We start with discussing the net-zero goals that entities have set. Net-zero targets represent a pivotal element of corporate climate strategy, signalling long-term commitments to GHG emissions in line with global climate objectives. Three companies did not commit to a net zero target. Most entities have set their net-zero goals for the year 2050. However, a significant portion of the sample has adopted more ambitious timelines, with target years ranging from 2045 to as early as 2038. Among these, the year 2040 emerges as a particularly common choice, indicating a growing trend toward accelerated decarbonization. On the other hand, two companies that stated to have an overall net zero target subsequently caveated this by formulating reduction goals that only comprise certain categories of Scope 3 emissions or the Scope 3 emissions for only part of their activities. The variations in target years are detailed in Table 3, which provides an overview of the net-zero timelines across the sample. Importantly, all entities explicitly state that their net-zero goals are aligned with the Paris Agreement.

It should be noted that the majority of companies report only absolute net zero emission targets whereas three companies use only carbon intensity or relative net zero targets²² and nine companies use a combination of absolute and relative net zero targets (e.g., per megajoules of energy produced). While relative goals (like CO2eq per passenger-kilometre for airlines) can be useful for benchmarking within industries, they complicate cross-sector comparisons. Moreover, ESRS E1-4 paragraph 34(a) explicitly requires entities to disclose GHG emission reduction targets in absolute terms, either in tonnes of CO2 equivalent or as a percentage reduction from a base year, and where relevant, in intensity terms.

All entities in the sample except two have also set intermediate emission reduction goals, which serve as important milestones for tracking progress. The majority of these targets are set for the year 2030, with a few exceptions extending to 2032, 2033, and 2035. While net-zero goals tend to be comparable due to their alignment with the Paris Agreement, intermediate goals are more difficult to compare across entities as they tend to be entity-specific, covering only some scopes or combinations of scopes, some categories and some activities or some parts of the emissions. Table 4 summarizes the findings in respect of absolute and intensity intermediate goals. Interestingly, some companies use a lot of intermediate goals (up to seven).

Most entities have set combined intermediate emission targets for Scope 1 and 2 emissions, while also defining separate goals for Scope 3. Specifically, 86% of the entities disclose intermediate Scope 1 & 2 targets, 60% report intermediate Scope 3 targets and 34% disclose intermediate targets for the total of Scope 1, 2 and 3. Nearly all of these goals are expressed as gross reductions, rather than net values. However, the magnitude of these reduction targets varies significantly across entities. As shown in Figure 1, the average reduction target for Scope 1 and 2 by 2030 is approximately 50%, while Scope 3 targets average around 35%. The boxplots illustrate the substantial variance in ambition levels, which we attribute to the large differences in emission reduction goals across entities. This variation likely reflects underlying differences in industry context, operational boundaries, and strategic priorities.

To benchmark progress toward their emission reduction targets, entities establish a base year, which serves as the reference point from which reductions are measured. These base years are critical for assessing the trajectory of climate goals and assessing whether entities are on track in meeting their targets. What stands out in this research is the considerable heterogeneity in the selection of base years across the sample. As illustrated in Figure 2, base years range from 2015 to 2023, with 2018 and 2019 emerging as the most frequently chosen. Two companies use two different base years for different targets. This considerable heterogeneity in chosen base years hinders the comparison between companies in analysing the effectiveness and magnitude of their emission reduction efforts to attain (intermediate) Paris targets.

Figure 3 illustrates a strong example of transparent GHG emission reporting, as presented by Eni. The figure begins with the disclosure of the baseline year, including the total GHG emissions recorded in that year. This establishes a clear reference point for evaluating progress. Eni then presents the results for 2024, showing both the absolute emissions and the reduction achieved relative to the baseline. This is visually juxtaposed with the projected pathway toward carbon neutrality, enabling readers to quickly assess whether the company is on track to meet its long-term climate goals. Importantly, Eni has applied this structure to each of the four climate-related goals it has set, offering a consistent and comprehensive view of its progress. Eni also discloses the GHG boundary used. The GHG boundary defines which emissions are included in an entity’s reporting scope and which emissions might be considered direct or indirect. Common boundaries used are ‘equity share’, ‘operational control’ and ‘financial control’.²³

Given the different potential outcomes of these boundaries, it is important that stakeholders are aware of the approach chosen.

Once companies identify emission reduction as a strategic priority, often linked to long-term value creation or regulatory compliance, they typically establish specific targets, which then guide the selection of measures and tools to achieve those goals. This subsection explores the measures taken to achieve these reductions, including absolute emission cuts, GHG removals and storage²⁴, and the use of verified carbon credits.

While most entities focus on operational reductions and offsetting strategies to meet their climate targets, the use of GHG removal and storage remains limited across the sample. Only 14% of the entities report employing GHG removal and storage as part of their emission reduction strategy. Given the central role that removals and storage are expected to play in achieving net-zero emissions, particularly in the final stretch toward meeting the 2050 Paris-aligned climate objectives, we would have anticipated a higher adoption rate. The current low uptake suggests that the path to 2050 may be challenging, as reliance on GHG removals is likely to increase significantly in the coming years. Moreover, when these measures are used, the reported volumes are relatively small, typically below 0.05 MtCO2eq, indicating that removals currently play a minor role in overall reduction efforts. However, in all cases where GHG removals are disclosed, the reductions are certified by third-party organizations.²⁵

Compared to GHG removal and storage, carbon credits are more commonly used by entities in the sample to offset emissions. Specifically, 43% of the entities report using carbon credits as part of their climate strategy. However, the magnitude of these credits is generally limited relative to their total emissions, indicating that credits play a supplementary rather than central role in current reduction efforts. The carbon credits are also validated by third parties. It is noteworthy that most entities have set gross reduction goals for 2030, meaning that carbon credits do not contribute to achieving these intermediate targets. Instead, credits are often positioned as a future tool to help meet net-zero goals, particularly for hard-to-abate emissions. Among entities not currently using carbon credits, a majority disclose intentions to adopt them in the near future, suggesting a growing reliance on market-based mechanisms as net-zero deadlines approach.

Figure 4 showcases Bayer’s approach to disclosing GHG removals and mitigation projects, aligning with the ESRSs. A key strength of Bayer’s reporting is the clear labelling of disclosures according to the relevant ESRS standard. In the context of GHG removals and mitigation, Bayer provides detailed data on the voluntary carbon credits it has purchased. Notably, the report includes comparative figures from the previous year, offering a valuable reference point for assessing progress. Bayer further specifies that all purchased credits originate from nature-based solutions. The report also clarifies that these projects are located outside the European Union, placing them outside the scope of compliance-based cap-and-trade schemes and confirming their classification as voluntary credits. To ensure credibility, Bayer discloses that all credits are verified by independent third-party standards.

As discussed before, of the 35 companies only one has no carbon reduction targets at all whereas two more have no net zero goal. Intermediate targets have been committed to by 33 of the 35 companies. Having concrete targets, particularly if they are SBTi approved, is a good step. However, meeting the target is more important. This raises the question whether these companies are on track to meet the targets they committed to. The answer to this question is not simple as the pathway to net zero (or even intermediate targets) may not be a smooth glidepath between a baseline emission in the base year to the 2024 actual emissions and then to a target emission in the target year. This is illustrated by Holcim when they explain that they expect the emissions to even increase till 2033 before they will decrease. However, one would expect the emissions relevant to the targets to reduce and to reduce directionally to the target in the target year.

In order to provide some indication of whether companies are on track we have looked at the targets as defined by the companies themselves, split between (i) intermediate and (ii) ultimate net zero targets and finally we added (iii) the goal to reduce total Scope 1, 2 and 3 emissions to net zero (allowing for a maximum of 10% offsetting of the residual emissions). The latter analysis allows us to apply a uniform method to assess whether a company is on track to meet the absolute net zero target by 2050 and is necessary because companies have used a range of different definitions of net zero targets (see end of section 4 and previous subsections). In each of these three cases of defined targets we compared the baseline amount, the amount in 2024 and the target amount, assuming a linear relationship over time. This allows us to assess whether the actual amount in 2024 is below or above the target line that can be drawn between the baseline amount in the base year and the target amount in the target year. We acknowledge that this is not a perfect method to determine whether a company is on track to meet its targets, but at least it provides some indication. In total we assessed this for each of the 102 intermediate and 32 net zero ultimate targets as defined by companies themselves as well as for the total Scope 1, 2 and 3 emissions of each of the 35 companies. This adds up to 169 analyses. However, not all companies provide the necessary information to make this assessment. Table 5 provides a summary of findings.

The table paints a diverse picture. However, some conclusions can be drawn.

In addition to voluntary measures, many entities are subject to regulatory frameworks such as the EUETS or other mandatory cap-and-trade schemes. This subsection evaluates whether entities participate in such schemes and the proportion of their Scope 1 emissions that fall under these regulatory caps.

The extent to which entities’ Scope 1 emissions fall under the EUETS varies significantly across the sample. As shown in Table 6, there is considerable heterogeneity. For some entities, nearly all Scope 1 emissions are covered by the EUETS, while for others, coverage is as low as 5%. Based on the underlying data used to construct Table 6, we find that, on average, approximately 60% of Scope 1 emissions is subject to the EUETS. This variation reflects differences in industry exposure, operational geography, and regulatory scope.

In addition to the EUETS, several entities are subject to other national or regional emissions trading schemes. Within the sample, entities report participation in ETSs from the United Kingdom, Switzerland, Canada, Brazil, Mexico, China, and South Korea. However, while EUETS participation is typically disclosed in detail, often within the emissions table per scope as required by ESRS E1-6, disclosures related to other ETSs are generally sparse or absent. There is clear room for improvement in this area, as comprehensive disclosure of all ETS-related emissions would enhance the credibility and completeness of sustainability reporting including the financial risks involved.

Figure 5 presents an example from Holcim’s integrated annual report, demonstrating best practice in disclosing exposure to emissions trading schemes. Holcim provides a clear graphical breakdown of its relative CO2 emissions, distinguishing between emissions covered under various ETSs and those that remain uncovered. This visual segmentation allows readers to quickly understand the extent to which Holcim’s emissions are regulated and the jurisdictions in which these regulations apply. The figure further enhances transparency by quantifying Holcim’s exposure to each ETS in relative CO2 terms. This enables stakeholders to assess the regulatory landscape the entity operates within and to evaluate potential financial and operational risks associated with carbon pricing.

This subsection investigates how entities account for ETS-related assets and liabilities, including the measurement of granted and purchased emission rights, the presentation of these rights in financial statements, and the recognition of liabilities when emissions exceed allocated rights. The accounting treatment is, in principle, influenced by the entity’s role within the ETS, whether as a regulated emitter, a provider of emission reductions, or a market participant, and by the specific rights and obligations assigned under the scheme.

Entities participating in the emissions trading schemes may receive emission allowances free of charge. Under the EUETS, this free allocation is primarily designed to prevent carbon leakage and support sectors that are energy-intensive or exposed to international competition. The system operates on a cap-and-trade basis, where a cap is set on the total amount of greenhouse gases that can be emitted by covered entities. Over time, this cap is gradually reduced to drive overall emissions down. Correspondingly, the number of allowances allocated for free also decreases, encouraging companies to invest in low-carbon technologies and improve energy efficiency.

The accounting treatment of these allowances under IFRS remains unsettled due to the absence of a dedicated standard.²⁷ Big Four accounting firms note that entities typically choose between accounting for allowances as intangible assets under IAS 38 or as inventories under IAS 2, with allowances received for free often measured at either fair value or a nominal amount, depending on the chosen accounting policy and interpretation of IAS 20 on government grants (PwC 2021; EY 2022; KPMG 2025). However, diversity in practice exists, particularly regarding subsequent measurement and recognition of liabilities when emissions exceed allowances.

If the emission rights are held by a broker/trader, we expect these to be presented as inventory with measurement at cost or fair value less costs to sell, with realised and unrealised changes in value recognised in P&L. If the rights are not held for trading, i.e., no broker/dealer, we find these rights to be presented as inventory, intangible asset or other assets and measured applying either the government grant model or net liability approach.

The government grant model recognises emission rights acquired for free at initial fair value with a corresponding liability to the same amount. Emission rights acquired in the market are initially measured at cost. Subsequent measurement is at cost (the revaluation model may be applied if emission rights are presented as intangibles, but we have not observed this policy in our sample and we expect this to be rare in practice).

The net liability approach recognises emission rights at cost (so zero for emission rights received for free) and only recognises a liability if emissions exceed emission rights received for free. In that case the liability is measured at the carrying amount of the emission rights in stock or at fair value to the extent the company does not possess the emission rights yet.

In ESMA (2024a), the ESMA does not prescribe a specific accounting methodology for the recognition and measurement of emission allowances received under the EUETS. In its public statements, ESMA encourages entities to develop accounting policies that are consistent with existing IFRS standards and tailored to the specific features of the carbon pricing scheme in which they participate. Entities are expected to clearly disclose the accounting treatment applied to emission allowances, including how assets and liabilities are recognized, measured, and presented in the financial statements.

The analysis of the financial statements reveals that 34% of the entities do not disclose their accounting policy related to emission rights. Given that the financial impact of emission rights can be material, especially in the near future, this lack of disclosure suggests that there is significant room for improvement in the quality and consistency of reporting practices.

With respect to allowances received for free under the EUETS, there is limited diversity in accounting treatment. Of the 21 entities that disclose their accounting policies, 19 recognize the allowances received for free at nil value upon initial recognition. Only two entities adopt the government grant approach, recognizing the allowances initially at fair value based on prevailing market prices.

With respect to the accounting treatment of purchased emission rights, all entities recognize these at cost upon initial recognition. However, there is considerable variation in how these instruments are classified in the statement of financial position, as can be observed in Table 7. One-third of the entities classify emission rights as inventory, consistent with the inventory model. A larger portion, 43%, classifies them as intangible assets, following the intangible asset model. The remaining 24% categorizes the emission rights under other assets, which is often due to the perceived immateriality of these rights in the context of the overall financial statements.

Only 51% of the entities discloses the impact of emission rights on their liabilities. This relatively low percentage is notable given that any entity exceeding its allocated emission rights is required to purchase additional allowances on the market, thereby incurring a liability. As such, we would have expected a higher proportion of entities to reflect this potential obligation in their disclosures. Apparently, companies either reduce their emissions sufficiently in order to avoid a shortage, or they buy additional credits in the market prior to the balance sheet date. Among those that do, all indicate that any liability arising from exceeding the emission cap will be measured at market value. Here too, classification practices vary, 78% of the entities classify such liabilities as provisions, while the remaining 22% report them as other or current liabilities.

These findings suggest that the comparability of financial statements is currently impaired by variations in classification and disclosure practices. While the amounts involved may not yet be material under the cost model, many entities expect the volume of purchased emission rights to increase in the future. As such, the financial significance of these instruments is likely to grow. To enhance comparability and transparency, it may be appropriate for standard setters to consider mandating a standardized classification framework for both emission rights and related liabilities.

Figure 6 highlights TotalEnergies’ approach to disclosing accounting policies related to instruments derived from the EUETS. Despite its brevity, the policy effectively communicates all essential elements of how TotalEnergies accounts for ETS-related instruments. The disclosure covers the treatment of emission allowances received free of charge, as well as those purchased on the market. It explains how these rights are presented in the financial statements, the conditions under which provisions are recognized, and the measurement used. Furthermore, TotalEnergies distinguishes between standard transactions and forward contracts, noting that these are accounted for differently, an important nuance for understanding the financial implications of carbon trading activities.

A key observation in current reporting practices is the limited integration between sustainability disclosures and the financial statements. In many cases, emissions allowances are only briefly mentioned in the accounting policies, with no further elaboration or quantitative disclosure in the notes to the financial statements. This lack of transparency hinders stakeholders’ ability to assess the financial implications of a company’s environmental impact. An exception to this observation is Repsol, which provides a more integrated approach. As illustrated in Figure 7, Repsol discloses a detailed movement schedule of its emissions allowances within the financial statements. This disclosure creates a tangible link between the financial and sustainability reports, enabling users to reconcile the financial treatment of emissions allowances with the entity’s reported GHG emissions. We recommend that entities adopt similar practices by including movement schedules for emissions allowances in their financial statements.

The quality of information is enhanced by third party assurance of that same information. In this section we discuss the assurance provided, if any, on the relevant sections of the sustainability report and the financial statements of the companies surveyed. As mentioned before, the CSRD requires the sustainability section of the management report to receive (at least) limited assurance from an external party.²⁸ Limited assurance is a lower level of assurance than reasonable assurance, which is required for annual financial statements of listed entities. Limited assurance is often referred to as a review whereas reasonable assurance is referred to as an audit.²⁹ The CSRD expects the assurance to be provided by the statutory auditor of the financial statements that accompany the sustainability report. However, Member States may allow a statutory auditor or an audit firm other than the one carrying out the statutory audit of the financial statements to perform the review of the sustainability report. Member States may also allow an independent assurance services provider (IASP) established in their territory to perform the review, provided such IASP is subject to equivalent requirements as statutory auditors.

Although this is the first time such requirement is introduced at an EU level, many companies already had certain sustainability information, amongst which emission data, reviewed or audited by an external party. This was done either voluntarily or due to national regulation.

All companies in our research received an unqualified opinion on the 2024 financial statements, which means that reasonable assurance was provided by the auditor on the information in the financial statements, including information about the emission rights and any other information in the financial statements referred to in this article.

In Table 8 an overview is provided of the level of assurance provided on the sustainability report of the surveyed companies, including the emission data. In all cases the assurance is provided by the statutory auditor of the financial statements.³⁰ This shows that although in some countries it is allowed to ask an external party that is not the statutory auditor of the financial statements to perform the limited assurance on the sustainability report, apparently this option is not used by the large, listed entities in our sample. Possible explanations are the fact that the statutory auditor of the financial statements is already familiar with the entity, allowing for lower fees compared to a party that has to start from scratch. Another reason could be that the statutory auditor already has to perform a number of procedures to make sure the management report, including the sustainability report, does not contain errors or inconsistencies with the financial statements. So, hiring another party to review the sustainability report inevitably leads to duplication of effort.

Most companies (21) applied the CSRD/ESRS framework and received limited assurance on the whole sustainability report. Seven companies that applied the CSRD/ESRS framework went beyond the minimum requirements and asked for reasonable assurance on selected disclosures. And five companies that were not subject to the CSRD (yet) and applying various reporting frameworks asked for limited assurance on selected sustainability information. To assess what this means for the assurance provided on the emission data referred to in this article we looked at the selected datapoints that received additional assurance to understand what level of assurance the emission data of the surveyed companies received. Table 9 provides a summary of the assurance provided for Scope 1, 2 and 3 emission data of the surveyed companies. In general, the emission data did receive assurance, be it limited assurance. Assurance on Scope 3 emissions is only slightly less than it is for Scope 1 and 2, despite the fact that Scope 3 emission disclosures are harder to collect and generally less reliable and more subjective, so harder to provide assurance on.

All audit reports on financial statements of the surveyed companies were unqualified and there were no ‘emphasis of matter’ (EOM) paragraphs³¹ on emissions or emission rights. However, 21 audit reports contained a total of 31 ‘key audit matter’ (KAM) paragraphs relating to the topics addressed in this article, i.e. carbon reduction commitments, carbon reduction requirements and/or emission rights. Table 10 provides the number of KAM’s per audit report and Table 11 provides an overview of the topics addressed in the KAMs.

Not surprisingly, the item addressed most is the impact of the energy transition including the related carbon commitments and carbon requirements on the impairment testing of non-current assets including goodwill, intangible assets, property plant and equipment as well as equity-accounting investments.

The findings show that 60% of the auditors included a KAM specifically addressing emissions and emissions rights. This is high but should not be a surprise given that the population covers companies with high emissions of greenhouse gases. This has led to sometimes elaborate explanations in the audit report on the risks involved and the process the auditor has gone through to assess the risks and the acceptability of the judgements and estimates applied by management. We want to mention in particular the KAMs of auditors that focus the reader’s attention to the judgements of management deviating from scenarios developed by international organisations and how the auditor got comfortable that this would not lead to a qualification. For example, the auditor of Repsol highlights that management used scenarios that are not consistent with the ‘Fit for 55’ package of measures announced by the European Commission and Repower EU Roadmap.³² The most elaborate discussion of the topic was found in the audit report of OMV where the auditor included five KAM’s covering 7 pages all relating to the impact of energy transition and carbon commitments.

All assurance reports on sustainability reports were unqualified. However, several auditors have raised the attention to specific elements in the sustainability reports. International Standard on Assurance Engagements 3000, which was generally used for limited assurance engagements, provides the possibility to include an EOM paragraph in the assurance report. We found EOM’s in nine assurance reports, eight of which were of French companies or of companies with a French background³³.

The topics mentioned in the EOMs include, amongst others: