41 State-owned enterprises and subsidies

Chapter 41 of 60 in the Bangladesh 2036 research base. Contents of the series.

The enterprise state is a second budget that returned 0.30 percent on its assets in FY23 and was kept liquid by prices the state itself sets

Chapter 04 showed that the state owned enterprise layer is a hidden budget and left it at the consolidated balance sheet; this chapter opens the ledger. The claim it defends about the decade to the FY36 horizon the chapter 15 scenarios assume is this: the enterprise state will either shrink as a fiscal risk through pricing, contract and closure reform that the 2023 adjustments proved feasible, or it will compound into arrears that migrate onto the balance sheets of the state owned banks chapter 06 documents, and the difference between those paths is a fiscal flow measured in multiples of the development budget's social sectors. The scale is easy to state. The consolidated non-financial SOE portfolio held 474,418.88 crore BDT of assets at end FY23 and paid its owner a dividend of 700.81 crore BDT, a return of 0.15 percent [MoF Budget 2023]. In FY25 the government's subsidies and incentives outlay ran at 215,555 crore BDT, 3.75 times the combined FY26 development allocations for health, education and agriculture [MoF iBAS 2026] [Planning Commission ADP 2025]. The 2023 price adjustments, which returned BPC to profit and nearly tripled the gas price charged to the power system in one step, proved the state can close the gap when it chooses; the record assembled here says it chooses mostly when a crisis forces the choice, which is why the decade's SOE and subsidy story is a story about crisis timing, not about policy knowledge.

The record: a portfolio that grew its sales 32 percent and its liabilities 29 percent in the same year

The consolidated ledger for the non-financial SOE sector, the Finance Division monitoring cell evaluation that chapter 04 introduced, shows a portfolio expanding fast in nominal terms and not earning its cost of capital. Sales revenue rose from 146,433.34 crore BDT in FY22 to 193,699.56 crore BDT in FY23, an increase of 32.28 percent; operating profit rose faster, from 6,483.94 crore BDT to 11,119.84 crore BDT [MoF Budget 2023]. The bottom line moved the other way: net profit fell 20.74 percent, from 1,813.04 crore BDT to 1,437.04 crore BDT, and the return on the asset base fell from 0.46 percent to 0.30 percent [MoF Budget 2023]. The wedge between operating profit and net profit is financial expense: the sector paid 4,931.32 crore BDT of debt service in FY23 against 4,172.55 crore BDT in FY22 [MoF Budget 2023]. Liabilities grew faster than everything else, 345,887.12 crore BDT at end FY23 against 268,937.18 crore BDT a year earlier, 28.61 percent in one year, taking the debt to equity ratio from 2.12 to 2.69 [MoF Budget 2023]. How many enterprises the Finance Division's own consolidation covers could not be confirmed from the sources this chapter draws on, with the Finance Division SOE evaluation the resolving source; this research base's SOE data lake itself carries individual operating series for about 20 major enterprises (Petrobangla, BPC, BPDB, Bangladesh Railway, Biman, BSCPLC, Carew and roughly a dozen others), not the full non-financial SOE population, so the chapter reads the portfolio through its consolidated aggregates and its largest measured members.

Two liquidity readings sharpen the picture. The consolidated current ratio improved from 1.39 to 1.70 [MoF Budget 2023], which looks like comfort until it is set against the FY23 price adjustments documented below: the sector collected cash because administered prices rose, not because working capital discipline changed. And the sector's payments to the exchequer dwarf its dividends: 6,807.85 crore BDT of taxes paid in FY23 against the 700.81 crore BDT dividend, down from 7,543.96 crore BDT of taxes in FY22 [MoF Budget 2023]. The enterprise state is primarily a tax collecting and price administering machine, secondarily a profit making one, and the distinction matters for reform: a state that values SOEs as tax collectors has an incentive to keep administered margins high even when the fiscal accounts would prefer cost based pricing.

Beneath the aggregate the portfolio splits into two industries. The profit centres sell priced hydrocarbons and infrastructure services. Petrobangla contributed 19,723.04 crore BDT to the exchequer in FY24, up from 10,077.47 crore BDT in FY20, paid a dividend of 5,773.14 crore BDT in FY24 against 433.96 crore BDT in FY23, and paid 11,667.99 crore BDT of supplementary duty and VAT against corporate tax of only 364.10 crore BDT, down from 3,071.05 crore BDT the year before [Petrobangla 2024]. Its production system runs below capacity, 2,626 million standard cubic feet per day in FY24 against a 3,313 capacity, and its customer book is frozen at 4,326,504 accounts in FY24 against 4,318,389 in FY19, 8,115 new connections in five years [Petrobangla 2024]. BPC sold 7.35 million tonne of petroleum in FY23, up from 6.30 million tonne in FY21, and held a current ratio of 2.44 [BPC 2023]; its operating profit of 5,858.95 crore BDT in FY23 is chapter 26's finding and is not restated here as new. The airports earn, chapter 56 shows, and the profitable SOE list has small strong members: the submarine cable company earned an operating profit of 367.24 crore BDT in FY23 [BSCPLC 2023], and the shipping corporation's FY25 profit is chapter 57's finding.

The loss centres are the utilities and the legacy agro processors. BPDB carried a current ratio of 0.80 in FY23, served 3,980,433 retail consumers, and cut distribution system loss to 7.92 percent from 8.5 percent in FY21, while its operating loss of 6,131.02 crore BDT and 192.7 days of accounts payable are the chapter 25 findings this chapter treats as given [BPDB 2023]. Bangladesh Railway is a departmental enterprise running a permanent operating deficit: revenue of 1,400.11 crore BDT against expenses of 3,308.03 crore BDT in FY23, an operating ratio of 236.27 percent, a deficit of 1,907.92 crore BDT funded by budget transfer, improved from 2,185.20 crore BDT and 290.07 percent in FY22, yet the two years the information book shows as worst are these last two [Bangladesh Railway 2023]. The two secondary water utilities chapter 28 carries, Khulna WASA with a 15.5 percent system loss and Rajshahi WASA at 0.94 percent, bracket what utility management quality is worth. The agro processors are a managed decline: Carew and Company, the state sugar mill, produced 2,381.5 tonne of sugar in FY23 against 5,883.0 tonne in FY21, at a recovery rate of 5.08 percent [MoF SOE 2023]. The jute industry's state segment no longer produces at all: all Bangladesh Jute Mills Corporation mills have been shut down since July 2020 [BTTC 2024], which makes the corporation a carrier of liabilities and worker claims rather than an operating loss line, a distinction the consolidated accounts blur.

The subsidy side of the second budget is the larger number. The subsidies and incentives outlay recorded in the government accounting system reached 215,555 crore BDT in FY25; the first quarter of FY26 recorded 35,397 crore BDT, an annualised run rate of 141,588 crore BDT on chapter arithmetic, about a third below the FY25 pace [MoF iBAS 2026]. For scale, the published FY26 Annual Development Programme gives health 18,148.14 crore BDT, education 28,557.43 crore BDT and agriculture 10,795.97 crore BDT, 57,501.54 crore BDT together and 25.00 percent of the 230,000.00 crore programme on the printed shares, against which one year of subsidies and incentives is 3.75 times as large [Planning Commission ADP 2025, ADP at a glance table]. Within agriculture, the fertilizer allocation labelled 2025 in the Finance Division snapshot is 16,500 crore BDT, down from 24,500 crore BDT at the 2024 label, after rising from 8,800 crore BDT at the 2022 label; the snapshot labels are carried as published, and the 2025 label sits close to but not exactly on the 17,000 crore BDT fertilizer and other agricultural activities figure the Finance Adviser named for the FY26 budget speech, so treat the crore level as not established against the Finance Division's own fertilizer subsidy line [MoF Subsidy 2026]. The macro frame is chapter 04's, though this chapter re-derives it directly from the primary database rather than repeating chapter 04's figure: fossil fuel subsidies on the SDG 12.c.1 measure rose from 1.98 percent of GDP in 2021 to 6.41 percent in 2022 and fell back to 1.78 percent in 2023 [UNSD 2026, ER_FFS_CMPT_GDP]. The subsidy bill, like the SOE accounts, swings with world prices and the exchange rate and contracts only through administered price increases.

Mechanism: administered prices set the sector's income statement, and the losses then migrate down a chain that ends at the state banks

The first mechanism is price administration. The enterprise sector's profit line is not set by markets but by the gazette: the 2022 to 2023 adjustment rounds moved the gas price charged to the power system from 5.02 taka per cubic metre at the June 2022 step to 14.00 taka at the February 2023 step and 15.50 taka at the May 2023 step, an increase of 178.88 percent in eight months on chapter arithmetic, while the industrial tariff went from 11.98 to 30.00 taka per cubic metre, up 150.42 percent [Petrobangla 2024]. The same rounds returned BPC to its 5,858.95 crore BDT operating profit and cut the measured fossil fuel subsidy ratio from 6.41 to 1.78 percent of GDP [BPC 2023] [UNSD 2026, ER_FFS_CMPT_GDP]. This is why the consolidated operating profit rose 71.50 percent in FY23 while net profit fell: the pricing rounds filled the operating lines, the interest bill consumed the gain, and the portfolio's asset growth ran ahead of everything. Pricing reform in Bangladesh is not a technical event but a sequence of politically timed steps, and the FY23 sequence is the template the decade can repeat.

The second mechanism is the arrears cascade chapter 26 names and chapter 25 prices. When administered prices sit below cost, the deficit does not appear as a budget line; it appears as receivables down a chain of state counterparties, generators unpaid by BPDB, BPDB unpaid by distribution and by the government's own subsidy promise, Petrobangla and BPC unpaid by their largest customers, each carrying the last as a receivable and the next as a payable. The consolidated current ratio of 1.70 is an average that hides the chain's weak links: BPDB's 0.80 is the cash flow signature of a utility paying its generators late, and the 192.7 days of payables chapter 25 records is the cascade in one number [MoF Budget 2023] [BPDB 2023]. The cascade is self limiting only when prices track cost; at any other price it is a loan the enterprises extend to the budget at zero interest, on which the consolidated debt to equity ratio of 2.69 is the accumulating statement [MoF Budget 2023].

The third mechanism is the bank channel. SOE losses that do not fit on enterprise balance sheets end up as credit risk in the banking system: the state owned commercial banks and the specialised agricultural banks are the lenders of last resort to the enterprise state, and chapter 06 records where that stands, a 32.77 percent gross non performing ratio at state owned banks in June 2024, capital of 5.44 percent against 10.64 percent for the industry, and both specialised banks below the minimum capital requirement. What no series in the data lake measures is the direct stock of SOE and SOE supplier exposure inside those bank portfolios; the figure could not be confirmed, with the Bangladesh Bank supervisory returns the resolving source. No current IMF Financial System Stability Assessment covers Bangladesh; the last is IMF Country Report 10/38 from 2010. The mechanism does not need the exact stock to bind: any consolidation of enterprise arrears into guaranteed debt, the step the IMF programme's arrears conditions are designed to force into the open, lands on banks whose provision coverage chapter 06 already measures at 69.64 percent for the state owned group [BB AR 2024, annual report FY24, Table 5.06].

The fourth mechanism is exit, and the record shows three exit modes with different fiscal tails. Closure: the jute mills shut in July 2020 [BTTC 2024], removing an operating loss and creating a legacy liability of worker claims and idle assets whose settlement value could not be confirmed, with the corporation's accounts and the ministry's settlement reports the resolving sources. Managed decline: the sugar mill's output cut by 59.52 percent in two years while the subsidy that keeps imported sugar cheap is carried elsewhere in the budget [MoF SOE 2023]. Recapitalisation: Biman, which chapter 56 shows restored to profit, took a share issue against government equity of 683.77 crore BDT in FY25, taking paid up capital from 2,082.41 crore BDT to 2,766.18 crore BDT [Biman 2025]. The state is therefore not divesting; it is choosing, case by case, between shutting, shrinking and funding its enterprises. Outright privatisation, the fourth mode in the reform vocabulary, has no measured record in the data lake: no series carries enterprises divested or proceeds, and the figure could not be confirmed, with the Privatization Commission's successor bodies and the Finance Division the resolving sources.

The decade ahead: three reform tracks and one programme clock decide whether the second budget shrinks by FY36 in the scenarios chapter 15 assumes

The first track is pricing convergence, the continuation of the FY23 template. The IMF programme, in the reviews chapter 04 cites as its fiscal anchor, makes tariff adjustment and arrears reduction structural conditions, and each further step that aligns administered fuel, gas and power prices with cost converts an enterprise loss into a transfer line the budget can see and target [IMF Art IV 2025]. The Q1 FY26 subsidy run rate of 141,588 crore BDT annualised shows the outlay already easing from the FY25 peak [MoF iBAS 2026]. The decision point is the next world price spike: the reform scenario assumes administered prices track cost through it, the stall scenario assumes the 2022 pattern of absorption and late correction.

The second track is subsidy redesign. The budget can pay an enterprise to sell below cost, or it can pay households not to need the discount, and the second design is targetable, digitisable and measurable through the G2P rails chapter 34 documents and the payment infrastructure chapter 37 carries. Fertilizer is the test case: an allocation at the 24,500 crore BDT level, the 2024 label, spread as a blanket input subsidy, versus a targeted farmer transfer at a fraction of the leak, and the snapshot's 16,500 crore BDT figure at the 2025 label, not confirmed to the exact taka, close to the Tk 17,000 crore the FY26 budget speech named, is consistent with the first correction of a world price spike already under way [MoF Subsidy 2026]. The decision point for the fiscal decade is whether the FY27 budget the programme window targets is the first to publish a cash transfer countermeasure alongside each administered price rise; that publication, not the price gazette, is what makes the reform auditable.

The third track is corporate reform without divestiture. The record of the transition years is corporatisation: an equity injection into a profitable national carrier [Biman 2025], fleet renewal at the shipping corporation under concessional finance that chapter 57 carries, loss reduction at the power utilities that chapter 25 prices, and the post 2024 reform commissions' proposals on enterprise governance that chapter 47 assesses. The decade's decision is whether the finance division's monitoring cell evaluation becomes an ownership policy, dividends targetable, board appointments professionalised, loss making enterprises given closure paths with funded worker settlement, or stays an annual report. The assets exist: a portfolio of 474,418.88 crore BDT [MoF Budget 2023]. A dividend yield of 2 percent would pay 9,488.38 crore BDT, thirteen and a half times the FY23 dividend, and lifting the 0.30 percent net return to 3.83 percent would fund the published FY26 health allocation of 18,148.14 crore BDT outright, on chapter arithmetic [MoF Budget 2023] [Planning Commission ADP 2025].

The programme clock runs over all three tracks. The IMF programme reviews are the forcing sequence through the window the reviews target [IMF Art IV 2025], and LDC graduation, dated 24 November 2026 but under UN General Assembly review after the CDP found an extension of the preparatory period appropriate, would tighten the external financing terms chapter 02 documents whenever it takes effect, raising the cost of the borrowing that today stands in for subsidy reform. An enterprise sector that cannot earn its cost of capital, 345,887.12 crore BDT of liabilities at end FY23 [MoF Budget 2023], becomes more expensive to carry in exactly the years the budget's other claims grow.

Risks and upside: the arrears regime, the price spike and the bank channel against the dividend lever, the efficiency spread and the exit tool

Risks. First, the arrears regime returns: a world fuel and fertilizer spike against an unadjusted administered price structure would reopen the 2022 gap at a weaker exchange rate, and the revealing indicator is the quarterly subsidies and incentives outlay, which would re-approach and pass the FY25 peak of 215,555 crore BDT [MoF iBAS 2026]. Second, contingent liability crystallisation: consolidated SOE liabilities grew 28.61 percent in FY23, and a recognition event, guarantees called or arrears formalised as debt, would jump the public debt ratio without a taka of new borrowing; the revealing indicator is the consolidated liabilities series against nominal GDP growth [MoF Budget 2023]. Third, the bank channel binds: SOE and SOE supplier paper sitting in state owned and specialised banks converts a sectoral problem into a systemic one, and the revealing indicator is the state bank non performing ratio chapter 06 carries, already 32.77 percent at June 2024 [BB AR 2024, annual report FY24, Table 5.03(a)].

Upside. First, the dividend lever: Petrobangla alone paid 5,773.14 crore BDT in FY24, 8.24 times the entire sector's FY23 dividend [Petrobangla 2024] [MoF Budget 2023], so the yield the portfolio can pay is already demonstrated inside the portfolio. Second, the efficiency spread: system loss of 7.92 percent at BPDB and the chapter 28 spread between 15.5 and 0.94 percent across water utilities are recoverable cash that needs no new law, and the revealing indicator is the utility loss series. Third, the exit tool: the July 2020 jute mill closure [BTTC 2024] and the sugar mill's managed decline [MoF SOE 2023] show the state can end loss lines, and each completed exit with settled worker claims converts a recurring support into a one time cost.

What to watch: five indicators whose thresholds mark the regime

1. Subsidies and incentives outlay. Current value 215,555 crore BDT in FY25, with 35,397 crore BDT in the first quarter of FY26 [MoF iBAS 2026]. Threshold: a quarterly run that annualises above 250,000 crore BDT marks the return of the absorption regime; a full year below 150,000 crore BDT confirms the pricing reform is holding.
2. Consolidated SOE net profit. Current value 1,437.04 crore BDT in FY23, a 0.30 percent return on assets [MoF Budget 2023]. Threshold: a negative sector print, the first in the series, marks the loss regime; a print above 5,000 crore BDT says the FY23 pricing gains are compounding.
3. Consolidated SOE liabilities. Current value 345,887.12 crore BDT at end FY23, up 28.61 percent in the year [MoF Budget 2023]. Threshold: growth above nominal GDP growth for two consecutive years is the compounding regime chapter 04 warns of; two years below 10 percent growth marks the arrears break.
4. BPDB current ratio. Current value 0.80 in FY23 [BPDB 2023]. Threshold: a sustained reading below 0.60 is the cash crisis that forces a government guarantee; a move above 1.00 sustained over two fiscal years is the strongest single sign the arrears cascade is unwinding.
5. Fertilizer subsidy allocation. Current value 16,500 crore BDT, not confirmed to the exact taka, at the 2025 label, from 24,500 crore BDT at the 2024 label, year labels as published in the snapshot [MoF Subsidy 2026]. Threshold: a return toward 25,000 crore BDT with world prices flat marks the pricing rollback; a budget line pairing the allocation with a named farmer cash transfer marks the redesign this chapter tests.

Sources used

[MoF Budget 2023] Finance Division consolidated SOE evaluation, SPFMS Monitoring Cell via bdpolicy.db, series: soe_consolidated_total_assets_bdt_m, soe_consolidated_total_liabilities_bdt_m, soe_consolidated_total_equity_bdt_m, soe_consolidated_debt_equity_ratio, soe_consolidated_current_ratio, soe_consolidated_sales_revenue_bdt_m, soe_consolidated_operating_profit_bdt_m, soe_consolidated_net_profit_bdt_m, soe_consolidated_dividend_paid_bdt_m, soe_consolidated_dsl_paid_bdt_m, soe_consolidated_taxes_paid_bdt_m, FY2021-22 and FY2022-23. [MoF iBAS 2026] Finance Division iBAS++ integrated budget and accounting system via bdpolicy.db, series: mof_subsidies_incentives_outlay, FY2024-25 outlay at 30 June 2025 and first quarter FY2025-26 outlay at 30 September 2025. [MoF Subsidy 2026] Finance Division agricultural subsidy allocations via the bdfacts_sector snapshot in bdpolicy.db, series: bdfacts_agri_subsidy_fertilizer_bdt_crore, year labels as carried in the snapshot; a secondary compilation, cross checked against a Tk 17,000 crore FY2025-26 fertilizer subsidy figure the Finance Adviser named in the budget speech as reported by The Business Standard, close but not identical to the snapshot's 16,500 crore BDT 2025 label, so the crore figure could not be confirmed to the exact taka against the Finance Division's own published subsidy annex. [MoF SOE 2023] Finance Division state owned enterprise evaluation tables via bdpolicy.db, series: carew_sugar_production_mt, carew_sugar_recovery_rate_pct, Carew and Company, FY2020-21 to FY2022-23. [Petrobangla 2024] Petrobangla annual report 2024 accounts and operations, and BERC tariff orders, via bdpolicy.db, series: petrobangla_exchequer_contribution_crore_bdt, petrobangla_dividend_paid_crore_bdt, petrobangla_sd_vat_crore_bdt, petrobangla_corporate_tax_crore_bdt, petrobangla_daily_gas_production_mmscfd, petrobangla_daily_gas_capacity_mmscfd, petrobangla_total_gas_customers_nos, petrobangla_power_tariff_bdt_m3, petrobangla_industry_tariff_bdt_m3. [BPC 2023] Bangladesh Petroleum Corporation annual report and audited accounts via bdpolicy.db, series: bpc_petroleum_sales_mt, bpc_current_ratio, bpc_operating_profit_bdt_m, FY2022-23. [BPDB 2023] Bangladesh Power Development Board annual report and audited accounts via bdpolicy.db, series: bpdb_current_ratio, bpdb_consumers_total, bpdb_distribution_system_loss_pct, FY2022-23. [Bangladesh Railway 2023] Bangladesh Railway Information Book 2023 via bdpolicy.db, series: railway_operating_revenue_bdt_m, railway_operating_expenses_bdt_m, railway_operating_ratio_pct, railway_net_operating_deficit_bdt_m, Tables 74, FY2022-23. [Biman 2025] Biman Bangladesh Airlines Limited audited consolidated financial statements for the year ended 30 June 2025, statement of changes in equity, share issue against government equity. [BSCPLC 2023] Bangladesh Submarine Cable Company Limited via the MoF SOE evaluation tables in bdpolicy.db, series: bscplc_operating_profit_bdt_m, FY2022-23. [BTTC 2024] Bangladesh Trade and Tariff Commission annual report 2023-24 via ocr_text/btc, jute goods anti dumping investigation, record of the July 2020 shutdown of all BJMC jute mills. [Planning Commission ADP 2025] Annual Development Programme 2025-2026, Programming Division, Bangladesh Planning Commission, June 2025, ADP at a glance sector table: health 18,148.14 crore BDT (7.89 percent), education 28,557.43 crore BDT (12.42 percent) and agriculture 10,795.97 crore BDT (4.69 percent) on the 230,000.00 crore programme approved by the National Economic Council on 18 May 2025, https://adp.plancomm.gov.bd/book/2025-2026-ADP-BOOK.pdf, accessed 6 September 2026. The lake series adp_health_cr_bdt, adp_education_cr_bdt and adp_agriculture_cr_bdt via bdpolicy.db are the FY2023-24 programme book, internally consistent with its 255,267.44 crore project sum, and are not used. [UNSD 2026] UN Statistics Division, SDG Global Database, Indicator 12.c.1 fossil fuel subsidies as a proportion of GDP, series ER_FFS_CMPT_GDP, pulled directly from the UN SDG API (unstats.un.org/sdgapi, Bangladesh geoAreaCode 50) on 2026-09-06, calendar years, source line IEA fossil fuel subsidies database, OECD.Stat and IMF Energy Subsidies Template. This chapter's earlier draft cited the same tag via the bdfacts snapshot at 2.32, 7.39 and 2.07 percent for 2021 to 2023; those values do not match the API and are not used here. [BB AR 2024] Bangladesh Bank annual report FY2023-24 via ocr_text/bb/annual_report, Tables 5.03(a) and 5.06, state owned and specialised bank non performing loans and provision coverage. [IMF Art IV 2025] IMF Article IV and ECF/EFF review documents for Bangladesh, structural conditions on tariffs and arrears, resolving source for review level targets.

---

Verified line by line against primary sources: 47 claims checked, 3 corrected.

Previous: 40 Tax administration and revenue reform

Next: 42 External debt and sovereign risk

All sixty chapters and the ten narrative chapters.