China’s self-driving car industry has taken off over the past decade. Previously a small and fringe sector, it is now receiving backing from both the government and industry heavyweights, cultivating a new generation of tech start-ups hoping to be the first to get a viable product on the road. Despite an uncertain timeline for real-world application, investors and developers alike remain optimistic that laying the technological foundation now will reap huge rewards in the future. In this article, we take a look at the industry’s development outlook, government regulations, and which areas of the industry are open to foreign investment.
The concept of self-driving cars has captured people’s imaginations for decades. Proponents of the technology paint a vision of how it will transform the way humans live and travel, helping to ease congestion, reduce emissions, and even save lives.
2021 was the oft-cited year for the arrival of self-driving cars, but this reality has failed to materialize. Even as the technology improves, the long-awaited launch date gets pushed further and further into the future as technological bottlenecks, irregular transport infrastructure, and public unease hamper the technology’s uptake.
Skeptics have argued that even if it becomes possible for fully autonomous vehicles (AVs) to safely drive on public roads, it will be many decades down the road. This uneven and unpredictable development of the industry and lack of a timeline for profit-making has made many investors think twice and resulted in an increasingly ambivalent media and public.
But whereas the hype surrounding AVs already peaked a few years ago in the U.S. and other western countries, in China, it is still on the rise. China’s self-driving car industry is brimming with optimism; a new generation of upstarts are receiving significant investor attention and established technology companies and traditional automakers alike are jumping on the bandwagon.
This optimism also has state backing – the Chinese government is actively promoting the industry through policy incentives and development initiatives, wanting to ensure that if and when autonomous driving takes off, China and Chinese companies will be on the front foot. Meanwhile, new regulations serve to legitimize the sector and standardize its products, providing a more concrete legal framework for both operators and investors to work within.
In this article, we provide an overview of China’s self-driving car market and the relevant regulations that are currently in place and explore which areas of this emerging sector foreign investors can participate in.
The potential of China’s self-driving car industry
Although autonomous driving presents a relatively high risk for investors, given the industry’s current lack of profitability, the market potential may be huge.
Management consultancy firm McKinsey has predicted that highly autonomous vehicles will see mass deployment in China in the next decade. The firm also expects that the uptake of AVs will cause a shift toward a service-oriented passenger vehicle market, focused on mobility services, and away from a product-oriented market, focused on private car sales.
The extremely high costs of developing and deploying the technology, as it currently stands, will also help to drive the AV market in the direction of mobility services. Most consumers will not be able to purchase a fully autonomous car but could still afford to enjoy its benefits through taxi or bus services, which would be able to run 24/7 as the human driver is taken out of the equation.
This reflects the current development of the industry, with many companies largely focusing on developing ‘robotaxis’ and autonomous passenger busses. Some companies have already begun rolling out trial robotaxi and bus services in urban areas in China.
This presents possibly the biggest opportunity for autonomous driving in China. Information provider and analytics firm IHS Markit predicts that China’s shared mobility market will reach a valuation of RMB 2.25 trillion (US$352 billion) by 2030, or a compound annual growth rate (CAGR) of 20 to 28 percent. If AVs can penetrate this market, the rewards may be enormous.
For lower-level AVs, adoption may be even higher. IHS Markit predicts that 80 percent of new vehicles will have some automated capabilities by the end of the decade. It will therefore be almost a prerequisite for a vehicle to have some autonomous capabilities in order to be competitive.
Government targets for China’s self-driving car industry
Autonomous driving and smart transport were named as key areas for digitalization in the 14th Five-Year Plan (FYP) – China’s highest-level initiatives for social and economic development – which calls for companies to:
“Develop autonomous driving and vehicle-road collaboration travel services. Promote smart road management, traffic signal linkage, and public bus priority access control. Build […] intelligent car parks.”
This is the first time autonomous driving technology was mentioned in an FYP, and, though brief, it confirms the importance that the central government places upon the development of the industry.
China has set several ambitious targets for the development of the autonomous driving and smart vehicle industries in recent years, ranging from targets for creating a regulatory framework to the roll-out of intelligent vehicles on roads.
These targets have been cemented through several policy documents. One such document is the Innovative development strategy for smart vehicles. Issued jointly by 11 government departments, including the National Development and Reform Commission (NDRC) and the Ministry of Industry and Information Technology (MIIT) in February 2020, the document sets targets and strategies for the development of autonomous driving. According to the document, by 2025, China aims to:
- Have ‘basically shaped’ China’s smart vehicle technology innovation, industry ecosystem, infrastructure, regulations and standards, and product supervision and data security systems.
- Realize large-scale production of smart vehicles capable of ‘conditionally automated driving’ and realize the commercial application of smart vehicles capable of ‘highly automated driving’ in specific environments.
- Achieve positive progress in the construction of equipment related to smart transport systems and smart cities, realize regional coverage of wireless communication networks for vehicles (such as LTE-V2X).
- Realize the gradual launch of new generation wireless communication networks for vehicles (such as 5G-V2X) in some cities and highways.
- Realize full coverage of high-precision space-time reference networks.
Another document, titled the Roadmap for Intelligent Connected Vehicle Technology 2.0, was released later in 2020. The document lays out concrete targets for where the government intends for China’s self-driving car industry to be in the coming years, outlining the following goals:
- Intelligent connected vehicles with partially automated driving and conditional automated driving capabilities should account for more than 50 percent of all vehicle sales.
- The assembly rate of new cars with Cellular Vehicle-to-Everything (C-V2X) terminals should reach 50 percent.
- Realize commercial application of vehicles with highly automated driving capabilities, initially in specific scenarios and restricted areas, before expanding to a wider scope.
- Realize the widespread operation of various connected vehicles with highly automated driving capabilities across vast areas of China.
Achieving these goals will require significant investment in technology R&D, testing, and production, as well as a robust regulatory environment in which innovative companies have space to maneuver.
The Innovative Development Strategy for Smart Vehicles calls for deepening international cooperation and encourages domestic and foreign companies to jointly work on basic research, technology development, and commercial application. It also says to encourage foreign-invested enterprises (FIEs) to actively participate in the industrial development of intelligent vehicles.
The document calls for:
- A higher level of international cooperation, urging to fully utilize mechanisms such as multilateral and bilateral cooperation and high-level discussions to establish a platform for international cooperation.
- Participation in the establishment and coordination of international and regional standards and strengthen the mutual recognition and trust of authentication and approval results.
- International communication and cooperation on rules and regulations for intelligent vehicles.
Where to invest in China’s self-driving car industry
The latest edition of the Catalogue of Encouraged Industries for Foreign Investment (2020 Edition), which is updated by NDRC and the Ministry of Commerce (Mofcom) on a yearly basis, lists several industries and subfields related to intelligent transport and autonomous driving.
There is a wide scope of fields currently open to foreign investment, including areas such as hardware development and manufacturing, autonomous driving technology R&D, security technology R&D, and development of testing and evaluation mechanisms, among others.
|Sectors Related to Autonomous Driving Encouraged for Foreign Investment|
|R&D and manufacturing of automotive electronic devices||Autonomous driving systems.|
|AVs component manufacturing||Transmission/safety/body/driving/information control systems, advanced driving assistance systems (ADAS), automatic driving control systems, and other key components of smart cars.|
|R&D and manufacturing of key hardware components||Sensors, auto chips, central processors, vehicle operating systems and information control systems, vehicle network communication system equipment, visual recognition systems, high-precision positioning devices, chassis-by-wire systems.|
|Technology R&D||New types of intelligent terminal modules, multicore heterogeneous intelligent computing platform technology, high-precision positioning and map technology for all-weather and complex scenarios, sensor fusion technology, core wireless communication technology for vehicles, basic cloud control platform technology.|
|Security R&D||New security isolation architecture, software and hardware collaborative attack recognition technology, terminal chips security encryption and application software security defense technology, wireless communication encryption technology, secure communication and verification authorization technology, data encryption technology.|
|Development of testing and evaluation mechanisms||Testing and evaluation system framework R&D, technology and authentication tools for simulations and real-life vehicle road tests, vehicle and system testing and evaluation methods, construction of basic testing databases.|
|Manufacturing of hardware components for L3/L4/L5 AVs||Lidar, millimeter-wave radar.|
|Source: Catalogue of Encouraged Industries for Foreign Investment, National Development and Reform Commission and the Ministry of Commerce|
Fully automated driving
As has been discussed above, fully autonomous vehicles are unlikely to hit the road for commercial use until at least the end of the decade. In its report, McKinsey cited technology bottlenecks and unstandardized transport infrastructure as reasons for this extended timeline. China also currently falls behind other countries in several fields, such as computing platforms, sensors, AV system integration, and more.
China is therefore keen to solve these technical issues and catch up with other markets, which will require a large amount of investment in software development and technological R&D, as well as rigorous testing and assessment. Many of these fields are also encouraged for foreign participation, as listed in the above catalogue.
The high cost and as-yet uncertain return on investment of this sector mean that the space is currently dominated by the internet giants (Baidu, Tencent, and Alibaba, to name some of the biggest players), established traditional automakers, or upstarts that are backed by these companies.
Baidu, for example, is one industry frontrunner, having launched free trials of its Baidu Apollo robotaxi service in Beijing, Shanghai, and Guangzhou, and earlier in 2021 becoming the first company to get approval to charge fares for a fully driverless robotaxi service in a closed park in Beijing. Didi Chuxing, China’s dominant ride-hailing platform, also established an autonomous driving arm in 2016 and has launched road tests in several Chinese cities
Alibaba, meanwhile, has thrown its weight behind AutoX, which has received licenses to conduct road tests in several cities across China, including Shenzhen, Guangzhou, and Shanghai, among others. Baidu Apollo, Didi Chuxing, and AutoX have all received licenses to conduct road tests in California.
Several foreign automakers are also investing in Chinese AV upstarts. Toyota currently backs Pony.ai, which was the first company in China to launch road tests, GM recently invested US$300 million in Momenta, and Nissan has bet on WeRide.
Low-level automated driving
With the large-scale commercialization of fully autonomous vehicles still uncertain, many players are dipping their feet in peripheral applications and industries.
Although not technically ‘autonomous’, low-level automation in passenger vehicles also forms an important part of China’s autonomous driving vision and is becoming increasingly common. Many new car models are offered with ADAS or autonomous parking functions. According to Auto Gasgoo, one of China’s leading outlets for news on the auto market, over 25 percent of new passenger vehicles sold in 2020 had low-level (L2 and below) automated driving capabilities.
This scene is dominated by automakers and electric vehicle companies. The most notable foreign company in this space is Tesla, whose China-made electric vehicles offer automated features such as driver-assist and automated parking.
In many cases, autonomous systems are developed for carmakers by third-party companies specializing in the technology. Baidu Apollo, for example, in addition to its robotaxi ambition, also provides its connected car offerings to Chinese carmaker WM Motor.
Autonomous trucks and delivery vehicles
Beyond autonomous cars, interest in automated transport and logistics is rising. This trend may also have been accelerated by the pandemic, which has increased demand for no-contact delivery.
The AV industry may also have a strategic role to play in China’s development, helping to answer China’s growing labor problem. China’s shrinking and increasingly well-educated workforce has already begun to impact industries that rely on low-skilled labor, including trucking and transport. Automating these sections of the logistics industry through the use of self-driving vehicles could serve to eventually replace the need for human drivers.
This movement has already begun. AVs have been deployed to deliver parcels in university campuses and transport containers in port areas. E-commerce companies, such as JD.com and Alibaba, have developed and begun deploying unmanned delivery vehicles to deliver parcels.
The driverless truck industry has also transitioned from the research stage to testing and deployment. Generally considered more commercially viable as trucks are mostly required to drive on highways, which are more uniform and better regulated than smaller roads, many companies are betting big on this field.
Autonomous truck start-up Plus recently conducted successful road tests on the Wufengshan Highway in the Yangtze River Delta, and plans to launch pilot runs of its service in 2022. Meanwhile, Chinese trucking start-up TuSimple recently partnered with U.S. chipmaker NVIDIA to develop autonomous trucks, with the companies saying in a press release that production would begin in 2024. JD Logistics, JD.com’s logistics arm, and meal delivery platform Meituan have also recently led a round of funding into upstart Inceptio, which reached US$270 million.
Microchips and sensors
China’s self-driving car industry is still reliant on imported technology for certain components as domestic companies play catch-up with global competitors. One of the most important of these is auto chips, a requisite component for even low-level autonomy, and one that is only becoming even more in demand as the technology matures.
Foreign companies that produce auto chips for the Chinese market include Germany’s Infineon Technologies, which has a manufacturing plant in Wuxi, Jiangsu Province, and the Chinese joint venture of the British microchip design company Arm, which recently announced it would start creating computing units for autonomous driving systems.
Sensors – the technology that enables AVs to ‘see’ their surroundings – is another field in which foreign companies may have an advantage. The technology in China is still years behind the global frontrunners – according to the McKinsey report – and some domestic developers turn to overseas suppliers. Earlier in 2021, China’s SAIC Motor announced it would use lidar sensors from U.S. start-up Luminar Technologies to power automated driving capabilities in its future car models.
However, that is not to say there are no Chinese companies in this field. With lidar set to become a very lucrative industry if AVs take off, tech giants such as Huawei and DJI, the Chinese drone maker, are getting in on the action. China is also home to an increasingly crowded start-up scene, with companies such as XPeng, RoboSense, and Hesai developing offerings.
Standards and regulations in China’s self-driving car industry
China’s self-driving car industry remains relatively unregulated when compared to traditional auto industries. There are currently no nationwide regulations on AVs and only recently did the government release standards on the classification of different levels of autonomous driving.
This is partly due to the fact it is still a relatively nascent industry, but the decisions may also be strategic. Authorities have largely chosen to adopt an ‘innovate now, regulate later’ policy to give companies freer rein to develop and test the technology. This is an important strategic choice that may give Chinese companies an advantage over their competitors in more heavily regulated markets, such as the U.S. and Europe.
However, several peripheral and regional regulations and guidelines have been released in recent years, laying the groundwork for the creation of wider laws that could help to focus both the industry and investors.
In September 2021, China released the country’s first-ever standards on autonomous driving classification. The Taxonomy of Driving Automation for Vehicles, formulated jointly by 11 major carmakers and suppliers and overseen by the MIIT, classifies six levels of automation, from Level 0 (L0), which has no automated driving capabilities, to Level 5 (L5), which is fully autonomous in all scenarios.
Previously, the industry had relied largely on the U.S. classification of automated driving levels. The new standards provide a clearer framework for development and lay a foundation for potential regulations and guidelines in the future.
It also sets standards for how vehicles can be marketed, requiring, for example, that L0 to L2 vehicles only be marketed as having ‘driver-assist’ functions. Many new car models produced and sold in China can be considered to be between L0 and L2 vehicles as they come with features such as emergency braking, park assist, and cruise control.
|Official AVs Classification in China|
|Level 0: Emergency response||The system cannot continuously perform lateral or longitudinal vehicle motion control during dynamic driving tasks but can continuously perform object and event detection and response during dynamic driving tasks.|
|Level 1: Partial driver assistance||Under conditions that it has been designed for, the system can continuously perform lateral and longitudinal vehicle motion control during dynamic driving tasks and can perform some object and event detection and response required for lateral and longitudinal vehicle motion control.|
|Level 2: Combined driver assistance||Under conditions that it has been designed for, the system can continuously perform lateral and longitudinal vehicle motion control during dynamic driving tasks and can perform some object and event detection and response required for lateral and longitudinal vehicle motion control.|
|Level 3: Conditionally automated driving||The system can continuously perform all dynamic driving tasks under the conditions it has been designed for.|
|Level 4: Highly automated driving||The system can perform all dynamic driving tasks and can automatically carry out minimal risk strategy under conditions it has been designed for.|
|Level 5: Fully automated driving||The system continuously executes all dynamic driving tasks under any driving conditions and automatically executes minimal risk strategy.|
|Source: Taxonomy of Driving Automation for Vehicles, Ministry of Industry and Information Technology and National Technical Committee of Auto Standardization|
The above standards will come into effect on March 1, 2022.
In July 2021, the MIIT, Ministry of Public Security, and Ministry of Transport (MOT) jointly issued trial management standards for road tests and demonstrations of intelligent connected vehicles.
The standards outline specific eligibility requirements for the operators of tests and demonstrations, as well as for the safety driver (which is required for all AV road tests) and the vehicle itself.
The standards require that a vehicle have a human-operated mode in addition to an autonomous mode and must ensure that a human can take over from the system quickly and easily in any scenario. They also have specific requirements for data collection and storage capabilities, outlining the types of data the vehicle is required to be able to collect and respond to, how quick its response to the data must be, and how long it must store the data for.
China has been greatly strengthening its data regulatory environment in recent years, releasing a series of landmark regulations, with the Cybersecurity Law (CSL), the Data Security Law (DSL), and the Personal Information Protection Law (PIPL) forming a foundation for regulating digital industries.
Data-guzzling intelligent vehicles are therefore becoming subject to increasing scrutiny and stricter regulations as the authorities seek to improve the safety of the technology and protect user data and national security.
In May 2021, the Cyberspace Administration of China (CAC) began soliciting opinions on a set of trial provisions titled Several Provisions on Vehicle Data Security Management (the ‘provisions’), which outline new requirements for manufacturers and operators of intelligent connected vehicles to protect personal information and ‘important’ data. The finalized trial version of the provisions was later released in August 2021 and took effect on October 1, 2021.
The provisions define ‘personal information’ as information from any stakeholders, including car owners, drivers, passengers, and pedestrians, as well as any information from which one can infer the personal identity of a stakeholder or describe their personal behavior.
‘Important’ data includes vehicle data collected from sensitive areas, such as military zones and government agencies, surveying and mapping data of higher precision than that released by official state maps, and any audio or video data collected from outside the vehicle, such as faces, voices, and license plates (this data is collected by vehicle cameras and sensors).
The provisions overlap significantly with the other data legislation. For instance, they require operators to inform users of the data they are collecting and delete it upon their request, while also outlining specific measures for storing the data to ensure its safety. The regulations, as is also stipulated in China’s data security law, require that any data collected in China must be stored in China, and must undergo a security assessment before it can be exported. These requirements are also stipulated in the PIPL, which comes into effect on November 1, 2021.
Existing data laws also place strict security requirements on entities that collect and store data, requirements that have now been expressly extended to intelligent vehicle operators through several official documents.
A set of guiding opinions on data management for makers of intelligent connected vehicles and product access released by the MIIT in August 2021 outlines requirements for the storage and protection of personal and core information collected from users in China. The document mandates that companies implement their own data security management systems, appoint personnel responsible for the security management, and carry out security assessments to identify potential threats and risks to cybersecurity. The guiding opinions also reiterate the ban on the export of data collected by vehicles in China without approval from the authorities.
These same requirements are stipulated in China’s Cybersecurity Law, Data Security Law, and PIPL, but were previously not expressly extended to vehicles.
Risks and limitations of China’s self-driving car industry
The tightening of data and internet regulations will impact almost all sectors that work in the digital space, and recent regulatory developments have ensured that automakers and AV operators will also have to adhere.
Given the industry’s heavy reliance on processing data to develop the technology, developers and automakers will have to make changes to their technical design and business structure to ensure compliance.
Industry players have already come under pressure to comply with newly tightened regulations. Several foreign automakers, including Tesla, Ford, and BMW, have built data centers in China to store data collected from Chinese users in order to comply with data regulations.
However, given the government’s eagerness to develop the industry, it remains unlikely that it will regulate it to the extent that it will seriously hinder development. At the same time, stronger data security management and governance may also help to improve the image of the industry and steer it in a direction of increased security and reliability.
Impact of auto chip shortage on China’s self-driving car industry
The global auto industry has been hit hard by a worldwide chip shortage, and China is no exception. Several leading automakers have been forced to cut output from their Chinese factories due to a lack of chips, which are indispensable for almost all modern vehicles, even those with no automated driving features.
In China, this issue has been exacerbated by the U.S.-China trade war, which has seen the U.S. ban exports of microchips to companies on the Entity List. The trade war may make it increasingly difficult for U.S. companies to supply chips to Chinese AV developers, further exacerbating the chip crunch and hampering industry growth. Forecasts for the industry’s development have already been pushed back a couple of years due to the impact of the Covid-19 pandemic, which is the primary cause of the chip shortage.
To counteract this, several chipmakers, including Taiwan’s TSMC, have pledged to open more fabs in China and increase production. At the same time, China is doubling down on its domestic microchip industry. Some players in the market are already developing their own auto chips, such as Baidu, which recently launched its second-generation Kunlun chip for autonomous driving.
Public opinion of China’s self-driving car industry
As is the case in other parts of the world, one of the largest headaches for China’s self-driving car industry is convincing consumers that the technology is safe. Despite China’s population generally being open and accepting of new technology, many people are still wary of entrusting the potential life-or-death task of driving a car to a computer.
Concerns over safety features have already impacted automakers, most notably Tesla, which received negative attention in April 2021 after a consumer alleged that a recent crash had been caused by faulty brakes.
Strong public relations will therefore be crucial to building trust with consumers, not only for AV operators themselves but also for companies providing core technology, as there is the potential that they also get thrust into the limelight should accidents occur.
Tighter government regulations and better oversight may also help to improve the image of AVs, as consumers are generally trusting of the government to ensure security and order. Companies that are seen to actively comply with laws and regulations may therefore be viewed favorably by the public.
A long-term investment with huge potential rewards
China is the world’s largest auto market. A large portion of this growth can be attributed to private car sales; however, China also has a huge demand for mobility services, which accounted for 10 percent of new car sales at the beginning of 2019, according to McKinsey, and is set to continue growing. This presents a huge opportunity for AVs that can penetrate this space.
The technology also has exciting proposals for China’s vast logistics industry, offering to make networks and supply chains more efficient and lower cost. In addition, its ability to solve some of China’s thorniest problems – shrinking labor forces, congestion, air pollution – has gotten the government’s enthusiastic backing.
Although the technology cannot yet be deployed for commercial use, the race is on to develop a product that can begin transporting passengers or goods without the need for any human intervention. Whoever wins this slow-but-steady race is set to reap generous rewards.
This article was first published by China Briefing, which is produced by Dezan Shira & Associates. The firm assists foreign investors throughout Asia from offices across the world, including in China, Hong Kong, Vietnam, Singapore, India, and Russia. Readers may write firstname.lastname@example.org for more support.