Construction Union Membership Tracker 2025: Rate, Earnings and States

This month in brief

• In 2025, 11.1% of U.S. construction wage and salary workers were union members (995,000 of 9,000,000), and 12.0% were represented by a union (BLS Table 3).
• The 2025 rate rose 0.8 points from 10.3% in 2024, the lowest construction rate in both the BLS series (since 2000) and the unionstats.com series (since 1983).
• Over the long run the private construction rate fell 16.5 points, from 27.5% in 1983 to 11.0% in 2025, a 60.0% decline; the number of members fell 12.4% while construction employment rose 119.0%.
• Construction remains about twice as unionized as the private sector: 11.0% vs 5.9% in 2025 (ratio 1.86), compared with 27.5% vs 16.5% in 1983 (ratio 1.67).
• Construction union members earned a median $1,585 a week in 2025 vs $1,132 for nonunion workers, a raw gap of 40.0%, about twice the 19.6% gap across all workers.
• State rates differ widely: pooled 2023 to 2025 rates run from 32.1% in Hawaii and 26.5% in Illinois to 0.8% in North Carolina and South Dakota.
• Union filings tell a different story from the household survey: 14 building trades unions reported 3,262,114 members in 2025 LM-2 filings, 7.6% more than in 2015, but these counts include retirees, non-construction and non-U.S. members.

Construction union membership in 2025: the latest reading

Construction union membership rose in 2025 after reaching its lowest recorded level in 2024. About one in nine construction wage and salary workers belonged to a union, and about one in eight was covered by a union contract.

In 2025, 995,000 of the 9,000,000 wage and salary workers in the U.S. construction industry were union members, a membership rate of 11.1% (BLS Union Members 2025, Table 3). A further 83,000 were covered by a union contract without being members, so 1,078,000 workers (12.0%) were represented by a union.

The 2025 membership rate was 0.8 points above the 10.3% recorded in 2024, and the number of members rose by 79,000 (8.6%). Construction wage and salary employment rose from 8,894,000 to 9,000,000, so membership grew faster than the workforce.

Construction was the second most unionized private industry group in 2025, after transportation and utilities (14.3%), and well above manufacturing (7.7%) and the private-sector rate of 5.9% (BLS Table 3; BLS Economics Daily). Construction held 13.5% of all private-sector union members while employing 7.2% of private-sector wage and salary workers.

Measured by occupation instead of industry, union membership is higher: 16.4% of workers in construction and extraction occupations were union members in 2025 (17.4% represented), up from 15.4% in 2024. The occupation group counts tradespeople wherever they work, including in manufacturing, utilities and government, and excludes office and management staff of construction firms.

Union member: A wage and salary worker who answers yes to: on this job, are you a member of a labor union or of an employee association similar to a union? Self-employed workers are excluded (BLS Technical Note).

Represented by a union: Union members plus workers who are not members but whose job is covered by a union or employee association contract.

For workforce context, see the construction labor shortage tracker and the report on immigrant construction workers.

Construction union membership rate since 1983

The construction union membership rate has fallen by more than half since 1983, but the number of union members has fallen much less, because construction employment more than doubled. The decline was steepest from 1983 to 2006 and resumed after 2017.

In 1983, 27.5% of private-sector construction wage and salary workers were union members; in 2025 the rate was 11.0%, a drop of 16.5 points or 60.0% (unionstats.com private construction series). The lowest year in the series was 2024, at 10.3%.

The number of private construction union members fell from 1,130,700 in 1983 to 990,600 in 2025, a decline of 140,100 or 12.4%. Over the same years private construction wage and salary employment rose 119.0%, from 4,109,000 to 8,996,800, so almost all job growth was nonunion.

The all private-sector rate fell from 16.5% to 5.9% over the same period (a 64.2% decline). The point gap between construction and the private sector narrowed from 11.0 to 5.1, but in relative terms construction became more distinct: its rate was 1.67 times the private-sector rate in 1983 and 1.86 times in 2025.

On the official BLS series, which begins in 2000, the construction rate fell from 17.5% in 2000 to 11.1% in 2025 (6.4 points), while the private-sector rate fell 3.1 points to 5.9% (BLS LU database). The 2023 to 2025 average (10.7%) remains below the 2017 to 2019 average (13.1%).

How construction compares with other industries

Construction and manufacturing were equally unionized in 1983. Since then construction has lost a smaller share of its union density and far fewer members, so it is now one of the most unionized private industries.

Counting all wage and salary workers in each sector, construction's union membership rate fell from 28.0% in 1983 to 11.5% in 2025 (down 58.9%), while manufacturing fell from 27.9% to 7.8% (down 72.0%) (unionstats.com sector series). Construction union membership fell 16.6% in headcount terms; manufacturing union membership fell 78.9%, from 5,353,700 to 1,131,500.

Transportation, communications and utilities remained the most unionized private sector group in 2025 (17.3%, down from 46.4%). Public administration, a government-heavy sector, was the only sector shown whose rate was higher in 2025 (30.7%) than in 1983 (30.0%).

Union vs nonunion construction earnings

Union members in construction have higher median weekly earnings than nonunion workers, and the raw gap is about twice as wide as across the whole workforce. The gap is not a controlled estimate of what a union adds to pay.

Full-time construction union members had median weekly earnings of $1,585 in 2025, compared with $1,132 for nonunion construction workers, a raw premium of 40.0% or $453 a week (BLS Table 4). Nonunion workers earned 71.4% of the union median. At 52 weeks, the weekly gap equals about $23,556 a year, a simple annualization of medians rather than an annual earnings estimate.

The raw gap is wider in construction than across the economy: 19.6% for all wage and salary workers ($1,404 vs $1,174) and 16.7% in the private sector ($1,364 vs $1,169) (BLS). In construction and extraction occupations the gap was 55.5% ($1,561 vs $1,004).

The construction premium has narrowed over time. It was 50.0% in 2000, peaked at 60.3% in 2007, fell to 35.1% in 2022 and stood at 40.0% in 2025. From 2024 to 2025, nonunion median weekly earnings rose 7.7% and union median earnings rose 3.6%, so the premium fell from 45.6% to 40.0%.

For pay levels by metro and occupation, see construction pay by metro.

Construction union membership by state

Construction union density follows the same geography as union density across all industries: highest in the Great Lakes, Hawaii, New York and the West Coast, lowest in the South and Mountain West. Small samples make three-year pooling more reliable than single-year rates.

Pooling 2023, 2024 and 2025, the highest construction union membership rates were in Hawaii (32.1%), Illinois (26.5%), Missouri (24.1%), Ohio (24.0%) and Michigan (22.6%) (unionstats.com state files). The lowest were North Carolina and South Dakota (0.8%), Florida and South Carolina (2.0%) and Vermont (2.6%).

Of the 51 jurisdictions, 23 had pooled rates above the U.S. pooled rate of 10.7%, and 15 were below 5%. The spread between the highest and lowest pooled rates was 31.3 points.

By headcount, California (111,700), New York (100,700), Ohio (66,400), Michigan (65,200) and Illinois (61,400) had the most private construction union members in 2025, together 405,400 or 40.9% of the U.S. total. Texas had 45,100 union members but a rate of 4.0% in 2025.

Union-reported membership in the building trades

Unions report their own membership to the Department of Labor each year. These counts are larger than the survey estimate and rose over the past decade, but they measure something different and are not comparable with the CPS rate.

The 14 national building trades unions tracked here reported 3,262,114 members in their LM-2 filings for fiscal years ending in 2025, up 7.6% from 3,031,211 in their 2015 filings (DOL OLMS yearly data). The largest were the International Brotherhood of Electrical Workers (740,182), the Laborers (606,450), the Carpenters (443,045), the Operating Engineers (413,007) and the United Association (388,205).

These are union-reported totals, not survey estimates. They can include retirees (the Laborers reported 82,464 retired members), members outside construction (the Laborers also reported 42,385 regular mail handler members, and SMART's full name covers air, rail and transportation workers) and members outside the United States. The United Association's 2025 filing, for example, lists 293,894 building trades journeymen and 60,321 building trades apprentices within its total.

• North America's Building Trades Unions says it represents more than 3 million craft professionals in the United States and Canada, and that its affiliates had net growth of 49,554 members and 314,958 registered apprentices in 2024 (NABTU; NABTU statement). Organization's own statements.
• The Institute for Construction Employment Research, cited by NABTU, estimates union density of 17.9% among construction craft workers and 29.9% among nonresidential skilled trades after excluding office occupations and adjusting for misclassification (NABTU statement; ICERES). These use a different definition from BLS.
• Associated Builders and Contractors, which represents merit shop (open shop) contractors, describes the 2025 BLS figure as 88.9% of construction workers not belonging to a union, after a record-high 89.7% in 2024 (ABC 2026; ABC 2025). Organization's own statements.

Union density matters for public procurement debates such as project labor agreements on federal work; see federal construction contracts by state for where that spending goes. This tracker takes no position on unionization.

When this tracker updates next

The tracker updates once a year, when BLS publishes the annual Union Members release and unionstats.com posts its matching files.

BLS publishes Union Members once a year, usually in late January: the 2023 data came out on 2024-01-23 and the 2024 data on 2025-01-28 (BLS release archive). The 2025 release came on 2026-02-18 (BLS 2026 calendar). As of 2026-10-06 the date for Union Members 2026 was not yet on the BLS release calendar.

• BLS Union Members 2026 (Table 3 membership, Table 4 earnings, Table 5 states): this tracker's headline rate, member count and premium.
• unionstats.com 2026 files (completed on 2026-01-23 for the 2025 data, before the BLS release): long series and state construction rates (unionstats.com).
• OLMS LM-2 filings for fiscal years ending in 2026: union-reported counts, which unions file after their own fiscal years close.

If CPS collection runs for all 12 months of 2026, the 2026 annual figures will again be full-year averages comparable with years before 2025.

Momentum and share analysis

Download the calculations (CSV)

Findings to track

1. In 2025, 995,000 U.S. construction wage and salary workers were union members, 11.1% of the 9,000,000 employed in the industry.^[2]
2. In 2025, 1,078,000 construction workers (12.0%) were represented by a union, including 83,000 covered by a contract without being members.^[2]
3. The construction union membership rate rose from 10.3% in 2024 to 11.1% in 2025, and union members rose by 79,000 (8.6%).^[2]
4. The 2023 to 2025 average construction rate (10.7%) is below the 2017 to 2019 average (13.1%), so the 2025 rise does not reverse the decline of the past decade.^[5]
5. The private construction union membership rate fell from 27.5% in 1983 to 11.0% in 2025, a drop of 16.5 points or 60.0%.^[14]
6. Private construction union members fell from 1,130,700 in 1983 to 990,600 in 2025 (12.4% fewer), while private construction employment rose 119.0%.^[14]
7. The gap between construction and the private sector narrowed from 11.0 points in 1983 to 5.1 points in 2025, but construction's rate relative to the private sector rose from 1.67 to 1.86 times.^[14]
8. Construction and manufacturing had near-identical union rates in 1983 (28.0% and 27.9%); by 2025 construction was at 11.5% and manufacturing at 7.8%.^[14]
9. Construction accounted for 13.5% of all private-sector union members in 2025 but 7.2% of private-sector wage and salary employment.^[2]
10. Full-time construction union members had median weekly earnings of $1,585 in 2025, vs $1,132 for nonunion workers: a raw premium of 40.0%.^[3]
11. The raw union premium was 55.5% in construction and extraction occupations, 19.6% for all wage and salary workers and 16.7% in the private sector in 2025.^[3][1]
12. Nonunion construction median weekly earnings grew 7.7% from 2024 to 2025, vs 3.6% for union members, narrowing the raw premium from 45.6% to 40.0%.^[3]
13. Hawaii (32.1%), Illinois (26.5%), Missouri (24.1%), Ohio (24.0%) and Michigan (22.6%) had the highest pooled 2023 to 2025 construction union rates.^[13]
14. California (111,700), New York (100,700), Ohio, Michigan and Illinois held 405,400 private construction union members in 2025, 40.9% of the U.S. total.^[13]
15. Fifteen states, including Texas (2.7%) and Florida (2.0%), had pooled 2023 to 2025 construction union rates below 5%.^[13]

Frequently asked questions

Methodology

Source selection: the headline 2025 figures come from the BLS Union Members 2025 release (Tables 3, 4 and 5) and the BLS LU database, which use the Current Population Survey (CPS). The long history from 1983 and the state construction rates come from the Union Membership and Coverage Database maintained by Barry Hirsch, David Macpherson and William Even (unionstats.com), an academic compilation of CPS microdata using BLS methods. Union-reported membership comes from Form LM-2 filings in the Department of Labor OLMS yearly data files. Trade association and union statements (ABC, NABTU, ICERES) are Tier 2 and are labelled as the organizations' own statements.

Inclusion and exclusion: the construction industry rate covers private wage and salary workers in the construction industry, including office staff; self-employed workers and independent contractors are excluded. NLRB election statistics were not used because the agency's petition and election data pages did not provide a construction breakdown that could be retrieved and checked. Commentary pages that restate BLS figures without new data were excluded.

Conflicting numbers: BLS (11.1%) and unionstats.com (11.0%) differ slightly for private construction in 2025, and by up to 1.7 points in 2000 to 2002, because of weighting and industry coding. BLS values are used for 2025 headline figures and for 2000 onward; unionstats.com values are used for the 1983 to 2025 series so the full history is on one consistent basis. The unionstats.com sector series (construction 11.5% in 2025) includes government workers classified to construction and is used only for cross-sector comparisons.

Derived metrics: point and percent changes, ratios, raw earnings premiums (union median / nonunion median – 1), three-year averages, pooled state rates (sum of members / sum of employment over 2023 to 2025) and LM-2 totals are calculated in scripts/compute.py, scripts/facts.py and scripts/tables.py from the unmodified raw files. Every formula and its input rows are listed in derived_metrics.csv.

Limitations: union data come from one-quarter of the CPS sample, so industry and especially state estimates carry sampling error; BLS does not report the statistical significance of the 2024 to 2025 change in construction. 2025 figures are 11-month averages that exclude October 2025. Earnings comparisons are medians that do not control for occupation, region, age or firm size and exclude benefits. LM-2 totals are union-reported, can include retirees and non-construction or non-U.S. members, and cover fiscal years that differ by union.

Last update: 2026-10-06. All sources were opened on that date; raw files are stored in raw/ and the pipeline can be re-run with scripts/pull.sh. This report is for information only and is not legal, financial or employment advice.

This report presents published data with its limitations. It is not legal, financial or investment advice.

Sources, ranked by quality

Tier 1: primary and official sources (16)

• [1] U.S. Bureau of Labor Statistics. Union Members 2025 (news release USDL-26-0229). 2025 annual average (released 2026-02-18). Accessed 2026-10-06.
• [2] U.S. Bureau of Labor Statistics. Union Members 2025, Table 3: Union affiliation of employed wage and salary workers by occupation and industry. 2024 and 2025 annual averages. Accessed 2026-10-06.
• [3] U.S. Bureau of Labor Statistics. Union Members 2025, Table 4: Median weekly earnings of full-time wage and salary workers by union affiliation, occupation and industry. 2024 and 2025 annual averages. Accessed 2026-10-06.
• [4] U.S. Bureau of Labor Statistics. Union Members 2025, Table 5: Union affiliation of employed wage and salary workers by state. 2024 and 2025 annual averages. Accessed 2026-10-06.
• [5] U.S. Bureau of Labor Statistics. Union affiliation database (LU), flat files lu.data.1.AllData and lu.series. 1983 to 2025 (industry series from 2000). Accessed 2026-10-06.
• [6] U.S. Bureau of Labor Statistics. Union membership rate 10.0 percent in 2025 (The Economics Daily). 2000 to 2025. Accessed 2026-10-06.
• [7] U.S. Bureau of Labor Statistics. Industries at a Glance: Construction (NAICS 23). 2022 to 2025. Accessed 2026-10-06.
• [8] U.S. Bureau of Labor Statistics. Union Members Technical Note. 2025 release. Accessed 2026-10-06.
• [9] U.S. Bureau of Labor Statistics. 2025 federal government shutdown impact on the Current Population Survey. October 2025. Accessed 2026-10-06.
• [10] U.S. Bureau of Labor Statistics. Release calendar 2026 (Union Membership (Annual) for Annual 2025). 2026 schedule. Accessed 2026-10-06.
• [11] U.S. Bureau of Labor Statistics. Archived news releases: Union Members (Annual). Release dates, 2009 to 2026. Accessed 2026-10-06.
• [12] Barry T. Hirsch, David A. Macpherson and William E. Even (unionstats.com). Union Membership and Coverage Database: union membership, coverage, density and employment by industry, 1983 to 2025. 1983 to 2025 (2025 compilation completed 2026-01-23). Accessed 2026-10-06.
• [13] Barry T. Hirsch, David A. Macpherson and William E. Even (unionstats.com). Union Membership and Coverage Database: union membership, coverage, density and employment by state and sector, 1983 to 2025. 2023 to 2025 files used. Accessed 2026-10-06.
• [14] Barry T. Hirsch, David A. Macpherson and William E. Even (unionstats.com). Union Membership and Coverage Database: membership by sector and demographic group, 1973 to 2025 (private construction, private sector and industry sector series). 1983 to 2025. Accessed 2026-10-06.
• [15] Barry T. Hirsch, David A. Macpherson and William E. Even (unionstats.com). Documentation for the Union Membership and Coverage Database. 2025 compilation. Accessed 2026-10-06.
• [16] U.S. Department of Labor, Office of Labor-Management Standards. Online Public Disclosure Room, yearly data download (Form LM-2 filings, 2015 and 2025 files). Fiscal years ending 2015 and 2025 (file dated 2026-10-03). Accessed 2026-10-06.

Tier 2: trade bodies, market research, company data (6)

• [17] Associated Builders and Contractors. Nearly 9 Out of 10 of US Construction Workers Are Not Union Members (news release, 2026-02-18). 2025. Accessed 2026-10-06.
• [18] Associated Builders and Contractors. ABC: A Record-High 89.7% of US Construction Workers Are Not Union Members (news release, 2025-01-28). 2024. Accessed 2026-10-06.
• [19] North America's Building Trades Unions. About NABTU. Current. Accessed 2026-10-06.
• [20] North America's Building Trades Unions. Building Trades Union Membership in the U.S. Continues to Rise (statement, 2025-01-29). 2024. Accessed 2026-10-06.
• [21] North America's Building Trades Unions. NABTU Points to ICERES Study with More Informed Union-Density Data in Response to New BLS Report (statement, 2024-01-25). 2023 (BLS) and ICERES estimates. Accessed 2026-10-06.
• [22] Institute for Construction Employment Research. ICERES Annual Report: Union Membership in the Skilled Construction Trades, 2013 to 2022. 2013 to 2022. Accessed 2026-10-06.