How Semiconductors Impact the World

Advanced chip designs hold the key for global technology and geopolitical dominance

Last month a frigid blast of winter weather across the US plunged Texas into an unusually icy emergency that knocked out power to more than 2 million people, froze 69 people to death, and hit the state with a$18 billion economic damage bill. It sent chills down global semiconductor supply chains, aggravating an already critical chip shortage for industries from automobiles to consumer electronics and the technology-media-telecommunication (TMT)sector. Consulting firm AlixPartners expects the shortage will cut $60.6 billion in revenue from the global automotive industry this year.

The shortage comes at a time when radically new chips are required to power 5G and the Fourth Industrial Revolution needs for every industry anywhere in the world. As every company becomes a technology company, chip supply chain disruptions can bring the global economy to its knees.

Austin, Texas is home to several important semiconductor fabs. Samsung has a large foundry that produces an estimated 90,000+ wafers per month, making a range of logic and flash memory chips using technologies as advanced as 14 nm (nano meters). This makes it one of the largest capacities and one of the most advanced fabs in the US. NXP Semiconductors, a major supplier of chips to the auto industry, has two locations on the south and east sides of town, and Infineon Technologies operates its Fab25 (a Former Cypress facility) there as well. All three companies shut down operations because of the power crisis.

The US$440-billion semiconductor industry has emerged as a vital geopolitical weapon in the war between the developed world and China. That apart, the exponential growth of computing processing required to power the Fourth Industrial Revolution hinges on technological breakthroughs in the semiconductor industry. At this point in time the US holds the key in this vital industry as it dominates chip design. The top-10 fabless chip makers worldwide, which designs the chip, includes four U.S. companies: Qualcomm, Broadcomm, AMD, and Nvidia. Taiwan has a 50% market share in fabricating the chips from the designs in its foundries.

The US has been choking off chip supplies to China by banning countries like Taiwan and others which uses US machinery in its foundries from selling to the communist country. This has forced Huawei to drop out of the 5G race and its fourth quarter witnessed a 40% decline in mobile phone sales, due to chip shortages. Analysts said that the outlook for Huawei’s mobile phone business will depend on chip supply – crucially, how the Biden administration will deal with chip ban when Huawei and other Chinese tech firms can develop and produce high-end chips.

Five Tectonic Shifts

The US administration is looking to invest $37 billion in funding to supercharge chip manufacturing in the United States as a shortfall of semiconductors has forced US automakers and other manufacturers to cut production. It sees five tectonicshifts that will define the future of semiconductors and the information & communication technology (ICT) industry.

1. Fundamental breakthroughs in analog hardware are required to generate smarter world-machine interfaces that can sense, perceive, and reason.

Annual investment need: $600M throughout this decade to pursue analog-to-information compression/ reduction with a practical compression/reduction ratio of 105:1 for practical use of information more analogous to the human brain

2. The growth of memory demands will outstrip global silicon supply, presenting opportunities for radically new memory and storage solutions.

Annual investment need: $750M throughout this decade to develop emerging memories/memory fabrics with >10-100X density and energy efficiency improvement for each level of the memory hierarchy. Discover new storage systems and storage technologies with >100x storage density capability.

3. Always-available communication requires new research directions that address the imbalance of communication capacity vs. data-generation rates.

Annual investment need: $700M throughout this decade for communication enabling data movement of 100-1000 zettabyte/year at the peak rate of 1Tbps@ <0.1nJ/bit. Develop intelligent and agile networks that effectively utilize bandwidth to maximize network capacity

4. Breakthroughs in hardware research are needed to address emerging security challenges in highly interconnected systems and AI.

Annual investment need: $600M throughout this decade for privacy and security hardware advances that keep pace with new technology threats and use cases (e.g., trustworthy AI systems, secure hardware platforms, and emerging postquantum and distributed cryptographic algorithms).

5. Ever-rising energy demand for computing vs. global energy production is creating new risk, and new computing paradigms offer opportunities to dramatically improve energy efficiency.

Annual investment need: $750M throughout this decade to discover computing paradigms/architectures with a radically new computing trajectory demonstrating >1,000,000x improvement in energy efficiency.

What About Others?

The Europeans too understand the need to invest in advanced semiconductor manufacturing if they are to stay in the race to dominate the technology landscape and not be held to ransom by other countries. It does have some players in the market, such as Infineon or STMicroelectronics, and the European Commission announced a joint initiative involving 19 Member States in December 2020, which could be worth EUR145bn. But the cloud sector shows that Europe is more likely to follow the localist than the protectionist option. GAIA-X was launched as a federated data infrastructure system focusing on Cloud services, with the aim to foster digital sovereignty. As it moved along, it started accepting non-European companies as partners, as long as they followed European values. We expect the same pathway with semiconductors; because Europe lacks the expertise and the scale to create a national champion – unless it is willing to spend and take the time to build one.

China’s approach is like the one it has used when it comes to Internet players or even telecoms equipment, with its overall aim to be the global technological leader in the 21st century. It realises chips are a weakness, as it is the biggest importer in the world, and it does not want to find itself in the situation where its access is reduced or even cut off through its reliance of foreign partners. It wants to become self-sufficient and independent in all technology sectors, and is willing to do whatever it takes, both in terms of time and money, to achieve its objective. It will take time for China to become a global leader, but if and when it does, it will look to export its expertise abroad, expanding the market for its own players, meaning that US-Chinese geopolitical tensions over technology will continue.