Diversification away from Taiwan chipmakers is necessary

During this period of chip shortage, the world’s reliance on Taiwan chipmakers became rather obvious
Image: Shutterstock

A global shortage of chips since last year has exposed Taiwan Semiconductor Manufacturing Company (TSMC)’s critical position in the global semiconductor supply chain. TSMC is the world’s largest chipmaker with a 53% share in the global semiconductor foundry market in Q2 2021 and its current market capitalization ranks 8th in the world. Together with three other smaller players: United Microelectronics Corporation (UMC), Powerchip Technology Corporation (PSMC) and Vanguard International Semiconductor Corporation (VIS), Taiwan is dominating the global market of chipmaking with a 63.3% market share in Q2 2021.

This ongoing global chip shortage was caused by several reasons. These include: a sharp increase of demand for consumer electronics when people need to work, study, and entertain themselves from home during the Covid-19 pandemic, a demand surge of high-performing mining machines as the price of bitcoin increased last year, and the demand rebound for cars during the global economy's recovery from the pandemic later last year. Other reasons include: the disrupted global supply chain when factories shut down and global transportation capability dropped steeply due to the pandemic, and geopolitical factors such as the trade war between US and China which resulted in stocking up of chips by China and other countries. According to a Goldman Sachs’ report, 169 industries were affected by this chip shortage, ranging from smartphones to computers, from automobiles to industrial controlling systems.

During this period of chip shortage, the world’s reliance on Taiwan chipmakers became rather obvious. Industrial players all over the world are now attempting to diversify away from Taiwanese chipmakers through several potential non-Taiwanese sources.

Firstly, in May 2021, South Korean government announced a plan by companies to invest $451 billion to boost the country’s chipmakers' competitiveness, among which, $151.1 billion came from Samsung, the world’s second largest chipmaker, and also South Korea’s largest company. Earlier in May 2020, Samsung had already begun the construction of a $116 billion cutting-edge chip production line intended to help it take on TSMC. Its output is expected to go toward applications that range from 5G wireless networks to high-performance computers.

Secondly, China, the second largest economy in the world, is determined to boost self-sufficiency in the semiconductor industry amid tensions with the U.S.. In this plan, Semiconductor Manufacturing International Corporation (SMIC), China’s largest and most important chipmaker, also the world’s 5th largest foundry, plays an essential role. In March 2021, SMIC started the building of a $2.35 billion new factory in Shenzhen which is partially funded by the government. Again in September 2021, SMIC announced its plan of building another $8.87 billion new plant on the outskirts of Shanghai. However, SMIC currently is mainly manufacturing 28-nanomater chips which is an old technology, while TSMC is manufacturing 5-nanometer semiconductors, the most cutting-edge chips which are used in smartphones. It is also testing the 3-nanometer chips which may be first adopted by Intel and Apple in the near future. Although SMIC could focus on industrial clients which do not require high-end chips, the technological gap between these two companies should not be overlooked if China would like to rise as a leading power in the global semiconductor industry.

Also in March 2021, Intel announced plans to invest $20 billion in U.S. chip facilities and its push to become a global foundry player.

Another U.S. chipmaker, GlobalFoundries, the fourth largest in the world, is also expanding its production capacity in South East Asia with $4 billion investment in its Singapore plant in June 2021. In a recent report by Nikkei Asia, Beh Swan Gin, chairman of the Singapore Economic Development Board was encouraged by GlobalFoundries’ decision to expand its plant in Singapore with $4 billion investment and commented that "The semiconductor industry is a key pillar of Singapore's manufacturing sector, and GlobalFoundries' new fab investment is testament to Singapore's attractiveness as a global node for advanced manufacturing and innovation." Singapore will definitely benefit from the new investment from GlobalFoundries. But compared with the production capacity of its Taiwan peers, the shift in supply chain in the chipmaking industry may not be as significant as GlobalFoundries has expected.

Despite all the attempts by global industrial players to diversify away from Taiwanese chipmakers, I believe in the short term, the global semiconductor industry will still continue relying on them.

There are a lot of reasons for this, such as the hefty capital needed in funding a new semiconductor foundry and the extreme complexity in building one. The technological capability in chipmaking, especially the high-end ones, will take companies years to accumulate. The semiconductor supply chain is also one of the most complicated among all industries, with more than two thousand steps, which requires the collaboration of dozens of industrial players all over the world. Making chips is difficult, diversification amid Covid-19 and US-China trade war is even more difficult.

According to a 2014 paper published in “Technology in Society” by two Taiwanese researchers, the structure of Taiwan’s semiconductor industry is a key reason for its dominance. TSMC was the first to adopt the independent chip foundry model, whereby the firm focused solely on manufacturing chips based on others’ designs. At the same time, other Taiwanese semiconductor firms also played highly specialized roles, each focusing on some specific stages in the chip manufacturing process. According to Adam Smith’s theory about “division of labour”, in doing so, they built leading-edge expertise. Therefore, over the years, Taiwan has established a very efficient and profitable ecosystem for chipmaking that is not easily imitated by others.

In a nutshell, diversification away from Taiwan chipmakers is a necessary attempt, but it is not easy, and it takes time.

Yan Li is Professor of the Information Systems, Decision Sciences and Statistics (IDS) Department at ESSEC Business School Asia Pacific