IC industry executives in China discuss the current semiconductor market and what needs to happen in order for the country to be successful in this space.
China is making progress becoming less reliant on foreign supply of semiconductors, according to Chinese IC industry executives, but the country’s efforts still require setting specific intermediate goals and prioritizing those goals.
The executives discussed these issues during a recent roundtable event called "Countdown: How Far is China from 40% IC Self-Sufficiency?" held during the 2019 China IC Leader Summit, jointly organized by AspenCore's EETimes China, EDN China, and ESM China Electronic Business.
China has been the largest consumer of integrated circuits (ICs) since 2005. In 2015, the State Council of China adopted an ambitious plan for the China electronics industry to become self-reliant. The "Made in China 2025" plan set a strategic goal to grow the country’s IC industry to achieve a self-sufficiency rate of 40% by 2020, increasing to 70% by 2025.
Three years later, China was importing more chips, not fewer. According to the 2018 National Imported Key Commodity Value Index, published by the State General Administration of Customs, China imported 417.57 billion integrated circuits in 2018, with a total amount of RMB 2058.41 billion (US $312 billion), an increase of 19.8% over 2017. This is the first time that China’s IC imports have exceeded US $300 billion.
Which markets should China try to win? What are the obstacles to China achieving chip self-sufficiency? Given that China lacks internal sources for basic device IP, electronic design automation (EDA) tools, and semiconductor manufacturing equipment, how can China create a path to achieve semiconductor self-sufficiency?
The guests from the Chinese semiconductor industry at the 2019 China IC Leader Summit roundtable discussed the unavoidable problems encountered in the development of China's IC industry — and offered advice from various perspectives — on how to proceed with developing a home-grown industry.
Which Markets Must be Won?
With the goal of improving self-sufficiency, which areas must be won by the domestic semiconductor industry chain? Dr. Wayne Dai, VeriSilicon CEO, believes that the key is to solve the bottlenecks, such as the huge amount of memory products required.
Andy Miao, deputy general manager and technical director of Shenzhen Micro Semicon agreed that in the storage market, China is making good progress, but this is a market that is easier to break into. By contrast, microcontroller (MCU) is a US $20 billion market, 99% of which is still in the hands of international manufacturers.
If Chinese companies are to reduce their reliance on imports, it will be a long-term process, including upgrading the level of analog technology and quality control, he said.
Invest in Foreign Tech, Keep it Low-Key
Miao also pointed out that the premise of the 40% self-sufficiency rate is that China would gain control of the technology needed in order to vault ahead, which could include directly purchasing foreign technology. However, the IC industry is a strategic technology for all countries, ownership is politically sensitive, and it is best to act low-key in terms of investments and acquisitions.
"At present, our most important goal is to accumulate basic skills, put the numbers aside, take it step-by-step, don't rush success," Miao said.
YanQiu Xia, director of strategy and business development of HiSilicon, believes that the 40% figure is more of a guideline from the government. It is still necessary to achieve results, however, and each company must set its own goals.
"It’s like when parents want their children to be admitted to a prestigious school. Can they all be right? Parents can’t decide, but they have to rely on their children’s internal drive. Just like a company’s goal, perhaps it’s going to be 60%. So why do you stick to 40%?" Xia said. "If a company starts from 0% and reaches 10%, it has already achieved great success — it should not be a goal set for a specific company. Enterprises should upgrade their own competitiveness, continuously evolve and develop, and strive to achieve this result."
Su Weimin, chairman of the ZigBee Member Group China, spoke from the perspective of standards bodies. "The chip is the basis of communication. Although 14 chip manufacturers are currently certified by the ZigBee Alliance worldwide, only one Chinese company is certified,” said Weimin. “Nonetheless, in the field of communications, Chinese chip manufacturers have made significant progress."
Wei Xu, secretary of the Shanghai Integrated Circuit Industry Association (SICA) and executive vice president of Shanghai HHGrace Semiconductor Manufacturing Corporation, hopes that the cooperation among electronics companies in China will bring about an increase in the self-sufficiency rate of chips.
"The IC industry chain is very long, from design, manufacturing, packaging, testing, to materials,” he said. “There is no single country that can dominate the world. China should consider the integrity and construction of the industrial chain, but in a completely closed environment. The country began to build fabs in 2014, so [there is progress] from the perspective of capacity. In 2020, the localization rate of chips will make great progress."
Open Up “No Man's Land,” Innovation is King
There might be an opportunity for Chinese electronics companies to compete with foreign manufacturers by joining them at the starting line of the "no man's land” — markets that are still so new. For example, artificial intelligence (AI) and driverless vehicles are markets in which industry market standards are relatively in their infancy. Can China’s industry exploit this opportunity to develop its own chip architecture and related technological ecosystem?
Dai said it could be profitable to look for opportunities at the “edges” of markets. For example, he believes that smart audio and wireless headphones could be an entry point. For example, wireless headphones can also be used as hearing aids. These new products are not yet widely available, they have bright prospects, and would represent large quantities.
Xu said that in the post-Moore’s Law era, application drivers could be an advantage for the Chinese electronics industry. China could leverage its huge internal market to set worldwide trends and determine the standards of the whole world. Future applications include areas such as smart cities and healthcare. A country with a large population is likely to do this in a creative mindset or atmosphere.
Miao agreed with Dai and Xu — look for opportunities on the edges. The biggest driving force for all design innovations comes from the terminal's performance and system requirements. China is very successful as a manufacturing country. Huawei, for example, creates Internet of Things (IoT) platforms and is big enough to set its own standards.
“Once a certain standard is formed, Chinese companies have a good chance to achieve leadership,” Miao noted. “Especially in the field of home appliances and industrial control, IoT will land first. This is the most direct place for human-computer interaction, and it is also the best field for Chinese manufacturers, including Haier, Midea, and Gree. Gree's home appliances have already taken the lead in the world.
"Systems and home appliance manufacturers need to cooperate with chip manufacturers in optimizing their systems,” Miao continued. “Chip manufacturers can help them improve system integration and reduce costs." He believes these companies will set new standards for the market.
Would these new standards apply only to the Chinese market, or could China set standards for the global industry?
Xia thinks that it is first necessary to determine where standards-setting power comes from. Historically, the Chinese government has participated in formal standards activities only in the IC industry. The supply chain in the communications industry is dispersed, however. Consequently, individual corporate standards could influence the progress of this industry.
Initiative Gets the Lead
"For example, Apple dared to cancel the earphone hole,” Xia said. “Huawei also changed the specifications of the memory card before. This is not what any government agency or industry organization forced them to do, but they took the initiative to do something to lead the industry. Other manufacturers follow."
Unilateral action does not automatically translate into market leadership, however. Sometimes other manufacturers don’t follow another company’s lead. "Several top manufacturers may not agree. For example, Midea and Haier might not agree to the same standard,” Xia said. “Whoever becomes the real standard depends on the outcome of market competition."
There are too many games involved in this, which is one of the reasons why so many people use Google's tensor processing unit (TPU). Few are concerned about Google's ability to continue leading the market. Meanwhile, a small company that open sources something might produce an innovative product, but it still might not get market traction. It's not because the product isn’t good, it’s the suspicion that a small company might not have the resources to continue innovating in the next year, the next year, and the year after that.
Xia added, "So when we set standards for new fields, don't think about not following international mainstream manufacturers at all. No one believes that Chinese manufacturers can defeat Apple or Google within 10 years, but we must aim at our own five-year plan. The goal is high, the action will not be deterred when the competition leads the entire industry chain to establish their so-called standards, in order to continue to compete with them.”
Weimin also believes that an emphasis on national autonomy might not be the most effective strategy in the context of globalization. Some domestically-enforced standards have forced Chinese companies to neglect cooperation with the international community. In a closed system, a country or an enterprise can set a standard. For example, the China high-speed railway system operates largely on its own.
Bold Cooperation Sets Standards
According to Weimin, "If Chinese companies can achieve leadership in a certain field, they should boldly cooperate with international companies, or establish a business alliance to set standards and directly target international companies. Conversely, if a company is in a weak position doing one's own standard will hurt the company."
Senhua Dong, deputy general manager of Empyrean Software Co. Ltd., believes that AI is just an enabling technology that can be used in different chips and systems. On the application side, it is more suitable for China's development stage, because China has a huge market. Talking about China surpassing the United States in AI actually has a negative effect on the healthy development of the industry.
"If we step back and think about it, we will find that most of the basic algorithms for AI are still in the hands of the United States. From a technical point of view, we are still in a relatively weak position," Dong said. "But from the perspective of applications and market, China's large number of home appliances and system manufacturers have strong demand, which can provide many improvements for our AI algorithm and chip design. This creates a synergistic combination can drive the development of domestic AI technology."
Liansheng Cao, vice president of Advanced Micro-Fabrication Equipment Inc. (AMEC), expressed his views from the perspective of the “no-man's land.” He believes that China's entire IC industry is still catching up with market leaders, with a few exceptions that include Huawei.
“In this kind of ecosystem, can IC practitioners have original things?" Cao went on to say, "I understand the no-man's land has two forms. One is 'there is no one that I can chase who is better than me.’ Then the question is: what should I do now? The other is the idea of ‘completely original,’ which is especially important for young people in the industry. Don't give up the spark of innovation. It's very difficult, but very valuable.”