Manufacturing Costs

Hourly factory pay rose 4.36%. Weekly earnings rose 6.15%. The workweek explains the difference.

August 2025 to August 2026 hourly earnings rose 4.36%, hours 1.71%, and their product 6.15%. The weekly dollar change decomposes into $52.07 hourly earnings, $20.37 hours and $0.89 interaction.

Editorial evidence cutoff: September 9, 2026. Published September 29, 2026. Observation periods are stated throughout; older figures are retrospective evidence.

Manufacturing production-worker hourly earnings rose from $29.10 in August 2025 to $30.37 in August 2026, a 4.36% increase. Multiply those rates by the corresponding average workweeks, however, and implied weekly earnings rise from $1,193.10 to $1,266.43, an increase of 6.15%.

The difference is not an unexplained payroll surcharge. Average weekly hours increased from 41.00 to 41.70. For a September labor review, that distinction matters because an hourly-rate headline alone does not describe how much the weekly earnings benchmark changed.

The multiplication has to use the same workers

Both components cover manufacturing production and nonsupervisory employees. The hourly series is CES3000000008; weekly hours are CES3000000007, also available as AWHMAN. Both are monthly, seasonally adjusted measures for the same population.

That alignment is essential. Multiplying hours for one employee group by earnings for another can produce a number that resembles a weekly wage without representing the published population. A site's familiar URL or a shortened chart label cannot change the underlying survey definition.

The calculation here is straightforward: $29.10 multiplied by 41.00 gives $1,193.10. The later $30.37 multiplied by 41.70 gives $1,266.429, displayed as $1,266.43. BLS defines average weekly earnings using that multiplication. The analysis preserves unrounded inputs until the final presentation.

Of the extra $73.33, about $21.26 came from the longer week

The weekly increase can be separated into three parts. Holding the workweek at its earlier 41.00 hours, the $1.27 increase in hourly earnings adds $52.07 per week. Holding the hourly figure at $29.10, the additional 0.70 hours adds $20.37.

There is also an interaction: the extra 0.70 hours is paid at the higher hourly average. That contributes another $0.889, displayed as $0.89. Together, $52.07, $20.37 and $0.889 equal the unrounded $73.329 weekly change.

The decomposition is accounting, not a causal model. It identifies how the product of two averages changed. It does not establish that employers chose longer schedules because wages rose, or that the same workers received each component. The result is still useful: approximately $21.26 of the weekly difference reflects the hours change and its interaction with the hourly change.

A 6.15% increase is not everyone's raise

These are aggregate estimates. Average earnings can change when the mix of workers, establishments or industries changes, even if an individual worker's rate stays the same. Average hours can shift for similar reasons.

The weekly comparison therefore does not mean a typical continuing employee received a 6.15% pay increase. It also does not identify how the distribution changed across shifts, occupations or wage levels. Answering those questions requires employee-level or appropriately detailed evidence.

The figures are gross earnings measures. They are not take-home pay after deductions, and they do not represent the fully burdened employer cost of labor. A September budget that needs benefits, payroll taxes, training or other employer expenses must add the relevant evidence rather than relabel this earnings calculation as total compensation.

Do not add an overtime premium a second time

Average hourly earnings already reflect the compensation captured by the survey, including the effects of premium pay within its definition. Adding a separate assumed overtime premium to the published average without a proper reconciliation can double-count part of the earnings.

The workweek also should not be read as a schedule shared by every production employee. An average of 41.70 hours does not establish that each worker logged 1.70 overtime hours. BLS's overtime measure concerns hours paid at a premium, which is a separate definition.

For a plant-level estimate, the clean approach is to use its actual staffing and pay rules. The national weekly calculation can provide context, but it should not replace the payroll structure needed to distinguish regular hours, premium hours and different employee groups within that operation.

Weekly payroll and cost per good unit answer different questions

A higher weekly earnings benchmark does not automatically mean labor cost per finished unit increased by the same percentage. Output, yield, rework, downtime and the number of workers involved can all change the denominator of a unit-cost calculation.

The present comparison contains no matching plant-output measure. It cannot determine whether the extra hours produced more good units, absorbed inefficiency or supported a different product mix. Inferring a productivity result from the earnings multiplication would introduce information that the calculation does not contain.

That distinction is particularly relevant when a supplier cites national wage growth in a price review. The benchmark may help frame the discussion, but the exposure depends on the labor share of the supplied item and the actual production process. A weekly earnings increase alone does not establish the appropriate percentage adjustment to the entire selling price.

Use the bridge in the September conversation

The hourly earnings history and weekly-hours history let readers inspect the matched observations. The earnings route retains older wording; this analysis uses the production-and-nonsupervisory population defined by the source.

A useful labor review can put the two rates and the three-part weekly bridge beside the plant's own payroll change. If the local experience differs, the next questions become concrete: did hours, occupational mix, premium pay or staffing change? That is more informative than forcing every difference into the hourly-rate headline.

The retrospective conclusion is clear. From August 2025 to August 2026, the hourly benchmark rose 4.36% and the weekly benchmark 6.15%. The longer workweek accounts for the extra growth in the multiplication. That explains a measurement difference without pretending to explain every worker's pay or every factory's unit cost.

Sources and calculation

The source series are BLS hourly earnings CES3000000008 and weekly hours AWHMAN. BLS calculation methods define weekly earnings as their product. The 2026 release schedule places August payroll data on September 4. August estimates were preliminary and the analysis uses the version available September 9.

Sources and evidence

Evidence period: August 2025 to August 2026; same manufacturing production/nonsupervisory population. 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/CES3000000008

fred.stlouisfed.org/series/AWHMAN

bls.gov/opub/hom/ces/calculation.htm

Published 2026-09-29.