IREDA — Pure-play green lender, growth premium meets its first NPA scare
Indian Renewable Energy Development Agency Ltd
1. Snapshot
IREDA is the Financing node of the Indian power value chain, but pointed at one slice of it: it is the country’s only dedicated, pure-play green-financing NBFC — a Navratna PSU, 71.76% owned by the Government of India, lending exclusively to renewable energy and energy-efficiency projects. Market cap ~₹35,719 Cr, price ~₹127, 52-week range ₹175 / ₹109, P/E ~19.1, P/B ~2.6× (book value ₹49.1). The animal: a specialist infrastructure-finance company (RBI classifies it as an IFC) — a borrow-wholesale, lend-to-RE-projects spread business that listed only in November 2023 and is still being priced as a growth story rather than a mature lender. As of 2026-06-20, from screener snapshot.
2. Business & position in the value chain
IREDA does one thing: it borrows money (bonds, term loans, lines from multilaterals and the bond market) and on-lends it to renewable energy projects — solar, wind, hydro, transmission tied to RE, equipment manufacturing, and energy-efficiency work — across the full project life, from conceptualisation through post-commissioning. It is the rupee plumbing under India’s clean-energy build-out: where PFC and REC finance the whole power sector (thermal, T&D, discoms, everything), IREDA is the pure-play RE financier. That purity is the entire investment debate — it is maximal exposure to the single fastest-growing node in Indian power, with no thermal or discom legacy to dilute it, but also no diversification when one RE borrower goes bad.
The loan book is ~₹93,075 Cr at end-FY26, up 22% YoY from ₹76,282 Cr (SaurEnergy). Disbursements were ₹34,946 Cr (+16%), sanctions ₹51,883 Cr (+9%). The income statement is a spread business: ₹8,310 Cr revenue against ₹4,905 Cr interest cost in FY26 (snapshot Annual P&L). Concentration risk is structural and double-edged: 100% RE exposure, a book skewed toward large project loans, and — as FY26 demonstrated — meaningful single-name exposures (Gensol, Hetero) that can move the GNPA line by a full point on their own.
3. Management & promoter quality
Promoter is the Government of India via the Ministry of New and Renewable Energy (MNRE), holding 71.76% after the FY26 QIP (down from 75% — see §6). The CMD is Pradip Kumar Das, in the chair since 2020, who took IREDA through its 2023 IPO and its 2024 upgrade to Navratna status (Wikipedia; Indian Bureaucracy). Navratna matters operationally: it gives a PSU more autonomy over investment and fundraising decisions without case-by-case government sign-off — useful for a lender that needs to move at the pace of the RE build-out.
The post-IPO track record reads well on growth and, until FY26, on credit: IREDA likes to point out it has written off only ~₹135 Cr cumulatively across nearly four decades of lending (Renewable Watch). That is a genuine underwriting credential, but it is also a number from the pre-listing era; the FY26 GNPA jump is the first real stress test of that discipline under public-market scrutiny.
PSU promoter dynamics cut both ways. Positives: sovereign backing keeps IREDA’s cost of funds low and its bond access deep, and the government wants this lender to grow because it is an arm of national RE policy. Negatives, all standard for a listed PSU: OFS / dilution overhang (the government has explicitly capped its own dilution to ≤3.76% of post-issue capital but is willing to dilute to fund growth — Business Standard); the company is an instrument of policy first and a profit-maximiser second; and governance/board independence is what it is for a government NBFC. No pledging concerns (sovereign promoter), and related-party noise is minimal. Even-handed read: competent, mission-aligned management with a clean long-run loss record, now being judged on whether that record survives a faster, larger, public-market book.
4. Financial trends
The growth has been genuine and the returns respectable for a lender, but the cash-flow and leverage shape is exactly what you’d expect of a young, fast-growing NBFC — it is consuming capital, not throwing it off.
| Metric | FY24 | FY25 | FY26 |
|---|---|---|---|
| Revenue (₹ Cr) | 4,965 | 6,755 | 8,310 |
| Net profit (₹ Cr) | 1,252 | 1,698 | 1,874 |
| Loan book (₹ Cr) | ~50k | 76,282 | 93,075 |
| ROE % | 15% | 18% | 16% |
| GNPA % | 2.66 (Q4) | 2.45 | 3.49 |
| Net NPA % | ~1.5 | 1.35 | 1.29 |
| Borrowings (₹ Cr) | 49,687 | 64,740 | 77,846 |
| Net worth (₹ Cr) | ~8,560 | 10,266 | 13,781 |
Sources: snapshot _snapshot.json; GNPA/net worth from SolarQuarter.
- ROE / ROCE — ROE 16% in FY26 (18% FY25, 15% FY24), healthy for an NBFC and improving on the leverage cycle, though it dipped a touch as the QIP added equity and provisioning bit. Screener’s ROCE of 8.69% is the misleading-for-a-financier number — for a lender, cost of funds sits inside “operating,” so ROCE understates the economics; judge on ROE and spread instead.
- Growth — revenue ~5x and the loan book ~3x over five years; PAT from ₹346 Cr (FY21) to ₹1,874 Cr (FY26). Loan-book growth has decelerated from ~28% (FY25) to 22% (FY26) — still fast, but the law of large numbers and a more cautious post-Gensol posture are showing.
- Margins — financing margin ran 32–35% through FY24–25, then wobbled to 16% in Q1 FY26 as the Gensol/Hetero provisioning landed (note the ₹427 Cr expense spike that quarter in the snapshot), before recovering to 34% in Q2–Q3 and settling at 29% in Q4. The Q1 air-pocket is the whole asset-quality story in one line.
- Balance sheet — borrowings ₹77,846 Cr against net worth ₹13,781 Cr: a leveraged book, as every spread lender is. The FY26 QIP lifted net worth ~34% and shored up capital adequacy ahead of the next leg of growth. Screener’s “low interest coverage” con is a category error for an NBFC — interest is the raw material, not a red flag.
- Cash flow — operating cash flow has been deeply negative every year (−₹14,477 Cr FY26), which for a growing lender is mechanically correct: disbursing loans shows up as operating outflow, funded by the financing line (+₹14,748 Cr). The CFO/OP of −192% is not a quality signal, just an artefact of the NBFC model. Dividend payout resumed at 20% in FY26 (final dividend ₹0.75/share for FY26 plus interim) after a multi-year pause to retain capital for growth (Business Standard).
Financier’s lens (book × spread × asset quality): the book is compounding at 22%, the spread is healthy in the low-30s% financing-margin terms once you strip the one-off provisioning quarter, and asset quality is the variable that just turned. Two of three legs are strong; the third is the watch-item the rest of this dossier circles.
5. Latest quarter
Q4 FY26, reported end-May 2026. Net profit ₹492.63 Cr, down 1.8% YoY and ~16% QoQ off the ₹585 Cr December quarter — the first time IREDA’s headline PAT has declined year-on-year as a listed company (Business Standard; SaurEnergy). Revenue ₹2,175 Cr (+14% YoY). The drop was margin pressure plus provisioning, not a demand problem — disbursements and the book both grew.
For the full year: record PAT ₹1,874 Cr (+10%), revenue ₹8,310 Cr (+23%), loan book ₹93,075 Cr (+22%). The flagged detail is asset quality: GNPA 3.49% (₹3,245 Cr) vs 2.45% a year ago; net NPA 1.29% (₹1,172 Cr), actually down from 1.35% (SolarQuarter). That net-NPA-down-while-gross-NPA-up shape is the key tell: it means IREDA has been aggressively provisioning the bad accounts rather than carrying them lightly. One small momentum flag worth noting — a lender that prided itself on reporting within a week of quarter-close took noticeably longer this time (SaurEnergy).
6. What’s happening now
The live wire is the GNPA wobble and its two named culprits. (HARD) In Q1 FY26, gross NPA spiked to 4.13% from 2.45% — the single ugliest print since listing — driven by two accounts going bad: ~₹730 Cr of exposure to Gensol Engineering (the EV/solar group whose alleged fund-diversion fraud blew up in 2025) classified as NPA, and a second ~₹783 Cr borrower migrating from Stage 2 to NPA. Stage-3 assets jumped 77% QoQ to ~₹3,302 Cr (Analytics Insight; Renewable Watch). IREDA flagged “fraudulent activities” by Gensol and Hetero Renewable Energy and began recovery/legal action — and crucially has already recovered ~₹100 Cr-plus of its ~₹700 Cr Gensol exposure (Renewable Watch). (HARD) Through the year GNPA then improved sequentially — ~3.97% → 3.75% by Q3 → 3.49% at year-end — so the trajectory is a spike that is being worked down, not a runaway. Read it as a digestible, idiosyncratic, two-name event rather than book-wide rot — but it is the first crack in the “₹135 Cr in 40 years” halo.
(HARD) Capital raises. A ₹2,005.9 Cr QIP closed in June 2025 at ₹165.14/share, oversubscribed 1.34× (PIB); this is what cut promoter holding from 75% to 71.76%. The board then approved a further ₹2,994 Cr QIP in February 2026 to keep funding book growth, with the government’s dilution capped at ≤3.76% of post-issue capital (SolarQuarter; Business Standard). Expect periodic equity raises to remain part of the IREDA story — a growing lender needs capital, and a PSU funds it via QIP rather than debt-only.
(SOFT) The tailwind. IREDA’s FY26 borrowing programme was ~₹30,800 Cr, and management continues to frame the company against India’s clean-energy target arithmetic — the 500 GW non-fossil capacity goal by 2030 implies a multi-year RE-financing runway that IREDA is the purest listed proxy for. That structural demand is the bull case’s foundation.
7. Expectations baked in
This is the crux. IREDA trades at ~19× earnings and ~2.6× book — and the comparison that defines the debate is with its sibling PSU lenders: PFC and REC trade at roughly 5–7× earnings and around or below book (EnrichMoney; AlphaSpread). So IREDA carries a 3–4× P/E premium over lenders that are larger, more diversified, more profitable on a cost-of-funds basis, and pay fatter dividends.
What is that premium paying for? Three things: (1) growth — a 22% book CAGR vs PFC/REC’s low-double-digits; (2) purity — the only listed way to own just the RE-financing tailwind; and (3) runway — a far smaller base (₹93k Cr book vs PFC’s ~₹5 lakh Cr) with more room to compound. The reverse-DCF feel: at ~2.6× book and 16% ROE, the market is implicitly underwriting many years of high-teens-to-20% book growth with asset quality normalising back toward the old sub-2% GNPA. That is a demanding bar — it prices IREDA as a growth compounder, not a utility-like annuity lender. The bear’s one-liner is simply: you are paying 3× the multiple of REC for a lender that just printed a worse GNPA than REC and earns a thinner spread. The bull’s: REC’s multiple is a value trap of slow growth, and IREDA is the only PSU lender with a genuine compounding runway. Both are defensible; the price has chosen the bull’s framing, which is why the GNPA line matters so much — it is the variable that can collapse the premium.
8. Rerating signals — up vs down
| Could re-rate UP if… | Could re-rate DOWN if… |
|---|---|
| GNPA keeps grinding down (3.49% → back toward 2%) and the Gensol/Hetero recoveries land, proving the spike was idiosyncratic | A third large RE account turns — proving Gensol was a pattern, not an outlier, and that the 40-year clean record doesn’t survive a bigger, faster book |
| Loan book sustains 20%+ growth as the 500 GW RE build-out accelerates, keeping IREDA the purest proxy | Book growth decelerates below ~15% as the base grows, undercutting the entire growth-premium thesis |
| Cost of funds falls (rating upgrades, multilateral lines, a rate-cut cycle), widening NIM and lifting ROE | Spread compresses — funding costs rise or competition (banks, PFC/REC moving into RE) squeezes lending yields |
| QIPs get absorbed at premium prices, keeping capital adequacy comfortable without overhang | Repeated dilution / a government OFS floods supply and caps the multiple; PSU dividend-extraction pressure |
| Net NPA stays low (1.29%), confirming aggressive, conservative provisioning | RE-specific stress (module/tariff disputes, PPA renegotiation, a discom-payment squeeze hitting RE developers) broadens the bad book |
| The premium-vs-PFC/REC gap is defended by superior growth, drawing in growth-mandate FIIs/DIIs | The market simply re-rates IREDA toward PFC/REC multiples — a 19→8 P/E de-rating needs no scandal, just a mood change on “expensive PSU” |
9. Conviction texture
The bull case in its strongest form: IREDA is the cleanest listed instrument for the single most durable tailwind in Indian power. India needs to roughly triple non-fossil capacity this decade, every megawatt of it needs financing, and IREDA is the only pure-play, sovereign-backed, Navratna lender pointed entirely at that flow. It compounds its book at 22%, earns a 16% ROE, has a four-decade near-zero-loss heritage, and sits on a base small enough to keep growing for a long time. On that telling the premium to PFC/REC is not expensive — it is the correct price for the only PSU lender that is actually a growth company.
The bear case in its strongest form is just as clean: you are paying ~19× earnings and 2.6× book for a spread lender — and spread lending is a commodity where the durable winners are the ones with the lowest cost of funds and the best diversification, which is PFC and REC, not IREDA. IREDA earns a thinner spread, just printed a 3.49% GNPA (worse than its larger peers), concentrated its first real losses in exactly the kind of trendy RE names a policy-driven lender is prone to over-fund, and dilutes shareholders periodically to keep growing. If GNPA wobbles again or growth merely normalises, the multiple has 50%+ of air to give back with nothing “wrong” happening — it just stops being priced as a compounder.
What the evidence actually supports: the FY26 NPA spike looks idiosyncratic and well-provisioned, not systemic — two named frauds, net NPA falling while gross rose, GNPA grinding down sequentially through the year, and real cash recovered on Gensol. That is a manageable scare, and screener’s “cons” (P/B, interest cover) are mostly NBFC category-errors rather than genuine warnings. But the snapshot also confirms the uncomfortable parts: the spread is narrower than peers’, the cash flow is structurally negative, and the valuation prices in years of flawless execution. The honest read is that IREDA is a good business riding a real tailwind whose price has already extrapolated the best case — so the texture of conviction lives almost entirely on one line. Watch GNPA quarter by quarter: if it keeps falling and the book keeps compounding, the premium defends itself; if a third large account turns, the whole growth-compounder thesis — and the multiple that rests on it — is what breaks first. No recommendation either way.