Condivergence: How the CHIPS Act affects the semiconductor industry – The Edge Malaysia
This article first appeared in Forum, The Edge Malaysia Weekly on June 3, 2024 – June 9, 2024
On Aug 9, 2022, US President Joe Biden signed the CHIPS and Science Act, which provided US$76 billion (RM358 billion) over a five-year period to boost onshore semiconductor production. The US manufactures only 12% of the world’s semiconductors, even though it designs 47% of semiconductors sold worldwide. Japan has plans to invest US$67 billion in the chip industry, far exceeding the US$47 billion announced so far by Europe. China has recently upped the ante by adding US$47.5 billion to a fund that will now have a US$140 billion war chest. The top three largest economies account for 81% of state intervention in the chip sector, according to the Hinrich Foundation, an Asia-based philanthropic organisation that works to advance sustainable global trade through research and education.
Since semiconductors are found in almost all “smart” products and are critical to cutting-edge military equipment and artificial intelligence, the Trump and Biden administrations want to secure American global power through sustaining technology leadership and manufacturing output. As China has become the main source of global manufacturing and trade, both America and Europe woke up to the fact that too much outsourcing of manufacturing makes them vulnerable to supply shortages, where rival powers can control supply chain choke points.
With the top four regions aiming to pour more than US$330 billion to subsidise a semiconductor industry that produced US$526 billion in revenue in 2023, will there be a glut in semiconductor chip supply in the coming years? Governments are giving not just direct grants, but also tax credits and tariff plus other protection to domestic champions.
The US money programme has brought back leading US and non-US technology companies such as Intel, Micron, GlobalFoundries, Samsung and Taiwan Semiconductor Manufacturing Corporation (TSMC) to expand semiconductor manufacturing capacity. In response, almost all aspiring industrial and military powers are investing in semiconductor technology, because of the dual-use aspect of semiconductor chips. Chips not only have a huge market as inputs into consumer products, the new generation of quantum computers and artificial intelligence (AI) microchips are vital in smart military equipment, such as drones and robot fighting machines.
Even though China’s high savings surplus, accounting for 28% of total global savings in 2023 or only 5% less than the combined savings of the US and the EU, enables China to invest in new technology, the combined West can now cut off China’s access to leading technology and key equipment to build semiconductors through a variety of sanctions, intellectual property rights controls, as well as access to both markets and suppliers. The US Department of Commerce’s Bureau of Industry and Security (BIS) has powerful tools to control access of any company or nation to semiconductor technology.
The BIS rule restricts transfer of “dual-use technology” from Nvidia, Advanced Micro Devices (AMD) and others to Chinese companies. Since the restrictions target China as a geographic destination, Chinese attempts to import AI technologies to their subsidiaries outside of China have met with brutal restrictions, such as the Netherlands and Japan’s restrictions on selling critical types of semiconductor manufacturing equipment to Chinese buyers, directly or indirectly. Chinese semiconductor manufacturers are specifically restricted from the procurement of smaller chips (16nm logic, 18nm DRAM and 128-layer NAND) from South Korea and Taiwan producers.
The rules are much tougher than meets the eye. The US Department of Commerce can prohibit any company that receives semiconductor funding support from spending US$100,000 or more, and from increasing capacity by more than 5% in China or any other high-risk country. This means that China and other manufacturers can only concentrate on the 28nm or larger nodes that remain widely use in commercial applications. Here, the consumer market is so large that producers with economies of scale have a durable competitive advantage.
These geopolitical actions mean that semiconductor producers in the Global South, such as those in Malaysia and Southeast Asia, must take into consideration the huge geopolitical risks in making their strategic investments in domestic semiconductor production. No company or country can afford to be caught in the geopolitical cross hairs that ban access to market or technology.
That is, other than huge semiconductor giants like Intel, AMD, Lam Research, Broadcom and so on, which are foreign owned. Malaysian semiconductor companies actually operate as original equipment manufacturers (OEMs), which manufacture semiconductor products or equipment based on foreign specifications and technology without involving significant domestic research and development (R&D). Malaysia is now home to a mix of OEMs and original design manufacturers (ODMs) such as Globetronics, Vitrox and others. This business model shift means that as domestic companies learn to invest in R&D and home-grown patents and intellectual property, there will be an increase in domestic value added, including upgrading of human talent and skills. It also means that to advance in technological capabilities and capabilities, talent upgrading and reskilling is critical.
Because Malaysia relies heavily on foreign direct investments (FDIs) that utilise its strategic business location and natural resources, our products contain a high concentration of foreign technology. About 40% of Malaysia’s exports come from the electronic and electrical engineering technology sector, with export earnings of RM575 billion in 2023. According to Deloitte’s, Malaysia is a net borrower of foreign technology with less than 20% of Bursa-listed companies categorised as technology companies at the end of 2023. Malaysia’s R&D to gross domestic product ratio is 1.08% of GDP in 2022, less than half of that of China (2.54%) and way below that of Organisation for Economic Co-operation and Development countries (2.72%), especially the US (3.36%).
This implies that if Malaysia is to break out of the middle-income trap, devotion to a knowledge-based economy that upgrades its technology and know-how continuously must be the way to go. Malaysia also cannot upgrade without a major overhaul of its education and labour-reskilling policies. You cannot attract high-value FDIs without offering highly skilled engineers and technicians. Cheap low-value labour and cheap land are no longer enough.
Economic statecraft in an age of intense geopolitical rivalry means that smaller countries must live by their wits, skills and nimbleness to shift gears and directions when needed for survival. Access to the major markets will always be available when we have niche technological advantages that give us competitiveness on intellectual property rights, rather than relying on natural resources alone.
It is time to act quickly on our R&D and human skills. Just providing funding is not enough, because it is the real economy transformation, with a cohesive partnership between academia, business and the civil service that will make the difference in a brutally competitive world.
Goodbye laissez-faire, hello smart intervention.
Andrew Sheng writes on Asian global issues. Loh Peixin is a research associate at George Town Institute of Open and Advanced Studies, Wawasan Open University. The writers are engaged in a major study of the tech industry in Penang.
Save by subscribing to us for your print and/or digital copy.
P/S: The Edge is also available on Apple’s App Store and Android’s Google Play.
Copyright © 1999-2023 The Edge Communications Sdn. Bhd. 199301012242 (266980-X). All rights reserved