\u3000\u3 China Vanke Co.Ltd(000002) 920 Huizhou Desay Sv Automotive Co.Ltd(002920) )
Intelligent driving has become the next outlet of industry reform. Vehicle manufacturers and suppliers have achieved differentiated layout. Many vehicle enterprises have entered the Bureau of intelligent driving, began to increase investment in intelligence, and the upstream and downstream cooperate to promote the rapid improvement of automobile intelligence. We believe that the different R & D investment of automotive OEMs in intellectualization will eventually lead to the difference in the path of realizing intellectualization. At present, there are three ways for vehicle manufacturers to get involved in intellectualization: ① Tesla invests more than 10 billion yuan every year to strive to realize the full stack of software and hardware; ② The new car building force represented by Xiao Peng has invested about 2 billion yuan in R & D every year, hoping to realize self-research at the level of software and algorithm. ③ Cooperate with external suppliers to improve the intelligent level of the whole vehicle. We assume that after the long-term development of the industry and the return of R & D expense rate to the steady state of the industry, the choice of path to realize product intelligence is ultimately determined by the amount of revenue. Therefore, for the above three paths, according to our calculation, the corresponding revenue volume of auto OEM to realize the whole stack of software and hardware (from chips, domain controllers, software, algorithms, etc.) needs to exceed 100 billion, the revenue volume to realize the whole stack of software and hardware self-research and external collaboration needs to be more than 40 billion, and the rest of the OEMs will cooperate with third-party suppliers to realize auto intelligence.
The concentration of the intelligent mobile phone market does not increase but decreases, and the downstream customers of intelligent Tier1 are more diversified. We believe that the proportion of auto enterprises adopting full stack / software self-research and third-party cooperation mainly depends on the concentration of the intelligent car market. By reviewing the changes in the concentration of the mobile phone market around 2010, we observed that smart phones did not lead to an increase in the concentration of the mobile phone market. On the contrary, at the initial stage of smart phones, the industry showed a trend of decentralization. In the final steady state, the concentration was also lower than before. We believe that the complexity of products and the number of downstream market segments are the fundamental factors that determine the concentration, while the factors such as high complexity of automobile, long industrial chain and long iteration cycle have not changed essentially due to intelligence. In addition, the car is the optional daily consumption with the highest unit price, and the diversified demand makes the market have many subdivided long tail markets. At present, automotive intelligence is still in its infancy. According to our statistics, Cr5 has shown a downward trend in recent years, with the addition of traditional car enterprises, Internet enterprises, ICT giants and new forces in car manufacturing. We judge that the concentration of the automotive market will not be greatly improved due to intelligence in the short term. For intelligent Tier1, the downstream customer base continues to expand and become more diversified.
Three roles of Tier1 in the future: automotive intelligent enabler, hardware solution supplier and OEM. We expect that based on the differences in the path of realizing intellectualization, the path of future automobile enterprises can be divided into the following three categories: embracing the third party, the full stack of software and hardware; The realization of these three development paths requires the help of partners with different roles, which also corresponds to three possible roles of Tier1 in the future intelligent era: automotive intelligent enabler, hardware solution supplier and OEM.
1) the car factory with the whole stack of software and hardware is expected to cooperate with the hardware OEM. Car companies that choose the full stack of software and hardware aim to achieve the optimal solution of experience and cost through the high coupling of algorithms and chips, but they also have the highest requirements for R & D and revenue volume of car companies. At the same time, in consideration of cost and efficiency, such car factories will probably seek OEM for hardware OEM. In this era, the core competence of the manufacturer lies in large-scale manufacturing and cost control.
2) the depot with full software stack is expected to cooperate with the hardware solution supplier. Car companies that choose the full stack of software will take intelligence as their selling point and create a competitive advantage through the continuous iteration ability of automatic driving algorithm. At the same time, it needs a partner to provide hardware and supporting underlying software solutions. For example, Xiaopeng P7 is equipped with Huizhou Desay Sv Automotive Co.Ltd(002920) intelligent driving domain controller ipu03 based on NVIDIA Xavier chip. We believe that different from the OEM, the hardware solution supplier should have the following four abilities in addition to the basic manufacturing ability: 1) in-depth understanding of the chip and board level design ability; 2) Understanding of vehicle level safety standards and functions; 3) Development ability of underlying software; 4) Localized service and rapid response capability.
3) automobile enterprises embracing third parties need to be empowered by intelligent overall solution suppliers. Car companies that choose to embrace third parties will only maintain intelligence at an average level, but may have a high market share in a certain segment by virtue of brand, cost performance and other advantages. We judge that in the future, a large number of car manufacturers will have a demand for the overall solution of software and hardware. Intelligent driving Tier1 is expected to enter and exit the "turnkey" business model and become an intelligent enabler of cars. We believe that the core competitiveness of automotive intelligent enablers lies in: 1) channel construction at the depot side; 2) Understanding of intelligent software and hardware architecture; 3) Diversity of solutions.
The proportion of the three roles determines the long-term space of Huizhou Desay Sv Automotive Co.Ltd(002920)
At this stage, Huizhou Desay Sv Automotive Co.Ltd(002920) 's role is still dominated by hardware solution supplier. Relying on the first mover advantage of in-depth cooperation with NVIDIA, in-depth understanding of vehicle regulations and functional safety levels, as well as the ability of rapid product landing, the company has obtained multiple fixed points in the past two years. In the long run, in addition to the continuation of the existing role, we believe that the company is expected to evolve into an enabler of automotive intelligence. The reasons are: 1) the company has a stable and comprehensive customer system; 2) In the past few years, the company has accumulated the software and hardware architecture through interaction with automobile enterprises; 3) The R & D centers established in various places have carried out forward-looking technology and algorithm research; 4) By investing in intelligent driving tier2, lidar manufacturers and other upstream enterprises to build an intelligent driving ecosystem, we have the core capabilities required for business model upgrading. Therefore, we suggest to pay attention to the progress of Huizhou Desay Sv Automotive Co.Ltd(002920) transformation of vehicle intelligent enablers, and remind us to pay attention to the improvement effect of the overall software and hardware scheme on the revenue side and the optimization of gross profit margin brought by the upgrading of business model.
Investment suggestion: the company is the core target of automotive intelligence, with obvious card position advantages and clear follow-up growth logic. We are optimistic about its expansion from intelligent cockpit to intelligent driving, and its transformation from automotive electronics supplier to automotive intelligent enabler. We expect that the company's revenue from 2022 to 2024 will be RMB 12.60/17.59/22.82 billion respectively, and the net profit attributable to the parent company will be RMB 1.138/16.66/23.46 billion respectively, maintaining the Buy-A investment rating. As the company is still in the growth stage, we give PE 70 times the net profit in 2022, corresponding to the six-month target price of 143.5 yuan / share.
Risk tips: 1) the risk that the development of intelligent vehicle market is less than expected; 2) The risk that the supply chain repair is not as expected due to the intensification of the epidemic; 3) The risk of intensified shortage of upstream chips; 4) The risk that the company's intelligent driving business is not progressing as expected; 5) The risk of intensified market competition; 6) The risk of inaccurate judgment on the evolution direction of automobile intellectualization; 7) The assumption is less than the expected risk.
