“Is ‘all made in China’ the right strategy?” questioned Prof. Wei Shaojun in his keynote speech at Aspencore’s 2020 Global CEO Summit earlier...
During such tumultuous economic times, “Is ‘All Made in China’ the right strategy?” questioned Wei Shaojun in his keynote speech at Aspencore’s 2020 Global CEO Summit held here this week.
In the midst of a worsening trade war with United States, “All Made in China” has become a popular rallying cry in China in recent months among local chip vendors. Chinese IC companies say they are committed to getting everything done in China — from design to manufacture and supply — with little reliance on technologies, products and suppliers from abroad.
Wei, professor at Tsinghua University and chairman of IC design branch of China’s Semiconductor Industry Association, cautioned in his speech: The idea of “’All Made in China’ is very hot, and is getting too hot.”
Given difficult environments the industry is in today — both internally and externally, he suggested that China should tone down the rhetoric, and explore alternative strategic plans, which would allow China to leverage its huge market and good existing foundation. The goal is to lead China to a bigger progress in the next five to 10 years.
It was at the Central Foreign Affairs Working Conference in June 2018 when President Xi Jinping first spoke about “major changes unseen in a century” China is facing today.
Although there are diverging opinions as to what exactly “major changes unseen in a century” are, there is a common thread: The progress and development of science and technology are driving “major changes.” It’s critical to understand that such unprecedented changes have far-reaching impact, accompanied by a lot of uncertainties.
At the summit, Wei showed a graph of “Global GDP Growth in the Past Fifty Years.” An analysis of the data found that in the 31 years from 1970 to 2001, the cumulative global GDP was US$568.3 trillion, an average of $18.3 trillion per year. In the 19 years from 2001 to 2019, the total global GDP reached $1,172.5 trillion US, an average of $65.1 trillion US per year, 3.6 times that of the previous 31 years.
In nearly 20 years, what caused such a high growth in global GDP? Is it oil, coal, transportation, agriculture or what?
None of the above.
After 2000, Internet technology, mobile communication technology, especially the combination of the two — mobile Internet technology gradually moved towards global unification, prompting the rapid development of the global economy.
We can see that the fastest growing in the past 20 years is the information industry.
Let’s look at another set of data.
In the 15 years from 1987 to 2001, the cumulative revenue of the global semiconductor industry was $152.4 billion, an average of $10.2 billion per year.
During the 18 years from 2002 to 2019, the cumulative revenue of the global semiconductor industry reached $5,221.3 billion, an average of $290.1 billion per year, 2.9 times that of the previous 15 years.
We can see that there is a strong correlation between the development of the semiconductor industry and GDP, and the semiconductor industry strongly supports the development of the information industry.
The same is true in China. China’s rapid economic growth in the past two decades has also come from the information industry.
China’s IC industry 2004~2019
From 2004 to 2019, Wei pointed out, China’s IC industry grew rapidly, with output value increasing nearly 14 times, with an average annual compound growth rate of 19.2%, far higher than the global average annual compound growth rate of 4.5%. In 2019, China’s integrated circuit industry continued to maintain double-digit growth, with annual sales reaching 756.23 billion yuan, a year-on-year increase of 15.8%. (Data source: publicly compiled data, China Semiconductor Industry Association, 2020)
In the past 15 years, the chip design industry has developed rapidly. It is the only link in the three industries where the annual growth rate has been positive in the past 15 years and has become an important locomotive for the development of my country’s integrated circuit industry, Wei observed. China’s design industry has surpassed Taiwan to become the world’s second largest design industry cluster, and its share in the global integrated circuit design industry has increased from 3.56% in 2004 to 42.99% in 2019.
In terms of chip manufacturing, the scale of the industry has maintained steady growth, with an average annual compound growth rate of 17.96%. Under the guidance of the “Outline” and the strong pull of the Big Funds, China’s IC manufacturing industry is ushering in a new round of rapid growth. Since 2014, the average annual compound growth rate of the manufacturing industry has been 24.72%.
Opportunities: ‘Strong demand, insufficient supply’
Wei said that our domestic chip products account for 10.3% of the global market. They are mainly IC design with almost no IDM, so high-end chips are still heavily dependent on external sources. We cannot get rid of the dependence on imports, Wei observed. Of course, this is also normal. We produce most of the world’s electronic products, so it is not incomprehensible that we use the world’s high-end chips, he explained, but what we definitely want to see is the development of our own industry.
Challenges: ‘natural disasters’ and ‘man-made disasters’
Coupled with the accelerating Sino-US conflicts, some people advocate the decoupling of industries. Both the US government and the Japanese government are actively promoting the return of manufacturing industries. However, realistically speaking, these moves did not play a big role. After all, China has a huge market and a relatively complete industrial chain supporting its ecosystem. What’s more, under the pandemic, China was the first to control it and fully resume work and production.
In contrast, the current situation of the foreign pandemic situation is still severe, especially in the recent past.
China and the United States have long been integrated into the global technology system, and it is not realistic to completely decouple.
IC strategies China’s industry should consider
In the context of the global pandemic, and the China-US technology war, China’s semiconductor industry faces both opportunities and challenges. Although we are under tremendous pressure, Wei said, I believe this is the right time for us to calm and clear our minds.
1. China is into the global technology system, it’s impossible to go back
China has integrated into the global technology system and it is impossible to go back. In the past two years, many domestic experts and industry executives have called for us to build another system. I think this idea is wrong, because we have already an integral part of the global technology system.
In the next-generation technologies such as quantum, space, etc., where there is no international technology system, so we can go our own way.
But in other electronic components, electric vehicles, consumer electronics, equipment, artificial intelligence, we play in almost all of them, an important part of international standards and international technology systems. Therefore, we cannot leave the global technology system and go back. It is impossible to build an alternative universe.
According to the research data from the McKinsey Global Institute, China is still highly dependent on many key technologies such as electronic components, electric vehicles, consumer electronics and the Internet, robotics, artificial intelligence, and next-generation technologies. The data shows that in many technologies, China has a very high proportion of adopting global standards.
Although the proportion of domestic suppliers is relatively high, the comparable ratio of Chinese suppliers is very low. This also means that the source of many technological innovations is not in China, and China needs to continue technological exchanges with the world.
From another perspective, we have a domestic market share of more than 50% in the fields of photovoltaic panels, high-speed rail, digital payment, smart phones, cloud services, robots, etc., Wei observed, but in foreign markets the market share, except for photovoltaic panels and smart phones, is very low.
Wei said, “We must continue to develop these technologies, but we can go global and seize the global market. If we don’t go outside, how can we develop and grow? In addition to semiconductors, we also need to go international, so internationalization is ours. We must adhere to the general direction.”
2. Decoupling ‘harms others and disadvantages ourselves’
According to data released by the American Semiconductor Industry Association, the value of integrated circuit products sold by American semiconductor companies in China in 2018 exceeded $100 billion, and sales in China accounted for half of the revenue of many American companies. Obviously, in the case of China-US technology decoupling, US semiconductor companies will not benefit from it. At the same time, China-US technology decoupling will greatly affect US semiconductor leadership. Therefore, the American Semiconductor Industry Association and many semiconductor companies in the United States also oppose the US government’s move to decouple from China.
According to a report made by the Boston Consulting Group to the American Semiconductor Industry Association, when the United States did not deliberately restrict China in technology in 2018, the global market share of American semiconductor companies was 48%, with revenues as high as $226 billion. Research and development investment reached $40 billion.
However, after the gradual decoupling of Chinese and American technologies, only according to China’s 2025 plan, China will replace 15-40% of American suppliers. Then the revenue of American semiconductor companies in the global market will be reduced to $205 to 220 billion. The share will fall to 43-46%, and the reduction in revenue will also reduce the R&D investment of US semiconductor companies to 36-39 billion yuan.
If 100% technology decoupling between China and the United States takes place, then the global revenue of US semiconductor companies will be further reduced to $143 billion, the market share will drop to 30%, and R&D investment will also fall to $16-28 billion. The leading position of American semiconductors will be greatly weakened and will be overtaken by the Korean semiconductor industry. In the long run, it may be overtaken by China.
3. China is already on a favorable track in IT, industry
In the field of information technology and industry in China, we are already on a relatively strong track, Wei said. He segmented the industry into Internet, mobile communications, artificial intelligence, autonomous driving, quantum technology, integrated circuits. (See the table below. The above categories are listed from left to right). A smiling face means good development, crying face means bad development. Of course, this is relatively speaking.
We have seen that China and the United States have the most smiling faces, with no crying faces.
China and the United States have a relatively good development trend in information technology. This has a lot to do with China’s large-scale development of information technology after 2000.
Of course, China and the United States have their own shortcomings. The United States is not as good as China in mobile communications networks, but the United States is indeed stronger than us in semiconductors, Wei pointed out. Therefore, we are on a favorable track, and this should not be confused.
4. China must guard against extremism and closed development
The Prism Project (Editor’s note: Edward Snowden claimed, under the NSA’s program called Prism, the US has been hacking servers in the Chinese mainland and Hong Kong for years), the ZTE incident, the Huawei incident, the Sino-US trade war… The reality has told us more than once, cybersecurity and national security have faced challenges, which have become the root causes of conflicts.
Every nation is concerned about some of their core technologies and equipment are being controlled by others. In this context, many nations see it is their best interest to pursue “localized substitution” and solutions that are “independent and controllable.”
So, what should replace “Made in China”?
There is no doubt that some foreign products that have occupied the Chinese market for decades are superior in performance and speed, and the process of localization has been long and painful. But we must insist on doing so, Wei argued.
If it is not for autonomous and controllable products, Chinese industry may be paralyzed within a day. This is not alarmist. Cyberspace has become the country’s “fifth frontier” beyond the four territories of land, sea, air, and sky. To ensure the security of cyberspace is to protect national sovereignty, and its importance has become increasingly prominent. If localized substitution is a war, then basic software and hardware such as CPUs, operating systems, and databases are autonomous and controllable “frontal battlefields” and are the foundation and guarantee of national network security.
5. Re-examine the five major sectors of the semiconductor industry centered on products
Wei pointed out:
In the past two years or so, our industry has been talking about “localization substitution” (substitution by home-grown products)
I think we must prepare for the limit and determine where the bottom line lies. We cannot replace our technology development with this kind of thinking. The main theme of the semiconductor industry should be “to open up and cooperate.”
There is no substitution for this.
I still think that the Sino-US semiconductor industry must compete in order to develop. We must take the product as the center and re-examine the five major sectors of the semiconductor industry: design, manufacturing, packaging, testing equipment and materials.
In future development, we must pay special attention to the balanced development of these five areas. At present, we are slightly ahead of the design sector, but the material sector is relatively weak. And yet, I believe that the time is still on our side and it is not that terrible.
In addition, Wei observed that we (which is to say, Chinese companies) were unable to produce integrated device manufacturers (IDM), because we thought foundry-and-design model was the best.
Now this view is changing. While continuing to pursue the “design-foundry” model, it’s time to vigorously develop IDM. The typical IDM industry like memory is slowly advancing in China.
A good example is Wuhan-based Yangtze Memory Technologies Co., Ltd. (YMTC) launched in July 2016. Their main three-dimensional flash memory (3D Nand) plant was completed in September 2017. The company entered trial production in October 2018, and it began small-batch mass production in 2019. The technical level has also reached 32-layer 8GB, 64Gb, and 64-layer 3D-NAND, with 128Gb on its way; a breakthrough will be made in 128-layer 3D-NAND in 2020; the overall technology has reached the international advanced level.
Another example is Hefei-based Changxin Memory launched in May 2016. With the main business of DRAM products, the plant was completed in January 2018, trial production began in June, the yield rate reached 20% in August, over 80% in October, and mass production in December. Its 19nm 8GB, DDR4 is at the international advanced level; in the third quarter of 2019, 8Gb LPDDR4 is mass-produced. The 17nm process will be completed in 2021.
6. Grasp business opportunities on the offensive
Finally, Wei said that China took the lead to get out of the impact of the new coronavirus epidemic, and China’s economy has recovered quickly. On the other hand, the US’s suppression of China’s technology industry has forced many Chinese companies to stock up large quantities of U.S. products. Of course, a large amount of inventory also has great risks. For example, after the new president of the United States takes office, there will be policy adjustments and the opening of some devices, which will make stockpiling risky.
Recently, some chip projects have caused widespread concern in the market. Recently, Meng Wei, spokesperson of the National Development and Reform Commission, responded to this at a regular press conference held by the National Development and Reform Commission. Meng Wei said that the enthusiasm of domestic investment in the IC industry is constantly rising. Some “three-nos” companies with “no experience, no technology, and no talents” have joined the IC industry. Some companies have insufficient knowledge of the law of IC development and blindly start projects. The risks of horizontal duplication of construction are apparent, and even the construction of individual projects is stagnant, and the factory buildings are empty, causing waste of resources.
In this regard, Wei said that we must respect the law of industrial development and overcome the rapid development shooting for quick success.
Since 2007, China’s wafer manufacturing capacity has increased rapidly in the world, far higher than other countries and regions. In 2019, China has 199 integrated circuit wafer manufacturing production lines (above 4 inches), of which there are 28 12-inch production lines and 35 8-inch production lines (including 1 pilot line). The enthusiasm for investment and construction of factories in various places is high, but a number of manufacturing projects are facing unfinished shutdowns. The blind impulse that violates the development law of the semiconductor industry is worthy of vigilance.
In addition, Prof. Wei also emphasized that China should humbly learn from the US semiconductor industry and increase investment in innovation. Since 1988, the United States has greatly outperformed other high-income countries in terms of productivity growth and real GDP growth. The leading position of technology has enabled American companies to establish a virtuous circle of innovation. Large-scale R&D brings excellent technology and products, which in turn brings higher market share and higher profit margins, so that they can invest more in R&D.