Summit Power valuation report: fair value BDT 7.6–18.5 (Oct 2026)

Thesis: Summit Power's PPAs run out by 2033, putting fair value at BDT 7.6-18.5, base 13.6

Summit Power Limited (DSE: SUMITPOWER) is Bangladesh's largest listed independent power producer (IPP) group by plant count. It runs 15 plants of 930.55 MW built since 2003, of which eight with 733 MW were operating in FY2025 [5 p.135] [3]. It earns money in two ways. Its HFO-fired subsidiaries sell power to the Bangladesh Power Development Board (BPDB) under 15-year power purchase agreements (PPAs). These pay a capacity charge for being available and an energy charge that passes the cost of furnace oil through. Its older gas plants sell to the Bangladesh Rural Electrification Board (BREB). The catch for shareholders is ownership. Summit owns only 20% of the 300 MW Kodda-2 plant, the group's biggest earner, 64% of Kodda-1, and 49% of Rupatoli and Madanganj-2 [2 p.18]. The parent, Summit Corporation, and partners own the rest. In FY2025 non-controlling interests took BDT 2,983mn of the BDT 3,391mn group profit, 88% [5 p.151] [5 p.199].

The latest period in this report is FY2026 (year to 30-Jun-2026), but the audited FY2026 statements are not yet filed. On 28-Oct-2026 the board announced consolidated EPS of BDT 1.04 (FY2025: 0.38), NAV per share of BDT 40.73, operating cash flow of BDT 5.81 a share and a 10.5% cash dividend (BDT 1.05), record date 19-Nov-2026 [3]. We label these 'DSE announcement, audited statements not yet filed'. Line items come from the FY2025 annual and the nine months to March 2026. The model will be updated when the FY2026 annual is filed.

We value Summit plant by plant. Each contract's equity cash flow runs to its PPA end, with no perpetuity, and is multiplied by Summit's share. We then add the 30% associate Summit Meghnaghat, the 17.64% KPCL stake, the parent's claims and idle-plant salvage. At a 16.7% cost of equity this gives BDT 14.42 a share. A dividend discount model (DDM) gives BDT 10.1. Weighted 80/20, base fair value is BDT 13.6, in a range of BDT 7.6-18.5 (probability-weighted 12.8). The price of BDT 15.7 on 29-Oct-2026 [4] sits inside the range, above the base. The base implies 8.3x FY2027F EPS and 0.33x the announced NAV. Book value overstates the cash the plants will return, because they are depreciated over 20-30 years against 15-year contracts [2 p.12].

Group: Summit owns 20-64% of the four HFO plants that earn most of its profit, and 30% of Meghnaghat

Summit Power was incorporated in 1997 and listed on the Dhaka and Chittagong stock exchanges in 2005 [2 p.10]. Its first plants were small gas engines at Ashulia, Madhabdi and Chandina, built for the Rural Electrification Board from 2003. A second wave of 33 MW gas plants followed in 2009 (Rupganj, Jangalia, Maona, Ullapara), then Madanganj-1, a 102 MW furnace-oil plant, in 2011. The third wave was large HFO plants for BPDB, built in separate companies with Summit Corporation and foreign lenders. These are Madanganj-2 (55 MW) and Rupatoli in Barisal (110 MW) in 2016, and Kodda-2 (300 MW) and Kodda-1 (149 MW) in Gazipur in 2018 [2 p.10]. Their capacity tariffs are set in USD, and the subsidiaries keep their books in USD [2 p.11].

Summit Corporation, the parent, holds the controlling stake in Summit Power. It also owns 51% of Summit Barisal and Summit Narayanganj II, 36% of Ace Alliance and 80% of Summit Gazipur II [2 p.42]. Summit Power consolidates all four because its management runs them, so the group accounts show their full revenue and debt. Most of their profit, however, belongs to Summit Corporation. Summit Power acquired its 30% of Summit Meghnaghat Power, a 335 MW gas and HFO combined-cycle plant, from Summit Corporation in 2014 by issuing its own shares [5 p.147] [2 p.18]. It also holds 17.64% of Khulna Power Company (KPCL), carried at market value, BDT 750mn at Tk 10.70 a share [2 p.20]. Furnace oil for the HFO plants is bought through Summit Oil & Shipping, a group company that receives advances. The plants are run by Summit's own staff. CRISL rated the company AA+ in July 2026, down from AAA a year earlier [3]. A new managing director, Maj Gen (retd) Dr Monirul Islam Akhand, took over in January 2025 [3].

Track record: EPS fell from BDT 5.25 in FY2021 to 0.38 in FY2025 as PPAs expired

Summit's record shows the cost of contract expiry. EPS was BDT 4.78-5.25 in FY2019-FY2021, when all the BREB and BPDB gas plants were inside their first contracts [6 p.141]. It fell to 3.87 in FY2022 and 2.07 in FY2023, when the BDT's fall against the USD made imported furnace oil cost more than the HFO revenue that BPDB repaid. It recovered to 3.13 in FY2024 as that mismatch eased [3]. Then it collapsed to 0.38 in FY2025, when seven plants with 234 MW were shut or ran on 'no electricity, no payment' (NENP) terms and BDT 1,380mn of impairment was booked [5 p.151] [3].

The dividend tells the same story: BDT 3.50 a share for FY2021, 2.00 for FY2022, then 1.00-1.05 a year since [6 p.201] [7 p.192] [3]. The payout has run above 100% of EPS in FY2025 and FY2026 because the parent pays out of retained earnings and subsidiary dividends. Owners' return on equity fell from about 15% in FY2021 to 1% in FY2025 (company figure) [5 p.131]. Our forecast's 4-5% owners' ROE on the announced NAV sits inside that range, near its low end. That is because the plants left are the 20-64% owned ones.

Cash flow is the better guide to dividends than EPS. Consolidated operating cash flow was BDT 7,516mn in FY2023 [6 p.173] and 10,447mn in FY2025 [5 p.199], but the parent's own operating cash flow was only 1,695mn in FY2025 against a dividend bill of about 1,121mn [5 p.199]. The parent relies on dividends from the subsidiaries and Meghnaghat: BDT 768.5mn in 9M FY2026 [2 p.35]. Every cut in the subsidiaries' earning years therefore reaches Summit's shareholders through those dividends, with a lag, and only at Summit's ownership share.

Plant registry: 614 MW of HFO contracts end between 2031 and 2033; the gas plants end by 2028

The registry below is taken from the plant table and tax-holiday note of the March 2026 interims [2 p.10] [2 p.48]. All four HFO plants are in subsidiaries that Summit consolidates because its management runs them, even where it holds under half (IFRS 10) [2 p.42]. The parent's own gas plants are on their second or third contract. Ashulia-1 and Madhabdi-1 hold NENP extensions to November 2028 but have had no gas connection [2 p.10]. Rupganj, Maona and Ullapara (77 MW) were shut by BREB from 1-Apr-2025 [9]. Jangalia (33 MW) has had no renewal since June 2024 [2 p.11]. Madanganj-1 (102 MW) is being sold for re-export to Sabson Energy of Dubai [3].

Utilisation: generation fell 22% to 2,051 GWh in FY2025 as BPDB called oil plants less

Net generation fell to 2,051 GWh in FY2025 from 2,622 GWh in FY2024 [5 p.135]. On 733 MW of operating capacity that is a 32% plant factor. BPDB dispatches by merit order and by how much fuel money it has. HFO plants are expensive, so it runs them least when demand is soft or when it is short of cash. Kodda-1 and Kodda-2 generated 565 GWh and 496 GWh, about 43% and 19% of capacity. Rupatoli made 88 GWh, a 9% factor [5 p.40]. Low dispatch matters less for an HFO plant than it looks. The capacity charge is paid for availability, and the energy charge passes fuel through. What falls with dispatch is the HFO pass-through revenue, plus the small margin on the operating and maintenance part of the energy charge.

The nine-month figures show the same pattern in FY2026. Group revenue fell 27% to BDT 23,601mn, but gross profit rose 3% to BDT 5,749mn [2 p.3]. HFO pass-through revenue fell to BDT 12,910mn from 19,654mn, while electricity (capacity and energy) revenue fell less, to BDT 10,529mn from 13,247mn. That fall is mostly the three BREB plants shut in April 2025 [2 p.34]. Gas plants are limited by Petrobangla's gas allocation. Ashulia-1 and Madhabdi-1 sit idle for lack of a gas connection [2 p.10], and in winter gas goes to fertiliser plants and households first.

Contracts: all subsidiary capacity revenue ends by FY2034, and NENP renewals pay energy only

The PPA end dates set Summit's value. Rupatoli's 15-year term ends on 4-Apr-2031 and Madanganj-2's on 28-Feb-2031. Kodda-2's ends on 9-May-2033 and Kodda-1's on 11-Jul-2033 [2 p.48]. The tax holiday ends on the same dates, so any post-PPA profit is taxed at 27.5%. Summit's own history shows what happens at expiry. Rupganj, Maona and Ullapara were renewed on 'no electricity, no payment' terms in 2024, ran for about ten months without capacity revenue, and BREB then shut them from 1-Apr-2025 [3] [9]. Madanganj-1 ran under NENP from May 2024 but was not dispatched after August 2024. It is now being sold [3]. Since the FY2023-24 budget, the government's policy is that renewals carry no capacity payment [12]. Private power producers' earnings are under pressure from that shift, as reported by the Financial Express [10].

So we follow the F23 post-PPA rule. Base: each plant runs one NENP year earning 25% of its contract EBITDA (energy only), then is sold for 10% of its gross cost. Bear: shut at expiry with 5% salvage. Bull: renewal for five years at half the old capacity rate (60% of contract EBITDA). One more risk sits inside the contract years. The interim government is reviewing IPP capacity terms. We take a 5% haircut on the capacity part of EBITDA in the base, 10% in the bear and none in the bull. A 25% haircut alone would cut fair value by BDT 2.53 a share (Ex. 30).

Receivables: BPDB owes BDT 18.6bn on the HFO plants, funded with BDT 15.7bn of short-term loans

BPDB paid slowly through FY2026. Its receivable on the four HFO plants rose to BDT 18,564mn in March 2026 from 13,853mn in June 2025. Kodda-2 alone accounts for 9,728mn [2 p.21]. That is about 260 days of their annualised revenue. Summit funds the gap with bank loans. Group short-term loans rose to BDT 12,422mn from 8,366mn, and deferred LC liabilities for furnace oil reached 3,283mn [2 p.29]. The parent borrows on behalf of the subsidiaries and passes the interest to them under the fuel supply agreements. That interest was BDT 1,201mn in nine months, about BDT 1.6bn a year [2 p.36]. In the model this carry cost falls on each subsidiary in proportion to its receivable. The bear case lifts it 25% (about 60 more days outstanding); the bull cuts it 25%.

The parent has two receivables we treat as claims, not cash. First, BDT 6,956mn of BREB invoices for three expansion plants, billed at the bulk-supply tariff. BREB pays only the original Tk 2.83/kWh. The Appellate Division ruled for BREB on jurisdiction in March 2023, and Summit's review petition is pending [2 p.21]. We assume 30% is recovered in FY2029 (bear nil, bull 60%). Second, BDT 2,427mn owed by BPDB for Madanganj-1, against which BDT 312mn is provided after an excess-fuel dispute [2 p.21]. We assume 70% is recovered. The auditor has added emphasis-of-matter paragraphs for both FY2024 and FY2025 [3].

Project debt is short and shrinking. USD project loans from IDCOL, ICD, OFID, SMBC and Clifford fell to BDT 13,913mn from 16,527mn in nine months, with final instalments in 2030-31, before the PPAs end [2 p.28]. Interest-rate swaps fix most of the Kodda loans. One related-party item needs watching. Advances to Summit Oil & Shipping, the group's own fuel importer, were BDT 3,064mn, plus BDT 730mn of other receivables [2 p.23] [2 p.21]. They fell from 7,560mn in June 2025.

Does the debt outlive the contracts? No. Every project loan matures in 2030-31, two to three years before the Kodda PPAs end, and the lenders hold sponsor guarantees from Summit Power (Barisal, Narayanganj II, Ace Alliance) or Summit Corporation (Gazipur II) [2 p.28]. The risk is in the short-term loans instead. They are rolled with local banks (City, Eastern, Bank Asia, BRAC, Mutual Trust, Prime, Standard Chartered) at market rates [2 p.29], and they grow whenever BPDB pays late. At the PPA ends we assume the receivables are collected and those loans repaid, which releases working capital to the subsidiaries' owners. The bear case collects only 85%. Dividend capacity at the parent is therefore the subsidiaries' dividends less head-office costs and tax, plus Meghnaghat. Summit's attributable equity cash flow in FY2027F is BDT 1.4bn, against a dividend of BDT 1.1bn at 10.5%.

Industry: the state stopped paying capacity on renewals, which caps every IPP's life at its first PPA

Bangladesh built a large fleet of private oil and gas plants in 2009-2018 on 15-year PPAs that guaranteed a capacity charge. BPDB is the single buyer, and the treasury funds its losses. In the FY2023-24 budget, the government said renewed rental and IPP contracts would carry no capacity payment [12]. Listed IPPs (Summit, United Power, Doreen, Baraka, Shahjibazar) have since reported falling earnings as their first contracts ended, as reported by the Financial Express [10]. Summit's own peer table puts the five listed IPPs' FY2025 EPS between BDT 0.36 and 20.66 [5 p.72].

For Summit this means two things. Renewals are worth an energy margin at best, which is why our base allows only one NENP year. And BPDB's cash position drives receivable days and short-term loan costs. Furnace-oil plants also face a policy push toward gas, LNG and renewables, which lowers their dispatch further. We have no T1 source for system installed capacity or total capacity charges in the corpus; we list it as a gap rather than quote a single outlet.

Three features of the market set the outlook for every listed IPP. First, dispatch: oil plants are called last, so they run at low plant factors, as Summit's 9-43% factors at its HFO plants in FY2025 show. Second, fuel: furnace oil is imported and priced in USD, so the falling taka raised the cost of HFO generation. Gas from Petrobangla is cheaper but scarce, and LNG imports depend on foreign exchange. Third, the buyer's cash: BPDB sells power below its cost, waits for government subsidy, and pays generators late. That is why Summit's HFO receivables rose by BDT 4.7bn in nine months even while dispatch fell. For a valuation this points one way. The remaining contract years are worth something, renewals are worth little, and collection delay is a cost that falls on the generator through its short-term borrowing.

Earnings quality: owners kept 12% of FY2025 group profit and 41% of 9M FY2026's

Three items separate Summit's reported profit from what shareholders own. First, non-controlling interests. In 9M FY2026 group profit was BDT 3,868mn but owners got 1,574mn (41%) [2 p.4]. Kodda-2, 80% owned by Summit Corporation, makes the most money. Second, IFRS 16 straight-lining. Capacity revenue is booked evenly over each PPA, and the cumulative difference, BDT 5,939mn, sits as an 'other asset' [2 p.20]. Kodda-1 and Kodda-2 have booked revenue ahead of billing (BDT 7.2bn); Rupatoli and Madanganj-2 the reverse. That asset unwinds to nil by 2033 and does not add cash. Third, impairments come at the year end. The BDT 1,380mn charge in Q4 FY2025 followed a third-party valuation of seven idle plants [5 p.151]. FY2026's Q4 loss of BDT 0.43 a share is unexplained until the annual is filed.

Cash covers profit at group level. Operating cash flow per share was 7.04, 6.13, 9.78 and 5.81 in FY2023-FY2026 against EPS of 2.07, 3.13, 0.38 and 1.04 [3]. But that cash is consolidated, and most of it belongs to the subsidiaries' other owners. Summit's own cash comes from subsidiary and associate dividends: BDT 768.5mn in 9M FY2026 [2 p.35]. Governance flags: the fuel importer is a sponsor affiliate holding BDT 3.1bn of advances, and the parent guarantees subsidiary project loans [2 p.28]. Dividends from tax-exempt subsidiaries are treated as tax-free on legal advice [2 p.48].

Forecast: EPS of BDT 1.63 in FY2027F, holding near 1.7 to FY2031F before the PPAs end

FY2026 first. The nine months gave owners BDT 1,574.3mn [2 p.4]. In the last two years Q4 earned about 0.92 times the average of the first three quarters, before impairments (FY2024: EPS 0.73 against 0.80; FY2025: profit 1,126mn excluding the impairment, against 1,215mn) [5 p.151]. That seasonal Q4 would add BDT 483mn. The DSE-announced EPS of 1.04 means owners' profit of BDT 1,111mn. That leaves a residual of BDT -946mn (-0.89 a share), which we show as its own labelled line, not spread across drivers. Its likely cause is a year-end impairment or provision, as in FY2025, but the audited statements are not yet filed.

The forecast is built plant by plant. Each HFO plant's EBITDA starts from its 9M FY2026 revenue less fuel, less the subsidiaries' other plant costs and overheads split by MW, times the full-year factor. Kodda-2 earns BDT 4.6bn a year, Kodda-1 1.8bn, Rupatoli 1.5bn and Madanganj-2 0.8bn. Then come the drivers: the contract-review haircut (5% base), dispatch on the energy part, project-loan interest at 6.5% on balances repaid by 2030-31, and the short-term loan carry. Each plant's profit is multiplied by Summit's share, and we add the parent's gas plants (ending FY2027-FY2029) and 30% of Summit Meghnaghat. Building up 9M FY2026 this way gives BDT 382mn more owners' profit than reported. The gap is parent tax (231mn), financing fees and loan-cost amortisation (151mn), straight-lining (+162mn), finance income (+106mn) and an allocation residual. We carry it at its actual 9M rate as a named reconciling line, scaled down as contracts end.

Result: owners' profit of BDT 1,744mn and EPS of 1.63 in FY2027F, against the 9M FY2026 run-rate of 1.93. The gap is the 5% capacity haircut (about BDT 0.15), the end of Madhabdi-2 and Chandina-2, and the reconciling line, offset by lower project-loan interest. EPS stays at 1.68-1.78 to FY2030F, falls to 1.62 in FY2032F after Rupatoli and Madanganj-2 end, and to 1.05 in FY2034F. Only Meghnaghat (BDT 0.83) remains after that, to its assumed FY2036 end. The model carries the reported depreciation rates. At expiry, Kodda-2 would still hold about BDT 10bn of book value against our salvage of 2.3bn, so a large impairment is likely around 2033. It is a book loss, not a cash cost.

Plant by plant, the forecast reads as follows. Kodda-2 (300 MW, 20% owned) earns contract EBITDA of about BDT 4.5bn a year after the haircut to FY2032, 3.8bn in FY2033 when its PPA ends in May, and 1.2bn in its NENP year. Summit's 20% of its equity cash flow is BDT 210mn in FY2027F. Kodda-1 (149 MW, 64%) earns about BDT 1.7bn a year to FY2033. Its project loan, the longest, is repaid by FY2031. Summit's share of its equity cash flow is BDT 179mn in FY2027F, rising as the loan runs off. Rupatoli (110 MW) and Madanganj-2 (55 MW), both 49% owned, earn BDT 1.4bn and 0.8bn a year until their 2031 expiries, then one NENP year. Their net working capital, about BDT 0.8bn of it Summit's share, comes back in FY2032. The parent's four running gas plants earn little: about BDT 0.5bn a year of plant EBITDA before head-office costs, falling to nil by FY2029 as Madhabdi-2, Chandina-2, Ashulia-2 and Chandina-1 reach their contract ends. The parent's cash flow is lumpy because of the claims: we assume BDT 1.5bn from Madanganj-1 and 2.1bn from idle-plant sales in FY2028, and BDT 2.1bn of the disputed BREB invoices in FY2029. That lumpiness shows in the DDM dividends, which jump in FY2028-29. Dividends follow cash, not profit: the FY2028F DPS of 2.68 exceeds EPS of 1.68 because depreciation, claim collections and the run-off of the contracts' working capital release cash that never passes through earnings.

The balance sheet follows the same drivers. Project loans of BDT 13.9bn are repaid by FY2031 from the plants' cash, and short-term fuel loans run off as BPDB pays and contracts end. The group should be close to debt-free at the subsidiary level by FY2034. Owners' equity changes little in the forecast: earnings of BDT 1.7-1.9bn a year are broadly paid out. When the plants are written down to salvage value around 2031-2033, NAV will fall towards the cash and claims that remain. That is why we do not value Summit on P/B: its NAV of BDT 40.73 is mostly plant book value that the contracts will not recover.

Valuation: plant SOTP and DDM give BDT 7.6-18.5; base 13.6 is 8.3x FY2027F EPS and 0.33x NAV

Method. Each plant's cash flow to equity (EBITDA, less tax after the holiday, maintenance capex, project-loan interest and principal, and short-term loan carry), plus working capital released after the PPA ends and salvage, is discounted at the cost of equity and multiplied by Summit's share. There is no terminal value: the contracts end and the base case shuts the plants after one NENP year. A separate sum of the parts is required here because the units differ in ownership (20-100%), contract end and risk. The skill's SOTP rule applies to that difference, and the parts are the plants themselves. Cost of equity is 16.7%: the 10-year treasury bond cut-off yield of 9.27% (25-Oct-2026) [8] plus a Blume-adjusted beta of 1.06 times a 7% equity risk premium. The raw weekly beta against the DSEX over 104 weeks is 1.09, R-squared 0.30 [4]. F23's 0.7 floor does not bind.

How the legs reconcile. The SOTP's per-share value comes mostly from cash the plants pay out before 2034, so it behaves like a bond with a shrinking coupon. At the base fair value, FY2027F EPS of 1.63 implies a P/E of 8.3x. That is low by DSE standards, and it should be, because earnings fall to nil within about eight years. A no-growth perpetuity of FY2027F EPS at 16.7% would come to BDT 9.8. The gap to our base is the value of the years after the PPAs end, which the market should not pay for. The DDM pays out only 60% of attributable cash and compounds the rest at 6.2%, so it sits below the SOTP. Its 20% weight reflects how DSE investors actually value Summit: on the dividend.

Summit Meghnaghat is the largest single piece, at BDT 4.25 a share. Its 30% share of profit was BDT 879mn in FY2025 [1 p.21] and BDT 710mn in nine months of FY2026, and it paid Summit BDT 600mn [2 p.18]. Its PPA tenor and statements are not in our corpus, so its FY2036 end date is a labelled assumption and a gap. Kodda-1, of which Summit owns 64%, is worth more to shareholders (BDT 1.93) than Kodda-2 (BDT 1.86), which earns 2.6 times as much but is 80% owned by Summit Corporation.

The DDM pays 60% of attributable cash earnings each year, the median of Summit's FY2021-FY2026 payouts. The retained 40% earns the 6.21% bank deposit rate [8] and is paid out in FY2039. It gives BDT 10.1, below the SOTP by 4.32, because cash retained at 6.2% loses value against a 16.7% cost of equity. We weight SOTP 80% and DDM 20%. F23 asks for 50/30/20 including a P/E leg, but there is no peer set on the same attributable basis from peers' own filings, so we drop that leg.

Post-PPA value is small. Setting every plant to shut at expiry, with 5% salvage, lowers fair value by only BDT 0.4 (2.9%). The working capital released after each PPA ends is larger, because BPDB's arrears and fuel stock come back as cash. The price of BDT 15.7 is about what the base gives with renewals at half the old capacity rate and a 1pp lower cost of equity. In other words, the market is paying for some renewal value or a better outcome on the BREB claim. At 38.5% of NAV and a 6.7% yield on the BDT 1.05 dividend, the stock trades on its dividend, not its book.

Risks: a 25% capacity haircut costs BDT 2.53 a share, losing the BREB claim 1.26

Each risk below names the model line it moves, and the bear case moves the same drivers by the same amounts. The bear also takes a 10% haircut, shuts plants at expiry, collects nothing on the BREB claim and only 40% of Madanganj-1, realises 25% of idle-plant book and 85% of working capital. Because Q4 has lost money two years running (the BDT 1,380mn impairment in FY2025 and the BDT 946mn residual in FY2026), the bear also carries a recurring Q4 charge of BDT 1,163mn, their average, in FY2027, fading to 775mn and 388mn and out by FY2030. It is a stack of plausible moves, and it is deeper than any single shock except the 25% haircut stress.

The largest single risk is a renegotiation of the Kodda contracts. A 25% cut to the capacity part of EBITDA on every remaining PPA year, and on Meghnaghat, lowers fair value to BDT 11.0. We treat this as a stress, not the bear, because the contracts are USD-indexed agreements with foreign lenders behind them. The second is the BREB claim. It is the parent's own asset, so it reaches shareholders at 100%. Losing it takes BDT 1.26 off fair value. The third is collection. If BPDB stretches payment by about 60 days for three years, the extra short-term loan interest costs BDT 0.39 a share. On the upside, a BPDB settlement through special bonds would cut that carry. A bull-case renewal of Kodda at half the old capacity rate would add about BDT 1.2 a share.

• Related-party fuel: BDT 3.8bn owed by Summit Oil & Shipping (advances plus receivable). A loss there would hit the subsidiaries' working capital, which is mostly owned by others; Summit's share is about 40%.
• Sponsor control: Summit Corporation owns 80% of Kodda-2 and 36% of Kodda-1, and decides subsidiary dividends.
• Book value: about BDT 10bn of Kodda-2 book would remain at PPA end; expect impairments around 2031-2033.
• Meghnaghat: its PPA end and own statements are not in our corpus; each year less of contract life costs about BDT 0.3 a share.
• Gas: Petrobangla's allocation limits the parent's gas plants; two units already sit idle.

What to watch: the FY2026 annual's Q4 detail, the 19-Nov-2026 record date and the BREB review petition

Fuel supply is the operational risk to watch. The HFO plants depend on imported furnace oil bought through Summit Oil & Shipping and paid for with letters of credit. When dollars were scarce in FY2023, the cost of HFO ran ahead of what BPDB reimbursed, and EPS fell to 2.07. In FY2025 the taka was steadier and the subsidiaries' operating profit rose BDT 694mn [5 p.135]. The gas plants depend on Petrobangla. Two units already sit idle without a gas connection, and gas shortages in winter cut output at the others. Because the gas plants are a small and shrinking part of value, the HFO fuel chain matters more.

• Record date 19-Nov-2026 for the BDT 1.05 dividend; AGM 30-Dec-2026, held digitally (DSE).
• FY2026 audited annual: the cause of the BDT 0.89 a share Q4 residual, the June-2026 BPDB receivable and any new impairment.
• Q1 FY2027 results (due about November 2026): BPDB collections and short-term loans.
• Madhabdi-2 and Chandina-2 contract ends (Nov-Dec 2026): renewal on NENP terms or shutdown.
• BREB review petition and the Madanganj-1 sale proceeds.

Signposts that would change the range: a BPDB settlement that cuts HFO receivables below BDT 12bn (bull carry); any renewal offer for Rupatoli or Madanganj-2 before 2031, and its capacity terms; a court outcome on the BREB invoices; a higher Meghnaghat dividend or disclosure of its PPA end; and the size of any further impairment in the FY2026 annual. A contract-review announcement that names capacity cuts for the Kodda plants would move us toward the bear case.

Model update note: this report uses the FY2025 audited annual, the 9M FY2026 interims and the FY2026 figures announced on DSE on 28-Oct-2026 (audited statements not yet filed). The model will be updated when the FY2026 annual is filed [3].

FAQ

What is Summit Power's fair value range?

BDT 7.6-18.5 a share, base 13.6 (probability-weighted 12.8), from a plant-by-plant SOTP to each PPA end at Summit's ownership (80%) and a DDM (20%) at a 16.7% cost of equity.

When do Summit Power's PPAs expire?

Madanganj-2 on 28-Feb-2031, Rupatoli on 4-Apr-2031, Kodda-2 on 9-May-2033 and Kodda-1 on 11-Jul-2033; the parent's BREB gas plants end between Nov-2026 and Nov-2028.

How much of Summit's subsidiaries does it own?

20% of Summit Gazipur II (Kodda-2), 64% of Ace Alliance (Kodda-1) and 49% each of Summit Barisal and Summit Narayanganj II; Summit Corporation owns most of the rest. It also holds 30% of Summit Meghnaghat and 17.64% of KPCL.

What did Summit Power earn in FY2026?

The board announced consolidated EPS of BDT 1.04, NAV of BDT 40.73 and a 10.5% cash dividend on 28-Oct-2026; the audited statements are not yet filed.

What is 'no electricity, no payment'?

A renewal term under which the plant is paid only for energy actually dispatched, with no capacity charge; Summit's Rupganj, Maona and Ullapara plants ran on it for about ten months before being shut in April 2025.

Sources

1. Summit Power Limited audited financial statements, year to 30-Jun-2025 · link · accessed 2026-10-31
2. Summit Power Limited un-audited financial statements, nine months to 31-Mar-2026 · link · accessed 2026-10-31
3. DSE news archive for SUMITPOWER, Nov-2024 to Oct-2026 (old.dsebd.org) · accessed 2026-10-31
4. AnalystBro chart_bars: SUMITPOWER and DSEX daily and weekly closes to 29-Oct-2026 · accessed 2026-10-31
5. Summit Power Limited Annual Report 2024-25 (full report) · link · accessed 2026-10-31
6. Summit Power Limited Annual Report 2022-23 (full report) · link · accessed 2026-10-31
7. Summit Power Limited Annual Report 2023-24 (full report) · link · accessed 2026-10-31
8. Bangladesh Bank treasury auction and scheduled-bank statistics pages, snapshot 29-Oct-2026 · accessed 2026-10-31
9. The Daily Star: 'Three Summit power plants to shut from April 1' · link · accessed 2026-10-31
10. The Financial Express: 'Listed private power producers face earnings pressure amid policy shift' · link · accessed 2026-10-31
11. AnalystBro model for this report (model.xlsx) · accessed 2026-10-31
12. The Business Standard: 'No capacity payment for contract renewal' (FY2023-24 budget) · link · accessed 2026-10-31