On August 17, Xu Daquan, executive vice president of Bosch China, wrote on moments: due to the increasingly serious epidemic in Malaysia, Bosch related chip supplier factories will be closed until August 21. The chip supply of many key components of Bosch has been directly affected and is expected to be basically cut off in August. The news was widely forwarded by the industry for a while, and Lao Li's team also interviewed relevant experts for the first time.

Since the second half of last year, the problem of lack of core has been hyped by the media from the extreme weather in Texas to the factory fire in Japan to the epidemic in Malaysia. Lao Li's team organizes interviews with automotive chip experts almost every month to determine the supply of chips. Although the semiconductor industry is relatively closed, researchers have set up a supply chain for automotive chips. Based on the current supply and demand situation, many researchers judge:Starting from Q4, the core shortage problem of automobile enterprises will be solved one after another, superimposing the effect of market peak season and prosperity, and the whole vehicle stock will usher in a new high.。

The end time of the lack of core given by the secondary market may exceed the expectations of many friends. Today, Lao Li will talk with you about the impact that the closure of the Malaysian chip factory will bring to China. What problems have been exposed in the year of lack of core in China? How will China's automotive chips develop?

Domino effect of industrial chain

In addition to the giant Taiwan Semiconductor Manufacturing Co Ltd, there is also an important supporting region for Asia's automotive semiconductor industry-Malaysia. Its semiconductor industry began in the 1970s and has been accumulated for nearly half a century. Malaysia has become an important part of global semiconductor closed testing. Malaysia accounts for 13% of the global back-end semiconductor production.

In the closed testing process of automotive semiconductors, Malaysia is even more isolated, accounting for more than 40% of the global market share of automotive chip closed testing. ST, Infineon, on and NXP all have factories in Malaysia, and the final flow of their factory chips is China:According to the statistics of the researchers, more than 60% of the passenger cars of Chinese European joint venture brands need Bosch ESP, while the core chip L9369 of Bosch ESP system basically comes from Malaysia's ST, and ST is the one most affected by the Malaysian epidemic.The company.

The situation is not as bad as expected.Although Malaysia entered a full blockade in June, most semiconductor companies have restored nearly 100% production capacityEarlier, many researchers predicted that the problem of missing cores in domestic cars would basically be solved from August this year, but accidents always happened inadvertently. COVID-19 infected the Malaysian factory of ST and stopped production, and it was reasonable for Bosch executives to send friends.

Interestingly, everyone is spreading the problem, but seldom studying the solution to the problem. Lao Li's team interviewed local experts in Malaysia and came to the conclusion that:The Malaysian chip factory has 80% of the labor force to be vaccinated, and the government allows the chip factory to be put into operation at 100%. According to the vaccination schedule and the time of resumption of production, the problem of core shortage is expected to be alleviated from September this year, and the supply chain will be fully restored to pre-epidemic levels from November.

The automobile semiconductor industry chain is long, and any problem in its link will lead to the lack of core downstream and the performance mode and time are different.

In February this year, the core shortage was due to Taiwan Semiconductor Manufacturing Co Ltd's lack of wafers. In the case of capacity shortage, Taiwan Semiconductor Manufacturing Co Ltd allocated wafer capacity mainly to consumer electronics customers, resulting in fewer wafers in the automotive sector and delayed release of capacity. Taiwan Semiconductor Manufacturing Co Ltd is also trying to solve the wafer allocation problem, starting from April this year to do capacity switching, because the wafer capacity switching is gradual, until this year Q4 can return the wafer capacity to the normal level.

Compared with the shortage of wafer capacity, the core shortage in Texas is due to natural disasters. Texas Semiconductor suddenly shut down during the production process. Generally speaking, if the semiconductor factory shuts down normally, the production capacity can be restored in a few weeks. If the semiconductor factory shuts down suddenly in production, the recovery cycle is much longer. The technical assessment process alone will take 2-3 weeks, and if there are problems such as lithography machines, it will take at least 2-3 months to repair.

When everyone focuses on the lack of core, the industry is quietly changing. From the demand side, the domestic demand for new energy vehicles is rising rapidly, with monthly market penetration of more than 10% in the last three months, while the market demand for fuel vehicles has declined and sales have been declining.

The supply side is also polarized. due to the fast response and continuous optimization of the chip supply chain, the problem of core shortage has been basically controllable, while other enterprises have superimposed the Malaysian epidemic because of their own supply chain problems. the safety factor of the short-term chip supply chain is declining sharply.

From the perspective of the secondary market, after entering the fourth quarter, the demand side of the domestic passenger car industry is generally better than the supply side, and the terminal often queues or even increases the price of second-hand cars. If the automobile sector pulls back in the near future, it is a better opportunity to cover positions. After the end of the chip impact in the fourth quarter, car companies with better demand will enter the inventory replenishment stage, with greater upward profit flexibility, in which independent brands have more advantages than joint venture brands.

What are the problems exposed by the lack of core?

The reasons for the lack of core are various, but the essence isSupply chain securityproblem.

Before the COVID-19 epidemic, the globalization of the automobile industry advanced rapidly, and various countries and enterprises had a relatively clear division of labor, forming a perfect supply chain. When there were changes in the upper reaches of the industrial chain, another country or enterprise thousands of miles away will soon be affected. Let's take Malaysia's lack of core as an example to sort out the transmission path:

In this supply chain, Chinese automobile manufacturing is completely "bound" by ST chips. To solve the problem of supply chain security, we need to start from several directions:

The first is the self-control of the core technology.Whether it is the neck problem of Huawei chip or the prohibition of high-end semiconductor equipment, the essence is that the core technology is uncontrollable, and the problem of lack of core in the car is also uncontrollable to a certain extent. Although the MCU produced by ST is not a high-end chip, except BYD, no enterprise can achieve the application of automobile specification level.

In the future, Chinese enterprises will always emphasize the autonomous controllability of chips, and some chip companies that have not been paid attention to in the past may usher in a larger opportunity in China. At present, domestic mainframe manufacturers have seized this direction, but the replacement process is relatively long. There is little chance of chip switching this year, and there may be an opportunity next year.

The supply chain management capabilities of Japanese enterprises are worth learning from, and this round of chip impact has little impact on Japanese enterprises, on the one hand, because the domestic sales and trend of Japanese brands are relatively stable, and they control sales and supply well; on the other hand, it is because Japanese OEM will position first-tier suppliers as Japanese domestic enterprises, and Japanese Tier 1 will choose domestic suppliers such as Renesas on the main chip to form coordination.

Second, there is a good supply chain management mechanism., includingSupply time, supply mechanism and supply price. For relatively complex devices such as the MCU, it takes 12-16 weeks for chips to be ordered to ship, while inertial sensors for vehicle stabilization systems take 26 weeks. Therefore, managing its dependence by carefully managing order arrangements and maintaining inventory balance is essential for just-in-time supply, which is vulnerable to unusual market factors such as the COVID-19 pandemic, as was the case in Malaysia.

From the perspective of supply time, car companies-Tier 1-chip enterprises have a fixed management mode, which is generally controlled by the LeadTime (lead time) index of the product. During the lack of core, the hoarding behavior of channel merchants and Tier 1 caused the LeadTime of automotive chips to change from the normal level of the past 16 weeks to 10 weeks, and the safety inventory from 7-10 days to a month. The problem of lack of core is that there is something wrong with the market mechanism.

There are also problems in the supply mechanism. After the recovery of production capacity in Malaysia, priority may be given to certain countries and enterprises, multinational enterprises will give priority to procurement, and Chinese enterprises may not give priority to procurement. It is impossible to judge which country has suffered or benefited.

In the process of global mainframe factory guaranteed chip supply, multinational companies such as Volkswagen, General Motors and Mercedes-Benz will give priority to European and American demand with better profitability. Chinese joint venture companies have less say in purchasing than their headquarters, so the problem of lack of core of joint venture brands will be more serious. Independent brands have more autonomy in chip supply issues because they have a higher say in chip control than joint ventures.

Generally speaking, there are two kinds of price systems for automotive semiconductors. The first is the big customer system. Big car companies negotiate prices according to annual procurement. Generally speaking, they can determine the demand and price of the following year with chip enterprises from September to November every year. The second year will be carried out in accordance with the agreement, so big Tier 1 and big Tier 2 will not increase prices midway. The second is the non-large customer system, for small customers, semiconductor enterprises and channel enterprises will generally increase prices, online price increases are basically derived from this system.

The automotive semiconductor industry is like a big river, half of the water is clear, the bottom is clear, and half of the water is muddy and unpredictable. From this perspective, it is reasonable to strictly check the price rise of semiconductors.

Is there a future for Chinese automotive chips?

The problem of lack of core makes Chinese automobile companies wonder, where is the future of China's automotive semiconductors? Everyone thought about it for a long time and thought about what to do, but they didn't figure out how to do it.

China sells 25 million cars a year, accounting for more than 30 per cent of the world, while in the global chip market, domestic production of automotive chips is less than 15 billion, accounting for only 4.5 per cent of global production capacity, and imports of key components such as MCU are more than 80 per cent to 90 per cent, so China imports more than 100 billion yuan of automotive chips every year.

Landing to products, automotive semiconductors is also a broad concept. According to their functions, automotive semiconductors are divided into automotive chips, power devices, sensors and so on. People often refer to automotive computing chips, which can be divided into MCU chips and SoC chips according to the integration scale. The main power devices are IGBT and SiC.

Domestic automotive semiconductor products are in supply security problems, the domestic lack of core is mainly MCU chips, the MCU suppliers on the market are NXP, TI, Renesas and other foreign enterprises, the domestic car specification-level MCU products are not many enterprises, BYD is one. Prior to this, the domestic IGBT also once had a rush for goods, and the global IGBT products were mainly supplied by Infineon in Germany, while China has only cultivated two enterprises, Starr Semiconductor and CRRC Corporation, and the road of product promotion is still very long.

Automotive chips will face the common problem of chips-contract manufacturing. Some experts told Lao Li that at present, about 70% of the world's automotive MCU production depends on Taiwan Semiconductor Manufacturing Co Ltd, which creates a production bottleneck for the whole industry. To make matters worse, the automotive chip business accounts for only 3% of Taiwan Semiconductor Manufacturing Co Ltd's total income, with a high market share but low value, so Taiwan Semiconductor Manufacturing Co Ltd does not attach importance to this business.

Although Taiwan Semiconductor Manufacturing Co Ltd said it would increase investment to support its car customers, it was a long-term move to cater to the industry's long-term trend towards high-performance computing, rather than to help solve current supply bottlenecks.

In the lack of core for more than a year, automobile enterprises, as the most downstream application enterprises, do not think about solving the problem of neck jam, but in various ways to "sweep goods" and increase inventory. Overall, BYD is in relatively good shape, and other automakers are under different pressure, but it is expected that all Q4 companies will get out of the core shortage:

Among the independent brands, BYD's overall inventory reserves are relatively strong, and the feedback from many experts is that the impact is not great; Geely, due to Malaysia's investment in Proton cars, has more resources available to the local government and production capacity, and is also actively responding; relatively speaking, Changan and Guangzhou Automobile have a certain degree of pressure.

Car companies are also working out strategies that are in the best interests of the company according to the supply of chips. Enterprises dominated by traditional fuel vehicles, such as the Great Wall, mainly produce the most profitable tanks and Harvard, giving priority to ensuring fuel vehicles in the case of lack of cores, and reducing the less profitable Euler production. NIO Inc., ideal, XPeng Inc., BYD and other new energy vehicle companies mainly guarantee the output of electric vehicles. These companies account for a relatively high or even 100% of new energy vehicles, and their market share and sales volume are relatively more important at the present stage. Shareholders will not examine profitability. In the first half of this year, ideal, XPeng Inc. and NIO Inc. basically do not look at the price, while traditional car companies, or those with certain profit pressure, If the price rises to a certain extent, the price of the chip will not be accepted.

For more than a year, China's automotive chip industry has experienced too much. Chips such as automotive MCU, which have a certain threshold value but are not high, are chicken ribs for many enterprises, but without their own "chicken ribs", China's automotive chips will always have the risk of being cut off.