Data guideRead at source, 9 September 2026

The Skilled Trades Shortage, Measured

Pulled from the Bureau of Labor Statistics today, the numbers do not say what the headlines say. Construction job openings are running at a 3.1% rate so far in 2026 — below the all-industry rate of 4.4%, and well below construction's own 4.9% peak in 2022. By that measure the trades are a looser market right now than the economy as a whole. What is real is the replacement arithmetic. Across seven trades, BLS projects about 376,500 openings a year through 2035, and between 85% and 97% of them come from replacing people who leave the occupation rather than from growth. The shortage is a pipeline and retention problem wearing a demand problem's clothes, and that changes what a recruiting budget should be doing.

Claim one: employers cannot fill the openings

The job openings rate is the cleanest available measure of unmet demand, and for construction it has been falling for four years.

Three things worth reading off that table. Construction's openings rate has been below the all-industry rate in every year since 2019, which is the opposite of what a uniquely acute shortage would look like. The hires rate has fallen every year since 2020, from 5.8% to 3.9%, so employers are not just posting less — they are hiring less. And the quits rate at 1.7% is the lowest in the series, which usually means workers see fewer places to go.

Manufacturing tells a similar story: an openings rate of 3.8% so far in 2026 against 6.2% in 2021, with hires down from 3.6% to 2.4% over the same period.

Claim two: it is a retirement problem

This one holds up, and it is the number that should drive the plan. BLS projects annual openings for each occupation over the 2025–35 decade, and separately projects how much employment will grow. The difference between the two is replacement: people transferring to another occupation or leaving the labour force.

Employment and median annual wage are the BLS May 2025 estimates. Openings a year and employment change are from the Occupational Outlook Handbook for 2025–35; growth per year is the projected ten-year change divided by ten, and replacement share is one minus growth over openings. Note that the Handbook's employment base differs slightly from the wage survey's, because they are different surveys with different reference periods — we have shown each as published rather than reconciling them.

Read the right-hand column and the recruiting problem changes shape. For welders, 97% of the roughly 40,300 annual openings exist because someone left, not because the work grew. Growth accounts for about a thousand jobs a year across an occupation of more than four hundred thousand people. You are not competing for a share of an expanding market. You are replacing your own leavers and someone else's, every year, forever.

Work out your own replacement load

Enter your headcount in a trade and your annual turnover. This applies the same arithmetic to your business, and puts the national projection next to it for scale.

Claim three: young people will not do the work

We have no primary data of our own on attitudes, and we are not going to borrow someone's survey and present it as fact. What the federal data does show is the entry requirement, and it is the strongest recruiting argument in the sector: for electricians, plumbers, carpenters, welders and industrial machinery mechanics, the typical entry-level education recorded by BLS is a high school diploma or equivalent. For HVAC and automotive technicians it is a postsecondary non-degree award. None of these seven occupations requires a degree, and the median pay in five of them is above $60,000 a year.

That is a genuinely good offer, and most trade job ads do not make it. They open with years of experience and a certification requirement, which is exactly the wrong end of a replacement problem. The skills-based hiring guide covers what to swap those for.

What the arithmetic implies for a recruiting budget

• Retention is a recruiting line item. If nine in ten openings are replacements, a point of turnover avoided is worth more than a point of applicant conversion gained, and it is usually cheaper. Work out what a leaver costs you in advertising alone before deciding the budget belongs at the top of the funnel.
• Build the entry route rather than bidding for the finished worker. Everyone in your market is chasing the same experienced journeyman, which is why that campaign is expensive. The pipeline nobody is competing for is the person who would start an apprenticeship if the ad told them the pay and the path.
• Advertise continuously, not in bursts. A replacement problem is a steady-state problem. Campaigns that switch on when a resignation lands pay a learning cost every time; a steady low-level presence in the market does not.
• Price the trade, not the sector. Median pay across these seven ranges from $50,620 to $64,520, and the pay you post is the single biggest determinant of who responds.

What it costs to reach them

Across the 891 Boostpoint-managed Meta campaigns in our 2026 Social Job Advertising Benchmark, the median cost per applicant was $13.88 and the volume-weighted blended figure across 1,334 campaign-months was $8.02. Skilled trades roles sit above the median and below the licensed-clinical end: they are not the cheapest applicants we buy, and they are nowhere near the most expensive.

Put that next to the replacement load. A shop of forty electricians losing a quarter of its people a year is hiring ten a year to stand still. At a realistic cost per applicant and a realistic applicant-to-hire ratio, that is a known, budgetable number rather than a crisis — which is the point of doing the arithmetic instead of reading the headline.

Trade-specific pages: construction labor shortage, welder shortage and recruiting, electrician recruiting and auto technician shortage. For the all-industry picture rather than the trades one, see labor shortage.

Frequently asked questions