ITIF Search
Washington Post Gets It Wrong on Manufacturing Jobs

Washington Post Gets It Wrong on Manufacturing Jobs

September 16, 2022

Washington pundits all too often feel compelled to educate the rubes who don’t understand real economics. In fact, all too often, the pundits don’t understand.

Case in point is Catherine Rampell’s recent Washington Post column, “The myth of the manufacturing comeback.” Rampell parrots the conventional “inside the beltway” wisdom about U.S. manufacturing:

Contrary to myths that we’ve stopped ‘making’ things in the United States, we already manufacture a lot of stuff here. In fact, we manufacture nearly the most ‘stuff’ on record, as measured by the inflation-adjusted value of those products.

First, the St. Louis Fed data she cites shows that the year of the most “stuff” was 2007, and 2022 manufacturing output was 4 percent below that peak. Moreover, real GDP grew 25 percent over that period. If manufacturing output remained at a constant share of GDP, it would have had to grow 29 percent more than it did.

Second, as ITIF and economists such as Susan Hausman point out, the overall output numbers overstate significantly actual output growth for two reasons. First, as offshoring has grown, the Bureau of Economic Analysis (BEA) capabilities of differentiating between value-added here vs. overseas have not kept up, and foreign value-added is incorrectly attributed to domestic, making it look like U.S. output is higher than it actually is.

Third, overall BEA output numbers have been significantly inflated because of how the BEA counts changes in value added in the computer and electronics sector (known in official tables as NAICS 334). With Moore’s Law at work and computing power doubling every 18 to 24 months, BEA treats this as if output doubled. Consider that from 1990 to 2010, the real gross output for computers and electronic products grew by an unbelievable 570 percent, 20 times faster than the growth rate for the rest of manufacturing. But nominal shipments of computer and electronic products from U.S. factories grew just 24 percent between 1992 and 2011, while between 2000 and 2010, they fell by about 70 percent. So while manufacturing value added increased by 13 percent from 2007 to 2021, leaving out NAICS 334, it increased just 3 percent.

Finally, no assessment of U.S. manufacturing is comprehensive unless it looks at other nations. As ITIF has recently shown in its Hamilton Index of Advanced-Industry Performance (that looks at U.S. performance in pharmaceuticals; electrical equipment; machinery and equipment; motor vehicle equipment; other transport equipment; computer, electronic, and optical products; and information technology and information services), U.S. performance is weak and declining. U.S. global market share in advanced industries has fallen since 1995 to 6 percent under the size-adjusted global average and 20 percent below when software is not included. China’s share in advanced industries is 34 percent higher than the global average, Japan 43 percent, Germany 74 percent, South Korea and Taiwan both more than double. To match the advanced-industry share of China’s economy, U.S. output would have to expand by nearly $680 billion (42 percent).

But Rampell is not deterred because she persists in the belief that, “We just happen to make that stuff with fewer workers than we used to, because technological advances have led to huge productivity gains.” That’s not really true. If high productivity were the cause of manufacturing job decline, how does she explain that U.S. manufacturing productivity gains were higher in the 1990s than in the 2000s, but manufacturing jobs declined just 3 percent in the 1990s compared to 33 percent in the 2010s? Its cause was China, offshoring, and the skyrocketing U.S. trade deficit.

I also have to note Rampell’s comment, “U.S. factories still make things, but those things are increasingly produced by robots.” In fact, when controlling for manufacturing wage levels, the United States ranks 16th in robots per 1,000 workers, behind number one Korea, which has not seen any manufacturing loss. Moreover, over the last 13 or so years, manufacturing labor productivity has grown more slowly than overall economy-wide productivity, hardly a sign of “increasingly produced by robots.”

It’s not as if this information is not known. Rampell could have easily found it had she done a simple Internet search. Rather, it is easier just to repeat conventional wisdom. After all, the conventional wisdom lets one maintain the comforting belief that globalization still works (even as China has distorted it with its mercantilist predation) and that America does not need a dreaded “industrial policy.”

Despite getting the narrative wrong, Rampell is right in one way: President Biden is wrong to tout manufacturing jobs as the key factor of success. The main reason manufacturing jobs have grown in the last decade has little to do with a more robust manufacturing sector and a lot to do with lagging productivity growth.

Jobs are not the principal reason to support manufacturing. The principal reasons are keeping the trade deficit under control (in absolute terms and as a share of GDP is at an all-time high) and ensuring U.S. economic power. Should we continue to lose advanced industry capabilities, especially to China, America’s global power will shrink, and our dependency on China will grow. The reality is that if we automated all of manufacturing (which is a worthwhile goal, for it would boost living standards and U.S. competitiveness), manufacturing would be no less important.

Arguing over these seemingly technical points may sound trivial but getting the narrative right is critical. If policymakers wrongly think that U.S. manufacturing is healthy, they have less motivation to take needed steps to fix the problem (e.g., institute an investment tax credit, boost apprenticeship programs, increase funding for NIST manufacturing programs, etc.). Likewise, if they think the goal is jobs rather than output and competitiveness, they will take the wrong steps (like enacting policies to limit automation).

Hopefully, Congress will realize that American manufacturing is weak—and America is on a deindustrialization path similar to the UK, especially with the sky-high value of the dollar—and instituting an industrial policy is critical.

Back to Top