Manufacturing Costs
Factory Openings Rose 35.51%. Hiring and Quits Rates Stayed Flat.
July 2025 to July 2026 openings increased from 428,000 to 580,000, while hiring stayed 2.30%, quits 1.40%, and CES employment moved from 12.625m to 12.622m. July openings remained 43.41% below the April 2022 archive peak.
Editorial evidence cutoff: September 9, 2026. Published September 29, 2026. Observation periods are stated throughout; older figures are retrospective evidence.
Manufacturing job openings increased from 428,000 in July 2025 to 580,000 in July 2026, a rise of 35.51%. Over the same comparison, the hiring rate stayed at 2.30% and the quits rate at 1.40%. Manufacturing payroll employment was almost unchanged.
For a September 2026 staffing review, the combination is more useful than a simple declaration that the labor market is heating or cooling. It shows more unfilled positions at the selected month-end without a corresponding change in two published turnover rates. It does not identify why those openings remained unfilled.
More vacancies did not mean a higher hiring rate
The increase in openings was 152,000 positions between the matched July observations. June 2026 openings had been 501,000, so the July count was also higher than the immediately preceding month. The annual increase is not merely a result of comparing a strong July with a weak latest month.
The hiring rate, however, was 2.30% in both July 2025 and July 2026. That rate describes hires during a month relative to the relevant employment base. It is not the fraction of the reported month-end vacancies filled that month.
Holding those definitions apart prevents an appealing but invalid calculation. Dividing 580,000 openings by 2.30% does not produce a time to fill or a recruiting backlog measured in months. It combines a count and a rate with different denominators and timing. The apparent precision would conceal a measurement mistake.
A vacancy is a stock; hiring is a flow
BLS measures job openings at the last business day of the month. Hiring covers additions to payroll throughout the month. A position can be filled before the final day and therefore contribute to hires without appearing in the closing vacancy count.
Openings also have a specific survey definition. The position must exist, work must be available, it could begin within the specified window and the establishment must be recruiting externally. The statistic is not simply a count of every online advertisement or an unrestricted measure of an employer's future wishes.
These distinctions help explain why the series can move differently without contradicting each other. They do not explain the actual July outcome by themselves. Establishing whether the change reflected expansion, harder recruiting, turnover or another mechanism requires evidence about the positions and employers involved.
Stable quits do not identify worker confidence
The manufacturing quits rate remained 1.40% in the two July observations. It records voluntary separations within the survey's definition. It does not directly measure how optimistic workers felt, the quality of outside offers or their reasons for leaving or staying.
A lower or unchanged quits rate can be consistent with several different circumstances. The aggregate alone cannot distinguish satisfaction, limited alternatives, workforce composition or the timing of separations. Those explanations need supporting evidence before being presented as the reason for the rate.
For September planning, the useful contrast is therefore descriptive: openings were higher, while the published hiring and quits rates were unchanged. That combination justifies closer examination of recruiting and retention information. It does not justify assigning a single psychological explanation to the entire manufacturing workforce.
Payroll employment adds context, not an accounting identity
Manufacturing payroll employment was 12.625 million in July 2025 and 12.622 million in July 2026, a difference of 3,000 jobs, or about 0.02%. Relative to the size of the employment base, that is a small net change.
The payroll estimate comes from the Current Employment Statistics program, while openings and turnover come from JOLTS. Differences in survey design, estimation and timing matter. The figures should be shown as related evidence, not forced into a simple identity in which one published count must exactly explain the other.
Nor does nearly unchanged net employment mean little hiring occurred. Additions and separations can offset each other. A factory can experience substantial recruiting work while its headcount changes little. The national endpoints do not disclose the gross movement at any particular establishment or the skill mix behind it.
Looking back to 2022 changes the headline again
The archived manufacturing openings peak was 1.025 million in April 2022. July 2026's 580,000 was about 43.41% below that high, even though it was 35.51% above July 2025. A large decline from an exceptional earlier peak can coexist with a substantial rise over the latest year.
That retrospective comparison is useful because both narratives can be numerically correct. Calling the market a collapse focuses on the old maximum. Calling it a surge focuses on the more recent starting point. Neither reference alone settles what a September employer should expect when recruiting a specific role.
The older peak should not be treated as normal, and the latest annual increase should not be projected forward automatically. Displaying the path with both reference dates lets the reader assess the change without making a selected baseline carry an unsupported forecast.
The next evidence belongs closer to the vacancy
A plant deciding how to staff in September needs details such as the occupations sought, offers made, acceptance rates, time spent in each recruiting stage and whether vacancies reflect new positions or replacement needs. Those measures can distinguish problems the national aggregate cannot.
The openings history, hiring-rate history and quits-rate history provide context for that review. Their different timing and units should remain visible when they are presented together.
The measured conclusion is narrow but valuable. The July vacancy stock was markedly higher than a year earlier, while two turnover rates and net payroll employment changed little. It identifies a mismatch worth investigating. It does not establish a labor shortage, an imminent hiring wave or a managerial reluctance story without evidence beyond those aggregates.
Sources and calculation
The analysis uses manufacturing JOLTS openings JTS3000JOL, JOLTS hiring and quits rates, and CES manufacturing employment, in the version available September 9. BLS definitions distinguish month-end openings from monthly flows. The release schedule places July JOLTS on September 1. Counts are seasonally adjusted estimates and subject to revision; no statistical-significance test or time-to-fill estimate is claimed.
Sources and evidence
Evidence period: July 2025 to July 2026; April 2022 historical reference. The frozen evidence record lists the source files and verified hashes available September 9, 2026. Source revision: 5e4fb7726c3d40060c0151c086c903baae856cab. Later live-data updates do not alter the historical evidence in this article.
fred.stlouisfed.org/series/JTS3000JOL
Published 2026-09-29.