Api Power Company Limited closed at Rs 345.00 at NEPSE's Friday, October 16 session, a price that values the hydropower company at roughly 45 times its core operating earnings per share of Rs 7.63। The earnings figure comes from the company's own Q4 FY 2082/83 filing, published at apipower.com.np and summarized by ShareSansar on August 14, 2026, and it is the profit that remains after the paper is stripped out: Rs 960.61 million of reported net profit, less a Rs 474.15 million fair-value gain, leaves about Rs 486 million of core operating profit over 63,797,242 listed shares। At the same arithmetic, the company's headline EPS of Rs 15.06 would put the stock at just 22.9 times earnings and looking cheaper still — but that is the wrong number to use, because nearly half the profit never touched cash।
This is the research desk's next institutional equity research note: a fundamentals-first note built on the company's filings published at apipower.com.np, CARE Ratings Nepal's published rating actions, the NEPSE listing record at nepalstock.com, and market data as of the October 16, 2026 close। What follows is what those documents say, what they do not say, and the valuation they imply।
A far-west hydropower company
Api Power Company Limited is a hydropower developer and operator focused on the far west of Nepal। It was established on June 19, 2003 as a private limited company, converted to a public limited company on July 21, 2013, and listed on NEPSE on November 18, 2015 in the Hydro Power sector, per CARE Ratings Nepal's rating rationales and ShareSansar's company record। Its corporate office sits at Trade Tower Nepal, Thapathali, Kathmandu, and Sanjeev Neupane has been managing director since 2012।
The operating portfolio is 62.5MW across four legs। Naugad Gad (8.5MW) in Darchula began commercial operation in August 2015, Upper Naugad Gad (8MW) about 5 km upstream in October 2019, and Upper Chameliya (40MW), also in Darchula, in August 2023, running under contingency until its RCOD in February 2024। A 6MW solar PV leg sits across Chandranigahpur (4MW), Dhalkebar (1MW), and Simara (1MW)। All four legs sell under power purchase agreements to the Nepal Electricity Authority, and each agreement has a fixed expiry: this is a company of wasting assets, and the valuation later in this note prices that fact directly।
One structural line deserves flagging early। ShareHub reports Api Power's shareholding as 0% promoter and 100% public as of October 16, 2026 — a 0% promoter holding is valid, meaning a fully public float, but it is single-sourced and no annual report was located to confirm the holder list, so the desk treats it as provisional। The company was privately held from 2003 to 2013, its IPO put out 4,000,000 units at Rs 100, and the equity base has been built up since by a long ladder of rights and bonus issues to today's 63,797,242 listed shares।
The record profit that is half paper
The headline is genuinely a record। The Q4 FY 2082/83 filing reports revenue from electricity sales of Rs 1,766.56 million, profit from operations of Rs 1,085.67 million, and net profit of Rs 960.61 million — more than double the prior year's Rs 427.65 million headline, and the best year in the company's recorded history। Earnings per share print at Rs 15.06 and book value per share at Rs 119.51, both from the company filing itself।
But Rs 474.15 million of that profit is an FVTPL fair-value gain — a mark-to-market revaluation under NFRS 9, booked on the company's 25% stake in Ingwa Hydropower, carried at Rs 343.80 a share। It is a paper gain of about 49% of reported PAT: no cash changed hands, and it will reverse if the valuation does। Strip it out and the core operating profit is about Rs 486 million, giving core EPS of Rs 7.63, not the reported Rs 15.06। The desk used the core figure for every earnings-based leg of this valuation, because capitalizing a one-off mark-to-market gain is precisely the error the regulator has already called out once।
The rest of the income statement is steadier than the headline suggests। Revenue grew about 7% year on year to Rs 1,766.56 million, profit from operations held at Rs 1,085.67 million, and return on assets came in at 6.15%। The margin headwind is real and already scheduled: Naugad Gad completed its 10-year commercial-operation tax holiday on August 19, 2025, per bankingnews and CARE, so its revenue is taxed at the 12.5% concessionary rate from FY 2082/83 — the tax line is no longer zero on the anchor plant।
The regulator already rejected this arithmetic once
The fair-value issue is not academic: the Energy Regulatory Commission has already drawn the line on it। For FY 2080/81, the board proposed 9.4737% — a 9% bonus plus 0.4737% cash — and on September 27, 2024 the ERC rejected it, because NFRS 9 fair-value gains are not distributable as dividends। The board canceled the proposal and came back with 5.2632%: 5% bonus plus 0.2632% cash, per ShareSansar's October 16, 2024 coverage। For FY 2081/82 the board repeated the same 5.2632%, the 22nd AGM endorsed it on January 13, 2026, and 30,37,963.91 bonus units listed on NEPSE on March 1, 2026।
The ex-bonus arithmetic went through cleanly in NEPSE's records, and the desk checked it against its own OHLC archive। Book closure was Friday, January 2, 2026, a non-trading day। The January 1 close was Rs 294.0; the January 4 open — the first session after closure — was Rs 280.0, which is 294.0 divided by 1.05 exactly: a textbook 5% ex-drop on the adjustment date। Over nine years the payout history averages 6.83% a year, per ShareHub's dividend record, and it is bonus-heavy with a token cash line: distributions grow the share count, not shareholder cash।
What peers charge for the same earnings
Set API against seven operating hydro peers on fundamentals compiled from ShareHub and ShareSansar aggregator data, and the multiple arithmetic is flattering। The peer median P/E is 67.16 times and the peer median P/B is 3.09 times। API at Rs 345 changes hands at about 45 times core earnings — a 33% discount to the peer median — and at about 2.9 times book value of Rs 119.51 against the peer median of 3.09। On peer earnings, the implied fair value is Rs 512.07 (67.16 times core EPS of Rs 7.63); on peer book, Rs 368.95 (3.09 times BVPS of Rs 119.51)। Both sit well above the Rs 345 price।
Three caveats keep this honest। First, the peer fundamentals are single-sourced from aggregators; NHPC was excluded from the P/E median for negative EPS, BPCL's EPS is heavy with dividend income rather than clean hydro operations, and UMHL's EPS is a Q3 not Q4 number। Second, the peer-implied band on the inflated reported EPS of Rs 15.06 would be Rs 1,011.43 — shown, not used, because capitalizing the FVTPL gain is the same error the desk refuses above। Third, and the point of this note: multiples price going-concern earnings against peers whose PPAs, debt loads, and asset lives differ, while a discounted cash flow prices API's actual finite-life, debt-heavy cashflows। The gap between the two legs is where the target gets interesting।
The DCF leg
The research desk built a finite-life free-cash-flow-to-firm model with no terminal value: each leg is projected to its own assumed PPA expiry — Naugad Gad around August 2045, Upper Naugad Gad around October 2049, Upper Chameliya around August 2053, and solar around 2045 — because a hydro BOOT has no perpetual cashflow to capitalize। The weighted average cost of capital comes out at 9.67%: a 2.60% risk-free rate anchored to NRB's 91-day T-bill auctions plus a tenor uplift, a Nepal equity risk premium of 8.89% from Damodaran's country risk premiums, a bottom-up beta of 0.878 relevered at a 25% market-value debt weight, and a pre-tax cost of debt of 8.50% reflecting the CARE-NP BB+ sub-investment-grade rating।
The model passes its own consistency check: at the base-case plant load factors, the driver tree implies FY 2082/83 revenue of Rs 1,764.66 million against the filed Rs 1,766.56 million, a difference of 0.11%। Three scenarios follow। The base case — mid load factors, Upper Chameliya at the market-norm tariff, steady operation — values the equity at Rs 83.31 a share। The bear case — weak hydrology, Upper Chameliya at the low bound of CARE's Rs 4.00–8.40 tariff range — gives Rs 10.91, with Rs 7,014 million of net debt absorbing nearly all firm value। The bull case — strong hydrology, tight operating costs, easier rates — gives Rs 149.23। Even the bull DCF sits below the peer-implied band, and the desk ran no terminal value in any scenario: the 0% terminal-value share of firm value is a feature of the hydro template, not an oversight।
The blend and the rating
The target blends 50% DCF with 25% of the P/B-implied Rs 368.95 and 25% of the P/E-implied Rs 512.07 on core EPS, with the pipeline carried deliberately at Rs 0 because none of it is modelable from verified inputs। The arithmetic lands at Rs 226 for the bear case, Rs 262 for the base case, and Rs 295 for the bull case: implied 12-month returns of negative 34.5%, negative 24.1%, and negative 14.5% against the Rs 345.00 close। The desk's rating rule is mechanical — buy at plus 20% or better, hold between negative 10% and plus 20%, reduce between negative 30% and negative 10%, sell below negative 30% — and it is applied without softening। On the base target, the implied negative 24.1% reads REDUCE। That is the research desk's view, reported as fact, not as an instruction to any reader।
A reader checking Nepse Decode's verdict engine will find a bounce-setup pattern in API's history, and it deserves a direct answer because it looks like a disagreement with this note। It is not। The setup the engine scores — a trough of at least 30% followed by a close back above the five-session high on heavy volume — has already played out: the December 2024 trough at Rs 249.50 has been followed by a 38% recovery to Rs 345, and the engine currently marks the setup inactive। What is active on the engine is a bearish climax-volume flag, the kind its backtests associate with five to ten sessions of weakness, and the engine's overall read is neutral on a twenty-session horizon। A twenty-session neutral and a twelve-month reduce are different questions, and they give compatible answers।
A second apparent disagreement is worth defusing: the headline price-to-earnings of about 22 times, which looks reasonable next to the sector। That multiple divides the price by the reported EPS of Rs 15.06 — earnings that include the Rs 474.15 million paper gain। On the Rs 7.63 the plants actually produced, the multiple is 45 times। The engine's fundamentals feed carries the reported number; this note carries the core one। Both are arithmetically right; only one of them can ever be paid out as dividend, as the ERC's 2024 rejection of the FY 2080/81 payout already demonstrated। The overvaluation call survives both checks।
The pipeline is real, and so are the obstacles
The option value the blend excludes is genuinely material। Middle Chameliya (28.3MW), via the Darchula Power subsidiary, was about 40% complete in mid-January 2026 per CARE's February 2026 rationale। Marsyangdi Besi (50MW) is the big one: the board decided in August 2025 to acquire Divya Jyoti Hydropower, signed the Troyer AG and Troyer Hydro equipment contract in November 2025, and then saw construction halted on November 22, 2025 after local protests, per Kantipur/HT syndication; ShareSansar reported about 25% physical progress in July 2026। Parwanipur 8MW solar remains listed as under construction in the Q4 FY 2082/83 filing, with a revised cost of Rs 534 million — CARE's expectation of a mid-April 2026 commissioning was missed, and no COD confirmation has surfaced।
Hanging over all of it is the pending 1:0.3 rights issue: 19,139,173 units at par, Rs 191.39 crore earmarked as equity for Marsyangdi Besi, board-proposed on December 23, 2025 per investopaper। It is not yet confirmed: AGM endorsement is unconfirmed and no SEBON approval has been seen as of October 17, 2026, so the note keeps it outside the base share count and treats the pipeline as option value at Rs 0। The desk notes the dilution maths plainly, though: if the rights were issued at par with proceeds ignored, the base DCF per share would fall to about Rs 64 — but the proceeds fund an unmodelable acquisition, so the note stays qualitative।
The price action
The tape runs 2,506 sessions back to the November 18, 2015 listing। The all-time high of Rs 888 intraday in June 2016 has never been revisited: at Rs 345, the stock sits 61.15% below it। Over the last year it has gained 27.68%, from Rs 270.20 on October 16, 2025। At the close, the 20-day simple moving average sat at Rs 337.06, the 50-day at Rs 333.56, and the 200-day at Rs 328.64 — the price rides above all three in bullish alignment। The RSI14 reads 63.33, upper-neutral and approaching overbought, and the MACD line at 1.91 sits above its signal at 1.30, short-term positive। Liquidity is genuine by NEPSE standards: about 211,000 shares and Rs 7.11 crore a day over the last three months, and Friday itself printed 443,463 shares for Rs 15.30 crore of turnover, roughly twice the average।
What the files do not yet say
Several items sit outside what the sources confirm, and this note keeps them there। The biggest DCF input gap is Upper Chameliya's exact PPA tariff: modeled as a scenario range from CARE's Rs 4.00–8.40 band, never hard-coded as the company's number — one verified disclosure would collapse the spread। The signed dates of the Naugad Gad and Upper Naugad Gad PPAs were not found, so 30-year terms are market-norm assumptions। The 0% promoter / 100% public line is single-sourced। Net debt of Rs 7,014 million is derived from CARE's 0.92 gearing, not a published debt schedule। The FY 2079/80 PAT carries a conflict between ShareSansar's Rs 6.80 million headline and Rs 44.20 million in later comparative figures, and the peer fundamentals are single-sourced aggregator data। FY 2082/83 figures are unaudited, and the rights issue is unconfirmed। None of these gaps flatters the price; all of them are stated so the reader can weigh them।
The unresolved question
Put the two pictures together and the tension is plain। On one side: 62.5MW of operating capacity with Rs 1,766.56 million of filed revenue, a regular if bonus-heavy dividend record, a genuine development pipeline, and a price that charges only 45 times core earnings in a sector where the median multiple is 67.16। On the other: nearly half of the record profit is a paper mark-to-market that the regulator has already refused to let shareholders distribute, every plant is a wasting asset toward a fixed PPA expiry, the tax holiday on the anchor plant has already ended, Rs 7,014 million of net debt sits against the firm value, and a finite-life DCF at a 9.67% WACC prices the operating equity at Rs 83.31।
Either the peer-median arithmetic is right and API is simply cheap at 45 times core earnings, in which case the DCF is too punitive on asset life and debt, or the finite-life maths is right and the multiple is pricing going-concern earnings the company will never collect in full। The numbers that would decide it — a verified Upper Chameliya tariff, a Middle Chameliya commissioning date, a Marsyangdi Besi resolution, the audited FY 2082/83 annual report, and SEBON's verdict on the 1:0.3 rights — do not exist yet। What breaks the debate, if anything does, will be one of those facts arriving, not the chart। Until then, the Rs 345 question stays open।
A note on what is verified here. Every figure above comes from a named source: the company's Q4 FY 2082/83 filing published at apipower.com.np (revenue Rs 1,766.56 million, profit from operations Rs 1,085.67 million, FVTPL fair-value gain Rs 474.15 million, net profit Rs 960.61 million, EPS Rs 15.06, BVPS Rs 119.51, ROA 6.15%, paid-up capital Rs 637.97 crore, 63,797,242 listed shares), ShareSansar's coverage of those filings, the FY 2080/81 ERC dividend rejection of September 27, 2024 and the FY 2081/82 board proposal of December 23, 2025, CARE Ratings Nepal's February 2026 reaffirmation of CARE-NP BB+/A4+ on credit watch with positive implications for facilities of Rs 7,120.54 million, bankingnews on the Naugad Gad tax holiday ending August 19, 2025, investopaper on the Marsyangdi Besi acquisition decision and the pending 1:0.3 rights, Kantipur/HT syndication on the November 2025 protest halt, ShareHub for the 0% promoter/100% public line, market cap and 52-week figures, Damodaran's country risk premiums for the Nepal equity risk premium of 8.89%, and market data from the October 16, 2026 session with moving averages, RSI14, MACD, and returns computed by the desk on its own OHLC archive of 2,506 sessions। The valuation, multiples, and scenario work were compiled by the research desk from those sources; no figure was invented, and no corporate-action percentage was calculated where the official notice did not state one।
This article is published for research and education purposes only। It is not investment advice, and nothing here should be read as a recommendation to buy, sell, or hold any security। Figures reflect publicly reported information at the time of writing; quarterly figures are unaudited and annual figures may be restated। Valuation sketches are illustrative। For transparency: the author holds shares of this company (API) in his personal portfolio। Always read the company's own filings and consult a licensed advisor before investing।
Frequently asked questions
What is API's P/E ratio?
About 45 times on core operating EPS of Rs 7.63 — reported PAT of Rs 960.61 million less the Rs 474.15 million paper fair-value gain, over 63,797,242 shares — at the October 16, 2026 close of Rs 345.00। On the reported headline EPS of Rs 15.06 it would be 22.9 times, but the desk considers that figure FVTPL-inflated and used the core number throughout। The hydro-peer median is 67.16 times।
What is the Rs 474.15 million paper gain?
An FVTPL fair-value gain under NFRS 9, booked in FY 2082/83 on the company's 25% stake in Ingwa Hydropower Limited, carried at Rs 343.80 a share। It is a mark-to-market revaluation: about 49% of reported net profit, with no cash behind it, and it reverses if the valuation falls। It is the same kind of gain that got the FY 2080/81 dividend proposal rejected by the Energy Regulatory Commission।
What is the research desk's target price for API?
The blended base target is Rs 262, with a bear case of Rs 226 and a bull case of Rs 295, from a 50% DCF, 25% peer P/B, 25% peer P/E blend on core earnings। That implies negative 24.1% against the Rs 345.00 close, and the desk's mechanical rating rule reads REDUCE between negative 30% and negative 10%। This is the research desk's published view, not a recommendation to any reader।
What about the pending 1:0.3 rights issue?
The board proposed a 1:0.3 rights issue of 19,139,173 units at par on December 23, 2025, with Rs 191.39 crore earmarked as equity for the Marsyangdi Besi acquisition। As of October 17, 2026, AGM endorsement is unconfirmed and no SEBON approval has been seen, so it sits outside the base share count। If issued at par with proceeds ignored, the base DCF per share would fall to about Rs 64 — but the proceeds fund an unmodelable acquisition, so the note stays qualitative।
Has Api Power paid dividends?
Regularly, though bonus-heavy। The last two years each carried 5.2632% — 5% bonus plus 0.2632% cash — and the nine-year average is 6.83% a year। The FY 2079/80 year carried none, and the original FY 2080/81 proposal of 9.4737% was canceled after the ERC rejected it over FVTPL fair-value gains। Cash payouts are token; distributions grow the share count, not shareholder cash।
Who researched this company analysis?
The underlying fundamental, technical, peer-multiple, and valuation work was compiled by the research desk from the company's filings at apipower.com.np, the SEBON and ERC records of its corporate actions, CARE Ratings Nepal's rating actions, the NEPSE listing record, and market data as of the October 16, 2026 close, with every metric source-tagged and gaps flagged rather than filled। This article reads those numbers against NEPSE pricing for educational purposes।