Financial services for low-income customer segment in India is a vast, but difficult terrain. NBFCs are alternative

Transcription

1 BFSI September 2016 Financial services for low-income customer segment in India is a vast, but difficult terrain Difficult to ride it through BANKS NBFCs are alternative Best to ride it on SFBs high cost model means limited reach better reach but limited offering built for deeper reach with wider offering Small Finance Banks: Made to measure Research Analysts: Ravi Singh ravi.singh@ambit.co Tel: Rahil Shah rahil.shah@ambit.co Tel: Pankaj Agarwal, CFA pankaj.agarwal@ambit.co Tel:

2 BFSI CONTENTS SECTORS SFBs: Made to measure..3 A new breed of banks 4 Market opportunity of Rs2trn to support 20% loan CAGR..7 After a near-term dip, RoEs set to recover 12 Sifting through SFBs for potential winners 21 Valuation: benchmarking with consumer lender 35 NBFCs indicates re-rating potential COMPANIES Equitas Holdings (BUY) 41 Ujjivan Financial Services (BUY)...57 September 22, 2016 Ambit Capital Pvt. Ltd. Page 2

3 BFSI POSITIVE THEMATIC SMALL FINANCE BANKS September 22, 2016 SFBs - Made to measure Substitution of high-interest informal credit to low-income group can support 20% AUM CAGR (FY17-21) for small finance banks (SFBs). SFBs are better placed to cater to the segment than NBFCs (higher cost of funds and fewer offerings) and banks (few low-cost field staff). RoEs would see a near-term dip but recover to ~18% by FY22 (similar to consumer lender NBFCs today) led by decreasing cost of funds and capital consumption. While consumer lender NBFCs are a good valuation benchmark, some discount is warranted given SFBs limited track record. Whilst we are BUYers on both, Ujjivan (implied 2.8x FY18 P/B) is marginally ahead of Equitas (implied 2.7x FY18 P/B) on brand, innovation-led culture and organizational strength. Equitas, highly focused on containing risks first, has a more conservative approach. Substitution of informal credit offers `2tn market for SFBs Stratified economic data shows that 24% credit penetration in bottom 142mn households with 60% market share of informal credit (~30% interest rate vs ~20% from SFBs) holds the growth opportunity (`2tn, US$30bn) for SFBs. Superior cost of funds (~200bps lower) and product offerings (lending along with savings, payments) place SFBs favorably against NBFCs; banks don t yet have operational model (lack frontline cadre and customer focus) to compete effectively in SFBs target low-income group customer segment. RoE to dip 400bps in near term and recover to ~18% by FY22 Near-term RoEs would dip by ~400bps, on an average, due to regulatory and operational costs associated with transformation. However, as one-off costs normalise, cost of funds decline (~200bps as SFBs have access to new funding avenues) and capital is consumed (driving financial leverage from 5x to 8-9x) RoEs can recover to ~18%, in five years, by FY22E. Ujjivan and Au Financiers stand out Ujjivan and Au, closely followed by Equitas, emerge neck-to-neck on superior risk-management (product/geographical diversification and ticket sizes) and RoEs. On our IBAS framework, we find Ujjivan making better progress on creating sustainable competitive advantages through a credible brand, innovation-led culture and well-knit organization. Equitas, on the other hand, appears highly focused on containing risks first (low ticket size of loans) and maintaining its conservative approach at this stage. Visibility on RoE recovery drives our BUY stance Benchmarking with other BFSI sub-sectors places SFBs closer to consumer lender NBFCs on steady state RoE and growth potential. Even with some discount to consumer lenders (2.7x FY19 BV with longer track records), current valuation of SFBs (2.1x FY19 BV) has re-rating potential. Margins, asset quality and progress of SFB transition are key catalysts for the re-rating. Our valuations, based on excess RoE model, indicate an upside of ~23% for both Equitas and Ujjivan. Worsening of market discipline in lending to the lowincome group, hardening of interest rates, and higher compliance and operational risk standards as banks are key risks to our thesis. Equitas and Ujjivan a quick comparison AUM (` bn) (Jun'16) RoE (%) P/B (X) EPS CAGR FY16 FY18E FY22E FY17E FY18E (FY17-21E) Equitas % 10.8% 18.4% % Ujjivan % 12.2% 18.6% % Key Recommendations Equitas Holdings BUY Target Price: `220 Upside: 23% Ujjivan Financial Services BUY Target Price: `500 Upside: 23% Share price performance May-16 May-16 Jun-16 Jun-16 Jul-16 Jul-16 Aug-16 Aug-16 Aug-16 Sep-16 Equitas Ujjivan Sensex Source: Bloomberg, Ambit Capital research Research Analysts Ravi Singh ravi.singh@ambit.co Pankaj Agarwal, CFA pankaj.agarwal@ambit.co Rahil Shah rahil.shah@ambit.co Ambit Capital and / or its affiliates do and seek to do business including investment banking with companies covered in its research reports. As a result, investors should be aware that Ambit Capital may have a conflict of interest that could affect the objectivity of this report. Investors should not consider this report as the only factor in making their investment decision.

4 BFSI A new breed of banks In September 2015, the RBI paved the path for the creation of a new segment in the Indian financial system by granting 10 small finance bank (SFB) licenses to existing NBFCs (mostly MFIs) and a local area bank. With the overt objective of financial inclusion, the rules regulating the new sector mandate a focus on small-ticket lending to customers at the bottom of the pyramid (see exhibit below). The entities, in return, are allowed to mobilise deposits and expand into other banking activities. Exhibit 1: Key features of the RBI s guidelines for small finance banks Activities Key guidelines Primarily basic banking activities of deposit-taking and lending for small-sized customers. Loans of size up to `.2.5mn will have to form at least 50% of the loan book. Distribution of mutual funds, insurance and pension products with the prior approval of the RBI. Forex dealing is allowed. At least 25% of branches need to be in unbanked rural centres (population <10k). The small finance bank CANNOT be a Business Correspondent (BC) for another bank. However, it can have its own BC network. After first five years of operation and a review by the RBI, the requirement of prior approval for annual branch expansion plans and scope of activities can be liberalized. Prudential norms Usual CRR and SLR apply (currently at 4% and 21.5%, respectively). Capital requirement Shareholding Source: RBI, Ambit Capital research PSL (priority sector lending) requirement of 75% (of adjusted net credit) will apply. 40% of adjusted credit under PSL will be subjected to similar sub-limit for sub-sectors as it is for existing banks. The remaining 35% of adjusted credit can be to any subsectors. An NBFC/MFI converting into an SFB will cease to exist and all its business which a bank can undertake should be brought under the bank and the activities which a bank cannot statutorily undertake should be divested or disposed of. Minimum paid-up equity capital of `1bn. Minimum Tier-I capital of 7.5%; minimum capital adequacy ratio of 15%. Minimum promoter shareholding of 40% for five years. Promoter shareholding to be brought down to 40% in first five years to 30% in next five years and 26% in further two years. Foreign shareholding as per FDI policy for private sector banks, i.e. currently total FDI limit at 74% with FII sub-limit of 49%. Listing within three years after reaching a net worth of `5bn. A high focus on financial inclusion and small-ticket lending led to eight microfinance companies (MFIs) receiving SFB licenses. Two other SFB license recipients Capital Local Area Bank (a local area bank) and Au Financiers (an NBFC) also focus on small-ticket lending in local markets. Exhibit 2: Brief profiles of small finance bank license recipients (end-fy16) Company Type Based in Number of branches Portfolio size (AUM, ` bn) Au Financiers NBFC (CV, SME, LAP) Jaipur Capital Local Area Bank Local Area Bank Jalandhar Disha Microfin Micro Finance Institution Ahmedabad Equitas Holdings Micro Finance Institution Chennai ESAF Microfinance and Investments Micro Finance Institution Thrissur Janalakshmi Financial Services Micro Finance Institution Bengaluru RGVN (North East) Microfinance Micro Finance Institution Guwahati Suryoday Micro Finance Micro Finance Institution Navi Mumbai Ujjivan Financial Services Micro Finance Institution Bengaluru Utkarsh Micro Finance Micro Finance Institution Varanasi September 22, 2016 Ambit Capital Pvt. Ltd. Page 4

5 BFSI These 10 SFB license recipients are in the process of launching their SFB operations (Capital SFB and Equitas SFB already launched). Two of these entities (Ujjivan and Equitas) have gone for public listing and a few more are likely to list. The initial reception of listed SFB license holders has been strong with 30-80% returns since listing with valuations at 2.6x FY17E BV being at ~110% premium to regional private sector banks but at ~30% discount to consumer financing/sme financing NBFCs. In this note, we attempt to answer the following questions: Market opportunity: What is the size of the opportunity for SFBs? What are the drivers of growth in terms of rising credit penetration in the target segment and market share shift from informal lenders? Journey of transition: Can new SFBs exploit this opportunity in a profitable and sustainable manner? Near term: What is the immediate net impact of regulatory and operating expenses linked with conversion to an SFB, change in cost of funds and improvement in balance sheet efficiency? Long term: What is the realistic sustainable profitability SFBs can aspire to as a result of liabilities franchise stabilizing with some amount of low-cost granular liabilities and asset mix undergoing change with addition on new asset classes? Who to pick? What differentiates one SFB from another? Right valuations for the sector: What is the fair valuation multiples for SFBs based on their long-term growth and ROE potential and whether SFBs should trade at a discount or premium to rest of the BFSI sub-sectors? September 22, 2016 Ambit Capital Pvt. Ltd. Page 5

6 BFSI Exhibit 3: Porter s analysis for small finance banks Bargaining power of suppliers Bargaining power of buyers HIGH With number of new banks, new fin-tech companies and existing bank vying for same talent pool, bargaining power of banking talent is high. MODERATE A large number of customers from low-income group are first time users of financial services and have little to choose from in terms of lenders. With little differentiation as deposits takers, SFBs main bargaining tool to mobilise retail or bulk deposits are interest rates. This reflects high bargaining of depositors The bargaining power against technology providers is evenly placed due to high competition among vendors and banks requirement to choose right technology. Given bulk of the lending is microfinance with no collateral, borrowers have little to lose in case of defaults. This gives lower bargaining power to borrowers in terms of pricing, but higher pricing power in case of default. With credit penetration rising, increasingly multiple lenders are chasing same customers which raises bargaining power of borrowers. Competitive intensity MEDIUM Competition from NBFC-MFIs/NBFCs is intense, however lower cost of funds, stable liability base and better brand and customer trust will give an edge to SFBs. A favorable operating model (large share of low cost front-line field staff) protects from competition from banks. Most upcoming SFBs are regional entities with little over-lap of home states among them, which limits the competition among SFBs. Barriers to entry HIGH SFB licenses are regulated by the RBI and even though the RBI could, at some stage, begin issuing on-tap licenses, the entry of new banks is likely to be some years away, which will allow first batch of SFBs to establish their presence. Banks are trying to enter into low income group customer segment through tie-up with MFIs and BC channel. However, the model is yet unproven and fraught with operational risks linked with customer selection, cash management and banks brand and reputation. Threat of substitution LOW Due to low financial literacy and technological awareness, these customers yet need extensive door-step servicing. Thus, these customers are unlikely to be catered by technological solutions offered by banks/fintech companies. Erratic cashflow requirements against uniform and low income pattern mean that customers will continue to need credit to meet the financial needs during fluctuating shortfall from income cashflow. Improving Unchanged Deteriorating Source: Ambit Capital research September 22, 2016 Ambit Capital Pvt. Ltd. Page 6

7 BFSI Market opportunity of `2tn to support 20% loan CAGR As a section of the least income earning population in India moves to higher income segments, SFB-like entities (MFIs, bank-linked SHGs) will find a large opportunity opening up. Stratified economic data of rural and urban areas show SFBs can cater to ~140mn households out of 265mn Indian households. However, even in the lowest strata, the credit penetration (~24%) is material already and comparable to richer households (~33%). But, the share of informal credit (at higher interest rate of ~30% and above) is much higher in the lower income group (~60% vs 30% in richer segments). Thus, taking market share from higher interest rate informal credit is a key opportunity for SFBs. This phenomenon, in our view, can support loan CAGR of 20% for SFBs, as a whole, in the coming 3-5 years. Regulatory norms force upcoming SFBs to focus on small-ticket lending to financially under-served and un-served segments (loans below `2.5mn will have to form at least 50% of loan book). This task covers a range of products under two main lending segments individual retail loans (including microfinance) and small business loans. While products offered would vary, the focus of the SFB format would be to keep the same specific customer segment at the center of their operating models. This customer segment is that of economically active urban and rural poor and small selfemployed customers. Given that eight MFIs dominate the list of 10 SFB license recipients, microfinance is likely to remain the central product initially for most new SFBs. Even over the longer term, as SFBs introduce new banking products, the core customer segment focus is not likely to change much. We build ~60% and ~35% dependence on microfinance for both Ujjivan and Equitas and expect 50-60% for the entire industry by FY20. The financial services entities currently active in these customer segments are banks, including co-operative banks (retail, agricultural and small business loans), NBFCs (small business loans, vehicle loans, home loans), and MFIs (group loans to individuals). The market opportunity for new SFBs thus depends on 1) rate of growth in size of their target customer segment; 2) level of credit penetration in that segment; and 3) market share shift from informal players. The regulatory norms force upcoming SFBs to focus on smallticket lending to financially underserved and un-served segments. Microfinance is likely to remain the central product initially for most new SFBs Changing population mix favors increasing demand; potential customer size is ~140mn households As per the 2011 census and population growth rates, India s population is estimated to be ~1.3bn and comprises ~265mn households. Urban centres account for 85mn households and rural centres the remaining 180mn households. The wide disparity in income and wealth distribution for the population spread across a large geographical area makes it necessary to look at stratified economic data to assess the size of the opportunity for microfinance. Economic data (as per NCAER, or National Council of Applied Economic Research) shows a long-term trend in changing mix of households by income categories (Exhibit 4 see note for definitions). The starkest shift, both in rural and urban markets, is in the segments below middle-class, where a large number of households has been moving out from bottom-most income segment ( deprived ) to slightly better segment ( aspirers ). This creates an opportunity for lenders operating in economically active urban and rural poor to address the financial services demand of a growing segment. We look at stratified economic data to assess the size of the opportunity for microfinance. A large number of households have been moving out from bottom-most income segment (deprived) to slightly better segment (aspirers). September 22, 2016 Ambit Capital Pvt. Ltd. Page 7

8 BFSI Exhibit 4: The long-term trend of households moving from the deprived segment to the aspirers segment 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0.1% 1% 2% 3% 0.4% 3% 6% 6% 15% 12% 23% 22% 30% 31% 82% 2002: Rural 63% 2010: Rural 56% 2002: Urban 60% 14% 2010: Urban 72% 2002: Total 2% 13% 34% 52% 2010: Total Rich Middle class Aspirers Deprived Source: NCAER; *NOTE - Definition (based on annual household income at 2002 prices): Deprived = below `90k, Aspirers= `90-200k, Middle-class = `200k-10mn; and Rich = over `10mn. Potential addressable market and credit penetration The decennial National Sample Survey (NSS) carried out by the Government of India for debt and investments pattern segregates Indian households according to asset holdings and analyses the pattern of credit penetration and debt obligations for these groups (see exhibit below). Exhibit 5: Credit penetration and average debt of rural and urban households segregated in deciles in increasing order of asset holding (2013) ` % 20% 30% 40% 50% 60% 70% 80% 90% 100% Avg. Average value of household assets Rural ,549 5,689 1,007 Urban ,248 2,001 3,513 14,560 2,285 Credit penetration Rural 19.6% 22.3% 27.1% 27.5% 31.0% 33.0% 32.7% 37.3% 42.6% 41.3% 31.4% Urban 9.3% 14.6% 20.2% 24.2% 21.7% 23.4% 23.8% 25.4% 29.4% 31.7% 22.4% Avg debt per indebted HH Rural Urban , Source: NSSO, Ambit Capital research Across various deciles of asset holdings, the data shows the level of formal or informal credit penetration and also the average indebtedness of the credit takers. While estimating the market size for microfinance using this data, we look at the average indebtedness of households in different deciles and compare that with loan ticket-sizes that are served by microfinance. For addressable market size, we have thus focused on customer indebtedness of up to `70k in rural markets. While average ticket sizes for microfinance are `15-30k, the addressable customer segment could have indebtedness from other source also, such as farm loans and mortgages. For urban markets, we have used indebtedness level up to `120k, higher than rural centres, due to higher presence of individual small business loans and housing loans. Exhibit 6 shows the distribution of household debt in urban and rural centres by purpose of loans. NSO data shows the level of formal or informal credit penetration and average indebtedness of the credit takers across various deciles of asset holdings. September 22, 2016 Ambit Capital Pvt. Ltd. Page 8

9 BFSI Exhibit 6: Mix of household credit by purpose of loans business and housing loans drive credit demand Rural Urban Total Expenditure in farm business 29% 2% 13% Expenditure in non-farm business 11% 16% 14% For housing 20% 59% 43% For other household expenditure 23% 12% 16% For education 3% 4% 3% For medical treatment 6% 2% 4% Others 8% 5% 6% Total 100% 100% 100% Source: NSSO, Ambit Capital research Thus, as highlighted (dotted red line) in Exhibit 5, for rural households, we define addressable market as bottom 60% in rural markets and bottom 40% in urban markets. While this is true that not all households in the bottom segment are bankable and are unlikely to be addressed by new SFBs any time soon, this trait reflects in lower credit penetration in lower deciles, which are likely to remain structurally lower than that for households in higher deciles. Adding target segments in this manner adds up to an addressable market of ~142mn households (40% of urban households and 60% of rural households). While ~142mn households, with ~24% credit penetration only, appears a large addressable market for SFB/microfinance/SHG lending, the actual opportunity could be materially lower as the microfinance model in India mostly relies on presence of economically active women in households to target these households for lending. Some part of this potential market is also likely to stay unapproachable in the foreseeable future due to spread in far-flung areas or adverse socio-economic reasons (e.g. tribal population in difficult terrain). Market share gain from informal markets NSO data further shows a significant presence of non-institutional lenders (see definition of institutional and non-institutional lenders in the notes to Exhibit 7) in rural and urban markets, particularly in the lower strata. The presence of noninstitutional lenders is higher in lower ranking households on asset holding overall and in rural markets. Bottom 60% in rural markets and bottom 40% in urban markets form addressable for SFBs. Non-institutional credit accounts for 46% of rural household credit and 26% of urban household credit. Exhibit 7: Share of institutional/non-institutional credit for indebted households; deciles arranged by asset holding (rural) Exhibit 8: Share of institutional/non-institutional credit for indebted households; deciles arranged by asset holding (urban) Non-Institutional Institutional Non-Institutional Institutional 100% 100% 80% 60% 36% 30% 36% 41% 37% 44% 50% 51% 57% 68% 80% 60% 34% 38% 47% 46% 40% 20% 0% 64% 70% 64% 59% 63% 56% 50% 49% 43% 32% 10% 20% 49% 30% 52% 40% 50% 60% 70% 80% 90% 100% 58% 65% 78% 84% 40% 20% 0% 66% 62% 53% 54% 51% 48% 42% 35% 22% 10% 20% 30% 40% 50% 16% 60% 70% 80% 90% 100% Source: NSSO, Ambit Capital research; *Note: Institutional lenders include banks, government, SHGs and NBFCs. Non-institutional lenders include landlords, moneylenders, relatives and friends. Source: NSSO, Ambit Capital research September 22, 2016 Ambit Capital Pvt. Ltd. Page 9

10 BFSI The average outstanding debt for indebted households in rural and urban markets indicates that share of non-institutional credit in household credit is 46% in rural market and 26% in urban market (exhibits 7 and 8). Further, data based on interest rate of loans shows that presence of non-institutional credit is highly skewed towards high-interest-rate lending (Exhibits 9 and 10). Exhibit 9: About 70% of informal rural household debt is at interest rate of 20% and higher Exhibit 10: For urban households ~60% of informal debt is at interest rate of 20% and higher 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1% 2% 16% 34% 16% 50% 35% 32% 47% 10% 36% 22% Institutional Non-Insti. All >30% 20-30% 12-20% <12% 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0% 1% 5% 6% 30% 40% 36% 28% 12% 58% 54% 30% Institutional Non-Insti. All >30% 20-30% 12-20% <12% Source: NSSO, Ambit Capital research Source: NSSO, Ambit Capital research Thus, non-institutional credit to Indian households with interest rate of more than 20% is estimated to be `3.1tn and more than 30% is `1.5tn. Assuming that penetration in the addressable market rises only gradually and microfinance, in the near term, focuses only on poaching market share from high interest rate informal credit on the back of flexible operating model and lower interest rates, the market potential is significant. Total AUM of MFIs/SHGs stands at `900bn (~1% of Indian bank s loan book) vs informal credit of `1,500bn (2% of Indian bank s loan book) with interest rate of more than 30%. Exhibit 11: Pulling it all together - current status of target market segment for SFBs Rural Total SFB target segment Above target segment Number of households (mn) Credit penetration 31% 27% 38% Average debt per household for indebted household (` k) Outstanding household credit (` bn) 5,863 1,627 4,236 Share of informal market 46% 61% 40% Urban Number of households(mn) Credit penetration 22% 17% 26% Average debt per household for indebted household (` k) Outstanding household credit (` bn) 6, ,048 Share of informal market 26% 54% 22% Total Number of households(mn) Credit penetration 29% 24% 33% Average debt per household for indebted household (` k) Outstanding household credit (` bn) 12,826 2,187 10,284 Share of informal market 35% 59% 29% Source: NSSO, Ambit Capital research September 22, 2016 Ambit Capital Pvt. Ltd. Page 10

11 BFSI Exhibit 12: Market opportunity five-year CAGR of ~20% looks reasonable Low case Base case High case Comments Total households (mn) 265 As at end Addressable market, in terms of households (mn) 142 Current credit penetration in addressable market 24% Estimated market that is being served already (` bn) 1,041 Assuming average ticket size of ~`30k per household. Target increase in credit penetration 2% 5% 10% Growth opportunity (` bn) Target market share gain from informal market 10% 20% 30% Growth opportunity (` bn) ,279 Inflation in ticket sizes 5% 8% 12% Growth opportunity (` bn) Total market size in five years (` bn) 1,840 2,595 3,540 Growth CAGR 12% 20% 28% Source: Ambit Capital research Credit penetration in upper end of customer segment is ~10% higher than the lower segment. Informal lending has ~60% market share in addressable market vs ~30% market share in upper customer segment. Reflective of growth in income level and average ticket sizes. In summary, the economic data suggests that 1) potential size of SFBs customer segment is ~142mn households; 2) but the key driver of potential growth is market shift from high-interest informal lending to SFB rather than a sharp increase in credit penetration as credit penetration is already significant. In five years, we believe lending to this customer segment could grow at ~20% CAGR to reach a market size of ~`2.6tn (current market size is ~`2tn). The key driver of potential growth for SFBs is market shift from highinterest informal lending rather than a sharp increase in credit penetration as credit penetration is already significant. September 22, 2016 Ambit Capital Pvt. Ltd. Page 11

12 BFSI After a near-term dip, RoEs set to recover While the market opportunity is significant, the key challenge in next 1-2 years for upcoming SFBs is transformation from NBFC to bank. The operational expenses and regulatory cost related to this conversion will hit RoE in the near term. However, the immediate reduction in cost of funds by bps will partly offset this impact. Overall, RoE could decline by bps between FY16 to FY18 before recovering from FY19. With opex normalizing after the phase of initial investment and funding costs further coming down, we expect RoEs of better run SFBs to normalize closer to 18% over 3-5 years (FY20-22). A frontline heavy operating model compared to banks and lower cost funds and better brand/trust would help SFBs strengthen their niche in low income customer segment and hence would help SFBs deliver these ROEs. Moreover, diversification of the asset base away from microfinance to individual MSE or housing loans would reduce concentration risk and would make business models more robust. Conversion related opex to impact near-term RoEs As the companies transform into SFBs, the components of overall profitability will undergo changes. The table below lists the positive or negative impact of various factors on profitability. While regulatory cost will be a drag on an ongoing basis, growth in opex around SFB conversion will likely spike due to one-off investments and will begin to significantly normalize from FY19. Offsetting these costs, cost of funds will also fall. Initially, the companies will benefit from newer avenues of funds (e.g. CDs (certificates of deposit), bulk deposits and CPs (commercial paper)) at lower cost that they will be able to avail as banks. Over the longer term, some mobilization of low-cost retail deposits could provide further easing of cost of funds, but the progress is likely to be gradual and will be associated with cost of investments in distribution network. Exhibit 13: Impact on profitability due to conversion into an SFB Factor Impact Comment Negative Opex incurred for branch conversion/technology/marketing Application of SLR Application of CRR Application of PSL Positive Fall in cost of wholesale funding Mobilisation of retail deposits Savings on optimisation of cash balance maintenance Increase in leverage Source: Ambit Capital research Significant and immediate Significant and immediate Negligible Negligible Significant and immediate Material in long term Material and immediate Material in long term In the next exhibit, we look at what could be the near-term RoE impact on Ujjivan and Equitas (two upcoming SFBs) purely due to change in form from NBFC to SFB. While regulatory cost will be a drag on an ongoing basis, opex will spike due to one-off investments around SFB conversion. Offsetting, these costs, cost of funds will also fall. Costs incurred in rolling out bank branches and other channels, related recruitment and infrastructure upgrade, and upgrade of IT systems SFBs will need to hold minimum SLR, same as universal banks. Thus, a portion of fund mobilised will have to be deployed in lower earning SLR investments than higher-yielding loans. While the usual CRR requirement applies, this is unlikely to be a major challenge given that these entities already carry significant cash on their balance sheets that can be moved to meet CRR requirements. While usual PSL requirements apply. This is unlikely to be a major drag as most of the existing lending already qualifies for PSL. Ability to access CDs/CPs/bulk deposits/refinance at lower rate as bank than current cost of funds. Over a longer term, mobilisation of retail term deposits and savings balances would lower cost of funds. Currently, the MFIs hold a large share of their assets as cash due to seasonality in funding received from banks, reliance on longer-duration market borrowing and lagged deployment as lending. After conversion to banks, the balance sheet can be used more efficiently through more effective treasury management. Current leverage (5-6x) is low in MFI; this is likely to gradually increase closer to banks leverage of 12-13x. September 22, 2016 Ambit Capital Pvt. Ltd. Page 12

13 BFSI Exhibit 14: Near-term RoE hit of ~200bps (due to change in the form from an NBFC to an SFB alone) on Ujjivan and Equitas (as of FY16 financials) Balance sheet (as % of assets) Liabilities Ujjivan Equitas Comments Current As SFB Current As SFB Capital 20% 20% 18% 18% Borrowings 77% 77% 76% 76% Other Liabilities 3% 3% 6% 6% Total Liabilities 100% 100% 100% 100% Assets Cash 8% 5% 13% 6% Investments 0% 15% 0% 13% Better options for cash management to help SFBs carry lower cash on their balance sheets. SLR investment 0% 15% 0% 13% SLR requirement leads to SFB deploying funds in SLR book. non-slr investment 0% 0% 0% 0% Loans & Advances 89% 77% 81% 74% Fixed & other assets 3% 3% 6% 6% Total assets 100% 100% 100% 100% Du-pont Interest income to assets 18.5% 17.7% 16.1% 15.9% Yield on loans 21.5% 21.5% 20.0% 20.0% Assuming yield on loans stay unchanged. Yield on investments NA 7.6% NA 7.6% Yield on investment to reflect G-sec yields. Interest expenses to assets 8.4% 6.7% 6.9% 5.8% Cost on borrowings 11.4% 9.4% 11.3% 9.3% Net Interest Income 10.1% 11.0% 9.2% 10.1% Other Income 1.9% 1.9% 1.6% 1.6% Total Income 12.0% 12.9% 10.8% 11.7% Operating Expenses 6.1% 7.6% 5.7% 7.2% Operating Profits 5.9% 5.3% 5.1% 4.5% Provisions 0.5% 0.5% 0.9% 0.9% PBT 5.4% 4.8% 4.1% 3.5% Tax rate 34.8% 34.8% 35.7% 35.7% Reduction in cost of borrowings due to access to new avenues of funds as SFBs. Increase in opex is reflective of the branch expansion plan and SFB conversion related costs in the first year. RoA (%) 3.5% 3.2% 2.7% 2.3% RoA is likely to see an impact of ~40bps. Leverage (x) While we have kept the leverage unchanged here. Increase in leverage would support RoE over the medium term. RoE (%) 18.3% 16.4% 13.3% 11.4% RoE is likely to see an impact of ~200bps. Cost to income 50.8% 58.7% 53.0% 61.7% As seen in the table above, for Equitas and Ujjivan, while SLR impact and high opex related to conversion to banks will adversely impact profitability, this will be partly offset by benefits from lower cost of wholesale funding. SFBs will benefit from newly available avenues like CDs, CPs, bulk deposits and refinance to raise funds at lower cost, compared to what borrowings/bank loans they raised as NBFCs. Both the entities are likely to see RoA/ROE compression of ~40bps/200bps purely due to conversion to SFBs. Beyond the SFB conversion, we also expect some normalization in credit cost from current cyclical lows and some NIM compression due to competition and change in loan mix towards lower yielding assets. In all, we expect RoEs of Ujjivan and Equitas to compress by ~ bps over FY17-18 before normalization in opex leads to RoE recovery from FY19. RoEs of Ujjivan and Equitas to compress by ~400bps before normalization in opex leads to RoE recovery from FY19. September 22, 2016 Ambit Capital Pvt. Ltd. Page 13

14 BFSI RoE to recover closer to ~18% by FY22E Diversification away from microfinance Eight of ten SFB license recipients have microfinance-heavy loan books (~60% of loan book), which expose these companies to concentration risk. While we have, in the past, discussed (in the notes dated 17 th March, 2016 & 30 th March, 2016) our concerns on sustainability of currently strong growth and profitability for Indian microfinance industry, there is no denying the fact that current trends for this asset class have been strong on both growth and asset quality. The periods of such booming growth see trends in repayments and operational efficiencies also becoming favorable. However, we don t believe a structural fix for high pro-cyclicality (everything looks good when the going is good and everything unravels as quickly if the cycle turns) in the microfinance business has been found. The world over, MFI lending sees cycle of boom and bust. In India too, even if the APlike deep crisis episodes can be averted (due to better regulation and industry coordination), it is unlikely that high cyclicality in microfinance has been eradicated. Unsurprisingly, SFBs are looking to diversify their lending away from microfinance. Ujjivan in its FY14 annual report highlighted that. At each of our branches, we have successfully disbursed group loans and we may be serving 5,000-7,000 families. However, in the same urban work-areas, there may be up to 100,000 families who are financially excluded. With group loans we have achieved a mere 5-7% penetration. Our objective in the next five years is to increase our penetration to 50%. We can do so by widening the range of our loan products, which meet the specific needs of different sub segments of the financially excluded population. Hence, the need of the hour is to innovate and design need specific products and processes to build a viable business on the Individual Lending (IL) platform. Entities converting to SFBs also indicate moderate growth in MFI book The cyclical nature of microfinance is said to be behind MFI companies quest for banking licenses. Thus, while overall growth of the MFI industry is rapid, the companies converting to SFBs are already indicating a moderation in MFI book growth. After growing in its MFI book at 53% YoY in FY16, Equitas now expects 25-30% loan growth for MFI AUM in the coming years. Similarly, after 57% YoY growth in FY16, Ujjivan is indicating a much sharper slowdown in MFI loan growth (sub-20%) for a year or two as it undergoes the transformation to an SFB. So, we expect Equitas and Ujjivan s MFI AUM growth to be slower (20-30% YoY) over FY17-21 than what was seen in recent years (50-60% YoY). Besides, concerns on over-heating of the MFI industry have also begun to re-surface. As a large part of the MFI industry shifts to banking, there is uncertainty on how market practices will evolve around safeguard rules (e.g., ceiling of two MFIs lending to one customer). Thus, emerging conservatism among companies partly also appears to be driven by fear of onset of a negative cycle in the broader microfinance industry. Equitas and Ujjivan recently in their results announcement also cautioned However, we are concerned with over-leveraging by clients in certain pockets and more so given that 70% of the MFI market comprising of eight SFBs & Bandhan will move away from the two MFI per client norm shortly. To mitigate this risk, there is some discussion amongst the industry players to see whether loans from Banks can also be counted for the purpose of this two MFI rule. However, the progress is slow. Equitas MD&CEO after 1QFY17 results announcement Risk mainly comes from operation. I mean large part of our risk is operational in nature and also it is because of some manmade or some natural calamity. So if we are able to geographically diversify our portfolio and if we are well spread than things going wrong in one place or two does not affect us in our overall portfolio. Ujjivan MD&CEO after 4QFY16 results announcement Even if the AP-like deep crisis episodes can be averted (due to better regulation and industry coordination), it is unlikely that high cyclicality in microfinance has been eradicated. Equitas expects 25-30% loan growth; Ujjivan indicates a slowdown too in MFI loan growth (sub-20%) for a year or two. September 22, 2016 Ambit Capital Pvt. Ltd. Page 14

15 BFSI SFBs early demonstration of diversification The existing NBFC businesses of upcoming SFBs have strong focus on regional lending or a customer niche. While microfinance currently accounts for the bulk of their loan books, these companies have also tried to extend understanding of their customer segment to personal loans, small business loans and vehicle loans. Our discussion with most upcoming SFBs indicate that, in their avatars as SFBs, these companies are looking to add 2-3 asset classes (business loans, home loans, personal loans) to bring more diversification and scale in their assets side businesses. As SFBs, companies will look to add 2-3 asset classes (e.g. business loans, home loans, personal loans) Exhibit 15: Equitas has diversified from MFI-only book and has added new asset classes Exhibit 16: Ujjivan too has begun to add asset classes such as MSE and housing loans 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 100% 88% 76% 60% 53% 54% FY11 FY12 FY13 FY14 FY15 FY16 Vehicle MSE Housing MicroFin 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 95% 96% 93% 90% FY13 FY14 FY15 FY16 Individual loans Group loans ; MSE is business loans The management guidance also indicates that these companies will be deemphasizing their microfinance books more and more in coming years. Over the coming years, Equitas expects vehicle, MSE/housing and microfinance to account roughly equal (i.e. one-third each) of the total loan book. Similarly, Ujjivan expects individual loans to grow to ~50% of total AUM over the next five year or so. We do not expect this trend of changing AUM mix to lead to significant pressure on AUM yields as yields for individual business/vehicle/housing loans (18-20%) are comparable to MFI loans (20-23%). This is so because the customer segment is same in terms of risk profile and security structure (lack of collateral or ease of realization). Exhibit 17: MFI to account for ~40% (currently 52%) Equitas AUM in five years (by FY22E) of Exhibit 18: MFI to account for ~58% (currently 87%) Ujjivan s AUM in five years (by FY22E) of non-mfi, unsecured, 5% Equitas non-mfi, unsecured, 12% Ujjivan MFI, 40% non-mfi, secured, 55% non-mfi, secured, 30% MFI, 58% The changing AUM mix will, however, have positive impact on cost-to-income as the ticket sizes in the non-mfi category are higher (for example, MSE loan of `150k vs MFI loans of `30k). September 22, 2016 Ambit Capital Pvt. Ltd. Page 15

16 BFSI Liabilities: Operating model is the differentiator SFBs can ensure funding from market borrowings and bulk deposits in the near to medium term to replace bank borrowings so as not to cause disruption in lending during the phase of conversion to SFB. Further, long-term benefits could accrue if they are able to also build a stickier (lower cost, if possible) retail liabilities franchise. To this end, SFBs will have to make inroads into the Indian deposits market. A differentiated operating model (frontline-staff-led doorstep banking) compared to banks allows SFBs to serve their customer segment better and offer opportunity to take deposits market share from PSU banks in non-urban centers. Further, SFBs could combine their deep understanding of and relationships with their customer segments, a highly flexible operating model (doorstep banking and active use of familiar banking correspondents), and bank license to create their own deposit bases which so far have not been captured by the formal banking channel (cash/chit funds). Exhibits 19 and 20 show the distribution of Indian bank deposits by geography and bank type. While urban centres dominate overall deposit base (exhibit 19), rural centres have meaningful contribution in low-cost CASA deposits. Given their overall size, PSU banks dominate the Indian deposits market (exhibit 20). This dominance is more pronounced in rural centres even as private banks have made a serious dent in the urban deposits market. Wholesale funding to ensure that there is no funding disruption in near term. Taking market share from PSU banks and informal segment is key opportunity. Exhibit 19: Urban (including metro) centres dominate deposits market Exhibit 20: PSU banks dominate deposits market, particularly in rural centres Rural Urban PSU Banks Private Banks Others Banks 100% 80% 60% 65% 81% 100% 80% 60% 10% 7% 12% 22% 40% 40% 78% 71% 20% 0% 35% CASA 19% Term Deposits 20% 0% Rural Urban Source: RBI, Ambit Capital research; Note: Data as on FY15 Source: RBI, Ambit Capital research; Note: Data as on FY15 Thus, the long-term liabilities-side challenges for new SFBs are threefold 1) get some market share from strong incumbent banks in urban centres; 2) make a dent in PSU banks deposits franchise in rural and semi-urban areas; and 3) create a deposits franchise of their own by effecting market shift from informal savings/investment market to SFBs. 1- Competing in urban and metro areas don t bet on this yet This is one of the toughest pieces in the liabilities game. Thanks to strong inroads on the back of superior technological offerings and innovative products (cash management, transaction services, and corporate salary packages), incumbent banks are strongly placed to protect their retail liability franchise. The only tool left for SFBs to compete on, therefore, is higher interest rates. It is highly likely that SFBs will offer a material premium in interest rates for their savings and retail term deposits. The examples of Bandhan Bank, IDFC Bank and mid-size private sector banks (Kotak Mahindra Bank, Yes Bank and IndusInd Bank) show that there is significant correlation between the pace of CASA deposit mobilization and interest rates (see Exhibit 21). SFBs will likely offer a material premium in interest rates for their savings and retail term deposits. September 22, 2016 Ambit Capital Pvt. Ltd. Page 16

17 Exhibit 21: New banks offered higher interest rates on savings and term deposits and hence were able to outgrow system deposit growth Total Deposits Deposit Mix (FY16) 3 Years CAGR (FY14-16) Rate of Interest FY16 - ` bn Current A/c Savings A/c Term Deposits Savings A/c Term Deposits Savings A/c^ BFSI Term Deposits* Kotak Mahindra Bank 1,386 17% 21% 62% 60% 33% 5%-6% 7.50% Yes Bank 1,117 10% 18% 72% 50% 14% 6%-7% 7.75% IndusInd Bank % 19% 65% 35% 16% 4%-6% 7.75% IDFC Bank 82 4% 1% 95% NA NA 4% 8.25% Bandhan Bank 121 2% 20% 78% NA NA 4.25%-5% 8.25% Banking system 101,600 9% 27% 64% 14% 10% ; Note: ^Historical rates offered; *Current term deposit rates. 2 - Nibbling on PSU banks deposit base in non-urban centres challenging but possible Attacking PSU banks deposits market share in non-urban centres could be an easier strategy in the overall liabilities game plans of new SFBs. We have seen how new private sector banks strongly dented PSU banks dominance in deposits market share in urban centres (Exhibit 23). However, given metro/urban centres dominate total deposits pie, this same competition to PSU banks was not extended to PSU banks domination in rural and semi-urban areas. Exhibit 22: Composition of the deposits market in rural centres Exhibit 23: Composition of the deposits market in urban centres PSU Banks Private Banks Others Banks PSU Banks Private Banks Others Banks 100% 80% 9% 11% 11% 12% 10% 7% 8% 9% 9% 12% 100% 80% 11% 9% 7% 7% 7% 7% 14% 22% 20% 22% 60% 60% 40% 85% 81% 80% 79% 78% 40% 82% 77% 72% 73% 71% 20% 20% 0% FY95 FY00 FY05 FY10 FY15 0% FY95 FY00 FY05 FY10 FY15 Source: RBI, Ambit Capital research Source: RBI, Ambit Capital research Given private sector banks are still, to a large extent, focused on urban and metro centres and PSU banks don t have the right operating model to serve customers in rural and semi-urban centres, these areas remain under-served. This offers a good opportunity for upcoming SFBs to gain deposits market share from PSU banks in these centres. SFBs can use a combination of better doorstep services, assets side relationships and higher interest rates to effect this shift. 3 Creating a new deposits market can create genuine differentiation This could be the most differentiated piece in deposits mobilization by SFBs. SFBs could combine their deep understanding of their customer segments, a highly flexible operating model, and bank license to create their own deposits base which has so far not been captured by the formal banking channel. The data below (exhibit 24) illustrates that the major elements of financial savings outside formal financial systems are cash at home (13% of savings, more so in rural market) and SHG (self-help groups)/chit-funds (2% of savings). As statewise data in exhibit 25 shows, in some states like Kerala, Andhra Pradesh and Tamil Nadu, the share of financial savings in SHGs/chit-funds is higher. Through combination of better doorstep services, assets side relationships and higher interest rates, SFBs can gain deposits market share from PSU banks. SFBs can create their own deposits bases, which so far have not been captured by formal banking channel, at all. September 22, 2016 Ambit Capital Pvt. Ltd. Page 17

18 BFSI Exhibit 24: Distribution of investment (financial) by household (%) Rural Urban All India Bank deposit PPF/PF Insurance Cash at home Post office Stock market SHG/ Chit Fund Other Source: NCAER NSHIE ; Note: PPF=Public Provident Fund, PF=Provident Funds, SHG=Self Help group Exhibit 25: Distribution of investment by households financial assets for by state (%) Cash at home Bank deposit Post Office PPF/PF Stock market Insurance SHG/ CF Other Jammu & Kashmir Himachal Pradesh Punjab Uttaranchal Haryana Delhi Rajasthan Uttar Pradesh Bihar Assam West Bengal Jharkhand Orissa Chhattisgarh Madhya Pradesh Gujarat Maharashtra Andhra Pradesh Karnataka Kerala Tamil Nadu Other States NE States All-India Source: NCAER NSHIE ; Note: PPF=Public Provident Fund, PF=Provident Funds, SHG=Self Help group, CF=Chit Funds Operating model key to capture savings market share The key question is how new SFBs will be able to capture financial savings in the informal market and even from PSU banks in non-urban markets, given that existing banks have not been able to do so. The answer lies in the operating model doorstep banking and active use of banking correspondents helped by higher familiarity with the customer segment due to SFBs asset-side relationships. The exhibit below shows the relationship between the required operating model and customer segments served by different bank and non-bank lenders. We have used employee expenses per employee as the determinant of the operating model. Lower employee expenses per employee, for example, show high use of frontline staff to reach out to customers at their doorstep. Placement of lenders across income as % of assets shows the segment they operate in. Low income yield indicates large-ticket corporate and commercial lending while high yield indicates small-ticket granular lending. Thus, players such as Equitas and Ujjivan clearly have differentiated operating models where banks do not compete. As these companies become banks they would have the distinct advantage of servicing their customer segments as NBFCs with similar operating models can t offer liabilities services. A suitable operating model doorstep banking and active use of banking correspondents could help SFBs. There is strong correlation between customer segmentation (income yields) and operating model (employee expense per employee). September 22, 2016 Ambit Capital Pvt. Ltd. Page 18