Workforce and Labor

Manufacturing Job Openings Have Collapsed. Wages Never Got the Message.

Job openings are roughly half what they were at the 2022 peak, which reads as a brutal cooling. Set them against the two decades before the pandemic and they are still far above normal. Wages, meanwhile, never had a down year. The labor market did not cool so much as return to a level that used to be called red hot.

Read the last four years alone and manufacturing looks like a labor market in freefall: job openings stand at 481 thousand (Jun 2026), about 53% below the April 2022 peak of 1,025 thousand. Now read the whole 26-year record and the story inverts. Openings averaged about 298 thousand across the two decades before 2020. Today's reading is roughly 62% ABOVE that norm. The collapse was real, and it was a collapse from an unprecedented bubble back toward a level that used to count as a tight market. That distinction is everything, and it explains the thing that has confused every forecaster: wages never flinched. Factory earnings at $30.35/hour have risen in every year of the record.

The cooling everyone predicted did happen

Commentators spent years forecasting a cooldown, then declared victory when openings halved. Both the forecast and the victory lap used the 2022 peak as the baseline, which is the single most misleading reference point available. 2022 was not normal; it was the tightest labor market in the history of this series, an artifact of reopening demand colliding with a shrunken workforce. Measuring the descent from a record is how you turn a normalization into a crisis. The honest baseline is the pre-pandemic decade, and against that baseline employers are still advertising more factory jobs than they did in almost any year of the 2000s or 2010s.

Why wages ignored it

Once you use the right baseline, the wage behavior stops being a puzzle. Wages never fell because the market never actually got loose; it got less absurd. Manufacturing wages are set by the replacement cost of a trained operator, not by the marginal job posting, and that cost did not fall when postings came off the boil. The people who can run a press brake, hold a tolerance, or program a cell remain scarce, because the constraint is the skill, not the vacancy count. Factory earnings are 178% higher than when the wage record begins in 1990, and not one of those 37 years was a down year.

Openings measure how many workers employers want. Wages measure what the ones they already have are worth. This cycle drove those two numbers in opposite directions, and almost every labor forecast assumed they could not diverge.

The churn numbers, with the caveat they deserve

The hires rate is 2.60% of employment and the quits rate 1.50% of employment (Jun 2026), and both sit high within the window we hold for them. I will not call those records, because unlike openings and earnings these two series carry only about a year of history in our archive, so their "highs" are a statement about recent months and nothing more. Read narrowly, they still say something useful: workers inside manufacturing are confident enough to move. A workforce that keeps quitting voluntarily while employers cut back on postings is a workforce that has not lost its leverage, even as the demand for new hires fell away.

What this means for your staffing plan

Stop waiting for a slack labor market to solve your labor cost, because on this record it has not existed since before the pandemic and it is not arriving now. Plan on paying market for the crew you need, and put the effort into retention, cross-training, and automating the specific tasks where the scarce skill is the bottleneck, because those are the levers that actually respond. Use the openings series as a demand signal rather than a wage forecast: the descent from 2022 is genuine information about how aggressively your competitors are expanding, and it argues for care on capacity commitments. It says nothing at all about what the next machinist will cost you.

Openings against the whole record, not just the peak

The 26-year record shows job openings making a full round trip, which is why point-in-time comparisons mislead so badly here. Their high came at the close of 2021 around 877, gave way over the following years to 401 by the end of 2024, and has climbed since to 481. That leaves it 45% below the peak and well off the floor, so whether today looks high or low depends entirely on which year you anchored to.

Use the labor cost calculator to model what churn and retention do to your annual labor line. Price your turnover

Published 2026-08-06.