The European Union’s “Made in Europe” Regime

I. The diagnosis: the Draghi Report and the decline of European industry

The intellectual origin of the Industrial Accelerator Act lies in the report The Future of European Competitiveness, prepared by Mario Draghi at the request of Commission President Ursula von der Leyen and published on 9 September 2024. The report, in two parts, found that closing the European Union’s competitiveness gap would require additional annual investment in excess of EUR 800 billion — approximately 5 per cent of Union gross domestic product.

Draghi’s diagnosis was organised around three axes: closing the innovation gap with the United States, reconciling decarbonisation with competitiveness, and enhancing security by reducing strategic dependencies. His findings on energy costs became a reference point repeated in every subsequent Commission document: industrial electricity prices in Europe are two to three times those in the United States and China, and natural gas prices three to five times those in the United States.

What matters most for present purposes is that the report expressly advocated the principle of “European preference”. Draghi argued for the strategic use of public demand — that is, for directing public procurement and public support towards production located in Europe — and gave institutional form to that idea in the European Defence Industry Authority he proposed for the defence sector. The origin preference introduced by the Industrial Accelerator Act in public procurement and support schemes is that same idea, transposed to civil strategic sectors.

It should also be noted that the report’s implementation record is weak. According to an assessment carried out one year after publication, only 11.2 per cent of its 383 recommendations had been fully implemented and 46 per cent were regarded as “in progress”; Draghi himself stated in September 2025 that every problem he had identified had worsened. The Industrial Accelerator Act is one of the institutional answers to that criticism.

II. The intermediate steps: the Competitiveness Compass and the Clean Industrial Deal

The Commission translated Draghi’s three axes into its own political programme by means of the Communication of 29 January 2025 entitled A Competitiveness Compass for the EU. That is also the document in which the Industrial Accelerator Act was first identified as a legislative instrument, although it appeared there under the name Industrial Decarbonisation Accelerator Act.

The Communication of 26 February 2025, COM(2025) 85, entitled The Clean Industrial Deal, gave the first concrete indication of what the instrument would contain. The instrument was to introduce “clean, resilient, circular and cyber-secure” criteria designed to stimulate demand for sustainable products manufactured within the Union; to accelerate permitting for decarbonisation and modernisation projects in energy-intensive industry, beginning with steel; and to establish a low-carbon labelling scheme starting with steel and extending to cement.

III. A change of name and of scope: from decarbonisation to industrial policy

Public consultation on the draft ran from 16 April 2025 to 9 July 2025, and the text was expected to be tabled in the final quarter of 2025 and, in later timetables, in December 2025. It was in the event tabled only on 4 March 2026; the reason for the delay, so far as the press has reported it, was internal disagreement over the “Made in Europe” provision.

Two fundamental changes occurred in that period. First, the word “decarbonisation” was dropped from the title. This was more than a matter of nomenclature; it signalled that the instrument’s centre of gravity had shifted from climate policy to industrial policy. Secondly, the scope both widened and narrowed: a text conceived as a labelling and permitting instrument centred on steel and cement became a far broader measure encompassing European preference in public procurement, foreign direct investment conditionality and cross-sectoral content requirements; conversely, the final draft leaked to the press on 4 March 2026 removed quantum technologies, artificial intelligence, semiconductors, biotechnology, robotics, sensing, space and propulsion technologies from its scope.

The political divide between Member States was decisive in shaping that outcome. France pressed for a more robustly protective text directed against State-backed competition from China and the United States, while Germany, Sweden, the Netherlands and Finland favoured a more measured, market-oriented approach. A similar division emerged within the Commission: the Commissioner for Industry, Stéphane Séjourné, argued for stricter origin criteria, whereas the Directorate-General for Trade preferred a broader approach extending equivalent treatment to free-trade-agreement partners.

The clearest trace of that internal debate is found in the revised draft leaked to the press on 12 February 2026. That draft softened the mandatory Union-origin approach for steel and introduced the third-country equivalence mechanism into the text. This mechanism, examined in detail below, is the provision that determines Türkiye’s position under the measure, and it entered the text at the last moment as a result of lobbying. Two consequences follow for Türkiye: its present position is an achievement, but an achievement capable of being withdrawn in the course of the legislative process.

IV. The Chinese context

The Chinese factor behind the measure is the cumulative product of three distinct legal developments.

The first is the anti-subsidy investigation concerning electric vehicles. Opened in October 2023, the investigation led to the imposition of provisional duties in July 2024, and definitive countervailing duties were adopted on 4 October 2024 and entered into force at the end of October 2024, for a period of five years, by Implementing Regulation (EU) 2024/2754. The rates were set at 35.3 per cent for SAIC, 18.8 per cent for Geely, 17.0 per cent for BYD, 7.8 per cent for Tesla’s exports from China and 20.3 per cent for other cooperating producers; these rates apply in addition to the Union’s standard 10 per cent customs duty on motor cars.

The second is Regulation (EU) 2022/2560 on foreign subsidies. Under that Regulation, in force since July 2023, the Commission opened its first ex officio investigation on 9 April 2024 in relation to subsidies for Chinese wind turbines, examining projects in Spain, Greece, France, Romania and Bulgaria.

The third is the structural deterioration in the trade balance. In 2024, EU–China trade in goods and services exceeded EUR 845 billion, leaving a deficit of more than EUR 300 billion against the Union. Between 2014 and 2024 the Union’s imports from China rose by more than 102 per cent, while its exports rose by approximately 47 per cent.

The Industrial Accelerator Act completes that set of instruments. Countervailing duties deter imports through price; the Foreign Subsidies Regulation scrutinises subsidised investment and participation in tenders; the Industrial Accelerator Act makes public demand and public support directly conditional upon origin. Read together, the three show the Union’s doctrine of “de-risking” towards China extending from trade defence law into industrial policy.

V. The legal precursor: the Net-Zero Industry Act

The Industrial Accelerator Act is not the first instance of origin-based preference in Union law. Regulation (EU) 2024/1735 of 13 June 2024, the Net-Zero Industry Act, already required that “sustainability and resilience” criteria — that is, non-price criteria — be assessed alongside price in public procurement and in renewable energy auctions.

The Industrial Accelerator Act extends that logic in three respects. As to scope, it broadens the field of application of non-price criteria from “net-zero technologies” to all decarbonisation projects in energy-intensive industry. As to intensity, it converts what was a flexible “assessment criterion” into numerical and binding thresholds. As to instrument, it adds public support schemes alongside public procurement — which, as will be seen below, is the genuinely problematic step under World Trade Organization law.

In addition to Regulation (EU) 2024/1735, the proposal would also amend Regulation (EU) 2024/3110 on construction products and Regulation (EU) 2018/1724 on the Single Digital Gateway. The amendment to the Construction Products Regulation is intended to prevent definitions of low-carbon concrete, mortar and steel from appearing in different forms in two separate instruments; the amendment to the Single Digital Gateway Regulation is intended to ensure that the accelerated permitting procedures are conducted through the existing digital infrastructure.

I. Legal basis and structure

The proposal rests on two legal bases: Article 114 TFEU, concerning the harmonisation of the internal market, and Article 207(2) TFEU, concerning the common commercial policy. The second basis is required for the chapter on foreign direct investment conditionality, since that chapter regulates the market access of third-country investors.

The text comprises six chapters and four annexes. Chapter I contains general provisions and definitions; Chapter II, the enabling conditions for industrial production and decarbonisation (permitting); Chapter III, the Union-origin and low-carbon requirements; Chapter IV, the conditions applicable to foreign direct investment; Chapter V, industrial manufacturing acceleration areas; and Chapter VI, common and final provisions. Of the annexes, the most important for present purposes is Annex III, which lays down the Union-origin requirements applicable to vehicles.

The measure’s declared macroeconomic objective is to raise the share of manufacturing in Union gross domestic product from 14.3 per cent in 2024 to 20 per cent by 2035. Manufacturing currently accounts for 18.3 per cent of Union employment and is responsible for 26.2 per cent of the Union’s greenhouse gas emissions.

II. Two criteria: Union origin and low carbon

The first step in analysing the proposal is to distinguish two separate criteria.

Union origin is a geographical and value-chain criterion. It requires that the product, or its components, be manufactured within the Union — or in a third country treated as equivalent.

Low carbon is a performance standard. It requires the product to fall below a given carbon-intensity threshold; where it was manufactured is immaterial.

That distinction is decisive both as a matter of trade law and for Turkish industry. For some product groups only the low-carbon criterion applies, while for others the two criteria apply cumulatively. According to the text of the proposal:

III. European preference in public procurement

The origin and low-carbon requirements are superimposed as an additional layer upon public procurement governed by Directives 2014/23/EU (concessions), 2014/24/EU (the classic sector) and 2014/25/EU (water, energy, transport and postal services). The proposal thus does not establish a new procurement regime; it operates by adding origin conditionality to the existing one.

The procurement provisions are to apply to procurement procedures launched on or after 1 January 2029. It should be stressed that this date is independent of, and later than, the entry into force of the proposal.

IV. Conditionality in public support schemes

This is the most radical provision of the proposal, and it has not attracted the attention it deserves. Origin preference in public procurement may rely upon an exception recognised in international trade law; no such exception exists for public support (see Chapter Five below).

The proposal requires Member States to make a given proportion of the national budget allocated to public support schemes in the relevant sectors conditional upon the origin and/or low-carbon requirements: at least 45 per cent of the national budget in support schemes relating to energy-intensive industry; 100 per cent of the national budget in support schemes relating to vehicles.

That the figure for vehicles is set at one hundred per cent is no accident. Purchase incentives for electric vehicles are the instrument that in practice creates demand for such vehicles across the Union; making the whole of that incentive budget conditional upon origin constitutes the real source of pressure compelling Chinese manufacturers to produce in Europe.

V. Third-country equivalence: the mechanism that determines Türkiye’s position

The equivalence mechanism laid down in Articles 8 and 9 of the proposal allows the Union-origin requirement to be treated as satisfied in respect of content from certain third countries. The mechanism is, however, constructed asymmetrically as between the two channels of public demand.

Equivalence for the purposes of public procurement is available to countries falling within one of three categories: (i) countries with which the Union has established a free-trade area; (ii) countries in a customs union with the Union; and (iii) parties to the World Trade Organization Agreement on Government Procurement (GPA).

Equivalence for the purposes of public support schemes is narrower: only countries in a free-trade-area or customs-union relationship with the Union benefit; being a party to the GPA alone does not suffice.

The consequence of that asymmetry for Türkiye is this: by virtue of the customs-union limb, Türkiye falls within the scope of equivalence for both channels. For a country that is not a party to the GPA this is a notable position, and it operates to the advantage of Turkish industry. That conclusion is nevertheless qualified by a second asymmetry, examined below: equivalence of origin is not the same thing as market access.

There is also uncertainty as to the nature of the equivalence. According to recital 16 of the proposal, the Commission is to assess at regular intervals whether the conditions for excluding a third country from the scope of equivalence are met. That formulation indicates that equivalence arises automatically by category, but that the Commission is endowed with a standing power of exclusion. Equivalence is, in other words, not an acquired right but a revocable status. This gives rise to a problem of legal certainty for Türkiye and is directly connected with the reciprocity condition considered below.

VI. Foreign direct investment conditionality

Chapter IV of the proposal is without precedent in Union law: it makes foreign investment conditional not on grounds of national security but on grounds of industrial policy.

Two conditions must be satisfied cumulatively for the regime to be engaged: the value of the investment must exceed EUR 100 million, and the investor must originate in a third country holding more than 40 per cent of global manufacturing capacity in the relevant “emerging strategic sector”. As already noted, the second criterion in practice captures China alone.

The emerging strategic sectors covered are battery technologies; electric and fuel-cell vehicles together with their electrification and digitalisation components; solar photovoltaic technologies; and the extraction, processing and recycling of critical raw materials.

The competent national Investment Authority may require, as a condition of approval, that at least four of six conditions be met: that the foreign investor’s holding not exceed 49 per cent of share capital or voting rights; that a joint venture be established with a Union undertaking, again subject to the 49 per cent ceiling; that intellectual property and know-how be licensed to the target undertaking; that at least 1 per cent of gross annual revenue be spent on research and development within the Union; that at least 50 per cent of the workforce consist of Union workers; and that at least 30 per cent of inputs be sourced from Union suppliers. For transactions exceeding EUR 1 billion, decision-making authority passes directly to the Commission.

It should be emphasised that these provisions do not replace the existing foreign investment screening regime under Regulation (EU) 2019/452 but operate in parallel with it. The same transaction may thus require two separate authorisations under two distinct instruments, applying concepts of “control” that do not fully coincide.

VII. Permitting and industrial acceleration areas

The proposal provides for a maximum permitting period of 18 months for covered manufacturing, decarbonisation and net-zero technology projects, together with a period of 45 days (extendable by 30 days) for notification that an application is complete. The permitting provisions will begin to apply one year after entry into force, and each Member State will be required to designate at least one industrial manufacturing acceleration area within twelve months of entry into force.

It should be noted that the proposal does not exempt acceleration areas from the Environmental Impact Assessment Directive or the Habitats Directive. Acceleration is achieved through procedural coordination — a single competent authority and a single point of contact — while substantive environmental protection standards are formally preserved.

VIII. Timetable and review

The timetable set out in the proposal may be summarised as follows.

I. Why the automotive sector merits separate treatment

Among the sectors covered by the Industrial Accelerator Act, the automotive sector occupies a special place in three respects. First, it is the sector with the highest origin threshold (70 per cent). Secondly, it is the only sector for which the conditionality rate in support schemes is set at one hundred per cent. Thirdly, and most importantly, the measure operates in the automotive sector not through the Industrial Accelerator Act alone but through three interconnected legislative files.

According to the Commission’s impact assessment, the greater part of the anticipated increase in value added also comes from this sector: by 2030, additional value added of more than EUR 600 million is projected in the steel, aluminium and cement industries, as against an increase of up to EUR 10.5 billion across the automotive value chain.

The condition of the sector likewise forms part of the rationale for the measure. The recitals to the proposal record that more than 100,000 job losses were announced in the automotive sector in 2024 and 2025, and that supplier profitability fell from 7.4 per cent in 2017 to 5 per cent in 2023. Presenting the proposal, Commissioner Séjourné stated that 200,000 jobs had been lost in energy-intensive industry and the automotive sector since 2024, and that 600,000 job losses were projected in car manufacturing over the present decade.

II. The anatomy of the 70 per cent rule

The vehicle origin rule laid down in Annex III to the proposal imposes two cumulative requirements on vehicles that are to benefit from public procurement or public support schemes.

The first requirement — the value threshold. At least 70 per cent of the ex-works price of the vehicle’s components must be of Union origin. The battery is excluded from that calculation. This is an origin test based on value, not on change of tariff heading or on weight.

The second requirement — final assembly. The vehicle must have undergone final assembly within the Union. Satisfying the value threshold is not of itself sufficient; the two requirements are cumulative.

As to the vehicles covered, the measure is directed at clean vehicles: battery electric vehicles, plug-in hybrids and fuel-cell electric vehicles. Vehicles with internal combustion engines fall outside the origin rule — which is unsurprising, since the lever employed is public procurement of, and public support for, clean vehicles.

The text of the proposal draws no distinction between vehicle categories (M1 passenger cars, N1 light commercial vehicles, M2/M3 buses, N2/N3 heavy commercial vehicles). As will be seen below, this is one of manufacturers’ principal complaints and is the source of ACEA’s call for differentiation by segment.

The exception for small vehicles. A more flexible, alternative test is provided for small electric vehicles of less than four and a half metres (category M1E, with a threshold of approximately 4.2 metres): such vehicles may satisfy the origin requirement either by meeting the 70 per cent rule or by meeting the requirement that their battery cells be manufactured within the Union. That exception forms part of the policy of the “small, affordable electric vehicle made in the Union” and is directly connected with the super-credits under the carbon dioxide fleet regulation discussed below.

III. The battery regime and its progressive tightening

The exclusion of the battery from the 70 per cent calculation is not an exemption but a deferral and a subjection to a separate regime. Given Europe’s dependence on China for battery cell manufacture, including the battery directly in the value calculation would have rendered the rule unworkable.

The proposal instead establishes a graduated regime based on the number of components. Initially, at least three main battery components, including the battery cells, must be of Union origin; after three years that number rises to five, with cathode active material and the battery management system added to the list. According to assessments by the supplier industry, a separate Union-origin threshold of 50 per cent for critical components and electric powertrains will take effect at the end of that same three-year transition.

A single comparison suffices to show how ambitious this regime is: today, Union-origin content stands below 50 per cent in battery electric vehicles, as against more than 80 per cent in vehicles with internal combustion engines. The measure aims to close the gap of some thirty percentage points in electric vehicles within five to seven years.

IV. Where the rule bites: four channels of demand

To understand how the origin rule will operate in practice in the automotive sector, it is necessary to identify the channels of demand through which it applies. One of those channels lies outside the Industrial Accelerator Act, in a separate legislative file.

The first channel — public procurement of vehicles. This lies directly within the Industrial Accelerator Act. Annex III lays down the vehicle origin requirements applicable to procurement procedures.

The second channel — national support schemes. Member States are required to make the whole of the national budget allocated to public support schemes relating to vehicles subject to the origin and/or low-carbon requirements. Purchase incentives for electric vehicles fall within this channel.

The third channel — corporate fleets. This channel does not lie within the Industrial Accelerator Act. The proposal for a Clean Corporate Vehicles Regulation, COM(2025) 994 of 17 December 2025, would require that from 2030 a specified proportion of new passenger car and light commercial vehicle registrations by large undertakings be zero- or low-emission. Corporate vehicles account for approximately 60 per cent of new passenger car registrations across the Union and for up to 90 per cent of light commercial vehicle registrations. That proposal would make national financial support for corporate vehicles conditional upon “Made in Europe” status from 2028, and refers to the Industrial Accelerator Act for the content of that condition. The two files are legally independent of one another but substantively interdependent.

The fourth channel — the carbon dioxide fleet regulation and super-credits. The automotive package of December 2025 amended Regulation (EU) 2019/631 to reduce the 2035 fleet target from a 100 per cent to a 90 per cent reduction, allowing the remaining share to be met through e-fuels, biofuels and credits for low-carbon steel melted within the Union. The same package introduced super-credits for small electric vehicles manufactured within the Union and restricted pooling with third-country manufacturers of pure electric vehicles. The critical point is this: the automotive package left the “Made in EU” criterion undefined, deferring its definition to secondary legislation and to the Industrial Accelerator Act. The origin criteria in Annex III will therefore bear directly not only on public procurement but also on compliance with the carbon dioxide fleet targets.

Taken together, these four channels show that the “Made in EU” criterion has become not a condition of market authorisation in the European automotive market but a condition of competitiveness. The rule does not prohibit the sale of vehicles falling outside its scope; it excludes them from public procurement, from purchase incentives, from corporate fleet support and from fleet compliance credits. In commercial terms the difference is, more often than not, a formal one.

V. The industry’s response

The industry’s response to the measure is far from uniform, and the division is at its sharpest along the manufacturer–supplier axis.

ACEA (the European Automobile Manufacturers’ Association) advanced seven principal demands in the position paper it published in early July 2026: (i) incentives strengthened sufficiently to offset the cost of localisation; (ii) limitation of the geographical scope to the EU 27 plus the United Kingdom, coupled with targeted protection for existing investments in Türkiye and Morocco; (iii) a “leaner and fairer” content calculation taking account not only of parts sourcing but also of the manufacturer’s contribution in research and development, engineering and skilled labour; (iv) alignment of battery content targets with the actual ramp-up timetable for Union cell production; (v) differentiation by segment; (vi) calculation of content at the level of the completed vehicle rather than component by component; and (vii) gradual introduction of the rule with targeted exemptions. The Association warned that, as it stands, the proposal risks disrupting established value chains and stranding investments made in good faith under the existing regulatory framework.

In its earlier position paper of September 2025, ACEA had argued that value-added content tests were unworkable, since they would require full transparency across the supply chain, and had proposed instead a criterion based on change of tariff heading. That paper also emphasised that approximately one-third of Union vehicle production is exported under preferential trade agreements and that strict rules of origin give rise to a risk of retaliation.

Volkswagen, Stellantis and Renault, representing approximately 60 per cent of Union vehicle production, submitted a joint proposal on 12 June 2026. The proposal would replace the vehicle-by-vehicle pass-or-fail test with a principle of fleet averaging: if 70 per cent of the vehicles a manufacturer sells within the Union satisfy the content requirement, the manufacturer’s entire fleet would be treated as “Made in Europe”. The three manufacturers further argue that the content calculation should reflect the “full value” of a car — research and development, engineering and production taken together.

Toyota, through the President of its European operations, Yoshihiro Nakata, called in June 2026 for the United Kingdom, Japan and Türkiye to be accorded equivalent status, arguing that their exclusion would weaken investment, employment and the regional scale required for global competitiveness.

BMW has been the most forthright opponent. Management board member Milan Nedeljković described the “Made in Europe” plan as dangerous, stating that it would lead to less innovation, lower growth and, ultimately, diminished prosperity in Europe, and that Europe should strengthen free trade and improve the business environment rather than mandate localisation.

CLEPA (the European Association of Automotive Suppliers) has taken precisely the opposite position. The Association, which has more than three thousand members, insists that the 70 per cent threshold be maintained, arguing that any relaxation would remove the incentive to manufacture in Europe. CLEPA further calls for the foreign direct investment threshold to be lowered from EUR 100 million to EUR 30 million — supplier investments being typically of smaller scale — and for the cost-differential exception to be raised from the range of 25 to 30 per cent to 35 per cent.

That division is significant in legal terms. Manufacturers are seeking to relax the origin rule and suppliers to tighten it, because the stringency of the rule represents a compliance cost for the manufacturer and a protected market for the supplier. The direction in which the text evolves in trilogue will be determined in large measure by the relative weight of these two lobbies.

VI. The repositioning of Chinese manufacturers

The most tangible effect the measure has produced before even entering into force is the change in Chinese manufacturers’ investment strategies. Notwithstanding the countervailing duties, the share of Chinese brands in new car sales in the Union reached 6.4 per cent in 2025 and 9.3 per cent in December 2025.

An interview with BYD’s European adviser, Alfredo Altavilla, reported by Reuters on 17 September 2026, illustrates the nature of that repositioning. According to Mr Altavilla, Chinese manufacturers have now turned from building new plants to acquiring existing assembly facilities, the reason being to begin production before the local content rules take effect. He stated that BYD requires three assembly plants and one battery plant in Europe in order to comply with the rule; that Spain and France are the most workable options; and that Italy is out of contention because Stellantis holds the available spare capacity. The same interview set out the moves of BYD’s competitors: Leapmotor and Dongfeng accessing Spanish and French capacity through their partnership with Stellantis, Geely in discussions with Ford for Spanish capacity, and Chery taking over the former Nissan facility in Barcelona.

The most notable development for Türkiye is BYD’s investment in Manisa. The billion-dollar plant announced by the company in 2024, with a projected annual capacity of 150,000 vehicles, had been suspended as at June 2026. According to a statement by BYD’s Vice-President, Stella Li, the reason for the delay is not customs duties — Türkiye already enjoys access through the Customs Union — but uncertainty as to how the “Made in EU” local value-added rules will treat Türkiye in respect of vehicles benefiting from public procurement and public support.

That example encapsulates in a single case the central thesis of this study: even as a proposal not yet adopted, the Industrial Accelerator Act is already determining foreign direct investment decisions concerning Türkiye. The uncertainty surrounding Türkiye’s legal position under the measure has itself become a deterrent to investment.

I. The exposure of the Turkish automotive industry

To appreciate Türkiye’s sensitivity to this measure, the scale of its exposure must first be set out.

In 2025 the automotive sector, with exports of USD 41.5 billion, was Türkiye’s largest export sector and accounted for approximately 15.2 per cent of total exports; the corresponding figure for 2024 was USD 37.2 billion. The target announced for 2026 is USD 43 billion, and exports of USD 20.8 billion were achieved in the first half of the year. The sector accounts for approximately 5 per cent of Turkish gross domestic product and provides employment to more than 550,000 people, of whom 250,000 are directly employed. Türkiye is Europe’s fifth-largest vehicle producer.

The critical figure, however, is the destination of those exports. Approximately 72.5 per cent of Turkish automotive exports go directly to the European Union, and close to 85 per cent to Europe as a whole. In the words of Cengiz Eroldu, Chairman of the Automotive Manufacturers Association (OSD), this is “not merely a trading relationship but a deep industrial integration encompassing production, investment and supply chains”.

It should also be noted that the integration runs in both directions. The Union’s vehicle exports to Türkiye rose by 24.8 per cent in 2025 to EUR 18.8 billion, making Türkiye the Union’s third-largest export market, taking 11.1 per cent of its vehicle exports. In the same year 713,476 passenger cars of Union origin were shipped to Türkiye, of which 17.4 per cent were battery electric; vehicles of Union origin account for 49 per cent of the Turkish new car market.

The roster of manufacturers producing in Türkiye is the institutional expression of that integration: Ford Otosan, Tofaş (Stellantis), Oyak Renault, Toyota, Hyundai Assan, Mercedes-Benz Türk, MAN Türkiye and Anadolu Isuzu, together with the domestically owned Karsan, Otokar, TEMSA and TOGG. Approximately 1,100 component manufacturers are active, and the proportion of domestic parts in vehicles produced in Türkiye ranges between 50 and 70 per cent.

II. The position conferred by the Customs Union and its limits

As noted above, the equivalence mechanism laid down in Articles 8 and 9 of the proposal treats content originating in countries in a customs union with the Union as being of Union origin. Türkiye falls within that category by virtue of Association Council Decision No 1/95 and — notwithstanding that it is not a party to the GPA — is covered by equivalence for the purposes both of public procurement and of public support schemes.

This is an important gain for Turkish industry. Concretely, a battery management system manufactured in Manisa, a transmission manufactured in Bursa or a body panel manufactured in Kocaeli may be counted towards the 70 per cent Union-origin calculation for an electric vehicle assembled in Germany. Türkiye’s inclusion within this framework was described by the Minister of Trade, Ömer Bolat, on 5 March 2026 as “a positive and constructive decision”; the President of TOBB, Rifat Hisarcıklıoğlu, while regarding the development as “strategically significant”, emphasised that the next step must be the modernisation of the Customs Union.

A second and less widely noticed asymmetry, however, comes into play at this point. Equivalence of origin is not the same thing as market access. The equivalence provisions of the proposal ensure that Turkish goods are treated as being of Union origin; they do not enable Turkish undertakings to participate directly as tenderers in Union public procurement. Indeed, according to a statement by a Commission spokesperson reported in the assessment note of the Economic Development Foundation (İKV), Union undertakings will be able to source components from Türkiye, but Turkish undertakings will not be able to participate directly as parties in Union public tenders; reciprocity safeguards apply at that point.

Assessments occasionally found in the Turkish press to the effect that “Turkish producers will be able to compete on equal terms in EU public tenders” are, in this respect, inaccurate and should be treated with caution.

III. The scope of Decision No 1/95

The legal reason for that asymmetry lies in the scope of the Customs Union itself. Association Council Decision No 1/95 establishes the free movement of goods in industrial products and processed agricultural products, places Türkiye under an obligation to align with the Union’s Common Commercial Policy, and imposes an obligation to approximate its legislation to the Union acquis in the fields of competition and State aid.

There are, by contrast, three fields the Customs Union does not cover: services, agriculture (other than processed agricultural products) and — decisively for present purposes — public procurement. As the European Parliament’s study on the subject expressly states, the EU–Türkiye Customs Union contains no common rules on public procurement.

The “customs union” limb of the equivalence provision therefore confers equivalence on Türkiye at the level of the origin of goods; but because the Customs Union does not deliver the reciprocal opening of public procurement required for equivalence at the level of market access, that second tier of equivalence does not arise for Türkiye. In short: the Customs Union is necessary, but not sufficient.

IV. The reciprocity condition and the 15 per cent price advantage under Law No 4734

The Commission’s power of exclusion, foreshadowed in recital 16 of the proposal, becomes a concrete risk at this point. Equivalence is conditional upon the third country concerned according national treatment to products and undertakings of Union origin.

Under Turkish law, Public Procurement Law No 4734 provides for a mandatory price advantage of 15 per cent in favour of tenderers offering domestic goods in designated supply tenders. That is an explicit, legislatively mandated discrimination against foreign — and therefore Union — tenderers and goods. By the very structure of the reciprocity test, it cannot be maintained that Türkiye satisfies the national treatment condition for so long as that provision remains in force.

A further structural shortcoming is that Türkiye is not a party to the World Trade Organization Agreement on Government Procurement. Türkiye holds observer status alone in relation to that Agreement and — since the asterisk by which the Organization marks observers engaged in accession negotiations does not appear against Türkiye — is not at present conducting accession negotiations either.

Read together, those two points permit Türkiye’s position to be summarised as follows: Türkiye enjoys equivalence of origin by virtue of the Customs Union, but that equivalence is subject to the Commission’s power of exclusion, and legislation capable of triggering that power — the domestic goods price advantage under Law No 4734 — is in force in Turkish law.

So far as the press has reported it, the Turkish side is alive to that tension. In July 2026 it was reported that an amendment to the Public Procurement Law was being prepared to confer on the President the power to extend the price advantage provided in favour of domestic goods to goods and tenderers of Union origin on a reciprocal basis. That is a narrower and more rapid route than more comprehensive solutions such as accession to the GPA or modernisation of the Customs Union. Whether that amendment has in fact been enacted requires separate verification against the Official Gazette.

Comparable movement has been observed in relation to the Communiqué on Domestic Goods. By an amendment of 7 February 2026 the Ministry of Industry and Technology added a transitional provision to the Communiqué providing that its terms would not apply, until 1 July, to the calculation of the domestic contribution ratio for finished products in the automotive sector. The Ministry gave as its reason that, given the large number of suppliers and the diversity of inputs in the sector, declarations of domestic contribution ratio for the finished vehicle risked causing interruptions in public procurement and in vehicle purchases by disabled citizens. It was further stated that the exception covered finished products alone and did not extend to component manufacturers.

V. The risk created by the ACEA formula

The most serious threat to Türkiye arises not from the present text of the proposal but from amendments proposed in the course of the legislative process.

The formula contained in ACEA’s position of July 2026 — limiting the geographical scope to the EU 27 plus the United Kingdom while affording targeted protection to existing investments in Türkiye and Morocco — appears at first sight to favour Türkiye; but in its legal structure it reduces Türkiye’s position from categorical equivalence to individual protection.

Were that formula to be adopted, Türkiye would not sit within the general framework; only the investments of European manufacturers already producing in Türkiye — that is, Ford Otosan, Tofaş, Oyak Renault and the like — would be protected. The decisive criterion would consequently become not whether the vehicle was manufactured in Türkiye but the identity of the manufacturing undertaking and the date on which the investment was established. While that draws a critical distinction as regards Chinese brands wishing to manufacture in Türkiye, it would also weaken the position of Turkish-owned manufacturers and — more importantly still — of Turkish component manufacturers exporting independently of any European manufacturer.

The position to be defended on behalf of the Turkish automotive industry is therefore clear: categorical equivalence must be preserved, and individual protection must not be accepted in its place.

VI. Türkiye’s institutional response and its options

The response of Turkish public and private sector institutions has developed along three axes since the publication of the proposal.

The first axis is the call for modernisation of the Customs Union. On 21 December 2016 the Commission sought a negotiating mandate from the Council to modernise the Customs Union so as to cover services, public procurement and agriculture; the General Affairs Council refused that mandate on 26 June 2018 and concluded that “no further work in this direction should be planned”. The file returned to the agenda at the meeting between the Commissioner for Enlargement, Marta Kos, and the Minister of Foreign Affairs, Hakan Fidan, on 6 February 2026, at which the parties agreed to continue work towards modernisation. Minister Bolat reported that 15 of the 29 identified problem areas had been resolved. In February 2026 representatives of the Turkish business community likewise called for modernisation expressly to cover services, agriculture, public procurement and e-commerce.

A qualification must be entered here: no primary source establishes a direct legal connection between the February 2026 talks and the Industrial Accelerator Act. The connection is political and inferential rather than documented.

The second axis is sectoral lobbying. The Machinery Manufacturers’ Association (MİB) submitted formal observations to the European Commission requesting that content of Customs Union origin continue to benefit from Union-origin equivalence both in public procurement and in support programmes, and proposing sustainability criteria based on performance rather than on origin. In June 2026 ESİAD, in observations submitted to the Ministry of Trade, sought express recognition of Turkish-origin components within “Made in EU” products under the Customs Union framework, recognition of the electric vehicle supply chain, mutual recognition of Turkish emissions accreditation, and the establishment in Türkiye of industrial manufacturing acceleration areas comparable to those in the Union.

The third axis is the alignment of Türkiye’s own legislation: the movement described above in relation to the Public Procurement Law and the Communiqué on Domestic Goods falls within this head.

In our assessment four options are open to Türkiye, and they are complementary rather than mutually exclusive.

The first is to open accession negotiations to the GPA. That is the most direct legal remedy, resolving the reciprocity objection at its root; but it is also the most costly administratively and politically, and it does not presently appear on the agenda.

The second is a reciprocity-based opening under Law No 4734. That is a narrower and more rapid route, although it is uncertain whether the Union would regard it as a “sufficient opening”.

The third is the modernisation of the Customs Union so as to cover public procurement. That is the most satisfactory solution in legal terms, but it requires the 2018 Council decision to be overcome and is not a process Türkiye can control on its own.

The fourth is direct lobbying in the legislative process to preserve the existing categorical equivalence and to prevent the adoption of the ACEA formula. In the short term this is the most critical of the four, since the Council’s timetable for political agreement falls between September and December 2026.

I. Framing the question

The risk that the Industrial Accelerator Act carries in international trade law is concentrated in the distinction between public procurement and public support schemes. As will be seen, that distinction arises from the text of GATT 1994 itself and has been sharpened by the case law of the Appellate Body.

II. GATT Articles III:4 and III:8(a)

Article III:4 of GATT 1994 requires that imported products be accorded treatment no less favourable than that accorded to like domestic products — the principle of national treatment. Any origin-based preference breaches that provision prima facie.

Article III:8(a) establishes an exception to that rule: laws and regulations governing the procurement by governmental agencies of products purchased for governmental purposes, and not with a view to commercial resale or with a view to use in the production of goods for commercial sale, fall outside the scope of Article III. That is a genuine government procurement exception.

The difficulty lies at the boundary of that exception.

III. The case law in Canada – Renewable Energy and India – Solar Cells

In Canada – Certain Measures Affecting the Renewable Energy Generation Sector / Canada – Measures Relating to the Feed-in Tariff Program (DS412/DS426), the Appellate Body narrowed the scope of the Article III:8(a) exception decisively. The dispute concerned the minimum domestic content requirements imposed in respect of wind and solar generation equipment under the province of Ontario’s feed-in tariff programme. The Panel report was circulated on 19 December 2012, and the Appellate Body report was circulated on 6 May 2013 and adopted on 24 May 2013.

The Appellate Body reversed the Panel and held that the Article III:8(a) exception did not apply. Its reasoning was as follows: the product purchased by the government was electricity, whereas the product discriminated against was generation equipment. Since there was no competitive relationship between those two products, an exception that shields discrimination only in respect of the product purchased could not immunise discrimination as to domestic content at the level of the equipment. The Appellate Body further confirmed that the minimum domestic content levels infringed Article 2.1 of the TRIMs Agreement and Article III:4 of GATT. As to characterisation as a subsidy, although the existence of a financial contribution was accepted, no conclusion could be reached on the element of “benefit” and consequently on Article 3.1(b) of the SCM Agreement.

The same approach was maintained in India – Solar Cells (DS456). The domestic content requirement applied by India to developers selling electricity to the State under the Jawaharlal Nehru National Solar Mission was found to be contrary to Article 2.1 of the TRIMs Agreement and Article III:4 of GATT in the Panel report (24 February 2016) and the Appellate Body report (16 September 2016, adopted 14 October 2016); India’s defence under Article III:8(a) was rejected in line with the Canadian case law.

IV. Applying that case law to the proposal

That case law draws a sharp line between the two limbs of the proposal.

As regards the public procurement limb, the position is comparatively defensible. Where the State itself purchases a “Made in EU” vehicle or low-carbon steel, the product discriminated against and the product purchased are one and the same; the Article III:8(a) exception is operative.

As regards the public support scheme limb, the position is precisely the reverse. When the State grants a purchase incentive for an electric vehicle it does not purchase the product; it subsidises its purchase by a third party. Measured against the “product purchased” test in the Canadian and Indian case law, the Article III:8(a) exception cannot apply here. The provision of the proposal making 100 per cent of national support scheme budgets for vehicles conditional upon origin is therefore directly exposed to a claim of infringement of Article III:4 of GATT.

V. The TRIMs Agreement

Article 2.1 of the Agreement on Trade-Related Investment Measures prohibits investment measures inconsistent with Article III or Article XI of GATT. Paragraph 1(a) of the Illustrative List annexed to the Agreement expressly treats as inconsistent with Article III:4 any measure requiring the purchase or use of products of domestic origin — whether specified in terms of particular products, of volume, of value or of a proportion of production.

The conditionality provision of the proposal relating to support schemes answers precisely to that description in its structure: the beneficiary’s entitlement to the subsidy is made conditional upon a given proportion of its inputs being of Union origin (or treated as equivalent). The structural identity with the type of measure condemned in the Canadian and Indian disputes is plain.

Other disputes point in the same direction (Canada – Autos, DS139/142; Brazil – Taxation, DS472/497; United States – Renewable Energy, DS510), although the content of those rulings should be separately verified against primary sources.

VI. Article 3.1(b) of the SCM Agreement

Article 3.1(b) of the SCM Agreement treats as a prohibited subsidy any subsidy contingent, whether solely or as one of several other conditions, upon the use of domestic over imported goods. That is a regime distinct from, and more severe than, Article III:4, since no commercial injury need be shown: the infringement arises of itself.

Because the Appellate Body was unable to complete its benefit analysis in the Canadian dispute, no binding case law has developed under that provision. That is a genuine lacuna in legal terms and leaves scope for each side to construe the provision in its own favour. The support scheme provision of the proposal, however, is a subsidy rather than a public procurement contract; it is accordingly deprived of the protection of Article III:8(a) and constitutes the most obvious target under Article 3.1(b).

VII. Article IV of the Government Procurement Agreement and the position of non-parties

Article IV of the GPA requires parties to accord national treatment to the goods, services and suppliers of other parties — within the scope of their own schedules of commitments — together with most-favoured-nation treatment as between parties.

The proposal’s adoption of GPA membership as one of the categories of equivalence is a coherent choice as a matter of Union law: equivalence in public procurement is conferred on countries that are in any event subject to a reciprocal obligation of national treatment. That choice, however, creates a sharp exclusionary edge for non-parties. Türkiye stands precisely on that edge: it obtains equivalence of goods origin through the customs-union limb, but because it is not a party to the GPA it cannot benefit from full equivalence of market access.

VIII. Criticism and China’s response

Bruegel’s policy brief on the subject records that the proposal’s linking of subsidies and consumer incentives to content of “Union origin or equivalent” carries a serious risk of legal challenge before the World Trade Organization; that the GATT exceptions are unlikely to justify the exclusion of particular countries; that the dispute settlement mechanism between the EU and China remains fully operative; and that failure to implement an adverse ruling would damage the Union’s standing. The institute recommends that support schemes be founded on the sustainability and resilience criteria of the Net-Zero Industry Act rather than on origin-based rules, and further warns that local content requirements may breach commitments given to allies such as Japan and the United Kingdom. The same institute notes that the 20 per cent manufacturing share objective repeats a target set in 2012 and never met, and that costs in solar energy procurement could double.

On 25 April 2026 China’s Ministry of Commerce submitted formal observations to the Union, and on 27 April 2026 stated publicly that the proposal imposed numerous restrictive conditions on foreign investment and might infringe World Trade Organization principles. The statement characterised the “EU origin” label in procurement and support policies as an investment barrier and as institutional discrimination, and intimated countermeasures should negotiations fail. The Commission spokesperson, Olof Gill, rejected the criticism.

IX. Interim assessment

In our view the position of the proposal under World Trade Organization law may be summarised as follows. The provisions relating to public procurement rest on defensible ground by virtue of the Article III:8(a) exception. The conditionality provisions relating to public support schemes, by contrast — and particularly at the rate of 100 per cent for vehicles — carry a serious risk of infringement under Article III:4 of GATT, under Article 2.1 of the TRIMs Agreement and paragraph 1(a) of the Illustrative List, and under Article 3.1(b) of the SCM Agreement alike. The existence of the third-country equivalence mechanism mitigates but does not eliminate that risk, since equivalence is confined to defined categories of country and is subject to the Commission’s discretionary power of exclusion; that power is itself capable of giving rise to argument under Article I (most-favoured-nation treatment).

The practical consequence of that finding for Türkiye is this: Türkiye’s legal position draws strength not only from Union law but also from the Union’s need to defend the measure before the World Trade Organization. The legal fragility of origin-based exclusion makes a broad scope of equivalence the preferable course for the Union as well.

I. The Council

The Irish Presidency aims to secure political agreement on the Industrial Accelerator Act at the Competitiveness Council of 24 September 2026 and to convert that agreement rapidly into a negotiating mandate at the level of the Permanent Representatives Committee. It is reported that the Presidency has begun work on a complete compromise text, including the most sensitive chapters on “European preference”, but that some regard that timetable as unduly ambitious. On the timetable envisaged, the Competitiveness Council of December 2026 would be the meeting at which agreement with the Parliament is celebrated.

II. The European Parliament

The file is under the joint competence of three committees in the Parliament: Industry, Research and Energy (ITRE, the committee responsible), Internal Market and Consumer Protection (IMCO) and International Trade (INTA). The rapporteurs appointed are Christophe Grudler (Renew Europe, France), Pierre Jouvet (S&D, France) and Anna Cavazzini (Greens/EFA, Germany); the committees examined the file jointly on 2 June 2026.

The division within the Parliament parallels that within the Council. Rapporteur Grudler criticises the geographical scope of the draft as too broad, on the ground that it extends to more than eighty countries through existing free trade agreements; the Swedish Member Jörgen Warborn (EPP), by contrast, warns that restricting World Trade Organization partners would reduce opportunities for growth and employment.

The significance of that division for Türkiye is plain: the narrowing approach advocated by Mr Grudler and ACEA’s “EU 27 plus the United Kingdom” formula point in the same direction. Türkiye’s position within the scope of equivalence is the provision that will come under the greatest pressure at the parliamentary stage.

III. Expected timetable

Final adoption is not expected before the middle of 2027. Since the origin requirements in public procurement are to apply to procedures launched on or after 1 January 2029, the period in which the origin regime will begin in practice to bite in the automotive sector is 2028 to 2030. The increase of the battery component requirement to five components and the entry into force of the 50 per cent threshold for critical components will follow three years later, that is, at the beginning of the 2030s.

Representatives of the Turkish automotive industry likewise expect the detailed negotiations to extend into 2027. That affords Türkiye both a risk and a margin for manoeuvre: the risk is that uncertainty is already affecting investment decisions, as the BYD Manisa case shows; the margin for manoeuvre is that time remains for modernisation of the Customs Union, for a reciprocity-based opening under Law No 4734, and for steps on the GPA.

Primary sources

• European Commission, COM(2026) 100 final — Proposal for a Regulation establishing a framework of measures for the acceleration of industrial capacity and decarbonisation in strategic sectors, 4 March 2026.
• European Commission, SWD(2026) 72 final — Executive Summary of the Impact Assessment.
• European Commission, QANDA_26_516 — Questions and Answers on the Industrial Accelerator Act.
• European Commission, COM(2025) 85 — Communication on the Clean Industrial Deal, 26 February 2025.
• European Commission, COM(2025) 994 — Proposal for a Clean Corporate Vehicles Regulation, 17 December 2025.
• Regulation (EU) 2024/1735 (Net-Zero Industry Act).
• Implementing Regulation (EU) 2024/2754 (definitive countervailing duties on electric vehicles originating in China).
• Regulation (EU) 2022/2560 (Foreign Subsidies Regulation).
• Regulation (EU) 2019/452 (Foreign Direct Investment Screening Regulation).
• Decision No 1/95 of the EC–Turkey Association Council, OJ L 35, 13.2.1996.
• Public Procurement Law No 4734.
• Communiqué on Domestic Goods and the amendment of 7 February 2026.

Dispute settlement reports

• Canada – Certain Measures Affecting the Renewable Energy Generation Sector / Canada – Measures Relating to the Feed-in Tariff Program, DS412/DS426, Panel Report 19.12.2012, Appellate Body Report 06.05.2013 (adopted 24.05.2013).
• India – Certain Measures Relating to Solar Cells and Solar Modules, DS456, Panel Report 24.02.2016, Appellate Body Report 16.09.2016 (adopted 14.10.2016).

Secondary sources and reports

• Draghi, M., The Future of European Competitiveness, European Commission, September 2024.
• Bruegel, The Flaws in the European Union’s Proposed Industrial Accelerator Act and How to Fix Them, 2026.
• ACEA, Position on the Industrial Accelerator Act, July 2026; Made in Europe Content Requirements, September 2025.
• CLEPA, Industrial Accelerator Act Position Paper, April 2026.
• Economic Development Foundation (İKV), Assessment Note on the EU Industrial Accelerator Act and the “Made in EU” Requirements, 22 April 2026.
• European Parliamentary Research Service, Industrial Accelerator Act — EU Legislation in Progress, EPRS_BRI(2026)789300, 21 May 2026.
• European Parliament, EU–Turkey Customs Union and Trade Relations: What Options for the Future?, EXPO_IDA(2021)653640.