BPDB Must Own Bangladesh’s Base Load Power Demand

When you don’t own one, renting a home is perfectly sensible when you need a place quickly and lack the capital to making an outright purchase.

Renting five homes when you have one family is how you end up in a financial disaster. For a decade, Bangladesh’s Power Development Board has been that tenant.

Many power contracts require payments for available capacity even when electricity is not dispatched, with substantial exposure to dollar-linked costs.

Bangladesh had 28,919 MW of grid-connected capacity, including imports, in May 2026. The Power Division’s September update records maximum generation of 17,201 MW on 20 May-not peak demand.

Yet Power Grid’s hourly records show 3,557 MW of load shedding at 3pm on 10 September. Capacity on paper is no guarantee of electricity in the socket.

The sector has two problems at once: excess installed capacity and dependence on imported fuel. The Power Division reported 32,688 MW on 15 September 2026, but that broader total includes captive and renewable capacity and should not be treated as grid capacity.

IEEFA estimated a 61.3 per cent reserve margin for FY2024-25, or 58.6 per cent using derated capacity; these are historical estimates, not a measure of fuel-backed reserves available today.

Primary-energy import dependence rose from 47.7 per cent in FY2020-21 to 62.5 per cent in FY2024-25, while average power-generation costs rose 83 per cent.

Zero Carbon Analytics estimates that the 2026 fossil-fuel import bill could rise by $2.8 billion, or 30 per cent over 2025, if prices remain at its year-to-date average. That is a conditional projection.

Capacity payments are a major part of the burden. A June 2026 report citing BERC’s assessment projected overall capacity payments and rental charges of Tk 526.08 billion in FY2026-27, against Tk 482.60 billion estimated for FY2025-26.

These are broader totals than payments to private and rental plants alone, and are not audited year-end outturns.

The revised FY2024-25 power-subsidy budget was Tk 620 billion; BPDB’s Tk 556.6 billion revenue shortfall for that year was a separate, pre-subsidy measure, not an additional loss after that entire budget allocation.

The Financial Express reported payment delays of around 300 days in April 2026. Bangladesh Bank’s reference exchange rate was Tk 123.0227 per dollar on 6 October. Against an illustrative Tk 80 baseline, the same dollar obligation costs about 53.8 per cent more in taka.

In my view, dependence on debt-heavy contracts and foreign-currency obligations cannot be the foundation of national base-load security.

According to internal BPDB sources, disputes have arisen between BPDB and some heavy fuel oil (HFO)-based independent power producers (IPPs) over the invocation of suspension clauses in their power purchase agreements.

These sources contend that failure by certain IPPs to follow the prescribed contractual procedures when invoking those clauses may open the door to negotiated termination of their contracts.

That position remains disputed. One way to reduce excessive capacity payments would be to retain a limited number of essential IPP plants and aim to utilise 100 per cent of their available capacity whenever demand, fuel supply and maintenance requirements permit, while ending surplus contracts through amicable settlement.

The objective should be to concentrate generation in the plants the system actually needs and reduce the recurring cost of redundant capacity.

Fuel is the second weakness. Private furnace-oil plants generated electricity at an average cost of about Tk 27.5 per kWh in FY2024-25, according to IEEFA; its estimate of Tk 9.5 per kWh in capacity payments is a component of that burden, not another charge to add to the total.

Gas shortages also idle plants: an April 2026 summer-planning report put supply to power plants at about 900 million cubic feet a day against roughly 2,524 million needed to run the then 12,204 MW gas fleet, limiting expected output to about 5,200 MW.

These were planning-period figures, not a current operating snapshot. Longer-term LNG costs remain exposed: at a projected 730 billion cubic feet of annual imports, reported scenarios put the bill near $8.5 billion at $12 per MMBtu, or above $14 billion at $20.

In July, Petrobangla told S&P Global that QatarEnergy deliveries could fall from around 40 planned cargoes to 20 in 2026. On 23 September, Bangladesh approved two additional Qatar Energy Trading cargoes for November-an approval, not proof of delivery.

A capacity contract cannot guarantee fuel, shipping access or foreign exchange.
Nor can the India interlink carry the load alone. BPDB lists 1,496 MW from Adani’s Godda plant and 1,160 MW through Bheramara and Tripura, plus 40 MW from Nepal: total import capacity of 2,696 MW, also confirmed in the Power Division’s September update. India’s 2,656 MW is roughly 9 per cent of the May grid-capacity figure.

A July report quoted BPDB’s chairman as putting Adani’s FY2025-26 purchase cost at Tk 13.83 per kWh and reported a subsidy request of Tk 55.87 billion for FY2026-27.

The request should not be confused with an approved payment. In late 2024, Adani cut deliveries by half amid reported arrears of about $850 million. India is an important partner and the interlink a useful source, but no country should let a single supplier dispute threaten the electricity reaching its hospitals.

Two programmes point the right way. The government’s September statement set 2030 targets of 5,500 MW of rooftop solar, 4,500 MW of ground-mounted solar and 450-550 MW from other renewable technologies: about 10,450-10,550 MW in total. Eligible solar equipment and components received a conditional 1 per cent total import-tax facility for 180 days under the September announcement.

Rooppur’s two 1,200 MW reactors are another major prospective source of continuous power. Unit 1 fuel loading began on 28 April; project officials reported a successful safety-valve pressure test on 30 September, and a nuclear-fuel consignment reached the site on 3 October.

These milestones do not establish commercial grid operation. The $12.65 billion construction contract must be distinguished from the $11.38 billion Russian credit agreement signed in 2016. Meanwhile, the IMF confirmed a request for a new programme in June and continuing discussions in July; that is not confirmation of a newly approved financing arrangement. Fiscal discipline remains essential.

So, who is accountable for base load at three in the morning during summer? Responsibilities already exist, but I believe accountability for dependable supply needs to be clearer.

Nuclear and large coal stations can provide sustained output, subject to maintenance, fuel availability and operating constraints. Public ownership alone does not remove those risks. BPDB should have a statutory duty to secure dependable base-load supply, with roughly 13 GW as my proposed planning benchmark, to be tested against hourly demand, outages and reserve requirements. Nuclear, coal and domestic gas would form its backbone.

I favour BPDB ownership of the core fleet, but Rooppur’s designated operator is NPCBL, and Power Grid’s NLDC has the system-operation role: any institutional change must preserve specialist operation and independent nuclear regulation. Fuel should be financed before idle capacity is paid.

New private capacity should be limited to peaking or merchant sales, linked to the taka and free of capacity payments. If Pakistan can bring down reliance on HFO plants from 60% to 1.9%, Bangladesh should be able to as well.

One can argue, BPDB is itself in financial trouble. But that is an argument for reform and funding, not for allowing responsibility to remain fragmented.

A mandate needs matching cash, tariffs that follow cost, transparent dispatch and a grid ready to absorb nuclear power. Give BPDB that, and Bangladesh stops renting rooms it cannot use and starts owning the house.

Javed Hosein is a member of the Board of Trustees of Independent University, Bangladesh (IUB)