Specialty Trade Contractors (U.S.) — NAICS 238

A Histometrics rollup primer for a general investing audience — relevant to both public-market and private investors. "NAICS" is the North American Industry Classification System, the standard the U.S. government uses to group businesses by activity. This is a three-digit subsector — the layer above the four-digit industry groups, and two above the leaf-level national industries. It sits inside NAICS sector 23, Construction, alongside two sibling subsectors: general building construction (236 — the homebuilders and general contractors) and heavy & civil engineering (237 — highways, bridges, pipelines, power lines). This page rolls up four industry groups — 2381, 2382, 2383, and 2389 — each of which is itself a rollup of trades. Its distinctive value is the contrast across those four: how they differ in size, growth, ownership, and how you can invest. For any single trade, read the child primers.

---

1. Overview

Specialty trade contractors are the subcontractors of the construction economy — the crews a general contractor (GC), developer, or homebuilder hires to perform one part of a building rather than manage the whole thing. If sector 23 is "construction," this subsector is the labor that actually builds: the excavators who clear and grade the land, the concrete and steel crews who raise the shell, the electricians, plumbers, and mechanical trades who wire and pipe the inside, and the drywall, paint, and flooring crews who finish it. Almost all of their work is bid from a GC or builder; a large slice of repair, remodel, and replacement (R&R) work is sold straight to property owners.[1]

Taken together this is one of the largest slices of the U.S. economy that most investors have never been able to buy directly. On this level's own federal figures it is roughly $1.28 trillion of annual receipts, about 506,000 firms, and nearly 5 million workers — the largest of construction's three subsectors by employment.[2][3] Three facts define it for an investor:

• It is overwhelmingly private and almost perfectly fragmented. The four largest firms in the entire subsector hold just 1.8% of revenue.[3] There is no dominant national player and, across most of the trades, no clean publicly traded pure-play at all. Public-market investors reach the theme mostly indirectly — through the material makers, distributors, diversified builders, and equipment lessors that surround the trades.
• It is labor. The value is skilled tradespeople putting work in place, and a persistent shortage of them — electricians, plumbers, welders, carpenters, equipment operators — is the binding constraint on the whole subsector.[9]
• The four children are not one business. They ride different demand engines, run at different sizes, and offer different ways in. The real payoff of looking at 238 as a group is seeing that contrast — so that is where this primer leads.

---

2. What's inside — the four child industry groups, and how they differ

NAICS nests: this three-digit subsector splits into four four-digit industry groups, each of which splits again into trades. The four are not competitors — an excavator does not bid against an electrician — so 238 is really four distinct labor markets stacked under one heading. What unites them is a common economic DNA: project-based, bid-driven, labor-intensive, thin-margin subcontracting. What divides them is where they sit in the building's life, how big the typical firm is, and how (if at all) public capital can touch them.

A useful way to hold the four in mind is the order they arrive on a jobsite — dirt first, finishes last:

Acronyms: PE = private equity; ESOP = employee stock ownership plan; HVAC = heating, ventilation, and air-conditioning; R&R = repair, remodel, and replacement; E&C = engineering and construction.

Four contrasts worth carrying forward:

1. One child is nearly half the subsector. Building Equipment (2382) alone is ~46% of receipts and ~48% of the workforce — bigger than the other three combined on employment. It is also the highest-paid (about $76,000 per worker versus a subsector average near $71,600) and has the biggest average firm, because licensed mechanical work is skilled and credential-gated. If you understand only 2382, you understand the plurality of the subsector.

2. They run on different clocks. The chronological arc is also a cyclical arc. Site prep (2389) and the structural trades (2381) are earliest-cycle — first to boom, first to be cut when a project slips. The inside systems (2382) and finishing (2383) come later and lean more on the defensive replacement and service work that keeps going regardless of the economy. The subsector as a whole is therefore steadier than any one child.

3. The one genuine oligopoly is buried inside the most fragmented average. Everything here is fragmented — but the elevator business inside 2382 is a global oligopoly (a handful of makers) sitting on a sticky, decades-long service annuity. Aggregating four non-competing groups dilutes that concentration to near-invisibility in the headline numbers (§3). Read the low concentration as "competition is fought locally, trade by trade," not "the whole field is uniformly open."

4. The public door is barely ajar — and only in two places. Across a $1.28-trillion subsector, there are essentially two clean listed pure-plays: elevators (inside 2382) and insulation (inside 2383). Everywhere else, listed exposure is one step removed. This is the single most important fact for a public-market investor, and it is covered in §4.

---

3. Size — this level's rollup figures

These are our ingested ground-truth federal statistics for NAICS 238 as a whole. They blend the 2023 County Business Patterns (CBP, the Census Bureau's annual count of employer establishments) with the 2022 Economic Census (its every-five-years benchmark for firms, receipts, and concentration), so this is a size profile across two reference years, not a single comparable annual series.[2][3]

A few things fall straight out. The average firm is small — about $2.5 million of receipts and roughly ten employees per establishment — and average pay runs near $71,600 per worker per year.[2][3] The subsector is seasonal: first-quarter payroll ($81.1 billion) is only ~22.7% of the annual total, below a flat 25% quarter — the fingerprint of trades that slow in winter (you cannot pour concrete, grade a site, or lay brick in a hard freeze; the indoor mechanical and finishing trades are steadier). And it is almost perfectly fragmented: the four largest firms hold just 1.8% of revenue, the top fifty only 7.1%, and the HHI — a 0–10,000 concentration index where anything under 1,500 is already "unconcentrated" — is 1.7, about as close to textbook perfect competition as any trillion-dollar corner of the economy gets.[3]

That subsector HHI of 1.7 is lower than every one of its four children (which range from 1.9 to 7). That is not a paradox: aggregating four largely non-competing groups dilutes any one firm's dominance across the whole. The biggest electrical contractor is a rounding error against plumbing, roofing, excavation, and drywall combined — and, crucially, the elevator oligopoly inside 2382 vanishes entirely into the blend. The low number confirms an owner-operator, private-market industry; it does not mean there are no concentrated pockets inside it.

Here is how the subsector breaks down by child, which makes the contrast quantitative. The children reconcile cleanly into the totals — establishments (521,315) and employees (4,986,759) sum exactly, and receipts and payroll to rounding; the firm count is a hair below the naive sum because a firm active in two groups is counted once here.[2][3]

Two structural differences stand out. Building Equipment (2382) is even more dominant in labor than in dollars — ~48% of the workforce on ~46% of receipts — the signature of a large, skilled, credential-heavy trade. And the smallest firms sit in Building Finishing (2383) (~$1.5 million average receipts, the most atomized painters, tilers, and carpenters), while 2382 and 2389 tie for the biggest average firm (~$3.0 million) for opposite reasons — 2382 because mechanical contracting scales, 2389 because moving earth requires owned fleets of heavy machines.

Undercount caveat — it matters a lot here. Every figure above counts only employer establishments with paid payroll. It excludes the self-employed, no-payroll sole proprietors, day-labor crews, and 1099 "independent" installers — and these trades are unusually skewed toward exactly those operators (owner-operator painters, carpenters, roofers, framers, single-machine excavators). The Census tabulates those "nonemployer" businesses separately, and our ground-truth dataset has no nonemployer count for 238, so we put no number on it and flag only the direction of the bias: the true population of operating businesses is a large multiple of the ~506,000 employer firms shown, and total economic activity exceeds the $1.28 trillion of employer receipts. On top of that, foundation, framing, or finishing work that homebuilders and GCs self-perform with their own crews is booked to them, not here. Read $1.28 trillion as the employer-core floor of a materially larger real economy, not a ceiling — the revenue totals are more reliable than the head-counts, and small/individual ownership dominates.

---

4. Investable universe — where value concentrates across the children

The single most important fact for a public-market investor is the same across the subsector: the trades themselves are private, and the listed exposure is almost always one step removed. Virtually every firm sits under the Small Business Administration's (SBA) size standard (roughly $19 million average annual receipts for most of these trades), so the industry qualifies almost entirely as small business.[5] Value therefore concentrates in three tiers an investor must treat differently — and only two spots in the whole subsector offer a clean listed pure-play. Tickers below are for orientation, not recommendations, and are held to this section and §10.

Tier 1 — the two clean pure-plays. Both are the exception that proves the rule, and both are anomalies within their child:

• Elevators (inside 2382). Otis Worldwide (NYSE: OTIS) is the liquid U.S. pure-play on the elevator "razor-and-blades" service annuity — the one place a near-pure listed contractor-cum-service business exists, precisely because elevators are an oligopoly rather than a fragmented trade. Foreign-listed peers KONE (Helsinki) and Schindler (Swiss) round it out.
• Insulation (inside 2383). Installed Building Products (NYSE: IBP) is the cleanest listed installer in the subsector — the #2 U.S. insulation contractor, ~94% of revenue from installation.

Tier 2 — diversified listed contractors that touch the trades as one line of business. These are partial handles — value the parent for what it actually is:

• Mechanical/electrical (2382): EMCOR (NYSE: EME), Comfort Systems USA (NYSE: FIX), IES Holdings (NASDAQ: IESC), MYR Group (NASDAQ: MYRG), Limbach (NASDAQ: LMB), Quanta (NYSE: PWR), MasTec (NYSE: MTZ), Dycom (NYSE: DY), APi Group (NYSE: APG), Chemed (NYSE: CHE, via Roto-Rooter).
• Site prep / structural (2389, 2381): Sterling Infrastructure (NASDAQ: STRL, the nearest thing to a listed site-development play), Granite Construction (NYSE: GVA), Construction Partners (NASDAQ: ROAD), Primoris (NYSE: PRIM), Tutor Perini (NYSE: TPC); plus Apogee (NASDAQ: APOG, glazing), INNOVATE (NYSE: VATE, steel erection), Concrete Pumping Holdings (NASDAQ: BBCP), Smith-Midland (NASDAQ: SMID, precast).
• The franchise proxy: FirstService (NASDAQ/TSX: FSV), which owns finishing-trade franchise brands (CertaPro Painters, California Closets) inside a larger property-services compounder.

Tier 3 — the supply chain and demand side ("picks and shovels"). Because the contractors are private, the deepest listed pool is the layer around them, with economics that differ sharply from thin-margin labor:

• Materials: aggregates/cement (Vulcan Materials VMC, Martin Marietta MLM, CRH plc CRH, Eagle Materials EXP); building products (Owens Corning OC, Carlisle CSL, Sherwin-Williams SHW, PPG, James Hardie JHX, Mohawk MHK); HVAC equipment (Carrier CARR, Trane TT, Lennox LII, A. O. Smith AOS).
• Distribution: Ferguson (NYSE: FERG), Watsco (NYSE: WSO), Builders FirstSource (NYSE: BLDR), Home Depot (NYSE: HD), Lowe's (NYSE: LOW), QXO (NYSE: QXO).
• Equipment makers & renters (sell into the whole fragmented base): Caterpillar, Deere, United Rentals, Herc; pool adjacents Pool Corporation (NASDAQ: POOL) and Latham (NASDAQ: SWIM).
• Demand-side proxy: homebuilders (D.R. Horton DHI, Lennar LEN, PulteGroup PHM, NVR). There is no pure specialty-trade ETF; the Global X U.S. Infrastructure Development ETF (NYSE Arca: PAVE) bundles several of the electrical, mechanical, and materials names.

The honest summary: public investors mostly buy the elevator annuity, the one insulation installer, the supply chain, and the demand side. The other ~$1.2 trillion of contractor value is a private-market opportunity.

---

5. How the money works

Across all four children the economic engine is the same: project-based, fixed-price (or unit-price) subcontracting. A contractor bids a lump sum — or a rate per cubic yard moved, per ton of steel erected, per square of roof, per square foot of drywall — then earns the spread between that price and its actual cost of materials, field labor, and overhead. Revenue is recognized over time as the job progresses. There is no recurring subscription, and in a field this fragmented, almost no pricing power. None of the specialized valuation frameworks used elsewhere in finance apply here — there is no regulated rate base, no store-count model, no reserves to deplete; the business is volume, labor productivity, and estimating accuracy.

Because the price is fixed at bid and much of the work is irreversible, the central operating risk is margin fade — labor overruns, weather delay, material inflation, or a mis-priced change order eating a job's profit. Two features shape returns in every child. First, it is a working-capital business: contractors buy materials and pay crews before the customer pays, and owners hold back retainage (commonly 5–10% of each bill) until completion, so cash flow lags accounting profit and mechanics-lien rights matter. Second, surety bonding capacity and a clean safety record effectively cap how large a project a firm can chase, which is why better-capitalized operators win the bigger, better jobs.

The one axis that genuinely differs across the four — and that drives valuation everywhere in the subsector — is the recurring-service share. The more a business leans toward maintenance, repair, and replacement rather than one-and-done new installation, the more defensive and highly valued it is:

• 2382 has the most of it — the elevator service flywheel (a machine sold once at low margin, then serviced under sticky contracts for 20–30 years; at the industry leader, service is ~60% of sales but >90% of operating profit), plus HVAC repair-and-replace and emergency electrical work.
• 2381 and 2383 have a defensive replacement base (re-roofing, re-siding, repaint, re-floor) even though the install work is one-and-done.
• 2389 is the most project-and-utilization-driven (an idle owned excavator costs more per day than it bills), with a pocket of recurring pool and industrial-maintenance service.

The right input-cost gauges are the Bureau of Labor Statistics (BLS) Producer Price Indexes for concrete, steel, copper, aluminum, lumber, and diesel — not the factory-capacity or same-store metrics used for manufacturers and retailers.

---

6. Demand drivers

Demand for the subsector is, at bottom, the volume of things being built and maintained — but the four children pull from different reservoirs, which is why the group is steadier than any one trade:

• Data centers and the AI buildout — the dominant current pull. The artificial-intelligence (AI) data-center wave is the fastest-growing construction segment, and it touches all four children in sequence: earthwork and pads (2389), concrete/structural steel/precast (2381), electrical power distribution and mechanical cooling (2382), and durable finishes (2383). Data-center power demand is forecast to climb steeply this decade, dragging the electrical and site-prep trades hardest.[7][8]
• Electrification and grid load growth. Rising electricity demand for the first time in a generation — EV charging, heat pumps, solar, storage, and grid modernization — is a structural, multi-year tailwind concentrated in 2382 (electrical).[8]
• Factory reshoring. Semiconductor and battery plants are massive site and structural jobs (2389, 2381, 2382), though the wave cooled from its 2023–24 peak.
• New residential construction. Foundations and framing (2381) and site work (2389) are the earliest and most rate-sensitive trades; single-family activity has been soft with mortgage rates near 6.5%.[6]
• Repair, remodel, and replacement — the defensive ballast. Re-roofing (~80% of roofing demand), re-siding, HVAC and elevator service, and interior remodeling ride an aging building stock and record homeowner-improvement spending, much of it non-discretionary and partly insurance-funded after storms. This is what keeps the subsector from moving as violently as new construction alone.[15][16]
• Regulator-forced turnover. The refrigerant transition (see §7) is forcing HVAC-equipment replacement; tightening energy codes pull insulation and glazing toward higher-performance work.[12]
• Public infrastructure. The 2021 Infrastructure Investment and Jobs Act (IIJA) supports site, structural, and utility work, with 2026 its final authorized year and a "funding cliff" risk after.[8]
• Financing conditions gate homebuying and debt-financed commercial development — which is why the new-build children are sharply cyclical and the replacement/service children are not.

Leading indicators worth watching: housing starts and permits, monthly construction spending, the American Institute of Architects' (AIA) Architecture Billings Index (below 50 signals contraction), contractor backlog, and Harvard's Leading Indicator of Remodeling Activity (LIRA) for the R&R side.[6][15]

---

7. Regulation

None of these trades is rate-regulated or entry-licensed at the federal level; regulation targets worker safety, building integrity, and the environment, and — importantly — the trade-by-trade licensing regime is itself a moat that keeps the subsector fragmented and local:

• Worker safety (Occupational Safety and Health Administration, OSHA, under 29 CFR Part 1926). The recurring hazards: respirable crystalline silica (concrete, masonry, drywall, tile, engineered-stone countertops); fall protection (roofing, framing, glazing, finishing); steel erection (Subpart R); and trenching/excavation (Subpart P, cave-in protection required at five feet).[10]
• Trade licensing and codes (state and local, but keyed to national standards). Electrical work must meet the National Electrical Code (NEC, published as NFPA 70); plumbing and mechanical trades are separately licensed; elevators follow ASME A17.1 (American Society of Mechanical Engineers) with mandatory periodic inspection; structural work references ACI 318 (concrete) and AISC steel standards. Each credential is a barrier that favors established firms.[10][12]
• Environmental (Environmental Protection Agency, EPA). Construction stormwater permitting (a permit is required for disturbing one acre or more); the Lead Renovation, Repair and Painting (RRP) rule on pre-1978 buildings; and the AIM Act refrigerant transition — Section 608 technician certification plus a global-warming-potential limit on new HVAC equipment that is forcing an equipment turnover.[11][12]
• Labor and wages. Davis-Bacon prevailing wages on covered federal work; state contractor licensing, bonding, and lien rules; and — acutely for this heavily foreign-born workforce — Form I-9/E-Verify employment-eligibility enforcement.[13]
• Trade policy. Section 232 steel tariffs (raised to 50% in 2025), plus cement, aluminum, and imported-tile duties, bear directly on input costs.[14]

The practical investment point: compliance depth — safety systems, bonding capacity, licenses, and certifications — is itself a competitive advantage that favors the larger, better-run operators.

---

8. Consolidation

NAICS 238 is one of the most fragmented parts of the U.S. economy (subsector CR4 of 1.8%, HHI of 1.7), which is precisely why it has become a magnet for consolidation.[3] Three forces are at work, running at different speeds across the four children:

• Private-equity (PE) roll-ups. Sponsors buy founder-owned local firms at mid-to-high-single-digit multiples of EBITDA (earnings before interest, taxes, depreciation, and amortization; specialty-trade deals have recently run roughly 6–8×) and centralize marketing, procurement, estimating, and bonding.[17] The pace varies sharply: electrical and HVAC (2382) and roofing (inside 2381) are furthest along (PE now drives the majority of deals in some of these trades); site prep, pools, and paving (2389) are accelerating; finishing (2383) has pockets (national insulation roll-ups, painting and closet franchises); and labor-heavy, seasonal trades like masonry and finish carpentry resist roll-up because scarce skilled labor, not capital, is the constraint.
• Consolidation already achieved at the top of one child. Elevators (inside 2382) are a standing oligopoly, and the headline capital-markets event of the cycle is a mega-merger there — KONE's ~$34 billion agreement to buy TK Elevator, which would create the world's largest elevator maker.[18]
• ESOPs and family successions. An aging owner base with thin succession pipelines makes ownership transfer a steady deal-supply engine; many of the largest concrete, steel, electrical, and mechanical firms are already employee-owned.

But two structural brakes apply everywhere. Heavy, hard-to-transport materials (wet concrete, structural steel, precast, moved earth) keep competition regional. And the real growth constraint is skilled labor, not demand or capital — a roll-up cannot conjure electricians. A separate force, prefabrication (factory trusses, wall panels, modular assemblies, precast, pre-finished millwork), is shifting value away from on-site labor and toward manufacturers, most visibly in framing and finishing. Roll-ups also carry genuine integration risk — estimating systems, union relationships, local licenses, and safety cultures do not merge as cleanly as financial statements.

---

9. Risks

The four children share one risk register — really the risk register of fixed-price construction subcontracting:

• Cyclicality and interest-rate sensitivity — heaviest on the new-build children (site prep, foundations, framing, structural steel), cushioned by the replacement and service children (roofing, siding, HVAC/elevator service, remodeling).
• Skilled-labor shortage — the binding constraint across the whole subsector, worsened by an aging workforce and immigration enforcement of a heavily foreign-born trade.[9]
• Fixed-price execution risk — one mis-estimated job on thin margins can erase profit, and much of the work is irreversible.
• Input-cost and tariff volatility — steel (50% Section 232 tariffs), copper, aluminum, cement, lumber, refrigerant, and diesel, hitting hardest on fixed bids without escalation clauses.[14]
• Thin margins and no pricing power in a near-perfectly-competitive market.
• Working-capital, retainage, and counterparty risk — contractors pay first and collect last; GC, developer, or insurer distress flows straight to the subcontractor.
• Data-center concentration / overbuild risk — the current boom leans on a handful of hyperscale buyers; a faster-than-expected cooling of AI capital spending would hit the fastest-growing trades (site prep, electrical, structural) first.
• Safety and warranty liability — silica, falls, trench and steel-erection hazards, and latent defects.[10]
• Substitution — factory prefabrication moving labor off-site, and cheaper materials taking share from field-laid trades.
• Consolidation / roll-up execution risk — integration failure and leverage against volatile cash flows.
• Policy cliffs — IIJA authorization ending after 2026, and expired residential energy-efficiency credits, are near-term headwinds for infrastructure- and retrofit-facing work.[8]
• Data and proxy risk for investors — employer-only federal statistics undercount the small-operator reality (§3), and nearly every listed proxy reflects the trades only partially (basis risk).

---

10. How to invest, and the outlook

Public-market investors should start by accepting that they cannot buy most of this subsector directly, then match the vehicle to the theme:

• For a clean recurring-service annuity, the elevator specialists (Otis) inside 2382 are the one near-pure listed contractor-cum-service business — high quality, but a good business can still be a poor investment at an excessive price.
• For electrical / mechanical growth (data centers, electrification), the diversified contractors (EME, FIX, IESC, MYRG, PWR, MTZ) — examined individually for segment mix, backlog quality, and data-center concentration — or the broad PAVE ETF.
• For the data-center / site-work theme, Sterling Infrastructure (STRL) and diversified E&C names (GVA, PRIM, TPC), plus aggregates and cement (VMC, MLM, CRH, EXP).
• For the defensive replacement theme, roofing/siding/insulation materials and distributors (OC, CSL, JHX, BLDR, QXO) and the one insulation installer (IBP).
• For housing-cycle exposure, the homebuilders (DHI, LEN, PHM, NVR) and HVAC distributors/makers (FERG, WSO, CARR, TT).

In every case, size the position to the actual exposure — a quarry, a shingle mill, an elevator maker, or a distributor is not a specialty contractor, and its margins, capital intensity, and pricing power differ.

Private investors — where the subsector genuinely lives — own or operate regional contractors, back or build PE roll-up platforms, buy into ESOP-style successions, purchase franchise territories, or provide equipment leasing and private credit against receivables and backlog. Small trades firms change hands at low-single-digit EBITDA multiples and fit SBA-backed acquisition financing. The best targets share the same fingerprints regardless of trade: disciplined job-cost and work-in-progress accounting, a clean safety record, adequate bonding, a trained crew in a labor-short market, low customer concentration, a meaningful recurring-service share, and a real succession plan. Because the subsector is so fragmented, sourcing and integration — not underlying demand — are the hard part, and the diligence priority is always job-level economics.

Outlook. A vast, fragmented, labor-scarce subsector that is two-speed into 2026–2027. The strongest pull is on the trades levered to data centers, electrification, and large commercial structural work — site prep, electrical and mechanical, concrete, and structural steel. The replacement- and service-driven work — elevator and HVAC service, re-roofing, re-siding, remodeling — offers the steadiest demand and the highest-quality economics, and consolidation should continue fastest where recurring revenue is richest. The soft spots are the homebuilding-levered trades (framing, single-family foundations, finishing) waiting on lower mortgage rates, and the trades facing substitution (prefabrication and cheaper claddings). Cutting across all of it: a deep skilled-labor shortage, volatile tariff-inflated input costs, and an accelerating PE-and-ESOP consolidation wave — all of which favor the larger, better-capitalized, more productive operators. For public investors the theme is best owned through the elevator annuity, the supply chain, and the demand side; for private investors, it remains a hands-on, owner-operator business that rewards discipline over passive capital. For the trade-by-trade detail, read the four child primers: 2381 Foundation, Structure & Building Exterior; 2382 Building Equipment; 2383 Building Finishing; 2389 Other Specialty Trade.

---

Sources

This is a rollup page; sources are consolidated from the four child primers (NAICS 2381, 2382, 2383, 2389). Federal figures for this level (§3) are our own ingested ground-truth statistics for NAICS 238.

1. U.S. Census Bureau, 2022 NAICS — Subsector 238 (Specialty Trade Contractors) and its four industry groups: definitions, scope, and cross-references. https://www.census.gov/naics/?input=238&year=2022
2. U.S. Census Bureau, County Business Patterns, 2023 — NAICS 238 (establishments, employment, annual and first-quarter payroll; employer-only coverage). https://www.census.gov/programs-surveys/cbp.html
3. U.S. Census Bureau, 2022 Economic Census — Concentration of Largest Firms, NAICS 238 (firms, receipts, CR4/CR8/CR20/CR50, HHI). https://data.census.gov/table/ECNSIZE2022.EC2200SIZECONCEN
4. U.S. Census Bureau, County Business Patterns methodology and Nonemployer Statistics (employer-only coverage; self-employed and no-payroll businesses excluded). https://www.census.gov/programs-surveys/nonemployer-statistics.html
5. U.S. Small Business Administration, Table of Small Business Size Standards (13 CFR 121.201) — NAICS 238 trades, ~$19 million average annual receipts. https://www.sba.gov/document/support-table-size-standards
6. U.S. Census Bureau, Monthly New Residential Construction and Monthly Construction Spending (2026) — starts, permits, and residential/nonresidential/public spending. https://www.census.gov/construction/nrc/current/index.html
7. Grid Strategies LLC, National Load Growth Report 2025; Construction Dive, Data center construction rolls into 2026. https://gridstrategiesllc.com/; https://www.constructiondive.com/
8. U.S. Department of Energy, Report Evaluating Increase in Electricity Demand from Data Centers (2024); Associated General Contractors of America, 2026 Construction Industry Outlook; American Institute of Architects, Consensus Construction Forecast (2026). https://www.energy.gov/; https://news.agc.org/economics/2026-construction-industry-outlook/
9. U.S. Bureau of Labor Statistics, Occupational Outlook Handbook — electricians, plumbers, HVAC mechanics, carpenters, and operating engineers (2025); Associated Builders and Contractors / National Association of Home Builders, 2026 construction workforce shortage analyses. https://www.bls.gov/ooh/construction-and-extraction/
10. U.S. Occupational Safety and Health Administration, Safety and Health Regulations for Construction, 29 CFR Part 1926 — silica (1926.1153), fall protection (1926.501), steel erection (Subpart R), trenching/excavation (Subpart P). https://www.osha.gov/laws-regs/regulations/standardnumber/1926
11. U.S. Environmental Protection Agency, Construction Stormwater (NPDES), Lead Renovation, Repair and Painting (RRP) Program, and Section 608 Technician Certification. https://www.epa.gov/npdes/stormwater-discharges-construction-activities
12. International Code Council, 2024 International Building/Residential and Energy Conservation Codes; National Fire Protection Association, NFPA 70 (NEC); American Society of Mechanical Engineers, A17.1 Safety Code for Elevators; U.S. EPA, AIM Act HFC Restrictions by Sector. https://codes.iccsafe.org/; https://www.nfpa.org/; https://www.asme.org/; https://www.epa.gov/hfcs
13. U.S. Department of Labor, Fact Sheet #66: The Davis-Bacon and Related Acts. https://www.dol.gov/agencies/whd/government-contracts/construction
14. U.S. trade-policy reporting, Section 232 steel tariffs (raised to 50%, 2025) and cement/aluminum/tile duty impacts on construction input costs. https://www.constructiondive.com/
15. Joint Center for Housing Studies of Harvard University, Leading Indicator of Remodeling Activity (LIRA) and remodeling-spending releases. https://www.jchs.harvard.edu/
16. Industry reporting, U.S. wind and hail residential insurance-claim volumes (2024); IBISWorld, U.S. specialty-trade contractor industry reports. https://www.ibisworld.com/united-states/
17. Capstone Partners, Construction Services / HVAC Services M&A Updates (2025) — NAICS 238 TEV/EBITDA ~6–8×; Cascade Partners / BMI, electrical-contracting M&A; Construction Dive, private-equity roll-up activity, 2024–2025. https://www.capstonepartners.com/insights/article-construction-ma-update/
18. CNBC, KONE to buy TK Elevator in ~$34.4 billion deal (2026); Roofing Contractor / AXIA Advisors, private-equity roofing-platform trackers. https://www.cnbc.com/; https://www.roofingcontractor.com/
19. Engineering News-Record Top 600 Specialty Contractors and related top lists (largest private operators by reported revenue, 2024–2025). https://www.enr.com/toplists
20. Company filings and investor relations for the listed proxies cited in §4 and §10 — Otis (OTIS), Installed Building Products (IBP), EMCOR (EME), Comfort Systems (FIX), IES (IESC), MYR Group (MYRG), Quanta (PWR), MasTec (MTZ), Sterling Infrastructure (STRL), Granite (GVA), Apogee (APOG), FirstService (FSV); materials, distribution, equipment, and homebuilder names (VMC, MLM, CRH, EXP, OC, CSL, SHW, JHX, MHK, CARR, TT, LII, AOS, FERG, WSO, BLDR, HD, LOW, QXO, POOL, DHI, LEN, PHM, NVR). U.S. Securities and Exchange Commission EDGAR: https://www.sec.gov/edgar