58 Labour rights, wages and due diligence

Chapter 58 of 60 in the Bangladesh 2036 research base. Contents of the series.

External anchors hold the labour rights regime, and the decade tests whether they survive their authors' exit

Bangladesh's labour rights regime is externally anchored: the standards are written abroad, the audits paid by buyers, and the domestic institutions that would make it self standing, unions, factory inspection and wage enforcement, are the weakest part of the system. The measured core is a gap between law and organisation: the ILO's textual compliance score for freedom of association and collective bargaining law, SDG indicator 8.8.2, rose from 4.97 in 2015 to 9.06 in 2024 [ILO 2026, SDG_0882_NOC_RT_A] while trade union density among employees fell from 26.3 percent in 2000 to 11.9 percent in 2018 [ILO 2026, ILR_TUMT_NOC_RT_A]. The claim this chapter defends: between 2026 and 2036 the external anchors are tested twice, first by the GSP+ application chapter 02 targets for its 2028 filing window and then by the European Union's corporate sustainability due diligence regime whose phase-in targets the end of the decade, and both tests ask whether compliance machinery that buyers built survives the buyers' transition to state run, law based enforcement. If the anchors transfer, the labour market acquires a compliance rent that supports the formalisation dividend chapter 08 prices; if not, the country competes as the cheap labour option, and the wage record, already down 1.9 percent in real terms between FY21 and FY23 on the wage rate index [BBS Wage 2024] [BBS CPI 2023], stays flat against the cost of living for another decade.

The record: wage floors set late and eroded fast, unions thin, safety institutions imported, and the child labour commitment missed

The wage record is the measurable core. The BBS Wage Rate Index moved from 180.83 in FY21 to 191.80 in FY22 and 205.30 in FY23, a rise of 13.5 percent over two years [BBS Wage 2024], while the consumer price index rose 6.15 percent in FY22 and 9.02 percent in FY23, a cumulative 15.7 percent [BBS CPI 2023], so the real wage rate index fell about 1.9 percent. The latest stored reading shows the same lag: the general index rose 7.80 percent in the year to March 2024 while rural consumer inflation ran at 9.68 percent and urban at 9.94 percent in that month [BBS Wage 2024] [BBS CPI 2024]. Chapter 08's only trade level series shows the same pattern: a Dhaka mason's day rate went from 496 taka in FY16 to 590 taka in FY22, a 19.0 percent nominal rise against 39.3 percent inflation, a 14.6 percent real fall across the construction boom [BBS Wage 2022] [BBS CPI 2022]. The economy wide real wage level beyond these series is not measured in the sources available to this chapter, the BBS labour force earnings modules being the resolving source, and any claim about it is not established.

Minimum wages are set sector by sector, the garment board the only settlement cycle with macro consequences, the point chapter 17 makes from the cost side. The ILO's recorded benchmark series carries two different bases across its own history, per its own metadata: 1,662 taka per month in 2006 and 2007 is scoped to garment employees, a sectoral reading, not a national floor, while 1,500 taka per month from 2008 through 2022 and 12,500 taka per month from 2023 are each classified by the ILO as a singular national minimum wage [ILO 2026, EAR_INEE_NOC_NB_A], the 2023 step matching the ready made garment board's December 2023 settlement. The garment gazette history runs 3,000 taka per month in 2010, 5,300 in 2013 after the Rana Plaza year, 8,000 in 2018 and 12,500 in December 2023, a 56.25 percent rise on the 2018 floor, not confirmed from the sources available to this chapter, the Minimum Wage Board gazette being the resolving source, the flag chapter 17 also carries; press accounts describe both as the lowest grade, seven grades collapsed to four in the December 2023 settlement, but the exact grade numbers and whether either figure is gross or basic pay stay unconfirmed, the same gap chapter 17 flags. Two structural facts matter more than the level: the five year settlement cycle puts the next garment round in the second half of the decade, so the floor is renegotiated once inside the window; and the boards cover formal trades, the agriculture, tea and domestic work floors sitting on irregular cycles below the garment floor.

The union record is the thin end of the system. Trade union density among employees was 26.3 percent in 2000, 14.1 percent in 2006, 11.3 percent in 2010 and 11.9 percent in 2018 on the stored ILO series [ILO 2026, ILR_TUMT_NOC_RT_A]; the series stops before registration politics became a bilateral trade issue, and a post 2018 reading is the first resolving fact available. The registration regime requires worker support thresholds, employer verification and a certification process in which rejections are common, counts not available to this chapter, the Department of Labour's registration statistics the resolving source, not confirmed. Export processing zones run under a separate 2019 law whose association structure falls short of the ILO's freedom of association standard, a design its supervisory machinery kept on file until the 2021 representation [ILO 2026]. The government's own account dates the roadmap that followed to the ILO Governing Body's 340th session of October to November 2019 and scopes it to labour law reform and labour inspection and enforcement, the areas Conventions 81, 87 and 98 govern [MoLE 2021]; its implementation status as of September 2026 is qualitative and could not be confirmed from the sources available to this chapter. The score that measures the law, not the practice, moved the other way from density, reflecting the 2013 and 2018 amendments to the Bangladesh Labour Act 2006 and the zones law that brought the statute book closer to the conventions.

Occupational safety is where Bangladesh built real institutions, after the Rana Plaza collapse of 24 April 2013 killed more than 1,100 workers [ILO 2026]. The Accord on Fire and Building Safety, the buyer funded inspection and remediation programme chapter 17 describes, and its successors under the RMG Sustainability Council and the International Accord made inspection a financed, contractually enforced activity in the export sector, with inspection counts and remediation closures not confirmed from the sources available to this chapter, the ILO roadmap documents being the resolving source. The state machinery that would replace them, the Department of Inspection for Factories and Establishments, last published a comparative record quantifying the catch up and the gap: approved manpower grew from 226 posts in FY06 to 1,156, approved inspector posts from 119 to 711 and working inspectors from 83 to 401, while registered factories and establishments grew from 20,934 to 104,660 and labour inspections from 7,000 to 41,062 in the FY23 reading to April 2023 [MoLE 2023], chapter arithmetic of roughly 261 registered establishments and 102 inspections per working inspector over the ten months of the reading. The stored survey record shows the residual risk inside the buyer architecture: 2,500 garment factory workers surveyed in November 2023 record a workplace injury prevalence of 44.52 percent, 17.52 percent with zero formal safety training, 22.0 percent under high occupational job risk, 28.12 percent with chronic health conditions and mean working hours of 48.03 per week, with the high risk share at 45.89 percent among machine maintenance workers and 32.49 percent in fabric dyeing [Zenodo 2023].

Child labour is the record where the country has measured itself honestly and missed its own commitment. The National Child Labour Survey 2022 counts 3.54 million working children aged 5 to 17, of whom 1.78 million fall in child labour and 1.07 million in hazardous child labour; against 2013, child labour rose 4.5 percent from 1.70 million while the hazardous count fell 16.6 percent from 1.28 million, the hazardous rate fell from 3.2 to 2.7 percent and the total rate held at 4.4 against 4.3 percent [BBS NCLS 2022]. The composition is the policy content: 99.0 percent of working children are informally employed, 52.2 percent of child labourers attend school while working, and 2.01 million children work unpaid as domestic workers with a further 0.08 million paid, overwhelmingly female [BBS NCLS 2022]. Boys dominate the counted total, 1.37 million against 0.40 million girls, partly because unpaid domestic work is undercounted by design. The geography inverts the income map: Chattogram division holds 643,547 child labourers, 36.2 percent of the national count, against 343,042 in Dhaka and 215,013 in Rangpur, with Barishal lowest at 71,979 [BBS NCLS 2022], a distribution chapter 14's regional record does not predict. Outside the counted workforce, a survey of 134 child workers in the tobacco and bidi chain records 56.72 percent in bidi manufacturing, a 39.1 percent school dropout rate, respiratory illness in 60.45 percent, skin conditions in 35.07 percent and a youngest working age of 6 years [Zenodo 2024]. The National Plan of Action to Eliminate Child Labour 2021-2025 carries the government's commitment to eliminate all forms of child labour within 2025 under SDG target 8.7 [MoLE 2021]; the only measurement inside the plan's cycle shows 1.78 million in 2022, so the commitment passed unmet.

Mechanism: enforcement is imported, informality sets the scale barrier, and wages move on political clocks

Enforcement is imported because the domestic enforcement market never priced it. A buyer funded inspection regime exists only where a buyer with contractual power sits, which is why the export garment sector has building safety institutions two decades ahead of the domestic garment, leather, construction and shipbreaking sectors, the last covered in chapter 57. The preference architecture converts that private enforcement into a public trade condition: GSP+ requires ratification and effective implementation of 27 conventions under the regulation currently in force, rising to 32 under the European Commission's proposed reformed scheme, the labour core of which are the ILO fundamentals on freedom of association, collective bargaining, forced labour, child labour and discrimination [EU GSP 2023]. The United States withdrew its own lever in 2013, suspending Bangladesh from the Generalized System of Preferences on worker rights grounds [USTR 2013], and its Department of Labor keeps Bangladesh on the list of goods produced by child labour or forced labour, placing named product chains inside the scope of United States customs forced labour enforcement [US DOL 2025]; the current product rows for Bangladesh could not be confirmed from the sources available to this chapter, the department's published list being the resolving source. The result is an asymmetric regime: the sector that earns the most foreign exchange is inspected the most, and the sectors that employ the most people are inspected the least, agriculture at 44.7 percent of employment in 2024 on chapter 08's numbers [ILO 2025], the tobacco and bidi chain the measured example [Zenodo 2024].

Informality is the scale barrier. Chapter 08's central number, 84.0 percent informal employment in 2024 [ILO 2025], is also this chapter's: labour law that applies to establishments reaches the 38.57 percent of workers in wage and salaried employment in 2024 and 38.96 percent on the 2025 reading [WB WDI 2026], and reaches them unevenly, while the own account and unpaid family majority sits outside registration, inspection and the wage boards. Union density of 11.9 percent [ILO 2026, ILR_TUMT_NOC_RT_A] and informality above 84 percent are two readings of the same structure, not two separate problems; the government's child labour plan says the same in its own words, child labour is more prevalent in the informal sector where little intervention exists and the Labour Law and its institutions have little or no role [MoLE 2021]. Any rights regime that works through enterprise representation and workplace inspection covers by construction less than a sixth of the labour force, and the household and farm economy where 75.4 percent of employed women work, again chapter 08's number [ILO 2025], is governed by nothing else.

The wage mechanism runs through the boards, not the market, and the boards run on political clocks. The garment settlements of 2013 and 2023 both followed unrest, and both reset the floor in steps the wage rate index then eroded between rounds; the taka slid from 83.9 to 110.0 per dollar between the end of 2018 and the end of 2023, a 31.1 percent depreciation [BIS 2026], which would have cut the dollar value of an unchanged 8,000 taka floor by about a quarter, to roughly 73 dollars, had the board not acted; the December 2023 settlement's 56.25 percent nominal rise to 12,500 taka instead pushed the floor's dollar value about 19 percent above its 2018 level, a real gain on the same unconfirmed gazette figures flagged above. The taka slid further after 2021, a separate 43 percent fall to 122.3 per dollar by end 2025 chapter 17 documents from the cost side [BIS 2026]. Indexation does not exist: no Bangladesh minimum wage is indexed to prices, so each round renegotiates the whole real wage history since the last round, and with FY23 inflation at 9.02 percent and March 2024 readings still above 9.6 percent [BBS CPI 2023] [BBS CPI 2024], the round that lands mid decade renegotiates against the largest accumulated price erosion in the record.

The post 2024 transition opened the domestic file. The reform commissions of the interim period included labour law, and the roadmap gave the external counterpart a benchmark list, but the implementation record as of September 2026 is a set of process claims without published counts, and every specific status reading here is not established. What is structural: the 2024 transition showed the labour market's political weight, the garment belt's disruption appearing in chapter 08's record as the fall of 1.91 million persons in employment between 2023 and 2024, from 70.98 to 69.07 million [ILO 2025]; any government prices labour law reform against that exposure.

The decade ahead: two external clocks, one tariff wall and one wage round

The GSP+ clock is chapter 02's, and its labour content is this chapter's: the application targets a filing in the 2028 working window the transition plan assumes [EU GSP 2023], and the conventions whose effective implementation the European Commission monitors include the association, child labour and forced labour standards this chapter has set out. Monitoring is biennial and permanent once granted, converting the roadmap's benchmark list into a standing treaty obligation; the decision is binary and its indicator is the filing date.

The due diligence clock is new in this decade, and it has already moved twice. Directive (EU) 2024/1760 on corporate sustainability due diligence entered into force on 25 July 2024. The 2025 stop the clock directive first targeted a July 2027 transposition and 2028 application; the finalised Omnibus I reform, Directive (EU) 2026/470 of 24 February 2026, then reset the transposition target to 26 July 2028 and set one targeted application date, 26 July 2029, for companies above 5,000 employees and 1.5 billion euro turnover, or 1.5 billion euro of EU turnover for a non-EU parent, replacing the original three size based waves [EU CSDDD 2024] [EU CSDDD 2025] [EU CSDDD 2026]. For the supplier base the mechanism matters before the legal deadlines do: garment is a flagship exposed sector, and the due diligence obligation, identify, prevent, mitigate and account for adverse human rights impacts across chains of activities, is already written into the buyer codes chapter 17's compliance margin records. Two effects follow. First, due diligence is the non tariff layer that survives graduation: it binds the buyer, not the preference, so it applies even in the most favoured nation tariff scenario chapter 02 sizes. Second, it concentrates orders: compliance documentation, remediation capacity and audit history are fixed costs, so the regime pushes orders toward the large compliant factories that already hold the 153 LEED certifications [BGMEA 2021] and away from the small firm tail that chapter 18's diversification agenda needs to survive. The chapter 15 scenarios assume this concentration in both the baseline and reform paths; the stall scenario assumes the tail exits through failure rather than upgrade.

The United States clock is a tariff clock with labour history attached. The preference lever has been unused since the 2013 suspension [USTR 2013], so the American labour posture operates through importer conduct and the customs forced labour enforcement the Department of Labor list feeds [US DOL 2025]. The tariff wall chapter 02 records, the Section 122 baseline of 10 percent from February 2026 with the announced intent to move to 15 percent and the statutory expiry of 23 July 2026 [US FR 2026], is priced independently of labour conditions, so the United States market charges the tariff without paying the compliance premium; that asymmetry is the decade's standing invitation to route the most price sensitive orders through the least inspected factories.

The domestic clock is the wage round. The five year garment cycle puts the next settlement inside the window, and this chapter's own arithmetic on the round is unforgiving: a floor of 12,500 taka per month [ILO 2026, EAR_INEE_NOC_NB_A] renegotiated after several years of 9 to 11 percent annual inflation [BBS CPI 2023] [BBS CPI 2024] needs a nominal rise on the order of 50 percent or more, depending on how many years elapse before the round, merely to hold its December 2023 real value. The chapter 15 reform scenario assumes the round delivers real convergence and the wage employment share reaches the mid forties; the baseline assumes the cost gap defence holds the rise near inflation; the stall scenario assumes a below inflation round followed by the unrest chapter 17's 2023 record already modelled. Child labour policy runs parallel: the missed 2025 commitment requires a successor plan, and the next National Child Labour Survey, whose timing could not be confirmed, is the only instrument that can measure whether the hazardous decline from 1.28 million in 2013 to 1.07 million in 2022 [BBS NCLS 2022] resumed or reversed.

Three risks print in monthly series and the upside converts compliance into coverage

Risks. First, preference failure without due diligence compensation: if the GSP+ application slips, the garment sector faces the most favoured nation tariff scenario on 52.4 percent of exports while remaining fully subject to the due diligence layer, a cost stack on margins chapter 17 measures at 19.73 USD per kilogram of 2024 unit value [BACI 2024] [BACI 2024]; the revealing indicator is the filing date and the amended GSP regulation text [EU GSP 2023]. Second, a real wage squeeze that ends in disruption: the wage rate index has lagged prices in every stored reading [BBS Wage 2024] [BBS CPI 2023], a below inflation settlement repeats the 2023 protest cycle in a sector that can stop the export economy, and the revealing indicators are the settlement number and the monthly export series in the quarter after it. Third, enforcement hollowing: the buyer programmes are transitional by design, and if the national inspection capacity does not replace them, the stored residuals, 17.52 percent of workers with no safety training and 44.52 percent injury prevalence [Zenodo 2023], are the floor the regime falls back to, not the ceiling, and the DIFE arithmetic of roughly 261 establishments per working inspector is the capacity the handover starts from [MoLE 2023]; the revealing indicator is the successor survey's training gap.

Upside. First, the compliance premium: an industry holding 153 LEED certified green factories [BGMEA 2021] and a documented safety architecture can price due diligence as a product feature, capturing the orders that pay for compliance, and the revealing indicator is the unit value series moving above the 19.73 USD per kilogram of 2024 [BACI 2024] while the certification count rises. Second, the formalisation dividend chapter 08 prices: every point of wage employment share, from 38.57 percent in 2024 and 38.96 percent in 2025 [WB WDI 2026], extends the reach of registration, contribution and inspection, so rights machinery and the pension arithmetic of chapters 04 and 15 compound; the revealing indicator is the wage share in the quarterly labour force series. Third, the child labour tailwind: the hazardous count fell 16.6 percent over a decade in which school attendance among child labourers reached 52.2 percent [BBS NCLS 2022], and the education expansion of chapter 09 plus the stipend architecture of chapter 34 are the levers that historically move children from work to school; the revealing indicator is the next survey round's hazardous rate against the 2.7 percent of 2022.

What to watch: five indicators whose thresholds mark the regime

1. ILO SDG 8.8.2 labour rights compliance score. Current value 9.06 in 2024, up from 4.97 in 2015, though the series is not monotonic, having peaked at 10 in 2023 before the 2024 reading [ILO 2026, SDG_0882_NOC_RT_A]. Threshold: a fall below 8 marks the legal rollback regime; a reading held above 9 while union density rises marks practice converging on the statute book.
2. Trade union density among employees. Current value 11.9 percent in 2018, the last stored year [ILO 2026, ILR_TUMT_NOC_RT_A]. Threshold: any published post 2018 reading below 10 confirms hollowing; a reading above 15 signals the representation regime changed; a new reading itself is the first resolving event.
3. National benchmark minimum wage. Current value 12,500 taka per month from 2023 [ILO 2026, EAR_INEE_NOC_NB_A]. Threshold: the next garment settlement above 20,000 taka per month signals real convergence; below 16,000 signals the cost gap defence won and the 2023 real value was consumed.
4. Hazardous child labour. Current value 1,068,212 children aged 5 to 17, 2.7 percent of the age group, in 2022 [BBS NCLS 2022]. Threshold: the next survey round below 2 percent signals structural decline resumed; a rate at or above the 3.2 percent of 2013 confirms the elimination commitment slipped a decade.
5. The safety training gap. Current value 17.52 percent of surveyed garment workers with zero formal safety training in November 2023 [Zenodo 2023]. Threshold: below 5 percent signals the national inspection regime has replaced the buyer programmes; a rise above the 2023 level signals the wind down outrunning the state.

Sources used

[MoLE 2021] Ministry of Labour and Employment, National Plan of Action to Eliminate Child Labour 2021-2025, via ocr_text/govtwin_min_labour: the SDG target 8.7 commitment to eliminate child labour within 2025 and the account of the time bound roadmap from the ILO Governing Body 340th session of October to November 2019. [MoLE 2023] Ministry of Labour and Employment, Department of Inspection for Factories and Establishments, DIFE in the Progress of Development, FY2005-06 and FY2022-23 comparative tables to April 2023, via ocr_text/govtwin_min_labour: the manpower, inspector, establishment and inspection counts tagged above. [ILO 2026] ILOSTAT harmonized indicators via the ILOSTAT API rplumber.ilo.org, series: ILR_TUMT_NOC_RT_A trade union density, EAR_INEE_NOC_NB_A monthly minimum wage in national currency, SDG_0882_NOC_RT_A SDG 8.8.2 compliance score; ILO supervisory documents for the Rana Plaza death toll, the zones law file and the 2021 representation. [ILO 2025] ILOSTAT modelled estimates via bdpolicy parquet as tagged in chapter 08, series: ilo_emp_nifl_sex_age_rt_a, ilo_emp_temp_sex_eco_nb_a. [WB WDI 2026] World Bank World Development Indicators via bdpolicy parquet, series: SL.EMP.WORK.ZS wage and salaried employment share, 2024 and 2025 readings. [BBS Wage 2024] BBS Wage Rate Index via the lake/labor/bbs_wage_rate_index_divisional_panel.parquet, national general index as tagged. [BBS Wage 2022] BBS daily wage rates for construction trades, via the lake/labor/bbs_daily_wage_rates_construction.parquet, as tagged in chapter 08. [BBS CPI 2023] BBS consumer price index annual panel via the lake/prices/bbs_cpi_historical_annual_panel.parquet, FY22 general inflation 6.15 percent, FY23 9.02 percent. [BBS CPI 2024] BBS consumer price index monthly sector panel via the lake/prices/bbs_cpi_monthly_sector_panel.parquet, March 2024 rural general inflation 9.68 percent, urban 9.94 percent. [BBS CPI 2022] BBS national general consumer price index via bdpolicy.db, series: bbs_cpi_national_general, as tagged in chapter 08. [BBS NCLS 2022] BBS National Child Labour Survey 2022, published March 2024, via bdpolicy.db series bbs_child_labour_ and bbs_working_children_, the lake/labor/bbs_child_labour_divisional.parquet and the ocr_text/bbs_surveys report text. [Zenodo 2023] Garment worker job risk survey, 2,500 workers, November 2023 fieldwork, DOI 10.5281/zenodo.17117330, via bdpolicy.db, series: rmg_workforce_injury_prevalence_pct, rmg_workforce_no_safety_training_pct, rmg_workforce_high_risk_pct, rmg_workforce_chronic_morbidity_pct, rmg_workforce_mean_weekly_hours_nos, rmg_machine_maintenance_high_risk_pct, rmg_dyeing_high_risk_pct. [Zenodo 2024] Tobacco and bidi child labour survey, 134 child workers, DOI 10.5281/zenodo.15031015, via bdpolicy.db, series: child_labour_tobacco_surveyed_workers_nos, child_labour_tobacco_bidi_factory_share_pct, child_labour_tobacco_school_dropout_rate_pct, child_labour_tobacco_respiratory_illness_pct, child_labour_tobacco_skin_disease_pct, child_labour_tobacco_youngest_working_age. [BACI 2024] CEPII bilateral trade at HS6 via the TradeWeave parquet, series: bd_bilateral and bd_hs6_trade for the EU27 share of 2024 goods exports, 52.4 percent (30,822,845 of 58,805,128 thousand USD), and the 19.73 USD per kg unit value, as computed in chapters 02, 17 and 44. An earlier 44.7 percent reading of this share was corrected in chapter 02 and inherited here. [EU GSP 2023] GSP regulation (EU) No 978/2012 as applied, EBA, GSP+ conditions and the 27 conventions under the regulation in force, 32 under the Commission's proposed reformed scheme, DG Trade. [EU CSDDD 2024] Directive (EU) 2024/1760 on corporate sustainability due diligence, in force 25 July 2024, EUR-Lex. [EU CSDDD 2025] Directive amending the CSDDD transposition and application calendar, stop the clock, April 2025, EUR-Lex, an interim calendar superseded by EU CSDDD 2026. [EU CSDDD 2026] Directive (EU) 2026/470 of 24 February 2026, the finalised Omnibus I reform of the CSDDD and CSRD, published in the Official Journal 26 February 2026, setting the CSDDD transposition target at 26 July 2028 and a single targeted application date of 26 July 2029 for companies with more than 5,000 employees and more than 1.5 billion euro turnover, EUR-Lex, fetched 2026-09-06. [USTR 2013] Office of the United States Trade Representative, June 2013 suspension of Bangladesh from the Generalized System of Preferences on worker rights grounds. [US DOL 2025] United States Department of Labor, List of Goods Produced by Child Labor or Forced Labor, Bangladesh named among the 82 listed countries. [US FR 2026] Federal Register tariff actions via the trade/bd_us_tariff_actions parquet, Section 122 baseline of 10 percent from 24 February 2026, statutory expiry 23 July 2026, announced 15 percent intent, record stamped 30 April 2026. [BIS 2026] Bank for International Settlements USD/BDT series via bdpolicy.db, series: bis_usd_bdt_monthly_eop, as tagged in chapter 17. [BGMEA 2021] Bangladesh Garment Manufacturers and Exporters Association publication via ocr_text/bgmea, 153 LEED certified green factories, as tagged in chapter 17.

---

Verified line by line against primary sources: 74 claims checked, 5 corrected.

Previous: 57 Blue economy and the maritime domain

Next: 59 Food inflation and the cost of living

All sixty chapters and the ten narrative chapters.