Click to read the latest China-US Focus Digest
Decoupling with China has been the flagship word of the day in Washington under the presidency of Donald Trump. The unilateral tariffs on Chinese goods imposed in 2018 and 2019 were only the initial maneuver in this direction. Since the outbreak of the coronavirus pandemic, we have seen an intensification both in words and deeds in the U.S. for ultimate decoupling with China.
Following repeated calls for American businesses to move back home, Trump claimed on May 14 that he had cut off all relations with China. The ban of TikTok, WeChat and Huawei’s 5G in the U.S., as well as an export ban of semiconductor chips, are key steps in de-coupling with China in cutting-edge technology.
The U.S. has also proposed the so-called Clean Network Initiative, which would exclude China from internet product shops, apps, the cloud, internet operators and the China-U.S. seabed fiber-optic network (CUCN). All these developments have aroused widespread concern in both countries.
China-U.S. de-coupling, whether feasible or not, can be classified into three different categories.
First are technologies and products for military use, or civil-military dual use but most likely for military.
For this category, China and the U.S. have been decoupled since the 1950’s. There was first COCOM in Paris, then the Wassenaar Arrangement. Decoupling in this category will continue. The real effects are, however, limited. China has developed almost all the modern defense and dual-use technologies it needs, including nuclear arms, intercontinental missiles (Dongfeng 21), the Chang’e 2 and the Beidou system.
Second are ordinary products for civil use. Decoupling efforts, including higher tariffs and calls for American businesses to move back home, seem useless in general.
During the first three quarters of 2020, Chinese exports of goods to the U.S. were 3 percent higher than the same period in 2017, before the tariffs were imposed. Exports to the U.S. in July, August and September 2020 soared by 15.6 percent, 22.4 percent and 19.4 percent respectively, year-on-year.
The average monthly export volume for this period was $ 44.16 billion, 10.5 percent higher than the monthly average of 2018 which was a historic high. In 2019, in computers and electronic products, 37.6 percent of the U.S. imports came from China, larger than the second-, third- and fourth-largest suppliers (Mexico, Malaysia and Chinese Taiwan) combined.
In electrical equipment, components and apparatus, 34.1 percent of U.S. imports came from China, equal to the second-, third- and fourth-largest suppliers (Mexico, Japan and Germany) combined. This pattern is not likely to change substantially.
There has been an interesting triangular trade pattern with China, Vietnam and the U.S. Since the U.S. imposed higher tariffs on Chinese products, there has been a sharp rise in U.S. imports from Vietnam, showing a good alternate supplier replacing China. In fact, part of the increased supply from Vietnam to the U.S. originated in China.
According to the U.S. Commerce Department, during the first half of 2020 U.S. imports from Vietnam increased by 8.5 percent over a year ago, from $30.44 billion to $33.03 billion. According to Chinese customs data, during the first three quarters of 2020, Chinese exports to Vietnam increased by 12.3 percent over a year ago, from $69.57 billion to $78.02 billion. The U.S. data also show that over the past three years to June 2020, U.S. imports from Vietnam increased by an aggregate 47.7 percent. Chinese customs data also show that over the past three years to September 2020, Chinese exports to Vietnam increased by an aggregate 56.4 percent.
Those interesting figures of parallel increases prove that Vietnam imported materials and components from China, further processed them into finished goods in Vietnam and then shipped them to the U.S. In other words, there has been little decoupling.
The outbreak of the pandemic has not pushed American companies to move back home. Instead, they have accelerated their investments in China. During the first seven months of this year, American companies invested in 860 new enterprises in China. As the Wall Street Journal put it, American businesses look at China as a “safe haven” for their capital.
Third are predominantly high-tech products and services, including 5G and semiconductor chips. In this category the issue will boil down to who will be de-coupled. Will it be China or the United States?
The U.S. high-tech ban on China, which is causing difficulties for Huawei and other Chinese companies in the short term, will result in just the opposite in the long run. Huawei has already made alternative arrangements, for example with its own Huawei Mobile Service. The Chinese government has mapped out an ambitious semiconductor chip development plan for 2020-25, with an investment of $1.4 trillion, both from government and the private sector, to attain 70 percent self-sufficiency. That could change the global high-tech landscape.
The U.S. ban is delivering a potentially mortal blow to its own high-tech industry, especially the semiconductor sector. The total world semiconductor chip market was worth $478.4 billion in 2018, with the Chinese market accounting for $158.4 billion, or 33.1 percent of the total — equal to the U.S. market ($103 billion) and European market ($43 billion) combined. The dependence of the top 10 U.S. chipmakers on China varied from 23 percent for Intel to 52 percent for Broadcom, 63 percent for Qualcomm and 80 percent for Skyworks Solutions. The loss of the Chinese market will threaten their survival.
A Boston Consulting Group report, “How Restrictions on China Trade will End the American Semiconductor Industry's Leadership,” said the R&D input of the U.S. semiconductor industry amounted to $312 billion over the past 10 years, with $39 billion for 2018 alone —twice that of the rest of the world. Half of semiconductor demand is for smartphones, PCs and consumer electronics, all with very short life cycles. Therefore, the U.S. semiconductor sector has no choice but to maintain intensive and huge R&D input to keep its competitive edge.
Because of its dominant power, the U.S. semiconductor industry has 48 percent of the world market share, 18 percent of which is China. If the U.S. semiconductor industry is totally decoupled from China, it will lose 18 percent of the world market, lowering its share to 30 percent in three years. By then the world’s largest semiconductor player will be South Korea or China, not the U.S. American semiconductor companies stand to lose 16 percent of their revenue and 25 percent of their R&D input. In that eventuality, they will no longer be able to lead the world in semiconductor technology.
The Asian Times on May 25, carried an article by David Goldman, a U.S. economist, titled “Who de-couples whom? Are we just doing the opposite?” The article said that while the U.S. is pondering decoupling with China, the real scenario looks more like Asia decoupling from America.
Recent developments are posing a real threat to the U.S. semiconductor industry. Samsung, the world’s largest semiconductor processor, announced on May 21 an investment plan of $8 billion to build the world’s largest chip processing factory, while expanding its memory chip factory in Xi’an, China. Toshiba of Japan announced a 195 billion yen ($1.9 billion) investment plan to build the world’s largest semiconductor factory, which would exclude U.S. technology completely.
Alarm bells can already be heard in Washington. Qualcomm, Intel and other U.S. companies have been lobbying fervently for a change in the decoupling policy, resulting in a certain relaxation of the tech ban, allowing Advanced Micro Devices and Intel to continue to supply Huawei.
Facts have already proved, and will continue to prove, that while individual cases of trade and production shifts may happen, and certain technology restrictions may remain, the substantive overall decoupling of economic relations and supply chains between China and the U.S. is out of the question. It runs counter to economic basics, hurts U.S. companies and will only end in failure.
