Apparel Manufacturing (U.S., NAICS 315) — a subsector primer

A NAICS subsector (3-digit) in the U.S. industry taxonomy. NAICS = North American Industry Classification System, the standard code set the U.S., Canada, and Mexico use to define industries. This page rolls up three child industry groups — 3151 (knitting mills), 3152 (cut and sew), and 3159 (accessories) — synthesized from their individual primers plus our ground-truth federal statistics for this level.[4]

1. Overview

Apparel Manufacturing is the part of the clothing chain that physically makes garments in the United States — knitting yarn into socks and sweaters, cutting and sewing cloth into shirts and pants, and stitching the accessories (hats, gloves, belts, ties) that finish an outfit. It is a factory business, not the brand-and-retail business most investors picture when they hear "apparel."

One fact frames everything below. Americans buy roughly $362–366 billion of clothing a year at retail, but only about 2.5% of it is actually made in America — the other ~97.5% is imported.[3] The country landed $79.3 billion of apparel imports in 2023, with China alone supplying $16.9 billion (~21%) and China plus Vietnam just under 40%.[7] So this entire subsector — about $9.9 billion of domestic receipts[1] — is the surviving U.S. manufacturing base, a small remnant of what the country buys. The famous apparel tickers are almost all brand marketers whose clothing is sewn offshore; there is no sizable, publicly traded U.S. pure-play garment manufacturer. For most investors, NAICS 315 matters two ways: as a private-market niche (family mills, contract shops, defense/uniform suppliers, "Made in USA" brands) and as a barometer of tariffs and reshoring.

The erosion is now documented at this exact level, not just inferred from the children: one industry index put U.S. apparel-manufacturing output down about 17% in 2025,[5] and the Bureau of Labor Statistics projects employment across NAICS 315 to fall from 84,500 in 2024 to 58,800 in 2034 — a 30.4% decline, one of the steepest projected declines of any U.S. industry.[6]

Because this is a rollup, its distinctive value is the contrast across the three children — so Section 2 leads with a side-by-side comparison before we cover the subsector as a whole.

---

2. What's inside — the three children and how they differ

NAICS nests: a 3-digit subsector (315) contains 4-digit industry groups. Subsector 315 contains exactly three, and they are three different economic animals:

• 3151 — Apparel Knitting Mills. Factories that knit yarn directly into finished garments — socks, hosiery, sweaters, tees, underwear — in one integrated operation. The most capital-intensive child (knitting machines, not hand-sewing), and the one with a scaled public name.[44]
• 3152 — Cut and Sew Apparel Manufacturing. The largest child by far: cutting and sewing finished cloth into clothing. It splits into a fee-for-labor contractor half (31521, ~27% of the group's receipts) and an own-the-goods manufacturer/brand half (31525, ~73%). This is where the recognizable apparel tickers classify — even though they manufacture offshore.[44]
• 3159 — Apparel Accessories and Other Apparel Manufacturing. The "everything else" corner — hats and caps, gloves, belts, neckties, trimmings. Its most defensible value sits in licensed headwear and men's-accessory licensing platforms.[44]

Two of the three (3151 and 3159) are effectively pass-throughs — each contains a single 5-digit child, so the group is that industry. (3151's single child, 31512/315120, is itself the product of a 2022 NAICS revision that merged the old hosiery-and-sock and other-knitting codes.) Only 3152 branches further, into contractors and manufacturers.[44]

(HHI = Herfindahl-Hirschman Index, a 0–10,000 concentration score; regulators treat anything under 1,500 as "unconcentrated." CR4 / CR50 = combined revenue share of the four / fifty largest firms.)[1]

The one-line read: 3152 is roughly three-quarters of the whole subsector on every measure — receipts, firms, and jobs — and it is where the tradable tickers live (as brand, not factory, plays). 3151 and 3159 are similar-sized tails (~12% of receipts each) but structurally opposite in shape: knitting is a few larger, capital-intensive mills (highest revenue per firm, most concentrated), while accessories is a cloud of tiny shops. All three are secularly shrinking, all three benefit only at the margin from tariffs and reshoring, and none offers a clean domestic-manufacturing public stock.

The newly visible contrast is head-weight. Read down the CR50 row and the three children stop looking alike: it takes the top 50 knitting firms 82.5% of receipts, the top 50 accessory firms 63.1%, and the top 50 cut-and-sew firms only 41%.[1] Knitting has a real middle tier of mid-size mills doing most of the domestic work; cut-and-sew has essentially none — it is a long tail with nothing on top of it. That single row explains why capital, ownership change, and any plausible roll-up thesis behave so differently in 3151 than in 3152.

And mandated demand is not spread evenly either. Defense Logistics Agency obligations in fiscal 2024 ran $624.3 million under the cut-and-sew contractor code against $101.0 million under the manufacturer code — roughly six times more defense sewing routed through the half of 3152 with no public exposure at all.[12] (Those are contract dollars awarded in a different year from the census figures above, so read them as a channel comparison, not a revenue share.) The knitting and accessories children describe Berry-compliant work qualitatively but do not report a comparable dollar figure.[44]

---

3. Size — the subsector's rollup figures

Our ingested ground-truth federal statistics for NAICS 315 (the three children combined):

What the numbers say:

• The children tie out cleanly to this total. Receipts ($1.22B + $7.52B + $1.20B ≈ $9.94B), establishments (178 + 3,430 + 729 = 4,337), employment (~7,656 + 47,582 + 7,925 = 63,163) and payroll all sum to the subsector figures — a useful cross-check, and confirmation that 3152 does about three-quarters of the work.[1][2]
• Micro-scale. Average revenue is ~$2.1 million per firm and average pay is ~$41,200 per worker — well below the U.S. manufacturing average, reflecting a base of small, labor-intensive shops.[1][2]
• More fragmented than any of its children. The subsector HHI of 31.6 is lower than every child's (48.8 / 151.4 / 273.8), and CR4 of 7.3% is below every child's (9.6% / 19.1% / 24.3%). That is not an error: pooling three already-fragmented industries measures each large firm's share against a bigger combined base, so the biggest apparel makers look even smaller here. It takes the top 50 firms combined just to reach 34.3% of the subsector — no one has pricing power.[1]
• Concentrated in place even while atomized in ownership. California holds roughly 1,332 cut-and-sew contractor establishments and 10,800 contractor employees — about 64% of the nation's contractor plants and 48% of its contractor employment[39] — and the Los Angeles garment cluster is put at ~45,000 workers on a trade-wide basis that spans more codes than these.[44] For the subsector's largest child, California labor law is the cost floor, not a regional footnote.

The children now carry second federal series, and they disagree on level while agreeing on direction. This is the most useful new caveat at this level, because the disagreement is systematic rather than isolated. In 3151, County Business Patterns records ~7,656 workers in 2023 while the BLS productivity series implies roughly 6,100 in 2024 — vintage and method, not contradiction, on a base whose hours fell 17.0% in that single year.[2][8] In 3159, the 2022 Economic Census puts shipments at $1.20 billion against BLS sectoral output of $1.10 billion for the same year, and CBP's 7,925 paid employees against BLS's 14,800 all workers (which includes the self-employed).[1][2][9][10] And at this level, BLS puts NAICS 315 employment at 84,500 in 2024 against CBP's 63,163 employer-establishment count for 2023.[6][2] The programs measure and deflate differently; treat the census figures as the narrow, employer-only floor and the BLS figures as the broader read, and note that every one of these series is falling.

Undercount caveat — three different kinds. These figures count only clothing physically produced in U.S. establishments, so they understate the category's real footprint. First, for knitting (3151) the gap is offshore, not hidden domestically: knit-sock and hosiery imports alone ran ~$2.63 billion in 2024, more than double all domestic knitting-mill receipts.[25] Second, for cut-and-sew contractors (3152) and accessories (3159), County Business Patterns counts only employer establishments — it misses home-based sewers, one-person contractors, custom milliners, and informal shops, the shadow segment behind the Los Angeles wage-theft cases; broader private counts run into the thousands more firms.[26] Third — the mechanism the accessories child adds — the category's biggest names design and market at scale but source production offshore, so their revenue never lands in a U.S. manufacturing code at all.[44] Read $9.9 billion as a floor on the surviving domestic base, not a measure of the ~$362–366 billion American clothing market.[3]

---

4. Investable universe — where value concentrates across the children

Value and tradable exposure are not spread evenly across the three children. They sit almost entirely on the 3152 side, and even there they are brand businesses, not factories. (Tickers appear here and in Section 10 only.)

• Public equity lives in 3152's "jobbers" — and their own filings now document how offshore they are. Ralph Lauren sources 96% of product value outside the U.S.;[31] G-III Apparel took ~76% of its fiscal-2025 product from Vietnam, China and Indonesia;[32] Levi Strauss sources nearly all product through independent contract manufacturers in roughly 32 countries;[33] VF Corporation sources from ~273 facilities in ~30 countries and Kontoor Brands makes only ~30% internally, buying ~70% from ~210 contract factories across 19 countries.[44] Judge the set — VF (VFC), PVH Corp. (PVH), Ralph Lauren (RL), Levi Strauss (LEVI), Columbia Sportswear (COLM), Kontoor Brands (KTB), Carter's (CRI), Oxford Industries (OXM), Guess? (GES), G-III Apparel (GIII) — as branded consumer-discretionary equities, not domestic-factory exposure.
• The one scaled "manufacturer" proxy sits in 3151 — and classifies in 3152 too. Gildan Activewear (NYSE/TSX: GIL) is a vertically integrated knitter of tees, fleece, underwear, and socks — FY2025 net sales of $3.62 billion at ~21.5% adjusted operating margin[28] — that in December 2025 acquired HanesBrands and roughly doubled its size.[27] It knits offshore (Central America/Caribbean, Bangladesh): a low-cost global commodity-knitting play, not U.S.-domestic exposure. HanesBrands is no longer a standalone ticker; treat that exposure as sitting inside GIL.[27]
• Uniform and workwear services are the closest public thing to domestic garment production — and still not the thing itself. Cintas (CTAS) runs five manufacturing facilities, and UniFirst (UNF) manufactured about 62% of the garments it placed in service in fiscal 2025; Superior Group of Companies (SGC) and Lakeland Industries (LAKE) sit nearby.[44] Their economics are recurring rental and service revenue, so they price like services, not merchant manufacturing.
• Two of the three children reach public markets mainly through Berkshire Hathaway. Fruit of the Loom (3151) manufactures 89% of its products in its own global facilities, with North American cloth production primarily in Honduras;[29] Fechheimer (3152) makes public-safety, postal and military uniforms.[30] Both are wholly owned and immaterial to the parent — a link, not an investment.
• 3159 has no public pure-play at all. Accessory exposure reaches public markets only as one small line inside diversified apparel houses. The category's defining businesses are private: New Era, which after its 2024 acquisition of '47 holds licences across all five major North American team-sport leagues,[36][44] and men's-accessory licensing platforms such as Randa.[37]
• Genuinely domestic-manufacturing exposure — in every child — is private and small: family sock/hosiery mills (Renfro, ~$540M revenue; Darn Tough),[38][44] the Los Angeles contract-sewing cluster, "Made in USA" direct-to-consumer (DTC) brands (American Giant, Los Angeles Apparel), licensed-headwear makers, and Berry-compliant defense/uniform suppliers such as ReadyOne Industries, Mills Manufacturing and Propper.[44] Verifiably U.S.-produced public exposure is very thin — micro-cap Digital Brands Group (DBGI), with Los Angeles production, is the exception that proves it.[44] Reaching the rest means private equity (PE), direct ownership, or a supplier relationship — not a ticker.

Bottom line: public routes are almost all brand plays (offshore-made); real domestic-factory exposure is private. The three child primers carry the full company-by-company maps.[44]

---

5. How the money works

Within this one subsector you have three different economic models, which is why "apparel manufacturing" as a single number tells you very little. The economics are those of commodity manufacturing and branded consumer goods — capacity utilization, input cost, and per-garment labor pricing — not utility rate base, real-estate rents, or resource extraction.

• Knitting (3151) is a capacity-utilization and input-cost business. Owners earn the spread between what a knit garment sells for and the cost to convert yarn into it. Yarn (cotton, polyester, nylon, wool, spandex) is the dominant cost, so margins swing with commodity cycles and large integrated players may hedge inputs 12–24 months forward; Census data puts annual payroll at roughly 26% of sales before benefits. Keeping capital-intensive knitting machines full is what turns thin gross margins into profit, and vertical integration (spin–knit–finish–sew, the Gildan model) captures more of the margin.[44]
• Cut-and-sew contractors sell labor and machine time. Priced per garment under CMT — cut, make, trim (the brand owns the fabric; the shop bills sewing labor plus a thin margin). The math is cost-per-minute × SMV — standard minute value, the industry's estimate of the sewing time a garment needs. Contractors typically add only ~10–15%, so profit lives or dies on utilization, labor efficiency, and product mix. The reason that margin cannot widen is buyer power: brands keep no long-term manufacturing commitments, so a domestic shop competes not with the shop down the street but with thirty-odd countries' worth of alternatives.[32][33][44]
• Manufacturers and brand "jobbers" sell the garment itself. Because U.S. sewing wages are a large multiple of Asian wages — roughly $15–$20/hour domestically against under $1/hour in Bangladesh — domestic makers can't win on price; they earn on speed/replenishment, small-batch runs too short to import economically, a "Made in USA" premium, and compliant (domestic-only) contracts.[43] The branded jobber earns differently again: value sits in brand, design, and intellectual property (IP), sewing is outsourced offshore, and healthy branded-apparel economics run roughly 60–70% gross margin at retail but only ~40–50% at wholesale — a distinction worth holding, since a mostly-wholesale name like Kontoor Brands posted a 44.5% gross and 13.1% operating margin in 2024.[44] Accessories (3159) follow the same split, with the richest niche being headwear royalties and licensing.

The common thread across all three is a labor-heavy cost stack that is getting less efficient. The children measure it differently but land in the same place: payroll is ~26% of sales in knitting, labor is 36.2% of total production cost in accessories, and average pay runs ~$36,700 in cut-and-sew contracting versus ~$46,000 on the own-the-goods side.[11][44] In accessories, where BLS publishes the series, the labor-productivity index fell from 77.7 in 2022 to 73.2 in 2024 while unit labor costs rose 12% — a labor-heavy structure getting worse at converting hours into output, in an industry that cannot pass costs through.[11]

So one subsector spans a ~10–15%-margin labor job shop, a thin-margin domestic factory, a capital-intensive commodity knitter, and a brand business at 40–70% gross margin depending on channel. Two ways to win recur across all three children: global low-cost scale (Gildan) or a protected/premium niche (military, medical compression, performance wool, licensed headwear, quick-turn private label).

---

6. Demand drivers

• Overall clothing spending, which tracks consumer income, employment, and confidence. Clothing is discretionary and cyclical, though basics (socks, underwear) are staples and less cyclical than fashion. U.S. households averaged $2,001 of apparel-and-services expenditure in 2024, down 2.0% from 2023.[42]
• Fashion and activity trends — the athleisure and performance boom lifts knit and technical categories; streetwear lifts headwear; formality cycles move neckwear.
• Import competition and the dollar — now the single biggest swing factor for domestic output specifically; with ~97.5% imported and $79.3 billion landed in 2023, domestic demand is a niche created by speed, small runs, and "Made in USA" preference, not by total clothing demand.[3][7]
• Trade policy and tariffs — the dominant near-term variable; average U.S. apparel-import tariffs rose from ~14.7% in January 2025 to ~35.1% by December 2025.[5] Section 301 duties add roughly 7.5–25% on Chinese goods on top of already-high most-favored-nation apparel rates.[22]
• Institutional / mandated demand — military uniforms and government workwear are a durable domestic-only floor, and the measurable dollars land disproportionately on cut-and-sew contractors ($624.3M of FY2024 DLA obligations).[12]
• Input costs — USDA's July 2026 outlook projects world cotton production of 117.3 million bales against mill use of 122.0 million, with ending stocks down 6% — a tightening fiber market that feeds straight into every child's cost line.[41]
• Seasonality — cold-weather gloves and winter hats, licensed sports caps, Q2 tees and fall/winter fleece, and holiday/back-to-school peaks whipsaw orders and inventory.
• Reshoring / nearshoring, best read as a portfolio tool rather than a wholesale shift: brands keep Asian factories for predictable high-volume programs while reserving domestic capacity for test runs and fast replenishment.[44]

---

7. Regulation

Apparel manufacturing is lightly capital-regulated but carries meaningful labeling, safety, labor, environmental, and trade rules — and trade policy is the regulatory story for the domestic base:

• De minimis repeal (2025) — and a genuine disagreement among the children on its scope and timing. The rule letting parcels under $800 enter duty-free (Section 321 of the Tariff Act) was curtailed in 2025, the biggest policy tailwind for domestic makers in years. The children date it differently: the knitting primer describes elimination for Chinese-origin goods in May 2025;[15] the cut-and-sew primer describes a 2025 suspension with full repeal pending in 2027;[16] the accessories primer describes a suspension effective August 2025.[17] These are three staged actions on the same policy rather than three contradictory facts, but the effective date that matters to any given importer depends on origin and shipment type — do not treat a single date as settled.
• "Yarn-forward" rules of origin. Under USMCA (U.S.–Mexico–Canada Agreement) and CAFTA-DR (Central America–Dominican Republic Free Trade Agreement), knit apparel gets duty-free access only if spinning, knitting, and assembly all happen inside the bloc — which is why so much U.S.-brand knitting sits in Central America.[14]
• Berry Amendment (10 U.S.C. §4862). The Department of Defense (DoD) must buy U.S.-grown-and-made clothing, fiber, yarn, and footwear — on the order of ~$2 billion a year.[13] It is a genuine moat, but the rollup detail corrects a natural assumption: the moat is mostly a cut-and-sew contractor's moat, with FY2024 DLA obligations of $624.3 million under the contractor code against $101.0 million under the manufacturer code.[12]
• Labeling and safety. The Federal Trade Commission (FTC) requires fiber content, country of origin, and care labeling; the Consumer Product Safety Commission (CPSC) enforces the Flammable Fabrics Act, with stricter children's rules under CPSIA (Consumer Product Safety Improvement Act). These apply to accessories exactly as they do to garments.[20]
• Forced-labor screening (UFLPA — Uyghur Forced Labor Prevention Act). Presumes goods tied to the Xinjiang region are barred from entry, with no de-minimis exception to the ban itself; enforcement is not theoretical — CBP detained 6,636 shipments in the first half of 2025 alone, and the entity list has expanded to 144 names. Apparel and cotton are priority sectors, so this raises import-compliance cost and modestly favors verifiable domestic supply.[21]
• State labor law. California's Garment Worker Protection Act (SB 62, 2022) banned piece-rate pay for garment work, required an hourly minimum, and made brands jointly liable for wage theft in their contractors' shops, with a $200-per-employee, per-pay-period penalty.[18] A Department of Labor survey of 50 Southern California garment contractors found violations in 80% of cases and recovered more than $892,000 for 296 workers.[19] With ~64% of the subsector's contractor plants in California,[39] this is the binding cost floor for the domestic sewing base — though note the 80% figure covers the garment-contractor sector, not knitting mills or accessory shops.[44]
• Environmental — a 3151-specific exposure the other children do not carry. Wet finishing brings wastewater and energy obligations: EPA textile-mill effluent guidelines regulate pollutants, and EPA is studying PFAS use in textile mills, a live source of future cost.[24]
• Extended producer responsibility — new and forward-dated. California's textile EPR program, with regulations taking effect no earlier than July 2028, will add fees, reporting, and product-accounting duties for brands and other statutory "producers" selling in the state.[23]

---

8. Consolidation

The subsector total (HHI 31.6, CR4 7.3%) hides two opposite dynamics:

• At the making stage — brutal and fragmenting, but not equally so. The binding constraint is a labor-cost gap scale cannot close, so the factory tiers of all three children are among the most fragmented manufacturing you will find. Where they differ is head-weight: the top 50 firms hold 82.5% of receipts in knitting but only 41% in cut-and-sew (and 63.1% in accessories).[1] In cut-and-sew, consolidation happens by exit, not acquisition — contractor establishments fell 28% in six years.[40][2] In knitting the pattern is ownership change rather than scale-building: Renfro was acquired by The Renco Group in 2021.[38] Survivors specialize — defense/uniform, technical/protective wear, fast-turn replenishment, premium "Made in USA," licensed headwear — rather than chase commodity volume. Weak commodity operators keep failing: Delta Apparel filed Chapter 11 in mid-2024 and was delisted; the children disagree on what its Salt Life brand fetched, one reporting ~$38.7 million[34] and the other $28 million[35] — a reminder to verify deal prices at the source before relying on them.
• At the brand stage — consolidation is the story. Value has migrated from factories to brands, and the branded owners keep buying scale. The defining event is Gildan's completed acquisition of HanesBrands on December 1, 2025 (~$2.2B equity / ~$4.4B enterprise value, ~$200M targeted synergies), creating a dominant vertically integrated activewear-and-innerwear player.[27] In accessories, New Era completed its acquisition of '47 in 2024, consolidating licensed headwear across all five major North American team-sport leagues — brand and licence consolidation, not domestic sewing capacity.[36][44]
• On the pure-manufacturing side, what roll-up activity exists is concentrated in uniforms and workwear private equity, not the fashion-facing base.

Expect continued bifurcation: bigger, sourcing-efficient brand platforms on top; a thinning tail of small domestic manufacturers and contractors below.

---

9. Risks

• Structural import competition is existential. ~97.5% import penetration, $79.3 billion of imports in 2023, and a persistent labor-cost disadvantage keep shrinking the domestic base — with BLS projecting a further 30.4% employment decline across NAICS 315 through 2034.[3][5][6][7]
• Thin, fragile margins. A small move in wages, rent, input costs, or utilization can flip a contractor or domestic factory to a loss.[1]
• Buyer power and customer concentration. Brands hold no long-term manufacturing contracts and source across dozens of countries, so one lost program can take a shop's utilization with it — and for independent knitters, a consolidating customer base means less pricing power still.[32][33]
• Tariff / trade-policy whiplash — now the dominant variable, cutting both ways (tailwind for domestic makers, direct margin hit for import-reliant brands); policy can change fast, and sourcing shifts between low-cost countries more readily than it returns home.[5]
• Input-cost volatility — cotton, oil-linked synthetics, and freight move factory costs directly, against a world cotton balance now drawing down stocks.[41]
• Rising unit labor costs. Where BLS measures it, productivity is falling and unit labor costs are rising (accessories: index 77.7 → 73.2, ULC +12%) — the squeeze is structural, not cyclical.[11]
• Labor and legal exposure — skilled-sewer scarcity, rising wages, UFLPA detentions, CPSIA/flammability liability, and SB-62-style wage-liability laws that reach up to brands, against an 80% violation rate in recent DOL garment inspections.[18][19][20][21]
• Environmental and compliance exposure — wet finishing faces EPA effluent rules and potential PFAS obligations; California textile EPR adds a forward-dated cost.[23][24]
• Small-firm fragility — most firms are tiny and thinly capitalized, exposed to demand shocks, credit tightening, and key-customer loss.[1]
• Reshoring may simply disappoint — mill, supplier and labor infrastructure takes years to rebuild, so tariff-driven intent converts slowly into domestic orders.[44]
• Fashion / brand risk (the public jobbers) — a jobber's value is its label, and labels are perishable.
• Illiquidity — genuine domestic-manufacturing exposure is private and hard to exit.

---

10. How to invest & outlook

Public routes are plentiful but almost entirely brand plays classifying in 3152 — judge the Section 4 equities as consumer-discretionary brand businesses (brand strength, gross margin, inventory discipline, DTC mix, tariff exposure), not domestic-factory bets, because their goods are mostly sewn offshore, as their own filings state.[31][32][33] The closest thing to a scaled manufacturer is Gildan (GIL), but it is a low-cost offshore commodity-knitting play.[28] The uniform-service names (CTAS, UNF) get closer to real domestic garment production than any apparel brand does, but they are recurring rental-and-service businesses and price like services.[44] Verifiably U.S.-produced public exposure is very thin (micro-cap DBGI); broad apparel or consumer-discretionary funds are the diversified alternative. Reserve any judgment on yields or valuation multiples for company-level analysis.

Private routes are where real domestic-manufacturing exposure lives — in any child: backing or building family knitting mills (3151), contract sewing shops or "Made in USA" manufacturers (3152), or headwear/glove/accessory makers (3159) — ideally in a defensible niche: Berry-compliant defense/uniform, quick-turn DTC, performance/technical, or licensed headwear, plus uniform/workwear PE roll-ups. Note where the mandated money actually lands: the $624.3 million of FY2024 DLA obligations under the contractor code says the compliance moat is strongest on the half of the largest child that has no listed comparable at all.[12] Many family mills also face succession questions — a classic small-cap/PE opening. The diligence variables that recur across all three children are customer concentration, machine age and utilization, operator retention, input-price pass-through, and environmental obligations. The thesis rests on speed-to-market, small-batch flexibility, compliance moats, and any durable tariff-driven reshoring premium — not on out-competing imports on price.

Outlook (forward-looking judgment). The multi-decade decline of commodity U.S. apparel making is unlikely to reverse, even though 2025 delivered the strongest policy tailwind in a generation (de minimis curtailment plus sharply higher tariffs). The evidence has, if anything, hardened: output down ~17% in 2025,[5] BLS projecting a 30.4% subsector employment decline by 2034,[6] and each child's second federal series confirming the trend even where it disputes the level.[8][9][10] Industry evidence is that it will take "more than tariffs" to reshore volume, given thin domestic spinning, knitting, capex, and skilled labor — and that sourcing shifts between low-cost countries far more readily than it returns home.[5][44] The most durable value sits where it already is: global low-cost scale (Gildan) at one end and legally or functionally protected niches (military, medical, performance, licensed headwear) at the other, with the undifferentiated middle continuing to erode. For most investors, NAICS 315 is best understood as a barometer of reshoring and trade policy and as a private acquisition or founder opportunity, with the liquid, tradable action sitting on the branded, import-sourced side.

→ For the full detail behind this rollup, read the three child primers: 3151 — Apparel Knitting Mills, 3152 — Cut and Sew Apparel Manufacturing, and 3159 — Apparel Accessories and Other Apparel Manufacturing.

---

Sources

1. U.S. Census Bureau. 2022 Economic Census — receipts, firms, and concentration ratios (CR4/CR8/CR20/CR50, HHI), NAICS 315 and children. 2022. Our ground-truth federal stats for this level. https://www.census.gov/programs-surveys/economic-census.html
2. U.S. Census Bureau. County Business Patterns 2023 — establishments, employment, payroll, NAICS 315 and children. 2023. https://www.census.gov/programs-surveys/cbp.html
3. AllAmerican.org. State of American Clothing Manufacturing Report — ~2.5% domestic / ~97.5% imported; ~$362–366B U.S. retail apparel market. 2024–2025. https://allamerican.org/research/clothing-manufacturing-report/
4. U.S. Census Bureau / NAICS Association. NAICS 2022 — Subsector 315, Apparel Manufacturing (definition and structure: industry groups 3151, 3152, 3159). 2022. https://www.census.gov/naics/
5. WWD / Sourcing Journal. US Apparel Manufacturing Fell 17% in 2025 (Kearney Reshoring Index; apparel-import tariffs 14.7%→35.1%). 2026. https://wwd.com/sourcing-journal/trade/kearney-reshoring-index-usa-domestic-apparel-manufacturing-1238945612/
6. U.S. Bureau of Labor Statistics. Industries with the fastest projected employment declines, 2024–2034 (NAICS 315: 84,500 → 58,800, −30.4%). https://www.bls.gov/emp/tables/industries-fast-decline-employment.htm
7. U.S. International Trade Commission. U.S. Imports for Consumption of Textiles and Apparel / Apparel Competitiveness (2023: $79.3B total; China $16.9B; China and Vietnam ~40%; Bangladesh/Cambodia/India/Indonesia/Pakistan $21.4B). 2024. https://www.usitc.gov/publications/332/pub5543_0.pdf
8. U.S. Bureau of Labor Statistics. Productivity and Costs by Industry — apparel knitting output, hours, and productivity trends (2024; 1987–2024 series). April 2025. https://www.bls.gov/news.release/archives/prin_04242025.htm
9. U.S. Bureau of Labor Statistics. All Workers — NAICS 315990 (14,800 in 2025; 17,000 in 2022), via FRED. https://fred.stlouisfed.org/series/IPUEN315990W200000000
10. U.S. Bureau of Labor Statistics. Sectoral Output — NAICS 315990 ($961.5M in 2025; $1.10B in 2022), via FRED. https://fred.stlouisfed.org/series/IPUEN315990T300000000
11. U.S. Bureau of Labor Statistics. Labor Share and Unit Labor Cost — NAICS 315990 (labor 36.2% of production cost, 2022; productivity index 77.7 → 73.2; ULC +12%), via FRED. https://fred.stlouisfed.org/data/IPUEN315990L030000000
12. Defense Logistics Agency. Clothing and Textiles Small Business Report, FY2024 (NAICS 315210 obligations $624.3M; NAICS 315250 obligations $101.0M). 2025. https://www.dla.mil/Portals/104/Documents/SmallBusiness/DB%20Flyers/2025%20Flyers/CandT%20SB%20-%20V-11-14-2025.pdf
13. U.S. International Trade Administration / Congressional Research Service (IF13001). The Berry Amendment (10 U.S.C. §4862) — domestic sourcing of clothing, fiber, yarn, footwear; ~$2B annual DoD textile/clothing procurement. 2024. https://www.trade.gov/berry-amendment
14. U.S. Department of Commerce (trade.gov) / U.S. Trade Representative. CAFTA-DR and USMCA textile "yarn-forward" rules of origin. 2024. https://www.trade.gov/summary-cafta-fta-textiles
15. Congressional Research Service (R48380). Imports and the Section 321 (De Minimis) Exemption — elimination for Chinese-origin goods, May 2025. 2025. https://www.congress.gov/crs-product/R48380
16. Forbes / CNN Business. Trump Suspends De Minimis Tariff Exemption for Goods Under $800 (2025 suspension; full repeal pending 2027). 2025. https://www.forbes.com/sites/zacharyfolk/2025/07/30/trump-suspends-de-minimis-tariff-exemption-for-goods-under-800/
17. U.S. Customs and Border Protection. De Minimis Guidance (suspension effective August 2025). 2025. https://www.help.cbp.gov/s/article/Article-1050
18. Remake / Epstein Becker Green (Wage & Hour Blog). The Garment Worker Protection Act (California SB 62) — piece-rate ban, joint brand liability, $200 per-employee per-pay-period penalty. 2021–2022. https://remake.world/stories/faq-the-garment-worker-protection-act-sb62/
19. U.S. Department of Labor. DOL Recovers More Than $892,000 for 296 Garment Workers in Southern California (50-contractor survey, 80% violation rate). 2024. https://www.dol.gov/newsroom/releases/whd/whd20240103-1
20. U.S. Federal Trade Commission (Textile Fiber Products Identification Act, Care Labeling Rule); U.S. Consumer Product Safety Commission (Flammable Fabrics Act, CPSIA). 2024. https://www.cpsc.gov/Business--Manufacturing/Business-Education/Business-Guidance/Flammable-Fabrics-Act
21. U.S. Customs and Border Protection / DHS. Uyghur Forced Labor Prevention Act (UFLPA) Enforcement Statistics and 2025 Strategy Update (6,636 shipments detained in H1 2025; entity list expanded to 144). 2025. https://www.cbp.gov/trade/forced-labor/UFLPA
22. White & Case LLP. United States Finalizes Section 301 Tariff Increases on Imports from China (~7.5–25% added duties). 2024. https://www.whitecase.com/insight-alert/united-states-finalizes-section-301-tariff-increases-imports-china
23. CalRecycle. Textile Extended Producer Responsibility Program (regulations effective no earlier than July 2028). 2026. https://calrecycle.ca.gov/epr/textiles/
24. U.S. Environmental Protection Agency. Textile Mills Effluent Guidelines — pollutant regulations and PFAS study. 2024. https://www.epa.gov/eg/textile-mills-effluent-guidelines
25. Observatory of Economic Complexity (OEC) / IndexBox. Knit Socks and Hosiery (HS 6115) trade — U.S. imports ~$2.63B in 2024. 2024. https://oec.world/en/profile/hs/knit-socks-and-hosiery
26. IBISWorld. Cut & Sew Apparel Contractors in the US — Industry Analysis (non-employer undercount). 2026. https://www.ibisworld.com/united-states/industry/cut-sew-apparel-contractors/339/
27. Apparelist / GlobeNewswire / SEC. Gildan Completes Acquisition of HanesBrands (closed December 1, 2025; ~$2.2B equity / ~$4.4B enterprise value; ~$200M synergy target). 2025. https://www.apparelist.com/2025/12/02/gildan-announces-completion-of-hanesbrands-acquisition/
28. Finviz / Yahoo Finance / SEC (Form 6-K). Gildan Activewear FY2025 results — net sales ~$3.62B, adjusted operating margin ~21.5%. 2026. https://finance.yahoo.com/news/gildan-activewear-gil-achieves-record-072420754.html
29. Fruit of the Loom / Berkshire Hathaway. Fruit of the Loom company profile (89% self-manufactured); Berkshire Hathaway Form 10-K (Honduras cloth manufacturing). 2024–2025. https://www.fotlinc.com/our-company/
30. Berkshire Hathaway Inc. Form 10-K (Fechheimer ownership — public-safety, postal and military uniforms). 2024. https://www.sec.gov/Archives/edgar/data/1067983/000095017024019719/brka-20231231.htm
31. Ralph Lauren Corporation. SEC Form 10-K (Fiscal 2025) — 96% of product value sourced outside the U.S. 2025. https://www.sec.gov/Archives/edgar/data/1037038/000103703825000011/rl-20250329.htm
32. G-III Apparel Group, Ltd. Fiscal 2025 Form 10-K (~76% of product sourced from Vietnam, China, Indonesia; no long-term manufacturing contracts). 2025. https://www.sec.gov/Archives/edgar/data/821002/000155837025003540/giii-20250131x10k.htm
33. Levi Strauss & Co. Fiscal 2025 Form 10-K (sourcing from manufacturers in ~32 countries). 2026. https://www.sec.gov/Archives/edgar/data/94845/000009484526000008/lvis-20251130.htm
34. ASI / PPAI. Delta Apparel Files for Chapter 11 Bankruptcy; Plans to Sell Salt Life Brand (~$38.7M to Iconix). 2024. https://www.ppai.org/media-hub/delta-apparel-files-for-chapter-11-bankruptcy-plans-to-sell-salt-life-brand/
35. Yahoo Finance / Bloomberg Law. Delta Apparel Files for Chapter 11 Bankruptcy (June 30, 2024); Salt Life sold for $28M. 2024. https://finance.yahoo.com/news/delta-apparel-files-bankruptcy-warns-203240207.html
36. PR Newswire / National Sporting Goods Association. New Era Completes Acquisition of '47. 2024. https://www.prnewswire.com/news-releases/new-era-completes-acquisition-of-47-302218138.html
37. PitchBook. Randa Apparel & Accessories — Company Profile (men's-accessory licensing platform). 2026. https://pitchbook.com/profiles/company/10044-55
38. PR Newswire / ZoomInfo. Renfro Brands acquired by The Renco Group (2021); ~$540M revenue legwear maker. 2021. https://www.prnewswire.com/news-releases/worlds-most-renowned-sock-manufacturing-company-renfro-brands-acquired-by-the-renco-group-301308934.html
39. California Employment Development Department. QCEW Industry Data, NAICS 315210 (~1,332 establishments; ~10,800 employees). 2022. https://labormarketinfo.edd.ca.gov/qcew/CEW-Detail_NAICS.asp
40. U.S. Bureau of Labor Statistics / Census Bureau. NAICS Revision Attachment — 2017 Benchmark Data for 315210 (2,921 establishments in 2017). 2020. https://downloads.regulations.gov/USBC-2020-0004-0028/attachment_1.pdf
41. USDA Economic Research Service. Cotton and Wool Market Outlook (July 2026; world production 117.3M bales, mill use 122.0M, ending stocks −6%). https://www.ers.usda.gov/topics/crops/cotton-and-wool/market-outlook
42. U.S. Bureau of Labor Statistics. Consumer Expenditures — 2024 ($2,001 average household apparel-and-services spending, −2.0%). 2025. https://www.bls.gov/news.release/cesan.nr0.htm
43. Sheng Lu (FASH455, University of Delaware) / U.S. Fashion Industry Association. State of U.S. Textile and Apparel Manufacturing, Employment and Trade (U.S. sewing labor ~$15–20/hour vs under $1/hour in Bangladesh). April 2025. https://shenglufashion.com/2025/04/07/state-of-u-s-textile-and-apparel-manufacturing-employment-and-trade-updated-april-2025/
44. Child primers: 3151 — Apparel Knitting Mills, 3152 — Cut and Sew Apparel Manufacturing, and 3159 — Apparel Accessories and Other Apparel Manufacturing, §§2–10 — source of the child-level structure and receipt splits, the branded-company sourcing disclosures (VF ~273 facilities / ~30 countries; Kontoor ~30% internal / ~210 contract factories in 19 countries), the uniform-service manufacturing figures (Cintas five plants; UniFirst ~62% of garments), Kontoor's 2024 margins (44.5% gross / 13.1% operating), knitting payroll at ~26% of sales, CMT/SMV mechanics and the ~10–15% contractor margin, the ~45,000-worker Los Angeles cluster estimate, the New Era five-league licence position, the Berry-compliant and "Made in USA" private supplier names, Digital Brands Group's Los Angeles production, the scope caveat on the DOL garment-contractor survey, and the reshoring-as-portfolio-tool framing.