26 Gas, LNG and fuel security

Chapter 26 of 60 in the Bangladesh 2036 research base. Contents of the series.

Thesis

Fuel security is the physical form of the macro constraint chapter 10 names, and the stored record shows the country running its engines on a domestic gas base that fell from a 981.8 bcf peak in 2017 to 711.4 bcf in 2024 against consumption of 955.2 [EIA 2025], a 243.8 bcf gap filled by imported LNG, imported oil and imported coal, with the import volumes now measured: 6.87 million tonnes of petroleum in 2023 [BPC 2023], 4.55 million tonnes of coal burned for power in 2023 against 0.41 in 2011 [BPDB 2023], and diesel use in generation up roughly twenty-three-fold from 11.93 to 274.75 million litres across 2020-2023 [BPDB 2023]. The claim this chapter defends is that the fuel decade to FY36, the horizon the chapter 15 scenarios assume, is an option-pricing problem: the state has written itself a portfolio of dollar-indexed obligations, LNG spot and term contracts, coal supply agreements, petroleum imports and cross-border power, whose total exposure no stored series aggregates, and the decade is decided by whether that portfolio is repriced toward term contracts, domestic exploration and efficiency before the next global price cycle reprices it from outside.

Where Bangladesh stands

The gas decline is the chapter's baseline. The EIA dry gas series shows production at its 981.8 bcf peak in 2017 falling to 711.4 bcf in 2024, while consumption ran 955.2 bcf, a 243.8 bcf gap that LNG imports and demand compression filled [EIA 2025]. The reserves census gives the reason: gas initially in place of 38,211 bcf, 2P recoverable reserves of 29,737 bcf, cumulative production to June 2024 of 21,082 bcf, and remaining 2P reserves of 8,655 bcf, 29.1 percent of the recoverable base [Petrobangla 2024], with the single Bibiyana field, its field-level figures not established here, carrying about half of output, a share that could not be confirmed from the sources this chapter draws on. The power sector's own gas burn confirms the squeeze: 267,768 million cubic feet in 2020 falling to 227,924 in 2023 while generation rose [BPDB 2023], the gas engines demoting from baseload to first-reserve duty as the fuel thins. The Bhola discovery prospects and the offshore blocks in the Bay of Bengal are the supply hopes, and their prospectivity, reserves and development timelines could not be confirmed from the sources this chapter draws on; the Bapex and Petrobangla exploration reports would resolve them; the bidding rounds' status is likewise not established.

The LNG bridge is now the system's load-bearing import. Two floating regasification terminals of 500 million standard cubic feet per day each were contracted in 2018, with purchase agreements with Qatar for 1.8 to 2.5 million tonne per year and Oman for 0.5 to 1.0 [Petrobangla 2018]. The split of LNG purchases between term contracts and spot cargoes could not be confirmed from the sources this chapter draws on; the Petrobangla procurement disclosures would resolve it, and that split is the chapter's central risk variable: the 2022-23 spot price spike, which chapter 10 shows forced the generation cuts that left 3,818 million kWh unserved in 2023 [BPDB 2023], ran through the spot-exposed share of the portfolio, and the term share is the hedge the state has been buying since. The regasification capacity itself, two terminals at 500 mmscfd, is a third of the daily consumption the production decline has pushed onto imports, and any terminal outage is an immediate rationing event.

The fuel import bill is the measured scale of the dependency. BPC imported 6.87 million tonnes of petroleum in 2023, up from 5.19 million in 2020, and sold 7.35 million tonnes in FY23 [BPC 2023]. The power sector's own fuel escalation is the sharpest curve in the record: coal burned for generation rose from 0.41 million tonnes in 2011 to 4.55 in 2023 [BPDB 2023], diesel from 11.93 million litres in 2020 to 274.75 in 2023 and furnace oil from 389.07 million litres in 2021 to 700.9 in 2023 [BPDB 2023], the diesel spike being the 2022-23 crisis signature, plants that were designed for gas burning imported oil at spot prices to keep the lights on. The fertiliser connection runs through the same gas: urea plants feed on domestic gas, and the fertiliser subsidy of 16,500 crore BDT in FY25 [MoF Budget 2025] is partly the gas shortage's fiscal shadow, imported urea and imported gas both.

The pricing layer is the state's reconciliation machine. BPC returned to an operating profit of 5,859 crore BDT in FY23 on administered price adjustments [BPC 2023], the template chapter 04 and chapter 10 cite, and the gas pricing formula, with its BERC adjustments to industrial and power tariffs, could not be confirmed in its current schedule; the BERC documents would resolve it. The arrears pattern runs through the energy SOEs: the Petrobangla-BPDB-BPC triangle of unpaid bills is the cash-flow mirror of the price gaps, sized in chapter 41's SOE account and chapter 04's consolidated liabilities at 3.46 trillion BDT [MoF Budget 2023]. The rationing record is the allocation mechanism's output: industry gas rationing episodes in 2022-2023, which chapter 18's ceramics and glass lines absorbed, could not be confirmed in their measured volumes; the Titas and distribution company documents would resolve them.

The industrial stakes are the rationing record's other side. Fertiliser, ceramics, glass, steel and the captive power plants behind every garment factory are the gas consumers whose allocation failed in 2022-23, and the captive generation shift, factories moving to self-supplied oil and solar, is the measured consequence in the power data: the diesel and furnace oil curves above [BPDB 2023] are largely captive and peaking demand, not utility choice. Captive capacity's total could not be confirmed from the sources this chapter draws on; the distribution company connection registers would resolve it, and its economics are the efficiency scandal of the system: gas-fired captive at subsidised gas prices, or oil-fired captive at world prices, both behind meters the system does not centrally optimise. The grid-and-market reform that prices captive gas at opportunity cost is simultaneously the fuel security measure and the renewables enabler chapter 27 frames, because the rooftop solar business case improves with every honest gas price.

The household and agricultural priority is the allocation rule's political core. Households cooking on gas, the census records 5.89 percent of households cooking with piped supply gas and 8.65 percent with LPG [BBS Census 2022] against the 67.38 percent wood-using majority chapter 30's air account carries, and the boro irrigation pumps of chapter 16 sit at the front of every allocation queue, and the industrial cuts were the price of that priority. The LPG substitution market, the fastest-growing fuel retail segment, is the market response to the piped shortage, and its import growth is inside the BPC petroleum totals [BPC 2023]. The policy question is whether the queue is managed by political priority or by price, and the decade's allocation regime is the answer.

Mechanism

The mechanism that created the exposure was a demand build on a declining base. Gas was the economy's cheap input, industry, power and fertiliser all priced off it, and the expansion of the 2010s assumed the base would hold; production peaked in 2017 [EIA 2025], the demand kept compounding, and the difference became imports with a lag the contracts wrote. The state hedged with volumes, two terminals, term agreements, and left the price risk open: the spot share bought cheap when the world was glutted and ruinous when the cycle turned, and the 2022-23 episode is the worked example. The rationing that followed was rational portfolio management under crisis: cut the users with no contracts, industry, and protect the users with political claims, households and agriculture, at the cost of the industrial output chapter 18 measures.

The pricing mechanism is the transmission from world cycle to Bangladesh CPI. Every imported MMBtu, tonne and litre enters the cost stack at the exchange rate of the day, and the pass-through ran at administered speed: fuel prices adjusted in steps that returned BPC to profit [BPC 2023], gas tariffs adjusted with lags, and the residual gaps accumulated as the SOE arrears the consolidated account hides. The May 2024 float [BIS 2026] changed the mechanism's gear: with the exchange rate market-set, the domestic price of fuel moves with the world price continuously unless the state re-imposes a wedge, and the wedge is the subsidy-plus-arrears machine the decade must decide about.

The efficiency mechanism is the demand side nobody prices. The gas burn data, less gas for more generation [BPDB 2023], shows the power fleet's efficiency improving under duress, and the industrial efficiency margin, waste heat, compressed air, boiler tuning, is not established in aggregate but is the cheapest fuel the country owns. The 243.8 bcf gap [EIA 2025] prices the option: every percent of demand-side efficiency is imported fuel the state does not buy.

The arrears cascade deserves the mechanics stated, because it is the risk with a banking-system tail. The chain runs: BPC and Petrobangla import at world prices, sell at administered prices, book the gap as receivables from the power and agricultural users, the users' payables to the energy SOEs age, and the energy SOEs' own borrowing from the state banks, the exposure chapter 06's bank-sovereign nexus carries, funds the gap. The consolidated SOE account's 3.46 trillion BDT of liabilities [MoF Budget 2023] is the stock this flow built, and the energy triangle is its fastest-growing corner. Every tariff step that closes a price gap is an arrears prevention, and the accounting reform that publishes the triangle's ageing is the transparency the decade needs before the next crisis.

The regional trade option closes the supply map. The Nepali hydro and Bhutanese surplus that chapter 27's cross-border account prices, the Indian grid links chapter 25 measures at 10,425 million kWh [BPDB 2023], and the potential Iranian and Central Asian routes the region discusses are the imports-by-wire alternative to imports-by-ship, and their relative cost shifts with every exchange rate move. The stored record prices only the Indian electricity line, and the chapter treats the wider regional trade as the named scenario whose precondition is the grid and settlement infrastructure the power sector chapter owns.

The decade ahead

The forces the FY36 scenarios assume are four. First, the production decline's slope: the 711.4 bcf of 2024 [EIA 2025] against the 8,655 bcf of remaining 2P reserves [Petrobangla 2024] implies the domestic base halves inside the window absent new discoveries, and the exploration decision, offshore blocks and Bhola follow-up, is the supply answer whose timeline is not established. Second, the contract architecture: shifting the LNG portfolio from spot to term, at the cost of paying above spot in glut years, is the insurance purchase the 2022 crisis priced; the term share is not established and is the number to publish. Third, the coal lock-in: the 4.55 million tonnes of 2023 [BPDB 2023] and the plant fleet behind it commit the system to imported coal through the decade, colliding with the buyer decarbonisation demands chapter 17 carries and the renewables path chapter 27 prices. Fourth, the pricing convergence: administered fuel prices that track cost are the fiscal reform chapter 04 frames, and every step removes an arrears generator.

No stored fuel-security projection targeting 2036 exists in the sources this chapter draws on; any figure is not established, and the named authors whose work would resolve it are the Petrobangla gas sector master plan updates, the IEA country work, and the IMF programme's energy sector reviews. The decision points: the exploration bid round, the LNG term-contract expansion, the pricing formula convergence, and the industry allocation rule that replaces crisis rationing with priced priority.

Risks and upside

Risks. First, the next price cycle: an LNG and oil spike against the current spot exposure recreates the 2022-23 rationing with a weaker taka, and the revealing indicator is the spot-versus-term split, which could not be confirmed here. Second, the production cliff: gas output below 650 bcf without new fields puts import dependence past one third of consumption, chapter 10's mark [EIA 2025], and the revealing indicator is the annual production series. Third, the arrears cascade: an energy SOE default into the banking system would chain chapter 06's cleanup to chapter 41's losses, and the revealing indicator is the SOE debt-to-equity ratio at 2.69 [MoF Budget 2023].

Upside. First, the exploration option: the Bay of Bengal's geology, the boundary settlements chapter 57 records, and modern seismic work make the offshore the only supply-side answer that strengthens the taka rather than testing it. Second, the term-contract insurance: the Qatar and Oman agreements [Petrobangla 2018] expanded are the hedge whose premium is known and whose payoff the 2022 crisis demonstrated. Third, the efficiency dividend: the industrial and power efficiency margins are the domestic production the country already owns, and pricing gas honestly is what makes firms buy them.

The fertiliser-gas linkage is the supply chain where fuel security becomes food security, and the chapter 16 account carries its subsidy face. Urea plants consuming domestic gas, the subsidised fertiliser at 16,500 crore [MoF Budget 2025], and imported urea during shortage years together make the gas allocation rule an agricultural price decision: every cubic foot rerouted from a ceramics kiln to a fertiliser plant is a farm input the budget does not import. The allocation table that would show the split could not be confirmed here; the Energy and Mineral Resources Division's allocation orders would resolve it, and publishing it is the transparency that makes the rationing rule contestable rather than discretionary.

What to watch

1. Domestic gas production. Current value 711.4 bcf in 2024 [EIA 2025]. Threshold: below 650 bcf without a new field on stream marks import dependence past one third of consumption; a new field lifting the series marks the exploration regime.
2. Coal burn for power. Current value 4.55 million tonnes in 2023 [BPDB 2023]. Threshold: stabilisation below 5 million tonnes with renewables rising marks the managed transition; growth past 8 marks the coal lock-in deepening.
3. Petroleum imports. Current value 6.87 million tonnes in 2023 [BPC 2023]. Threshold: a falling trend at constant activity marks the efficiency turn; growth against flat output marks the fuel intensity trap.
4. Oil products in generation. Current values 274.75 million litres of diesel and 700.9 million litres of furnace oil in 2023 [BPDB 2023]. Threshold: a return below the 2020 levels marks the crisis fuel retiring; persistence marks the expensive bridge becoming permanent.
5. LNG term-contract share. Current status could not be confirmed from the sources this chapter draws on; the Petrobangla procurement disclosures would resolve it. Threshold: a published term share above 70 percent marks the security regime; continued spot dependence marks the exposure compounding.

Sources used

[EIA 2025] US Energy Information Administration via the energy/eia_international_bd parquet, series: eia_international_bd, dry natural gas production and consumption. [Petrobangla 2018] Petrobangla LNG terminal and purchase agreements, cited from chapter 10. [Petrobangla 2024] Petrobangla Annual Report 2024 / Hydrocarbon Unit gas reserves census via bdpolicy.db, series: petrobangla_national_gas_giip_bcf, petrobangla_cumulative_gas_production_bcf, petrobangla_national_gas_remaining_2p_bcf. [BPDB 2023] Bangladesh Power Development Board fuel series via bdpolicy.db, series: bpdb_fuel_gas_mmcft, bpdb_fuel_coal_m_ton, bpdb_fuel_diesel_ml, bpdb_fuel_furnace_oil_ml, bpdb_power_capacity_generation_panel. [BPC 2023] Bangladesh Petroleum Corporation import, sales and operating series via bdpolicy.db, series: bpc_petroleum_import_total_k_mt, bpc_petroleum_sales_mt, bpc_operating_profit_bdt_m. [MoF Budget 2025] Finance Division fertiliser subsidy series via bdpolicy.db, series: bdfacts_agri_subsidy_fertilizer_bdt_crore. [MoF Budget 2023] Finance Division consolidated SOE account via bdpolicy.db, series: soe_consolidated_total_liabilities_bdt_m, soe_consolidated_debt_equity_ratio. [BIS 2026] Bank for International Settlements USD/BDT series via bdpolicy.db, series: bis_usd_bdt_monthly_eop. [Not confirmed, chapter 26] LNG spot versus term split, Bhola and offshore prospectivity, rationing volumes, gas pricing schedule; resolving sources Petrobangla, Bapex, BERC and distribution company documents.

---

Verified line by line against primary sources: 27 claims checked, 5 corrected.

Previous: 25 Power sector

Next: 27 Renewables and the energy transition

All sixty chapters and the ten narrative chapters.