Turkish carbon pricing after Climate Law No. 7552: an assessment of the CBAM and the National Emissions Trading System

One of the main risks the European Union (EU) faces as it tightens its climate targets under the European Green Deal is carbon leakage, that is, the relocation of carbon-intensive production to jurisdictions with laxer climate rules (European Commission, 2019). The Carbon Border Adjustment Mechanism (CBAM) is the EU’s answer: importers are charged for the carbon embedded in their goods, so that a tonne produced abroad faces a cost comparable to a tonne produced inside the Union. Financial obligations under the mechanism began on 1 January 2026 with the start of the definitive phase. Small importers, those below an annual threshold of 50 tonnes, were taken out of the system by the 2025 Omnibus amendment [Regulation (EU), 2023a/956; Regulation (EU) 2025/2083] ^[1][2].

Türkiye’s trade position illustrates what is at stake. Roughly half of Turkish goods exports go to the EU market, and Türkiye ranks among the EU’s principal suppliers of iron and steel, aluminium, cement and fertilisers, all of which fall within the initial scope of the CBAM (Çimen, 2024). UNCTAD (2021) estimates that the mechanism could divert trade worth billions of dollars away from developing countries, while the Asian Development Bank (2023) expects marked cost pressures for exporters with high carbon intensity.

In the Turkish literature, the potential impacts of CBAM have been examined descriptively from the perspectives of exports and competitive strength (Saraçoğlu and Kutlu, 2025); it has also been argued that a carbon tax should be adopted as a carbon pricing instrument (Artun, 2024). However, none of these studies analyse the legal relationship between the Climate Law No. 7552, which came into force on 9 July 2025, and the design of the National Emissions Trading System (ETS) it established, on the one hand, and the offset mechanism under Article 9 of the CBAM Regulation, on the other. This study aims to fill this gap.

The research question is this:

The answer is negative. Because the Article 9 deduction is recognised only for the carbon price actually paid, and because any rebate or other compensatory adjustment that reduces that price must be taken into account [Regulation (EU), 2023b/956, Article 9(1)], Turkish installations will pay no effective carbon price on covered emissions during a pilot phase built around 100% free allocation; the current design will therefore fail to fulfil its intended protective function.

The article builds on Boute (2024) analysis of the “carbon price effectively paid” in two respects. Firstly, it applies that framework to a concrete regulatory design in a third country, namely, the free allocation regime of the TR-ETS pilot, which did not yet exist when Boute wrote. Secondly, it links free allocation to the rebate and compensation adjustment in the second sentence of Article 9(1), drawing on the CJEU’s free allocation case law, the parallel wording of the Union Customs Code and WTO subsidy disciplines.

The methodology is doctrinal and comparative. The analysis deliberately concentrates on the Article 9 deduction; questions calling for separate treatment, such as the general WTO compatibility of the CBAM, are touched upon only within the limits of the SCM Agreement’s approach to free allocation. The remainder of the article addresses, in turn, the legal architecture of the CBAM and Article 9 (Section 2); Türkiye’s carbon pricing architecture and the comparative framework (Section 3); the intersection problem, its empirical scale and an indicative burden calculation (Section 4); and deduction-optimised design proposals (Section 5).

The CBAM rests on Regulation (EU), 2023a/956 and initially applies to cement, iron and steel, aluminium, fertilisers, electricity and hydrogen (Article 2 and Annex I). Between 1 October 2023 and 31 December 2025 importers only reported embedded emissions [Regulation (EU), 2023b/1773]; in the definitive phase, which began on 1 January 2026, authorised CBAM declarants must buy and surrender certificates matching the embedded emissions of the goods they import, at a price indexed to EU ETS auctions (Article 20 et seq.). Because the charge tracks the EU carbon price in this way, part of the literature describes the instrument as a tax in disguise (Klotz et al., 2022). Practical guidance on the design of border carbon adjustments had in fact been developed well before the CBAM was enacted (Cosbey et al., 2019).

The CBAM complements the EU ETS. Free allocation granted to EU producers in CBAM sectors is being phased out between 2026 and 2034 through a “CBAM factor” that starts at 2.5% and reaches 100% in 2034 [Regulation (EU), 2023a/956, Article 36; Directive (EU), 2023/959]. The underlying logic is symmetry: to the extent that domestic producers bear no actual cost because of free allocation, importers are not asked to surrender certificates either. How the free allocation adjustment is reflected in detail is governed by Regulation (EU) 2025/2620.

Article 9 of the Regulation is of central importance for third-country producers. It allows an authorised CBAM declarant to claim a reduction in the number of certificates to be surrendered on account of the carbon price paid in the country of origin. The article is blunt on one point: the claim is available only where that price has been effectively paid there [Article 9(1)]. A carbon pricing system on paper counts for nothing; payment does. The declarant must in addition keep records proving payment and have it certified by an independent person [Article 9(2)].

For the argument advanced here, everything turns on the second sentence: any rebate or other form of compensation available in the country of origin must be netted off, because it reduces the carbon price [Article 9(1), second sentence]. Do free allowances fall within that clause? The text leaves the question open. Pursuant to Article 9(4), the rules governing the conversion of the average annual carbon price actually paid into a credit will be established by the Commission’s implementing regulation. The Commission published a draft implementing regulation on this method on May 13, 2026, and the public consultation period closed on June 10, 2026 (European Commission, 2026). Although the regulation had not yet been adopted as of the date of this writing, the draft outlines the method’s main features (for the previous call, see European Commission, 2025).

The second sentence of Article 9(1) carries the doctrinal weight of this study. Read literally, “any rebate or other form of compensation” is an anti-avoidance clause of considerable breadth. Its concern is simple: a carbon price that has nominally been paid should not keep its “effectively paid” label if the payer recovers the money through some other channel. Union customs law has long worked with the same formula. “Price actually paid or payable” in Article 70 of Regulation (EU) No 952/2013 (Union Customs Code) defines value by economic reality, not by form, and there is no obvious reason to read the identical wording in the CBAM differently.

A contextual reading points the same way. Within the EU ETS, free allocation, that is, benchmark-based allocation under Article 10a of Directive 2003/87/EC, allows an installation to receive part of the allowances it needs free of charge instead of buying them on the market (CJEU, 2019). In Arcelor and in Borealis the CJEU stressed that free allocation is not an entitlement-based advantage but an allocation shaped by mathematical and normative criteria (CJEU, 2008; CJEU, 2016). What this line of case law yields is the following: free allocation is an economic compensation that lowers the installation’s actual payment per unit of emissions.

Any allowance handed out free of charge within an ETS narrows the effective base on which an emitter actually pays for carbon. Boute (2024) has developed this point for the CBAM, showing that the protection EU industry enjoys through free allocation blunts the price signal the ETS is meant to send. The present article turns that analysis into a doctrinal test for a specific TR-ETS design, a pilot with 100% free allocation, to which the second sentence of Article 9(1) can be applied directly.

WTO law offers a parallel line of reasoning. Whether free allocation amounts to a financial contribution or a subsidy under the SCM Agreement is contested; yet reading Article 1.1 and Article 3.1(a) of the Agreement together with Annex I shows that international trade law treats domestic measures which compensate for price burdens as a category of their own (WTO, 1994; Espa et al., 2022). That parallel supports a teleological reading of the second sentence of Article 9(1), that is, a reading anchored in economic reality. Border instruments of this kind have also been examined in international tax scholarship through the lens of climate clubs (Pirlot, 2024).

This interpretation is no longer merely a doctrinal inference. The Commission’s draft implementing regulation dated May 2026 explicitly lists emissions associated with free allowances received by operators among the “reimbursements or other forms of compensation” to be taken into account in determining the effective carbon price [European Commission, 2026, Article 8(1)(b)(i); see also Recital 14]. Thus, the economic reality-based interpretation advocated in this study-which draws a parallel between the CJEU’s case law on free allowances and the Customs Code-has found its counterpart in the rules proposed by the Commission. The draft also makes a subtle distinction: while emission reduction subsidies financed by carbon price revenues and meeting certain conditions are excluded from netting [Article 8(2)], free allowances are explicitly included within the scope of netting. If these provisions of the draft are retained in the final text, the debate will shift from the interpretive plane to the plane of positive law.

A counter-argument might insist that a “rebate” presupposes a positive cash transfer, which free allocation does not involve. That view, however, drains the words “any […] other form of compensation” of their content and sits uneasily with the purpose of Article 9, which is to measure the carbon price differential by economic reality. This study therefore prefers the reality-based interpretation on literal, contextual and teleological grounds alike.

In line with its updated Nationally Determined Contribution and the 2053 net-zero target (Republic of Türkiye, 2023), Türkiye has put in place a comprehensive legal framework for climate action through Climate Law No. 7552, published in Official Gazette No. 32951 of 9 July 2025. Article 9 of the Law provides for the establishment of an ETS by the Presidency for Climate Change, the preparation of a national allocation plan and the principles governing allocations; Article 9(6) stipulates that free allowances are to be distributed in proportion to historical emissions data or benchmark values (Climate Law No. 7552, 2025). The market architecture comprises primary and secondary markets, with flexibility tools such as a market stability mechanism, banking and offsetting.

The Draft TR-ETS Regulation, the implementing legislation under the Law, provides for a pilot phase covering 2026 and 2027 and applies to Category B and C installations in CBAM sectors whose annual emissions exceed 50,000 tonnes of CO₂e. Transitional Article 1(4) of the draft states in terms that covered installations will receive 100% free allowances during the pilot phase on the benchmarking method [Draft TR-ETS Regulation, 2025, Transitional Art. 1(4)]. Benchmark values are published in the National Allocation Plan, and advance allowances may be used to launch the pilot. According to the explanatory memorandum, the pilot is meant to ease adaptation to the system and build capacity.

The picture sharpens once Türkiye’s current effective carbon price is considered. The OECD (2026) reports that Türkiye applies no explicit carbon price and that its net effective carbon rate stood at an average of 9.37 euros per tonne in 2023, driven largely by the implicit price created by fuel excise duty. Although List II of Excise Duty Law No. 4760 attaches CO₂ emission thresholds to the taxation of motor vehicles, these thresholds do not function as a general carbon price. Over the same period the EU ETS auction price mostly fluctuated between 60 and 90 euros per tonne of CO₂e (World Bank, 2025). Once the 100% free allocation of the pilot phase is added, the carbon price actually paid by CBAM-covered installations in Türkiye effectively approaches zero.

Comparative experience shows that the free allocation of all allowances was not unusual in the early stages of the development of emissions trading schemes; however, this approach does not fully align with the reform trend currently prevailing in emissions trading schemes. In the EU ETS, auctioning has been adopted as the primary allocation method since 2013; free allocation, meanwhile, has been progressively restricted to the provision of criteria-based support for sectors exposed to the risk of carbon leakage. Similar systems in the UK and New Zealand also combine auctioning with targeted free allocation for emission-intensive and trade-exposed sectors. South Korea, however, offers a more appropriate point of comparison for Türkiye’s proposed system. The K-ETS initially relied almost entirely on free allocation: in Phase One, all allowances were distributed free of charge; with the gradual introduction of the auctioning method, the proportion of free allocation fell to approximately 97% in Phase Two and to 90% in Phase Three (ICAP, 2025). Consequently, the Turkish model – which envisages 100% free allocation based on criteria during the pilot phase – resembles the early development stage of the K-ETS rather than the current models in the EU or Switzerland. The South Korean experience demonstrates that an emissions trading system starting with almost entirely free allocation can, over time, evolve into a structure that places greater emphasis on auctioning. However, this experience also highlights the limitations of such a transition model in the context of the CBAM: allowances obtained free of charge do not, in themselves, constitute an effectively paid carbon price and therefore cannot serve as a basis for a claim for a reduction under Article 9 of the CBAM Regulation unless a genuine carbon cost is borne.

Switzerland repays close attention here. As Holzer (2021) records, Annex II of the proposed CBAM Regulation listed Switzerland among the countries excluded from its obligations. Why? Because the Swiss ETS has been linked to the EU ETS since the start of 2020, and the EU recognised that linking. Linking has a hard consequence: Swiss producers pay the same carbon price as their EU competitors. The exclusion honours that price, not the mere fact that an ETS exists on Swiss soil.

Turned around, the Swiss case yields an a contrario argument for Türkiye. Holzer (2021) himself warns that the CBAM bites hardest where production is carbon-intensive and no effective carbon price is charged. Türkiye has an ETS, so at first sight it escapes that description. Look at the price, though, and the picture changes: with allowances handed out free in full during the pilot, covered installations pay next to nothing, and in economic terms the country slides towards the no-carbon-price group. Switzerland earned its exclusion with an equivalent price. A pilot that produces no such equivalence can expect neither a comparable exclusion nor much of an Article 9 deduction. However, the scope of the Swiss comparison must be clearly defined: Switzerland’s exclusion is based not merely on price equivalence but on full system interconnection, including mutual recognition, and Türkiye is not seeking such a connection during the pilot phase. Consequently, the a contrario conclusion drawn from the Swiss example is limited to the following proposition: the path to both exclusion and a meaningful Article 9 reduction lies not in the existence of the system but in the price actually paid. The primary comparison between unconnected systems, however, is the South Korean example discussed in the previous sub-section.

Set the two frameworks side by side and the problem can be stated in two lines. Article 9 rewards only a carbon price that has actually been paid. The TR-ETS pilot, with its 100% free allocation, sees to it that almost none is. The deduction a Turkish exporter can secure through its CBAM declarant is accordingly minimal. The rebate and compensation clause supplies a second, independent ground for the same conclusion: free allocation compensates the installation for precisely the carbon cost it would otherwise have borne. Indeed, this second basis is now also supported by the Commission’s own draft rules: the netting of free allowances as compensation is explicitly provided for in Article 8(1)(b)(i) of the May 2026 draft (European Commission, 2026).

Data compiled from the TurkStat Foreign Trade Statistics Database convey the economic scale of the issue (TurkStat, 2026; see note 1). Table 1 shows the value and volume of Türkiye’s exports of goods covered by Annex I of the CBAM Regulation to 26 EU Member States over 2023–2025. From 1 January 2026 these flows are subject to direct financial obligations under the full CBAM regime.

Table 1 shows that the issue is anything but marginal: iron and steel together with aluminium account for roughly 5bn euros of exports to the EU in 2025, and the CBAM-sector total reaches 5.56bn euros in that year. This volume carries the direct certificate obligation from 1 January 2026.

Steel, the leading export, lends itself to an indicative calculation (see note 2). Türkiye shipped some 2.69 million tonnes of iron and steel to the EU in 2025 (Table 1). Take a deliberately conservative emission intensity of 1.4 tCO₂e per tonne, a figure reflecting how much of Turkish steelmaking runs on electric arc furnaces, and an indicative EU ETS price of 80 euros. The embedded emissions then come to about 3.8 million tCO₂e, and the potential certificate bill to about 300m euros a year. Under the current TR-ETS design virtually none of this can be deducted: the price actually paid is close to zero, so the burden lands on Turkish products almost in full.

Run the counterfactual and the stakes become visible. Auction allowances for half of those emissions and a real price is paid on every tonne; the Article 9 deduction would then claw back somewhere near half of the potential bill in Türkiye’s favour. At an intensity of 1.9 tCO₂e per tonne, the level primary production can reach, the forgone deduction grows larger still. For one product alone, then, the allocation choice moves hundreds of millions of euros. The effect is anything but marginal.

The EU is meanwhile paring back its own free allowances in CBAM sectors: the CBAM factor opens at 2.5% in 2026 and climbs to 100% by 2034 [Regulation (EU), 2023a/956, Article 36]. EU producers will therefore keep receiving substantial free allocation in the early years, and the asymmetry between the two sides lies in the direction of travel rather than in current levels. But the directions differ sharply. A price is taking shape inside the EU. In Türkiye, no reference price emerges while the pilot lasts.

Türkiye’s Customs Union with the EU (Decision No. 1/95) provides a legal frame, but because the CBAM is not a customs duty it yields no exemption. Moreover, the Article 9 deduction takes legal effect as a reduction of the EU importer’s certificate obligation; whether that advantage is passed through to the Turkish exporter depends on contract terms and market structure, and a frictionless pass-through cannot be assumed. Against this background, the finding of a long-run empirical study of the EU ETS that there is no evidence of carbon leakage in European manufacturing (Naegele and Zaklan, 2019), together with the IMF’s observation that leakage rates differ sharply across countries (Misch and Wingender, 2021), shows how thin the empirical case is for 100% free allocation as an anti-leakage device.

Two objections may be raised. The first is that the pilot phase is temporary and free allocation a legitimate tool for giving industry time to adapt. This does not answer the deduction problem: the pilot years 2026 and 2027 are precisely the years in which CBAM financial obligations begin, so the deduction is at its weakest when protection is needed most. Worse, the carbon cost left uncollected at home is paid into the EU budget as CBAM revenue, whereas auctioning would keep an equivalent amount inside Türkiye. The point matches the literature on the distributional effects of the CBAM on developing-country exports (Gergondet, 2021); Falcão (2021) likewise advises developing countries to adopt designs that retain carbon revenue within their own borders instead of ceding it to the EU.

The second objection is that establishing an ETS is itself a signal of commitment which the Commission might reward. Article 9, however, asks a narrower question: has a price actually been paid? The existence of a system does not answer it, and the Commission enjoys no discretion on the point. A more permissive approach in the forthcoming implementing act is conceivable in theory, but the text of the Regulation sets the outer limit [Regulation (EU), 2023b/956, Article 9(4)].

What follows from the analysis is a design agenda: the TR-ETS should be reviewed with a view to maximising the Article 9 deduction.

Allowances bought at auction produce a carbon price that is actually paid, and they therefore ground the Article 9 deduction; putting even part of covered emissions under auctioning during the pilot would make the deduction operational. The EU ETS is auction-based, and the UK ETS applies a reserve price (ICAP, 2025). For Türkiye the recommendation is to auction at least 15%–20% of allowances in the second pilot year (2027) and to set a reserve price indexed to the EU ETS.

A carbon tax generates a paid carbon price beyond doubt and so reinforces the Article 9 deduction (Artun, 2024). Article 73 of the Constitution frames the design through the ability-to-pay principle: the Constitutional Court has consistently held that a tax must be general, levied according to financial capacity and directed at covering public expenditure, and it has read “financial capacity” as “ability to pay” (Constitutional Court, 1995). Although a carbon tax takes environmental harm as its base, the criterion adopted in the same ruling, namely that a burden is unconstitutional only where it exceeds ability to pay and renders fundamental rights ineffective, can be reconciled with the steering function of environmental taxation (Kılıç, 2018). To avoid overlap with excise duty, the emissions-weighted component of fuel excise should be credited against the new tax. Even if a “similar financial obligation” model were preferred, a constitutional basis exists: the Constitutional Court (2013) has defined the concept as public revenue collected by virtue of public authority and required to be established by law, and the classification of the recovery contribution fee as such an obligation offers a precedent for carbon-based charges (Environmental Protection Act, 2026).

Türkiye’s free allocation phase-out should be synchronised with the EU’s CBAM factor timetable. Reducing allocation in step with the EU would both enlarge the deduction and prepare the ground for future equivalence or linking talks.

Meeting the documentation and verification requirements of Article 9(2) calls for an MRV infrastructure and a record-keeping system capable of proving the carbon price paid. The notion of an independent verifier should be defined in clear terms, and a catalogue of minimum standards for carbon payment documentation should be drawn up.

A pilot built on 100% free allocation could also invite a subsidy dispute under the WTO Agreement on Subsidies and Countervailing Measures, particularly as regards measures supporting export-oriented production (WTO, 1994; Espa et al., 2022). The compatibility of the CBAM itself with Articles I, III and XX of the GATT has been debated at length (Dias et al., 2020; Arko, 2021). For both CBAM alignment and SCM risk, the direction of travel for the Turkish design is the same: towards auctioning.

Climate Law No. 7552 and the national ETS were significant steps. Against Article 9 of the CBAM, this article has argued, they do not yet protect. The flaw sits in the design rather than in the existence of the system: full free allocation during the pilot pushes the carbon price actually paid by covered installations towards zero, and once the “effectively paid” condition and the rebate and compensation adjustment are applied, little remains for Turkish exporters to deduct.

The policy consequence is uncomfortable. In 2026 and 2027, the very years in which protection matters most, Turkish exporters will carry most of the CBAM burden, and the carbon cost that goes uncollected at home will accrue to the EU budget instead. The remedy proposed here, a deduction-optimised design, raises the auction share, adds an explicit carbon tax where needed, aligns the free allocation timetable with the EU’s, and brings MRV and verification standards up to the level Article 9(2) demands; nothing in it conflicts with climate policy, trade policy or constitutional tax law. The May 2026 draft of the implementing regulation for Article 9(4) has already made it clear that free allocation will be offset [European Commission, 2026, Article 8(1)(b)(i)]; once the final text and the rules on the default carbon price are adopted, the conclusion reached in this study will cease to be a forecast and will become a determination of the law in force.

No external funding was received for this study.

This study does not involve human participants or animal subjects; ethical approval was not required.

The foreign trade data analysed in this study are publicly available from the Turkish Statistical Institute (TurkStat) Foreign Trade Statistics Database.