The Real “Reality” of America’s Deindustrializing Economy
I feel bad for free market ideologues who believe as a matter of faith that markets almost always get it right and are always superior to government action. It must be hard to maintain such a Panglossian view in the face of economic reality—in this case, the fact that U.S. manufacturing has declined significantly. If policymakers were to fully recognize this reality—which implies that markets did not generate the optimal outcome for the nation—then all sorts of horrible things follow, such as industrial policy (gasp!), or even worse, protectionism. (The horror!)
So, to convince policymakers they should remain sanguine and do nothing other than embrace free trade (including with China) and open the borders to even more to low-skill immigration, one either has to claim that manufacturing is irrelevant (as if the business of producing cars is the same as the business of renting them), or simply deny the reality that U.S. manufacturing has declined. Otherwise, policymakers might take action to fix the serious national problem of deindustrialization.
CATO Institute analyst Colin Grabow has chosen the latter path, writing in “The Reality of American ‘Deindustrialization’” that: “Manufacturing in the United States has not disappeared but has been transformed and very much remains a vital part of the country’s economic fabric.” To be able to spin this yarn with a straight face requires taking liberty with statistics.
Let’s go through each of his claims.
Claim 1
The United States remains a manufacturing powerhouse, accounting for a larger share of global output than Japan, Germany, and South Korea combined.
I would hope so, because the U.S. economy is 80 percent larger than those three economies combined. Despite that size advantage, U.S. manufacturing value added is just 13 percent larger than theirs. No legitimate economist compares factors between economies without controlling for size. It would be equivalent to saying the U.S. health-care system is a disaster because we have twice as many people dying per year as Japan, Germany, and South Korea combined.
Claim 2
In key industries such as autos and aerospace, the United States ranks among the global leaders and is the second-largest manufacturing economy overall.
Yes, the United States leads in aerospace, but that’s the only manufacturing sector we lead. Japan produces more autos on a proportional basis—only 15 percent fewer than the United States, despite having an economy that is just 23 percent as big as the U.S. economy. In 2021, China produced almost four times as many autos as the United States, even though its GDP is smaller. Chinese manufacturing sector overall is also larger than the U.S. manufacturing sector, even though U.S. GDP is larger.
Claim 3
While it is undeniably true that certain manufacturing industries—particularly labor-intensive, low-tech ones—are no longer primarily located in the United States, many other, more advanced ones have flourished. Thus, factories producing consumer staples such as textiles and furniture, for example, have made way for facilities that produce products less often found in retail stores, such as chemicals and machinery.
According to the U.S. Bureau of Economic Analysis (BEA), inflation-adjusted machinery output grew just 12 percent from 2005 to 2022—22 percentage points slower than U.S. GDP growth. Chemicals performed even worse, growing just 9 percent—25 percentage points slower than GDP growth. The issue is not whether an industry grew, but whether it kept pace with the rest of the economy. ITIF’s Hamilton Index uses an analytical statistic known as a “location quotients” (LQ) to assess national output in key industries relative to the rest of the world. (The LQ is a ratio calculated as an industry’s share of a country’s economy divided by the global industry’s share of the global economy.) The U.S. LQ for machinery and equipment was just 0.59 in 2020 (down from 0.80 in 1995), meaning that the United States was 41 percent less specialized in machinery production than the global average. America’s performance in chemicals was slightly better, but still well below the global average at 0.74 in 2020 (down from 0.94 in 1995).
Claim 4
Productivity gains unleashed by automation and other technologies have enabled manufacturing output to remain near record highs even as direct manufacturing employment has declined.
As noted, according to BEA, real manufacturing value added declined from 12.6 percent of GDP in 2005 to 11.8 in 2022. Moreover, according to the Bureau of Labor Statistics (BLS), labor productivity growth in U.S. manufacturing actually dropped from 2010 to 2019, likely the first time in American history that has happened over a sustained period.
Claim 5
The country’s industrial prowess is also evidenced by foreigners’ appetite for investment in U.S. manufacturing… 2021 also saw $121.3 billion of new FDI flow into domestic manufacturing—an amount greater than any other industry.
But this counts acquisitions of U.S. companies as the same as building a factory in the United States. Since the early 1990s, foreign direct investment (FDI) to establish or expand operations in the United States has fallen by more than 90 percent to the point where the value of acquisitions as a share of FDI was 99 percent in 2022.
Claim 6
In 2019, U.S. firms exported over $1.3 trillion in manufactured goods.
Who cares? The issue is whether that is a lot or a little given the size of our economy. In 2019, the United States also imported $2.3 trillion in manufactured goods, so a healthy manufacturing sector would be exporting at least $2.3 trillion.
Claim 7
U.S. manufacturing still maintains high output despite such workforce reductions is largely due to robots, computers, process improvements, etc.
The U.S. lags behind other major advanced economies in measures like industrial robot adoption. According to the International Federation of Robotics, the United States ranked eighth in industrial robot density in 2021 (measured as industrial robots per 10,000 workers), behind South Korea, Singapore, Japan, Germany, China, Sweden, and Taiwan. And when controlling for the fact that wages are much lower in China, U.S. manufacturers use 12 times fewer robots than Chinese manufacturers.
Claim 8
A 2015 study found that 88 percent of manufacturing job losses from 2000 to 2010 can be attributed to improved productivity.
The study Grabow is referring to here, like many others, does not account for the fact that BEA output numbers have been significantly inflated because of how the BEA counts changes in value added in the computer and electronics sector (categorized 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. Meanwhile, nominal shipments of computer and electronic products from U.S. factories grew just 24 percent between 1992 and 2011, and between 2000 and 2010 they fell by about 70 percent. So, manufacturing value added increased by 13 percent from 2007 to 2021, but leaving out NAICS 334, it increased just 3 percent. As such, studies that control for this find that over half of the job losses were due to loss of U.S. global market share in manufacturing.
Moreover, from 1965 to 2000, manufacturing employment was more or less stable, hovering between 19.5 and 16.5 million jobs. But the 2000s saw the number of manufacturing jobs plummet by more than 5 million—a rate faster than in the Great Depression. In fact, the rate of manufacturing jobs loss was 10 times greater in the 2000s than in 1990s (32 percent versus 3 percent). Yet, manufacturing productivity grew faster relative to the rest of the economy in the 1990s than it did from 2000 to 2012.
Claim 9
Manufacturing’s share of employment nearly halved in Germany from 1973 to 2016 while Australia witnessed a two-thirds reduction over the same period.
Sure, but the share of manufacturing employment in Germany is still 80 percent larger than in the United States. Australia should not be mentioned as a model because its performance in manufacturing is even worse than America’s.
~ ~ ~
John Maynard Keynes once famously said, “When the facts change, I change my mind. What do you do?” Sadly, I don’t think we should hold our breath for free-market ideologues to change their minds, even if the performance of manufacturers continues declining. Although, they might just switch from “manufacturing is healthy” to “manufacturing is irrelevant,” a position some ideological free traders now assume.
