While Taiwan TSMC reports record profit by producing more than 90% of most advanced chips, USA’s Intel reports a loss in making chips with two-generation older technology (as of the 2nd quarter of 2022). Hence, the USA and China want to control Taiwan to regain and gain silicon supremacy—sowing the seed of chip war.
In an article in New York Times, Edward Wong and John Ismay depicted a picture of the USA’s mission of turning Taiwan into a weapon depot. To give a further vivid picture of the impending war around silicon chips, Chris Miller has come up with a book—Chip War. As silicon chips have been powering both incremental advancement and reinvention of all kinds of products, maintaining an edge on it is crucial for sustaining the USA’s superpower status. On the other hand, China would like to be at the edge of semiconductor chips. It’s essential for pursuing the mission of developing an innovation economy and enhancing military might. Therefore, the USA would like to regain its edge and prevent China from having it—creating a warlike situation.
Despite invention and early progress, how did the USA partially lose silicon supremacy to Taiwan, South Korea, Japan, and the Dutch? Is it due to tax differentials and subsidies Taiwan now makes better chips than American companies like Intel? Hence, by offering subsidies and decoupling the US-led chip supply chain from China, will the USA regain the edge and prevent China’s rise? To answer these and other pertinent questions, we need to review the history of the rise and fall of the dominance of geographies in different layers of the semiconductor value chain.
- The growing importance of semiconductors in gaining innovation edge—fuelling chip war
- The invention of the Semiconductor industry by the USA—gave birth to the Chip War
- USA industry became obsessed with military and space missions
- Rise of Japan as a global semiconductor powerhouse due to the reinvention of consumer electronics also caused a chip war
- USA regained silicon edge due to the PC wave
- The USA lost its silicon edge to Taiwan due to overlooking the rise of fabless models and smartphones–causing the current Chip War
- China became a factory of electronics—skyrocketing semiconductor import
- Distributed monopolization of the semiconductor value chain
- Winning chip war through subsidies and regulation runs the risk of failure
- Focus on the next wave for profitably winning the chip war
As we move to the next semiconductor technology node, chips get better performing, less power-consuming, and less costly. Hence, from weapons and manufacturing equipment to autonomous vehicles, superior performance depends on the underlying semiconductor chips. Consequentially, to gain and sustain an innovation edge in both military and commercial applications, the strategy is to attain the edge in chips and prevent adversaries from having it. Therefore, the USA and China are at loggerheads-triggering Chip acts in the USA and Europe, and restrictions on technology access.
The invention of the Semiconductor industry by the USA—gave birth to the Chip War
The semiconductor is a material that is neither insulator nor a conductor. Hence, scientists developed a technique to turn it into an insulator (off) or conductor (on). They did it by changing the polarity of an electric field to an impure semiconductor—making it a solid-state switch, known as Transistor.
In 1947, Dr. Shockley and two of his colleagues made this critical invention to replace the electromechanical telephone switch. They received Nobel Prize for it in 1956. However, despite its growing success, it began the journey in a humble form. It was noisy and fragile and expensive to make—one by one. Hence, in April 1952, Bell Labs issued its license to 40 companies that paid a $25,000 patent licensing fee each for further advancement and commercialization. In addition to large American firms, a small Japanese firm Sony paid for the license. This event led to rivalry among competing firms in making transistors increasingly better performing and cheaper for exploiting profit-making prospects. By the way, although American firms mostly became busy serving the military market, Japanese firms, like Sony, focused on consumer electronics products.
USA industry became obsessed with military and space missions
Although in 1954, Texas Instruments designed and manufactured the first transistor radio, American companies primarily focused on the defense and space markets. Due to far smaller weight, space, and power footprints, fighter planes and space vehicles found transistors as better alternatives to vacuum tube devices. For example, IBM bought all the transistors of the first batch that Fairchild produced for $150 apiece for building an onboard computer for the B-52 bomber.
Companies like Fairchild and Texas Instruments focused on miniaturization to make transistors increasingly preferable to vacuum tube devices. Consequentially, in 1959 at Fairchild, Robert Noyce invented the first monolithic integrated circuit (IC). Jack Kilby at Texas Instruments also made similar advancements during that time. As a result, Transistor gained further ground. For example, for building an onboard guidance computer for the Moon mission, in 1962, the MIT instrumentation lab estimated that a computer made out of Noyce’s chips would make the same computer one-third smaller and lighter than a computer made out of transistors. Subsequently, the decision to use Noyce’s ICs for onboard computers for NASA’s lunar mission led to Fairchild’s sell ballooning from $50,000 in 1958 to $2 million two years later.
Like Fairchild, Texas instruments and other American firms looked at the defense and space market to profit from Transistor. For better computers to hurl nuclear warheads to strike the Soviet Union, Texas Instruments proposed Kilby’s IC-based, less heavy, and more powerful computer to US Air Force’s Minuteman program. By the end of 1964, Texas Instruments supplied 100,000 ICs to Minuteman program alone for building an onboard computer to guide missiles. By 1965, as high as 20% of all the ICs delivered that year went for the Minuteman program. Hence, silicon edge significantly boosted the US military power over the Soviets during the cold war era.
Rise of Japan as a global semiconductor powerhouse due to the reinvention of consumer electronics also caused a Chip War
Like Americans, Japanese companies did not have the option of profiting from licensed transistors in updating war equipment. Hence, Sony and a few other Japanese companies focused on the consumer market. But unlike the military and space programs, civilians were not eager to pay a premium price for less heavy transistor-based radios, televisions, or music players. Furthermore, transistors were noisier than vacuum tube devices in processing analog signals, resulting in poor-quality sound and video reproduction. But Japanese companies had no other options to profit from Transistor.
Sony led Japanese companies focused on reinventing consumer electronics products. The journey began with Sony’s pocket radio as an inferior alternative to USA’s and Europe’s radio receivers. Hence, Sony and other Japanese companies focused on refining the design of transistors and production processes. Consequentially, Japanese transistor-based electronics products started gaining momentum, resulting in the growing demand for semiconductors. Consequentially, by 1957, Japan became the largest producer of semiconductor devices. And Japan maintained this position till 1986.
Contrary to the common belief of subsidies and tax deferential, Japan outperformed America due to superior performance in advancing transistors. As opposed to imitating and copying, Sony led Japan focused on scientific discoveries for technology refinement, leading to winning Nobel Prize by one of Sony’s Engineers in 1974. Hence, the USA’s first loss of silicon edge was due to Japan’s better performance on incremental advancement for pursuing creative destruction waves in consumer electronics. Therefore, in addition to being the top producer of semiconductors, Japan unleashed disruptive innovation effects on RCA and a few other western firms. Japanese firms also started gaining traction out of silicon chips in business computing opening new doors. For example, Busicom’s assignment of developing ICs for calculators led to Intel’s Microprocessor 4004. It had been a precursor to 8086/8088, used in IBM PC.
USA regained silicon edge due to the PC wave
In the early 1980s, Japanese NEC was the largest global chip producer. Japanese companies also specialized significantly in wafers, fine chemicals, and process equipment by this time. Hence, the USA felt threatened, applying punitive measures against the Japanese, forming SEMATEC, and giving grants to American firms. Notably, upon receiving grants, several American photolithography firms failed to survive, turning Japanese Canon and Nikon into global leaders.
The turnaround of the USA’s chip industry did not emerge from punitive measures and subsidies. It happened due to the Intel-led rise of PC wave for IBM’s selection of Intel 8088 for its PC. And Apple chose Motorola’s microprocessor for its Macintosh. Due to the rapid growth of the PC wave, American firms like Intel, AMD, Motorola, and a few others started accelerating the production of memory and logic circuits. Consequently, Silicon Valley became the world’s most significant and finest chip producer. Along the way, American firms like Cadence, Mentor Graphics, and Synopsis surfaced in offering design automation software to chip designers. Intel and many others perfected their integrated device maker (IDM) model.
Due to the growing popularity of chips for incremental advancement and reinvention of an increasing number of electronic products, demand for the 3rd party chip design and processing services started to grow. But American IDMs like Intel were not interested in it, as the volume and profitability were meager. By the way, this is a typical response of dominant firms riding a matured wave (PC-led IDM)–sowing the seed of failure. Subsequently, this trend grew and formed a new model—fabless chip companies and foundry service providers. Consequentially, it has grown as a creative wave of destruction, unleashing disruptive innovation effects on IDMs like Intel.
The USA lost its silicon edge to Taiwan due to overlooking the rise of fabless models and smartphones–causing the current Chip War
While fabless IC firms started growing in the USA, Taiwan’s ITRI spun off TSMC (in 1987) to process their designs into chips, forming the foundry model. Unlike the PC, fabless IC companies started showing a growing interest in the requirement of mobile handset makers. They were under the stress of increasing computational capability while making chips less power-hungry. Hence, there was a demand for increasingly more sophisticated process nodes for making chips for mobile handsets. Notably, the emergence of the software-centric design of smartphones, leading to the iPhone in 2007, led to an explosion of sophistication of mobile handset processors.
But American IDMs like Intel kept overlooking it. Even Intel rejected Apple’s request to make the processor for its first iPhone. Hence, Apple was compelled to go to Samsung to get its chip using older 90 nm technology, while Intel was churning out chips with its latest 65 nm process node. Intel’s such rejection sowed the seed of the rise of the Silicon edge of Asia and the fall of America. Hence, it was not due to tax differentials or subsidies the USA got into the declining trend.
The rapid growth of smartphones led to accelerated sophistication of foundries. Due to superior process R&D performance and customer relation management, TSMC kept outshining, reaching the edge. Consequentially, while TSMC was processing silicon for Apple, AMD, and many others with its 5nm process node, Intel was struggling to migrate from 10nm to 7nm. Hence, Intel kept falling behind due to its poor performance and miss of the smartphone waves, reporting a loss in the 2nd quarter of 2022. On the contrary, due to riding on the smartphone wave with the most sophisticated process technology, TSMC reported a rise in profit and revenue by 76% and 36%, respectively.
China became a factory of electronics—skyrocketing semiconductor import
Over the decade, along with Apple, multinationals have flocked to China to get their product manufactured. As a result, the import of Chips into China has accelerated. Consequentially, China’s semiconductor imports reached $277 billion in 2021. Furthermore, China heavily relies on Taiwan’s TSMC; It imported US$79.4 billion worth ICs during the first half of 2022 from Taiwan. Besides, China would like to acquire an innovation edge in diverse sectors, starting from cellular communication to Robotics. Hence, China urgently needs to increase the volume and sophistication of domestic production of chips. For a very logical reason, America finds itself a security threat. Therefore, all measures are underway to block China from attaining silicon edge and prevent the import of high-end chips from Taiwan. Furthermore, due to the growing security threat, the USA would like to protect Taiwan militarily.
Distributed monopolization of the semiconductor value chain
Semiconductor or chip production requires inputs from diverse suppliers. For producing high-end chips, those supplies must come from specialized suppliers. Due to growing efficiency and productivity improvement, each of the suppliers has attained hard to imitate edge. As a result, each significant layer of the semiconductor value chain has a high-level monopoly. For example, Japan has achieved a monopolistic situation in wafers, chemicals, and process equipment.
American firms have unbeatable strengths in chip design software, designs, and fabless companies. Notable ones are Cadence, Mentor Graphics, and Synopsis in EDA tools. Besides, US companies like KLA, Applied Materials, and Lam Research have a strong position in fab equipment, enabling companies like TSMC to produce sub-10-nm chips. And Dutch firm ASML has an absolute monopoly in high-end (EUV) photolithography machines. On the other hand, TSMC and Samsung have an edge in the latest process node. Furthermore, Taiwan also has a strong foothold in chip testing, bonding, and packaging.
Winning a chip war through subsidies and regulation runs the risk of failure
Due to the race to offer increasingly sophisticated chips and unfolding innovation waves, the semiconductor value chain has become highly monopolized. It has resulted from the accumulation of intellectual assets and supply chain optimization. Hence, the common belief of tax differentials or subsidies is not the underlying reason. There is no denying that they may have played a role in creating intellectual assets and supply chain optimization. Therefore, tax incentives or subsidies alone will not likely succeed in dealing with competitiveness. If chips are poorer due to older technologies, any amount of tax difference will do nothing.
Yes, China’s close to $70 billion for acquiring foreign firms and developing domestic capacity has opened access to technology and production capacity. But that is not sufficient for China to grow as a dominant player. The success will depend on how well Chinese firms outperform the competition and attain price-setting capability due to quality. For example, ASML is a success story as no one else can deliver a similar lithography machine. Similarly, Intel has been falling behind as it cannot keep pace with TSMC in assimilating and fine-tuning the next process node.
Focus on the next wave for profitably winning the chip war
Chip war is not about imitation or replication game, which could be won through subsidies and trade restrictions. To be a winner, an actor must outperform the competitors through superior innovation performance—sustaining over decades. For example, Nikon and Canon became a duopoly in the 1990s in lithography as American firms kept failing. On the other hand, ASML has become a global monopoly as no one else could deliver EUV machines for sub-10 nm nodes. Similarly, Intel has been falling behind as TSMC has been showing superior performance.
Due to the high amenability of making chips increasingly better and cheaper through intellectual assets, winning firms must show superior performance in incremental advancement. Furthermore, unfolding waves have been reshaping the dominance. Hence, aspiring firms should pursue reinvention opportunities for unleashing creative waves of destruction. Besides, the R&D need for further advancement has been exponentially growing. Consequentially, it has been demanding a growing market for reaching the minimum efficient scale point. Hence, the USA’s strategy of dividing the supply chain as RED and BLUE channels runs the risk of weakening competitiveness.
Perhaps, Chip war is based on a weak thesis. It’s time to pursue a smarter approach to win the race. By the way, high-level monopolization in each layer indicates that the chip war has already been won. Victory has been achieved by not a single country but by many.
...welcome to join us. We are on a mission to develop an enlightened community for making the world a better place. If you like the article, you may encourage us by sharing it through social media to enlighten others.
- Semiconductor Value Chain–globally distributed ecosystem
- Semiconductor Monopoly Due to Winning Race of Ideas
- Intel Falling Due to PC and Mobile Waves
- ASML Lithography Monopoly from Sustaining Innovation
- Taiwan’s Semiconductor Monopoly – How did it arise?
- ASML TSMC Nexus Fuels Semiconductor Monopoly
- ASML Monopoly in Semiconductor — where is magic?