Talk of a manufacturing revival is in the air. America has, in fact, gained a quarter-million industrial jobs (source) since the start of 2010. Unfortunately, this is less than 15 percent of the number lost during the recession. Furthermore, after this teasing uptick, U.S. manufacturing output seems to be stalling again. So it worth revisiting a much denied fact I have written about before here and here: American manufacturing is in a state of profound crisis.
To get past the slew of analysis out there claiming everything is fine, it is crucial to understand why the usually quoted statistics that seem to show that American manufacturing is healthy are wrong.
First off, looking at aggregate manufacturing output, as most of these analyses do, obscures the fact that total output has only been stable (or close to it) because of a few sectors which have grown enormously. The rest of the manufacturing economy has been declining. According to a recent report from the Information Technology and Innovation Foundation,
Most manufacturing sectors actually shrank in terms of real value-added from 2000 to 2009. In fact, from 2000 to 2009, fifteen of nineteen U.S. manufacturing sectors saw absolute declines in output; they were producing less in 2009 than they were at the start of the decade. There were declines of:
Food, beverage, and tobacco products - 0.2 percent
Electrical equipment - 2 percent
Chemicals - 3 percent
Machinery - 14 percent
Printing - 15 percent
Wood products - 16 percent
Motor vehicles - 18 percent
Fabricated metals - 27 percent
Nonmetallic minerals and primary metals - 28 percent
Paper - 28 percent
Plastics - 31 percent
Apparel - 40 percent
Furniture - 43 percent
Textiles - 43 percent
The bottom line? Fifteen manufacturing sectors, comprising nearly 80 percent of U.S. manufacturing output, produced less in 2009 than in 2000.
What were the wonder sectors that made up for all this decline? Mineral fuels (coal, oil, gas) and computers. Unfortunately, there are good reasons to believe that the apparent soaring of American fuel output is illusory. Coal output was unchanged 2000-2010, according to the Energy Information Agency, and gas output declined somewhat, so oil must have boomed spectacularly for these numbers to be right. (It hasn't.) Most of this increase in output is simply the rising price of oil. In any case, classifying oil extraction (not production!) as a manufacturing sector is dubious, for obvious reasons.
What does our manufacturing sector look like if we correct for these distortions? If we assume no real increase in oil production, and assume that the computer sector expanded by a more-realistic 50 percent during this period, American manufacturing's real (inflation adjusted) output declined by nine percent. (Source.) Even if we bump up our assumptions about the computers and electronics sector considerably, we still get decline.
To be fair, other analyses of the problem have produced different numbers. This is to be expected, as not all these analyses measure exactly the same things. But their general conclusion is consistent. For example, economist Susan Houseman has reported that while total manufacturing output grew 1.18% per year from 1997 to 2007, it grew by just 0.46% per year once the computers and electronics are taken out of the picture. That's anemic.
Computers are fine things, and it's understandable that they would be a growing part of our economy. But this is hardly a picture of a healthy manufacturing sector. It's an image of broad-based decline covered up by a boom in one industry.
Consider now manufacturing employment, as opposed to output.
Isn't the decline in U.S. manufacturing employment simply due to the relentless march of factory automation, and therefore a good thing? No. If the decline in manufacturing employment were due simply to the endless march of automation, we would expect to see slowly declining employment in this sector since a peak shortly after WWII. But instead, what see is a relatively stable employment level, but then things fall off a cliff after Y2K. See the chart below (source):
But there was no revolution in manufacturing technology in Y2K that suddenly started radically reducing the number of workers needed, which is what would have to be true for the above decline to be due to technological progress. So these numbers are a sign that outright decline, especially a yawning trade deficit, is responsible, not gradual technological change.
In any case, Luddite mythology aside, automation per se doesn't hurt overall manufacturing employment--as suggested by the fact that Japan, which leads the world in number of robots, also has a higher percentage of its workforce in manufacturing than the U.S.
If you think about it, this makes sense, as if automation enables nine workers to do what ten used to do, those nine are now a better bargain--which increases the incentive to hire them. (In energy economics, this fact is called Jevons' Paradox.)
So don't blame technology for our job losses. If anything, it's a lack of workplace technology, compared to our rivals, that is costing us jobs.
This lack of technology ultimately traces, of course, to a failure to invest in upgrading the manufacturing workplace. If companies continue to invest in manufacturing, whether this takes the form of physical plant or intangibles like research and development, their manufacturing operations will tend to remain healthy. If they don't, they will gradually exit the manufacturing business as their existing plant and know-how become obsolete over time. They may survive (or not!) as designers and packagers of goods manufactured by others, but they will no longer be manufacturing companies.
This means that the writing is on the wall for American manufacturing, as it is falling behind our competitors in the investment race. From 2000 to 2008, our capital investment in manufacturing as a percent of GDP was lower than that of most of our major peer economies. Indeed, between 2000 and 2009, capital investment within the U.S. by American manufacturers went down by more than seven percent. As a result, most American manufacturing industries are now less well capitalized than they were a decade ago. (Sources.)
American companies are not only running down their own productive capacity here at home, they are also building up the capacity of foreign nations. From 2000 to 2009, their manufacturing investment abroad averaged 16 percent higher than manufacturing investment at home. (Source.)
It is no accident that many foreign nations are simply not having the same experience of industrial decline that we are. Despite the myth that manufacturing necessarily declines in advanced nations, the truth is that, over the last decade, many other developed nations have seen manufacturing as a percent of their GDP remain stable, or even increase. In the "stable" category belong Germany, the Netherlands, and Norway. In the "increase" category go Sweden, Austria, Switzerland, Finland, the Czech Republic, Poland, Slovakia, Hungary, and South Korea. (Source.)
The final blemish on the supposed manufacturing revival in America is the fact that the few industrial jobs that are returning to the U.S. are returning at much lower pay scales than before. For example, the Suarez Corp. is reopening a former Hoover plant in North Canton, Ohio to produce EdenPure space heaters, vacuums, air purifiers and other small appliances it previously made in China. But while the Hoover plant used to pay its workers around $20/hr before it shut in 2007, the new jobs will pay $7.50/hr. (Source.)
This is not the formula for a middle-class economy, now or in the future.