Plus-Size Clothing Manufacturer for Small Brands (2026): 45+ Data Points on MOQ Per Colour, Size-Run Splits, and Per-Piece Cost

Key Takeaways

1. 14W-40W — the size range ASTM’s plus women’s body-measurement standard covers — 14 size steps, all graded at one standard height of 165 cm (5 ft 5 in), with the 14-step count calculated by us on the published range (ASTM International, D6960/D6960M-16e1 Standard Tables for Body Measurements for Plus Women’s Figure Type, T1).
2. Misses 16-18 (a 20W) — the clothing size the average American woman wears, from NHANES body measurements mapped onto the ASTM Misses and Women’s Plus tables — US women overall, mapped from 1988-2010 NHANES body measurements, not a plus-size customer base (Christel & Dunn, Washington State University, International Journal of Fashion Design, Technology and Education (2017), T1).
3. 95.7 cm vs 129.2 cm — median versus 95th-percentile waist circumference among US women aged 20 and over — a 55.4 cm spread from the 5th to the 95th percentile that a size run has to absorb in a handful of steps, a spread calculated by us from the published percentiles (CDC National Center for Health Statistics, Anthropometric Reference Data for Children and Adults: United States, August 2021-August 2023 (Series 3, No. 50), T1).
4. Sizes 10 to 30, average 18 — the size span and the centre of the customer base at the largest US plus-size specialty retailer, as reported to the SEC (Torrid Holdings Inc., Form 10-K for the fiscal year ended 31 January 2026, T1).
5. 4.0 out of 5 — China’s rating on minimum order quantity in the 2026 USFIA survey of 30 US fashion companies — the highest of any country in the study, against Cambodia at 3.0 and Bangladesh at 2.6 — a subjective 1-5 rating by mostly large buyers, not a minimum-order figure in pieces (United States Fashion Industry Association, 2026 Fashion Industry Benchmarking Study, T1).
6. 16.803 m to 19.513 m — fabric consumed by a 12-piece marker once 3XL joins a mixed XS-XXL ratio, as efficiency drops from 89.05% to 85.35% — 16% more fabric for the same number of garments, calculated by us and measured on a men’s long-sleeve shirt pattern rather than plus-size womenswear (Md Nazmul Haque, Southeast University Dhaka, International Journal of Textile Science 5(5) (2016), T1).
7. +1.848 m — extra fabric a single-size XXL marker consumed against the XL marker in the same experiment, at 1.36 points lower marker efficiency — the measured answer to whether bigger sizes cost more to cut, in the same single-pattern men’s long-sleeve shirt experiment (Md Nazmul Haque, Southeast University Dhaka, International Journal of Textile Science 5(5) (2016), T1).
8. 73% vs 3% — share of the variation in fabric consumption per garment explained by marker efficiency, against the share explained by how many sizes share the marker, across 258 modelled markers — shares of explained variation rather than percentages of fabric saved, in a straight-size men’s shirt simulation (Hanoi University of Industry, HaUI Journal of Science and Technology (May 2026), T1).
9. 0.9% to 28.6% — the span of US general duty rates across the two women’s apparel headings, 6104 and 6204. The rate turns on fibre content and construction — no line in either heading is graded by size (US International Trade Commission, Harmonized Tariff Schedule of the United States, 2026 Revision 19, T1).
10. 12.5% — the Section 301 forced-labor additional duty on products of China, applicable to goods entered for consumption on or after 24 July 2026 (Office of the United States Trade Representative, 91 FR 47318 (28 July 2026), T1).
11. -7.33% — year-on-year fall in operating income across China’s 13,684 above-designated-size garment enterprises, January to October 2025, with output down 3.20% over the same window (China National Textile and Apparel Council / China National Garment Association, compiling National Bureau of Statistics and China Customs data, T1).
12. 611 — US establishments making women’s, girls’ and infants’ cut-and-sew apparel in 2023, employing 7,413 people between them, with 50.6% employing fewer than five people, calculated by us from the published counts (US Census Bureau, County Business Patterns 2023, NAICS 315240, T1).

1. Why a Small Plus-Size Order Gets Quoted Away

A 100-piece plus-size order lands in an industry running on a 3.58% operating-income profit rate. That was the figure across China’s above-designated-size garment enterprises in January and February 2026, 0.23 percentage points below the same window a year earlier (CNTAC / CNGA Economic Operation Brief, January-February 2026, relayed by Texnet). At that margin, a four-size run of 100 pieces is not a small job. It is a changeover: a line stop, a fabric lay and a separate marker for every size.

The scale explains the reflex. 13,820 above-designated-size enterprises made 204.62 billion garments in 2024, and 20.07% of them lost money (China National Garment Association, Jan-Dec 2024 brief). Income is falling too: operating income across 13,684 such enterprises fell 7.33% from January to October 2025, with garment output down 3.20% over the same window (CNTAC / CNGA, Jan-Oct 2025 brief). Each of those figures covers a different window and a different enterprise panel, so none of them is chained into a trend here.

The domestic alternative is thin. 611 establishments in the United States make women’s, girls’ and infants’ cut-and-sew apparel, employing 7,413 people between them, and 50.6% of them have fewer than five staff (US Census Bureau, County Business Patterns 2023, NAICS 315240; the 50.6% is calculated by us from the published counts). US apparel manufacturing employment stood at 72,000 in August 2026 on a preliminary reading, down 68.5% from 228,800 in the twenty years to August 2026, a change we calculated from the published series (US Bureau of Labor Statistics, series CES3231500001). Whoever you are comparing a China quote against, you are usually comparing it against a workshop.

Stated buyer plans have shifted, a little. About 20% of the 30 US fashion companies USFIA surveyed between April and June 2026 planned to increase sourcing from China over the next two years, up from 6% a year earlier, though 40% still planned to reduce (USFIA, 2026 Fashion Industry Benchmarking Study). That is a stated plan from mostly large companies, not an outcome and not a small-brand sample.

What that leaves a small brand with is a need to understand how a minimum is counted before comparing two quotes. That arithmetic is the rest of this guide, and it is the same one behind our own plus-size production terms.

At a 3.58% operating-income profit rate, a four-size run of 100 pieces is not a small job. It is a changeover.

Every China figure here is cumulative year-to-date across a different window — full-year 2024, January-October 2025, January-February 2026. The enterprise panel changes between them, so each figure is quoted with its own period and none is chained into a trend.

2. What a 1X-4X Run Physically Spans

1X to 4X is a retail label. The measurement standard underneath it uses a different scale entirely: ASTM D6960/D6960M-16e1 lists plus women’s body measurements for sizes 14W through 40W — 14 size steps, a count we calculated from the published range — and it builds every one of them at a single standard height of 165 cm (ASTM International, D6960/D6960M-16e1). A size run grades girth. It never grades height.

That is a real simplification, because height does not hold still. US women aged 20 and over span 58.9 inches at the 5th percentile and 67.9 inches at the 95th, with a median of 63.5 inches (CDC National Center for Health Statistics, Series 3, No. 50), and the standard answers that spread with one number.

Girth is what the run has to absorb. Median waist circumference among US women aged 20 and over is 95.7 cm, with the 90th percentile at 121.5 cm and the 95th at 129.2 cm; hip runs 106.8 cm at the median and 131.5 cm at the 90th (NCHS, Anthropometric Reference Data, August 2021-August 2023). That is 25.8 cm of waist and 24.7 cm of hip between the median woman and the woman at the 90th percentile, both calculated by us from the published percentiles, and four size steps to cover it in if you launch 1X-4X. Waist is measured at the uppermost lateral border of the hip crest, so it is not a garment waist; none of these body figures carries ease or is a tech-pack specification.

Where those 14 steps actually fall, and what the grade increment between them should be, is a grading decision rather than a standards one, and it is a separate cost line. We do not rebuild it here: grading and fit blocks for extended sizes is its own guide.

The reference material is also older and patchier than the confidence with which size charts get quoted. D6960’s tables were built from US Department of Commerce data, the CAESAR study, the 2004 SizeUSA scan survey and Alvanon scans (ASTM D6960/D6960M-16e1, Introduction). SizeUSA scanned over 10,000 subjects in 12 US cities with over 200 standard measurements each (TC2, operator of SizeUSA) — thorough work, two decades ago. And the previous plus standard, D6960-04, stopped at 32W and was withdrawn in 2013 with no replacement listed on the ASTM record (ASTM D6960-04, Withdrawn 2013).

ASTM grades 14 plus sizes at one fixed height of 165 cm. Your size run grades girth. It never grades height.

One limit on all of the above: the ASTM measurement tables are paywalled and we read a four-page publisher preview, which carries the title page, scope and introduction only. Nothing on this page attributes a bust, waist or hip value or a grade increment to D6960. What the standard supplies here is its size range, its single 165 cm standard height, the datasets behind it and its edition history; every measured body figure comes from NHANES.

3. The Minimum Is Per Colour, Not Per Style: How 100 Pieces Splits Across a 1X-4X Run

Start with the word that does the damage. Our minimum on stock fabric is 100 pieces per colour, not per style. One style, one colour, 100 pieces. The same style in three colours is a 300-piece commitment before anyone has drawn a size ratio.

Now split it. A 1X-4X run is four sizes, so one colour at the minimum averages 25 pieces per size. Three colours across those four sizes is 300 pieces spread over 12 size-colour SKUs, still averaging 25 pieces each. Widen to six sizes and the same 100 pieces per colour averages 16 or 17 per size.

Read every one of those numbers as an average, because each assumes the 100 pieces split evenly across the run. That even split is an assumption rather than a recommendation or a plan. The arithmetic still decides whether an extended range is a product or a gesture, and these are our own terms worked through, not an industry benchmark.

There is no published size curve to plan against. We looked. No institution, trade body, retailer, standards organisation or peer-reviewed paper publishes a unit or revenue split across 1X, 2X, 3X and 4X. The percentages that circulate online come from market-research vendors that name no underlying measurer and do not reconcile with one another. Anyone who hands you a plus-size ratio is handing you a habit.

Why the colour is the unit at all is a dye-house question rather than a sewing-line one: a dye lot sets the floor. That economics sits in our guide to per-colour fabric minimums rather than being rebuilt here.

What the market does publish is how inconsistently minimums get counted. Four suppliers, four different bases, and every one of them sales copy rather than measurement. Hongyu Apparel says on an undated page that customised brands “usually have to start with 300 pieces per item per color”, while the same page elsewhere gives a per-design figure, so its own basis is internally inconsistent. DocShipper, a 2022-origin freight-forwarder guide carrying a March 2026 page-update stamp, lists Alibaba supplier examples of 150-300 pieces per colour for standard colours, 750-1,000 per colour for custom colours, 750-1,000 per order and 150-250 per size — four bullets, four bases, no survey and no methodology behind any of them. Shanghai Garment (Fumao Clothing) puts “100-300 pcs per color/style” on a page whose own scope is T-shirts in cotton or poly blends, while hoodies on the same page start at 300-500 pieces. ZEKA Apparel advertises 50 pieces per design with sizes and colours mixed inside that quantity, which is a structurally different offer again. Do not rank those four numbers against each other, and do not rank ours against them: they are not counting the same thing. Ask which unit a quote uses before you compare two quotes.

There is one institutional reading of China’s position on minimum order quantities. USFIA’s 2026 survey of 30 US fashion companies rated China 4.0 out of 5 on minimum order quantity, the highest of any country in the study, against Cambodia at 3.0 and Bangladesh at 2.6 (USFIA, 2026 Fashion Industry Benchmarking Study, Figure 2-11). It is a subjective 1-5 rating by mostly large buyers rather than a piece count, and it says nothing about any individual factory’s terms, including ours. We use it only as framing for why China still gets asked for varied work.

One word decides the order. Per colour, not per style — three colourways across 1X-4X is 300 pieces and 12 SKUs, averaging 25 pieces each.

4. Fabric Yield and the Marker: What the Top of the Size Run Actually Costs

Does the top of a size run cost more fabric? In one men’s long-sleeve shirt marker experiment, yes. A single-size XXL marker ran 1.36 points less efficient than the XL marker while consuming 1.848 metres more fabric (Md Nazmul Haque, International Journal of Textile Science 5(5), 2016). The bigger pieces stop nesting.

The larger effect shows up when the top of the run has nothing to nest against. A 12-piece mixed marker across XS-XXL, on the same men’s shirt pattern, ran 89.05% efficient at 16.803 metres. Add 3XL to the same ratio and efficiency falls to 85.35% while consumption climbs to 19.513 metres — 16.1% more fabric for the same 12 garments, calculated by us from the two published consumption figures.

What does not move the number much is how many sizes share the marker. Across 258 modelled markers in a May 2026 simulation, marker efficiency explained 73% of the variation in fabric consumption per garment, the number of garments per marker explained 21.5%, and the number of sizes in the marker explained about 3% (Hanoi University of Industry, HaUI Journal of Science and Technology 62(5), May 2026). Those are relative-importance shares of the model’s R-squared — shares of explained variation, not percentages of fabric saved. In the same paper’s simulated results table, efficiency sits flat at roughly 87-88% for every marker with two or more garments whatever the size mix, against 80.392% for a single garment. Efficiency comes from quantity in the lay, and at an average of 25 pieces per size a thin extended run gives the marker very little to work with.

Both studies are single-pattern experiments on men’s long-sleeve shirts, not plus-size womenswear, and one of the two is a software simulation rather than a cutting floor. Read the direction, not the magnitude. No published experiment cuts markers for a plus-size womenswear range, and the yardage premiums quoted around the top of an extended run trace to unnamed industry consensus with no measurer behind them, so none of them is used here.

Where the size steps themselves come from, and what the pattern work costs before a marker is ever made, sits in what grading actually costs.

Add 3XL to a 12-piece men’s shirt marker and the same 12 garments eat 16.1% more fabric.

None of this is a plus-size measurement. Both experiments cut a men’s long-sleeve shirt — one on a cutting floor in 2016, one in software in 2026 — and no published experiment cuts markers for a plus-size womenswear range. What transfers is the mechanism: a size with nothing to nest against costs fabric, and a shallow order gives the marker too little to work with. What does not transfer is any particular percentage, which is why the yardage premiums that circulate for the top of an extended size run are not quoted anywhere on this page.

5. Per-Piece Cost Across the Size Run: The Duty Line Is Fibre-Based, Not Size-Based

A useful negative finding first: US duty is not graded by size. General rates across the two women’s apparel headings run from about 0.9% on silk lines to 28.6% — cotton knit dresses at 11.5%, synthetic dresses at 16%, cotton woven trousers at 16.6%, synthetic trousers at 28.2% knit and 28.6% woven — and every one of those splits turns on fibre content and construction (USITC, Harmonized Tariff Schedule of the United States, 2026 Revision 19). A 4X and a 1X of the same garment enter under the same line at the same rate. If a quote raises the price of your top sizes and points at tariffs, the tariff is not the reason.

A separate 2026 development is the China-specific Section 301 action. A 12.5% ad valorem Section 301 forced-labor additional duty on products of China applies to goods entered for consumption on or after 12:01 a.m. eastern time on 24 July 2026, published at 91 FR 47318 on 28 July 2026 (USTR, Notice of Actions in Section 301 Investigations, 91 FR 47318). Whether a particular 6104 or 6204 line is exempt has to be checked against the notice’s Annex I and Annex II. Those annexes are published as images, we could not read them, and this page asserts nothing about coverage in either direction. Check your own line before you cost an order.

For a sense of level rather than of your quote: the average US-reported customs value per women’s garment imported from China in 2025 was $3.52 under HS 6104 and $5.54 under HS 6204, both calculated by us from published customs values divided by item counts (UN Comtrade, US imports from China, 2025). Treat both as indicative. They are heading-wide averages across dresses, trousers, skirts, shorts, jackets and suits, and US- and China-reported values for the same heading differ by as much as 3.65 times, so the two reporters do not reconcile (UN Comtrade, China-reported exports to the USA, 2024). Neither figure is a factory quote or an FOB price.

The same data show China’s share of those headings tilted towards units rather than value: 14.1% of US import value but 19.5% of items under HS 6104, and 15.7% of value against 23.7% of items under HS 6204 in 2025, calculated by us from a World-partner denominator and a China-partner numerator (UN Comtrade, US imports from the World and from China, 2025). Combined import value from China across both headings fell 30.2% between 2024 and 2025, again our arithmetic on two published years (UN Comtrade, US imports from China, 2024). 2025 saw substantial changes to US duty treatment of China-origin goods; the data do not say why volumes fell, and we are not going to supply a cause they do not carry.

China-origin apparel import prices, meanwhile, have barely moved. The BLS import price index for China-origin apparel read 107.4 in August 2026, roughly 4.7% above its January 2016 level and 8.6% above the June 2025 trough, both calculated by us from the published index levels (US Bureau of Labor Statistics, series EIUCOCHNZ315). Under five percent over the decade to August 2026. What a small brand feels as cost pressure sits in duties and in order size, not in what import prices show per garment — which is exactly why the per-colour minimum, rather than the per-piece price, is the number worth negotiating.

A 4X and a 1X enter the United States under the same tariff line at the same rate. If a quote blames the top sizes on tariffs, it is not tariffs.

How an FOB price is built — materials, labour, overhead, margin — and how it moves between 100 and 1,000 pieces is a separate question with its own guide. This section answers only the size-band question: what changes in the cost stack when the run gets wider at the top.

6. Where the Depth Goes: Sell-Through and Markdown Risk Across an Extended Run

Two public sources put a number on where a size range centres, and they describe different populations. Torrid, the largest US plus-size specialty retailer, tells the SEC that its own customers wear sizes 10 to 30, with an average of size 18 (Torrid Holdings Inc., Form 10-K, fiscal year ended 31 January 2026). Separately, a peer-reviewed mapping of NHANES body measurements taken between 1988 and 2010 onto the ASTM tables put the average American woman — all US women, not a plus-size customer base — between a Misses 16-18, which corresponds to a Women’s Plus 20W (Christel & Dunn, International Journal of Fashion Design, Technology and Education, 2017).

Our reading of the two: each sits below the midpoint of the range it describes. Neither gives you a split, because — to say it a second time — nobody publishes one. Any read on where depth should go is a judgement you are making, not a number you are citing.

That matters most where depth is thinnest. At an average of 25 pieces per size, selling through one size early while another sits until markdown is not an inventory accident. It is what thin depth does, whichever size turns out to be which. Days inventory outstanding across listed fashion companies reached an all-time high of 168 days in 2024, against the 2016-2019 average of 147 days (McKinsey & Company and The Business of Fashion, The State of Fashion 2026) — global listed companies across all categories rather than US plus-size.

Fit is where the money leaks back out. The average online apparel return rate across 100 surveyed US brands and retailers was 24.4% in 2023 (Coresight Research, March 2023, base of 100 US apparel decision-makers), while Torrid, a plus-size specialist, reported a 9% e-commerce return rate for 2020 (Torrid Holdings Inc., Form DRS/A, 2021) — different surveys, different years and different bases, so read the pair as an order of magnitude and never subtract one from the other. There is no published plus-size return rate. Our own view, which neither figure establishes, is that fit is the avoidable cost on an extended range, so the fit sample earns its money before a fifth colourway does.

One risk is worth naming because a public filing names it. Torrid’s 10-K for the year ended 31 January 2026 lists GLP-1 medication use as a demand risk to its plus-size assortment, with inventory management challenges, markdowns and lost sales as the stated consequences (Torrid Holdings Inc., Form 10-K, Risk Factors). It is a filed risk factor with no numeric size-mix figure attached, and not evidence that demand has shifted. What it does mean is that a first extended run is a bet on a distribution nobody publishes — and an even-split average of 25 pieces per size says nothing about which sizes would be the ones left over.

The largest US plus-size retailer’s average customer wears an 18 inside a 10-to-30 range. Nobody publishes how many 1X, 2X, 3X or 4X pieces that becomes.

Nothing in this section is a size curve. Two readings, of different populations, say roughly where a size range centres; neither says how many 1X, 2X, 3X or 4X pieces will sell, and no source found in this run does.

7. Producing Plus-Size Factory-Direct in Jiaxing: NewWay’s Terms

Our minimum on stock fabric is 100 pieces per colour, for dresses, sportswear and plus-size. Knitwear on stock yarn is 300 per colour. Custom-dyed fabric or yarn is 1,000. These are our terms rather than industry benchmarks, and the unit is the colour every time.

Samples take about 14 days, and bulk runs about 30 days after the pre-production sample is approved — counted from that approval rather than from order placement (NewWay published production terms, which carry both lead times). FOB sits between $6.50 and $26.80 depending on garment, fabric and construction; that range is not published on our public site, so it carries no outbound citation here. Every piece passes in-line inspection at the partner factory and a second inspection at our own facility in Jiaxing before packing. We hold BSCI and GRS certification, and we confirm current audit status at quote stage rather than printing a date that ages.

For a first extended-size run, the arithmetic in Section 3 is the conversation we would rather have up front. Four sizes in one colour averages 25 pieces each — an average, because an even split is an assumption and no published curve exists to weight it with. If that depth is too thin to trade, the fix is almost always fewer sizes with more depth in each, or fewer colours. It is rarely a bigger first order.

The rest of the process — tech pack in, lab dips, pre-production sample, the two inspection points — is set out in how we work.

Four sizes in one colour averages 25 pieces each, assuming an even split. If that is too thin to trade, take out a size before you add a colour.

Plus-Size Manufacturing for Small Brands by the Numbers: 20 Key Data Points (2026)

Methodology and Sources

This guide holds 49 data points: 40 from external sources and 9 first-party production figures from NewWay’s factory in Jiaxing, Zhejiang, each labelled in-table. Source-quality percentages are calculated on the 40 external stats only, so that our own numbers cannot flatter the score: 85.0% are Tier 1 (the primary measurer’s own document, page or API fetched and the figure confirmed in it), 5.0% Tier 2 and 10.0% Tier 3-flagged, giving a Tier 3 combined share of 10.0% against a 15% cap. Counting the first-party figures as primary-measurer data, 87.8% of all 49 points come from a primary measurer; counting them as unverified, the strict external Tier 1 share of the full set is 69.4%. Forty-nine candidate statistics were dropped during verification and assembly: every number that a search summary attributed to a page which did not contain it, all market-research size-band revenue splits (no named measurer, and they contradict other market sizings by a factor of six), most supplier MOQ self-claims, and several figures already owned by neighbouring NewWay articles. Every percentage, share, spread and per-piece value derived from customs data, census counts or published percentiles is our own arithmetic and is labelled as calculated, never as a published statistic. One gap is worth naming plainly: no institution, trade body, retailer, standards organisation or peer-reviewed paper publishes a unit or revenue split across 1X, 2X, 3X and 4X. Where this guide needed that number, it says so instead of supplying one.

NewWay is the primary measurer of its own minimum order quantities, lead times, FOB range and quality-control process. Those nine figures are one factory’s terms, not industry benchmarks, and they are excluded from the external tier percentages above. The per-colour MOQ figures and the FOB range rest on our internal production terms and are not currently published anywhere public, so those rows carry no outbound link rather than pointing at a page that does not support them. Some of our own public pages still describe the minimum as ‘per style’; that copy is wrong, it is being corrected, and the per-colour terms stated throughout this page are the authoritative ones. Nine uses sit inside the 8-10 uses-per-source cap that applies to external sources, so no cap exception is claimed. The most-used external source is CNTAC/CNGA material at five rows, spread across three separate published briefs.

Derived figures. Every percentage, share, spread and per-piece value on this page that comes out of customs data, census counts or published percentiles is our own arithmetic and is labelled as calculated where it appears: the customs unit values and the China import shares, the 2024-to-2025 import change, the percentile spreads, the 16.1% fabric increase, the establishment percentages, the employment percentage changes, and the per-size depth arithmetic in Section 3. None of them is a published statistic. A size-band split published by a market-research vendor was excluded rather than used, because it names no underlying measurer, covers all genders globally and implies a total market several times smaller than every other published sizing; its numbers are not reproduced here.

Recency notes

• Most recent available data: the plus-size body-measurement standard in force is ASTM D6960/D6960M-16(2023), reapproved in 2023 on a 2016 base edition. The body-scan datasets behind its tables, including SizeUSA, were collected around 2004, so the reference bodies are about two decades old.
• The NHANES anthropometric tables cover August 2021 to August 2023 and were published in June 2025. They are the newest measured US body-measurement reference available.
• The mapping of the average American woman onto a 20W uses NHANES body data from 1988 to 2010 and was published in 2017. It is the most recent published mapping of national body measurements onto the ASTM size tables that we could reach.
• The marker experiment behind the 3XL and XXL fabric figures was published in 2016 and is the most recent measured single-pattern comparison we could reach; the Hanoi University of Industry simulation is from May 2026.
• Coresight’s 24.4% apparel return rate is a March 2023 survey of 100 US decision-makers, and Torrid’s 9% is calendar 2020. Neither has a newer public equivalent.
• The four supplier pages quoted in the minimums section are not dated measurements. Hongyu Apparel and ZEKA Apparel carry no publication date, DocShipper’s guide was first published in June 2022 and its page is update-stamped March 2026, and Shanghai Garment’s page dates to about July 2025. They are quoted for the counting basis each one describes — per item per colour, per colour versus per order versus per size, per colour or style, and per design with sizes mixed — and not as current prices, as a benchmark, or as evidence about any factory’s terms today.
• China’s garment-industry figures are cumulative year-to-date across different windows: enterprise counts, revenue and output are January to October 2025, the profit rate is January and February 2026, and the production totals are full-year 2024. Each is quoted with its own period because the enterprise panel changes between them.
• US apparel manufacturing employment for August 2026 is a Bureau of Labor Statistics preliminary estimate and will be revised.
• County Business Patterns data are for 2023; the 2024 national file had not been published when this article was compiled.
• The Harmonized Tariff Schedule reading is from the 2026 Revision 19 release, effective 15 September 2026. Duty rates change between revisions; check the current release before costing an order.
• The days-inventory-outstanding figures are calendar 2024 data, reported in the 2026 edition of McKinsey and BoF’s State of Fashion, which was published in November 2025.
• NewWay’s own production terms — the per-colour minimums, lead times, FOB range and quality-control process — are current as of this page’s September 2026 update.

What we could not reach

• astm.org returns 403 to automated requests. The D6960 scope, standard height and source datasets were read from a four-page publisher preview, which does not contain the measurement tables. No bust, waist or hip value in this article is attributed to ASTM.
• The Section 301 notice publishes its Annex I and Annex II product lists as images, so we could not confirm whether HS 6104 and 6204 lines are exempt, nor how the new duty stacks with the older Section 301 China lists.
• OTEXA (trade.gov) was unreachable, so no US textile and apparel category or unit-price data from that source appears here.
• NPD/Circana press releases on plus-size market share were reachable only through the Internet Archive, and the Statista pages relaying them are paywalled. Those figures were excluded rather than relayed.
• A market-research size-band revenue split was reachable but names no underlying measurer, covers all genders globally, and implies a total market roughly six times smaller than every other published sizing of the same market. It was excluded, and its numbers are not reproduced anywhere on this page.
• The McKinsey State of Fashion 2026 PDF did not download from mckinsey.com in this session and was read from an Internet Archive copy.

Last updated: September 2026. We update this page quarterly.