Manufacturing workforce
US manufacturing jobs: 17.1 million then, 12.6 million now
US manufacturing employed 17.1 million people at the start of the 21st century. Today the figure is 12.6 million, after bottoming near 13.7 million earlier in the decade.
Those numbers are the ground truth under two opposite slogans, that American manufacturing is dying and that it is roaring back. Both borrow from the same series, and both usually drop the detail that makes it make sense.
This page traces the employment figures to a Congressional Research Service report and shows what the decline did and did not do to the kind of work that remains.
Data covers US manufacturing employment, 2000 to the late 2010s (CRS). Last reviewed by a human editor before publication.
The figures and where they come from
Each figure is rated for how safely you can cite it today. Ratings judge current usability, not whether a number was ever correct.
| Figure | What it is | Source | Citation Confidence | Notes |
|---|---|---|---|---|
| 17.1 million | Manufacturing employment, start of the century | [A] | High | US manufacturing employment at the start of the 21st century, the high point most 'decline' framings start from. |
| 12.6 million | Manufacturing employment, recent | [A] | High | Current manufacturing employment in the CRS report. Down about 4.5 million from the start-of-century figure. |
| 13.7 million | Post-recession low | [A] | Medium | The level manufacturing employment had fallen to before a partial recovery. The 'revival' is real but modest against the longer decline. |
| down 11 points | Production-worker share drop, 2000 to 2017 | [A] | Medium | The share of manufacturing workers classified as production workers fell 11 percentage points, evidence the jobs that remain skew away from the line. |
Why the numbers disagree
The same employment series feeds two opposite stories. Start at the 17.1 million peak and the fall to 12.6 million looks like collapse. Start at the post-recession low of 13.7 million and the partial recovery looks like a revival. Both are true movements in one series; which one you hear depends on where the speaker starts the clock.
The composition changed as much as the count. The production-worker share of manufacturing employment dropped 11 percentage points between 2000 and 2017, so the jobs that remain lean more toward engineering, maintenance, and technical roles and less toward the assembly line. A headcount alone hides that shift.
None of this settles the cause. Automation, trade, offshoring, and demand all moved over the same period, and the employment figures on their own cannot apportion blame among them. The numbers describe what happened to the count, not why.
How to cite these figures
Cite the span, not a single number: US manufacturing employment fell from 17.1 million at the start of the century to 12.6 million, with a partial recovery from a low near 13.7 million, per CRS.
Say where you start the clock. A decline framing and a revival framing use the same data from different baselines, and naming the baseline keeps the claim honest.
If the point is the nature of the work, use the production-worker share drop of 11 percentage points, which shows the mix shifting away from the line.
Where people go wrong
Quoting the peak-to-now fall as proof of collapse without noting the partial recovery, or the recovery without noting the longer decline. Both cherry-pick the baseline.
Reading the headcount as a story about automation alone. Trade, offshoring, and demand moved too, and the count cannot separate them.
Ignoring the composition shift. Fewer manufacturing jobs is not the same as the same jobs minus some; the remaining mix leans away from production work.
How we checked
The employment figures come from a Congressional Research Service report on the manufacturing revival, a nonpartisan reference written for Congress. We retrieved the report and located each figure, the 17.1 million and 12.6 million endpoints, the 13.7 million trough, and the 11-point production-worker shift, in its text.
We chose a CRS source because it assembles the Bureau of Labor Statistics employment data into one auditable place, and because the BLS employment pages themselves block automated verification. The figures are BLS-derived; CRS is the readable door to them.
The report does not assign a single cause to the decline, and neither do we. Where the count cannot separate automation from trade and demand, the page says so rather than picking one.
Full source list
Primary sources, with live links. Every figure above traces to one of these.
- [A]Congressional Research ServiceMay 2018
Congressional Research Service, R41898, "Job Creation in the Manufacturing Revival" (Marc Levinson)
https://www.everycrsreport.com/files/20180515_R41898_871083a3797be5051d9cc083978f38fcc107db8c.pdf
Common questions
- How many people work in US manufacturing?
- About 12.6 million, according to the Congressional Research Service, down from 17.1 million at the start of the century and up from a post-recession low near 13.7 million.
- Is US manufacturing dying or recovering?
- Both framings use the same data from different baselines. Employment fell sharply from the 17.1 million peak, then recovered partially from a low near 13.7 million. Naming the baseline is what makes the claim honest.
- Did automation cause the job losses?
- The employment figures cannot say. Automation, trade, offshoring, and demand all moved over the same period, and a headcount alone cannot apportion the cause.
- Are the remaining jobs the same as before?
- No. The production-worker share of manufacturing employment fell 11 percentage points between 2000 and 2017, so the mix leans more toward technical and engineering roles and less toward the assembly line.
More data, traced to source
- The 2.1 million missing manufacturing jobs: what that number actually is
The claim that U.S. manufacturing will have 2.1 million unfilled jobs by 2030 is everywhere. It comes from a 2021 survey of about 800 executives, not a government projection, and it is usually mis-attributed. Here is exactly what the number is.
- 'One robot destroys 5.6 jobs': what the study really measured
The viral statistic that one robot destroys 5.6 jobs comes from a real, careful study. But the 5.6 figure is a local-labor-market estimate, not a claim that each robot nationwide eliminates 5.6 jobs. Here is what the researchers actually measured.
- '47% of jobs at risk': what Frey and Osborne actually measured
The famous claim that nearly half of jobs will be automated traces to one 2013 study. It estimated the probability that occupations are susceptible to computerisation, not a prediction that 47% of jobs will vanish. Here is what it really said.