Case C-616/17 Blaise and Others : The precautionary principle and its role in judicial review – Glyphosate and the regulatory framework for pesticides

A. The precautionary principle and the limits of validity review of legislative measures

First of all, one should note that Blaise is extraordinary in the sense that it is one of the rare occasions in which a citizen managed to challenge the validity of EU secondary legislation. This review was only made possible by using the vehicle of the preliminary ruling procedure, due to the limited standing in the act of annulment procedure. ⁴⁶ In this regard, the Blaise case is an example of the fact that in order to challenge the validity of European law, individuals – due to the strict standing requirements – have to essentially breach the law in order to get a chance to challenge the legal framework, ⁴⁷ as was also criticized by AG Jacobs in the UPA case. ⁴⁸

Even though this might be welcomed by advocates of enhanced access to judicial review in front of the CJEU, the Court’s ruling in Blaise has been criticized as legalistic and too deferential. ⁴⁹ In addition to the wording and scope of the referred questions, this has two contributing factors: the margin of discretion of the legislator with the resulting limited scope of judicial review; and the (limited) role of the precautionary principle in judicial review.

Generally, the Court has been known for its deferential approach to review in the face of discretion, whereas one can draw a distinction between ‘political’ and ‘technical’ discretion. ⁵⁰ While political discretion concerns the freedom of the legislator to make political choices weighting diverse interests, technical discretion concerns the wide margin of appreciation granted to administrative bodies where they have to take decisions in the face of complexity. In the Blaise judgment, the deferential approach in the face of political discretion finds expression in para. 50: ‘(…) in view of the need to strike a balance between several objectives and principles, and of the complexity of the relevant criteria, judicial review by the Court must necessarily limited to whether the EU legislature, in adopting Regulation No 1107/2009, committed a manifest error of assessment.’

Although the questions referred by the French court concerned the PPP Regulation as such, the elephant in the room was of course the glyphosate renewal. The Court, as well as AG Sharpston, however stressed, that a finding of non-compliance of the Regulation with the precautionary principle could not be based – solely – on the circumstances of a particular case, such as the glyphosate approval procedure. ⁵¹ This is based on the principle that ‘the validity of a provision of EU law is to be assessed according to the characteristics of those provisions themselves and cannot depend on the particular circumstances of a given case’. ⁵² Thus, the Court was prevented from examining how the Regulation was applied in practice in the context of the glyphosate renewal, and whether this renewal is in accordance with the precautionary principle. ⁵³ This, however, has to be nuanced, as was also clarified by AG Sharpston stating that the validity of the Regulation could be questioned where ‘concerns relating to glyphosate are shown to be representative of a systemic and fundamental failure undermining the PPP Regulation’. ⁵⁴ Nonetheless, this was not argued in the questions referred to the Court and not discussed in the judgment. It remains debateable whether – in a case as highly politicized as glyphosate, concerning the world’s most used pesticide, ⁵⁵ and where all institutional actors were very aware of the significance of the procedure – an (allegedly) flawed risk assessment would not indicate systemic flaws in the regulatory framework. ⁵⁶

Thus, the Court was limited to review the validity of the PPP Regulation in the abstract, without taking into account how it was applied in practice in approval and authorization procedures, being restrained to only declare it invalid in the face of ‘manifest errors of assessment’. The question is therefore not if there is room for improvement, but if there are fundamental problems that would call into question the validity of the Regulation. As also AG Sharpston stated: ‘Most laws are capable of improvement; and the PPP Regulation is probably no exception to that general rule.’ ⁵⁷ The benchmark set by the referred questions is the assessment of alleged manifest errors of assessment in securing the adherence to the precautionary principle, which again set certain boundaries to the judicial review.

In this regard, one of the main criticisms voiced with regard to the precautionary principle is that it is not sufficiently defined in EU law and, therefore, is claimed to be of limited use for application in practice. ⁵⁸ A certain ‘vagueness’ should not be surprising given that it is a principle, not a rule. ⁵⁹ However, as explained before, the Court used its established definition of the principle. Still it was faced with determining which requirements the precautionary principle sets for legislation, and in which cases those where manifestly violated. This is not an easy task as generally the precautionary principle is invoked in cases where either the EU institutions or the Member States have adopted measures to preventing a specific risk from manifesting, like the ban on British beef in the BSE crisis. ⁶⁰ In such case the Court is tasked with reviewing a concrete risk assessment process or risk management measure, as opposed to the whole regulatory framework. In such cases the definition of the precautionary principle as established by the Court is used to review whether the measure taken is based on an appropriate risk assessment and whether it is proportionate. ⁶¹ These cases often revolve around the standard of proof and how much or which type of evidence of risk is required to apply the precautionary principle. ⁶²

This limited review due to a wide margin of discretion, which is not only to be found in EU law but also for example in common law courts, leads to a review that places substantive matters largely outside the remit of the review, while it focuses on procedural matters. ⁶³ As was identified in a review of the case law of the Court on the precautionary principle since 2000 in the context of the EU Horizon 2020 REconciling sCience, Innovation and Precaution through the Engagement of Stakeholders (RECIPES) project, in cases where the Commission or a Member State took a measure based on the precautionary principle, a challenge of the validity of such measure in the large majority of cases was not successful. ⁶⁴

In the rare cases where the challenge was successful, this was due to procedural matters, such as a missing impact assessment. ⁶⁵ The opinion of AG Sharpston in the Blaise case seems to confirm this, as she explained that the principle can be used ‘to challenge an act that is deemed too restrictive, as opposed to an act that is deemed not to be restrictive enough’. ⁶⁶ In essence AG Sharpston argues that the precautionary principle can be used to challenge measures that are ‘too precautious’ in the sense of being disproportionally strict, while it cannot be used – at least on its own – to challenge measures that are ‘not precautious enough’. This deferential review is criticized for example by Alemanno in his annotation to the Gowan case concerning the approval of the pesticide active substance fenarimol, criticized the Court for ‘failing to surround the invocation of the precautionary principle with a set of procedural guarantees’ ⁶⁷ and called for a ‘predictable heightened scrutiny aimed at counterbalancing the broadening of discretionary powers inherent in precautionary action’. ⁶⁸

As elaborated previously, the task of the Court in Blaise, however, was not related to one specific risk or substance of concern, but asked for judicial review of a general regulatory framework. Although the adoption of a pre-authorization scheme itself and the requirements this imposes on applicants constitute a precautionary measure, ⁶⁹ the question in the Blaise case was whether the Regulation is ‘precautious enough’, looking at certain concrete regulatory features that the questions referred to.

At this point, it becomes clear that the political discretion of the legislator and the technical discretion of the administration, and how they are reviewed by the Court, are connected. The review standard that the Court used in the Blaise case to review the validity of the PPP Regulation in essence asks whether the regulatory framework established by the PPP Regulation facilitates that in concrete risk assessment and risk management processes the precautionary principle is safeguarded. The Court established that in single case decisions in the approval and authorization procedures a comprehensive risk assessment, based on the latest and most reliable international scientific data, ⁷⁰ has to be carried out. Generally, in cases involving administrative decision making in the face of technically complex assessments, while the substance of the decision is not subject to judicial review, the Court compensates this with the review of sound administration and the duty of care, focusing on reviewing if the assessment was based on complete and accurate information. ⁷¹ This is also how the General Court approached a claim that the EU ban on neonicotinoid pesticides would violate the precautionary principle. ⁷² These review standards applied by the Court to individual decisions in the context of the application of the precautionary principle, shape the validity assessment of the legislation as carried out in the Blaise case. Concretely the judicial review boils down to examining if the PPP Regulation facilitates the risk assessment of pesticides and their active substance based on sufficient and high-quality scientific evidence. ⁷³

To summarize, given the margin of discretion of the legislator in drafting the PPP Regulation and the resulting limited scope of judicial review, as well as the – mostly procedural – benchmark set by the precautionary principle, the review of the Court remained relatively abstract. This approach to judicial review meant that the Court resorted to analysing the legislative text and not looking much further into the application of the legislation in practice, limiting itself to manifest flaws. However, the ruling was bound to disappoint anyone who awaited revolutionary findings from the Court, especially those who had hoped for an assessment of the glyphosate approval. ⁷⁴

B. A confirmation of the regulatory framework for pesticides in the EU?

As far as the actual review of the validity of the PPP Regulation is concerned, the Court’s position on the issues raised is very relevant, as some of the questioned provisions were also subject to criticisms in the legal literature, ⁷⁵ in a European Citizens Initiative ⁷⁶ and also by EU bodies. ⁷⁷

In the assessment of the ‘cocktail effect’, the Court’s stance is quite theoretical. While the Court is correct to state that, in the approval of active substances and especially in the authorization of PPPs on national level, the cocktail effect of a formulation is assessed, it should also be acknowledged that such an assessment is not easily carried out in practice: EFSA, for instance, has tried to establish harmonized risk assessment methodologies for chemical mixtures since 2006 and only finalized its guideline containing the harmonized framework for such assessment in 2019. ⁷⁸ Moreover, while the intentional mixing of chemicals in plant protection products may be covered by the Regulation, this is different with regard to unintentional and coincidental mixtures. Where the question referred to the Court only covered the mixture of active substance(s) and further chemicals in one pesticide, in the practical application on the fields, a variety of pesticides is used in one area, and this is much harder to assess and not sufficiently taken into account. ⁷⁹ Thus, the Court remained quite legalistic and did not examine whether, beyond the legislation, the risk assessment procedures foresee an examination of the cocktail effect. This would not have required the review of concrete assessments (like the one for glyphosate), which was deemed outside the scope of review, but an examination of the implementation of the legislation in scientific and technical guidance on the level of EFSA or the national regulatory bodies. This is arguably not necessarily included in the validity review; however, it would allow the Court to examine the regulatory framework more comprehensively, while still reviewing the general rules rather than individual cases.

The third question concerned a core feature of the regulatory framework: in accordance with the precautionary principle, the burden of proof has been shifted and producers of pesticides will have to prove the safety of their product. Such prior-approval schemes are used in several policy areas in the EU, including food, chemicals and pharmaceuticals. ⁸⁰ In accordance with the Commission Communication on the precautionary principle, such authorization schemes are used exceptionally with regard to ‘substances deemed “a priori” hazardous’. ⁸¹ This also based on the logic that public money should not be spend on commissioning studies that will be commercially beneficial for an industry marketing a product, especially given the very high costs involved. ⁸² However, it also leads to an information asymmetry between the applying companies and the risk assessing public authorities. The Blaise ruling confirms the validity of the choice of the legislator to base the risk assessment on industry funded studies and strengthens the arguments of the Commission and EFSA, ⁸³ proving that this practice is not only in accordance with the precautionary principle, but actually part of the application of it.

Here, in addition to discussing the ruling in Blaise, one should note that concerns were raised over whether this shift of the burden of proof sufficiently guarantees correct data and an independent and transparent risk assessment. ⁸⁴ These doubts were fuelled in the glyphosate assessment by the Monsanto papers scandal, ⁸⁵ which showed that Monsanto actively interfered in the supposedly objective scientific debate by ghost-writing scientific articles and intruding in peer review process. In the wake of this scandal, the Commission and also EFSA, very much in accordance with the Court’s analysis in the Blaise case, stated that in the risk assessment process the studies submitted by industry are verified through peer review, and note that their correctness or interpretation is not taken for granted. ⁸⁶ Moreover, both emphasized that these studies have to conform to up-to-date legal data requirements and comply with test guidelines developed by the Organisation for Economic Co-operation and Development (OECD) as well as with Good Laboratory Practice (GLP) standards, which ensures the quality and reliability of industry commissioned studies. ⁸⁷

What is remarkable in this regard in the Blaise ruling is the statement the Court makes in paragraph 94: ‘(…) it is the duty of the competent authorities, in particular, to take account of the most reliable scientific data available and the most recent results of international research and not to give in all cases preponderant weight to the studies provided by the applicant.’ At first sight, this paragraph follows logically from the previous paragraphs, in which the Court sets out the obligation to carry out the assessment on the basis of the state of the art scientific knowledge. However, upon closer inspection, the obligation to take into account international research might in practice be delimited by the previously mentioned standardization of scientific studies. In the risk assessment of pesticides, the authorities use the so-called ‘weight of evidence’ approach, which focuses on whether the studies are reproducible, how many studies support a conclusion and also how these studies are designed and conducted. A big emphasis is placed on the studies submitted by the applicant as those comply with the Good Laboratory Practice (GLP), while academic studies are often designed and carried out according to more original and less standardized designs, usually without GLP certification. ⁸⁸ Thus, although the Court confirms the duty to take into account international research beyond the industry studies, in practice the ‘trade-off between “regulatory science” and “research science”’ ⁸⁹ might stand in the way of independent studies counterbalancing the studies submitted by the applicant. However, this is a problem that is difficult to address in the abstract review of the regulatory framework, as opposed to the judicial review of a concrete authorization or approval decision, where it could be easier to question if all relevant studies have been sufficiently taken into account.

Of course in some regards, the Court’s confirmation of the validity of the Regulation constitutes more of an afterthought, as changes have already been introduced through the adoption of Regulation 2019/1381 on the transparency and sustainability of the EU risk assessment in the food chain. ⁹⁰ This reform addresses concerns regarding the transparency of the dossier submitted through making it publicly available – subject to certain delimited exceptions for commercially confidential information – and enables closer examination of the industry studies by third parties in public consultations. ⁹¹ Moreover, it allows the Commission to ask EFSA to commission additional studies in exceptional cases. ⁹² In some ways, this Regulation even closes gaps that the Court did not identify in Blaise, such as the possibility that applicants might stop or not submit their studies where they would provide evidence for risk to human health or the environment. This is now addressed by a mandatory register for studies commissioned for regulatory purposes. ⁹³ However, as noted also by other scholars, it does not change the fundamental choice to place the burden – and also opportunity – to prove the safety of a pesticide on the industry, ⁹⁴ which always means that the industry data is in a certain sense privileged and to large degree shapes the risk assessment. In this regard, the Blaise case is relevant as a confirmation of this regulatory choice and its compliance with the precautionary principle.

Finally, the clarification provided by the Court to the fourth question – that PPPs are not exempted from long-term toxicological studies – was welcomed by some, although doubts are voiced as to whether this will be sufficiently implemented in the Member States. ⁹⁵ However, overall, the Blaise case is certainly not the catalyst for a radical reform in the regulatory approach to pesticides that might have been desired by some actors.