Engineered Wood Member Manufacturing (U.S.) — NAICS 321215

An investor's primer. Figures marked to U.S. federal statistics where available; forward-looking statements are framed as judgments, not facts.

1. Overview

Engineered wood members are the load-bearing skeleton of a modern building, made not from a single sawn tree but from wood that has been cut into veneers, strands, or small boards and then glued back together into stronger, straighter, more predictable structural pieces. The category covers I-joists (the H-shaped beams that carry most new floors), laminated veneer lumber (LVL) and other structural composite lumber (SCL) used for beams and headers, glued laminated timber (glulam) arches and beams, and prefabricated wood roof and floor trusses. These products let builders span longer distances, waste less wood, and frame faster than with ordinary dimensional lumber.[1][2]

This is a mid-sized U.S. manufacturing industry — roughly $19.8 billion in annual shipments and about 47,000 workers — and a deeply cyclical one. Its fortunes rise and fall with new-home construction, so an investor's first question is always where we are in the housing cycle.[3][4]

Ways in differ sharply by product. The capital-intensive engineered-lumber side (LVL, I-joists, glulam) is dominated by a handful of large, mostly public forest-products companies, so public-market investors can get exposure through their shares. The truss-and-component side is a fragmented world of hundreds of small, largely private local fabricators — a classic small-business and private-equity hunting ground rather than a stock-market one. Both sit under the same NAICS code, and understanding that split is the key to the whole industry.

2. What it is and how it's structured

NAICS (North American Industry Classification System) code 321215 covers establishments that make fabricated or laminated wood arches, wood roof and floor trusses, and other fabricated or laminated wood structural members.[1] In the 2022 NAICS revision the Census Bureau merged two older codes — 321213 (Engineered Wood Member except Truss) and 321214 (Truss Manufacturing) — into this single code, partly because establishments frequently made products from both categories; historical data before 2022 is split across those two lines.[2][5]

What's in scope:

• I-joists (LVL or solid-sawn flanges with an OSB or plywood web)
• Laminated veneer lumber (LVL), parallel strand lumber (PSL), and laminated strand lumber (LSL) — collectively structural composite lumber (SCL)
• Glued laminated timber (glulam) beams and arches
• Finger-jointed structural lumber
• Prefabricated roof and floor trusses[1]

What it excludes (adjacent NAICS codes):

• 321211 / 321212 — Hardwood and Softwood Veneer and Plywood Manufacturing (the veneer and plywood feedstock)
• 321219 — Reconstituted Wood Product Manufacturing: oriented strand board (OSB), particleboard, medium-density fiberboard (MDF), and hardboard. OSB is a huge, separate industry even though it feeds I-joist webs.
• 321113 — Sawmills (dimensional lumber)
• 321918 — Other Millwork
• 321992 — Prefabricated Wood Building Manufacturing (prefab buildings or building sections)[1]

A note on newer mass-timber products: glulam sits squarely in this code, but cross-laminated timber (CLT) — large structural panels made by gluing layers of boards at right angles — is a younger product whose Census classification is less settled (classification depends on the establishment's primary activity), and many mass-timber plants make both glulam and CLT side by side. Treat "mass timber" as overlapping this code rather than fully inside it.[6]

Ownership mix. The industry is really two businesses under one roof. Engineered lumber (LVL, I-joists, glulam) is made in a small number of large, capital-heavy plants, typically owned by integrated forest-products firms that also control timber and veneer — a concentrated, mostly public group. Trusses and components are made to order in hundreds of small, freight-limited local shops, many family- or builder-owned. The federal data below blends the two.

Operating models differ sharply. Industrial engineered-wood-product mills make standardized, branded structural products in capital-intensive plants. LVL production generally entails peeling logs into veneers, drying and grading them, applying waterproof structural adhesive, laying the veneers parallel, curing them in a heated press, and cutting the resulting billet. I-joists combine OSB webs with LVL or solid-lumber flanges. Glulam bonds graded lumber laminations into straight or curved structural members. These processes reward mill scale, fiber integration, press utilization, technical certification, and nationwide distribution.[7][8]

Truss plants, by contrast, are closer to local design-and-fabrication businesses. A plant receives building plans, calculates loads under the applicable local code, optimizes lumber cuts and connector-plate placement, cuts dimension lumber, lays the members out on fixtures, presses metal plates into the joints, inspects the assembly, and delivers a bulky, job-specific product. Finished trusses have an economical freight radius (~150 miles) and generally cannot be redirected easily to another job, so responsiveness, design capability, builder relationships, and delivery execution matter alongside manufacturing efficiency.[9][10]

3. How big it is

Our ground-truth federal figures:

Those payroll and employment figures imply average pay of roughly $59,000 per worker, and the shipments and firm counts imply average sales of about $28 million per firm — but that average hides enormous dispersion between a billion-dollar LVL producer and a 20-person truss shop.[3][4]

Undercount caveat. Federal business statistics here are reasonably complete on the manufacturing side, but they understate the true economic footprint of trusses and components in two ways. First, a large share of trusses and wall panels is fabricated inside builder- and dealer-owned component plants whose parent companies are classified under construction or building-materials distribution codes, not 321215 — so their truss output lands elsewhere in the statistics. Second, the count of very small, single-location fabricators is inherently hard to capture. The $19.8 billion shipments figure should be read as the manufacturing-classified core of a larger structural-components economy.

Historical comparability caveat. The 2022 NAICS merger makes pre- and post-merger revenue, employment, and concentration series non-comparable unless both predecessor codes are reconstructed. On the predecessor boundaries, 2017 data reported $1.99 billion of shipments in non-truss engineered members and $6.61 billion in trusses — roughly $8.6 billion combined.[5]

Private market-research estimates size individual product niches — for example, the global I-joist market at roughly $1.5 billion in 2025 and the global CLT market near $1.3 billion in 2024 — but these are third-party estimates on different scopes and geographies; prefer the federal U.S. figures above for the industry's true size.[12][13]

4. The investable universe

There is no pure-play public "engineered wood member" company. Engineered wood is a segment inside larger forest-products or building-products companies, so public investors buy diluted exposure. The largest engineered-lumber and mass-timber producers are private.

Public companies with meaningful exposure:

Major private / other owners: Roseburg Forest Products (RFPI I-joists, RigidLam LVL), Pacific Woodtech (I-joists, LVL, glulam — owned by Japan's DAIKEN), RedBuilt (commercial trusses, I-joists, glulam), Freres Engineered Wood (Mass Ply Panel), SmartLam (CLT, Montana and Alabama), Nordic Structures (I-joists, glulam, CLT), 84 Lumber (component plants and EWP centers), US LBM (truss operations, jointly owned by Bain Capital and Platinum Equity), plus hundreds of independent regional truss fabricators.[21][23][24][25]

5. How the money works

Owners make money in two very different ways, and the drivers differ by product.

Engineered lumber (LVL, I-joists, glulam) — a capacity-and-spread business. These are continuous-process plants with high fixed costs. Profitability is driven by:

• Capacity utilization / operating rate — running the mill full spreads fixed costs; running it half-empty destroys margins.
• Price-cost spread — the gap between selling price and input costs: wood veneer and fiber, resins/adhesives (petrochemical-linked), energy, and labor. Prices for branded engineered lumber are stickier than the wild swings of random-length dimensional lumber, but they still fall in downturns.
• Volume, which tracks housing starts almost one-for-one.

Margins expand sharply when housing is strong and plants run flat-out, and compress fast when starts drop. Boise Cascade's experience illustrates the volatility: Wood Products segment operating income fell from $337 million (2023) to $232 million (2024) to just $5.8 million (2025) as EWP prices and volumes weakened. LVL production declined from 29.1 million cubic feet (2024) to 27.2 million (2025); I-joist production fell from 234 million to 215 million equivalent lineal feet.[14][15] Products are complementary and sold as systems: I-joists use LVL flanges, so makers cross-sell a whole "framing package."

Cost structure (indicative, from Boise Cascade). In 2025, wood fiber represented approximately 37% of Boise's Wood Products materials, labor, and other operating expense (excluding depreciation), with logs accounting for ~80% of that fiber cost. Purchased OSB (chiefly for I-joist webs) was about 5% and resins/glues about 6% of the same expense base. These are company-specific figures for a segment that also makes plywood — not industry-wide shares — but they indicate the cost drivers.[14]

Trusses and components — a made-to-order, local, pass-through business. Trusses are engineered and built to each project, shipped only a limited radius (freight kills the economics beyond ~150 miles). Capital needs are lower; the value added is design/engineering plus fabrication markup over the lumber cost, which is largely passed through. Margins are thinner and depend on local permit volume, labor, backlog, and working capital. Each order includes design, material optimization, engineering documentation, fabrication, and timed delivery. Nevertheless, local builder concentration can confer buyer power, and an incorrectly designed or late truss package can impose costs far exceeding the manufacturer's invoice — the business carries engineering and execution risk alongside ordinary manufacturing risk. This is why the segment is fragmented and local.

Distribution runs through pro dealers and two-step distributors, or direct to large national builders. Vertically integrated players capture margin at more than one stage — Boise Cascade makes and distributes (routing ~75% of its EWP volume through its own distribution segment in 2025), while Builders FirstSource builds and sells.[14][19]

6. What drives demand

• New residential construction is the number-one driver. Floors and roofs are where these products go: I-joists and LVL dominate new floor systems; trusses frame most new roofs. U.S. single-family starts totaled about 943,000 in 2025, down 6.9% year over year, while multifamily starts rose 17.4% — a worse mix for residential I-joists and trusses than a single-family-led expansion. Total 2025 starts were 1.36 million, down 0.6% from 2024.[26]
• Repair and remodeling (R&R). A steadier, growing tailwind; Harvard's Joint Center for Housing Studies projects homeowner improvement spending near $520 billion in 2026.[27]
• Substitution away from dimensional lumber, steel, and concrete. Engineered members span farther, are more dimensionally stable, and waste less fiber — a structural share-gain story, strongest when framing labor is scarce and prefabrication (trusses, panels) speeds job sites.
• Mass timber and decarbonization. Wood stores carbon and can replace steel/concrete in mid-rise buildings; this is a real growth vector off a small base. USDA reports that U.S. mass-timber construction used 179 million board feet of lumber in 2024, with another 48 million board feet in hybrid light-frame/mass-timber projects.[28]
• Interest and mortgage rates. Rates gate housing affordability and starts; elevated rates have dampened engineered-wood demand in recent years.[13]
• Nonresidential/light commercial construction adds truss, glulam, and mass-timber demand.

Countervailing trends: Boise identifies expanded floor-truss capacity, slab-on-grade construction (which eliminates many floor-system members), and firefighter-safety provisions affecting exposed I-joists as potential adverse factors.[14]

7. Regulation

• Building codes and product standards. Engineered wood members must meet consensus standards and carry code-recognized ratings before they can be specified: ASTM D5055 for I-joists, ANSI A190.1 for glulam, ASTM D5456 for structural composite lumber, and ANSI/APA PRG 320 for CLT. APA – The Engineered Wood Association provides much of the third-party certification and grade-stamping, and products carry evaluation reports (e.g., ICC-ES, the International Code Council Evaluation Service) for use under the International Building Code (IBC) and International Residential Code (IRC).[29]
• Tall mass-timber code changes. The 2021 and 2024 IBC created three new construction types — Type IV-A, IV-B, IV-C — permitting mass-timber buildings up to 18, 12, and 9 stories respectively, with the 2024 edition allowing more exposed timber. Dozens of states have adopted these provisions, unlocking a market that was effectively closed to wood before.[30]
• Formaldehyde / adhesives. EPA's TSCA (Toxic Substances Control Act) Title VI (40 CFR Part 770) caps formaldehyde emissions from composite wood — chiefly hardwood plywood, particleboard, and MDF. Importantly, structural plywood, structural panels, structural composite lumber, prefabricated wood I-joists, and finger-jointed lumber are generally exempt from TSCA Title VI when they meet specified standards. Structural engineered wood typically uses phenol-formaldehyde or no-added-formaldehyde resins. That exemption does not eliminate other air-emissions, worker-safety, building-code, or product-certification obligations.[31][32]
• Trade. Canadian softwood lumber — an input and a competing product — carries antidumping and countervailing duties that rose to roughly 35% in 2025, plus a new Section 232 tariff (+10%) on softwood timber and lumber effective October 2025. These raise input costs but can also shift demand toward domestic producers.[33]
• Workplace safety. Wood dust presents respiratory, fire, and explosion hazards, requiring collection, ventilation, housekeeping, and ignition control. OSHA provides specific guidance on wood-dust hazards.[34] Standard manufacturing oversight also applies: air permits for adhesive/dryer emissions, and sustainable-sourcing certification (FSC, SFI).

8. Competitive dynamics and consolidation

By the federal concentration data, the overall industry is unconcentrated: a Herfindahl-Hirschman Index (HHI) of just 340.8, with the top 4 firms holding 31.8% of receipts, the top 8 45%, the top 20 57.2%, and the top 50 70%.[3] But that low HHI is an artifact of blending two structures. Trusses are highly fragmented, which pulls the index down; the engineered-lumber sub-segment (LVL, I-joists, glulam) is effectively an oligopoly. Boise Cascade alone claims roughly 40% of North American LVL production and 37% of I-joist production — a level of concentration far above what the combined-code HHI suggests.[15]

Barriers to entry split the same way. Building an LVL or I-joist mill takes major capital and, ideally, captive veneer and timber — high barriers, few new entrants. Truss fabrication needs a saw, jigs, a design office, and a yard — low barriers, constant local entry and exit.

Consolidation runs in two directions: forest-products majors integrate upstream (timber → veneer → engineered lumber), while building-products distributors and homebuilders integrate downstream into component manufacturing (Builders FirstSource and UFP Industries rolling up regional truss shops). In mass timber, Mercer's ~$81 million purchase of Structurlam's assets in 2023 made it North America's largest producer.[18][19][20]

9. Risks

• Housing cyclicality and rate sensitivity — the dominant risk. Revenue and margins swing hard with starts and mortgage rates; a downturn spreads fixed costs over fewer units. The 2025 experience — Boise's segment income dropping from $232M to $6M on modest volume declines — shows how quickly margins compress.[14]
• Input-cost volatility — veneer and wood fiber, petrochemical-linked resins, energy, OSB (for I-joist webs), steel connector plates, and labor can compress the price-cost spread quickly. Western fiber can be constrained by public-land policy, seasonal access, wildfire, and long haul distances.
• Substitution both ways — when commodity lumber is cheap, some framers revert to dimensional lumber or open-web alternatives; engineered products must keep proving value. Boise specifically identifies floor-truss capacity expansion as a potential competitor to I-joists.[14]
• Overcapacity risk — announced 2025–27 capacity additions (e.g., Boise Cascade's Thorsby, AL I-joist and Southeast expansions) could pressure prices if they land into a soft housing market.[13][14]
• Trade and tariffs — softwood-lumber duties and Section 232 tariffs raise input costs and add policy uncertainty.[33]
• Mass-timber adoption slower than hyped — project economics remain challenging and the segment has seen high-profile stumbles (Structurlam's insolvency).[20]
• Labor and execution risk — local labor availability at truss plants; errors in loads, connector specification, manufacturing, or installation create product-liability and warranty exposure. Certification, in-plant quality control, and compliance with local building codes are part of the product.[10]
• Environmental and safety compliance — wood-dust hazards, air permits for adhesive/dryer emissions, fire/explosion risk.[34]

10. How to invest and the outlook

Public routes. There is no pure play, so choose the exposure profile you want:

• Boise Cascade (BCC) — the closest to a category bet, though roughly half its business is distribution.
• Weyerhaeuser (WY) — timber-REIT dividend plus engineered-wood exposure; more a land-and-cycle play.
• UFP Industries (UFPI) and Builders FirstSource (BLDR) — the components/trusses and downstream-manufacturing angle inside diversified building-products businesses.
• Mercer International (MERC) — optionality on mass timber within a pulp company.
• West Fraser (WFG) / Canfor — Canadian lumber majors with glulam and engineered-wood operations.
• Simpson Manufacturing (SSD) — picks-and-shovels exposure via connectors, plates, software, and automation equipment.

Tickers, dividends, and valuation multiples belong here, not in the industry description above; in all these names engineered wood is one segment, so model it as a slice, not the whole company.

Private routes. Because the largest engineered-lumber and mass-timber producers (Roseburg, Pacific Woodtech, RedBuilt, Freres, SmartLam, Nordic) are private, and because the truss world is hundreds of small shops, private investors have more direct paths: buying or backing regional truss/component fabricators (well within the SBA's 500-employee threshold), building-products private-equity roll-ups (US LBM's model), timberland and real-asset funds, or supplier/customer positions in the mass-timber supply chain.[21][23][25][11]

Diligence differs by sub-segment. Truss diligence should emphasize local housing permits, builder concentration, backlog quality, design liability, software and plate-supplier dependence, delivery radius, lumber-price pass-through, and the repeatability of production-builder plans. EWP-mill diligence should emphasize fiber security, press and dryer condition, resin systems, certification, utilization, product mix, channel relationships, maintenance capital, and the cost of downtime.

Outlook (forward-looking judgments, not guarantees). Near-term demand hinges on mortgage rates and single-family starts, which declined in 2025 and are expected to recover only slowly.[26] The structural case is more attractive than the cyclical one: a chronic U.S. housing shortage, ongoing substitution toward engineered products, prefabrication favored by scarce framing labor, and mass timber unlocked by expanding tall-wood building codes.[30] Tariffs on Canadian lumber cut both ways — higher input costs but a possible relative edge for domestic mills.[33] The clearest watch-item is timing: capacity additions arriving into a still-soft housing market could pressure prices before the next up-cycle rewards them. This remains a business to buy for the cycle, not in spite of it.

---

Sources

1. U.S. Census Bureau / NAICS Association, "NAICS Code 321215 — Engineered Wood Member Manufacturing (2022)," 2022. https://www.naics.com/naics-code-description/?v=2022&code=321215
2. Ask Kodiak, "321215 — Engineered Wood Member Manufacturing (NAICS 2022 change note: 321213 + 321214 collapsed)," 2022. https://naics.askkodiak.com/naics/2022/321215
3. U.S. Census Bureau, "2022 Economic Census — Concentration ratios and receipts, NAICS 321215" (receipts $19.80B; firms 715; CR4 31.8%, CR8 45%, CR20 57.2%, CR50 70%; HHI 340.8), 2022. https://www.census.gov/programs-surveys/economic-census.html
4. U.S. Census Bureau, "County Business Patterns 2023 — NAICS 321215" (establishments 983; employment 46,822; annual payroll $2.757B), 2023. https://www.census.gov/programs-surveys/cbp.html
5. BLS/Census, "NAICS 2022 revision submission — 321213/321214 merger rationale and 2017 baseline data," 2020. https://downloads.regulations.gov/USBC-2020-0004-0053/attachment_1.pdf
6. APA – The Engineered Wood Association, "ANSI/APA PRG 320: Standard for Performance-Rated Cross-Laminated Timber," 2019/2025. https://www.apawood.org/ansi-apa-prg-320
7. American Wood Council, "Wood Products — Floor / I-joist Girders," 2025. https://awc.org/resources/wood-products/floor/i-joist-girders/
8. American Wood Council, "I-joists," 2025. https://awc.org/resources/wood-products/i-joists/
9. Structural Building Components Association, "About Us," 2025. https://www.sbcacomponents.com/about-us
10. Structural Building Components Association, "SBCA In-Plant QC FAQs," 2025. https://www.sbcacomponents.com/sbca-in-plant-qc-faqs
11. U.S. Small Business Administration, "Table of Small Business Size Standards — NAICS 321215 (≤500 employees)," 2023. https://www.sba.gov/document/support-table-size-standards
12. Market Report Analytics, "Engineered Wood I-joist — Market Insights and Forecasts 2025–2033," 2025. https://www.marketreportanalytics.com/reports/engineered-wood-i-joist-56183
13. GlobeNewswire / Research and Markets, "North America Glue Laminated Timber Industry Report 2024–2032," 2024. https://www.globenewswire.com/fr/news-release/2024/05/10/2879460/28124/en/North-America-Glue-Laminated-Timber-Industry-Report-2024-2032-Non-corrosive-Eco-friendly-Durable-and-Lightweight-Characteristics-Spurs-Adoption.html
14. Boise Cascade Company, "Form 10-K FY2025" (Wood Products sales $1.61B; segment operating income $5.8M; cost structure; EWP production volumes; distribution share), 2026. https://www.sec.gov/Archives/edgar/data/1328581/000132858126000006/bcc-20251231.htm
15. Boise Cascade Company, "Investor Presentation June 2025" (40% LVL market share, 37% I-joist market share estimates), 2025. https://www.sec.gov/Archives/edgar/data/1328581/000132858125000082/bcc-investordeck3.htm
16. Boise Cascade, "Engineered Wood Products (EWP)" (LVL/I-joist plants, Alexandria LA and White City OR). https://www.bc.com/ewp/
17. Weyerhaeuser Company, "Form 10-K FY2025" (engineered solid section $649M; I-joists $343M; production volumes), 2026. https://www.sec.gov/Archives/edgar/data/106535/000119312526051422/wy-20251231.htm
18. UFP Industries, "Fourth Quarter and Fiscal 2024 Results" (net sales $6.7B; Construction segment trusses/components), 2025. https://ufpi.com/ufp-industries-announces-fourth-quarter-and-fiscal-2024-results/
19. Builders FirstSource, "Form 10-K FY2025" (manufactured products including trusses, wall panels, engineered wood), 2026. https://www.sec.gov/Archives/edgar/data/1316835/000119312526054643/bldr-20251231.htm
20. Mercer International / Mercer Mass Timber, "Mercer Mass Timber Successfully Acquires Structurlam" (~$81.1M; largest N. American mass-timber producer), 2023. https://www.businesswire.com/news/home/20230615483120/en/Mercer-Mass-Timber-Successfully-Acquires-Structurlam-Expanding-Operations-and-Strengthening-Market-Presence
21. Anthony Forest Products (a Canfor company), "Engineered Wood Products" (largest SYP glulam producer); Roseburg, "Engineered Wood." https://www.anthonyforest.com/ewp/; https://www.roseburg.com/engineered-wood/
22. Simpson Manufacturing Company, "Form 10-K FY2025" (Wood Construction sales $1.97B), 2026. https://www.sec.gov/Archives/edgar/data/920371/000162828026012920/ssd-20251231.htm
23. RedBuilt, "Engineered Wood Products for Commercial Buildings"; Pacific Woodtech via Snavely Forest Products. https://www.redbuilt.com/; https://snavelyforestproducts.com/product-detail/pacific-woodtech/
24. DAIKEN Corporation, "Group Report 2025" (Pacific Woodtech ownership), 2025. https://www.daiken.jp/about/groupreport/pdf/annual/2025/all_daiken_report2025e_w.pdf
25. US LBM, "US LBM Announces Joint Ownership Agreement with Bain Capital Private Equity and Platinum Equity," 2024. https://news.uslbm.com/us-lbm-announces-joint-ownership-agreement-with-bain-capital-private-equity-and-platinum-equity/
26. National Association of Home Builders / Census/HUD, "Overall Housing Starts Inch Lower in 2025" (total 1.36M; single-family 943K, −6.9%; multifamily +17.4%), 2026. https://www.nahb.org/news-and-economics/press-releases/2026/02/overall-housing-starts-inch-lower-in-2025
27. Builders FirstSource / Harvard JCHS, "Housing Market Going into 2025: Remodeling" (2026 improvement spending ~$520B), 2025. https://www.bldr.com/resources/blog/housing-market-early-2025-home-sales-rentals-and-remodeling
28. USDA Forest Products Laboratory, "Mass Timber Construction" (179M board feet in 2024; 48M in hybrid projects), 2026. https://research.fs.usda.gov/sites/default/files/2026-01/fpl_tour_info_masstimber_accessible.pdf
29. APA – The Engineered Wood Association, "Product standards — ASTM D5055 (I-joists), ANSI A190.1 (glulam), ASTM D5456 (SCL), ANSI/APA PRG 320 (CLT)," 2018–2025. https://www.apawood.org/ansi-apa-prg-320
30. WoodWorks – Wood Products Council / Construction Executive, "Tall Mass Timber under the 2021 & 2024 IBC (Type IV-A/B/C, up to 18 stories)," 2021–2024. https://www.woodworks.org/learn/mass-timber-clt/tall-mass-timber/; https://constructionexec.com/article/tall-mass-timber-buildings-now-possible-under-2021-ibc-code-changes/
31. U.S. Environmental Protection Agency, "Formaldehyde Emission Standards for Composite Wood Products (TSCA Title VI, 40 CFR Part 770)," 2016/2024. https://www.epa.gov/formaldehyde/formaldehyde-emission-standards-composite-wood-products
32. U.S. Environmental Protection Agency, "Frequent Questions for Regulated Stakeholders About Implementing Formaldehyde Standards" (structural product exemptions), 2024. https://www.epa.gov/formaldehyde/frequent-questions-regulated-stakeholders-about-implementing-formaldehyde-standards
33. CBC News / Congressional Research Service, "U.S. duties on Canadian softwood lumber (~35%) and Section 232 softwood tariff (+10%, Oct 2025)," 2025. https://www.cbc.ca/news/canada/british-columbia/canada-softwood-penalties-1.7604876; https://www.congress.gov/crs-product/R48781
34. U.S. Occupational Safety and Health Administration, "Wood Dust," 2025. https://www.osha.gov/wood-dust