Business

Know the Business

Reliance Power is, in practice, two thermal coal plants plus a deleveraging story: the 3,960 MW Sasan UMPP and the 1,200 MW Rosa plant generate essentially all the cash, the 5,160 MW thermal core is locked into 25-year PPAs and runs at world-class load factors, and the equity story turns on whether a 2.5 GWp solar + 2.5 GWh BESS pipeline can be funded and built without re-leveraging the parent. The market is being asked to buy three things in one wrapper — a high-quality regulated thermal asset, an unfunded renewable option, and a sub-book governance turnaround — and the share price (P/B ≈ 0.74 vs Adani Power at 7.4×) tells you which of the three it is pricing in.

1. How This Business Actually Works

The economic engine is one plant — Sasan — operated as the cheapest source of coal-fired electricity in India because it sits on its own coal mine.

The core mechanic: Sasan UMPP runs on the principle of mine-mouth integration. Reliance Power owns and operates the Moher and Moher-Amlohri coal blocks adjacent to the plant — 18.12 million tonnes mined in FY25, feeding the boilers on a captive basis. There is no railway haul, no e-auction premium, no imported-coal blending tariff. That is why Sasan ran at a 90.6% Plant Load Factor in FY25 versus an India thermal average of 69%. The plant sells at a fixed levelised tariff of ~₹1.20/kWh under a 25-year PPA to 14 distribution companies across seven states. The fuel-cost structure is essentially engineering the coal out of the ground at production cost rather than buying it at market — a structural cost advantage that only ends if the captive coal block runs short or environmental compliance forces a shutdown.

Rosa is the opposite model: a regulated cost-plus return on equity (15.5% on regulated equity per CERC), where the Uttar Pradesh DISCOM pays whatever fuel costs run plus a fixed-charge recovery for being available. Profitability is mechanical if availability stays above the normative threshold (96% in FY25) — the only operating risk is a payment delay from UPPCL.

Where incremental profit actually comes from. Once Sasan and Rosa are running, fixed-charge recovery is the dominant lever; revenue at the parent has barely moved (₹7,503 Cr → ₹7,583 Cr from FY22 to FY25). What has moved profit is the interest line falling in lock-step with debt repayment — operating income has stayed in the ₹1,160–₹2,735 Cr range while interest expense fell from ₹3,206 Cr (FY19 peak) to ₹2,056 Cr (FY25). Each ₹1,000 Cr of debt repaid adds roughly ₹100 Cr of pre-tax profit. That is the entire near-term earnings algorithm: not new MW, not pricing, just deleveraging the balance sheet that was over-built in the 2010–2015 capex binge.

For most of FY16–FY22 the two lines were within a coin-flip of each other; pre-tax income was a residual after interest. That is the definition of an over-levered asset, and it is what drove three years of net losses (FY19, FY20, FY22, FY23, FY24). The crossover is happening now.

2. The Playing Field

Among India's listed power generators, Reliance Power is the smallest by market cap, and it is the only name in the peer set still digging out of a financial hole.

What the peer set reveals. The line of best fit through the Indian power universe is brutal: returns on capital map almost linearly to book multiples. Adani Power earns 22.5% ROCE and trades at 7.4× book; Torrent Power earns 16% and trades at 4.7×; the PSUs (NTPC, NHPC) earn 7–11% and trade between 2–2.1× book. Reliance Power earns 6.2% ROCE and trades at 0.74× book — the only sub-book name in the set. Two readings are possible: (a) the market is mispricing the deleveraging that is now arithmetically forcing ROCE higher, or (b) the market is correctly pricing the SEBI/ED legal overhang and the going-concern note at RSTEPL that the operating numbers do not capture.

The right comparator is Adani Power, not the PSUs or Tata. Both are private-sector thermal IPPs that built in the same 2010–2015 cycle. Adani consolidated coastal plants under aggressive merchant pricing and PPA renegotiation; Reliance Power chose project-finance and let interest costs eat earnings for a decade. The result is the cleanest counterfactual in Indian power: same vintage, same fuel, same regulator — 22.5% ROCE vs 6.2%. The gap is execution and capital structure, not market access.

3. Is This Business Cyclical?

Power is not cyclical at the demand level — India's electricity demand has compounded for 40 years and FY25 set a 250 GW peak record — but it is intensely cyclical at the fuel cost, regulatory tariff, and DISCOM payment levels, and Reliance Power has been on the wrong side of every one of those cycles since 2014.

Where the cycle actually hits.

Fuel cost cycle (2018–2020): Imported coal prices spiked, and Reliance Power's gas-based Samalkot plant (2,400 MW commissioned, never fired commercially because domestic gas allocation never came) was permanently impaired. FY19's ₹2,952 Cr loss and FY20's ₹4,271 Cr loss were largely driven by impairment of stranded assets — not operating losses on the running plants. The gas allocation that was supposed to come never did. SMPL is now being monetized one combined-cycle module at a time to a Bangladesh JV with JERA Co. of Japan.

Regulatory cycle (2015–present): The 2015 environmental norms required significant emission-control retrofits at every coal plant in the country. CERC allowed these as Change-in-Law pass-through, but the cash outlay led the tariff recovery by several years. The Ministry has now extended the compliance deadline by three more years.

DISCOM payment cycle (chronic): Debtor days swelled from 80 (FY14) to 156 (FY22) when state distribution companies — the universal bottleneck of Indian power — fell behind on payments. UDAY and the late-payment surcharge rules pulled this back to 73 days by FY25. Each 30 days of receivables drag costs the company roughly ₹620 Cr of working capital — meaningful at this size.

Renewable disruption (structural): Solar tariffs in India fell from ₹17/kWh in 2010 to ₹2.40/kWh in 2024. CSP — the technology RSTEPL bet on — was overrun by photovoltaic. RSTEPL's accumulated losses of ₹3,326 Cr and going-concern qualification are the residue of a single technology call made fifteen years ago.

The pattern is consistent: this is a balance-sheet-cyclical business dressed up as a utility. Earnings volatility comes from the ratio of fixed cost (interest, depreciation) to a regulated revenue stream, not from sales volatility.

4. The Metrics That Actually Matter

Most utilities are valued on dividend yield and rate-base growth. Reliance Power is in a different stage: it is being valued on whether a buried equity stake can be re-floated. The metrics that explain value creation here are not the usual ones.

Why these and not the textbook ratios. P/E is meaningless here — FY25 net income of ₹2,948 Cr was created by a ₹3,234 Cr exceptional gain on deconsolidation of subsidiaries; ex-exceptional, the continuing operations made a small loss. ROE swings ±20 percentage points period to period because of impairment cycles. Operating margin swings between 14% and 47%. What does not swing is plant load factor at Sasan and the rate of debt paydown — those two numbers explain almost every quarter's earnings.

The single number that would change my mind on this stock is the parent's net debt at the next reporting date. Borrowings have fallen from ₹30,043 Cr (FY15) to ₹15,153 Cr (FY25), and management says the standalone parent is now debt-free. If consolidated debt continues to roll off at ₹2,500–₹3,000 Cr a year, interest expense alone gives ~₹250–₹300 Cr of incremental pre-tax profit per year — without a single new megawatt being built.

The metric I would not trust is the renewable pipeline guidance. NU Suntech's 930 MW solar + 465 MW BESS PPA with SECI is a real contract, but the project is at the SECI fake-bank-guarantee centre of the ED case; the BESS would be the largest in Asia outside China; and equity funding sources have not been disclosed at the parent. Until the project's first 100 MW is energised on time and on budget, treat it as an option, not a forecast.

5. What I'd Tell a Young Analyst

The whole valuation puzzle here collapses if you can answer two questions. First: how much is Sasan UMPP worth as a stand-alone asset, divorced from the rest of the holdco? A 3,960 MW plant operating at 90.6% PLF on a captive coal mine, 14 years into a 25-year PPA, generating roughly ₹3,000–₹4,000 Cr of EBITDA, would clear ₹40,000–₹60,000 Cr in a private auction in India today — multiples of the entire current market cap of the listed parent. Second: what is the parent-level liability if the SEBI forensic audit and ED-PMLA cases find against the company? Nobody can model that, including management. You do not have to be right — you have to size the position so that being wrong does not blow you up.

Three things to watch quarter by quarter, in this order:

1. Net debt at the parent. This is the only metric where the company controls the outcome and where the math compounds in the equity holder's favour.
2. Sasan PLF and Rosa availability. Either operational mishap (a coal block exhaustion, a CERC tariff cut, a major outage) and the cash engine cracks. So far both numbers have been remarkably stable.
3. Status of the SECI/ED matter and SEBI forensic audit. Read every Regulation 30 disclosure on the day it comes out. The fundamental analysis cannot price the legal tail.

Three things the market may be missing. The deleveraging is now arithmetic — ₹3,000+ Cr of operating cash flow pre-interest against ₹2,000 Cr of interest, on falling debt, is a coiled spring. The Sasan asset has no real Indian peer in mine-mouth integration economics, and that is durable. And the Bangladesh JV with JERA, if it actually gets to FID on Phase 1 (718 MW), turns the long-stranded SMPL equipment from a liability into recovered capital.

Three things that should make you walk away. Anil Ambani's track record at Reliance Communications and Reliance Capital is what it is, and the SEBI forensic audit is not the first time this group has had a regulator inside its books. The CSP plant (RSTEPL) has been propped up on going-concern accounting for years and nobody can tell you what its impairment will eventually look like. And the renewable pipeline depends on capital this entity has historically been very bad at raising on equity-friendly terms.

The honest description of this stock: it is a leveraged, governance-discounted call option on Sasan UMPP, with embedded tail risks that mostly are not in the cash flows. If that is the bet you want to make, fine. But never confuse it with owning a utility.