Portfolio Outstanding & Active Accounts
Mar 2020 – Mar 2026 · POS in ₹ Lakh Crore · Accounts in Lakh
₹16.5 L Cr
POS Mar-26
+12.9% YoY
1,224 L
Active Accounts Mar-26
-4.1% YoY (from Dec-25 peak)
₹1.35 L
Avg Outstanding / Account
vs ₹0.56L (Mar-20)
3.1x
POS growth FY20→FY26
4x account growth same period
Portfolio Outstanding (₹ L Cr) vs Active Accounts (Lakh) — Mar 2020 to Mar 2026
Account count peaked in Dec-25 at 1,303L — a slight pullback in Mar-26 signals lender caution on unsecured exposure
Originations — Value & Volume
Annual FY2020–FY2026 · Value in ₹ Crore · Volume in Lakh loans · Annual series
📈
FY26 originations: ₹11.45L Cr / 1,726L loans — biggest single-year jump. Originations/POS ratio of 69% vs gold's 163% confirms this is a genuine 18–24 month term product, not a revolving facility. The risk pool renewal cycle is multi-year, meaning stress from any one origination cohort takes 2–3 years to surface fully.
Annual Originations Value (₹ L Cr)
FY20–FY26 · 14x growth in 6 years
Volume (Lakh loans) & Avg Ticket Size (₹K)
Volume 4x in 6 years; ticket sizes compressing sharply
Average Ticket Size — The Fintech Effect
ATS trend from FY2021 through Q1 FY2026 · ₹ (absolute)
⚠️
Ticket size halved in 5 years: From ₹1.25L (FY21) to ₹58.5K (Q1 FY26). Fintechs and BNPL platforms drove millions of sub-₹50K personal loans for consumption. This is not a sign of broader access — it's a different credit product entirely. The average ₹58.5K loan at 18–24% interest costs the borrower ₹8–12K in total interest, funding a phone purchase or travel. The concentration risk is in how many of the 1,726L FY26 originations were to overleveraged thin-file borrowers.
Average Ticket Size at Origination (₹)
Steep secular decline driven by fintech volume, not traditional bank growth
Asset Quality — A Split Picture
PAR 31-90 stable · PAR 180+ rising persistently — two different parts of the book telling different stories
🔴
PAR 180+ at 5.3% and rising — the long-tail delinquency bucket that doesn't resolve. PAR 31-90 at 1.6% looks clean, but PAR 180+ has climbed from 2.4% (Mar-20) to 5.3% (Mar-26) without a break. This means stressed accounts are building up in the deep delinquency bucket faster than they're being resolved through write-offs or recoveries. For every ₹100 of personal loan outstanding, ₹5.30 is 6+ months overdue.
PAR Trend — All Buckets Mar 2020 to Mar 2026
31-90 stable; 180+ rising every year — a structural stress signal
PAR 31–90% Across Products (Mar-26)
Personal loans near the middle — but 180+ bucket is an outlier
| Period | PAR 31–90% | PAR 91–180% | PAR 180+% | Combined 31–180 |
|---|---|---|---|---|
| Mar-2020 | 1.6% | 0.8% | 2.4% | 2.4% |
| Mar-2021 | 2.3% | 1.3% | 3.2% | 3.6% |
| Mar-2022 | 2.2% | 0.8% | 3.2% | 3.0% |
| Mar-2023 | 1.9% | 0.8% | — | 2.7% |
| Mar-2024 | 1.9% | 0.9% | — | 2.8% |
| Jun-2024 | 1.8% | 1.1% | 4.5% | 2.9% |
| Mar-2025 | 2.0% | 1.1% | 4.7% | 3.1% |
| Sep-2025 | 1.6% | 1.1% | 5.6% | 2.7% |
| Mar-2026 | 1.6% | 0.8% | 5.3% | 2.4% |
Geographic Distribution — Q4 FY26
Personal loans have strong BT100 penetration — but also highest concentration risk in metro overleveraging
Originations Value by City Tier — Q4 FY26
BT100 cities contribute ~42% — digital channels enabling rural reach
BT100 Share — Personal Loans vs Other Products
Among highest BT100 penetration after CD loans — fintech reach effect
Hidden Insights
Non-obvious findings from cross-year pattern analysis — critical for unsecured retail strategy
Insight 01 · Delinquency Structure
The 31-90 PAR Is a Distraction — Watch the 180+ Bucket
PAR 31-90 at 1.6% looks stable. But PAR 180+ has risen every single year: 2.4% (FY20) → 3.2% (FY22) → 4.7% (FY25) → 5.3% (FY26). This is the "zombie cohort" — accounts that moved past early delinquency and never recovered. In a portfolio growing at 13% annually, a rising deep-delinquency rate means write-offs are not keeping pace with new stress formation. At ₹16.5L Cr portfolio, 5.3% PAR 180+ = ₹87,500 Cr of deep delinquency — larger than the entire two-wheeler loan portfolio.
PAR 180+: 2.4% (FY20) → 5.3% (FY26) — every year higher
Insight 02 · Product Transformation
The Personal Loan Product of 2026 Is Not the Same as 2020 — Ticket Compression Changes the Risk Model
In FY21, the average personal loan was ₹1.25L — a meaningful credit facility for home improvement or medical emergency. By Q1 FY26, it was ₹58.5K — a consumption top-up. These are fundamentally different credit profiles: different purpose, different repayment discipline, different borrower cohort. Lenders benchmarking current PAR against pre-FY23 history are comparing apples to oranges. The ₹50K-and-below loan cohort (likely 40%+ of FY26 volume) has no comparable stress history to validate credit models against.
ATS decline: ₹1,25,000 (FY21) → ₹58,500 (Q1 FY26) = -53%
Insight 03 · Volume vs Borrower Count
1,726 Lakh Loans in FY26, But Active Accounts Fell — Borrowers Are Churning, Not Growing
FY26 originated 1,726L loans but active accounts at Mar-26 (1,224L) are actually down from Dec-25 (1,303L). This means the incremental originations are going to existing borrowers (repeat loans, rollovers) rather than new-to-personal-loan borrowers. The addressable new borrower pool is contracting. Lenders reporting "strong originations" may be recycling existing credit-stressed borrowers at a faster pace — a sign of market saturation rather than credit expansion.
FY26 orig volume: 1,726L loans vs 1,303L peak accounts = 1.3x churn ratio
Insight 04 · Covid Cohort Hangover
The FY22-FY23 Origination Vintages Are Still Seasoning — PAR 180+ Has 18 More Months to Peak
Personal loans originated in FY22-23 (when volume jumped from 728L to 1,065L — a 46% YoY surge) are now entering the 36-48 month delinquency window where deep defaults crystallize. Typical personal loan stress peaks at 30-36 months post-origination. The FY22-23 cohort will hit that window in FY25-26. This mechanically explains the rising PAR 180+ and suggests it has not peaked. Provisioning models built on pre-FY22 vintage experience will underestimate forward credit costs.
FY22-23 vintage: +46% volume surge now entering peak-stress window
Insight 05 · Market Structure
Originations Recovering After FY25 Dip — The Correction Was Real, The Recovery Is Selective
FY25 originations (₹8.88L Cr) dipped from FY24 (₹9.08L Cr) — the first YoY decline in the decade-long data series. This was a deliberate pullback by banks and NBFCs responding to RBI risk-weight increase (Nov-23) on unsecured consumer credit. FY26's 29% recovery (₹11.45L Cr) shows selective re-entry: larger ticket, better credit score, existing-customer focus. The mid-market thin-file borrower segment — which drove the FY18-FY24 volume surge — remains effectively locked out of the formal credit system as lenders recalibrate underwriting.
FY25 originations dip: first-ever decline. FY26 recovery: +28.9% but quality-skewed