In short
Bangladesh’s garment figures depend on who counts. The 2024 revision cut FY2022-23 and FY2023-24 exports by about US$9 billion and US$10 billion; the revised figures then match US and EU import records within about 4% and 10%. Counts of export factories run from under 1,800 to over 7,000, workers from about 2.6 to 5.1 million, women’s share from 35% to 65%.
Want a costing for your own style? Send us your enquiry
This chapter is part of a research guide to the Bangladesh garment trade, which has the literature review, the method and the full bibliography.
How big is Bangladesh’s garment industry? It depends on who is counting, when they counted and what they counted. Export figures for two recent fiscal years were cut by roughly a fifth in 2024. Published counts of factories, each made for a different purpose, run from under 1,800 to over 7,000, counts of workers from about 2.6 million to 5.1 million, and the share of women in the workforce from about 35% to about 65%. Every one of those numbers has a source, and most of them are honest. They measure different things.
This chapter asks two questions. What do the main published series say about exports, factories and workers, and how far apart are they? And why do they differ, so that a researcher can choose the right one for a given question and cite it properly? For exports I did not stop at reading other people’s comparisons. I downloaded what the importing countries themselves recorded, from UN Comtrade, Eurostat and the US Census Bureau (through Comtrade), and set it beside Bangladesh’s figures for the same months. For factories and workers I put each published count next to its date and its definition. The story of how the industry grew is in CMB’s cited history, and the latest monthly figures are in our exports guide and our industry statistics guide; this chapter is about the measuring itself.
A word on where I stand. I co-founded CMB Sourcing Ltd., which makes knitwear and caps in its own factory (the factory of our sister company, Collective Studio Ltd., where I am Head of Growth) and sources other garments as a buying house. A firm like mine has an interest in the industry looking large. Nothing here comes from my firm’s records: the chapter is a desk review of published sources and public data.
Who publishes Bangladesh’s export figures
Two kinds of record sit behind every figure for Bangladesh’s garment exports. The first is the shipment. Each consignment leaves with an export declaration lodged with customs, run by the National Board of Revenue (NBR), and the Export Promotion Bureau (EPB) publishes export figures based on those customs data (Ahmed 2024), which BGMEA’s tables show by product and by market (BGMEA n.d.-b). The IMF’s list of Bangladesh’s data sources names Bangladesh Bank and the EPB together for “Export data by goods”, published monthly within four weeks of the month’s end (IMF 2025, 113). The second record is the money. Bangladesh Bank compiles the balance of payments and sees the export proceeds that come back through the banks. When the two were compared closely in 2024, they did not match, as the next section shows.
The figures most people quote come from neither agency directly. The Bangladesh Garment Manufacturers and Exporters Association (BGMEA) republishes EPB data in long tables by fiscal year (July to June), by calendar year and by market, and each table ends “Data Source Export Promotion Bureau, Compiled by BGMEA” (BGMEA n.d.-b). Our guide to the two associations explains who belongs to which. The World Trade Organization’s country figures for Bangladesh are the same national series again, as I show below.
A shipment record and a receipt record can both be correct and still differ. Proceeds arrive weeks or months after the ship sails, depending on the payment terms (see our guide to payment terms). Buyers take discounts and claims, orders are cut after shipment, and unsold stock is sold off at a loss; our guide to order problems covers who bears those losses. Each of these makes receipts smaller than shipments without anyone making a mistake. Sadiq Ahmed lists two of them, the gap between the letter-of-credit value and the proceeds and losses from stock-lot sales, discounts and commissions, among the six causes of the 2024 gap (Ahmed 2024).
What “ready-made garments” covers
BGMEA’s tables divide ready-made garments (RMG) into “woven” and “knit” (BGMEA n.d.-b). I could not find an EPB document that lists which customs headings make up each. The customs classification most researchers use is the Harmonized System: chapter 61 covers articles of apparel and clothing accessories, knitted or crocheted, and chapter 62 the same articles not knitted; chapter 61 excludes, for example, worn clothing, which goes to heading 63.09 (WCO 2022a, chap. 61, note 2). Bangladesh’s own reports to UN Comtrade stop early: the public interface returned its exports of chapters 61 and 62 for 2015 to 2018 and nothing for 2014 or for 2019 to 2024 (UN Comtrade 2026). For three of those four years they match BGMEA’s calendar-year knit and woven totals closely. In 2016 Bangladesh reported US$13,735.5 million under chapter 61 and US$14,927.1 million under chapter 62, against BGMEA’s knit US$13,736.95 million and woven US$14,931.33 million (UN Comtrade 2026; BGMEA n.d.-b). In 2017 both chapters are about 9% to 10% above BGMEA’s figures, and I do not know why. The WTO uses a third definition, clothing as division 84 of the Standard International Trade Classification (WTO 2023, 46). The three overlap heavily, but a researcher who mixes them should say so.
The 2024 revision and the breaks it left
In July 2024 Bangladesh Bank published revised figures that put export earnings for July 2023 to May 2024 at US$40.7 billion, against US$51.5 billion reported by the EPB for the same months (Ahmed 2024). The Daily Star, reporting Bangladesh Bank’s balance-of-payments release of 3 July 2024 for July to April, put the gap at nearly US$14 billion (Daily Star 2024); I could not open Bangladesh Bank’s release to reconcile the two figures. The IMF later described the cause as “discrepancies between export shipment data from the Export Promotion Bureau and the actual export receipts”, and put the revision at about 2% of GDP in each of FY2022-23 and FY2023-24 (IMF 2025, 8). EPB officials, as reported in September 2024, pointed to the difference in method: the central bank counts proceeds received, the EPB counts shipments. The EPB stopped publishing export data from July 2024 (Daily Star 2024). BGMEA’s page, read on 22 November 2026, again carries monthly figures to October 2026 under the same EPB source line (BGMEA n.d.-b).
Ahmed, vice-chairman of the Policy Research Institute of Bangladesh, sets out what the investigation found on the customs side: duplicated serial entries in customs data, fabric valued wrongly, samples counted as exports with a value, sales by firms in export processing zones counted twice, and the two receipt-side gaps mentioned above. All of these concerned ready-made garments (Ahmed 2024). His table also shows that the gap did not start in FY2022-23.
| Fiscal year | EPB, as first published | Bangladesh Bank, from export receipts | BGMEA’s series now | EPB minus Bangladesh Bank |
|---|---|---|---|---|
| FY2021-22 | 42.6 | 38.4 | 42.61 (not marked ®) | 4.2 |
| FY2022-23 | 47.0 | 38.0 | 38.14 (marked ®) | 9.0 |
| FY2023-24 | 46.1 | 36.4 | 36.15 (marked ®) | 9.7 |
BGMEA revised FY2022-23 and FY2023-24 to figures close to the central bank’s and left FY2021-22 as it was (BGMEA n.d.-b). On Ahmed’s numbers the FY2021-22 figure was US$4.2 billion higher than receipts. The IMF’s own technical assistance pointed the same way: in the government’s letter to the Fund of June 2025, Bangladesh Bank committed to take “full responsibility for compiling trade statistics”, citing the revisions for FY23 and FY24 and “the findings of the IMF’s Technical Assistance Mission on the need to revisit the published export data between FY19 and FY22” (IMF 2025, 82–83). The same report gives two deadlines for that commitment: end-December 2025 in the staff’s text and in both tables of structural benchmarks, and end-September 2025 in the paragraph of the government’s letter (IMF 2025, 24, 49, 83, 89). In January 2026 the IMF was still writing of “meeting the Structural Benchmark” as a step to come (IMF 2026, 54). I found no revised series for FY2018-19 to FY2021-22.
Where the series breaks
Three breaks follow. First, the fiscal-year series is on the revised basis from FY2022-23 and on the old basis before it, so the fall from US$42.61 billion in FY2021-22 to US$38.14 billion in FY2022-23 mixes a real change with a change of method. Second, BGMEA’s calendar-year table marks 2023 and 2024 as revised but not 2022 (BGMEA n.d.-b), although July to December 2022 belongs to FY2022-23, which is marked. I could not tell from the page whether those six months were revised. Third, the WTO’s tables carry the break into international comparisons. Its 2026 tables give Bangladesh’s clothing exports a change of −21.5% in 2023 (WTO 2026), which is exactly BGMEA’s revised 2023 (US$35,886.61 million) over its unmarked 2022 (US$45,709.15 million); the WTO’s changes for 2024 (+7.2%) and 2025 (+0.9%) match BGMEA’s calendar figures in the same way (my calculation). A reader of the WTO tables sees a fall of a fifth in one year, and part of it is bookkeeping.
What importers recorded: the mirror test
Every international shipment is recorded twice, once by the country that sends it and once by the country that receives it. Comparing the two is an old method in trade economics. Fisman and Wei used the gap between Hong Kong’s recorded exports to China and China’s recorded imports from Hong Kong, product by product, to measure tariff evasion (Fisman and Wei 2004), and Gupta, Roy and Ahmad, studying Bangladesh’s exports across importing countries, found evidence of under-invoicing that rose after export subsidies were introduced (Gupta, Roy, and Ahmad 2012). For garments, the importers’ side has one practical advantage: it is the value customs assesses. In the United States, for example, the customs value is the value appraised under the Tariff Act of 1930 (US Census Bureau 2022). I used it here for a simpler purpose, as an outside check on the size of Bangladesh’s exports before and after the revision.
The world total, and why it needs care
Adding up every country’s reported imports of chapters 61 and 62 from Bangladesh gives a figure 27% to 47% above Bangladesh’s own for 2021 to 2025 (UN Comtrade 2026; BGMEA n.d.-b). Most of that gap is not real. In UN Comtrade, each EU member’s figures come from its national statistics office, and Comtrade’s own guidance says that “intra-EU trade needs to be deducted from trade reported by the EU member states” (UN Comtrade n.d.). Eurostat explains why: in the EU’s harmonised data, the partner for goods arriving from another member state is the member state that sent them, but “for national statistics it may be the country of origin” (Eurostat n.d.). A carton of Bangladeshi T-shirts cleared in Rotterdam, held in a Dutch distribution centre and later trucked to Vienna can therefore appear as an import from Bangladesh twice, in the Dutch figures and again in the Austrian ones. When I replace the 27 national EU figures with Comtrade’s figure for the EU as a whole, which counts only goods entering the EU from outside, the gap falls to between 6% and 20%.
| Year | Bangladesh’s exports (EPB, calendar year) | Partners’ imports, every reporter added up | Partners’ imports, EU counted once | Partners reporting | EU-once figure above Bangladesh’s |
|---|---|---|---|---|---|
| 2021 | 35,811.87 | 45,982.6 | 37,858.9 | 156 | 5.7% |
| 2022 | 45,709.15 | 58,234.4 | 49,185.7 | 154 | 7.6% |
| 2023 | 35,886.61 | 49,979.3 | 41,484.9 | 155 | 15.6% |
| 2024 | 38,482.08 | 51,932.0 | 42,395.1 | 139 | 10.2% |
| 2025 | 38,824.71 | 56,942.6 | 46,574.0 | 123 | 20.0% |
The EU effect is large enough to deserve its own table. For 2023 I set each member state’s figure under the EU’s harmonised method (Eurostat) beside the same country’s national figure as published in Comtrade, and beside what Bangladesh recorded as exports to that country in FY2023-24.
| Importing country | Eurostat, EU method (€ million) | Eurostat, EU method (US$ million) | National figure in Comtrade (US$ million) | Bangladesh’s record of exports to it, FY2023-24 |
|---|---|---|---|---|
| Germany | 4,533.3 | 4,901.7 | 8,350.8 | 4,522.58 |
| Spain | 2,977.7 | 3,219.7 | 3,782.0 | 3,374.68 |
| France | 1,988.9 | 2,150.6 | 3,820.1 | 2,027.19 |
| Netherlands | 1,871.0 | 2,023.1 | 1,219.9 | 1,723.15 |
| Italy | 1,519.7 | 1,643.2 | 1,643.2 | 1,462.60 |
| Poland | 1,352.4 | 1,462.3 | 3,048.1 | 1,548.56 |
| Belgium | 627.3 | 678.3 | 352.1 | 518.79 |
| Czechia | 171.3 | 185.2 | 934.6 | 266.29 |
| Austria | 113.2 | 122.4 | 980.3 | 42.97 |
| EU, 27 members | 17,449.8 | 18,867.9 | 27,295.9 (sum of national figures) | 18,067.9 |
Austria’s national figure is eight times its EU-method figure and Czechia’s five times; the Netherlands’ and Belgium’s national figures are lower than their EU-method ones, which fits Eurostat’s note that national figures leave out “quasi-transit”, goods cleared at a port and sent straight on (Eurostat n.d.). Italy’s two figures are identical. In my reading the national figures attribute goods to Bangladesh by origin wherever they physically arrive from, while the EU method attributes them to the member state that first imported them. Bangladesh’s own record presumably follows the destination known at shipment (I could not open the EPB’s definitions), and in most rows it sits much closer to the EU method. For research the rule is simple: for the EU, use Eurostat’s EU-level data, and never add up member states’ national figures from Comtrade.
Matched months: six markets
A fair test needs the same months on both sides. For the six largest markets I added up the importers’ monthly figures for each July-to-June year and set them beside BGMEA’s fiscal-year figures for that market. The United States records two values: the customs value, which excludes freight, insurance and duty, and a CIF value, the landed value at the first port of arrival (US Census Bureau 2022). The customs value is the closer match to Bangladesh’s free-on-board shipment value. The EU records CIF only (Eurostat n.d.).
| Market and fiscal year | Bangladesh’s record (BGMEA) | Importer, value without freight | Importer, CIF value | Importer above Bangladesh, without freight | Importer above Bangladesh, CIF |
|---|---|---|---|---|---|
| United States, FY2023-24 | 6,631.09 | 6,595.4 | 6,903.7 | −0.5% | 4.1% |
| United States, FY2024-25 | 7,545.81 | 7,860.4 | 8,233.0 | 4.2% | 9.1% |
| United States, FY2025-26 | 7,744.61 | 7,655.2 | 7,972.4 | −1.2% | 2.9% |
| EU, FY2023-24 | 18,067.9 | Not recorded | 18,419.6 | Not applicable | 1.9% |
| EU, FY2024-25 | 19,712.0 | Not recorded | 21,630.6 | Not applicable | 9.7% |
| EU, FY2025-26 | 19,060.0 | Not recorded | 20,661.4 | Not applicable | 8.4% |
| United Kingdom, FY2023-24 | 4,194.59 | Not recorded | 3,790.4 | Not applicable | −9.6% |
| United Kingdom, FY2024-25 | 4,349.04 | Not recorded | 4,271.5 | Not applicable | −1.8% |
| United Kingdom, FY2025-26 | 4,388.55 | Not recorded | 4,776.4 | Not applicable | 8.8% |
| Canada, FY2023-24 | 1,161.98 | 1,358.5 | Not recorded | 16.9% | Not applicable |
| Canada, FY2024-25 | 1,302.24 | 1,590.5 | Not recorded | 22.1% | Not applicable |
| Canada, FY2025-26 | 1,343.90 | 1,694.7 | Not recorded | 26.1% | Not applicable |
| Japan, FY2023-24 | 1,086.25 | Not recorded | 1,230.9 | Not applicable | 13.3% |
| Japan, FY2024-25 | 1,185.47 | Not recorded | 1,388.3 | Not applicable | 17.1% |
| Japan, FY2025-26 | 1,134.72 | Not recorded | 1,301.5 | Not applicable | 14.7% |
| Australia, FY2023-24 | 800.80 | 834.5 | 854.5 | 4.2% | 6.7% |
| Australia, FY2024-25 | 813.62 | 908.3 | 932.6 | 11.6% | 14.6% |
| Australia, FY2025-26 | 739.23 | 828.1 | 850.2 | 12.0% | 15.0% |
For the two markets that took about seven-tenths of Bangladesh’s garments in these years (BGMEA n.d.-b), the revised figures pass the test. US customs recorded within about 4% of what Bangladesh recorded in each of the three years. The EU’s CIF figures run 2% to 10% above Bangladesh’s; freight and insurance explain part of that (they added 3% to 6% on the US route, as the next section shows), and the weeks at sea move shipments between years, which may explain why single years drift apart and back. FY2024-25 is the year to look at more closely in both markets. The other four markets are less tidy. The United Kingdom swings from 10% below Bangladesh’s record to 9% above it in three years, which looks like timing or attribution rather than a steady bias. Australia runs 4% to 12% above on a value without freight. Japan runs 13% to 17% above on CIF, and Canada 17% to 26% above on a value that excludes freight, so freight cannot explain Canada’s gap. A matched test for FY2021-22, the year Ahmed argues was also overstated, needs BGMEA’s or the EPB’s market breakdown for that year, which BGMEA’s current page does not show; I have not done it.
The larger gap in the world totals therefore lies mostly outside the EU and the United States: in Canada, Japan and a number of smaller markets. Several smaller importers record far more from Bangladesh than Bangladesh records sending them: in 2023 Mexico recorded US$588 million, the United Arab Emirates US$645 million, China US$400 million and Saudi Arabia US$350 million (UN Comtrade 2026), against Bangladesh’s FY2023-24 records of US$283 million, US$260 million, US$217 million and US$154 million for the same countries (BGMEA n.d.-b). The years differ, so these are only orders of magnitude. Goods that are shipped to one country and sold on to another, for example from a North American distribution centre into Canada or Mexico, or attributed to Bangladesh by origin after passing through a hub such as the United Arab Emirates, would produce this pattern, but I have not been able to test it.
Why the two sides of a trade disagree
The reasons are well known to trade statisticians. Here is how each applies to Bangladesh’s garments, with the size of the effect where I could measure it.
Valuation: free on board against cost, insurance and freight
Exports are usually valued free on board (FOB), at the port of loading, and imports CIF, including cost, insurance and freight to the importer’s border; that is the EU’s own rule (Eurostat n.d.), and the United States publishes a CIF series beside its customs value (US Census Bureau 2022). Canada and Australia value imports FOB (WTO 2026). For US imports from Bangladesh, the CIF value was 3.3% above the customs value in 2019, 5.5% in 2022, 4.6% in 2023, 4.9% in 2024 and 4.3% in 2025 (UN Comtrade 2026). That is a direct measure of the freight and insurance margin on this route. It was highest in 2022; I did not test how closely it follows freight rates.
Timing: the voyage and the year
Bangladesh records a shipment when it leaves; the importer records it when it arrives, several weeks later by sea. Bangladesh’s official year runs July to June, while most importers and the WTO publish calendar years. Comparing a fiscal year with a calendar year compares two different sets of months. Even with matched months, December shipments land in January.
Attribution: origin, destination and re-exports
Importers usually attribute goods to the country where they were made; the United States, for example, credits imports to the country “where the merchandise was grown, mined, or manufactured” (US Census Bureau 2022). Bangladesh, as far as I can tell without the EPB’s definitions, records the destination known at shipment. Goods sent to a distribution centre in one country and sold on to another are recorded by Bangladesh once and by importers twice, and the EU example above shows how large that effect can be.
Definitions: what counts as a garment
Harmonized System chapters 61 and 62, SITC division 84 and Bangladesh’s “RMG” are close but not the same, and I could not find a published statement of which headings the EPB includes in RMG. Whether it counts any made-up textile goods from other chapters, such as home textiles, is not stated anywhere I looked. When a national agency publishes “apparel” in groupings of its own, its figures will match none of the three.
Currency
Bangladesh and the WTO publish in US dollars, Eurostat in euro. Converting the EU’s 2023 imports at the year’s average rate gives US$18,867.9 million; converting month by month gives a slightly different figure for any twelve months. Comtrade converts with its own rates. The difference is small for a year but matters for monthly work, and a paper should say which rate it used.
Errors and misinvoicing
Some differences are mistakes. The 2024 investigation found duplicated customs entries and double counting on the Bangladesh side (Ahmed 2024). Others may be deliberate: under-invoicing moves money out of a country, and over-invoicing can capture export incentives. Gupta, Roy and Ahmad found evidence of under-invoicing in Bangladesh’s exports (Gupta, Roy, and Ahmad 2012), but I found no peer-reviewed study that applies mirror data to Bangladesh’s garments after 2018. The gaps in the tables above cannot, on their own, show misinvoicing, because valuation, timing and attribution explain most of what I can measure.
Who reports
Bangladesh’s own reports of garment exports in UN Comtrade stop in 2018 (UN Comtrade 2026), so recent work with Comtrade has to use partners’ records. Among the partners, the most recent years are always incomplete: 155 countries reported imports from Bangladesh for 2023, 139 for 2024 and 123 for 2025 when I downloaded the data. Missing reporters make world totals look lower than they will be, so the last two rows of the calendar-year table are likely to rise.
Where Bangladesh ranks among clothing exporters
Bangladesh is often called the world’s second-largest clothing exporter; BGMEA’s newsletter calls it “the world’s second-largest garment supplier” (BGMEA 2023, 4). Whether that is true depends on the year, the edition of the WTO’s tables and how the European Union is treated. The WTO lists the EU as one exporter twice over: once with trade between member states included, and once for extra-EU exports only.
| Data year | China | EU, all | EU, extra-EU | Bangladesh | Viet Nam | Source |
|---|---|---|---|---|---|---|
| 2020 | 142 | 125 | 38 | 28 | 29 | WTO 2021, 78 |
| 2021 | 176 | 151 | 43 | 34 | 31 | WTO 2022, 78 |
| 2022 | 182 | 156 | 45 | 45 | 35 | WTO 2023, 80 |
| 2025 | 157.17 | 177.04 | 44.20 | 38.82 | 37.51 | WTO 2026 |
Among single countries Bangladesh was second to China in 2021, 2022 and 2025, and third behind Viet Nam in 2020. Counting the EU as one exporter changes the order. With trade between member states included, the EU was ahead of China in 2025, and Bangladesh is third; working back from the same table’s annual changes, the EU had already moved ahead of China in 2024 (my calculation). Counting extra-EU exports only, the EU’s US$44.2 billion in 2025 is still ahead of Bangladesh’s US$38.82 billion, so Bangladesh is third on that measure too; in 2022 the two were level at US$45 billion (WTO 2023; WTO 2026). Bangladesh’s 2025 value in the WTO table, US$38.82471 billion, is BGMEA’s calendar-year figure to the last digit (WTO 2026; BGMEA n.d.-b), so the WTO ranking inherits the revision. Its share of world clothing exports was 7.9% in 2022 on the unrevised figure and 6.95% in 2025 on the revised series (WTO 2023, 80; WTO 2026). The lead over Viet Nam in 2025 was US$1.3 billion, smaller than the revision to a single fiscal year. Any statement of rank should give the data year, the WTO edition, and whether the EU is counted as one.
Counting factories
No official count of Bangladesh’s garment factories could be opened for this chapter. The Department of Inspection for Factories and Establishments (DIFE) licenses factories, and the Bangladesh Bureau of Statistics (BBS) runs manufacturing surveys, but neither website could be reached from where I worked on 22 November 2026. What can be read are counts made for other purposes: membership lists, safety programmes and an independent census.
| Count | Date | What is counted | Source |
|---|---|---|---|
| Around 4,000; around 4,500 | Undated, read 2026 | BGMEA “registered garment factories” and “member factories”, both on the same page | BGMEA n.d.-a |
| About 2,540 | Undated, read 2026 | BKMEA knitwear member factories | BKMEA n.d.-a |
| 4,000 to 4,500 | Before Rana Plaza (April 2013) | Earlier estimates of the size of the sector, as summarised by the Stern Center | Labowitz and Baumann-Pauly 2015, 4, 16 |
| 3,508 | 2015 | Export-oriented factories on the list for safety inspection after Rana Plaza | ILO 2015a |
| 3,660 | October 2015 | Factories inspected: 1,475 by the government with ILO support and 2,185 by the Accord and the Alliance (my sum) | ILO 2015b |
| 7,179 | 2015 | The Stern Center’s “official list”, merged from five public lists, direct and indirect exporters | Labowitz and Baumann-Pauly 2015, 4, 16 |
| About 3,200 | 2015 | Direct exporters on the associations’ “UD lists” of factories that applied for a utilisation declaration in the year (BGMEA’s list about 2,100) | Labowitz and Baumann-Pauly 2015, 20, 35 n. 24 |
| At least 2,000 fewer than 7,165 | 2016 | Anner and Bair’s estimate of closed, duplicate and domestic-market factories in the Stern database | Anner and Bair 2016, 1 |
| 3,223 | December 2020 | Export factories found by Mapped in Bangladesh in four districts | Mirdha 2020 |
| 3,374, of which 672 non-members | 2021 | Export-oriented factories on the Mapped in Bangladesh map, collected “through census method” | Choudhury et al. 2021 |
| 3,722, of which 984 non-members | 2024 | Export-oriented factories on the map, which also tracks new and non-operational ones | BRAC University CED 2024, 2 |
| 3,315: 2,053 BGMEA, 739 BKMEA, 807 neither (284 in both) | June 2025 | Export-oriented factories in a nationwide census | BRAC University CED 2025, 6–7 |
| 1,773, of which 1,680 active | June 2026 | Factories covered by the Bangladesh Safety Agreement and inspected by the RMG Sustainability Council | RSC 2026, 1 |
| About 1,500 | May 2026 | Factories under National Action Plan monitoring supervised by DIFE, as a BGMEA vice-president put it, reported by BGMEA | BGMEA 2026 |
Five things separate these numbers. The first is the unit. An association counts members, and a member can be a factory that has closed or stopped exporting; BGMEA’s page gives about 4,000 and about 4,500 on the same page, neither dated (BGMEA n.d.-a). The second is overlap: 284 of the factories in the 2025 census belong to both associations (BRAC University CED 2025, 6), so adding BGMEA’s and BKMEA’s totals counts them twice. The third is activity. The associations’ “UD lists” are a better guide to who is exporting than membership is, because a factory appears on one only if it has applied for and received a utilisation declaration (UD) during the year, the permit the association issues for each export order so that its materials can be imported duty-free; in 2015 BGMEA reported its list stable at about 2,100 (Labowitz and Baumann-Pauly 2015, 20). Chapter 3 explains how the bonded system that uses the UD works. The fourth is scope: direct exporters only, or the subcontractors who sew for them too. The Stern Center counted more than 7,000 factories including indirect suppliers, and Anner and Bair answered that the database was inflated by at least two thousand closed, duplicate and domestic-market factories (Labowitz and Baumann-Pauly 2015, 4; Anner and Bair 2016, 1); chapter 1 sets out that dispute. The fifth is purpose. Safety programmes count the factories they cover, which are a subset of the industry (RSC 2026, 1).
Mapped in Bangladesh, run by BRAC University’s Centre for Entrepreneurship Development, is the only source here that publishes a census method and dated results. Its authors began it because, they wrote, there was “no actual count of either the number of firms or workers” (Choudhury et al. 2021). Its counts still move in ways its publications do not explain: 3,722 export-oriented factories in the impact report of November 2024, 3,315 in the census as of June 2025 (BRAC University CED 2024, 2; BRAC University CED 2025, 6). I found no statement of the reason. Our guide to garment factories in Bangladesh covers where they are and how they are organised.
Counting workers
Worker counts divide into two families. Establishment counts ask factories how many people they employ. Household surveys, such as the BBS Labour Force Survey, ask people where they work. The two answer different questions, and both are useful.
| Workers | Year | What is counted | Source |
|---|---|---|---|
| “Around four million” | 2023 | The industry, in BGMEA’s newsletter; no source or method given | BGMEA 2023, 4 |
| 5,017,652 | 2024 | BGMEA’s member database (3,317,397) plus BKMEA’s (about 1.7 million), as the state minister for labour told parliament | UNB 2024 |
| 1.9 million direct jobs | FY2022-23 | BKMEA’s figure for the knitwear sector; no method given | BKMEA n.d.-b |
| 5.1 million | 2015 | Employment on the Stern Center’s official factory list, disputed by Anner and Bair, who found a data-entry error adding more than 335,000 | Labowitz and Baumann-Pauly 2015, 27; Anner and Bair 2016, 2 |
| 2,562,383 | December 2020 | Workers in 3,223 export factories in four districts, Mapped in Bangladesh | Mirdha 2020 |
| About 3 million | 2024 | Workers in the export-oriented industry, Mapped in Bangladesh | BRAC University CED 2024, 2 |
| 4,316,000 | 2022 | Labour Force Survey, “garment sector” as the minister reported it, which equals textiles plus wearing apparel (ISIC divisions 13 and 14) | UNB 2024; ILO 2026 |
| 3,459,745; 3,500,525; 3,706,077 | 2022; 2023; 2024 | Labour Force Survey, wearing apparel only (ISIC division 14), all employment | ILO 2026 |
The minister’s Labour Force Survey figure for 2022, 4,316,000 workers of whom 37.51% were women (UNB 2024), can be rebuilt from the ILO’s tables: 3,459,745 people in wearing apparel and 856,153 in textiles make 4,315,898, which rounds to the minister’s figure, and the women among them make 37.50%, against the minister’s 37.51% (ILO 2026). So the survey’s “garment sector” in that answer includes spinning, weaving and dyeing. Wearing apparel alone employed about 3.5 million people, and the survey counts everyone who makes clothes, including tailors and workers who sew for the home market. That total is only about half a million above Mapped in Bangladesh’s 3 million for export factories alone (BRAC University CED 2024, 2), although the survey also covers clothing made for the home market. One possible reason is that a household survey codes each person to one industry, so knitting and dyeing workers in vertical export factories may be counted under textiles rather than apparel. I could not find a study that reconciles the two sources.
The association figures are the hardest to use. The BGMEA and BKMEA databases together gave just over 5 million in 2024 (UNB 2024), but factories in both associations appear in both, and member lists include factories that have stopped. The “4 million” in BGMEA’s own publications comes with no method (BGMEA 2023, 4). For a count of workers in export factories, the census is the better source; for employment in clothing making nationally, the Labour Force Survey is.
Women’s share: why the estimates run from 35% to 65%
The industry’s reputation as an employer of women rests partly on a figure of about 80% that is still repeated. A 2023 study by BRAC University, funded by GIZ and the Ethical Trading Initiative, charts that figure for 1980 and 1994 and attributes it to BGMEA, without a document or method, and its text repeats it for the 1980s (Ethical Trading Initiative 2023, 4), and I found no original measurement either. CMB’s history page covers the trend briefly; here the question is method.
| Year | Women’s share | What is counted | Source |
|---|---|---|---|
| 1980; 1994; 1995; 2005 | 80%; 80%; 76%; 70% | Attributed to BGMEA in a chart; no method or original document | Ethical Trading Initiative 2023, 4 |
| 2010 to 2020 | 53.2% to 63.4% | Seven estimates by UN Women, CPD, SANEM, Farole and others, and Mapped in Bangladesh, as charted in the same study (I did not open the originals) | Ethical Trading Initiative 2023, 4 |
| 2015; 2018; 2021 | 54.22%; 53.89%; 53.65% | One survey of 337 factories, carried out in 2022, giving figures for three years | Ethical Trading Initiative 2023, 3–4 |
| 2021 | Woven 60.64%; knit 51.84%; sweater 43.90% | The same sample, by factory type | Ethical Trading Initiative 2023, 4 |
| December 2020 | 58.3% | Census of 3,223 export factories in four districts | Mirdha 2020 |
| June 2025 | 57% | Census of 3,315 export-oriented factories | BRAC University CED 2025, 6 |
| 2024 | 52.28%; 62% | BGMEA’s and BKMEA’s member databases, as the minister reported them | UNB 2024 |
| FY2022-23 | About 65% | BKMEA, knitwear sector; no method given | BKMEA n.d.-b |
| 2022 | 37.51% | Labour Force Survey, textiles and wearing apparel together | UNB 2024 |
| 2017; 2022; 2023; 2024 | 46.2%; 39.0%; 39.7%; 35.2% | Labour Force Survey, wearing apparel only | ILO 2026 |
Most of the spread comes from who is counted. Establishment counts in export factories give 52% to 58% in recent years; the household survey, which includes tailoring and textile mills, gives 35% to 40%. Within export factories, the share depends on the product: in the BRAC University sample women were 60.64% of workers in woven factories and 43.90% in sweater factories, and only 15.88% in knitting and 2.70% in dyeing, against 64.57% in sewing (Ethical Trading Initiative 2023, 4). It also depends on place and size: 71.47% in Chattogram against 48.44% in Narayanganj, and 61.23% in small factories against 52.05% in large ones (Ethical Trading Initiative 2023, 4). Since knitwear has grown faster than woven garments (BGMEA n.d.-b) (see chapter 5), some of the decline in the overall share would happen even if every factory kept the same mix of men and women. The knitwear association’s own figures, 62% and about 65% (UNB 2024; BKMEA n.d.-b), sit well above the BRAC sample’s 51.84% for knit factories, and neither states a method.
For a trend, a series that uses one method throughout is worth more than a longer series stitched from different sources. The BRAC University sample (2015 to 2021) and the Labour Force Survey tabulations (2017 to 2024) are the two I found. They agree on the direction, down, and disagree on the level by about 8 percentage points for 2017–18 and about 15 for 2021–22, because they count different people. The Labour Force Survey series also carries breaks: the ILO flags its 2010 and 2013 apparel figures as a change of methodology (ILO 2026). Why women are leaving is a different question, which the study for the Ethical Trading Initiative examines through surveys of current, former and potential workers (Ethical Trading Initiative 2023, 3).
Which series to use, and how to cite it
From the comparisons above, this is what I would advise a student or researcher.
- Bangladesh’s export value over time: the EPB series as compiled by BGMEA, by fiscal year (BGMEA n.d.-b), with the 2024 revision stated and FY2021-22 and earlier flagged as unrevised. The full series from FY1983-84 is a CSV download on our data page.
- Comparisons with other exporters: the WTO’s tables for clothing (SITC 84), naming the edition and how the EU is counted (WTO 2026), or partner data for HS 61 and 62 from UN Comtrade with the EU taken from Eurostat.
- Bangladesh’s share of a single market: the importer’s own statistics, such as Eurostat for the EU and US Census data for the United States, because they use the same method for every supplier.
- Macroeconomic work (balance of payments, GDP, reserves): Bangladesh Bank’s figures, which carry the revision the IMF describes (IMF 2025, 8).
- Factories: the Mapped in Bangladesh census with its date, for export-oriented factories (BRAC University CED 2025); association totals only as membership, never added together.
- Workers: the census for export factories; the Labour Force Survey, wearing apparel only (ISIC 14), for national employment in clothing (ILO 2026); never the sum of the association databases.
- Women’s share: one series with one method, with the population stated.
When citing a figure, give the publisher, the table, the date you read it, whether the year is fiscal or calendar, the valuation (FOB or CIF), the product definition (RMG, HS 61–62 or SITC 84), and whether the figure is revised. For downloaded data, give the query. The tables in this chapter can be downloaded as CSV files under CC BY 4.0, and the queries behind them are listed in my research notes so that anyone can repeat them.
Known breaks in the series
- FY2022-23 and FY2023-24 are marked revised, and later years were published after the revision; FY2021-22 and earlier are not marked revised, although the IMF’s technical assistance pointed to FY2018-19 to FY2021-22 (IMF 2025, 82–83).
- BGMEA’s calendar series marks 2023 and 2024 as revised and leaves 2022 unmarked (BGMEA n.d.-b), so the WTO’s 2023 change of −21.5% joins two bases (WTO 2026).
- The EPB stopped publishing in July 2024 (Daily Star 2024); check that any monthly series for 2024 is continuous.
- Bangladesh’s own reports to UN Comtrade end in 2018 and change from HS 2012 to HS 2017 in 2017 (UN Comtrade 2026).
- The EU lost the United Kingdom in 2020; use Eurostat’s EU27_2020 series for a consistent EU (Eurostat 2026a).
- The Labour Force Survey’s apparel series is flagged as a break in 2010 and 2013 (ILO 2026).
- Mapped in Bangladesh grew from four districts in 2020 to a nationwide census, so its early and late counts cover different areas (Mirdha 2020; BRAC University CED 2025, 6–7).
What the research does not yet tell us
- The true size of exports from FY2018-19 to FY2021-22. The IMF says these years need revisiting (IMF 2025, 82–83), and Ahmed’s table puts FY2021-22 US$4.2 billion above receipts (Ahmed 2024). Until a revised series is published, every growth rate across 2022 and 2023, including the peak year, carries an unknown error. A matched mirror test for those years, market by market, would help, and needs the EPB’s market data for them.
- A bridge between shipments and receipts. No public table splits the gap into timing, discounts and claims, cancelled orders, unpaid bills and recording errors. That split would also say something about who bears the cost of buyers’ purchasing practices, the subject of chapter 2.
- The EPB’s product definition. Without a published list of the customs headings in “RMG”, researchers cannot be sure that Bangladesh’s series and an HS 61–62 series measure the same goods.
- The gaps outside the EU and the United States. Canada records 17% to 26% more than Bangladesh on matched fiscal years without freight, Japan 13% to 17% more with it, and Mexico, the United Arab Emirates, China and Saudi Arabia far more again. Re-exports through hubs, attribution by origin and misreporting would each leave a different trace in transaction-level data, which is not public.
- An official count of export factories. DIFE’s licensing records and the BBS manufacturing surveys could give one, with a stated method; I could not open either. Mapped in Bangladesh fills the gap but cannot explain every change in its own totals.
- Informal subcontracting. The share of export production done in unregistered workshops has been disputed since 2016, from “more than 7,000” factories to a correction of at least two thousand (Labowitz and Baumann-Pauly 2015, 4; Anner and Bair 2016, 1). I found no new count since.
- Reconciling the worker counts. A census of export factories and a household survey of all apparel work give totals only about half a million apart, and women’s shares about 15 to 22 points apart. A study linking the two would settle how many people work in export garments and how many of them are women.
- Whether Bangladesh Bank now compiles trade statistics. The IMF benchmark was due in 2025 (IMF 2025, 24, 83); I found no public confirmation that it has been met.
How I researched this chapter
I carried out this review with the help of AI research tools, which searched for, read and summarised the sources and drafted the text to my brief; a separate AI check then reopened the sources and tested every figure, quotation and citation. I set the questions and the rules, and I am responsible for what is published here.
This is a desk review of published sources and public data, done on 22 November 2026. I did no interviews or fieldwork, and no figure comes from CMB’s records. Every source in the reference list was opened that day. I read in full: BGMEA’s export and membership pages and its November–December 2023 newsletter; BKMEA’s two pages; the IMF’s Country Reports 25/150 and 26/24 (as PDFs, searched for every passage on export data); Ahmed’s article, including the table in the printed edition; the Daily Star’s and UNB’s reports; the four WTO publications; both Mapped in Bangladesh documents (the 2025 booklet’s worker ratio is a picture, which I read from the rendered page); the ETI summary report; the Stern Center report and Anner and Bair’s critique; the RSC’s June 2026 report; the two ILO news pages; Eurostat’s and the US Census Bureau’s method pages; and Comtrade’s help page on the EU. For Fisman and Wei and for Gupta, Roy and Ahmad I read the abstracts only, and say so in the references.
The importer-side figures are my own downloads: UN Comtrade’s public preview interface (annual data for all partners, the EU aggregate and Bangladesh as reporter; monthly data for the United States, the United Kingdom, Canada, Japan and Australia), Eurostat’s Comext interface (annual and monthly EU data) and the ECB’s exchange rates. I added them up with a short script and checked that Eurostat’s monthly sums equal its annual figures, that Comtrade’s monthly and annual figures for the five other markets agree within 1%, and that Comtrade’s EU aggregate matches Eurostat within 0.4%. The queries are in the research notes for this chapter. An independent reader checks every figure in the tables against its source before the chapter is published.
What I could not open: Bangladesh Bank’s website, which refused automated and browser requests, so its own July 2024 release is known here only through the IMF and Ahmed; the EPB’s website; the US Office of Textiles and Apparel (OTEXA), whose “apparel” totals use their own definition and are quoted in the press; the BBS website, including the Labour Force Survey 2022 report, for which I used the ILO’s tabulations; and DIFE’s website. I also could not find out why Bangladesh’s own Comtrade figures for 2017 exceed BGMEA’s by about 9.5%, or why Mapped in Bangladesh’s count fell from 3,722 to 3,315 between 2024 and 2025.
References
Chicago author-date style. Web pages were read on 22 November 2026 unless the entry says otherwise.
- Ahmed, Sadiq. 2024. “What Are the Implications of Revised Export Figures?” The Daily Star (Dhaka), July 24, 2024. www.thedailystar.net/opinion/views/news/what-are-the-implications-revised-export-figures-3660881.
- Anner, Mark, and Jennifer Bair. 2016. The Bulk of the Iceberg: A Critique of the Stern Center’s Report on Worker Safety in Bangladesh. Research Brief, February 10, 2016. University Park, PA: Penn State Center for Global Workers’ Rights. ler.la.psu.edu/wp-content/uploads/sites/4/2021/11/CGWR-Critique-of-Stern-Report.pdf.
- BGMEA. 2023. “Paradigm Shift in Ensuring Workers Rights and Welfare.” Cover story. The Apparel Story, November–December 2023. Dhaka: BGMEA. bgmea.com.bd/uploads/newsletters/apparel-story-november-december-2023.pdf.
- BGMEA. 2026. “BGMEA Leaders Hold Meeting with DIFE on Factory License Renewal and SME Challenges.” News, May 18, 2026. bgmea.com.bd/page/BGMEA_Leaders_Hold_Meeting_with_DIFE_on_Factory_License_Renewal_and_SME_Challenges.
- BGMEA. n.d.-a. “BGMEA at a Glance.” Dhaka: BGMEA. Accessed November 22, 2026. www.bgmea.com.bd/page/aboutus.
- BGMEA. n.d.-b. “Export Performance: Comparative Statement on Export of RMG & Total Export of Bangladesh.” Dhaka: BGMEA. Accessed November 22, 2026. www.bgmea.com.bd/page/Export_Performance.
- BKMEA (Bangladesh Knitwear Manufacturers and Exporters Association). n.d.-a. “BKMEA at a Glance.” Narayanganj: BKMEA. Accessed November 22, 2026. www.bkmea.com/about/at-a-glance.
- BKMEA. n.d.-b. “Bangladesh Knitwear Industry.” Narayanganj: BKMEA. Accessed November 22, 2026. www.bkmea.com/about/knitware.
- BRAC University CED (Centre for Entrepreneurship Development). 2024. Mapped in Bangladesh: Impact Report 2022–24. Dhaka: BRAC University. ced.bracu.ac.bd/wp-content/uploads/2024/11/MiB-Impact-Report-2022_24_Final.pdf.
- BRAC University CED. 2025. Mapped in Bangladesh: Booklet 2025. Dhaka: BRAC University. ced.bracu.ac.bd/wp-content/uploads/2025/07/MiB-Booklet_2025_V11.pdf.
- Choudhury, Afshana, Fahim S. Chowdhury, Sadril Shajahan, and Faria Ahmad. 2021. “Beyond Digital Mapping of RMG Sector.” The Financial Express (Dhaka), September 6, 2021, updated January 28, 2023. thefinancialexpress.com.bd/views/economic-trends-and-insights/beyond-digital-mapping-of-rmg-sector.
- Daily Star. 2024. “Corrected Export Data Soon: EPB.” The Daily Star (Dhaka), September 2, 2024. www.thedailystar.net/business/news/corrected-export-data-soon-epb-3693096.
- Ethical Trading Initiative. 2023. Where Are the Women? A Study on the Declining Number of Female Workers in the Bangladesh RMG Industry. Summary Report. London: Ethical Trading Initiative, with BRAC University and GIZ. ethicaltrade.org/sites/default/files/shared_resources/ETI_GIZ%20Study%20on%20Women%20Workers.pdf.
- European Central Bank. 2026. “US Dollar/Euro Reference Exchange Rate,” monthly and annual averages (series EXR.M.USD.EUR.SP00.A and EXR.A.USD.EUR.SP00.A). ECB Data Portal. Accessed November 22, 2026. data.ecb.europa.eu/data/datasets/EXR/EXR.M.USD.EUR.SP00.A.
- Eurostat. 2026a. “International Trade of EU and Non-EU Countries since 2002 by HS2-4-6” (DS-059341). Comext, annual data, updated November 13, 2026. Luxembourg: Eurostat. Accessed November 22, 2026. ec.europa.eu/eurostat/databrowser/view/ds-059341/default/table?lang=en.
- Eurostat. 2026b. “EU Trade since 1988 by HS2-4-6 and CN8” (DS-045409). Comext, monthly data, updated November 13, 2026. Luxembourg: Eurostat. Accessed November 22, 2026. ec.europa.eu/eurostat/databrowser/view/ds-045409/default/table?lang=en.
- Eurostat. n.d. “International Trade in Goods: Detailed Data (ext_go_detail).” Reference metadata. Luxembourg: Eurostat. Accessed November 22, 2026. ec.europa.eu/eurostat/cache/metadata/en/ext_go_detail_sims.htm.
- Fisman, Raymond, and Shang-Jin Wei. 2004. “Tax Rates and Tax Evasion: Evidence from ‘Missing Imports’ in China.” Journal of Political Economy 112 (2): 471–96. Abstract read, in its NBER Working Paper 8551 version. doi.org/10.1086/381476.
- Gupta, Pranav Kumar, Devesh Roy, and Kaikaus Ahmad. 2012. Close Eye or Closed Eye: The Case of Export Misinvoicing in Bangladesh. IFPRI Discussion Paper 1157. Washington, DC: International Food Policy Research Institute. Abstract read. hdl.handle.net/10568/153827.
- ILO (International Labour Organization). 2015a. “National RMG Factory Inspection Initiative Hits 1,000 Mark.” News, June 24, 2015. www.ilo.org/resource/news/national-rmg-factory-inspection-initiative-hits-1000-mark.
- ILO. 2015b. “Safety Inspections for Targeted Bangladesh Export-Oriented RMG Factories Complete.” News, November 9, 2015. www.ilo.org/resource/news/safety-inspections-targeted-bangladesh-export-oriented-rmg-factories.
- ILO. 2026. “Employment by Sex and Economic Activity: ISIC Level 2 (Thousands), Annual” (EMP_TEMP_SEX_EC2_NB_A), Bangladesh, Labour Force Survey, ILO processing of the microdata. ILOSTAT. Accessed November 22, 2026. rplumber.ilo.org/data/indicator/?id=EMP_TEMP_SEX_EC2_NB_A&ref_area=BGD&format=.csv.
- IMF (International Monetary Fund). 2025. Bangladesh: Third and Fourth Reviews under the Extended Credit Facility Arrangement and the Arrangement under Extended Fund Facility; Press Release; Staff Report; and Statement by the Executive Director for Bangladesh. IMF Country Report 25/150, June 2025. Washington, DC: IMF. www.imf.org/-/media/files/publications/cr/2025/english/1bgdea2025001-print-pdf.pdf.
- IMF. 2026. Bangladesh: 2025 Article IV Consultation; Press Release; Staff Report; and Statement by the Executive Director for Bangladesh. IMF Country Report 26/24, January 2026. Washington, DC: IMF. www.imf.org/-/media/files/publications/cr/2026/english/1bgdea2026001-source-pdf.pdf.
- Labowitz, Sarah, and Dorothée Baumann-Pauly. 2015. Beyond the Tip of the Iceberg: Bangladesh’s Forgotten Apparel Workers. New York: NYU Stern Center for Business and Human Rights. bhr.stern.nyu.edu/wp-content/uploads/2024/02/beyond_the_tip_of_the_iceberg_report-1.pdf.
- Mirdha, Refayet Ullah. 2020. “Women Garment Workers on the Decline.” The Daily Star (Dhaka), December 14, 2020. www.thedailystar.net/business/news/women-garment-workers-the-decline-2011601.
- RSC (RMG Sustainability Council). 2026. Remediation Progress Rate Report: Bangladesh Safety Agreement. Data as of June 30, 2026. Published on the International Accord website. internationalaccord.org/wp-content/uploads/2026/08/RSC-BSA-Progress-Rate-Report_JUN-2026_FINAL.pdf.
- UN Comtrade. n.d. “EU in UN Comtrade.” UN Comtrade Help Center, Metadata. Accessed November 22, 2026. comtradewiki.azurewebsites.net/?p=113.
- UNB (United News of Bangladesh). 2024. “State Minister Informs Parliament of Number of Workers in RMG Sector.” June 23, 2024. unb.com.bd/category/bangladesh/state-minister-informs-parliament-of-number-of-workers-in-rmg-sector/137930.
- United Nations Statistics Division. 2026. UN Comtrade Database. Annual and monthly merchandise trade, HS chapters 61 and 62, Bangladesh (code 50) as reporter and as partner, queried through the public preview API (comtradeapi.un.org/public/v1/preview). New York: United Nations. Accessed November 22, 2026. comtradeplus.un.org/.
- US Census Bureau. 2022. “Description of the International Trade Statistical Program.” Guide to Foreign Trade Statistics, last revised October 4, 2022. Washington, DC: US Census Bureau. www.census.gov/foreign-trade/guide/sec2.html.
- WCO (World Customs Organization). 2022a. “Chapter 61: Articles of Apparel and Clothing Accessories, Knitted or Crocheted.” In Harmonized System Nomenclature, 2022 Edition. Brussels: WCO. www.wcoomd.org/-/media/wco/public/global/pdf/topics/nomenclature/instruments-and-tools/hs-nomenclature-2022/2022/1161_2022e.pdf.
- WTO (World Trade Organization). 2021. World Trade Statistical Review 2021. Geneva: WTO. www.wto.org/english/res_e/statis_e/wts2021_e/wts2021_e.pdf.
- WTO. 2022. World Trade Statistical Review 2022. Geneva: WTO. www.wto.org/english/res_e/booksp_e/wtsr_2022_e.pdf.
- WTO. 2023. World Trade Statistical Review 2023. Geneva: WTO. www.wto.org/english/res_e/booksp_e/wtsr_2023_e.pdf.
- WTO. 2026. “Table 22: Top 10 Exporters and Importers of Clothing, 2025.” In World Trade Statistics: Statistical Tables 2026 (spreadsheet, saved November 17, 2026). Geneva: WTO. www.wto.org/english/res_e/statis_e/world_trade_statistics_e.htm.
Sources
Checked . Rules and figures change, so confirm anything that affects your pricing.
- BGMEA, Export Performance (EPB data by fiscal year, calendar year and market) (opens in a new tab)
- IMF Country Report 25/150 (June 2025): the FY23 and FY24 export revision and trade statistics (opens in a new tab)
- UN Comtrade database (partners’ imports of HS 61 and 62 from Bangladesh) (opens in a new tab)
- Eurostat Comext DS-059341: EU trade by HS2-4-6 (opens in a new tab)
- WTO World Trade Statistics, Statistical Tables 2026 (Table 22, clothing) (opens in a new tab)
- BRAC University CED, Mapped in Bangladesh: Booklet 2025 (opens in a new tab)
- ILOSTAT: employment by economic activity, ISIC level 2, Bangladesh (Labour Force Survey) (opens in a new tab)
- Ethical Trading Initiative (2023), Where Are the Women? Summary report (opens in a new tab)
- Labowitz and Baumann-Pauly (2015), Beyond the Tip of the Iceberg, NYU Stern (opens in a new tab)
- Anner and Bair (2016), The Bulk of the Iceberg, Penn State CGWR (opens in a new tab)