Scotty Wristen ran a tight crew in Abilene, Texas, until five of his electricians left for a data center job paying $35 an hour to his $20. "Some of them were guys that I've had for eight years, five years," he told The Texas Tribune in April. "They came, they got trained, and they left to go out there." He replaced them with teenagers straight out of high school. No tools, no experience, and about five months of costly mistakes ahead.
Wristen's story is a single data point, but it landed in my inbox the same week a much larger number arrived, one drawn from the kind of dataset that forces you to stop treating anecdotes as anomalies and start treating them as symptoms.
One in five workers gone, every single year. Not fired. Gone. And the data behind that number, released March 31 in Bridgit's 2026 Construction Workforce Benchmark Report, suggests the problem is considerably worse than the headline figure implies.
Running to Stand Still
Bridgit calls it the "treadmill effect," and the math is brutal. In 2025, 71.7% of contractors grew their headcount. Sounds encouraging until you learn that 46% of contractors achieved zero net growth after attrition ate their gains. A firm targeting 100 net new hires at the median 18.7% attrition rate needs to make 125 total hires just to break even. At the uglier end of the distribution, where attrition hits 35%, that number climbs to 154 hires for every 100 you actually keep.
That is a staggering amount of onboarding, a staggering amount of lost institutional knowledge about local codes and inspector quirks and which lumber yard actually delivers on time, and a staggering number of job sites where the person finishing a task is not the person who started it and has no context for what came before.
Randstad's parallel analysis of 150 million U.S. job postings tells the same story from the demand side. Their 2026 report found that for every 100 young workers entering the trades, 102 are walking out, a net decline of 1.72% per year that compounds relentlessly because manufacturing, one of the feeder sectors for construction labor, is bleeding workers faster than it replaces them and the pipeline is not just narrow but actively running in reverse.
What This Means for Your Kitchen Remodel
Residential construction has always relied on subcontractor rotation, and that is an important caveat. Your framing crew is supposed to leave when the framing is done. Electricians, plumbers, drywallers, each arrive for their phase and move on. What matters to a homeowner is not turnover between trades but within them.
Consider a $400,000 home addition, a six-month project with a general contractor running a ten-person crew across multiple trades. At the industry median 18.7% annual attrition rate, the math says you lose roughly one worker over that timeline. At the higher end of the distribution, 35%, you lose two. On larger projects using twenty-person teams, the math predicts three to four departures mid-build.
Now factor in the "rookie ratio," another metric from the Bridgit report. Across the dataset, 36.4% of crew members have been with their contractor for less than one year. On larger teams of fifty-plus, that number hits 56%. More than half your crew might be learning the company's systems, the foreman's expectations, and each other's working rhythms while they are also learning your house.
| Crew Size | Expected Departures (6 months, 20.7% attrition) | Rookie Ratio | Workers with <1 Year Tenure |
|---|---|---|---|
| 5 (small reno) | ~0.5 | 25% | ~1 |
| 10 (major addition) | ~1 | 36% | ~4 |
| 20 (new build) | ~2 | 36% | ~7 |
| 50+ (production) | ~5 | 56% | ~28 |
Sources: Attrition prorated from Bridgit 2026 annual rate of 20.7%. Rookie ratios are direct from Bridgit's report segmented by team size. These are industry-wide contractor-level figures, not residential-specific breakouts.
Who Actually Stays
Senior project managers. Their attrition rate is 3.6%, with a median tenure of 5.6 years. Senior superintendents run similar numbers: 4.1% attrition, 7 years median tenure. These are the people who know the local code inspector by name, who can sequence a complex remodel without a Gantt chart, who have the subcontractor relationships that get your project priority scheduling instead of a "we'll fit you in" appointment three weeks from now.
Their junior counterparts turn over at more than double those rates and have median tenures of 3.7 to 3.8 years. Between a senior superintendent who has been through 200 projects and a junior one on project 15, the experience gap is not incremental but categorical, the difference between catching a framing error at rough inspection and catching it after the drywall is up, the cabinets are hung, and $22,000 of work needs to come back out.
Clark Lowe, president of O'Connor Company and a Marine Corps veteran who has watched this cycle play out for decades, put it more bluntly in a 2025 piece for the Construction Users Roundtable: "Workers don't want to be hired onto a sinking ship. They want job environments that make sense, a clear system for executing projects, safe working conditions, and managers who respect their time and input." When you keep hiring people into a system that burns them out, the new hires quit as fast as the old ones in what Lowe calls "a never-ending cycle."
It Is Not About Pay. Except When It Is.
Residential construction workers earn an average of $39.40 an hour, according to the Home Builder Institute's analysis of federal data, while broader construction averages $39.70. On paper, the gap is small, almost negligible, but it understates the real disparity because commercial and data center work offers longer project timelines, steadier employment, overtime that actually gets approved, and per diem benefits that residential work typically does not.
Edward Brady, CEO of the Home Builder Institute, told Homes.com the dynamic is simple. "If they don't have the [residential] work, they're going to leave the residential industry and go to the commercial. Then, residential construction suffers." He warned that when housing demand returns, the workforce will not be there to meet it.
Single-family housing starts hit 930,000 in April 2026, down 9% from March and 2.4% below the prior year. With a 1.5-million-home deficit and a workforce that cannot grow faster than it shrinks, the math suggests your next project will cost more and take longer than your last one did. Not because materials got expensive, though they did. Because the experienced human beings who convert materials into a house are an increasingly scarce commodity that the residential sector is losing the bidding war to retain.
What You Can Do About It
If you are planning a build or renovation, you now have specific questions to ask your general contractor. How long has your core crew been with you? What is your turnover rate over the past twelve months? Do the same subcontractors show up project after project, or are you sourcing new trades for every job? Are your lead carpenters and electricians employees or day-labor pickups?
No bid package will contain those answers. No contractor will voluntarily disclose a 30% attrition rate, but you can infer it from how they talk about their people. A builder who names his framing crew by first name, who can tell you his plumber has been on the last twelve jobs, who has a superintendent with gray hair and a reputation in the permit office, is a builder operating at the stable end of the attrition spectrum. A builder who quotes a lower number but shrugs when you ask who will actually be on site is telling you something important with that shrug.
Bridgit's CEO Mallorie Brodie framed it this way: "Most workforce decisions in construction still happen on instinct and spreadsheets. The companies pulling ahead are the ones thinking ahead." The top fifty ENR firms plan 6.8 years into the future while everyone else plans 4.7. That two-year gap in strategic horizon is the gap between a contractor who retains experienced people and one who runs on the treadmill, hiring and losing at the same speed, indefinitely.
Limitations
Bridgit's data comes primarily from large ENR-ranked contractors. How closely it maps to a five-person residential framing outfit in suburban Phoenix is genuinely uncertain. Residential-specific attrition rates are not broken out in the report, and smaller firms may run higher or lower depending on local conditions. Subcontractor rotation is also structurally different from employee attrition. A framing crew leaving after framing is complete is not turnover. A framing crew member quitting mid-project for a data center job is. Bridgit's figures capture both, and it is difficult to separate the signal. As a planning tool, the rookie ratio benchmark is useful but not a direct measure of quality. A first-year worker with strong mentoring can outperform a ten-year veteran coasting on autopilot. Finally, the correlation between attrition and project defect rates remains unstudied at scale. Logically the connection is strong, but the empirical evidence connecting specific turnover metrics to specific punch-list lengths does not yet exist in published research.
The Strongest Case Against Worrying
Residential construction has always been transient, a world where subs move between builders and journeymen chase the next job while seasonal swings send workers home every winter and scatter them across new crews every spring. A builder in 1995 could have told you the same story about crew instability, and houses still got built. Bridgit's numbers capture a chronic condition, not a new crisis, and chronic conditions can persist for decades without collapsing the system. Perhaps a 20% attrition rate is simply what construction looks like in a labor market that has always treated its workers as disposable, and the houses get built anyway, on budget and on time, often enough to keep the industry running.
That argument has real force, and it deserves to be stated at full strength. It just happens to depend on a housing market where demand was closer to supply, labor was less geographically concentrated in data center corridors, and the experienced workforce was not retiring at a rate of 20,000 electricians per year with a pipeline running at negative replacement. But the conditions that made 20% attrition tolerable in 1995 are not the conditions of 2026.