Sean Daily

Updated: October 18, 2026. Based on the Federal Reserve Board’s G.17 Industrial Production and Capacity Utilization release of October 16, 2026, and coverage from October 16–17. September figures are preliminary and will be revised; check the official sources below.

〰

US industrial production September 2026: the numbers

If you only saw the headline on Friday, October 16, you saw a strong month. US industrial production for September 2026 rose 0.7%, according to the Federal Reserve Board’s G.17 release, lifting the index to 104.0 (2017 = 100). Economists had expected 0.4%, First Trust noted, and it was the sixth monthly gain in a row.

Earlier months also looked a little better. The Fed revised August to a 0.1% gain from flat, and raised July’s index level as well. Output was 2.2% higher than in September 2025, and for the third quarter as a whole it grew at a 2.8% annual rate, down from 4.8% in the second quarter.

Now look underneath. Manufacturing managed only 0.3%. The bigger swing came from utilities, and the weakest spot was the auto industry.

Measure	Sept. 2026	Change from August	Change from Sept. 2025
Total industrial production (index)	104.0	+0.7%	+2.2%
Manufacturing (index)	98.4	+0.3%	+1.1%
Mining (index)	124.5	+0.2%	+2.8%
Utilities (index)	117.3	+3.5%	+8.7%
Motor vehicles & parts	—	−3.0%	−0.8%
Selected high-tech industries	—	+1.5%	+17.5%
Business equipment (index)	101.0	0.0%	+6.7%
Consumer goods (index)	98.1	+0.7%	−0.1%
Capacity utilization, total	76.9%	from 76.4%	+1.2 pts
Capacity utilization, manufacturing	75.9%	from 75.8%	+1.0 pt

〰

Utilities and chips carried the month

Utility output jumped 3.5% in September, with gains at electric utilities, the Fed said. Over the third quarter it grew at a 14.8% annual rate. First Trust points out that utility output has been climbing since 2023 after almost two decades of little growth, and ties that to the power demands of data centers.

The same story shows up inside manufacturing. The Fed’s selected high-tech industries rose 1.5% in the month, and semiconductors and related components rose 2.1%. In its quarterly summary, the Fed said durable goods output grew at a 3.8% annual rate in the third quarter, driven in part by a 28.4% gain in semiconductors and related electronic components. Nondurable manufacturing, by contrast, fell at a 1.6% rate.

〰

The side story: autos fell, and stocks fell the same day

Not everything followed the data-center trend. Motor vehicles and parts dropped 3.0%, and vehicle assembly alone fell 5.4%. First Trust calculates that auto output has shrunk at a 15.1% annualized pace over the past three months, and describes the sector as volatile and exposed to tariffs. Strip autos out and manufacturing rose 0.6%.

Here is the odd part. On the day the Fed showed chip and power output booming, Wall Street sold off tech. The Rio Times reported that the S&P 500 fell 1.09% on October 16 and the Nasdaq about 2.1%, as bond yields for data-center operator CoreWeave rose to about 15% that week. Investors were asking whether heavily indebted AI infrastructure companies can keep repaying debt while rates stay high. Strong factory data and nervous credit markets can both be true at once, and the next few months will show which matters more.

〰

Why the Fed is watching this report

For the Federal Reserve, the most useful number may be capacity utilization. It rose to 76.9%, above the 76.5% that roic.ai says analysts expected, but still well below the long-run average of 79.4%. Manufacturing utilization of 75.9% also sits under its 78.2% average. In plain terms, factories have room to produce more before running hot.

That matters because inflation is already high. The Rio Times summarized the September consumer price index at 3.7%, and roic.ai noted September producer prices were up 6.3% from a year earlier. A fixed-income analyst quoted by roic.ai read the report as growth holding up rather than price pressure building.

Markets are not expecting a move this month. Prediction-market odds compiled by the Rio Times late on October 16 put the chance of no change at the October 27–28 meeting at 86.5%, while a quarter-point hike in December was priced at 76.5%. Those are bets, not decisions, and Kiplinger notes Fed officials are in their quiet period from October 17 to 29, so no fresh guidance is coming before the vote.

〰

What this means for you

You can’t vote at the Fed, but you can decide how exposed you are to its next step. Each choice below has a cost:

• AI and chip stocks: The output data supports the build-out story. The bond market is questioning how it gets financed. Adding exposure bets on the first; trimming it guards against the second but may mean missing further gains.
• Variable-rate debt: If December brings a hike, credit card and home equity rates would likely follow. Paying debt down now cuts that risk but shrinks your cash cushion.
• Electricity bills: Rising utility output reflects heavy power demand. Locking in a fixed-rate energy plan, where available, adds certainty but can cost more if prices ease.
• Buying a car: Auto production is falling. Waiting could bring better deals if demand also cools, or tighter inventory if it doesn’t.

One more thing to keep in mind: the Fed’s annual revision to industrial production is scheduled for November 24, 2026, according to roic.ai. September’s strong number could look different after that.

〰

Frequently asked questions

Q: How much did US industrial production rise in September 2026?

It rose 0.7% from August, according to the Federal Reserve’s preliminary estimate released October 16, 2026. That beat the 0.4% consensus and left output 2.2% above September 2025.

Q: What drove the increase?

Utilities rose 3.5%, the biggest gain among major groups. Manufacturing added 0.3% and mining 0.2%, while motor vehicles and parts fell 3.0%.

Q: What is capacity utilization, and why does it matter?

It measures how much of the nation’s industrial capacity is in use. At 76.9% in September, it remains below the 79.4% long-run average, which suggests limited strain on factories and less direct pressure on prices.

Q: Will this change the Fed’s October decision?

Prediction markets on October 16 still saw an 86.5% chance of no change at the October 27–28 meeting. A single month of output data rarely decides policy on its own.

This article is general information, not financial advice. Preliminary data are revised and market odds change quickly, so check the official sources or consult a qualified professional before making decisions.