US Industrial Production July 2026 — Up 0.2%

The Fed's July G.17 painted a two-speed picture: total output rose 0.2% for a second straight monthly gain, but the growth was carried by business equipment and defense and space-related production while both durable and nondurable consumer goods output contracted.^[1][2][3]

At 103.0 (2017=100), total industrial production was 1.1% above its year-earlier level. Capacity utilization ticked up to 76.3% for total industry and 76.0% for manufacturing, both still well below long-run averages — slack that argues against industry-driven inflation pressure even as output grinds higher.^[1][3]

Industrial production is revised more than most readers expect, and the reason is structural rather than a sign of sloppiness. The Federal Reserve explains that for the first estimate of a month's output, «about 78 percent of the source data (in value-added terms) are available». The rest is estimated and replaced later.^[4]

Coverage then fills in month by month — «86 percent for estimates in the second month… 92 percent in the third month, 95 percent in the fourth month, 98 percent in the fifth» — and each figure «is subject to revision in each of the subsequent five months». So a headline published today is a five-month work in progress, not a final reading.^[4]

The practical consequence is that small monthly moves should be held loosely. The Fed publishes its own reliability estimates showing the likely range after the final revision, and for a month reported as a modest gain, that range can extend into negative territory. A reported +0.2% is not the same as a confirmed increase.^[4]

The July index reading of 103.0 is stated against a 2017 base — that is, 2017 average output = 100. That anchor is about to move. The Federal Reserve has said it plans to publish its annual revision in the autumn of 2026, and that «the base year for the revised indexes will be 2022».^[1][4]

When that happens, the same underlying production will be expressed as a different index number, because the denominator changes. Growth rates are largely preserved but levels are not comparable across the switch. Anyone comparing a pre-revision figure with a post-revision one will see a discontinuity that has nothing to do with factories.^[4]

The revision also reaches backwards. The Fed notes that method changes «will affect the index from 1972 to the present», so the historical series is restated as well — which is why an old chart and a new one can disagree about the same past month.^[4]

July capacity utilization of 76.3% was reported as 3.1 percentage points below its long-run average, where the long-run window is 1972–2025. That window is extended each year, so the benchmark itself shifts slightly over time — a comparison against 'the long-run average' is not against a fixed number.^[1][4]

The Fed's stated design goal for these rates is consistency over time, so that «a rate of 85 percent means about the same degree of tightness that it meant in the past». Utilization is therefore a measure of slack in installed capacity — how tight production is running — rather than a verdict on the business cycle. The G.17 release reports the level and the gap and stops there; it does not characterise the economy, and readings below the long-run average are common outside recessions.^[4]

The Federal Reserve publishes G.17 release dates for the year in advance. The August 2026 data were released on 18 September 2026 as scheduled, following the July release on August 18 — see the section below for what they showed. Monthly releases land around the middle of the following month.^[4]

This page covers July. The Federal Reserve released the August figures on 18 September 2026 (9:15 a.m. EDT). Industrial production was unchanged in August after the 0.2 percent rise in July, and manufacturing output decreased 0.3 percent. Mining "ticked up" 0.1 percent and utilities rose 1.8 percent.^[5]

The total index moved from 103.0 in July to 103.1 in August, and output was 1.4 percent above its year-earlier level. Capacity utilization was unchanged at 76.3 percent — still 3.1 percentage points below the 1972-2025 average, so the slack described on this page did not close.^[5]

Note what changed and what did not: the July gain came from capital spending rather than household demand, and in August the headline held flat while manufacturing itself fell. The figures on this page for July are unchanged and are kept as published.^[5]