Author | Zheng Senhong
In the first quarter of 2026, CATL (Contemporary Amperex Technology Co., Limited) delivered a financial report: revenue of 129.1 billion RMB, net profit of 20.738 billion RMB, and a daily net profit of 230 million RMB.
During the same period, the average profit margin of the domestic OEM (Original Equipment Manufacturer) industry was approximately 1.5%. For reference, the overall average profit margin of the manufacturing industry typically hovers around 5%.
Car manufacturing has become one of the least profitable businesses in the manufacturing sector.
Batteries are just one part of the equation. During the same period, Huawei Yinwang's gross margin reached 55%, while Momenta and Horizon Robotics posted even higher figures.
Although the business structures and revenue recognition criteria of these three companies vary, the trend is consistent: capital is flowing into these specific segments.
During the most brutal phase of the automotive price war, the profit curves of leading suppliers in the battery and intelligent vehicle sectors remained virtually unshaken, exerting pressure on automakers.
However, this is not the universal reality for all suppliers. Traditional Tier 1 suppliers focusing on wiring harnesses, interior and exterior trims, and stamped parts are equally suffering from the squeeze of annual price reductions and extended payment terms.
Chen Shihua, Deputy Secretary-General of the China Association of Automobile Manufacturers (CAAM), bluntly stated that for a 200,000 RMB car, the automaker's revenue is roughly 3,000 RMB. This is merely a static estimate and does not represent the actual profit of any specific model. But the money is not being made by traditional upstream component makers; it is being captured by those involved in intelligent technologies, batteries, and chips.
This contrast has spawned a popular narrative: automakers are working for suppliers, and the money is being siphoned off upstream.
This statement is half right. The correct half is that profits in the automotive industry are indeed concentrating in a few specific segments.
The incorrect half is that interpreting it as a zero-sum game of "suppliers bleeding automakers dry" will cause one to miss more critical shifts.
Because looking solely at the gross margin on paper cannot explain why the localization rate of Volkswagen's ID. UNYX models has reached 95%.
A high localization rate indicates a high proportion of local procurement, which does not mean that core technologies and core profits are all retained domestically; these are two different things. Meanwhile, if the domestic supply chain were merely "earning the spread" from Chinese automakers, it would not explain why global automakers are proactively relocating the starting point of their technical validation to China.
The zero-sum game narrative falls apart the moment the domestic supply chain begins supplying global automakers.
Therefore, a more accurate question is not "who is taking the money," but rather three related yet distinct issues:
Which segments are actually making money?
Why Chinese suppliers?
And how far are the profitable Chinese suppliers from achieving true industry dominance?
01. Where the Money Is Flowing: Three Distinct Types of Businesses
If you tear down an electric vehicle, where do the profits go? The answer is simpler than one might expect.
But first, let's draw a boundary: it is not the "supplier" category as a whole that is making money, but only a small fraction of it.
The ones truly sustaining high profitability are three structurally distinct types of businesses: scale manufacturing, platform reuse, and software functionalities.
The cost focus of ICE (Internal Combustion Engine) vehicles lies in the engine and transmission, whereas EVs (Electric Vehicles) have replaced these two with batteries and intelligent systems.
For pure electric models equipped with large-capacity batteries and advanced autonomous driving systems, these two items combined account for over half of the BOM (Bill of Materials) cost.
For PHEVs (Plug-in Hybrid Electric Vehicles) and entry-level models, this proportion is typically much lower.
More interestingly, for this portion of the cost, most automakers can only rely on external procurement. To be precise, it is the battery cell segment that mostly relies on external sourcing. In-house development of battery packs, joint development, and joint venture factory construction are matters on another level, with the joint venture between Li Auto and Sunwoda serving as an example.
Category 1: The Business of Scale Manufacturing.
When it comes down to car manufacturing, the first to get rich are the battery sellers.
CATL's net profit last year reached 72.2 billion RMB, three times Tesla's net profit for the same period. The two companies have completely different business structures, capital expenditures, and regional revenues; this comparison highlights a contrast, not a rule. A more effective frame of reference for this figure is that it exceeds the annual revenue of most OEMs.
But there is a hidden thread behind this number: CATL's domestic market share dropped from 52.1% in 2021 to 43.42% in 2025.
The companies that absorbed this market share are CALB (China Aviation Lithium Battery), Gotion High-Tech, and EVE Energy, with market shares reaching 6.98%, 5.65%, and 4.11% respectively. Sunwoda and SVOLT also saw growth rates exceeding 30%.
The driving force behind CATL's declining market share is the automakers themselves.
Apart from Maextro, all other brands under HIMA (Harmony Intelligent Mobility Alliance) have introduced tier-2 battery makers like CALB and Gotion High-Tech.
A small detail is that Gotion High-Tech customized 40GWh of capacity for Huawei in Wuhu, with Huawei involved throughout the entire process from factory planning to production line engineering.
Li Auto went even further. Last year, it established a joint venture battery company with Sunwoda. The battery packs for the Li Auto i9 are produced in-house by Li Auto, while the battery cells are supplied by both Sunwoda and CATL Jiangsu.
Li Auto elevated the status of its second-source supplier from mere procurement to a joint venture partnership.
From price control to cost control, the motivations of Huawei and Li Auto are identical. For battery packs of the same specifications, quotes from tier-2 battery makers are about 10% lower than CATL's. With annual sales of a million vehicles, this can save billions of RMB.
No automaker fails to understand this math; XPeng, Leapmotor, Tesla, BMW, and Volkswagen are all following suit.
However, the effect of breaking down these barriers is far smaller than imagined. Despite CATL's market share dropping by 8% over the past three years, its net profit actually surged from 30.7 billion RMB to 72.2 billion RMB.
This is not solely due to economies of scale. The overall market is still expanding, and a drop in market share does not equate to a decline in installed capacity; the growth of the energy storage business, the increase in the proportion of overseas revenue, and the decline in raw material prices are all driving up profits.
Yet CATL's annual installed capacity is more than ten times that of any single automaker's self-built production line, and seven times that of CALB. The most fundamental rule of manufacturing is that, given a sufficiently high capacity utilization rate, the larger the scale, the lower the cost.
CATL's products now cover 150 countries and regions globally, serving 176 brands and 529 passenger car models.
Although automakers can slash procurement prices by 10% by supporting second-source suppliers, they still cannot match the costs amortized by CATL's scale in the short term. The battery business is one where scale determines survival, and this fact remains unshaken.
Category 2: The Business of Platform Reuse.
Seres has paid Huawei 75 billion RMB over three and a half years. Channel service fees account for 8%, technology licensing for 2%, and the rest goes to cockpit and autonomous driving hardware. The channel fees go to the Consumer BG, while the technology foundation goes to Yinwang.
Huawei Yinwang generated 45 billion RMB in revenue last year, with a gross margin of approximately 55% and a profit of over 5 billion RMB. This money flows down three paths.
HIMA contributes 25.6 billion RMB, with the Turing Chassis, Giant Whale Battery, HarmonyOS Cockpit, and Qiankun ADS packaged for different vehicle model tiers.
The HI (Huawei Inside) model contributes 11.4 billion RMB, selling only technology and solutions. Now, a new HI Plus model has been added, launching the "Jing" brand through deep technological customization.
Components contribute 8 billion RMB, with LiDAR, cockpits, electric drives, and motors sold separately.
It is evident that HIMA is the ballast stone for Huawei Yinwang, delivering nearly 590,000 vehicles in 2025, with a target of 1 to 1.3 million vehicles this year. The HI model serves as the growth engine, with an expected delivery of over 400,000 units this year.
The essence of ecosystem fees is the scaled reuse of standardized technology. With the same underlying platform, the more automakers that use it, the lower the R&D cost amortized per company, and the more Huawei earns.
Category 3: The Business of Software Functionalities.
Research by JLAN Road shows that autonomous driving ranks third in car purchase demands for the 200,000 to 300,000 RMB segment, and enters the top two for segments above 300,000 RMB. As user demands shift, suppliers profit accordingly.
The optional installation rate for Horizon Robotics' HSD V2.0 reaches as high as 77%. Momenta's latest installed volume has surpassed one million units, currently covering over 57 mass-produced models.
The snowballing speed of demand cannot be reflected solely on the gross margin curve.
Installed volume and optional installation rates can indicate rising demand, but to explain profits, one must look at figures such as revenue per vehicle, the proportion of software licensing, and the amortization of R&D investments prior to mass production.
Horizon Robotics and Momenta posted gross margins of 64.5% and 71.6% respectively last year, whereas the ceiling for traditional Tier 1 suppliers is less than 25%.
A reality check is necessary here: high gross margin does not equal profitability. The gross margin in the software business is naturally high because the marginal cost approaches zero.
The true test lies in the R&D expense ratio. Autonomous driving companies are generally still trading massive R&D investments for vehicle model nominations, and most have not yet achieved stable net profits. The gross margin curve looks attractive, but the profitability model has not yet been fully proven.
These three types of businesses illustrate one fact: buying is better than building.
This holds true for most automakers, but not necessarily for a few top players with massive scale and long-term technological accumulation (such as BYD and Tesla). They precisely demonstrate that vertical integration can also be financially viable at a sufficient scale.
Domestic suppliers have used the three yardsticks of scale, reuse, and technology to measure the cost and time gaps on the automakers' path to in-house development.
02. Capabilities Forged by Hyper-Competition
Profit is merely the result. What is more worth questioning is why Chinese suppliers can win these orders, including those from global automakers.
"True competition is not between individual companies, but between supply chains." This quote by supply chain expert Martin Christopher has found its validation in the EV industry.
FAW-Audi replaced Bosch with Huawei; Mercedes-Benz entrusted its autonomous driving to Momenta; BMW deeply bound itself with CATL; and for the Nissan N7, from batteries to autonomous driving, a complete Chinese solution is implemented from start to finish.
From BBA (BMW, Benz, Audi) to the top three Japanese automakers, mainstream joint venture automakers are significantly increasing their procurement proportions from Chinese battery, autonomous driving, and cockpit suppliers.
The reason is simple: in global competition, the domestic supply chain has forged three core capabilities that cannot be replicated.
First is comprehensiveness. To be precise, the density of this supply chain in segments such as battery materials, battery cells, and intelligent cockpits is rarely matched globally.
From lithium mines to battery cells to battery packs, from chips to algorithms to OS (Operating Systems), and from sensors to domain controllers to complete vehicle platforms, China has formed one of the most complete and densest NEV (New Energy Vehicle) supply chains in the world.
However, "comprehensive" does not mean "fully autonomous." Certain processes for automotive chips, EDA (Electronic Design Automation) tools, and several core pieces of equipment remain embedded in the global system.
When it comes to "assembling an electric vehicle within a radius of a few hundred kilometers," there is no second option.
Imagine the supply chain of a power battery. The lithium comes from Ganfeng's mines, the electrolyte from Tinci Materials, the separator from Enjie's production lines, the battery cells roll off the production line at CATL's factory, and finally, the battery packs are transported to the OEM's workshop for installation.
This comprehensiveness brings about the response speed of the supply chain.
Around Tesla's Gigafactory Shanghai, a cluster of over 400 core suppliers has been formed. For all components including batteries, motors, motor controllers, thermal management, and chassis, the time from order placement to delivery is strictly controlled within 4 hours.
Second is speed.
Speed is built upon comprehensiveness and has been pushed to the extreme by the life-and-death pace of automakers.
The development cycle for a completely new platform by European automakers is five years, and just shipping the components from Europe to China takes over three months.
The domestic supply chain compresses the time from receiving requirements to solution delivery to within 12 months. Before the launch of a new Leapmotor vehicle, CATL mobilized its core technical team to urgently retrofit the production line, reducing the mass production preparation time from three months to two.
In the first five months of 2026, the domestic automotive market launched 542 new models, averaging 3.6 models per day. The frequency of new car launches has become almost as frequent as the frequency of meals.
But this is just the standard speed. What truly pressures global competitors is the suppliers' instantaneous response in handling the capacity for hit models.
A supplier manufacturing plastic cockpit components, in order to keep up with the delivery pace of the AITO M9, relocated directly from Zhejiang to Chongqing. The new factory is less than a five-minute drive from the Seres factory, delivering a batch of goods every two to four hours.
When the race for speed reaches its peak, every link in the entire chain is compressing time simultaneously.
Third is resilience.
One of the most underestimated capabilities of the domestic supply chain is resilient production.
On the same battery production line, CATL can simultaneously provide battery cells with different specifications, different chemical systems, and different packaging forms for dozens of automakers.
A single production line at CATL's Changan Yibin factory simultaneously runs battery cells for three brands: Deepal, Changan Qiyuan, and AVATR.
This flexible switching is not limited to batteries. Based on the same underlying architecture, Huawei's Qiankun ADS is divided into four solutions: SE, Pro, Max, and Ultra, covering vehicle models priced from over 100,000 RMB to over 1 million RMB.
From a market performance perspective, SE captures scale, Pro secures profits, Max builds reputation, and Ultra sets the standard. Four tiers on the same platform, each fighting in its own battlefield.
Resilience is forced by the fragmented demands of the market. Dozens of automakers sharing the same battery factory and the same autonomous driving company have honed the suppliers' physique to flexibly scale between extreme volume levels.
This ability to cater to both large and small volumes and to elastically scale enables the domestic supply chain to not only possess scale advantages but also respond to demand battles with agility.
The trinity of comprehensiveness, speed, and resilience is the strongest card in the hands of the domestic supply chain, forming a practical constraint in the procurement decisions of global automakers: not using the Chinese supply chain means longer cycles, higher costs, and slower iterations.
For global automakers, partnering with the domestic supply chain is no longer just an option for cost reduction, but an inevitable choice to maintain technological competitiveness and market response speed.
03. How Far from Becoming the Supply Chain Leader?
The domestic supply chain has transformed from a supporting role to the leading role.
But regarding the term "supply chain leader," we must first clarify what it entails: the ability to define technical standards, control key interfaces, possess pricing power, organize upstream and downstream coordination, and ultimately, the willingness to bear market risks.
Let's look further with this yardstick in mind.
Beneath the surface, the domestic supply chain has completed two structural reversals.
The first change: For some models targeting the Chinese market, the starting point for technical validation of global automakers' next-generation technologies has shifted.
The R&D and validation for the batteries, autonomous driving, and cockpits of Volkswagen's new CEA architecture are all completed in China. Volkswagen Anhui, which implements this architecture, has achieved a localization rate of 95% for its first batch of models.
On the other hand, Volkswagen has also deeply bound itself with Horizon Robotics through CARIZON. The Journey 6M has already been mass-produced and installed in the ID. UNYX 06/07. By the end of this year, 2026, it plans to launch the advanced version of Journey 6H, enabling urban NOA (Navigate on Autopilot).
It is understood that over 700 engineers at CARIZON are simultaneously developing the next-generation SoC (System on Chip) in-house, with the goal of reserving computing power and system capabilities for more advanced autonomous driving functions.
Volkswagen's shift is not an isolated case. BMW has laid out the largest R&D network outside of Germany in China, currently boasting over 3,000 engineers and four major R&D bases.
BMW Chairman Oliver Zipse bluntly stated that the Neue Klasse models cannot be realized without the support of Chinese suppliers.
This is no longer simple localization. Localization used to mean taking German solutions to China for adaptation, but now it is completely reversed. R&D, validation, and supply chain coordination are moving closer to China's manufacturing bases because the supply chain is here, the demand is here, and the iteration speed is here.
The second change: Upgrading from selling parts to selling complete solutions.
In the ICE vehicle era, domestic suppliers were mostly concentrated in manufacturing segments with lower technical barriers and added value, while high-value-added systems were long dominated by international Tier 1 suppliers.
Wanxiang Qianchao supplied Volkswagen with drive shafts for decades, and Tuopu Group manufactured rubber vibration dampers for half its existence, yet their profits remained paper-thin.
In 2021, the revenues of Wanxiang Qianchao and Tuopu Group were 14.3 billion RMB and 11.46 billion RMB respectively, while during the same period, Bosch's automotive business revenue in China alone reached 96.7 billion RMB.
In the NEV era, the rules of the game have completely changed.
What the domestic supply chain is truly selling is no longer processed materials, but packaged capabilities.
Desay SV is a typical case. Initially an OEM for Audi CD players, it is now the world's largest intelligent cockpit supplier, with its net profit growing from less than 300 million RMB to 2.454 billion RMB.
Even with the same supplier role, the profits change along with the products of different eras.
The old label of the "world's factory" is insufficient to define today's domestic NEV supply chain. A more accurate statement is that China is becoming one of the most important manufacturing and innovation centers in the global NEV supply chain.
The dividing line between the two lies in bargaining power and strategic influence. This is the most valuable chip in the industrial chain, determining the direction of technology and profit distribution.
But this influence is not pushed up by gross margins; it depends on harder metrics such as the magnitude of annual price reductions, the length of payment terms, the proportion of sole sourcing, and customer switching costs.
Measured by the yardstick of a supply chain leader, Chinese suppliers have gained some parts but still lack others:
The power to define the complete vehicle remains in the hands of automakers. Suppliers have no access to branding, channels, user data, or final procurement decisions.
Annual price reductions and extended payment terms have not disappeared. The automakers' moves to support second-source suppliers prove that the ceiling for procurement prices always exists, as evidenced by CATL's domestic market share dropping by nearly nine percentage points over four years.
The profitability model for the software business has not yet formed a closed loop. Beneath the high gross margins, most autonomous driving companies have not yet moved past the stage of trading R&D investments for design nominations.
Globalization is constrained by geopolitics. No matter how high the penetration rate in China, it cannot be automatically converted into market access in Europe and the US. Overseas capacity, certifications, and regulations are all new slopes to climb.
Over the forty years of the ICE vehicle era, domestic suppliers did one thing: they pushed manufacturing capabilities to the extreme. In any corner of the world, a more complete automotive component industrial cluster than the domestic one cannot be found.
In the EV era, this manufacturing network has been reactivated. With the same cluster density and manufacturing capabilities still intact, the technological content of the products has leaped forward by an era.
However, the strength of leading suppliers does not mean all suppliers are strong. High gross margins on paper do not mean the money has truly been pocketed. A high localization rate does not mean the power to make the rules has changed hands.
A more accurate statement is: the Chinese supply chain has upgraded from a "manufacturing supporter" to a "key technology solution provider," and is currently competing for the right to participate in and define the rules in core segments such as batteries, cockpits, and autonomous driving. Becoming a supply chain leader is an ongoing process, not an established fact.
The skeleton built over forty years has suddenly been endowed with new flesh and blood.
The skeleton is hard to replicate, and the flesh and blood are still growing. This climb from supplier to rule participant is halfway done. The remaining half will be decided in the arenas of pricing power and the global market.