EU tariffs on Chinese cars are fuelling the growth of Chinese factories in Europe

Europe is discovering that, whilst taxing a Chinese car is relatively straightforward, reducing its dependence on the industry that produced it is far more complex. Since 30 October 2024, when Brussels imposed countervailing tariffs — on top of the standard 10 per cent — the total duty burden on BEVs manufactured in China has risen to around 27 per cent for BYD, 28.8 per cent for Geely and 45.3 per cent for SAIC. Yet, the result has not been a straightforward retreat by Chinese manufacturers, but rather a shift in their mode of entry, which has entailed less reliance on finished vehicles exported from China and, at the same time, greater use of joint ventures, European brownfield sites and production capacity located directly within the protected markets. Consequently, the focus of the industrial debate has shifted from customs duties to ownership of the factory, the battery, the software and the platform.

The first consequence of the European regime – which is measured by the discrepancy between the product’s country of origin and the manufacturer’s nationality – has become clearly evident in the market data. The share of ‘Made in China’ BEVs in European electric car sales fell from around 22 per cent at its peak in 2024 to around 17 per cent in the first quarter of 2026; during the same period, however, Chinese brands continued to grow, with BYD expanding rapidly, whilst SAIC Motor suffered more due to the higher tariffs. Part of the statistical reduction in the Chinese share also stems from Western groups such as Tesla, BMW and Volvo shifting production destined for Europe out of China: although customs figures are improving, this does not necessarily imply a decline in China’s competitive presence. The tariffs apply to complete vehicles imported from China, but do not have the same effect on cars assembled within the European Union, nor on many PHEV configurations, CKD and SKD kits, or batteries imported separately. When the cost of tariffs, logistics and political risk exceeds the annualised cost of a local factory, the company brings production in-house within the protected perimeter. The result is that traditional import statistics are increasingly failing to measure what really matters to industrial policy. A BYD manufactured in Szeged in Hungary, whilst being European in terms of customs origin, may still be Chinese in terms of capital, platform, software and a significant part of the battery; conversely, a Western car produced in Shanghai is classified as Chinese in origin statistics, even though it is owned by non-Chinese capital. These two variables must now be treated separately.

The International Energy Agency estimates that, by 2025, China’s ICE-EV production capacity outside China will amount to around 1.7 million vehicles, compared with around 29 million within the country. Although the balance is still heavily skewed towards the domestic market, the overseas figure no longer represents a collection of marginal projects, but rather describes a production network linking South-East Asia, Latin America, Europe, Central Asia and other markets. In 2025, Chinese BEV plants in Thailand were operating at around 20 per cent capacity utilisation, whilst those in Indonesia were below 15 per cent. This is the point most often overlooked in the political debate: whilst, on the one hand, adding up the announced capacities creates the impression of an imminent industrial invasion, on the other, ignoring these plants simply because they are still underutilised means failing to recognise their future capacity. The project portfolio also reveals very different circumstances. BYD has its Rayong plant up and running in Thailand; Camaçari in Brazil is using the former Ford site, with an initial phase of 150,000 units and a much higher future target; Subang in Indonesia, which opened on 3 September 2026, has an initial capacity of 150,000 units; Szeged in Hungary has entered the start-up phase, although mass production has been postponed until late 2026; Manisa, in Turkey, has been put on hold; finally, the standalone project at Tanjong Malim in Malaysia has been abandoned in favour of a CKD model with a local partner. Speaking generically of ‘Chinese factories abroad’ therefore obscures crucial differences between actual production, nominal capacity, suspended investment and mere plans.

In Europe, Chinese expansion is particularly efficient as it often does not require the construction of a plant from scratch. Barcelona has reactivated the former Nissan site through the Chery-Ebro joint venture; Valencia is part of the Geely-Ford strategy; Stellantis has utilised Tychy and is relocating Leapmotor to Zaragoza; Rennes has been identified as part of potential cooperation between Stellantis and Dongfeng; in Galicia, an SAIC presence has been mooted for the latter part of the decade. For the host country, the outcome may be positive, as, rather than losing a site, it retains jobs, suppliers and infrastructure. The economic impact, however, depends on what is transferred alongside production: if only components, batteries and electronic systems designed elsewhere arrive, whilst the European plant carries out bodywork, painting and final assembly, the region retains manufacturing activity but not control over the most profitable technologies. This is why the old dilemma of ‘factory open or factory closed’ is no longer sufficient. The correct question is: who decides which model to produce, who approves the suppliers, who owns the software, who controls the updates, who manufactures the cells, where the e-drive is designed, and who retains the revenue from intellectual property. The most critical issue remains the battery. According to the IEA, in 2025 China controlled over 80 per cent of global cell production, around 85 per cent of active cathode materials and over 90 per cent of anode materials. Even when a car changes its country of origin, therefore, the core of the industry may remain largely tied to a Chinese supply chain. This dependence does not automatically disappear simply by opening a gigafactory in Europe. If the European plant belongs to a Chinese group, uses Chinese process technologies and remains dependent on cathode and anode materials from the same supply chain, Europe gains jobs, production capacity and logistical resilience, but not necessarily full industrial autonomy. In fact, the geography of production changes more rapidly than the ownership of the technology.

For Italia, the problem is more pressing because, although the industrial sector retains expertise in the automotive, mechanical and powertrain sectors, it has lost a significant part of its planned trajectory in the field of battery cells. In February 2026, Stellantis confirmed that Automotive Cells Company had initiated the process to halt the planned gigafactory projects in Italia and Germany, removing Termoli from the industrial roadmap originally envisaged as the third major European ACC hub. Termoli nevertheless retains production prospects for eDCTs, with a target of 300,000 electrified transmissions per year, but a hybrid transmission is no substitute for a national battery cell base for BEVs. The difference is substantial: Italia can remain competitive in advanced mechanical engineering and hybridisation, whilst the segment with the greatest potential for industrial growth is concentrated elsewhere. This is precisely why any new Chinese investment in Italia should be assessed not by the number of vehicles assembled, but by the depth of localisation. A plant that imports cells, modules, e-drives and software, and carries out only bodywork, painting and assembly in Italia, would improve production and employment, but would not bridge the strategic gap.