Winners and Losers in US Manufacturing

Manufacturing in the US has recently been doing quite well, as I've noted before. But as I continue to ponder exactly what is driving US manufacturing growth, I thought it might be interesting to get a better sense for exactly which types of manufacturing activities have been doing particularly well and particularly poorly.

Of course we all know that manufacturing employment in the US has been on a secular downward trend for about 3 decades now. This has been driven largely, I would argue, by technological advances that have dramatically increased manufacturing productivity. As illustrated by the picture below, improvements in productivity mean that the US manufacturing sector now requires a third fewer workers to produce the same output compared to the year 2000. That largely explains the 33% drop in manufacturing employment in the US between the years 2000 and 2010.


But some sectors of US manufacturing have done better than others. To get a better understanding of this, I put together a table showing the change in manufacturing employment in the US during the 2000s by type of manufactured product. Here's the result:


As the table above shows, lumber, primary metals (e.g. steel), motor vehicles, and textiles & apparels have done particularly poorly. But other sectors have done relatively well, such as the aerospace industry, some types of fabricated metal products (chiefly industrial equipment and intermediate inputs to other manufactured goods), medical equipment, food & beverages, and chemicals.

Can we generalize about the differences between sectors that have performed well compared to those that haven't? It's a very old and well-investigated research question (far predating the famous Leontief paradox, which provided a surprising answer to that question in the 1950s), but it's still interesting to think about.

To some degree, I think you can see the impact of international trade here. Some of the products that seem to be more sensitive to international competition (e.g. textiles & apparel) do worse than products that are probably less vulnerable to international competition (e.g. food and beverages). But I think a more consistent theme is that the strongest sectors in US manufacturing are those that require more highly-skilled labor, and/or are more highly automated. Not a surprising result, but perhaps reassuring nonetheless.