you.

On behalf of AU Small Finance Bank, that concludes this conference. Thank you for joining us, and you may now disconnect your lines.