Construction (United States) — NAICS 23

A Histometrics sector-level rollup primer for a general investing audience — relevant to both public-market and private investors. This is the two-digit sector, the top of the construction branch of the North American Industry Classification System (NAICS — the U.S. government's standard scheme for grouping businesses by what they do). It holds the entire private construction economy, split into just three children: 236 Construction of Buildings, 237 Heavy and Civil Engineering Construction, and 238 Specialty Trade Contractors. The value of this page is the contrast among those three — which is the biggest, which is growing, who owns the builders, how concentrated each is, and how (or whether) you can invest. Figures are reported federal facts with citations; statements about the future are labeled as judgments.

---

1. Overview

NAICS code 23, Construction, covers the firms that build, renovate, and repair the physical fabric of the country: the homes, offices, factories, and data centers people occupy; the roads, bridges, pipelines, and power and water lines that connect them; and the specialized trade work — concrete, steel, roofing, wiring, plumbing, drywall — that goes into all of it. [1] At roughly $2.9 trillion of annual contractor receipts and 7.7 million paid workers, construction is one of the largest goods-producing corners of the U.S. economy, on the order of 4% of gross domestic product. [2]

The sector divides into exactly three subsectors, and the division follows the natural sequence of a job:

• 236 — Construction of Buildings. The general contractors (GCs — a GC takes responsibility for delivering a finished structure) and builders who put up whole vertical buildings, both homes and everything that isn't a home. [3]
• 237 — Heavy and Civil Engineering Construction. The firms that build the big horizontal things — roads, bridges, utility lines, pipelines, water and sewer systems, ports, dams, and rail. [4]
• 238 — Specialty Trade Contractors. The subcontracted trades who do the actual hands-on work — pouring foundations, raising steel, roofing, wiring, plumbing, heating and cooling, drywall, painting, flooring, and site preparation — mostly working under a GC rather than for the owner directly. [5]

Two of these are roughly co-equal giants (buildings and specialty trades, each around $1.2–1.3 trillion), and one is a smaller but strategically central third (heavy-civil, about $416 billion). But the more useful way to read the sector is not by size alone — it is by how differently the three behave. They answer to different customers, attract different owners, run on somewhat different economics, and are reachable through completely different investment routes. That contrast is the whole point of this page.

---

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

The hierarchy narrows step by step: sector 23 (this page) → three subsectors (236, 237, 238) → industry groups (four-digit) → industries → the six-digit national industries where the leaf detail lives. The meaningful division is right here, at the top.

The contrast is the point. All three run on the same broad DNA — project-based, thin-margin, cyclical, skilled-labor-driven, and overwhelmingly private — but they diverge sharply on scale, on where the labor sits, on who owns the builders, and on how an outside investor can touch them.

Three things the table makes unmistakable.

• Buildings and specialty trades are near-equal in dollars, but nothing like equal in bodies. Construction of Buildings (~42% of receipts) and Specialty Trade Contractors (~44%) book almost the same revenue, yet 238 does it with three times the workforce (4.99 million vs. 1.62 million) and twice the firms (505,580 vs. 246,042). That is the signature of the general-contractor model: a building GC books the whole contract value with a thin coordinating staff (revenue per worker ~$754,000) and pays most of it out to specialty subs, who are the actual labor (revenue per worker ~$257,000). Roughly two of every three construction workers in America are specialty-trade subcontractors. [2][3][5]

• The smallest subsector has the biggest firms — and the deepest public menu. Heavy-civil is only ~14% of receipts, but its firms are the largest on average (~$11.9 million) and best-paid (~$91,300 a worker) because moving earth, erecting bridges, and stringing transmission lines is capital- and equipment-heavy work, often unionized. It is also, despite its size, the richest hunting ground for listed contractors — the diversified infrastructure names that have become the market's cleanest way to own the build-out theme. [4]

• Public-market depth is thin and unevenly spread. The single most important fact for a stock investor: this $2.9-trillion sector contains very few large, listed, pure-play builders. The only large pure-play building companies are the for-sale homebuilders inside 236. Nonresidential GCs, the biggest heavy-civil builders, and the vast bulk of specialty-trade firms are private. Public exposure to most of construction is therefore indirect — through diversified contractors, the supply chain, and the demand side — while the operating industry itself is a private-market opportunity. [3][4][5]

---

3. Size — this level's rollup figures

These are our ground-truth federal statistics for NAICS 23, with the three subsectors broken out so the mix is visible. [2] Receipts, firm counts, and concentration come from the 2022 Economic Census (EC); establishments, employment, and payroll come from County Business Patterns (CBP) 2023. The two programs use different years and definitions, so this is a size profile, not one clean annual series. Reassuringly, the three children reconcile to the sector: establishments and employment match to the person; receipts and payroll to within rounding.

Rollup column and CR4/HHI from our ground-truth extract for NAICS 23; [2] subsector columns from the 236, 237, and 238 material. [3][4][5] The firm-count row does not sum exactly — 246,042 + 34,900 + 505,580 = 786,522 versus the sector's 785,917 — because a firm operating in more than one subsector is counted once at the sector level.

Four quick reads:

• It is enormous and labor-heavy. Nearly 7.7 million paid workers and a ~$587 billion payroll. Average pay of about $76,600 sits above the private-sector average, reflecting skilled trades, operators, engineers, and project managers — but it blends a lower-paid specialty-trade base (~$71,600) with better-paid building (~$82,400) and heavy-civil (~$91,300) work. [2]
• It is seasonal. First-quarter payroll (~$134.8 billion) is only about 23% of the annual total — below a flat 25% quarter — the fingerprint of outdoor trades that slow in a northern winter (you cannot pour concrete or lay asphalt in a hard freeze). [2]
• It is near-atomistic. The four largest firms hold just 2.6% of receipts (top 50, only 11.4%), and the HHI — a 0–10,000 concentration score where anything under 1,500 is already "unconcentrated" — is 3.9. This is one of the least-concentrated large sectors in the entire economy: hundreds of thousands of firms compete and no one is remotely dominant. [2]
• The sector is less concentrated than two of its three children — but not the third. The rollup HHI (3.9) sits below Construction of Buildings (18.2) and Heavy-Civil (24.2): combining subsectors whose leaders don't overlap — the biggest homebuilder, the biggest line contractor, and the biggest roofer never bid against each other — dilutes any one firm's share of the whole. The exception is Specialty Trades, whose HHI of 1.7 is lower than the sector's, because that subsector is so vast (505,580 firms) and so perfectly competitive that it is more fragmented than construction as a whole. [2][5]

Read-with-care caveats — two of them matter a great deal at this level.

• Summed receipts double-count the pass-through — this is the big one. When a building GC (236) signs a $100-million project, it books the full contract value as receipts, then pays most of it out to the specialty-trade subs (238) and heavy-civil subs (237) who do the work — and they book their slices as receipts too. Summing the three subsectors therefore counts the same construction dollar more than once. The $2.92 trillion is gross contractor receipts, not the value of construction actually put in place. The government's separate "value of construction put in place" series (external context, not from our ground-truth file) runs materially lower — on the order of ~$2.1 trillion a year — precisely because it strips out that inter-contractor double counting. Treat $2.92 trillion as the industry's gross revenue, not as national construction output. [2][3][5]
• Small operators are undercounted, heavily. CBP and the Economic Census count only employer firms with paid payroll; they exclude the self-employed, no-payroll sole proprietors, and 1099 day-labor crews — and construction is thick with exactly those. The gap is largest in residential building (236) and in the specialty trades (238), especially roofing, framing, and painting, where private trackers count several times the employer-firm population. Read the 785,917 firm count as "employer firms," not everyone who builds for a living; the revenue totals are more reliable than the head-counts, and small/individual ownership dominates. [3][5]
• No official sector margin or backlog. Federal data publish no NAICS-23 operating margin, order backlog, or win rate. Do not infer any from a single company's results.

---

4. Investable universe — where value concentrates across the subsectors

There is no single public company whose revenue equals this sector — the work is local, project-based, and overwhelmingly private and municipal. Public depth is thin and clusters in a few distinct places, and where it sits across the three children is the useful map. The golden rule is to judge every listed name by the share of its revenue genuinely tied to construction — and which subsector — not by headline size.

• 236 — the only large listed pure-plays in the whole sector. A dozen-plus listed homebuilders (the for-sale/merchant model — they own the land, carry the finished house as inventory, and sell it with the lot) are the sole large, direct, exchange-traded building companies. The rest of residential (custom homes, apartments, remodeling) and all of nonresidential general contracting — Turner, Bechtel, Whiting-Turner, DPR, Kiewit — are private, reached indirectly through apartment REITs (real estate investment trusts — companies that own income property and pass most of their income to shareholders), building-products makers, home-improvement retail, and homebuilder ETFs (exchange-traded funds — baskets of stocks that trade like a single share). [3]
• 237 — the richest listed menu, and the cleanest "infrastructure" theme. The diversified heavy-civil contractors that self-perform across roads, utility lines, and site work are the market's best single-ticker proxies for the build-out; a shorter list also gives near-pure-play utility, road-materials, and marine exposure, and broad infrastructure ETFs bundle them. Even so, the largest jobs — the biggest dams, tunnels, and pipelines — are built by private giants. [4]
• 238 — mostly indirect, with a few real listed handles. Because the trades are overwhelmingly private, listed exposure runs through the specialty mechanical and electrical contractors (the cleanest way in, with the most direct data-center leverage and steadier service margins than GCs), the one liquid elevator service business, one listed insulation installer, and — one step off the jobsite — the building-products makers and distributors that sell into the whole fragmented contractor base. [5]

The through-line for value. Two patterns hold across all three subsectors. First, defensible margin sits away from plain general contracting — in the capital-intensive for-sale homebuilders, the specialty and diversified contractors, the recurring-service trades (elevator maintenance, HVAC replacement, roofing repair), and the supply chain (aggregates, cement, building products, distribution). Second, a short list of diversified specialty and infrastructure contractors — the mechanical/electrical and heavy-civil names — appears in two or more children at once, which is exactly why the sector is investable as a theme even though no single company equals it. Tickers are reserved for Section 10. [3][4][5]

---

5. How the money works

Construction runs on one dominant profit model plus two genuine exceptions — and it is worth being explicit that the specialized frameworks used elsewhere in finance do not apply here. These are builders, not asset owners: there is no regulated-utility rate base, no real-estate funds-from-operations, and no mining cost-per-ounce measure to reach for. Value them as contractors.

• The dominant model: project-based, fixed-price contracting. A contractor wins work by competitive bid — a lump sum, or a unit price (dollars per cubic yard poured, per square of roof, per ton of steel, per crane-hour) — 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 (percentage-of-completion), so accurate estimating drives reported earnings and a single mis-bid surfaces as a margin write-down. Backlog — signed, not-yet-built work — is the most-watched forward gauge, though it can be delayed, resized, or canceled. Margins are thin (mid-single digits net is normal), volume is the primary lever, and there is almost no pricing power in a field this fragmented. [4][5]
• Exception one — the for-sale homebuilder inventory model (inside 236). Merchant builders own the land, carry the home as inventory, and earn a gain on selling finished land-and-house. This is capital-intensive and highly cyclical, valued on price-to-book and normalized through-cycle earnings rather than a single-year multiple. [3]
• Exception two — the land-subdivision spread (inside 237). Land developers buy raw land, win entitlements, install horizontal infrastructure, and sell finished lots to homebuilders for more than their all-in basis — a real-estate business judged on net asset value (NAV) and land basis, not backlog. [4]

What unites all of it: thin, cyclical margins; volume as the main margin lever; and working-capital timing. 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 a paper-profitable firm can be sunk on liquidity even in a good year. Surety bonding capacity effectively caps how large a project a firm can chase, which is a real barrier to scaling and a structural advantage for the better-capitalized. The higher-quality slice everywhere is the recurring, service-and-replacement end — elevator maintenance, HVAC repair, re-roofing — which is steadier and better-valued than one-and-done new installation. [3][4][5]

---

6. Demand drivers

Construction is derived demand — it tracks how much other people decide to build — and the whole sector is deeply cyclical and interest-rate-sensitive. But the three children are wired to different customers and are currently on different clocks, and two structural themes cut across all of them.

• The common master variable is the cost of money. Interest and mortgage rates set homebuyer affordability, developer and commercial-project financing, and the appetite for debt-funded capital budgets. A decisive fall in rates would lift all three subsectors together; today's elevated rates are the single biggest brake, felt most on the residential and land-development side. [3][4]
• The AI data-center and power build-out is the strongest current tailwind, and it touches all three. Data-center construction has been the fastest-growing construction segment, and it lands on 236 (the nonresidential GCs building the facilities), 237 (the utility and power-line work feeding them — the largest utility-capital wave in history), and 238 (the concrete, structural steel, electrical, mechanical, and site-prep trades that actually build them). This one theme is masking softness elsewhere across the sector. [3][4][5]
• Housing is the second structural driver — and it is rate-gated and depressed. A structural shortage of several million homes, an aging housing stock, and a mortgage "lock-in" effect (owners with cheap loans won't sell, freezing resale supply and channeling demand into new homes and remodeling) all support 236, the finishing and framing trades in 238, and land subdivision in 237 — but the segment is waiting on lower rates to re-accelerate. [3][5]
• Public infrastructure and reshoring add lumpier layers. The 2021 Infrastructure Investment and Jobs Act (IIJA) drove roads, bridges, water, and utility work, but its surface-transportation authority expires September 30, 2026 — a funding cliff that hangs over the government-funded parts of 237. Factory reshoring behind the 2022 CHIPS Act (Creating Helpful Incentives to Produce Semiconductors) added a wave that appears to have peaked. [4][5]

Judgment. The sector is not on one clock. Residential is depressed and rate-gated; heavy-civil is near record but hinges on 2026 reauthorization; the specialty trades are bifurcated between a booming new-build/data-center segment and a defensive replacement segment. The two forces that would move the whole sector at once are lower interest rates and the durability of the data-center capital cycle. [3][4][5]

---

7. Regulation

There is no single federal regulator for construction. The binding rules apply at the point of work, and the same stack reaches all three subsectors, with segment-specific overlays.

• State and local rules are the biggest swing factor. Zoning, land use, density and permitting, building and energy codes, and contractor licensing all sit at the local level, and the U.S. has no single national building code. On the residential side, local zoning and entitlement is the single biggest determinant of whether anything gets built. [3][4]
• Federal overlays reach the job site. Worker safety under the Occupational Safety and Health Administration (OSHA, 29 CFR Part 1926) — silica, fall protection, trenching, and steel-erection rules recur trade to trade; environmental permitting under the Environmental Protection Agency (EPA) and the National Environmental Policy Act (NEPA), which can add months or years to a large project; and, on public and subsidized work, prevailing-wage rules (Davis-Bacon), domestic-content requirements (Buy America), and surety bonding. [3][4][5]
• Trade and labor policy bear directly on costs. Section 232 tariffs on steel and aluminum and duties on cement and tile flow straight into fixed-price bids; and because the workforce is heavily immigrant, employment-eligibility (I-9/E-Verify) enforcement is a direct swing factor on both labor supply and cost across the whole sector. [5]
• Segment-specific regimes. Utility, water, and pipeline work is shaped by the Safe Drinking Water and Clean Water Acts, FERC, and state utility commissions (237); the refrigerant transition (the AIM Act) and electrical and elevator safety codes gate the mechanical trades (238); and lead-paint, warranty, and lien law bear on residential remodeling (236). [3][4][5]

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

---

8. Consolidation

The pattern across the whole sector is fragmented at the base, consolidating at the top. The base is hundreds of thousands of small local firms — work stays local because crews, equipment, permits, bonding, and site knowledge travel poorly. The top is a growing set of national consolidators, private-equity (PE) platforms, and employee-owned firms buying up regional specialists for their crews, licenses, service contracts, and bonding capacity. The sector's very low concentration (CR4 2.6%, HHI 3.9) shows how early this still is — the runway is long. [2]

Where and how consolidation runs differs by child:

• 236 — Buildings. The for-sale homebuilders are consolidating fastest (the ten largest now capture a record share of new single-family closings), because land, capital, and marketing reward scale. Custom building and remodeling stay stubbornly fragmented. Nonresidential general contracting has largely resisted roll-ups. [3]
• 237 — Heavy-Civil. National utility consolidators and PE platforms are rolling up regional specialists; materials-side roll-ups (aggregates, paving) run through roads; and a few protected niche oligopolies exist (dredging). [4]
• 238 — Specialty Trades. This is where consolidation runs hottest — the heaviest PE roll-up wave in the sector, concentrated in roofing, HVAC, and electrical, where recurring service revenue and higher margins support it; ESOPs (employee stock ownership plans) are the dominant succession tool; and the one genuinely concentrated pocket, the global elevator oligopoly, is itself consolidating further (a pending ~$34 billion merger of two of the majors). [5]

The unifying logic: scale compounds where land, capital, marketing, recurring service, or specialized engineering dominate; it stays elusive where the work is bespoke, local, and relationship-driven. And the binding brake everywhere is skilled labor, not demand or capital. [3][4][5]

---

9. Risks

The three subsectors share one risk register, tilting differently within it.

• Cyclicality and interest-rate risk. Construction is among the most cyclical activities in the economy; it turns with financing conditions — mortgage rates for residential, capital budgets and project financing for the rest. [3][4][5]
• Skilled-labor shortage — the binding constraint. A structural, aging shortage caps how much work any firm or roll-up can take on and pushes up wages, worsened by immigration enforcement of a heavily foreign-born workforce. The industry needs hundreds of thousands of additional workers a year just to stand still. [4][5]
• Thin margins and fixed-price execution risk. One mis-priced job or a cost overrun on a lump-sum project can erase profit; input-cost inflation and tariffs (steel, aluminum, cement, lumber, diesel) erode already-thin margins, hardest on fixed bids without escalation clauses. [3][4][5]
• Theme concentration. Recent growth leans heavily on data centers and, until recently, reshoring; a cooling of the AI capital cycle would hit the fastest-growing segments (nonresidential building, utility power, site prep, and electrical) hardest. [4][5]
• Working-capital, retainage, and counterparty risk. Contractors pay first and collect last; the distress of a GC, developer, or owner flows straight down to the subcontractor. [3][5]
• Policy cliffs and reversals. The IIJA surface-transportation expiry (September 30, 2026), the winding-down CHIPS build-out, tariffs, immigration, and prevailing-wage rules are all politically contingent and bear directly on demand and cost. [4][5]
• Roll-up and leverage risk. PE-backed platforms across all three children carry integration and leverage risk against volatile cash flows. [5]
• Measurement risk. Federal employer statistics omit a large nonemployer tail (heaviest in residential and the specialty trades), and summed subsector receipts double-count pass-through work — so top-down sizing is approximate. [2][3][5]

---

10. How to invest and outlook

How to invest. The sector splits investors cleanly by market.

• Public-market investors have genuine direct depth in only one place — listed homebuilders (the for-sale model in 236), the sole large pure-play builders in the whole sector; value them on price-to-book and normalized through-cycle earnings, not a peak price-to-earnings multiple. Everything else is indirect, and it is best organized by subsector. For the heavy-civil / infrastructure theme (237), the diversified contractors that self-perform across roads, utility lines, and site work are the cleanest single-ticker proxies — Quanta Services (PWR), MasTec (MTZ), Primoris (PRIM), Granite Construction (GVA), Sterling Infrastructure (STRL), Tutor Perini (TPC) — alongside road-materials producers (Vulcan VMC, Martin Marietta MLM) and broad infrastructure ETFs (e.g., PAVE; check current holdings). For the specialty-trade theme (238), the listed mechanical and electrical contractors are the most direct handle — EMCOR (EME), Comfort Systems USA (FIX), IES Holdings (IESC) — plus the elevator service annuity (Otis, OTIS), a listed insulation installer (Installed Building Products, IBP), and the building-products and distribution layer (Ferguson FERG, Watsco WSO, Home Depot HD, QXO QXO). For the residential themes (236), apartment REITs, home-improvement retail, building-products makers, and homebuilder ETFs. In every case, weigh backlog quality, service mix, and margin trajectory over headline revenue, and size the position to the actual construction exposure — a quarry, a shingle mill, or a distributor is not a contractor, and its economics differ. [3][4][5]
• Private-market investors find the operating industry in everything the public market can't reach — which is most of it: nonresidential general contractors and the biggest data centers and fabs; the private heavy-civil giants; and the vast base of specialty-trade firms. The routes are private equity, private credit and project finance, ESOP participation, real-assets and digital-infrastructure strategies, and — for government-funded work — the municipal and state bonds that finance the projects. Returns come from operating skill, backlog and job-cost discipline, and cash conversion, not index beta. [3][4][5]

Outlook (mid-2026; forward-looking judgment). Construction is a fragmented, thin-margin, execution-driven, ~$2.9-trillion (gross) sector, and its three parts are on different clocks. Buildings (236) are two-speed and mostly rate-gated — residential held up by a multi-million-unit housing shortfall and frozen resale inventory, nonresidential carried almost entirely by data centers while offices, retail, and hotels stay soft (the American Institute of Architects' consensus expects nonresidential building spending roughly flat in 2026 before a modest 2027 recovery). Heavy-civil (237) is near record but funding-gated — the utility/power super-cycle is the most durable tailwind in the sector, while highways hinge on a 2026 federal reauthorization. Specialty trades (238) are bifurcated — the data-center-levered trades (concrete, steel, electrical, mechanical, site prep) surging, the replacement trades (roofing, HVAC service) defensively steady, and the homebuilding-levered trades (framing, drywall) waiting on rates. The two variables that would move the whole sector at once are the cost of money and the durability of the data-center capital cycle. Across all three, the durable lesson is the same: money is made through backlog and inventory discipline, risk control, skilled-labor retention, and cash conversion — and, on the public side, through the listed homebuilders, the diversified infrastructure contractors, and the specialty mechanical/electrical names, rather than the private general-contracting giants. [3][4][5]

For the full segment detail — company rosters, contract economics, and the four-, five-, and six-digit breakdowns — see the three subsector primers: [3] Construction of Buildings, [4] Heavy and Civil Engineering Construction, and [5] Specialty Trade Contractors.

---

Sources

This is a three-child sector rollup. Sector-level figures come from our ground-truth federal statistics for NAICS 23; all synthesis, contrast, and forward judgments are drawn from the child material (236, 237, and 238), which themselves aggregate the industry-group and six-digit primers and their federal and industry sources. The three children reconcile to the sector totals (establishments and employment to the person; receipts and payroll to within rounding). Note: the NAICS 238 subsector rollup primer was not yet available at the time of writing, so the 238 column and commentary are synthesized from our ground-truth NAICS 238 statistics and the four NAICS 238 industry-group primers (2381, 2382, 2383, 2389).

1. U.S. Census Bureau. 2022 NAICS Definition — Sector 23, Construction (scope; three subsectors: 236 Construction of Buildings, 237 Heavy and Civil Engineering Construction, 238 Specialty Trade Contractors). 2022. https://www.census.gov/naics/?input=23&year=2022
2. Histometrics ground-truth federal statistics — NAICS 23 (receipts $2,920,771,250 thousand and firms 785,917, Economic Census 2022; establishments 814,557, employment 7,659,459, annual payroll $586,559,516 thousand, Q1 payroll $134,815,706 thousand, County Business Patterns 2023; CR4 2.6%, CR8 4.0%, CR20 6.8%, CR50 11.4%, HHI 3.9, Economic Census concentration 2022). The three subsectors reconcile to these totals. External context on "value of construction put in place" (~$2.1 trillion, a separate Census series that excludes inter-contractor double counting) is noted as context, not part of this ground-truth extract.
3. Histometrics rollup primer — NAICS 236, Construction of Buildings (receipts $1,223.2B, firms 246,042, establishments 254,630, employment 1,621,439, annual payroll $133.7B, Q1 payroll $32.4B, CR4 6.1%, HHI 18.2; two children — 2361 Residential and 2362 Nonresidential Building Construction — plus the GC/for-sale economics, the listed-homebuilder roster, demand drivers, regulation, consolidation, risks, and the mid-2026 outlook). 2026.
4. Histometrics rollup primer — NAICS 237, Heavy and Civil Engineering Construction (receipts $416.05B, firms 34,900, establishments 38,612, employment 1,051,261, annual payroll $95.99B, Q1 payroll $21.28B, CR4 7.2%, HHI 24.2; four industry groups — 2371 Utility System, 2372 Land Subdivision, 2373 Highway/Street/Bridge, 2379 Other Heavy & Civil — plus contract economics, the diversified-contractor and materials investable roster, IIJA and utility-capex demand, regulation, consolidation, risks, and outlook). 2026.
5. Histometrics ground-truth federal statistics — NAICS 238, Specialty Trade Contractors (receipts $1,281,526,572 thousand, firms 505,580, Economic Census 2022; establishments 521,315, employment 4,986,759, annual payroll $356,906,230 thousand, Q1 payroll $81,149,017 thousand, County Business Patterns 2023; CR4 1.8%, CR8 2.7%, CR20 4.6%, CR50 7.1%, HHI 1.7, Economic Census 2022) — plus the four NAICS 238 industry-group rollup primers, from which the 238 column is synthesized (its own subsector rollup was not yet available): 2381 Foundation, Structure & Building Exterior Contractors; 2382 Building Equipment Contractors (electrical, plumbing/HVAC, elevators); 2383 Building Finishing Contractors (drywall, painting, flooring, tile, carpentry); 2389 Other Specialty Trade Contractors (site preparation and residual trades). 2026.

For the underlying company-level detail, contract economics, and full source lists, see the three subsector primers (236, 237, 238) and the industry-group and six-digit primers beneath them.