Power Finance Corporation — Quasi-sovereign power lender mid-merger
Power Finance Corporation Limited
1. Snapshot
Power Finance Corporation is the financing node of the Indian power sector — a Maharatna PSU NBFC (RBI-classified Infrastructure Finance Company) that lends to generation, transmission, distribution and, increasingly, renewables. It is the parent of REC (holds 52.63% since the 2019 acquisition), and the two together carry the largest NBFC loan book in India at ~₹11.64 lakh crore. Market cap ₹1,42,234 Cr, current price ₹431 (52-week range ₹330–₹486), trading at a P/E of ~5.5 and roughly 1.07x book, throwing off a ~3.7% dividend yield. The animal: a quasi-sovereign, fixed-spread wholesale lender — closer to a regulated annuity machine than a growth NBFC, now in the middle of being merged with its own subsidiary. As of 2026-06-20, from screener snapshot.
2. Business & position in the value chain
PFC doesn’t generate an electron or string a wire — it lends the rupees that let everyone else do so. It sits at the money-supply chokepoint of the entire power value chain: project term loans to generators (thermal, hydro, nuclear, increasingly renewables), short- and medium-term funding to state distribution companies (DISCOMs), refinancing of commissioned assets, plus non-fund products (guarantees, letters of comfort). Standalone loan book closed FY26 at ~₹5.8 lakh crore; the consolidated group (with REC) is ~₹11.64 lakh crore.
The mix has shifted decisively. Management (CMD Parminder Chopra) described the arc on the May 2026 concall: from ~75% conventional generation at one time, the book is now ~50% generation (of which ~16% is renewable, the rest conventional), with distribution the next-largest slice. The group renewable book alone is ~₹1.65 lakh crore — among the largest in India — and PFC has now started financing the new technologies the grid actually needs: ~₹16,000 Cr cumulatively sanctioned toward battery and pumped-storage projects, an early-mover position in firm/dispatchable green power.
Concentration is structural and double-edged: PFC’s borrowers are overwhelmingly the Indian power sector, much of it state-owned. That means its asset quality is hostage to DISCOM health and the government’s reform appetite — but it also means an implicit sovereign backstop runs through the book in both directions (borrowers and the lender are largely the same family).
3. Management & promoter quality
The promoter is the Government of India, holding 55.99% — rock-steady across every quarter shown (no creeping OFS overhang in the recent shareholding table). This is the defining fact about PFC, and it cuts both ways.
The good. PFC is a Maharatna with a long, boring, reliable dividend record — DIPAM policy mandates ~30% payout, and PFC has paid consistently (FY26 total dividend ₹18.55/share; payout ~29%; ~3.7% yield). Asset quality has been cleaned up impressively under the current team: gross NPA fell from 3.66% (Mar-23) to 0.66% (Mar-26), net NPA from 1.03% to 0.13% — 80% of the peak NPA book resolved, with the remainder largely under liquidation and 100%-provisioned. That is genuine execution, not accounting cosmetics. The team is technically credible (CMD plus a newly appointed Director-Finance with 31 years across treasury/banking/fundraising). The “three T’s — Timeliness, Trust, Transparency” framing is PSU boilerplate, but the disclosure quality on the concall (project-by-project Stage-3 detail, hedging structure, spread guidance) is above-average for a PSU.
The flags. This is a government instrument first and a shareholder-value engine second. The state extracts dividends, can direct lending toward policy priorities (DISCOM bailouts, Atmanirbhar manufacturing, EVs, bioethanol — all name-checked on the call), and ultimately controls the strategic agenda. The merger itself is a top-down, Budget-announced decision, not a board-originated one. Capital allocation is therefore only partly the management’s to optimise. And the lender’s fortunes are tethered to the same state ecosystem it lends into — a circularity that is comforting in a crisis and limiting in a boom.
4. Financial trends
PFC is a financier, so the lens is book growth × spread × asset quality, not multiple expansion.
| Metric | FY24 | FY25 | FY26 | Read |
|---|---|---|---|---|
| Net profit (consol, ₹ Cr) | 26,461 | 30,514 | 33,625 | Highest-PAT NBFC in India; ~13% YoY growth |
| EPS (₹) | 59.88 | 69.67 | 78.49 | Compounding ~14% |
| ROE | — | — | 20.7% | High for a wholesale lender |
| ROCE | — | — | 9.71% | Low optically — it’s a lever business, ROCE is the wrong gauge |
| Gross NPA | 3.02% | 1.64% | 0.66% | Dramatic clean-up |
| Net NPA | 0.85% | 0.38% | 0.13% | Near-pristine |
| Spread (standalone) | — | — | 2.46% | Guided 2.40–2.50% for FY27 |
| NIM (standalone) | — | — | 3.55% | Stable |
| CRAR | — | — | 23.44% | Tier-1 at 21.93% — large headroom |
| Dividend payout | 23% | 23% | 29% | DIPAM ~30% discipline |
ROE/ROCE. The 20.7% ROE on ~1.07x book is the headline tension: a 20% return business priced at barely above book. ROCE of 9.71% looks weak only because PFC is a leveraged lender (7.7–7.8x) — for a financier, NIM and spread are the operating-quality gauges, and both are stable. The ROE has been helped recently by provision reversals (₹1,800 Cr in FY26: ~₹800 Cr from resolving Sinnar Thermal and TRN Energy, ~₹1,000 Cr from DISCOM rating upgrades feeding through the ECL model). That tailwind is largely spent — management was candid that “going forward, we will not have the benefit” of resolution write-backs.
Growth. PAT has compounded strongly (FY21 ₹15,716 Cr → FY26 ₹33,625 Cr, >16% CAGR), but the loan book grew only ~7% in FY26 vs a guided 10–11% — the miss came from disproportionate prepayments in a falling-rate environment, as banks aggressively refinanced PFC’s commissioned assets. Revenue 5-year growth screener flags as “poor” (~10%) — true for a wholesale book in a rate-cut, high-competition window.
Balance sheet. Net worth crossed ₹1 lakh crore (+13% YoY). Borrowings ₹4,88,500 Cr, 80:20 domestic:foreign, 65% fixed-rate — which is the key to PFC’s stability: rate moves hit it slowly (5–6 year average liability life), not suddenly. 97% of the USD-10.3bn foreign book is hedged.
Screener’s auto-flags, interpreted. “Low interest coverage ratio” and “might be capitalizing interest cost” are screener mis-reading a lender’s P&L — interest is the cost of goods, not a solvency stress. The “poor sales growth” flag is real but reflects the prepayment/competition dynamic, not deterioration. The genuine pros — book value parity, dividend yield, payout discipline — are accurate.
5. Latest quarter
Q4 FY26 / full-year FY26, reported at the May 13, 2026 investor meet. Standalone net profit hit a record ₹20,051 Cr (+16% YoY), driven by ~13% net interest income growth plus the ~₹1,800 Cr provision reversals. Consolidated PAT ₹33,625 Cr. Spread 2.46%, NIM 3.55%, yield 9.96%, cost of funds 7.50%.
CMD Chopra summed up FY26 as “a year where resilience met results.” The two live narratives: (1) the prepayment squeeze — “the prepayments were disproportionate to that which was factored in… as banks aggressively refinanced these assets,” dragging book growth to 7% vs 10–11% guided; and (2) forex pain — “FY26 was one of the most volatile years for global currency markets,” with rupee depreciation pushing the cost-of-hedging reserve to ~₹34bn (from ~₹1bn in FY25). FY27 guidance: ~10% loan growth, spread 2.40–2.50%, and a view that prepayment pressure should moderate now that RBI has paused (repo held at 5.25% for two consecutive meetings, neutral stance). Management does not expect further FY27 rate cuts.
6. What’s happening now
HARD — the merger is real and advancing. Announced by FM Sitharaman in the Union Budget on 1 February 2026, REC is to be merged into PFC. Both boards have given in-principle approval; legal advisor, transaction advisor, merchant bankers and registered valuers are appointed; valuation and the draft scheme are in progress. Per news reporting, the President of India’s approval (required under PFC’s Articles) was communicated around 10 June 2026 — a major milestone clearing — though the final share-swap ratio and scheme are still to be filed. Target for the merged entity to come into existence: 1 April 2027, subject to MCA, RBI, SEBI and Cabinet approvals. (Angel One, Ventura, Whalesbook)
The catch (the thing markets are chewing on). A plain share-swap dilutes the government’s stake in the merged entity to ~42% — below the 51% threshold for “government company” status under the Companies Act, and below covenant thresholds on PFC’s bonds. The government has committed to preserving government-company status, but the “how” is undecided; estimates suggest a ~₹25,000 Cr infusion (or a structured swap) may be needed. Reported swap-ratio chatter has ranged from 6 PFC : 7 REC to 8 PFC : 9 REC — none official; the valuers will decide. (Angel One merger timeline, IndMoney)
HARD — RE-financing pivot. ~₹16,000 Cr sanctioned to battery/pumped-storage; FY26 disbursements ₹1,65,414 Cr with renewables tilting to hybrid solar-wind. Group RE book ~₹1.65 lakh crore.
HARD — asset-quality clean-up essentially complete. Sinnar Thermal (₹3,001 Cr, 42% principal recovered) and TRN Energy (₹1,139 Cr) resolved in FY26; Stage-3 now ₹6,323 Cr across 19 projects, 86% covered, 16 of them 100%-provisioned.
Sector tailwind backdrop. India installed capacity past 530 GW; 55 GW non-fossil added in FY26 (a record); peak demand hit an all-time 256 GW in April-26; DISCOM AT&C losses down to 15.04% and DISCOMs reporting their first all-India positive PAT (~₹2,700 Cr) — directly de-risking PFC’s distribution book.
7. Expectations baked in
At ~5.5x P/E, ~1.07x book and a ~3.7% yield, the market is pricing PFC as a cheap PSU financier in run-off-ish growth — not as a compounder. A 20.7% ROE business at ~1.07x book implies the market doubts that ROE is durable: it expects spread compression (competition + rate cycle), a fading of the provision-reversal tailwind, mid-single-digit-to-~10% book growth, and a government that will keep extracting dividends rather than reinvesting for growth. In rough reverse-DCF feel, ~1x book on a 20% ROE bakes in either a structural fall in ROE toward the low-teens, or a permanently high cost of equity (the “PSU discount” — political risk, policy-directed lending, dilution overhangs).
The “cheap PSU financier” debate, plainly: bulls say a 20% ROE, 0.13% net NPA, 23% CRAR lender at book value is mispriced and the merger plus DISCOM recovery should narrow the discount. Bears say the ROE is flattered by one-off reversals, the spread is structurally drifting toward 2.4%, growth is hostage to prepayments and bank competition, and the PSU discount exists for reasons (the government’s hand on lending, dividends and now a dilutive merger) that won’t vanish. The honest read: PFC is priced for nothing to improve and something to mildly deteriorate — which is why the asymmetry sits in the catalysts below.
8. Rerating signals — up vs down
| Could re-rate UP if… | Could re-rate DOWN if… |
|---|---|
| Merger swap ratio lands fairer-than-feared and the government-company status is preserved cleanly (capital infusion confirmed), removing the overhang | Swap ratio dilutes PFC shareholders more than expected, or “government company” status is jeopardised (bond-covenant + index/ownership uncertainty) |
| Loan book growth reaccelerates to/above 10% as prepayments moderate now that RBI has paused | Prepayment squeeze persists or worsens — banks keep refinancing commissioned assets, dragging growth below guidance |
| Spread holds at 2.45–2.50% and the merged entity realises cost/scale synergies on a ₹11.6 lakh crore book | Spread compresses below 2.40% as competition + forex hedging costs bite; cost of funds re-rates up |
| Asset quality stays pristine and DISCOM reform (RDSS, smart meters, positive DISCOM PAT) keeps feeding ECL write-backs / lower credit cost | Credit cost normalises higher with the reversal tailwind exhausted; a new large power-sector NPA emerges |
| PSU re-rating wave + merger creating a single ~₹2.4 lakh crore “single-window power financier” draws institutional flows | Sharp rupee depreciation drives another large hedging-reserve / translation hit; merger execution slips past April-2027 |
| Dividend yield stays ~3.7–5% with payout discipline, anchoring valuation as a yield play | Government raises payout extraction or directs policy lending that erodes returns; OFS supply if the state needs to fund the infusion |
9. Conviction texture
The bull case in its strongest form: this is the closest thing India has to a sovereign-backed bond that compounds at 20% ROE, trades at book, yields ~3.7%, and has just finished scrubbing its balance sheet to a 0.13% net NPA with 23% capital adequacy. It funds the single most structurally favoured sector in the country — power, where demand keeps hitting records and the renewable/storage build-out is only starting — and it’s about to merge with its own subsidiary into a ~₹2.4 lakh crore single-window financier with scale, lower duplication, and an early lead in battery/pumped-storage lending. If the market ever decides the PSU discount is too harsh, the re-rate is mechanical: even 1.3–1.5x book on a sustained 18%+ ROE is a different stock.
The bear case in its strongest form: the 20% ROE is partly borrowed from the future — ~₹1,800 Cr of FY26 profit was provision reversals that management openly says won’t recur, and the underlying spread is drifting toward 2.40% under bank competition and forex cost. Growth is the soft spot: a wholesale lender that grew its book 7% while guiding 10–11%, because its best (commissioned) assets get refinanced away the moment they de-risk, has a structural moat problem — it originates the risky construction-phase loan and loses it once it’s safe. And the merger, however strategically sensible, is a government-driven event with an unresolved dilution/ownership puzzle hanging over the price. The PSU discount is not a mistake; it’s the market pricing the fact that this company answers to the state before it answers to minority holders.
What the evidence actually supports: the asset-quality clean-up is real and verifiable (GNPA 3.66%→0.66% over three years). The capital cushion is real (23.44% CRAR). The dividend discipline is real (DIPAM ~30%). The growth concern is also real and management-confirmed (the prepayment drag). The spread is stable-to-gently-eroding, not collapsing. So PFC is genuinely cheap and genuinely constrained — both things are true at once. The thing to watch to know which way it breaks is the merger scheme: the swap ratio and the government-status mechanism will either lift the overhang and let the DISCOM-recovery / scale story re-rate the multiple, or confirm that minority shareholders are second in line and entrench the discount. Until that scheme is filed, the price is doing exactly what a 5.5x P/E says it is — waiting.
Sources: local screener snapshot (2026-06-20) and PFC Q4 FY26 investor-meet transcript (May 13, 2026); merger status from Angel One, Angel One timeline, Ventura, IndMoney, Whalesbook; valuation/peer context from Tickertape and Screener. No buy/sell recommendation.