US-China Decoupling: A Reality Check
I was a decoupler before decoupling was cool. I have advocated selective decoupling of the US economy from China for the past four years, arguing that the United States requires absolute self-sufficiency in strategic areas such as defense electronics. That, Dr Henry Kressel and I argued in a November 2016 Wall Street Journal op-ed, requires a massive national effort to bring computer chip fabrication back onshore. We wrote: “Washington should also enforce strict US content rules for sensitive defense technology. Many of the Pentagon’s military systems depend on imported components. That’s a concern on security grounds alone. Procurement rules should be changed to require that critical components be manufactured in the US.”
The suddenly popular idea of total decoupling of the American and Chinese economies, however, is not a policy, but a tantrum.
Self-sufficiency in strategic goods is expensive, but national security is like JP Morgan’s yacht: If you have to ask how much it costs, you can’t afford it. Self-sufficiency in all manufacturing is a different matter; imports from China now amount to roughly a quarter of total US manufacturing output, and the cost of domestic substitutes would be far greater than that, because the US doesn’t have the skills to replace a great deal of Chinese production.
China shipped $70 billion of smartphones to the US in 2018 and $45 billion worth of computers. Here is Apple CEO Tim Cook on why Apple makes I-phones in China: “China has moved into very advanced manufacturing, so you find in China the intersection of craftsman kind of skill, and sophisticated robotics and the computer science world. That intersection, which is very rare to find anywhere, that kind of skill, is very important to our business because of the precision and quality level that we like.
“The thing that most people focus on if they’re a foreigner coming to China is the size of the market, and obviously it’s the biggest market in the world in so many areas. But for us, the number one attraction is the quality of the people… It’s not designed and sent over, that sounds like there’s no interaction. The truth is, the process engineering and process development associated with our products require innovation in and of itself. Not only the product but the way that it’s made, because we want to make things in the scale of hundreds of millions, and we want the quality level of zero defects.”
Cook added: “The products we do require really advanced tooling, and the precision that you have to have, the tooling and working with the materials that we do are state of the art. And the tooling skill is very deep here. In the US you could have a meeting of tooling engineers and I’m not sure we could fill the room. In China you could fill multiple football fields.”
The US doesn’t have the engineers to make a smartphone. In fact, we don’t have enough engineers to expand US manufacturing output by any significant margin. As of 2015, China graduated six times as many engineers as the United States, according to the National Science Foundation. That was five years ago. In the meantime China’s university system, enriched by tens of thousands of American-educated doctoral candidates, has come up to par with US universities in most STEM fields. Four out of five US doctoral candidates in electrical engineering and computer science are foreign students, and the largest cohort by far is Chinese. And most Chinese engineers go home when they get their degree, because only 5% of American college students major in engineering, and there aren’t enough faculty jobs around to hire new PhDs.
Meanwhile, Russia, China’s de facto partner in a range of high-tech industries, graduated nearly twice as many engineers as the United States in 2015. Russian engineers are first rate, as the Israelis well know; mass immigration of Russian Jews brought about 150,000 scientists and engineers to Israel, and turned the small country into a pocket superpower. Together, China and Russia have an eight-to-one advantage over the United States in engineering graduates.
Whether the United States could train up enough tooling engineers to produce I-phones onshore, and how long it would take if we could, is hard to answer. The US graduates barely over 30,000 mechanical engineers per year. If we waste our limited talent by replacing production of consumer electronics imports from China, we will lose the race for pre-eminence in the Fourth Industrial Revolution. The United States should concentrate on forcing breakthroughs in frontier technologies that China does not yet dominate, rather than chasing after China’s production of existing products.
In January of this year the Trump Administration asked Taiwan Semiconductor Manufacturing Corporation, the world’s most advanced fabricator, to build a plant in the United States. TSMC has shown no interest in doing so. Remarkable is that the fact that the Trump Administration did not ask Intel or another US chip fabricator to build capacity to replace what we now import from Asia. US chip designers lost interest in the capital-intensive business of fabrication two decades ago. TSMC and South Korea’s Samsung now dominate fabrication of newer, more-efficient 7 nanometer chips, while Intel is struggling to come up to speed in production of top-of-the-line chips. American manufacturers still provide most of the world’s fabrication equipment, but the Dutch firm ASML dominates the market for the newest chips. America’s share of world semiconductor shipments has fallen from a quarter in 2015 to a tenth in 2020, and what remains onshore consists mainly of older and soon-to-be obsolete plants.
It is not surprising that TSMC is coy about building plants in the US. The Trump Administration is considering a ban on TSMC’s exports of chips to Huawei, which accounts for 10% of its total sales, because TSMC uses some American chip-making equivalent. The United States could, of course, hire Taiwanese engineers to build our own fabrication plants, but it would come late to the game. A tenth of Taiwan’s production engineers presently are working in Mainland China at double pay, in China’s crash effort to build its own fabrication capacity, according to Nikkei Asian Review.
Nonetheless the United States has no intelligent choice except to bring chip fabrication onshore. A back door or a time bomb can be planted anywhere among the 20 billion transistors on a seven-nanometer computer chip, as University of Michigan researchers showed in 2016. There is no way to secure the electronics of the American military except by fabricating the components in secure facilities in the United States. That will strain our resources and cost upwards of $100 billion. A number of promising new technologies are still in embryo that might transform the industry. Instead of etching transistors on silicon with light, it may be possible to build chips up from the molecular level, at a fraction of the cost.
American technological capacity is hollowed out, even in fields at which we believe ourselves to excel, for example, information science. Apple and Google announced this week that a smartphone app would be forthcoming in May that would inform the user whether they had been in contact with a person infected with coronavirus. Alipay and Tencent had such an app available in mid-February; the South Koreans and Israelis had similar apps available in early March. Israel used the terrorist tracking system of its formidable security police, the Shin Beth, to track actual and suspected carriers, while China employed its comprehensive surveillance capability to match coronavirus test results to the location and body temperature readings of hundreds of millions of people.
Writing in The American Interest, Andrew Michta of the George Marshall Center for European Studies proposes a “long hard road to de-coupling,” and avers: “We need a massive reinvestment in STEM curricula in our high schools and in science and engineering programs at our colleges and universities, so as to expand the available labor and management pools for our re-shored companies. Again, it will take a concerted effort by Congress, the US Department of Education, and especially parents and alumni donors to restore colleges to their proper place of teaching and learning, which at one time decades ago produced the best professional and managerial classes in the world.” That is what America requires, but the lead time will be a generation.
The obstacles are formidable. As Edward Dougherty, distinguished professor of engineering at Texas A&M University, wrote last year in Asia Times, “The mathematics necessary for this modern engineering developed rapidly through the 1950s and was required for graduate engineering students in good programs. This requirement has been dropped in most of today’s American universities. Instead, engineers are groping around trying to find solutions by playing with a computer. On the other hand, in Iran, students are required to study the relevant mathematics at the undergraduate level. As a nation, we have, with forethought, decided that our children should have inferior educations to Iranian and Chinese children.” If China provokes us to impose a culture of competence on our educational institution, and to reward real achievement instead of the inculcation of self-esteem, it will have done us a greater service than any of our allies at any time in our history.
The divergence of American and European views on China is noteworthy. Reviewing mentions of decoupling in the German press during the past month, I found nothing but commentary to the effect that it is a bad idea. As I reported April 3 in Asia Times, Europe’s big pharmaceuticals firms are lining up to start joint ventures with Tencent, Huawei and other big Chinese IT companies, in order to exploit China’s vast wealth of data. In the European view, China’s initial blunders in reporting the coronavirus outbreak are of less interest than China’s use of information technology to control the epidemic, and the artificial intelligence applications of China’s vast database.
That raises a troubling question: If America really were to attempt to decouple from China, where would America’s allies stand? Most of America’s allies rejected the Trump Administration’s urgent demand to exclude Huawei from the rollout of their new 5G networks. As noted, Europe’s pharmaceuticals firms see in China’s handling of the coronavirus not an insidious cover-up by a wicked Communist Party, but rather a technological advance in which they hope to invest. If the world divides into economic zones, all of Eurasia may wind up in a zone dominated by China, standing off against North America and Japan.
To some extent this already has happened. Huawei is the world leader in telecommunications equipment in part because it hired 50,000 foreign employees, mostly Europeans, and mostly engaged in basic R&D. One doesn’t need to rehash outdated geopolitical theories to observe that a China-led Eurasia would outweigh the United States several times over in human as well as physical capital.
None of this need happen, to be sure. As I wrote in a 2019 discussion on the future of NATO at Law and Liberty, “If the United States reasserts its leading role in frontier technologies, our NATO partners will queue up to join the effort, if only because breakthroughs in military technology also have game-changing civilian applications.” The US requires a set of “Manhattan Project” initiatives focusing on quantum computing, technological spinoffs of mobile broadband, defense against the new generation of hypervelocity missiles, chip-making technology, and other game changers. Whether America’s penchant for innovation still is powerful enough to compensate for the sheer number of Chinese and Russian engineers remains to be seen. As an American, I would hate to lose by default.
It’s important to remember that every single invention of the digital age, from the microchip to the semiconductor laser, to the graphic user interface, to light-emitting diodes, to plasma displays, and to the Internet itself, started with a Defense Department research grant to one of the great corporate labs. And without exception, every one of these technologies was discovered when someone was trying to do something else. The semiconductor laser, the foundation of optical networks, started with a Defense Department grant to RCA Labs to find a way to illuminate battlefields at night, for example. The United States invented the digital age not because it set out to do so, but because it developed such a density of innovative basic research that the practical discoveries vastly exceeded the original expectations of the projects that funded them.
America’s triumph in the Cold War came from our willingness to deal with the unknown unknowns. China’s weakness is that its vast army of engineers is harnessed to specific projects with fixed goals. America’s advantage at the peak of our prowess was our ability to look for the unexpected. The US will succeed to the extent it looks forward to the next generation of innovations. It’s pointless to speak of restoring “lost manufacturing capacity.” Manufacturing may be unrecognizable in 20 years, as low-latency 5G communications allows industrial robots to devise new production solutions without human intervention, and 3D printing makes all manufacturing local. The world will look very different a generation from now. The question is who will lead it.