On this page
Construction Labor Supply: A Multi-State Contractor-License Census
Hiring demand tells you where the work is going; it does not tell you whether anyone is credentialed to do it. This report reads the other side of the market — the supply of licensed contractors who are legally allowed to perform the work — and where that base is deep enough to staff a build versus too thin to find a crew at any wage. It is built from roughly 1.8 million state contractor-license records across ten states and the District of Columbia: the credentialed firm-and-tradesperson base, counted directly from the public licensing registries that govern who may legally contract construction work. It is the structural counterweight to the Lab's Hiring Demand Signal.
Key metrics
The size of the base matters less than its structure and where it concentrates. The figures below pair each headline number with the context that makes it a supply signal — every figure is drawn directly from the covered registries.
| Metric | Value | What it means |
|---|---|---|
| Total license records | 1,795,385 | Across 10 states + DC — a large sample, not a national total |
| Distinct firms | ~1,491,000 | After cross-registry de-duplication of the ~1.8M records |
| Multi-state firms | 8,374 | Hold credentials in more than one covered state |
| Deepest commercial pool | ~224,924 | California distinct firms at a 96% active share |
| Largest raw registry | 413,335 | Connecticut license records; mostly individual-held (317,159) |
| Active-status range | 22%–100% | CT/DC (~22%) to OR/VA/AR (100%) — partly definitional |
Why this signal leads
A contractor license is a credential to operate: a public, dated record that a firm or tradesperson has met a state's requirements to perform construction work. Aggregated across states, license registries become a supply-side map of the construction market — the depth of the credentialed base in a given place, its composition, and its renewal health. The underlying spine holds about 1.8 million license records from ten state registries plus the District of Columbia, normalized to a common schema and de-duplicated into roughly 1.49 million distinct firms.
It is the structural counterpart to the demand and exposure briefs: where those read hiring activity and operational risk, this reads the standing credentialed capacity that any hiring market draws from. Demand signals tell you where roles are being posted; the supply census tells you whether the credentialed base in that market is deep enough to fill them. That is what makes it a leading structural read — the base is fixed in the near term, so a thin pool is a binding constraint long before a wage print reveals it.
Three properties of licensing data govern how much weight any single number deserves, and they are stated up front because they shape every figure that follows.
The credentialed supply base
Across the covered jurisdictions, the spine holds 1,795,385 license records resolving to roughly 1.49 million distinct firms. A small but strategically important slice of those firms operate across more than one state.
Geographic concentration
The table ranks covered jurisdictions by total license records, with the active-status share and the business- vs individual-held split shown separately. The composition differs sharply by state — a direct consequence of how each registry is structured, and the reason cross-state totals should be read with care rather than summed naively.
| State | License records | Active | Active share | Business-held | Individual-held | Distinct firms |
|---|---|---|---|---|---|---|
| Connecticut | 413,335 | 90,410 | 22% | 96,176 | 317,159 | 291,280 |
| Florida | 280,548 | 264,136 | 94% | 130,849 | 149,699 | 276,488 |
| Minnesota | 273,816 | 135,988 | 50% | 57,075 | 216,741 | 224,346 |
| California | 244,440 | 233,655 | 96% | 128,041 | 116,399 | 224,924 |
| Colorado | 227,422 | 62,512 | 27% | 17,755 | 209,667 | 171,455 |
| Washington | 161,008 | 75,419 | 47% | 127,103 | 33,905 | 149,299 |
| Virginia | 84,912 | 84,881 | 100% | 50,594 | 34,318 | 82,418 |
| Oregon | 55,944 | 55,944 | 100% | 55,944 | 0 | 45,392 |
| Washington DC | 34,255 | 7,454 | 22% | 22,271 | 11,984 | 19,482 |
| Arkansas | 19,705 | 19,705 | 100% | 15,467 | 4,238 | 17,756 |
Connecticut and Florida anchor the covered set on raw record count; California and Florida anchor it on active, business-held depth — the unit most relevant to a firm trying to hire or subcontract. California alone contributes roughly 224,000 distinct firms with a 96% active share, the deepest clean commercial supply pool in the sample.
License structure: firms vs tradespeople
The business- vs individual-held split is the single most important structural variable in the data, because it determines what a state's license count actually represents.
- Firm-dominant registries — Oregon (entirely business-held) and Washington (127k business vs 34k individual) license at the company level; their counts read as a firm census.
- Tradesperson-dominant registries — Connecticut (317k individual), Colorado (210k), and Minnesota (217k) license many individual trades; their counts read closer to a credentialed-worker census.
- Mixed registries — California and Florida carry substantial counts of both, making them the most directly useful for reading firm depth and trade depth together.
The practical implication: when comparing supply across states, compare like with like — business-held against business-held for firm depth, individual-held against individual-held for trade depth. Summing the two across dissimilar registries produces a number that does not mean anything.
Active-status and renewal health
Active-status share — the fraction of records a registry currently reports as active — ranges from about 22% (Connecticut, DC) to 100% (Oregon, Virginia, Arkansas). Part of that spread is real renewal and lapse behavior; part is definitional, because states differ in whether they retain and surface expired, inactive, and historical records.
Because the definition is not uniform, active share is most defensible as a within-statesignal — a renewal-health read for a given registry over time — rather than as a cross-state ranking. Where a registry maintains a clean active flag (California, Florida, Virginia), the active business-held count is the most reliable proxy for currently-operating supply.
Labor implications
The credentialed base is the denominator behind every construction-hiring decision: it is the pool any employer, developer, or subcontractor draws from, and it cannot be expanded faster than the states issue and renew licenses. That makes the Contractor Capacity Census™ the supply-side anchor of the AlphaHire framework — the structural read that the demand-side signals must be set against to mean anything operationally.
- Depth, not just demand. A deep posting pipeline in a market with a thin active, business-held base is a structural shortage in the making — the roles get posted but the credentialed firms to fill them are not there at any wage.
- Read in the right unit. Firm-dominant registries (Oregon, Washington) tell you about subcontractor and prime depth; tradesperson-dominant registries (Connecticut, Colorado, Minnesota) tell you about credentialed-worker depth. The binding constraint differs by which unit your scope needs.
- Active share gates real capacity. Where a registry keeps a clean active flag, the active business-held count — led by California (96%), Florida (94%), and Virginia (100%) — is the most reliable proxy for currently-operating supply you can actually engage.
- Multi-state operators are the mobile slice. The 8,374 firms credentialed in more than one covered state are the part of the base that can follow work across markets — the capacity most likely to chase the deepest demand.
This is where the supply census connects to the rest of the framework. Set it against the supply–demand balance to locate the markets where deep demand meets shallow active supply; against the contractor-workforce-pressure read to see where that imbalance is already biting; and against the role-demand report to map which credentialed roles the thin pools most constrain. The census measures the standing capacity; the demand reports measure the claim on it.
How to apply this
- GCs & EPCs: use the census as a depth check — how deep is the active, business-held supply in the states you operate in, and is it firm-dominant or tradesperson-dominant for the scope you need to subcontract?
- CFOs & finance leaders: flag the markets where the credentialed base is thin enough that hiring or subcontracting will be structurally hard regardless of wage — that is where labor cost and schedule risk compound.
- Workforce-planning leaders: read supply in the right unit — firms vs tradespeople — for the decision in front of you, and watch the cells that pair deep demand (from the Hiring Demand Signal) with shallow active supply.
- Investors, PE operating partners & lenders: underwrite in-region craft-and-firm availability as a named execution risk where the active, business-held base is shallow against the work coming into the market.
Methodology
Source. Public state contractor-licensing registries for ten states (CA CSLB, FL DBPR, CT DCP, MN DLI, CO DORA, WA L&I, VA DPOR, OR CCB, AR CLB) and the District of Columbia, loaded May 30–31, 2026. Four further target states remain blocked by registry anti-scraping controls and are excluded. Normalization (CLS-v1.0). Records are mapped to a common license spine, classified as business- or individual-held, assigned an active-status flag where the registry provides one, and de-duplicated into distinct firms across registries. Privacy. Only aggregate counts are published here — no licensee names, addresses, license numbers, or any individual record is reproduced on the public surface. Comparability. License taxonomies and status definitions are not standardized across states; figures are reported per state and should not be summed across dissimilar registries. See the Methodology page for the Lab's confidence-handling and directional-framing standards.
What this report shows & doesn't
- What this report shows. The standing credentialed base of the construction market in the covered states: how many firms and tradespeople hold the licenses required to operate, how that base is structured (business- vs individual-held), and how much of it is currently active. It is the supply-side counterpart to the demand and exposure briefs, and the input to the supply–demand balance read.
- What this report does not show. It is not a forecast, not a headcount of working construction labor, not a national total, and not a skills inventory. Coverage is ten states and DC only (four target states are scraper-blocked); license records are credentials to operate, not active headcount; license possession does not guarantee availability, capacity, or willingness to take new work — the most-constrained firms are often the busiest. Cross-state specialty comparison is limited by non-standard taxonomies, and active-status definitions vary by jurisdiction. Because this is a point-in-time May 2026 snapshot, no long-run historical year-over-year series is included.
- Confidence level. High on the empirical record counts (public state license registries, ten states + DC). Moderate / directional on cross-state composition reads and the de-duplicated distinct-firm and multi-state-firm figures (AlphaHire CLS-v1.0 normalization). Trust the within-state depth and structure; read cross-state totals as a structural map rather than a precise national count.