�ԹϺ�����

The European Union's commitment to the fight against climate change is pursued in part through budgetary mainstreaming, which aims to integrate key policy priorities, such as climate, into every stage of the budget process, from design and preparation to implementation and evaluation across all spending programmes. The overarching targets are the allocation of 30 percent of the current Multiannual Financial Framework (MFF) and 37 percent of the Recovery and Resilience Facility (RRF) to investments in climate change mitigation and adaptation. 

Despite its commendable aims, the EU’s mainstreaming framework is marked by design flaws, excessive complexity and fragmented implementation. These weaknesses raise doubts about its effectiveness, risk greenwashing through exaggerated climate benefits, allow environmentally harmful spending and make it impossible to accurately measure the true impact of mainstreaming. These are not minor technicalities but fundamental challenges undermining the EU’s climate ambitions. Substantial reforms in climate mainstreaming are essential to enhance the greening impact of the MFF.

The structure of climate mainstreaming architecture

The climate mainstreaming framework consists of several components (Figure 1). The overarching minimum targets for climate change mitigation and adaptation are broken down into specific programme targets (Figure 2). These targets are supported by expenditure tracking systems. The Climate Adjustment Mechanism at programme level is a tool to monitor the progress toward achieving sectoral spending targets and make corrections if necessary. 

Moreover, as part of its green commitment, the EU has pledged that its activity will do no harm to its climate and environmental goals (Beltrán Miralles et al, 2023), leading to the adoption of the Do No Significant Harm (DNSH) principle, a key pillar of the climate mainstreaming methodology.

Figure 1: The structure of climate mainstreaming in the 2021-2027 MFF

^{Source: Bruegel}

Figure 2: Climate mainstreaming targets by various instruments

Tracking green spending is notoriously difficult

The expenditure tracking methodology assigns one of three coefficients (0, 40 or 100 percent) to categorise each activity funded by the EU (European Commission, 2022a). Projects with a 100 percent weight are expected to play a major role in achieving EU climate goals, either directly through initiatives such as renewable energy, or indirectly via areas such as research and development. A 40 percent coefficient is assigned to projects anticipated to contribute positively but less substantially to climate objectives. Activities assigned a 0 percent weight are judged to have no impact on climate (positive or negative), consistent with the DNSH principle. The coefficients are then used as weights in calculating the share of spending devoted to mitigating and adjusting to climate change (ECA, 2022).

In most cases, activities are not directly assigned a coefficient but are instead grouped under ‘intervention fields’, each of which has a predetermined coefficient. There are currently four lists of intervention fields, which, while significantly overlapping, are not identical ¹ There are three defined lists for programmes, covering the RRF, Cohesion policy (under the Common Provision Regulation) and InvestEU, along with an additional list outlined in the document establishing the Climate Mainstreaming Architecture (Beltrán Miralles et al, 2023). . Several of these intervention fields are quite broad. For example, some RRF intervention fields include “Energy efficiency renovation of existing housing stock, demonstration projects and supporting measures” with a weight of 40 percent and “Reconstructed or modernised railways, Trans- European Transport Network” with a weight of 100 percent (Beltrán Miralles et al, 2023).  

However, the assignment of coefficients is not justified with appropriate documentation, even though the methodology of this assignment is crucial for the effectiveness of the mainstreaming system and so should rely on sound scientific reasoning. Moreover, not all EU budget instruments define intervention fields, giving the overall procedure a lack of transparency. For example, the Common Agricultural Policy (CAP) is classified under just four broad activity types, each assigned either a 40 percent or 100 percent weight ² The weights for the four CAP activities are: “(a) 40 percent for the expenditure under the basic income support and the complementary income support (…); (b) 100 % for expenditure under the eco-schemes (…); (c) 100 % for expenditure for the interventions referred to in Article 93(1) other than those referred to in point (d) of this paragraph; (d) 40 % for expenditure for natural or other area-specific constraints (…).” See Article 100 of Regulation (EU) 2021/2115 . Some interviewees of Begg et al (2025) described these generous CAP climate weights as resembling greenwashing. This is a significant problem, given that CAP accounted for a half of the reported climate spending in the 2014-2020 MFF.

The climate mainstreaming target of the 2014-2020 MFF was 20 percent, and the European Commission (2021) reported that it was marginally exceeded (20.1 percent). However, the European Court of Auditors (ECA) (2022) challenged this finding, identifying activities assigned 100 percent despite lacking a direct connection to climate goals. Additionally, they noted discrepancies in how similar projects were assessed across different programmes. After accounting for these issues, the ECA concluded that the climate-relevant share of the budget reached only 13 percent, rather than 20.1 percent, of the total.

Over-reporting climate spending is not specific to the previous MFF but appears to be an inherent issue with the tagging system. The coefficients used by the Commission are essentially qualitative indicators supplemented by quantitative weights to approximate the overall volume of green investments. Interviews revealed that the practical application of these coefficients often depends heavily on the subjective judgment of national coordinators responsible for implementation, including the challenging process of assigning various projects to specific intervention fields (Begg et al, 2025).

A further challenge of the tagging system is that the assignment of coefficients to intervention fields is based on the expected effects, not the actual progress achieved towards those objectives. The actual and expected effects may differ, but there is no estimate of the impact of climate expenditure on actual climate objectives (Nesbit et al, 2021). 

An alternative system would not only tag investments based on expected impact but factor in results. The national coordinators we interviewed as part of Begg et al (2025) expressed differing views on the merits of results reporting. Some emphasised the administrative burden, arguing that outcomes should be ensured through proper ex-ante measures. Others suggested that ex-post evaluation would be more relevant, as many projects function within broader systems designed to achieve specific objectives. To support this approach, EU countries should be incentivised to develop such systems through some form of ex-post reporting requirements.

The DNSH principle: harm may still slip through

Another major challenge arises with the application of the DNSH principle, which lacks a uniform framework, with considerable variance in implementation across programmes (CEE Bankwatch Network, 2024; Beltrán Miralles et al, 2023). This includes different types of simplified evaluation mechanisms which exist under the RRF, InvestEU and EU taxonomy, as well as programme-specific conditions that supplement the DNSH principle, which are not fully clear. 

There is also an issue with compliance assessments across EU countries. The ECA (2024) assessed 24 measures across four countries and concluded that while some always conduct a substantive DNSH assessment, others consistently opt for a simplified approach, even when the measures in question pose a risk to certain climatic or environmental goals. 

However, both types of assessment are flawed. CEE Bankwatch Network (2024) analysed the practical implementation of the DNSH principle in the RRF using six case studies. In some cases, the simplified procedure relied solely on uncritically accepting the countries' declarations. Meanwhile, assessments conducted under the standard procedure are neither publicly accessible nor consistently rigorous, often resulting in superficial or short-sighted evaluations. 

The complexity of the practical implementation of the DNSH rule creates a lack of transparency surrounding the mechanisms the Commission has put in place to guarantee that every intervention funded by the EU is fully compliant, which makes it difficult to assess the true effectiveness and enforcement of the DNSH principle. Additionally, the legal basis for its application is not always clear.

The lack of transparency is worsened by the Commission’s failure to report projects with adverse environmental impacts, despite the European Court of Auditors’ (2022) recommendation. The European Commission (2022b) has rejected the recommendation, insisting that all programmes comply with the DNSH principle. 

The assumption that no EU projects harm environmental goals is demonstrably incorrect. It is possible to track biodiversity projects which are harmful – biodiversity is one of the areas to which the DNSH principle is applied. The French government’s environmental protection body, L’Inspection générale de l’Environnement et du Développement durable (2023) reported that around €6.5 billion of EU funding has been channelled to activities harmful to biodiversity as part of the EU’s CAP. Similarly, the World Wide Fund for Nature (2024) estimated that at least 58-60 percent of CAP spending, about €32 billion annually, is harmful to biodiversity. These activities directly contradict the DNSH principle. 

An interviewee of Begg et al (2025) suggested that part of the problem is that the taxonomy's definition of significant harm is not a good guide.  The taxonomy’s definition was originally created for a different purpose, tagging green private investments (Beltrán Miralles et al, 2023), but since then, numerous regulations applied this principle to EU spending, sometimes under changed definitions. When an investment project encompasses various economic and other activities, it is difficult to apply the taxonomy definition. 

Finally, certain interventions are exceptions to the DNSH principle (Beltrán Miralles et al, 2023). For example, under the RRF regulation, energy infrastructure investments intended to improve security of supply can be funded without undergoing DNSH assessment if certain additional conditions are met, whilst under InvestEU, operations below €10 million are not evaluated. Further scrutiny would be needed to assess the potential environmentally adverse impacts of these exceptions and the trade-offs they create.

Recommendations for a greener MFF

While the EU's commitment to allocate 30 percent of its MFF and 37 percent of the RRF to climate is commendable, ensuring quality spending is critical. The overall success of the mainstreaming framework depends on the accuracy of climate-related investment tracking across programmes and countries, and the extent to which this spending is effectively linked to measurable climate outcomes.

The existing framework for climate mainstreaming faces significant challenges. The architecture is overly complex, as various regulations set corresponding rules which are not properly harmonised. Special provisions undermine its coherence, while the impacts of exceptions are not analysed. Furthermore, poor documentation makes the architecture hard to understand and apply.  Some of the Commission’s methodologies are undisclosed, while the application of climate mainstreaming remains inconsistent, with varying interpretations and implementations across programmes and countries further hindering its effectiveness and uniformity. 

Intervention fields are defined too broadly – potentially including activities with varied climate impacts – and are applied inconsistently due to system complexity. Moreover, it is often unclear how the coefficients of these intervention fields were established in the first place, implying that overall estimates of climate spending in the EU budget are questionable. There is a risk of exaggerating positive climate impacts while overlooking harmful projects.  These issues highlight the need for significant reform to the EU’s climate mainstreaming framework.

Our proposed improvements include:

1. Consolidating the mainstreaming architecture – including the DNSH principle – into a single legal act with a unified list of intervention fields applied to all EU investment, to eliminate inconsistencies across programmes.
2. Narrowing the scope of intervention fields, increasing their number and defining specific fields for the CAP to replace the current categories.
3. Developing a clear, publicly available, scientifically grounded methodology for assigning climate coefficients to intervention fields, potentially by bodies external to the Commission.
4. Introducing a legal requirement to report on projects harmful to biodiversity and the climate, ensuring transparency on the climate impact of EU spending. The Commission could refer to the French tagging system (see Nesbit et al, 2021).
5. Amending the three-tier coefficient system (0, 40 and 100 percent) to a more granular scale (eg 0, 25, 50, 75 and 100 percent) to narrow broad gaps between categories, with the addition of negative scores ³ The European Commission rates country-specific recommendations in European Semester on a five-level scale: no, limited, some, substantial, or full progress (0, 25, 50, 75 and 100 percent). .
6. Complementing the tagging of expected impact with ex-post analysis of actual achievements, with the Commission reporting estimates of green spending impact.

7. Minimising the disparity between countries conducting substantive DNSH assessments and those using a simplified approach. Substantive assessments should be consistently rigorous, whilst simplified assessments should be subject to proper scrutiny.

This analysis has been produced with financial support from the European Climate Foundation. The analysis builds on research we conducted as part of Begg et al (2025) for which we interviewed several representatives of EU countries, the European Commission and the European Court of Auditors.

References

Begg, I., F. Corti, A. Liscai, Z. Darvas, M. Krystyanczuk, K. Sekut, A. Fiore, J. Bachtler, S. Kah, C. Mendez, O. Van Der Valk (2025) Performance and mainstreaming framework for the EU budget - Empirical evidence, analysis and recommendations, Study requested by the Budget Committee of the European Parliament, Policy Department for Budgetary Affairs, Directorate-General for Internal Policies, PE 767.500, available at  

Beltrán Miralles M. M., T. Gourdon, I. Seigneur,  M. Arranz Padilla and  N. Pickard Garcia (2023) ‘The implementation of the 'Do No Significant Harm' principle in selected EU instruments’, JRC Science for Policy Report, Joint Research Centre of the European Commission, available at

CEE Bankwatch Network (2024) From theory to practice: A case-based analysis of the EU’s ‘do no significant harm’ principle, Central and Eastern European Bankwatch Network, available at

ECA (2022) 'Climate Spending in the 2014-2020 EU Budget- Not as high as reported', Special Report 09/22, European Court of Auditors, available at  

ECA (2024) ' Green transition – Unclear contribution from the Recovery and Resilience Facility', Special Report 14/2024, European Court of Auditors, available at ;

European Commission (2020) ‘Regulation - 2020/852’, available at ;

——— (2021) ‘Annual management and performance report 2020’, available at  

——— (2022a) ‘Climate Mainstreaming Architecture in the 2021-2027 Multiannual Financial Framework’, Commission Staff Working Document, SWD(2022) 225 final, available at,

——— (2022b) Replies of the European Commission to the European Court of Auditors’ Special Report on Climate Spending in The 2014-2020 EU Budget, European Commission, available at

Nesbit, M., T. Stainforth, M.  Kettunen, and E. Blot (2021) Review of approaches to tracking climate expenditure: A report for the National Audit Office of Finland, Institute for European Environmental Policy (IEEP) available at

L’Inspection générale de l’Environnement et du Développement durable (2023) Le financement de la stratégie nationale pour la biodiversité (SNB) pour 2030, available at

WWF (2024) Can Your Money Do Better? Redirecting Harmful Subsidies to Foster Nature & Climate Resilience, World Wide Fund For Nature, available at