Workforce and Labor
Factories Are Maxing Out Hours Instead of Hiring. That's a Tell.
When demand rises, a plant can add hours or add people, and which one it chooses reveals what it believes. Right now factories are pushing hours toward a range high while holding back on hiring, and that choice is a signal worth decoding.
There are two ways to make more product: work the people you have harder, or hire more people. The choice between them is one of the more revealing decisions a manufacturer makes, because adding hours is fast and reversible while adding headcount is a bet on durable demand. Right now the data shows factories leaning hard on the first lever: average weekly hours sit at 41.7 hours/week (Jul 2026), up about 1.5% from a year ago, near the top of their archived range, with overtime at 4.1 hours/week per week. Hours are stretched. The question is why plants are stretching them instead of hiring, and the answer changes the read entirely.
Two stories, opposite conclusions
Maxed-out hours are consistent with two very different worlds, and telling them apart is the whole game. In the bullish version, demand is strong and durable, the plant is capacity-constrained, and the stretched workweek is the last step before it commits to hiring or capital investment, a leading sign of expansion. In the bearish version, demand is decent but management does not trust it, so it squeezes overtime out of the existing crew rather than take on the fixed cost and firing risk of new hires, a defensive, late-cycle crouch. Same hours number, opposite meaning, and the surrounding series are what break the tie.
- Weekly hours (Jul 2026): 41.7 hours/week
- Quits rate (Jun 2026): 1.50% of employment
- Job openings (Jun 2026): 481 thousands
The tie-breakers: quits and openings
To tell the bullish stretch from the bearish crouch, read the labor market around the hours. The quits rate, at 1.50% of employment (up about 7.1% from a year ago), is the confidence vote: a high quits rate says workers believe better jobs exist, consistent with a hot, expanding market where the hours stretch is a prelude to hiring. Job openings, at 481 thousands (up about 23.0% from a year ago), say whether firms are even trying to add people. High hours plus high openings plus a firm quits rate is the expansion story; high hours plus falling openings and a sinking quits rate is the defensive one. Employment itself, at 12,611 thousands of employees, is the slow confirmation that lags both.
A stretched workweek is a question, not an answer. Whether it means expansion or anxiety is written in the quits rate, not the hours.
What it means to plan around
For a manager, the hours signal is a prompt to make the hire-or-invest decision consciously rather than by default. If the surrounding data says the demand is durable, the stretched workweek is expensive insurance, overtime premiums and fatigue, that a second shift or a machine would relieve, and the math usually favors committing once hours are pinned at a high. If the data says demand is shaky, holding off on fixed labor cost is the prudent crouch. Either way, hours at a range high are the signal that the decision can no longer be deferred, and the labor-market series tell you which way to jump.
The workweek across the whole record
- 1939: 38.4 (Archive begins 1939; selected years shown)
- 1943: 44.6 (The wartime peak, the highest in the record)
- 1944: 45.4
- 1949: 39.4
- 1960: 38.4
- 1971: 40.2
- 1982: 39.0
- 1993: 41.4
- 2004: 40.6
- 2015: 41.8
- 2026 (latest): 41.7 (Firm, and squarely mid-range over the full record)
The 88-year record shows the factory workweek making a full round trip, which is why point-in-time comparisons mislead so badly here. Its high came at the close of 1944 around 45.4, gave way over the following years to 38.4 by the end of 1960, and has climbed since to 41.7. That leaves it 8% 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 capacity planning calculator to test whether stretched hours justify a second shift or a capital purchase. Model the decision
Published 2026-08-06.