Use Rooppur to break the capacity-payment trap

Bringing Rooppur into commercial operation expeditiously can give the BNP government the leverage to restructure a power sector that is draining public finances, consuming scarce foreign exchange and raising manufacturing costs.

Its strategic value extends beyond generating electricity. By reducing dependence on expensive US dollar capacity payment based power plants, Rooppur can strengthen the government’s hand in negotiating amicable exits from unnecessary contracts and better terms for plants the country still needs.

And the best time for this negotiated settlement is this winter when demand drops considerably in Bangladesh. Unfortunately, while not entirely deliberate, commissioning delays of Rooppur is pushing the current government into dangerous territories by next summer.

Without contractual reform, Bangladesh risks paying for nuclear electricity while continuing to finance idle oil-fired plants. The objective should be clear: turn additional power supply into lower total costs for households, industry and the national budget.

The scale is substantial. The government-commissioned White Paper identified 41 independent power producer contracts, representing 6,308 megawatts, awarded under the Quick Enhancement of Electricity and Energy Supply (Special Provisions) Act, 2010.

Separately, the National Review Committee estimated that inflated capacity charges were costing Bangladesh up to $1.5 billion annually. That is a committee estimate, not an adjudicated recovery-but it demands a serious response.

The government should distinguish two procurement groups.
Projects awarded through genuine competitive international tenders under Bangladesh’s established procurement framework deserve recognition for the stronger transparency, fairness and price discovery that competition provides.

If changing demand makes such a plant uneconomic, the government should negotiate a voluntary commercial settlement while protecting confidence in properly conducted procurement.

Projects awarded under the 2010 special law, particularly without competitive international bidding, require more intensive scrutiny of demand assumptions, pricing, ownership and risk allocation.

The law’s repeal in November 2024 did not automatically extinguish its contractual legacy. Nor does a special-law award alone prove corruption.

The question is whether the project served credible grid needs at reasonable cost, or whether politically connected interests-the “winning coalition”-displaced the public interest.

Actual award records must determine classification; a tender for construction equipment does not establish that the power-purchase agreement itself was competitively awarded.

For contracts whose continuation demonstrably harms the national interest, amicable termination should be the preferred route where it delivers net savings.

Necessary plants should instead face negotiations over fixed charges, currency indexation and payment structures. Every settlement must account for verified arrears, outstanding debt, lender consent, compensation and replacement supply. The public needs a cheaper electricity system, not an expensive arbitration followed by another bill.

Pakistan demonstrates that negotiated change is possible. In October 2024, it agreed early exits for five plants totalling 2,463 megawatts.

The announced saving was PKR411 billion over their remaining contract lives-not every year. In January 2025, its cabinet approved revised agreements with another 14 IPPs, projecting annual savings of PKR137 billion.

These figures describe separate government announcements and should not be casually added together.

Pakistan also shows why labels can mislead. Its revised Nishat Power arrangement retained part of the equity return as fixed remuneration and linked further entitlement to output.

It was a hybrid settlement, not the abolition of every capacity payment. Bangladesh should negotiate actual tariff components rather than assume that calling an agreement “take-and-pay” removes its liabilities.

Dollar indexation is the nail in the coffin for a power sector already burdened by expensive fuel and excessive capacity commitments.

The sharp depreciation of the taka has turned fixed dollar-linked charges into escalating domestic liabilities.

The exchange rate rose from Tk84.80 per dollar at end-June 2021 to Tk122.77 at end-June 2025: an unchanged dollar obligation became 44.8 per cent more expensive in taka, before inflation adjustments and without buying any additional electricity. Removing unjustified dollar indexation must therefore be central to negotiated reform.

The supplied CLC agreement illustrates the mechanism. Its original years 1-10 capacity tariff includes dollar-denominated components converted into taka, alongside a local component.

Applying those terms to 108 megawatts, with inflation factors held constant, produces an illustrative annual capacity bill of Tk124.8 crore at the earlier exchange rate and Tk176.2 crore at the later one. The Tk51.4 crore difference buys no additional capacity.

These are model calculations, not CLC’s actual invoices.

The foreign escalable component also adjusts for contractual inflation, compounding the currency effect. Where debt repayments and underlying costs are genuinely in taka, dollar indexation can increase revenues without a corresponding foreign-currency expense.

BPDB should audit each plant’s financing and negotiate taka-based domestic-cost components. Local ownership alone does not prove taka financing. Nor does freezing conversion at today’s rate reverse previous increases: immediate savings require a lower negotiated base payment.

Careful contract administration matters equally. CLC Section 13.2(j) permits suspension of the producer’s performance when qualifying invoices remain unpaid beyond ten business days after their due date, while preserving capacity payments during that suspension.

The clause contains no express separate suspension-notice prerequisite, and the agreement protects against waiver through delayed enforcement.

Deliberate nonpayment is therefore a poor bargaining strategy. BPDB should reconcile invoices, disputes, payments, tested capacity and operating records, then negotiate from verified facts.

Contracts should also be grouped by remaining life. Near-expiry plants are candidates for non-renewal or energy-only extensions. Mid-life agreements may justify discounted buyouts or reduced fixed payments. Longer commitments require careful financing and tariff restructuring.

Original terms suggest CLC’s expiry around February 2032 and Summit Gazipur II’s around May 2033, subject to amendments and confirmation of their current contractual status.

Running fewer plants more intensively can help, but only if the unnecessary payment obligations also end. Moving the same generation from two plants to one does not reduce the combined capacity bill when both contracts remain payable.

Nor should BPDB burn expensive HFO simply to spread a fixed charge over more units. Dispatch should follow avoidable operating costs, subject to reliability and network constraints.

Rooppur can materially change that calculation. At an assumed mature-year capacity factor of 85 per cent, its two units, each rated at approximately 1,080 megawatts net, could generate about 16.1 terawatt-hours annually.

That is an operating scenario, not a commissioning forecast or proof that an equivalent amount of HFO capacity can immediately retire. Dependable output, transmission readiness and reserves must come first.

For short daily peaks, batteries increasingly challenge the claim that HFO is indispensable. IRENA reports that installed utility-scale storage costs fell 93 per cent between 2010 and 2024, reaching $192 per kilowatt-hour.

BloombergNEF reports stationary-storage battery-pack prices fell 45 per cent in 2025 to $70 per kilowatt-hour. These are different measures: packs exclude substantial installation and project costs.

Both nevertheless demonstrate why Bangladesh should test storage through competitive tenders.

Bangladesh should stop offering conventional, long-term, full year-round capacity-payment packages to HFO plants needed only briefly each day. It should procure the required peak service from storage, demand response and other dependable resources.

An illustrative 2,000-megawatt battery fleet with 8,000 megawatt-hours of usable energy could cover a four-hour peak, provided charging supply and network capacity are available. Batteries have charging losses, financing costs and degradation; they cannot solve a prolonged energy shortage merely by being installed.

Essential standby capability still has value. Any availability payment should therefore meet a demonstrated reliability need, be competitively priced and carry enforceable performance obligations.

Storage must compete on its full delivered cost, while existing HFO contracts must be amicably settled so that their fixed charges do not survive the replacement.

Rooppur offers the government a stronger alternative at the negotiating table. Bangladesh should use it to secure affordable power through transparent, differentiated and amicable settlements.

Simultaneously, government should seriously consider terminating US dollar indexation because US dollar saved is US dollar earned.