Analyst Meet for Bharat Financial Inclusion India Conference Call

Transcription

1 Analyst Meet for Bharat Financial Inclusion India Conference Call Note: 1. This document has been edited to improve readability. 2. The information contained in event transcripts is a textual representation of the analyst meet and while efforts are made to provide an accurate transcription, there may be material errors, omissions, or inaccuracies in the reporting of the substance of the discussions held in the analyst meet. 3. This document may include forward looking statements, including discussions about the management's plans and objectives and about expected changes in revenues and financial conditions. Forward-looking statements about the financial condition, results of operations, plans and business are subject to various risks and uncertainties that could cause actual results to differ materially from those set forth in this document. You should not construe any of these statements as assurances of financial performance or as promises of particular courses of action. Page 1 of 24

2 Good afternoon. We thank each one of you for being here. Let me introduce the team present here Dilli Raj - President Ashish Damani Chief Financial Officer Ashish Pipaliya EVP Investor Relations & New Initiatives Pratap R Deputy Chief Financial Officer Highlights for the quarter are as follows: 1. Strong Operational Performance led by growth in Customers: Non-AP Borrower base crossed a milestone of 50 Lacs (Increase of 10% QoQ and 38% YoY) Non-AP Portfolio growth of 10% quarter on quarter and 16% on monthly average basis 2. Continued efficiency Cost of Borrowing: Marginal cost for the quarter maintained at sub 10% level. It is higher from the last quarter of 9.3%, predominantly due to the first quarter phenomenon (i.e. post the peak funding received and concomitant reduction in marginal cost of borrowing in fourth quarter of last year). The reduction is pronounced in terms of WAHC (Weighted Average Historical Cost), which reduced to 10.7% in Q1FY17 from 11.9% in Q1FY16. We foresee this to be at 9% for the full financial year. Operating Efficiency: We have seen reduction in Cost to income to 45.7% from 47.5% (QoQ) and our OPEX to GLP came down to 6.3% from 6.5% (QoQ). 3. Profit for the period and MAT credit: Our Operating profit was Rs. 107 crore in Q1FY17 compared to Rs.84 crs in Q4FY16 (i.e. 27% QoQ growth). As communicated earlier, the Page 2 of 24

3 recognition of MAT credit of Rs.32 crs for Q1FY17 takes the PAT to Rs.139 crs. (i.e. 65% QoQ growth). We also had a transitory provision of Rs. 97 crore as of March 2016, which was additionally recognized in Q1FY17 and thus reported PAT was. Rs.236 crs. For FY17 and FY18, the tax outgo (on cash basis) will be restricted to MAT, but we will also be availing MAT on the books. The tax expense in the P&L statement will be function of MAT payout, MAT credit and utilization of carried forward loss. Analysis of Industry s Growth: We know industry growth is engaging the attention of several stakeholders. Here are takeaways on our growth vis-à-vis the Industry s growth: 1. 5 year CAGR for the sector and us are 17% and 13% respectively (Slide no.19): The sector growth (CAGR) for Non-AP markets over last 5 years is around 17% i.e. from Rs.28,300 crs in October 2010 to Rs.68,645 crores as on March 16 (Including Bandhan's loan portfolio). Even for Bharat Financial Inclusion Limited, you may say that we had 84% growth in FY16 i.e. from Rs.4,171 crs to Rs.7,677 crs. But, again if you see the growth rate (CAGR) over 5 year period, it was 13% i.e. from Rs.3,942 crs in September 2010 to Rs.7677 crs in March Consolidation in the sector: There has also been re-distribution of market share. We had number of MFIs (more than 300) but post the AP MFI crisis and RBI s regulatory intervention, there has been a strong consolidation leading to the current top ten MFIs accounting for 80% of market. The reasons for consolidation include few MFIs not able to sustain the supply side shock and many other MFIs could not keep up with enhanced supervisory standard of RBI post the Malegam Committee recommendations. Few of our peers who had a Page 3 of 24

4 large AP market share, also had a good non-ap market share and they had to give away the market share in Non-AP markets to new wave of MFIs during that crisis. Hence, this is more of each division of market share. A particular player may have grown at 100% to 200% growth, but on an overall basis, the five and half year CAGR is 17% for the sector and 13% for BFIL. 3. Our Growth pattern vis-a-vis Industry (Slide no ): The growth pattern for Bharat Financial Inclusion is totally incongruent to the sector. To start with, you will see on slide no. 20 that in six out of top ten states, our growth is lower than the sector average. And even if you look at it slide 21, if we do a net contribution analysis of states for the growth, three examples I would quickly sight: Tamil Nadu has contributed to 15% of incremental growth for the whole sector and we do not even have a single loan production center in that state. Orissa contributed only 6% to the net contribution of the growth for the sector and that state s contribution to our growth is 19%. Bihar contributed 6% growth towards the whole sector and 11% for us. Hence, the important takeaway is that although our growth rate is on par with that of the sector, but we are following a different growth pattern. Thus, we are consciously avoiding some of the red oceans and focusing more on blue oceans. 4. Industry growth skewed towards urban, whereas we remain rural focused (Slide no.22) There is always this urban versus rural divide. Over the last three years, the urban contribution for the sector has increased from 33% in 2013 to ~60% in But, we still have 75% of our centers in rural areas. Page 4 of 24

5 5. Our borrower growth exceeds ticket size growth for all periods (Slide no.23) If you do an analysis of drivers of contribution to our growth, the ticket size growth has played a limited role with CAGR of 10% over last 3 years. But, the borrower growth has contributed more than ticket size across all time periods and the CGAR for the same was 21% for last 3 years. Even, if you look at last quarter s AUM growth of 10% QoQ, that completely mirror images the growth in the borrower base i.e. 46 lacs in Q4FY16 to 51 lacs in Q1FY Our loan offtake is lowest among the Top 5 MFIs (Slide no.24) If there is a concern about higher growth, it has to be looked through the prism of leverage of the customers. I do not think analysts doubt risk management capability of us or our peers or ability to digest the credit risk. But, the fact that as we deal with the sensitive segment of the society, over leveraging of the customers is a key concern. On this front, our ticket size is lowest and loan outstanding per borrower is the second lowest among the top 5 companies. As we articulated earlier, we do not play ticket size as a growth driver. There is a 10% growth (CAGR) over three years (FY14-16) and even this year, we could look at 15% annualized increase in the loan outstanding per borrower, but that is not a conscious driver of growth. Assessment of Regulatory standards: There is a lot of discussion whether there would be changes in the regulatory regime, as some of our peers transform into a small finance bank. This is really the prerogative of the regulator, but our limited view is as follows: RBI pursues single set of regulation for a set of players under single form. Hence, we don t think NBFC-MFI guidelines will apply to the Page 5 of 24

6 Bharat Financial Inclusion Limited small finance banks, because they have been subjected to the banking guidelines i.e. CRR, SLR, Capital adequacy, PSL etc. But, as this topic is engaging the attention of analysts and others (including regulators), the regulator would find a way to avoid any irrational exuberance or exploitation by both SFBs and NBFC-MFIs. The key discussion point in the recent period has been whether SFBs can be the 3 rd lender as NBFC-MFIs regulations of not more than 2 MFIs per borrower doesn t apply to SFBs. There could be some sort of tension in that transition period of three to six months, but thereafter I think the conventional credit risk management will prevail. As these companies convert into SFBs, there will be close monitoring by the regulator; thus without changing the regulatory template or superimposing the NBFC MFI guidelines on SFB, the overall exuberance will be tempered. After initial transition pain of three months, things would settle down. All entities whether a MFI, SFB or Bank will be reporting to credit bureau and you can take a conscious call on how many lenders and customer leverage, before you lend. Hence, I am very confident that with RBI s close scrutiny and with all serious and responsible players, the conventional credit management will prevail. Interest rate trend: In terms of interest rate, we have reduced the interest rate by 3.8% in the last year and have come down to 19.75% and thus, continue to the lowest interest rate MFI in the world and India too. Our operating spread cap for the FY16 was 9.6%; thus, we are well below that mandated 10%. There is no pressure or requirement for us to reduce the interest rate further, unless there is a huge reduction in the anchor rates, which gets mirrored into our cost of borrowing. Page 6 of 24

7 Guidance for FY17: In Q4FY16, we have clearly communicated that guidance of Rs.450 crs profit for FY17 is without taking MAT credit into account. Now, we availed the transitory provision credit of Rs.97 crs as on March 16, Rs.32 crs of MAT credit for Q1FY17, and you can estimate the MAT credit over the next three quarters. Hence, the numbers can be easily readjusted by you and there is no need for us to revise the guidance. Now, with that I will throw it open for questions. Andrey Purshottam: Sometimes ability to pay by a borrower can be a problem in the case of an event risk such as a cattle disease, flood, drought etc. I was just wondering that by now we may be able to model and asses this risk and are able to quantify; Can you share your insights on this? No, absolutely the event risk (or what we call as external events) is something which we have been alluding to whenever there is a question about the 1% credit cost we model in. We always make it clear that it is without providing for external events. In terms of getting some hard data around it, there have been several incidents over the last 10 years in the Indian market. There was an issue in Kolar district with concentration of economic activities into a particular activity i.e. silkworm farming. The floods have been happening whether it is Kurnool or Kosi. The answer to above is to keep the ticket size lower and you emphasize on the income generation. There could be some consumption smoothening because cash is fungible at the end of the day, but if large part of it is for income generation (say 75%), people will come back on the feet and start repaying. Page 7 of 24

8 Further recent example is of floods in Tamil Nadu for few MFIs (excluding us as we don t have exposure in Tamil Nadu). There were floods in five large districts of Tamil Nadu, in such circumstances, we announce a moratorium of four to six to eight weeks, depending upon the situation and the local management is fully empowered to assess the situation. Often times, we see that people quickly come back and actually start repaying from the second or third week. In Chennai floods, the eight weeks moratorium was given, it was not that everybody would come back and paid on third week, but they started paying from third week and by the eighth week it was back to that 99% collection efficiency. The painful truth is that as the economic activities of our members are involved in catering to things which are essential to everyday living, If there is flood, then they go to the so-called rescue center (community school etc.) and after 2-3 days they have to vacate and with a situation of nowhere to go back, they will come back to start the economic activities. We did not had the mad cow disease or cattle related risk in a big way playing out in India. We did look at insurance coverage for the same, but that is not available. The real answer to your question is, apart from following the Grameen model, avoiding the concentration risk by having Pan-India presence and being present in 100,000 villages and towns. There was so much concern about monsoon deficit over the last one and half years. But, when we analyzed data of top 100 rain deficit districts, our exposure in those 100 districts was only 10%. Hence, if you are talking about a big issue like monsoon deficit for a rural player like us, that is just 10% of our portfolio. Hence, avoiding concentration risk helps us to contain this. Andrey Purshottam: We don t have statistical data on hand to try and quantify this event? Page 8 of 24

9 No, lot of that is harnessing on local knowledge base, I will give an example. Data is available based on the events to estimate the risk. But, the very relevant risk management practice here is harnessing the local knowledge. Based on my personal experience, few years back, when we were talking about branch expansion, a very young guy said not to open a branch at a particular location as there will be Koshi flood. I thought he resisted expanding branches. But, when the Koshi flood happened and I checked back my notes, I found it was exactly in the places where that person was advising me to not open branches. If we have an institutional mechanism to harness the local knowledge, the field staff will pick up and tell. The answer to your question is, data analysis is not going to help you because all these events which have happened in the last 10 years is so specific to a particular region or with a particular reason. Hence, modeling is not going to help you. I will take one more example and explain. We have the data capability and every customer data is digitized, so people often ask about even credit scoring also, but we rely primarily on group underwriting, followed by check with credit bureau. The credit sensitivity here is, the women's husband becoming alcoholic during that year, but how is credit bureau going to tell you that. We have very capable data analytics in our office, who can come back with phenomenal conclusions. But, the top management is trying hard not to fall prey to that and the keep group underwriting effective. Andrey Purshottam: My other question was that, are you seeing a risk in these women, as they start coming to you for repeat loans and increasing their loan ticket size? Taking the loans for purposes that they are not pursuing but their husbands are pursuing; and what are your mechanism for ensuring that the loan is used for the specific purpose that it was designed for? Page 9 of 24

10 This is a very-very important question because even during AP crisis, when our non-ap markets had repayment rate of 97%, a major lesson was that if the large part of loan goes for income generation, then no matter, customer will repay; that is how the ever greening myth got eroded. To answer the question, we always have a process called loan utilization check. The big change that happened after the AP crisis was that earlier it was just a process, it was neither incentivized nor penalized. Now, post AP crisis, there is an incentive and the penalty is based on the external checks such as the outbound calls made through our in-house Member helpline team, then the sample check done by our internal audit with their own metrics and bankers also check. If ever there is a difference in the LUC done by that Sangam manager and the conclusions by any of these external check, it is a zero tolerance item. The process of LUC is that the Sangam manager goes and verifies at the customer place, whether the purpose for which the loan has been granted, has been undertaken or not. It could be an existing economic activity or might be a new one. In any case, the loan has to be first approved by the centre. If the women says I want this for a grocery store, the Sangam Manager doesn t do the stock verification. The approach here is more to confirm whether she has got a shop or not. Now, for buffalo, someone may show the same buffalo to us and the other lender, but we are fine, because we know we are not fully meeting her requirement. A buffalo costs nothing less than Rs.35,000 and the best ones go beyond Rs.65,000, while we are giving Rs.20,000 maximum as first loan (even our average offtake is Rs. 16, 612). In the initial two loan cycles, we actually want to co-share the risk and it is fine with us if she shows the buffalo to two people. The worry is when there is no buffalo. This is how the LUC process is done. The underlying premise of the Grameen model is, that if there is an economic activity she will find her way to repay the loans. And of course, you facilitate that by making a convenient installment, keeping the ticket size Page 10 of 24

11 low and having a conducive environment for the group underwriting to perform itself. Jhanvi: This is Jhanvi from Motilal Oswal. Sir, earlier you spoke about the regulatory arbitrage between MFIs and SFB players regarding the two MFI norm, what is your take on the interest rates cap of 10%? Do you think that also will apply to SFBs eventually or you think RBI will give them some levy given the Opex and the other costs that the SFBs might have to incur? No, as I said earlier, it is regulators prerogative. But, I have a personal view that RBI will not subject small finance bank to any of the norms that are applicable to NBFC-MFIs. RBI has always demonstrated that they like to follow a single form of presence and one set of regulations for one set of players. Hence, I do not think SFBs would be bound by the 10% margin cap. But having said that, let's look at what has happened, let s take Bandhan. They were charging 22.4% even after becoming a universal bank. It is amply evident that the spread cap does not apply to them. When we reduced the rate by 3.8%, brought it down to sub-20 at 19.75%, Bandhan has reduced the rates by 2.5% and now they are at 19.9%. Hence I think, at the end of the day, market forces and customer care will play out a role. Kashyap Zaveri: Kashyap Zaveri from Capital72 Advisors. I have three questions, one is on the other OPEX which has been going significantly lower than the overall asset growth. What more squeezing can be done over there, or would it start matching the overall asset growth as we go from here? Second question is on your entitlement on MAT credit, you have mentioned that this does not form part of your Rs.450 crores PAT guidance but does it form part of your Tier-I capital reported as of the quarter? And will MAT add to capital over a period of time, in which case what happens to the capital raising program? Page 11 of 24

12 And third is on the growth, we have seen till now growth being for last two years pretty aggressive, if you could give any guidance on broad CAGR over let's say next three to five years? Ashish Damani: In terms of OPEX, we have come down to 6.3% and aiming for it to be 6% in next 2 years. We will get more operational efficiencies, given that our growth is primarily driven by adding new loan officers and not addition of branches. Hence, we would see the operating cost growing around 25-30%, but in terms of portfolio, the growth will be 45% to 50%. Kashyap Zaveri: So you were talking about both employee plus other OPEX, right? Ashish Damani: That is right. Kashyap Zaveri: Could you break it up between the two, let's say other administrative or other OPEX? Ashish Damani: ~70% of our Opex is contributed by personnel expenses. The other OPEX would again be primarily rent and followed by vehicle fuel for the loan officers. Pratap R: MAT is similar to advance tax that is paid, so it is cash paid out and is available as credit to be adjusted against normal tax liability. Hence, due to this, it gets credited to the P&L and it comes on as an asset on the balance sheet, by that treatment it also becomes part of the networth and also considered for Tier-I capital; The reported capital adequacy of 23.2% includes the MAT credit recognition. And going forward, any MAT credit we recognize will continue to form part of Tier-I capital. Kashyap Zaveri: And on growth over period of three to five years? Ashish Damani: Our medium term guidance (two to three years) is to continue 50% growth rate. Post this, the CAGR should be around 35% for three to five year period. Page 12 of 24

13 Amit Monti: Hi, I am Amit Monti from 2point2 Capital. You had mentioned about the consolidation that has happened over the last few years post the AP MFI crisis, but if you look at FY16 numbers and the MFIN data of the 50 MFIs that they have listed out, 49 of those 50 have been growing in FY16, most of them have been growing at fairly rapid rates, maybe 40% - 50% plus. And there is no variance in terms of our smaller MFIs, which are also going probably faster than the larger MFIs. So in FY16, in some ways we have seen the reversal of the consolidation trends where even the smaller players are becoming more aggressive, we have the likes of Muthoot Finance who have in a short period of time got to Rs.650 crores kind of a portfolio. And so what do you think of this reverse consolidation trend, will it sustain over the next three to four years and what are the implications? What I meant earlier by consolidation was what followed immediately the AP MFI crises and the turnaround. If you look at Non-AP market share of MFIs under CDR and of some MFIs who just fell by the side, they also had a fairly large non-ap market share. Your point is also well taken, if you look at micro meter (Industry publication) last year, most of them have grown at 50%. But the point, I am making about consolidation is that ~ 80% of Rs.68,645 crores of AUM is in the hands of top 10 MFIs (including Bandhan). To answer your question on their growth path, is because of operating spread cap of 10%, somewhere you need to have minimum economic size of Rs.2,000 crores of AUM. ( I am just throwing a number, maybe one would say Rs.1,500 crs or other would say Rs.1,800. crs). Irrespective of your corporate health background, etc., the initial Public sector banks would insist on 1:1 leverage. Hence, unless you are talking about group or a corporate or an individual who can come with Rs.500 crores capital, complying with 10% operating spread cap is difficult. Hence, I am saying that for the names you gave, they may be new in terms of microfinance but they have their own strength. Page 13 of 24

14 And one more thing, lets even take this norm of not more than two MFI lenders. If the concern is about over leveraging of the customer, theoretically it does not make a difference whether you borrow Rs. 60,000 from five people or two people or 10 people, but why then there is such regulation? I do not want to explain more of the great thinking of RBI, by now you would have figured out. Hence, the operating spread cap of 10% to MFI lenders, have very consciously built quality barriers to ensure consolidation happens and the industry is filled with serious players of size and scale, so that the customers get lower rates, and most importantly the regulatory bandwidth of the regulator is also spent in an optimal manner. Jigar: Sir, Jigar from OHM. Sir, first question is on the fund raise, if you can highlight on the proposed mix of domestic and foreign? We have got shareholders approval for upto Rs.750 crores. As of 30 th June 16, we have 75% foreign equity and 25% domestic equity. There is no constraint on us to do only domestic issuance, because this is a 100% FDI/FII sector. Hence, there is no straightjacket around our shoulders. At today's price, if we take the entire Rs.750 crores as foreign, the foreign holding will only go up by ~1.7%. Hence, if we can manage 75%, we can manage 76.7% as well. The approval is valid for next 12 months. Also, it is not that we are gasping for capital, we have 23% capital adequacy. Hence, whenever there is a good window of opportunity we will effectuate. Jigar: Sir, we appreciate the fantastic presentation and just a question on that, if you can give that for the first bucket or the new customer, what is the average ticket size for IGL? Rs.19,410 for Q1FY17 (Note: We are mentioning this data point now, in the transcript). Page 14 of 24

15 Aman: This is Aman from Astute Investment Management. Sir, I have three set of questions. 1. Whether LUC is done for a group or for an individual? Is the external checks based on random or done for everyone? 2. What is the attrition rate of your company vis-à-vis the industry? And because there are so many MFIs how do you manage to retain the top talent? 3. The ratio of rejected versus approved loan applications and how you are seeing the trend? 4. The money chasing the same set of people is the biggest risk in the industry and you have the lowest amount of per person loan. But, it has still increased around 10% for this year or over a period of time. Have you seen any change in the repayment profile of such customers? 1. LUC - Group formation is a slow process. The center becomes functional with just three groups and will go up to 10 groups. To answer your question, LUC is never done for the whole group, but for individual members. The woman has to state the economic activity in the application. And the Sangam Manager does the LUC, post the disbursement (disbursement to a new group is done on staggered basis). Here, let s say, we have given loan for a buffalo and then do not find any during LUC. In such case, Sangam Manager will stop disbursement for other members in the group and there the peer pressure sets in. If there is a disconnect such as money used for consumption (marriage etc.), the woman is not penalized, but asked to return back the money, otherwise further disbursement to the group member is stopped. 2. Attrition - (Refer slide no. 73), we have 27% attrition rate for our Sangam Managers (Loan officers). We do not know the number for peers and it is not in public domain. In terms of retaining the best talent among our loan Page 15 of 24

16 officers, who are the only growth engine, we just provide support. We insist on 10th pass as the qualification. Our loan officers earn around Rs.16,000 per month (around their local lieu), which is a reasonably good sum for that geography. The loan officer is probably touching the tablet for the first time and hence there is good amount of training required. This job still commands a decent respect in the village. Having said that, it is a tough job because the guy has to report to the branch at 6.30 in the morning, take the cash, do five center meetings, come back to branch. Tablet has made things easier and he does have today 2-3 hours of rest during the day. We do not lose sleep over 27% annualized attrition, but what we focus on the following things At what point in time he is leaving the Organisation, I think he is figuring out that this job is not meant for him and we have also not invested heavily in him. (~50% of loan officers leave the job within 6 months) Data shows that the staff who leave makes lesser incentive than the average pool, thus, we would largely be losing underperformers. 4. Then, your question on the PAR or NPA. If you look at peers, who have businesses other than microfinance, such as individual loans etc. Their NPA at 1% in microfinance and has gone up to 4.6% on individual loans or 4.1% for old trucks or 3.9% for Affordable housing. We have 0.1% NPA in microfinance, but in the model we take credit costs at 1% as it includes standard asset provisioning also. Ashish Damani: 3. Slide no.25 has data on rejection rates. If you look at the rejection trend, the percentage has increased from 8% to 25% over the last 2 years, because the quality of the KYC has really improved. Now, we have 80% of our loan applications with KYC as Aadhar and that has helped us find the customer in the credit bureau database easier than before. Page 16 of 24

17 Aman: 3. Earlier inefficiencies were related to multiple KYC submission, but in October 2014, the MFIN code of conduct and we as an individual unit took a decision to put lot of focus on Aadhar. It is still not the single mandatory item but there is lot of focus on that. How often is the external check done? Apart from loan officers, we have outbound calls to customers on random sample basis to find whether the customer answer matches with the intended loan application usage. There are ways to get into that, the outbound call center team will start with an enquiry, are you happy, what you are doing, how much you are earning etc. Internal audit approach is more detailed because they have risk based supervision done on monthly basis for each branch. The difference is based on the grading, For example, depending on the grade of the branch, the time internal auditor spends on the field to do random LUC and various other things to detect ghost borrowers etc., is more than the time they would spend in the branch back office. If you look at it over a period of one year, in almost every branch this process of LUC external check by our own internal auditors would have been done. Aman: Do you still believe apart from external risk, the biggest risk to the industry is the increasing size of loan or? No, absolutely there is no denial that the over leveraging of borrowers is adverse to the sector, most importantly given the sensitivity of the customer segment. But, as I said earlier even at an industry level, there will always be pockets of excellence and pockets of inefficiency even in terms of players. Our loan outstanding is Rs.16,612 and we are very comfortable with that going up by 15% annualized for the next three years. Page 17 of 24

18 Andrey Purshottam: Andrey Purshottam again from Cogito Advisors. I just wanted to ask a question about your IT expenses and the degree of digitization. When you enter the data at village level, how long does it take you to capture a head office? Do you envisage your future IT expenses to be fairly incremental in nature or will it be fairly dramatic in terms of overhauling the entire structure? On the digitization, the tablets are making a huge difference; the software pre-populates data on affirmative basis. Loan officer comes back from the field and connects his tablet to the system, it is a store and forward model. Out of 1,200 branches, there will be only branches sending data on the next day or thereafter. If they do not sync the data, we have escalation metrics to handle this (although it could be genuine reasons such as no power supply, but we need to find it). Now on the IT budget, we largely have the IT architecture in place. Unless we spend money on a network of retail distribution points (i.e. cash points), we don t see any radical change. But having said that, the economic life of a tablet given to loan officers in a rural area is 2 years. I think these kind of IT replacement expenses will be around Rs.16 crores every 2 years. We have partnered with IBM and Samsung on the same. Sangam: Hi, this is Sangam from Subhkam. As you said that over leveraging of the existing customer is one of the very risky part of this business going forward, given that SFBs are coming and there the regulation is not there with regards to two MFIs being restricted to per customer. So how do you see the situation, say maybe couple of years down the line, the existing customer having exposure to may be an SFB as well. We would that restrict our exposure to that customer, assuming that it is one MFI and one SFB? And how from a risk mitigation perspective what are the steps that you would take in this regard and whether this will have any effect on your growth profile as well going forward? Page 18 of 24

19 As we mentioned earlier, my personal view is that the SFBs will not be subjected to these two lenders, or for that matter any other NBFC-MFI guidelines as a template. But it is not a constraint for you for growth. Today out of the 25% rejection in loan applications, 85% of the reason is more than two lenders and I am sure in most of those cases, one is a proposed SFB. The opportunity exists with both NBFC-MFIs and SFBs to ignore or include. NBFC-MFIs may ignore SFBs for this norm and vice versa. But, whether you are an MFI, SFB or a bank, you need to share the data with all the four credit bureaus. And I think these are responsible players of some size, with some track record, have credit rating, governance rating. If data is tearing into your eyes, will you make use of this flimsy excuse of becoming the third lender, my hunch is that will not happen and that will not be allowed to happen as well. Sangam: Today morning in CNBC, if I am not wrong, Equitas mentioned this point wherein being an SFB we are not subjected to that, so that opens up the opportunity with the same existing customer. That is a big risk in terms of leveraging of the particular customer with the current cash flows. I completely conclude with the analysis, but slightly differing with the conclusion. I have been repeatedly saying from the beginning that the two MFI lender or 10% operating spread cap or any of the NBFC MFI operative guidance will not apply to an SFB. Hence, there is no difference in that analysis. Now, if the data is available with you whether it is two loans or three loans, she has got 1 lakh and if your credit and risk systems are not comfortable with that, will you go on and increase exposure? I will be candid, if you want to throw caution to the window and use excuses to expand your book, night will follow the day. We are very conscious and we have reinforced. Page 19 of 24

20 The other comfort I have it is especially during the transformative period, players will be under complete scrutiny of Reserve Bank of India. Whilst the template will not be thrown at the SFBs, but the exploitation of that will not be allowed and RBI will have close monitoring in the initial stages of any bank transformation. I am trying to put it in different language called scrutiny, moral suasion etc. I have huge faith in RBI, they will not allow this to happen. Sangam: In terms of securitization, what percentage will we be comfortable? We are actually fine with the highest proportion coming from securitization, obviously because that is the lowest cost source for us, but some sort of a natural constraint is there. A large part of securitization is on Agri & Allied activities. Although we do not have major direct-agri exposure, but the livestock and some of the other businesses are included. Hence, about 42% of our portfolio is eligible for being termed as Agri and allied. Sangam: What are the components of AUM growth of 50%? Ankit: We are comfortable with 15% to 20% Annualised increase in ticket size from here, but a large part of increase would come from borrower growth i.e. 25 to 30%, which will be driven by 30% addition of Sangam managers annually. Hi, my name is Ankit. I just have a couple of questions on operations and one specific question on financials. On the operational side, how much of the 5 million customers have finished more than two or three borrowing and repaying cycles? That is the first question. Second is, how many of the branches have reached the maximum exposure on a per branch level? You may have a Rs.3 crores to Rs.4 crores cap on per branch, so what is basically capacity utilization level? Page 20 of 24

21 And the third question is, for the next nine months, is it safe to assume that the PBT will be equal to PAT, because we will be taking a MAT credit for the provision that you do. Ashish Pipaliya: In terms of borrower base, about 56 % of our customers are in cycle 2 and above. Pratap R.: Ankit: Pratap R.: Ankit: Coming to MAT, similar to this quarter whatever we have as MAT expense will get recognized as MAT credit and with that PBT equals to PAT, as long as we continue to pay MAT and recognize MAT credit thereof. Just as a corollary to that question, you have a deferred tax asset of about Rs.308 crores, so when is the timing difference reversing, so when would you start paying the normal Income Tax? We have carried forward tax loss of around Rs.790 crores, until we utilize that, DTA will continue to remain. Okay, so about mid of next year I would say? Pratap R.: Ankit: Pratap R.: Ankit: Yes, based on what your projections are. After that you will start utilizing this MAT credit, so till that you will pay 20% only? Yes, there will be a cash outflow of minimum of 21% and the differential will get adjusted from MAT credit. Now, just with this addition of tax assets to your net worth, will there really be a need to raise capital because you will be more than happy to get the growth done for the next one or two years because you have got the cushion of tax assets? Page 21 of 24

22 Ashish Damani: Actually, We have put it in the public domain that we are going to recognize MAT and have already factored in this our plan. The capital adequacy is comfortable, but it is our self-imposed financial obligation to maintain 5% excess capital above the mandated ratio. Now, if we continue with this 50% growth, somewhere in Q3FY17 or Q4FY17, we will touch the 20% capital adequacy. Hence, we will still like to go ahead with the capital raise. There was one question on branches, how branches are faring out. We have opened about 50 branches in last quarter or so, barring that all our branches would be above the break-even level. Pratap R: Savi Jain: Our concentration risk exposure norms suggests that a branch exposure ideally should be less than 1% of the net worth (with a tolerance range of 5% branches can go up to 2%). There are only 11 branches between 1% and 2% of networth. My name is Savi Jain. Is there an overlap in the customers who take microfinance loan and gold loans? I am asking this question because now your interest rates are lower than what the gold finance companies are charging, so is there a segment of customer who is moving from taking gold loans to microfinance or is the profile of such customers are very different from someone who could take up microfinance loan? I do not have data, but just a gut feel that overlap of target and loan usage is minimal. There are two factors, the first is that microfinance addresses low income group, whereas the gold loan is more for addressing irregular income. I do not have deep understanding of gold finance, but whatever little I know, the difference is that gold loan is more for irregular income, financial emergency (For example: There is school fees to be paid for the kids, there is a medical emergency, sudden travel plans etc.). Hence, those kind of requirements (what I call is irregular income group) largely rely on gold loan and ultimately there is a natural constraint of your gold holding. Also, if we look at branch network, there is a huge difference, the gold loan Page 22 of 24

23 branch would be close to a Mofussil bus stand and they are not deep into the village like microfinance institutions. The other key difference is in microfinance, customer is repaying on a weekly basis. I do not have data, but little bit of anecdotal evidences suggest that just because microfinance is cheaper I do not think a gold loan borrower would change into microfinance. And the reverse is also true, that even if the gold loan interest rates comes down, the microfinance borrower cannot take gold loan, unless she has gold holding. Savi Jain: Now between SHG and all the microfinance companies, the credit outstanding to this profile of customer has almost reached Rs.1 lakh crores. If these people are indeed generating economic activity out of this money why this is not showing in incremental GDP or even if I have to look at companies which are focused on these rural customers, I do not see the results reflecting these kind of economic activity being created. Basically it goes back to way the GDP is captured, if there are economic activities, rural incomes are part of it, I do not think there is separate statistics on breakup of contribution from microfinance, SHG or informal borrowing. In fact, there is a great discussion about Mr.Nandan Nilekani talking on what is the biggest trend in the next 10 years and he beautifully articulated that to be formalization of the economy. Hence, the economy exists but it may more in favor of informal borrowing today, just like your labor market which is largely an informal economy. Savi Jain: But consumption should have increased, so the consumption should come in the formal economy because there will be banks, there will FMCG As they say one man's poison is other man's food, somebody's income is someone else' consumption. If our borrowers have economic activities catering to basic consumption like food (basically 33% livestock) plus Page 23 of 24

24 grocery stores, food catering etc. that data will get captured. But, there is no specific data around this, or for that matter how much of that is formal or informal. Let me take liberty and reframe your question, is there a credible third party impact assessment done of what difference a microfinance loan does? There are plenty of studies. For example, the SIDBI s seven years longitudinal study, mentions impact assessment of microfinance loans. Even post Andhra Pradesh crisis there was a study done by MicroSave on what happens in case of no access to microfinance loans. Ashish Damani: Thank you. Please join us for the high tea. Page 24 of 24