Key Takeaways
- NIM jumped 60 bps QoQ to 6.4%, driven by sustained spreads and cash optimization, even as the company continued its branch-led expansion push.
- Asset quality is healing loan-by-loan: 90+ DPD fell ~30 bps, collection efficiency is back to pre-COVID levels, and the company is resolving delinquencies through property sales rather than ARC fire sales.
- Rate hikes are a pass-through story, not a demand risk: Management expects to pass on 50–70 bps of higher borrowing costs gradually and sees no muting of demand from genuine end-user homebuyers.
- The expansion engine is scaling up: Physical branches, virtual locations, and connector productivity are all rising in parallel, with an ambition to double the town footprint to 400 in 2–3 years.
- What to watch next quarter: Whether the co-lending partnership with Union Bank moves past teething issues to meaningful volumes, and how quickly the behavioral cheque-bounce gap narrows as repair work takes hold.
Margin & Spread Dynamics
- NIM expansion levers: The 60 bps QoQ jump to 6.4% came from sustained on-book spreads and further optimization of cash sitting on the balance sheet.
- Spread sustainability thesis: As the company expands deeper into smaller towns over the next 2–3 years, it expects to sustain spreads, with a creeping mix shift toward higher-yielding loan against property products providing additional headroom.
- Rate hike pass-through mechanics: Against a 40 bps repo hike, roughly 50% is expected to pass through over the next two quarters as MCLR resets, with an overall borrowing cost increase of 50–70 bps anticipated this year. The company will pass this on to customers gradually, and sees the yield floor rising (e.g., an 11% floor moving up), which narrows the band and helps bridge spread compression.
- Securitization income stability: On the new book, securitization spreads are in line with on-book spreads with no pricing arbitrage. On the existing book, earlier securitization was done at much higher prices, so that benefit continues and management does not foresee a P&L risk from the legacy securitization pool.
- Marginal cost of borrowing was flat at 7.2% overall, with the Q4 marginal cost at 5.8% due to an NHB drawdown; ex-NHB, the marginal cost stood at 7.5%.
Nutan Gaba Patwari: "We have done the model to see what is the risk, and at what point does the curve starts to give us more hit in the P&L. We are not foreseeing that because that given our securitization was at a much higher price earlier, we still continue to carry that benefit."
Asset Quality & Collections
- 90+ DPD improvement: The metric fell ~30 bps QoQ (post-RBI classification: 2.6% to 2.3%; previous NPA classification: 1.7% to 1.3%, a 40 bps decrease). The improvement is a lagged effect of 1+ and 30+ buckets reducing.
- The rollback overhang: Customers reclassified as NPA under the RBI circular are paying additional installments but must remain classified as 90+ until they reach zero DPD. This is an after-effect of COVID and will take time to fully clear, with the below-NPA pool expected to take 6–9 months to start reducing meaningfully.
- Resolution engine: Real improvements in delinquency are driven by property sales in a good market, with the entire resolution done in-house on a loan-by-loan basis. Over 200 properties were resolved or sold in the quarter, with write-offs of only ~₹2 Cr from partial write-offs on those sales.
- Customer payment behavior shift: More customers are slipping out of NPA by making more than one payment per month to return to zero DPD, a phenomenon expected to continue for 3–6 months. The company will handhold these rollback-case customers for as long as it takes (6, 9, or 12 months) rather than taking hard legal action, since they are actively trying to cure.
- Cheque bounce gap: The 9% gap between cheque bounces and 1 DPD (vs 6% pre-COVID) is explained by two behavioral shifts: ~3–4% of customers became casual about on-time payment during the moratorium and second wave, and another 2–3% is driven by digital payment ease, where customers bounce on the 4th and pay via UPI on the 5th–8th. Repair work is underway, but the behavioral gap may sustain for a few quarters.
- Provisioning conservatism: Credit cost was lower at 0.2%, with ECL provision at 1.1% of total POS. PCR stands at 47.1%, but prior to the NPA reclassification per the RBI circular, PCR was at 83.6% (vs 69.1% in Q3 FY22). Management is comfortable guiding to pre-COVID credit cost levels of 30–50 bps for the coming year.
- Restructured book slippages: ~20% real NPA slippage and ~10% in reclassification were observed from the restructured pool.
- SARFAESI limitation: SARFAESI cannot be used on rollback cases because the NPA circular has been postponed to September, and even from October 1, old CIBIL reporting rules continue, so SARFAESI cannot be imposed on rollback cases under the current understanding.
Manoj Viswanathan: "Also, from the customer friendliness perspective it is not the right thing to do because these customers are actually trying hard to come out of delinquency and come back to zero DPD so does not make sense to take a hard legal action on them. So we would like to handhold these customers for as long as it takes maybe six, nine months or twelve months to get them back to zero."
Demand & Rate Sensitivity
- Genuine end-user demand is resilient: The customer base consists of genuine end-users whose planning for purchasing or building a house happens over a long period, so rate increases generally do not mute or subdue demand. Customers recalibrate their requirements when there is a marginal rate change rather than abandoning purchases.
- BT Out dynamics are structural, not cyclical: BT Outs flow to larger banks, nationalized banks, and large HFCs offering rates of 7.5%–8.5% for prime customers. The differential is high and difficult to retain, but this pattern has been unchanged for 10 years. As rates rise across the system, the excitement around very low rates (6.5%–7.5%) subsides, and BT Out is expected to stay in the 4%–5% range with a seasonal skew in March.
- Rate pass-through confidence: Management does not see a challenge in retaining spreads because ground-level demand is very strong, and the pass-through will be gradual over 3–6 months so there is not too much lag.
Manoj Viswanathan: "We do not see a much of a challenge in retaining the spreads, and because the demand on the ground is very strong and customers generally recalibrate their requirements when there is some marginal rate change. So, we do not really see a muting of demand or subduing of demand because of the rate hikes."
Expansion & Distribution
- Branch build-out: Physical branches increased to 80 from 76 QoQ, with total distribution points rising to 200 from 187. The company is adding 4–5 physical branches every quarter, alongside virtual branches and digital locations in larger numbers. The ambition is to reach 400 towns in the next 2–3 years, up from the current 200, with 100 small locations added in the previous year alone.
- Virtual-then-physical model: A virtual setup starts generating business first, and a physical branch is added only when AUM reaches ₹10–20 Cr. Smaller branches show higher AUM growth due to the base effect.
- Connector channel scaling: Active connectors are expected to increase by 100–200 per quarter, with a productivity push from ~2 to ~3 loans per connector per month. The sourcing channel mix is comfortable at 70:30 (connectors:other channels), and digital and alliance channels are anticipated to move up from the ~3% mark as COVID impact recedes.
- Geographic mix: Maharashtra and Karnataka are expanding into smaller towns, with progress in the South and other markets currently sharper and offsetting Maharashtra's progress. The top-five-state disbursement mix is likely to remain at 70%–75% with slight variation.
- Self-employed mix: As the company expands into smaller towns, the likelihood of the self-employed sector going up slightly more than historically exists, but the mix is currently still in the 70:30 range.
Borrowing & Liquidity
- Rating upgrade progress: India Ratings assigned AA- with Stable Outlook to the long-term rating this quarter. Work has started on aligning the ICRA rating (currently A+ positive), and once aligned, the benefit from banks will start to pass through.
- NCD program initiated: The company filed an initial declaration with stock exchanges for 25% incremental borrowing through NCDs and has started the process of placing NCDs with mutual funds. A two-year block is allowed, and borrowing on the incremental portion will start this year.
- NHB refinance headroom: Sufficient headroom is available based on the internal formula linked to networth.
- Liquidity posture: The company will hold about two months of liquidity (~₹650 Cr) as an optimal level and will not run it down totally. Despite rising T-Bill and G-Sec yields, the overall comfort and ability to price well continues.
- Co-lending with Union Bank: Decent progress has been made but teething issues persist; substantial progress is expected this quarter, with more positive news anticipated on the next call. Co-lending will be accretive to the overall margin profile, with income likely in the other income line since assets are off-balance sheet.
Operating Costs & Efficiency
- Opex to Assets rose 20 bps QoQ to 3.0%, and cost-to-income increased to 35.7% from 33.0% QoQ, driven by people addition and compensation (including ESOPs in the prior quarter) with no one-off or variable pay element. The metric is expected to remain in the 3.0%–3.2% range going ahead as the company focuses on expansion.
- Centralized underwriting conviction: Management remains a strong believer in the centralized model and does not plan to localize it, arguing that information arriving in digital format is strengthening the ability to underwrite centrally.
- Digital adoption traction: Customer app registration reached 80% (vs 76% in Q3), e-stamp adoption at 65%, e-NACH at 51%, and e-sign at 20%. The Connector App was enhanced with easier usability and custom notifications.
- Ticket size shift: A secular increase in ticket size is underway, not a conscious decision; some co-lending program loans may be showing up in higher ticket sizes.
Manoj Viswanathan: "We are strong believers and we are also getting the benefits out of it so we do not plan to localize it at all and in fact with a lot of this information coming in a digital format our ability to underwrite better through a centralized mode is becoming stronger."
Guidance & Commitments
| Commitment | Specifics | Timeline |
|---|---|---|
| AUM growth | ~30% growth, extrapolating ₹200 Cr+ disbursal run rate with slight H2 uptick | FY23 |
| Opex to Assets range | 3.0%–3.2% | Going ahead |
| Credit cost | Pre-COVID level of 30–50 bps | Coming year |
| ROE | Mid-teen ROEs | In a couple of years |
| Rate pass-through | 50–70 bps overall borrowing cost increase passed on gradually | This year |
| Branch expansion pace | 4–5 physical branches per quarter | Ongoing |
| Town footprint | Reach 400 towns (from 200 currently) | Next 2–3 years |
| Connector addition | 100–200 active connectors added per quarter | Ongoing |
| Below-NPA pool resolution | Customers moving into lower buckets and getting fully resolved | 6–9 months before pool starts reducing |
| Rollback customer handholding | Handhold customers to zero DPD without hard legal action | 6, 9, or 12 months as needed |
| BT Out range | 4%–5% with seasonal skew in March | Going forward |
| Co-lending progress | Substantial progress expected | This quarter |
Key Metrics
| Metric | Q4 FY22 |
|---|---|
| AUM | Crossed ₹5,000 cr |
| Annual disbursements | Crossed ₹2,000+ cr for the first time |
| Quarterly disbursements | ₹641 cr, up 12.5% QoQ and 42% YoY |
| Net Interest Income | Up 51.6% YoY and 15.6% QoQ |
| NIM | 6.4%, expanded from 5.8% in Q3 FY22 |
| PPOP | ₹66 cr |
| Adjusted PAT | ₹48 cr, up 4.9% QoQ |
| ROA | 4.0%, flat QoQ |
| Annualized ROE | 12.5% on Q4 numbers |
| AUM growth guidance | ~30% for FY23 |
| Gross Stage 3 | 2.3%, down 30 bps from 2.6% in Dec'21; adjusted for RBI Nov'21 circular: 1.3%, down 40 bps from 1.7% |
| 1+ DPD | 5.3%, improved from 6.5% |
| 30+ DPD | 3.7%, improved from 4.7% |
| Credit cost | 0.2% |
| Cost of borrowing | 7.2% flat; marginal COB 5.8% (due to NHB drawdown); ex-NHB 7.5% |
| Opex to Assets | 3.0%, up 20 bps QoQ |
| Cost to income | 35.7%, vs 33.0% in Q3 FY22 |
| BT Out | 6.5% of loans in Q4 |
| Physical branches | 80, up from 76 QoQ; total distribution points 200, up from 187 |
| CRAR | 58.6% total; Tier 1 58.0% |
| Networth | ₹1,574 cr vs ₹1,381 cr as on Mar'21 |
| Book Value per share | ₹179.6 as on Mar'22 |
| Restructured book | ₹28 cr, 65 bps of overall book |
| 1 DPD on FY22 originations | 0.7% |
| PCR | 47.1%; prior to NPA reclassification per RBI circular: 83.6% vs 69.1% in Q3 FY22 |
| ECL provision | 1.1% of total POS |
| Write-offs | ~₹2 cr (partial write-offs from properties sold to settle accounts) |
| LCR | 125% vs requirement of 30% |
| Cash and investments held | ~₹650 cr, optimal level of about two months of liquidity |
| Direct Assignment volume (FY22) | ₹465 cr vs ₹300 cr last year |
| Loan against property share of AUM | 7% |
| Self-employed vs salaried mix | 70:30 range |
| Floating rate loan share | ~90% |
| Active connectors per quarter | ~1,500, increasing by 100–200 per quarter |
| Sourcing channel mix | 70:30 (connectors: other channels) |
| Top 5 states disbursement share | ~70%–75% (Gujarat, Maharashtra, Tamil Nadu, Andhra+Telangana) |
| e-stamp adoption | 65% of loans in Q4, 41% in FY22 |
| e-NACH adoption | 51% of loans in Q4, 38% in FY22 |
| e-sign adoption | 20% of loans in Q4, 16% in FY22 |
| 48-hour TAT for loan approval | 92% of customers, up from 90% in Q3 FY22 |
| Customer app registration | 80% of customers, up from 76% in Q3 FY22 |
| Payments via app growth | Up 82% YoY |
| Service requests via app growth | Up 42% YoY |
| Borrowing mix | 45% Banks (22% public, 23% private), 27% NHB Refinance, 23% Direct Assignment; zero Commercial Paper |
| Credit cost guidance | 30–50 bps for FY23, pre-COVID level |
| Disbursement run rate | ~₹200 cr per month |
| Cheque bounce rate | 14%–15% vs pre-COVID 10%–11% |
| Slippages from restructured book | ~20% real NPA slippage, ~10% in reclassification |
| Securitization spread guard | ~8% |
| Branch addition pace | 4–5 physical branches per quarter |
| New branch AUM growth (first 2 years) | 75%–100% |
| Vintage branch AUM growth | 20%–30% |
| Target town profile | 2–3 lakhs population, demand for 300–400 loans per month |
| Current towns | 200, ambition to reach 400 in 2–3 years |
| Yield operating range | 11%–13.5%, blended 12.7%–13% |
| Share of ₹0.15–0.25 cr loans and higher ticket size | Up 3% points over last one year |
| Maharashtra and Karnataka combined share of book | ~24%–25%, growing below overall portfolio growth |
| Properties sold/resolved in Q4 | More than 200 |
| Stage 3 breakup (Dec'21) | ₹102 cr total: ₹68 cr actual NPAs (1.7%) + ₹34 cr below 90-days RBI classification (0.9%) |
| Stage 3 total (Mar'22) | ₹101 cr |
| RBI NPA pool increase (rollback cases <90 DPD) | ₹10 cr increase |
| Connector productivity | ~2 loans per month, target to push to 3 |
| Direct marketing and other channels sourcing share | ~3% for last few quarters |
| FY22 disbursements | ~₹2,000 cr |
| Cheque bounce vs 1 DPD difference | ~9% currently vs ~6% pre-COVID (Q4 FY20) |
| Customers paying immediately after bounce date | 3%–5% |
| Below-NPA pool (rollback cases) | ₹44 cr (referenced as ₹57 cr in movement context) |