Analysis on the Current Situation of China's Semiconductor Industry

"Making chips costs a lot of money. The government supports me this time. In fact, my understanding is that money is not important, but it brings us warmth when we are dying. I hope that after seeing our report card today, whether the Zhongguancun, Haidian District and Beijing municipal government can give us more support."

In 2015, when it was rumored that Xiaomi was the most popular for "core making", there were different opinions on Lei Jun's determination to make chips in the market. At that time, it was said that Xiaomi made chips because Lei Jun was unconvinced and had been criticized for its weak technology R & D ability; Others say that it is for the patent issue in the Indian market; Some competitors say that the role of its brand is greater than the actual role; Some even say that it is the government's support.

If Lei Jun had not made a straightforward statement this time, many people would think that the statement of "government support" was very funny; The media, partners and fans on and off the stage of "confession" also saw Lei Jun's laughter after the confession PPT, which was also meaningful.

After all, the support of the 2 million pilot fund is not even a little finger for Xiaomi, which has spent 1 billion yuan to build a core. However, at the surging S1 press conference, Lei Jun publicly appealed for more support from Zhongguancun, Haidian District and Beijing municipal government, which made us have to ponder the current situation of China's semiconductor industry.

The government invested hundreds of billions of dollars in the semiconductor industry

In June 2014, the outline for promoting the development of national integrated circuit industry was officially released and implemented; During the 13th Five Year Plan period, the central government's fiscal and tax preferential policies for the semiconductor industry mainly continued the Guo Fa (2011) No. 4 document and Cai Shui (2012) No. 27 document during the 12th Five Year Plan period, which became a positive factor for the semiconductor industry.

However, during the period of the 13th five year plan, the central government has restricted the qualification and support of IC enterprises to varying degrees compared with the period of the 12th Five Year Plan. Instead, it gives financial support to domestic semiconductor enterprises or assists in the acquisition and merger of international large factories by directly taking shares in the semiconductor industry investment fund (hereinafter referred to as the big fund).

The first round fund of the national integrated circuit industry investment fund Co., Ltd. (ciciif), known as the "big fund" of the national team, is about US $20 billion, but according to market estimates, the total investment of the local government and state-owned enterprises will exceed US $100 billion in the first round. As of September 2016, about 60% of the $10 billion fund approved by ciciif had been invested in chip manufacturing, 27% in chip design, 8% in packaging and testing, 3% in equipment, and 2% in materials.

Policy objectives and policy support of "made in China 2025" mainland semiconductor industry

As can be seen from the above figure, the most important policy goal during the 13th Five Year Plan period is that the self-sufficiency rate of domestic core basic components and key basic materials will reach 40% in 2020 and further increase to 70% in 2025. However, considering that the self-sufficiency rate of domestic IC market in 2015 was less than 20%, during the 13th Five Year Plan period, in addition to the significant expansion of wafer foundry and packaging test capacity, domestic IC design enterprises need to invest more R & D in key core products.

There is no shortage of money, but overseas M & A is blocked

IC insights, a market research institution, pointed out that to achieve the goal of 70% IC self-control rate in the 13th five year plan of the Chinese government, we need to rely on two basic elements: capital and technology. At present, with the support of large government led funds, funds will not be a problem.

However, since 2014, China has tried to acquire technology practices by acquiring foreign semiconductor companies, such as ISSI and omnivision. Now, the acquisition strategy is not effective. Most foreign governments are very vigilant about China's ambition in the integrated circuit industry. It is very difficult for Chinese capital to acquire foreign IC companies. IC insights even believes that the peak of opportunities for China to acquire technology through the acquisition of foreign IC companies has slowed down.

Just a few examples:

In February 2016, Fairchild rejected the US $2.6 billion bid from two Chinese buyers, China Resources Microelectronics and China Chuang, and chose on semiconductor, which offered a lower price than Chinese companies. The reason given was also that it was worried that the US regulatory authorities would block the transaction. Also in February, Ziguang shares under Ziguang group announced that it decided to terminate the acquisition of Western data, an old American storage company, for us $3.775 billion, also because the US overseas investment committee wanted to intervene in the review.

In December 2016, China Fujian macro core Fund issued a statement on its website, withdrawing its offer for the acquisition of German semiconductor enterprise aisqiang and returning the previously purchased shares of aisqiang. Although the German Federal Ministry of economy withdrew the approval order for China Fujian macro core fund to purchase aisqiang company for 670 million euros, some media said, The Obama administration played a crucial role in blocking the deal.

It can be seen from the above that western countries are becoming more cautious and even "restrictive" towards China's investment in the semiconductor industry. Peng Hongbing, deputy director of the Electronics Department of the Ministry of industry and information technology, once said in an interview with the Wall Street Journal:

"The United States has too much unnecessary anxiety and does not want these conflicts between the United States and China. At the same time, it emphasizes that China must reduce its dependence on semiconductor chip imports."

According to customs statistics, in 2016, the import of integrated circuits was 342.55 billion pieces, a year-on-year increase of 9.1%; The import amount was US $227.07 billion, a year-on-year decrease of 1.2%. In the same period, China's crude oil import was only 607.8 billion. China has spent nearly twice as much on semiconductor chip imports as on crude oil.

In addition, according to the data of Bain & Co, China's annual consumption of semiconductors is worth more than US $100 billion, accounting for nearly 1 / 3 of the total global shipments, but China's semiconductor output value accounts for only 6% 7% of the world. Many imported chips are assembled in personal computers, smart phones and other equipment and then exported overseas. However, there is still a huge gap between the number of semiconductors produced by Chinese chip manufacturers and the number of semiconductors consumed by China itself.

Investment fever is becoming the "catalyst" of the semiconductor industry

In order to change the situation of relying on others to provide semiconductor chips for a long time, China has taken various measures to support it since 2014. In the past, it used to promote the development of semiconductors by "widely casting a net". During the same period, it invested in 130 semiconductor factories in more than 15 provinces, but the effect was poor. With the further overweight of policies and the significant upgrading of China's semiconductor industry infrastructure, global semiconductor companies have taken the initiative to determine their China strategy to cooperate with Chinese manufacturers so as not to be excluded. Gartner, a market research firm, predicts that in the next five years, some international fabless semiconductor enterprises may shift as much as 50% of their wafer procurement demand to Chinese foundries.

At present, there are two major semiconductor investment driving engines in China, the "big fund" and the Ziguang group, and they are not soft on the investment in wafer factories. Recently, there have been successive projects such as SMIC international, Changjiang storage and Nanjing Ziguang. Not only do Chinese people spend money on building factories, but also foreign enterprises actively spend a lot of money on chip manufacturing investment in China, presenting a rare situation over the years "Domestic and foreign" investment is hot.

From Samsung's additional investment in Xi'an memory chips, the start of SK Hynix Wuxi No. 2 plant in South Korea, to the formal finalization of the "grid core" of the 12 inch wafer factory of grofangde GF and its transfer to Chengdu, it spans two major fields: memory chips and wafer foundry, and its geographical scope is all over central, East and West China.

Specifically, in the field of wafer foundry, grofangde's competitors: TSMC's 8-inch plant in Songjiang, Shanghai and 12 inch plant in Nanjing; liandian's 12 inch plant in Xiamen; not to mention SMIC's new 12 inch plants in Beijing B2, B3 and Shanghai. The Chinese market seems to be an indispensable investment place for the layout of wafer foundry. Naturally, grofangde is no exception. It should actively seek investment in China Find a place to set up factories. It will not hand over China's great market to competitors.

Obviously, the domestic semiconductor industry is ushering in a wave of investment boom under the joint promotion of policies, Chinese manufacturers and foreign players.

However, two uncertainties also need to be noted.

On the policy, the tough measures taken from the Obama administration, blocking domestic capital into related fields, and the new US President trump (Donald Trump)'s constant attacks on the mainland's trade practices, advocating the return of manufacturing industry and other policies, had to be vigilant.

In terms of technology patents, with the increase of Chinese local manufacturers and overseas expansion, Samsung, Hynix, micron, Intel and Toshiba may make an issue of patents. Because the above-mentioned memory manufacturers have been manufacturing DRAM and NAND flash memory for several 10 years, there are many applications for memory technology patents and a wide range of product lines, few new manufacturers can not infringe the existing ones In the case of patent, develop new DRAM and NAND technologies.

Xiaomi has taken a new step on behalf of domestic manufacturers, but China still has to cross these thresholds and challenges to move towards high-end manufacturing.