The piece argues that maintaining manufacturing capacity is essential to preserve and develop technical expertise: designing complex artifacts follows making them, not the other way around. Remote design and academic training alone cannot substitute for proximity to production; examples include semiconductor design, which depends on nearby fabrication, and robotics, where hands-on work at scale matters. The author pushes back against the claim that automation, rather than offshoring, explains the collapse of U.S. manufacturing employment, saying that losing factories removes the tacit knowledge needed to innovate and design new generations of products.
The argument is supported with concrete U.S. data: manufacturing employment fell from 17.3 million in 2000 (a peak of 19.6 million in 1979) to 11.5 million in 2010, rising only to 12.6 million by August 2026 - about 5.8 million jobs lost over a decade with only ~1 million recovered. Manufacturing output (2017=100) rose from 51 in 1985 to 93 in 2000, peaked near 105 in 2007, then stagnated (93 in 2010, 99.1 in 2026), while labor productivity jumped from ~70 in 2000 to ~100 by 2010 and has since flattened. These patterns point to a geographic shift in production - outsourcing abroad - rather than pure productivity-driven job loss, with the practical consequence that without local factories the expertise to design and improve products erodes.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.