Given the insouciance of America's government and most of the business community about the ongoing decline of American manufacturing, one thing is clear: the U.S. has staked its economic future on being a post-industrial society. So it behooves us to take a long, hard look at whether this is a viable economic strategy for a nation with 300 million mouths to feed. We can't all be investment bankers.
The most succinct formulation of postindustrialism is this:
Manufacturing is old hat and America is moving on to better things.
The postindustrial economy is considerably less attractive today than it was only a few years ago, thanks largely to India’s success in computer software and business process offshoring. This discredited the (rather odd) idea that our competitors were only going to compete in manufacturing. But it is still America's core economic strategy, if only by default, and the idea played a large role during the 1980s and 1990s in getting Americans to accept deindustrialization. It has been promoted by writers as varied as futurist Alvin Toffler, capitalist romantic George Gilder, techno-libertarian Virginia Postrel, and futurist John Naisbitt. Newt Gingrich seized upon it as the supposed economic basis of his Republican Revolution of 1994.
Unfortunately, the core ideas of postindustrialism don’t stand up well to empirical evidence. Above all, a declining share of manufacturing in GDP is not an automatic correlate of economic progress. Between 1947 and 1966—a period of rapidly advancing technology and rising prosperity—manufacturing actually went up slightly as a share of our GDP. Manufacturing’s share of GDP has indeed fallen in recent years, with services expanding to fill the gap. But this merely reflects the fact that inflation has been lower in manufacturing than in services, due to higher productivity growth in manufacturing. (This is itself a clue that manufac-turing might have its advantages!) If one adjusts for the inflation dif-ferential, manufacturing’s share has actually been quite stable for the last 30 years or so, and only began to decline around 2000. This is far too late for transition to a postindustrial economy to explain it, but entirely in line with our burgeoning trade deficit in manufactured goods. And if one looks at the trend not in America’s production of manufactured goods, but in our consumption, there has been no decline at all. The gap between production and consumption is (as explained in the previous chapter) just our trade deficit. So using postindustrialism to justify our trade deficit in manufactures just presupposes what it is trying to prove.
Nevertheless, postindustrialism remains popular in some very important circles. In the 2006 words of the prestigious and quasi-official Council on Competitiveness, a group of American business, labor, academic and government leaders:
Services are where the high value is today, not in manufacturing. Manufacturing stuff per se is relatively low value. That is why it is being done in China or Thailand. It’s the service functions of manufacturing that are where the high value is today, and that is what America can excel in.
Unfortunately, the above paragraph is simply not true, and manufacturing is not an obsolescent sector of the economy. Low grade “screwdriver plant” final assembly manufacturing is indeed primitive, and can increasingly be done anywhere in the world, making it an intrinsically low-wage activity. But the manufacturing of sophisticated high-tech products is a different matter and remains concentrated in advanced industrial nations. The “Made in China” stamped on the outer casing of fax machines, cellular phones, and other high-tech products often just means that final “kit” assembly took place there. The key internal components, which make up a large percentage of the finished product’s cost, are frequently still made in high-wage nations like Japan. In the case of fax machines, this is the electro-optical read-write head. In the case of printers, it is the print engine. In the case of watches, it is the movement. Apple’s iPod, for example, is assembled in China, but its display module is made in Japan, its video processor chip in Taiwan or Singapore, its memory chip in South Korea, and its central processing unit in the U.S. or Taiwan—all nations whose average incomes are multiples of China’s.
Even more important than the value of these components is their value per man-hour of labor required to make them, as this is the ultimate basis of high wages. For example, of the 28,556 jobs created by the iPod outside retailing and distribution, 19,190 were production jobs, of which China captured the most (11,715). But 9,366 were professional jobs, of which high-wage Japan (with 1,140) and the U.S. (with 6,101) captured the lion’s share. For products whose production cost mainly consists of technology and capital, not low-skilled labor, low-wage nations have no advantage, as technology and capital are not cheaper there.
The table below gives a breakdown of the cost structure of the average U.S. manufacturer:
Advertising & Marketing
Research & Development
Health & Safety
Land & Rent
Source: Peter Navarro, “Report of The China Price Project,” Merage School of Business, University of California at Irvine, July 2006, p.5.
It has been estimated that direct labor is under 20 percent of production cost for half of U.S. manufacturers. The average cost disadvantage of U.S. manufacturers versus their opposite numbers in low-wage nations is, in fact, estimated to be only 17 percent, a difference obviously often within the reach of smart strategies.[xi] This is why manufacturing still thrives in some other high-wage nations.[xii] It is why America could be successfully defending positions in manufacturing industries when we are failing and why some other developed nations are succeeding at this better than we are. High U.S. wages (and/or unions) certainly aren't the problem: Germany has higher manufacturing wages and more unions than we do, but it remains competitive in these industries.
About the only thing postindustrialism gets right is that selling a product with a high value per embodied man-hour almost always means selling embodied know-how. But know-how must usually be embodied in some physical package before reaching the consumer, and manufactured goods are actually a rather good package for embodying it in. Exporting disembodied know-how like design services is definitely an inferior proposition, as indicated by the fact that since 2004, America’s deficit in high technology goods has exceeded our surplus in intellectual property, royalties, licenses, and fees.
That some individual companies like Apple make a success out of keeping design functions at home and offshoring the manufacturing does not make this a viable strategy for the economy as a whole. Apple is a unique company; that is why it succeeds. And even fabled Apple is not quite the success story one might hope for, from a trade point-of-view. Due to its foreign components and assembly, every $300 iPod sold in the U.S. adds another $140 to our deficit with China.[xiv] If sophisticated American design must be embodied in imported goods in order to be sold, it will not help our trade situation.
Meanwhile, other companies are shutting their U.S. design centers and moving them closer to actual production and the know-how that accumu-lates where it takes place. As Douglas Bartlett, chairman of the printed circuit board manufacturer Bartlett manufacturing in Cary, Illinois, puts it:
Anyone who knows anything about real-world manufacturing knows that the factory floor and the lab form a continuous feed-back loop. Unfortunately, virtually none of our trade and economic policymakers know anything about real-world manufacturing.
So the erosion continues, industry by industry. For example, in March 2007, Chrysler closed its Pacifica Advanced Product Design Center in Southern California, following the closure of nearby centers owned by Italdesign, American Specialty Cars, Porsche, Nissan, and Volvo. Of GM’s 11 design centers, only three are still in the U.S. In the words of Eric Noble of The Car Lab, an automotive consulting company, “Advanced studios want to be where the new frontier is. So in China, studios are pop-ping up like rabbits.” This trend bodes extremely ill for the future; as Stephen Cohen and John Zysman explain in their book Manufacturing Matters:
America must control the production of those high-tech products it invents and designs—and it must do so in a direct and hands-on way...First, production is where the lion’s share of the value added is realized...This is where the returns needed to finance the next round of research and development are generated. Second and most important, unless [research and development] is tightly tied to manufacturing of the product...R&D will fall behind the cutting edge of incremental innovation...High tech gravitates to the state-of-the-art producers.
Neither are individual technological or entrepreneurial genius going to save America, no matter how impressive they may look on the cover of glossy magazines. Richard Florida and Martin Kenny have documented the limited (albeit real) value of stand-alone inventive genius in their book The Breakthrough Illusion.[xx] Despite the impressive U.S. record in pure innovation, innovations actually fail to translate into mass production (and thus high employment) industries here as well as they do in Japan and elsewhere. The fragmentation of America’s high-tech research into thousands of small companies in Silicon Valley and elsewhere may be optimal for innovation itself, but it is not optimal for mass commercialization. Indeed, it has the unfortunate side effect of making it exceptionally easy for foreign companies to buy up American innovations à la carte. Among other things, this has helped make Japanese, rather than American, companies the ultimate commercial beneficiaries of much recent Pentagon-funded research.
A small American company named Ampex in Redwood City, California, encapsulates everything that is wrong with postindustrialism. This leading audio tape firm invented the video cassette recorder in 1970 but bungled the transition to mass production and ended up licensing the technology to the Japanese. It collected millions in royalties all through the 1980s and 1990s and employed a few hundred people. Its licensee companies collected tens of billions in sales and employed hundreds of thousands of people.
So when someone like self-described “radical free trader” Thomas Friedman writes that “there may be a limit to the number of good factory jobs in the world, but there is no limit to the number of good idea-generated jobs in the world,” (emphasis in original) this is simply false.[xxiv] There is nothing about the fact that ideas are abstract and the products of factories concrete that causes there to be an infinite demand for ideas. The limit on the number of idea-generated jobs is set by the amount of money people are willing to pay for ideas (either in their pure form or embodied in goods) because this ultimately pays the salaries of idea-generated jobs.
The final killer of the postindustrial dream, of course, is offshoring, as this means that even if capturing primarily service industry jobs were a desriible strategy, America can’t reliably capture and hold these jobs anyway. The caliber of jobs being offshored—which started with fairly mundane jobs such as call centers—is relentlessly rising. According to a 2007 study by Duke University’s Fuqua School of Business and the consulting firm Booz Allen Hamilton:
Relocating core business functions such as product design, engi-neering and R&D represents a new and growing trend. Although labor arbitrage strategies continue to be key drivers of offshoring, sourcing and accessing talent is the primary driver of next-gener-ation offshoring…Until recently, offshoring was almost entirely associated with locating and setting up IT services, call centers and other business processes in lower-cost countries. But IT out-sourcing is reaching maturity and now the growth is centered around product and process innovation.
Among sophisticated business functions, product development, including software development, is now the second-largest corporate function being offshored. Offshoring of sophisticated white-collar tasks such as finance, accounting, sales, and personnel management is growing at 35 percent per year. Meanwhile, despite a few individual companies bringing offshored call centers back home, offshoring of call centers and help desks continues to grow at a double-digit pace.
It is no accident that, as noted in the Foreword, some of America’s corporate elite are now starting to question postindustrialism, about which they were utterly gung-ho only a few years ago. In the February 2009 words of General Electric’s chairman, Jeffrey Immelt:
I believe that a popular, 30-year notion that the U.S. can evolve from being a technology and manufacturing leader to a service leader is just wrong. In the end, this philosophy transformed the financial services industry from one that supported commerce to a complex trading market that operated outside the economy. Real engineering was traded for financial engineering.
Immelt has since argued that the U.S. should aim for manufacturing jobs to comprise at least 20 percent of all jobs—roughly double their current per-centage. Only a few years ago, this idea would have been dismissed as an ignorant and reactionary piece of central planning, especially if it had not been proposed by a respected Fortune 500 CEO.
First, don't bet on America's long-term economic prospects as long as we remain committed to postindustrialism. Don't assume that everything will be fine as soon as the U.S. bounces back from recession. The weak U.S. manufacturing sector will remain an ongoing drag on U.S. performance.
Second, don't assume we will remain committed to it. If postindustrialism isn't a viable strategy, at some point there will be a revolt against it in Washington.
Third, take seriously companies that have managed to remain competitive in manufacturing domestically. As other high-wage nations like Germany and Japan have shown, this can be a winner.
Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.