Manufacturing Costs

Factory Orders Rose 9.95%. Physical Output Rose 1.27%.

July 2025 to July 2026 orders rose 9.95%, physical output 1.27%, and utilization 0.09 percentage points. Matched May to July averages show orders up 6.60% and output 1.50%.

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

Manufacturers' new orders were 9.95% higher in July 2026 than in July 2025. Manufacturing industrial production was only 1.27% higher, while capacity utilization increased by 0.09 percentage points. The three measures point to different magnitudes of change because they measure different things.

For a September 2026 planning meeting, that distinction is more useful than forcing them into one verdict about a factory boom. Orders record nominal bookings. Industrial production tracks physical output. Utilization compares output with estimated capacity. The gap between their growth rates does not automatically measure inflation, unfilled work or excess demand.

The order book grew faster than the production index

Manufacturers' new orders increased from $603.586 billion in July 2025 to $663.616 billion in July 2026, based on a source reported in millions of dollars. The nominal difference was $60.030 billion.

The manufacturing industrial-production index moved from 98.07 to 99.31 when displayed to two decimals. Its 1.27% growth calculation uses the unrounded source observations. The much smaller increase is not evidence that one series is wrong; it reflects a different measurement target.

Subtracting 1.27% from 9.95% produces an 8.68-percentage-point difference in growth rates. That arithmetic can summarize the contrast. Calling it an inflation rate would require a compatible price-and-quantity framework that these two aggregates do not provide. Calling it a backlog would require data on unfilled orders or another direct measure of work awaiting production.

Utilization supplies a third, quieter reading

Manufacturing capacity utilization was 75.87% in July 2025 and 75.96% in July 2026. The unrounded increase was 0.0914 percentage points, displayed as 0.09. It should not be presented as a 0.09% growth rate.

The small change indicates that output relative to estimated capacity changed little between those endpoints. It does not prove every manufacturing industry had ample spare capacity or that every supplier could absorb another order without difficulty.

Utilization is an aggregate estimate, and an operation can encounter a bottleneck in labor, tooling or one process while other capacity remains available. A national measure cannot identify the lead time for an individual product. Its contribution here is to keep a large nominal-orders headline from standing alone as a complete description of the production environment.

Average three months and the contrast remains

Monthly orders can move substantially with the timing and composition of bookings. To test whether the July comparison was doing all the work, we compared the average of May, June and July 2026 with the same three months of 2025.

Nominal orders grew 6.60% on that basis, rather than July's 9.95%. Manufacturing industrial production grew 1.50%, rather than 1.27%. The magnitude of the gap narrows, but orders still increased more rapidly than the physical-output index.

Average utilization moved from 75.61% to 75.86%, an increase of about 0.25 percentage points. This check is descriptive and contains only three months in each year. It does not remove all product-mix or timing effects. It shows that the difference is not confined entirely to the two July observations chosen for the opening comparison.

A booking is not a unit leaving the factory

An order can precede production, and different products can have different timing between agreement and delivery. Nominal order values can also change with price and product mix. Those features make the series relevant to manufacturing activity without making it interchangeable with current physical output.

The scope and weights of the orders and production measures differ as well. Their growth rates cannot be decomposed mechanically into one common volume component plus one residual price component. A correct explanation needs to preserve those measurement boundaries.

Several plausible mechanisms could contribute to the observed gap: a shift toward higher-value products, price changes, the timing of large bookings or work scheduled for later production. This analysis does not identify their separate contributions. Listing them as possible explanations is appropriate; selecting one as the measured cause would require additional evidence.

The missing evidence determines the next investigation

To argue that backlogs expanded, inspect a compatible unfilled-orders measure and its industry detail. To argue that the difference mainly reflects price changes, use a suitable deflator with scope and weights aligned to the nominal measure. A steel index cannot deflate every manufacturing order.

To connect the figures to supplier lead times, bring in delivery commitments or an appropriate survey rather than assume that the nominal-orders change predicts them. To understand a particular plant, compare its own booked work, scheduled production and demonstrated capacity on consistent definitions.

Those additional questions are not a reason to discard the original comparison. They are its value. The contrast identifies where a simple headline is insufficient and shows exactly which evidence would distinguish competing interpretations before a September investment or purchasing decision.

Put all three readings in the planning document

The new-orders history, manufacturing production history and utilization history should appear with their units and common July endpoint visible.

A reader can then see that the nominal booking value changed more than physical output, while utilization changed little. The three-month check adds context by showing that the difference remains when a single month's timing has less influence.

The retrospective conclusion is specific enough to guide the next conversation: the order-value measure rose faster than the production measure in the information available by September 9. It does not justify labeling the difference inflation or backlog. September planning becomes stronger when the competing measures are reconciled, rather than when one is chosen to tell the most dramatic story.

Sources and calculation

Sources are Census/FRED manufacturers' new orders AMTMNO, Federal Reserve manufacturing production IPMAN and manufacturing utilization MCUMFN. The article uses the September 9 data version, including July production 99.314 and utilization 75.9621 before later revisions. Annual comparisons use matched months; the sensitivity check uses three-month averages. Growth differences are not deflators or backlog estimates.

Sources and evidence

Evidence period: July 2025 to July 2026; May to July matched-quarter sensitivity. 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/AMTMNO

fred.stlouisfed.org/series/IPMAN

fred.stlouisfed.org/series/MCUMFN

Published 2026-09-29.