The glyphosate saga and the fading democratic legitimacy of European Union risk regulation

1. Introduction

What lies at the core of the glyphosate case, and what triggered the heated debate on the European Union renewal of approval of this active substance? What course of action did the European Commission (hereafter, ‘Commission’ or ‘EC’) follow, and how did it ultimately decide to manage the uncertain risks posed by the use of glyphosate? And what role has been played by the ‘Ban Glyphosate’ European Citizens’ Initiative?

This article endeavours to explore the unfolding of the glyphosate saga through the prism of a socially acceptable risk approach to EU risk regulation. Section 1 outlines the main features as well as the inner rationale of socially acceptable risk approaches, setting the stage for the following enquiry into the glyphosate controversy. Sections 2 and 3 focus on the risk assessment phase, analysing the position of the agencies involved against the background of the ‘sound’ science versus scientific uncertainty dichotomy. Sections 4, 5 and 6, on the other hand, provide an overview of the Commission’s narrow approach to risk management, emphasising how the latter has ultimately failed in its political task to reconstruct an EU-wide understanding of ‘intended level’ of health and environmental protection. Against this overall backdrop, Section 7 draws some conclusions on the shifting EU balance of functional and democratic legitimacy – highlighting how the Commission’s inability to flesh out a truly precautionary approach to the governance of glyphosate risks has undermined the very democratic legitimacy of EU risk regulation.

2. The boundaries and remit of EU risk regulation: The ‘socially acceptable risk’ approach

Despite the hegemonic transnational legal narrative on evidence-based risk regulation ¹ and the increasing – albeit discontinuous – scientification of the Court’s judicial review, ² a socially acceptable risk approach to the governance of uncertain risks is somehow encoded in the DNA of EU risk regulation. This paradigm lies at the very heart of the ‘constitutional’ nature of EU risk governance, ³ as originally envisaged.

If it is certainly true that both an evidence-based and a precautionary – or socially acceptable risk ⁴ – soul coexist under EU risk regulation, ⁵ it is equally true that no plausible legal reason exists for the former to trump the latter. Were the results of technical risk assessment meant to be the only – or the most relevant – factor at stake under EU risk regulation, the European Food Safety Authority (EFSA) would be an EU version of the United States Food and Drugs Administration Agency (FDA); however, this is certainly not the case. ⁶ Were technical risk assessors meant to set the relevant threshold of acceptable risk, ⁷ there would be no use for the risk managers to politically weigh and balance all interests at stake, identifying the EU-wide intended level of public health and environmental protection. ⁸ In fact, as most famously argued in Pfizer, ⁹ the distribution of authority between technical risk assessors and political risk managers epitomises the EU struggle to complement and reconcile functional and democratic legitimacy.

Against this backdrop, and irrespective of whether evidence-based risk governance is seen in a positive or a negative light, the foundations of EU risk regulation are rooted in a ‘socially acceptable risk’ approach. This is liable to accommodate a prudential risk assessment as well as a political evaluation of all available risk management options, with a view to ensuring the achievement of the intended level of EU health and environmental protection.

Starting off from the risk assessment phase, ‘socially acceptable risk’ approaches are epistemologically grounded on a pluralistic – rather than monolithic and universalistic – conception of science. ¹⁰ If evidence-based or ‘sound’ science approaches focus on what has been scientifically proved and quantified throughout the phase of technical risk assessment, a socially acceptable risk approach postulates that due consideration should – or may as well – be given to what science cannot prove or measure, in conditions of scientific uncertainty. As testified by the Commission Communication on the Precautionary Principle, a prudential risk assessment ¹¹ is perfectly compatible with – if not inherent to – EU risk regulation. The underlying rationale of the precautionary principle, as enshrined in article 191(2) of the Treaty on the Functioning of the European Union (TFEU), ¹² is that science should in fact be ‘consulted less for the knowledge that it has to offer, than for the doubts and concerns that it is in a position to raise’. ¹³

From this perspective, in the face of persisting scientific uncertainty, technical risk assessment is understood as a necessary but not sufficient condition for the enactment of risk regulation ¹⁴ ; in other words, the results of risk assessment are merely meant to inform the technical knowledge of risk managers. This is somehow a logical corollary of the EU model of risk regulation, with all its specificities. If the determination that a risk exists were to be based on ‘sound’ science, political risk managers would have no role to play. If the notion of acceptable risk were deemed to be ‘neutral’, ‘objective’ and ‘universal’ ¹⁵ – rather than socially embedded ¹⁶ – technical experts would be in charge with EU risk management, and fit for the purpose.

This consideration paves the way for a further analysis of risk management. Under a socially acceptable risk approach, political risk management is meant to encompass the results of the risk assessment phase, the overarching tenets of the precautionary principle ¹⁷ and any ‘other legitimate factors’ at stake. ¹⁸ The risk manager is thereafter called upon to establish whether and how to regulate uncertain risks. ¹⁹ At the core of the socially acceptable risk approach is the question whether, in the face of persisting scientific uncertainty, the risk is acceptable and worth running. A variety of different factors will be taken into consideration and weighed and balanced throughout the risk management process. These include the importance of enhanced public health and environmental protection, as postulated by the preventive philosophy underlying the precautionary principle; the intended level of protection, as pursued in the specific regulatory field; the extent and pervasiveness of any potential adverse effects; ²⁰ public perception of risk; ²¹ an evaluation of the relevant risk management options, including their availability, efficacy and impact; ²² the costs and benefits associated with the relevant risk; ²³ the underlying socio-economic distributional stakes; ²⁴ and any available alternatives to the potentially hazardous product or process, in accordance with the substitution principle. ²⁵

On these grounds the results of the risk assessment phase, the margins of scientific uncertainty, and the threshold of acceptable risk are iteratively interpreted in the light of the intended level of health and environmental protection and of the ‘other legitimate factors’ at stake. ²⁶ Risk management is understood as an inherently value-laden political process, encompassing a range of entangled considerations; the notion of acceptable risk thus unavoidably results from the interaction of cognitive (technical–scientific) assessments and normative (political, socio-economic and ethical) evaluations. ²⁷

If this is the legal and policy-making background, there is more to the socially acceptable risk soul of EU risk regulation. In highly controversial cases, where public health and environmental protection stakes are significant and public opinion is in favour of precautionary regulatory standards, evidence-based approaches are liable to undermine the political component of risk governance, dismantling the very democratic legitimacy of EU risk regulation. In turn, this has a range of far-reaching implications.

Firstly, citizens, whose fights for precautionary standards of protection are frustrated by a narrow adherence to ‘sound’ science, feel increasingly disempowered and disenfranchised. Secondly, a link is automatically established between a narrow evidence-based approach, and pro-market policies.

In the face of persisting scientific uncertainty, ‘sound’ science is liable to affect environmental protection and allocate burdens to consumers, shifting the costs of precautionary risk regulation away from market actors. ²⁸ Most uncertain risks materialise throughout production and trade processes, so that evidence-based risk regulation obviously benefits market stakeholders. Indeed, it is not a coincidence that the Commission Communication on the Precautionary Principle calls upon the political risk manager to strike a fair balance between precautionary public health and environmental protection, and the regulatory burdens placed on market actors. ²⁹ For this reason, when in highly controversial cases EU risk managers disregard both scientific uncertainty and vocal public opinion, a straightforward association is made between ‘sound’ science, on the one hand, and corporate power, on the other.

How does this overall analysis connect to the glyphosate case? This article argues that the glyphosate saga epitomises a patent failure by the EU institutions to strike a fair balance among scientific evidence and precautionary health and environmental protection, science and public perception of risk, technical expertise and democratic legitimacy. It thus exemplifies a further shift towards evidence-based risk regulation, and an utter disregard of what EU-wide public opinion clearly held to be the threshold of acceptable risk and the intended level of protection.

In February 2017, a European Citizens’ Initiative (ECI) was launched with the aim of requesting the Commission to ‘Ban Glyphosate and Protect People and the Environment from Toxic Pesticides’. ³⁰ Pursuant to Article 11(4) Treaty on European Union (TEU), ³¹ at least one million EU citizens from a significant number of EU Member States may request the Commission, within the scope and framework of its powers, to submit a legislative proposal to the Member States ‘on matters where citizens consider that a legal act of the Union is required for the purpose of implementing the Treaties’. The ‘Ban Glyphosate’ ECI put forward three requests. For the purposes of this article, the first and third requests are most relevant: ³² the campaigners respectively asked the Commission ‘to ban glyphosate-based herbicides, exposure to which has been linked to cancer in humans and has led to ecosystems degradation’, and ‘to set EU-wide mandatory reduction targets for pesticide use, with a view to achieving a pesticide-free future’. ³³

As the next sections document, the Commission, in its answer to the ECI campaigners, concluded that there are currently neither scientific nor legal grounds to submit a proposal to ban the use of glyphosate. ³⁴ Further than that, it held that any specific restrictions, conditions or risk-management measures should be adopted by the Member States, pending the authorisation process for glyphosate-based plant protection products; ³⁵ in other words, the Commission refused to unequivocally commit to enact any EU-wide mandatory reduction targets for glyphosate-based herbicides.

Nonetheless, was it appropriate on the Commission’s part to argue that there are currently neither scientific nor legal grounds to ban or phase out glyphosate? And was its political choice to elude the ECI’s request for EU-wide mandatory targets at all suitable? The following sections analyse the main events in the glyphosate saga, together with the applicable regulatory framework, through the prism of a socially acceptable risk approach. The conclusion is that the Commission had scientific ³⁶ and legal grounds, ³⁷ as well as compelling political reasons, ³⁸ to accept the requests put forward by the ECI campaign.

3. The Plant Protection Product Regulation and glyphosate’s risk assessment

To begin with, it is worth remarking that glyphosate is an active substance, ³⁹ rather than a plant protection product (PPP): ⁴⁰ in other words, it is a constituent substance and component of a range of plant protection products, commonly known as ‘pesticides’. Glyphosate has herbicide features, and is thus used as an active substance in herbicide products – most famously, Monsanto’s ubiquitous ‘Roundup’, ⁴¹ Dow AgroSciences’s ‘Accord’ ⁴² and Syngenta’s ‘Touchdown’. ⁴³ Indeed, glyphosate is the most used herbicide active substance in the EU and worldwide. ⁴⁴

The legal base for its controversial renewal of approval is to be found in Regulation (EC) No. 1107/2009, ⁴⁵ concerning the placing of plant protection products on the market (hereafter, ‘PPP Regulation’). Active substances are approved at EU level, whereas it is the responsibility of the single Member States to authorise the placing of any specific plant protection products on their national market. ⁴⁶ The procedure for EU-wide approval or renewal of approval of active substances is laid out in Chapter II, Section 1 of the PPP Regulation. ⁴⁷

Pursuant to Article 4(1) of the PPP Regulation, ⁴⁸ an active substance shall be approved if, in the light of current scientific and technical knowledge, it may be expected that the PPPs containing it will meet the requirements of Article 4(2) and 4(3). On the one hand, according to Article 4(2), the residues of the relevant PPPs shall not have any harmful effects on human or animal health or any unacceptable effects on the environment upon ‘an application consistent with good plant protection practice and having regard to realistic conditions of use’. ⁴⁹ On the other hand, as provided for by Article 4(3), the relevant PPP ⁵⁰ shall be sufficiently effective, shall not have any immediate or delayed harmful effects on human or animal health, shall not cause unnecessary pain and suffering to vertebrates, and shall not have any unacceptable effects on plants, plant products and the environment, having particular regard to its distribution and potential contamination of the environment, its impact on non-target species, and its effects on biodiversity and the ecosystem. ⁵¹ Additionally, the technical assessment shall establish that the relevant active substance satisfies the criteria enshrined in Annex II to the PPP Regulation. ⁵² Thus, an active substance can only be approved if it is not or must not be classified as carcinogenic, mutagen or toxic for reproduction, ⁵³ unless the exposure of humans and residue levels are negligible; and if it is not considered a persistent organic pollutant, a persistent bio-accumulative toxic or a very persistent and very bio-accumulative substance.

The origins of the glyphosate saga date back to 2012. ⁵⁴ In that year the application for the renewal of approval of this active substance was allocated to the German Federal Institute of Risk Assessment (BfR). Pursuant to the PPP Regulation, the competent scientific authority in the so-called ‘Rapporteur’ Member State shall assess the applicant’s dossier ⁵⁵ and submit a Draft Assessment Report to the Commission and the EFSA, evaluating whether the active substance is expected to meet the approval criteria. ⁵⁶ Upon circulating the Draft Assessment Report to the other Member States and the applicant, and after making it available to the public for the submission of written comments, the EFSA shall adopt its conclusions, stating whether the active substance is expected to meet the criteria enshrined in the PPP Regulation and, if appropriate, addressing any risk mitigation options set out in the Draft Assessment Report. ⁵⁷ The BfR’s Draft Assessment Report concluded that, in the light of current technical-scientific knowledge, glyphosate fully complies with the requirements enshrined in the PPP Regulation. ⁵⁸ At the end of October 2015 the EFSA adopted its own opinion on glyphosate, concurring with the conclusions of the BfR. ⁵⁹

Nonetheless, and just a few months earlier, the World Health Organisation’s (WHO) International Agency for Research on Cancer (IARC) had published its findings that glyphosate may in fact be linked to the development of cancer in humans; ⁶⁰ in the light of these results, the IARC decided to classify glyphosate as a Group (2a), ‘probably carcinogenic to humans’ substance. Although the Joint Food and Agriculture Organisation (FAO) and WHO Meeting on Pesticide Residues (JMPR) later concluded that there is no direct scientific evidence of glyphosate’s carcinogenicity, ⁶¹ the conclusions drawn by the IARC are structurally different from the ones reached by the BfR and EFSA. On these grounds, and quite clearly, conflicting bodies of scientific opinion coexist on the assessment of glyphosate’s carcinogenic potential.

In the wake of the remarkable impact of the IARC’s findings on EU public opinion, and in accordance with the specific provisions on carcinogenic active substances set out in the PPP Regulation, ⁶² the Commission rightfully decided to ask the European Chemicals Agency (ECHA) to assess the potential carcinogenicity of glyphosate, as provided for within the framework of the Classification, Labelling and Packaging (CLP) Regulation. ⁶³ On 15 March 2017, the ECHA released its opinion. ⁶⁴ The ECHA’s Risk Assessment Committee concluded by consensus that there is currently no technical–scientific evidence to link glyphosate to the development of cancer in humans; specifically, the Committee decided that glyphosate should not be classified as a carcinogenic, mutagen, or toxic-for-reproduction substance. ⁶⁵ On 7 September 2017, the EFSA published a further opinion on glyphosate, reaching the conclusion that this active substance does not have any endocrine-disrupting properties; ⁶⁶ the latter opinion was – appropriately – requested in the face of persisting scientific uncertainty on glyphosate’s toxicity for reproduction.

How did the Commission interpret the results of the risk assessment phase, and how did this element reflect on the Commission’s narrative on glyphosate?

4. The scientific grounds: ‘Sound’ science…anywhere to be seen?

In its response to the ECI’s request ‘to ban glyphosate-based herbicides, exposure to which has been linked to cancer in humans and has led to ecosystems degradation’, ⁶⁷ the Commission straightforwardly maintained that, as a matter of risk assessment, there are no scientific grounds to justify a ban on glyphosate. ⁶⁸

Starting from the environmental risks posed by glyphosate, the Commission emphasized that a full risk assessment had been conducted to take account of the levels of glyphosate in soil, water and air, as well as of its impact on non-target organisms. The Commission concluded that no evidence had emerged to indicate ecosystem degradation, when glyphosate ‘is used in accordance with the conditions of authorisation and in line with good agricultural practices’. ⁶⁹ This answer is a clear example of what is technically known as ‘boundary work’: ⁷⁰ namely, the analysis of a risk in a technical-scientific ‘vacuum’, whereby the mitigation of complex risks is relegated to the phase of risk management and the availability, efficacy, and impact of risk management measures ⁷¹ is disregarded. The consequence is that the risks which materialise in practice turn out to be unavoidably neglected. Section 6 will analyse this point in greater detail, explaining how the ECI campaigners claim that the faulty implementation of risk management measures at Member State level affects the environmental risks that glyphosate poses in practice; this factor lies at the core of the ECI’s request for EU-wide mandatory targets for the reduction of glyphosate-based herbicides.

Turning to public health risks, the Commission claimed that the available scientific evidence does not support the conclusion that glyphosate is carcinogenic; for this reason, it maintained that there are no scientific grounds to propose a precautionary ban on glyphosate. However, is this position an unquestionable one? Or is the Commission’s argument the mere reflection of an evidence-based perspective? A closer look at the Commission Communication on the ‘Ban Glyphosate’ ECI suggests that the latter is the case.

The Commission started off by underlining that the IARC is the only scientific authority which has so far reached the conclusion that glyphosate is a potentially carcinogenic substance. Further than that, the Commission maintained that the only reason why the IARC drew such a conclusion lies in the latter organisation’s choice to ‘look at both glyphosate – the active substance – and glyphosate-based plant protection products, [whereas] the EU assessment considered only glyphosate, as Member States are responsible for evaluating each plant protection product’; ⁷² and to only consider ‘published studies, whereas the EU assessment also considered the studies submitted by applicants as part of their dossiers’. ⁷³ In this light, the Commission contended, the assessment conducted by the EU authorities was more comprehensive than the one undertaken by the IARC. ⁷⁴

The Commission’s subtle suggestion that the IARC’s minority view is not reliable exemplifies a sound science approach to technical risk assessment. By arguing that the studies conducted at EU level are more comprehensive than the findings of the IARC’s monograph, the Commission Communication indirectly meant to undermine the scientific validity of the IARC’s minority opinion, as if one could clearly discern between ‘sound’, ‘best’ or even ‘fake’ science. This strikes a stark contrast with socially acceptable risk approaches, which rely on a nuanced vision of scientific uncertainty and scientific pluralism.

The way risk is framed and shaped, throughout technical risk assessment, affects all relevant findings and results. A scientist’s choice of one model for risk assessment, rather than another – for instance, the use of a linear, rather than a threshold model for a dose-response function – is critical to the quantification of a risk and the identification of its acceptable level; ⁷⁵ in this perspective, technical risk assessment is neither ‘neutral’ and ‘objective’, nor ‘universal’. Indeed, the case of glyphosate provides a very good example. The IARC’s choice to assess the carcinogenic potential of glyphosate as an active substance and as contained in glyphosate-based PPPs resulted in the classification of glyphosate as ‘probably carcinogenic for humans’; on the other hand, the EU authorities’ limited focus on the assessment of glyphosate as an active substance led to the opposite conclusion.

Despite the Commission’s hint that the EU assessments were based on ‘better’ science, it is legitimate to suggest that the IARC’s choice to assess both glyphosate and glyphosate-based herbicides, together with the latter organisation’s classification of glyphosate as ‘probably carcinogenic to humans’, reflect a genuinely prudential approach to the uncertain risks posed by the use of this active substance. Indeed, the complex case of neo-nicotinoids – also regulated under the PPP Regulation – shows that serious risks to public health and the environment may only be scientifically proved after years of exposure to hazardous substances. ⁷⁶

Against this backdrop, and looking at the case through the prism of a socially acceptable risk approach, it is fair to argue that the margins of scientific uncertainty and the relevant values at stake in the case of glyphosate may have triggered a more thorough reflection, on the part of the Commission, about the entity of the uncertain risks posed by the use of this active substance.

As Section 1 has shown, a prudential risk assessment ⁷⁷ is perfectly compatible – if not inherent to – the institutional architecture and overarching rationale of EU risk regulation. The precondition for the application of a precautionary approach to risk management is the acknowledgment of scientific uncertainty over a potential risk. In the words of the relevant Commission Communication, ‘the precautionary principle is relevant only in the event of a potential risk, even if this risk cannot be fully demonstrated or quantified or its effects determined because of the insufficiency or inconclusive nature of the scientific data’. ⁷⁸ Further than that, scientific uncertainty is defined as a situation where scientific information is insufficient, inconclusive or uncertain, and where the potential risk may be inconsistent with the chosen level of protection of the values at stake. ⁷⁹

Although the technical-scientific assessments conducted by the BfR, EFSA and ECHA did not directly and straightforwardly identify any relevant scientific uncertainties, the Commission could have still legitimately relied on their divergence with the IARC’s opinion. Thus, the margins of scientific uncertainty highlighted by the IARC could have certainly prompted the Commission to ask the WHO to conduct a review of the diverging IARC’s and JMPR’s assessments, as one Member State explicitly suggested during the Comitology process. ⁸⁰ Indeed, the level of scientific controversy surrounding the glyphosate case is still very high: in June 2017, for instance, the state of California decided to add glyphosate to its list of carcinogenic chemicals, with Monsanto’s ‘Roundup’ herbicide now being labelled throughout California as ‘cancer causing’. ⁸¹

It is also worth getting back to the point that, under EU risk regulation, technical risk assessment is acknowledged to be a necessary and yet not sufficient condition ⁸² for regulating uncertain risks, and a necessary but not sufficient component of a more encompassing decision-making process. The results of technical risk assessment are meant to inform the scientific knowledge of risk managers and to underpin, corroborate and substantiate the following phase of risk management; however, and in the face of scientific uncertainty, the precautionary principle mandates that political decision-makers shall ensure enhanced standards of health and environmental protection. ⁸³ For this reason, although it could have – arguably – been difficult for the Commission to enact a straightforward and total ban on glyphosate, there were certainly sufficient scientific grounds for the latter institution to refer to the IARC’s prudential risk assessment and propose a gradual phasing out of glyphosate, as recommended by the European Parliament. ⁸⁴ This point will be further explored throughout the following sections.

In the light of these considerations, it is fair to conclude that the Commission did not give any substantial weight to the margins of scientific uncertainty surrounding the glyphosate case. The narrative underlying the Commission’s response to the ECI campaigners is permeated by a sound science perspective on the results of EU risk assessment; the public perception of the uncertain risks posed by glyphosate and the public opinion backlash against the use of this substance in the EU are deemed to be simply irrational.

The Commission’s position that there were no scientific grounds to ban – or phase out – glyphosate is therefore highly questionable, if the case is analysed through the lens of a socially acceptable risk approach. The denigrators of the precautionary principle have defined precautionary approaches as an ‘infantile disease’ ⁸⁵ of risk regulation, claiming that ‘regulatory priorities should be directed toward the most important risks – which are not necessarily those that are politically most salient’. ⁸⁶ A similar argument has been formulated by reference to behavioural economics, and developed on cognitive grounds; from this perspective, all precautionary approaches rely on cognitive mistakes and failures that are typical of the individual dimension. ⁸⁷ Nonetheless, both ‘sound’ science and precautionary approaches are, in fact, exposed to biases and inconsistencies.

Any narrow adherence to the results of a ‘sound’ technical risk assessment carries the risk that the applicants, or the industry whose profit-making prospects are at stake, may manipulate scientific data and distort scientific evidence by commissioning ad hoc studies. ⁸⁸ In other words, ‘sound’ science approaches are systemically exposed to the risk of market-driven scientific bias. Precaution may be the expression of the ‘laws of fear’ ⁸⁹ as much as scientific ‘soundness’ may be directly exploited by market actors; the procedural consistency of both models is thus largely falsifiable. This point is in fact perfectly exemplified by the glyphosate saga. Whilst any in-depth analysis of the allegations against Monsanto, ⁹⁰ the BfR, ⁹¹ or the EFSA ⁹² would go beyond the scope of this article, it is still worth noting that both the scientific evidence provided by Monsanto and the risk assessments conducted by these authorities have been fiercely and repeatedly challenged. ⁹³ In the face of scientific uncertainty and complexity, ‘sound’ science is nowhere to be seen in the glyphosate case.

5. Risk management: Comitology and the position of the European Parliament

Upon risk assessment, the ball is in the Commission’s court: the latter must adopt a Regulation providing that the approval of the relevant active substance is renewed, subject to any appropriate conditions and restrictions, or is not renewed. ⁹⁴ The Commission must take into consideration, along with the Draft Assessment Report and the EFSA’s conclusions, ⁹⁵ any ‘other factors legitimate to the matter under consideration and the precautionary principle, where the conditions laid out in Article 7(1) of Regulation (EC) No. 178/2002 ⁹⁶ [‘GFL Regulation’] are relevant’. ⁹⁷ Article 7(1) of the GFL Regulation maintains that when scientific uncertainty persists, following a technical risk assessment, as to the possibility of any harmful effects on health (or the environment), ⁹⁸ provisional risk management measures may be adopted with the aim of safeguarding the high level of protection chosen in the Union. ⁹⁹

In accordance with the Comitology rules on the examination procedure, ¹⁰⁰ the Commission must then submit a draft Implementing Regulation to the Standing Committee on Plants, Animals, Food and Feed (PAFF Committee), which shall vote by Qualified Majority Voting. If the Standing Committee delivers a positive opinion, the Commission shall adopt the draft Regulation. ¹⁰¹ If the Standing Committee delivers a negative opinion, the Commission shall not adopt the draft act; however, if an implementing act is deemed to be necessary, it can either submit the same draft Regulation to the Appeal Committee, or submit an amended draft Regulation to the Standing Committee. ¹⁰² Equally, if no Qualified Majority has formed in the Standing Committee either in favour or against the adoption of the draft Regulation, the Commission can either submit the same draft Regulation to the Appeal Committee, or submit an amended draft Regulation to the Standing Committee. ¹⁰³ When the draft Regulation reaches the Appeal Committee, the latter shall also vote by Qualified Majority Voting. ¹⁰⁴ If the opinion of the Appeal Committee is negative, the Commission is precluded from adopting the Regulation. ¹⁰⁵ If the opinion is positive, the Commission shall adopt the draft regulation. ¹⁰⁶ If no Qualified Majority has formed in the Appeal Committee, the Commission may adopt the draft Regulation; ¹⁰⁷ in other words, and unlike under the Old Comitology rules, ¹⁰⁸ the Commission does not find itself under an obligation to adopt the draft Regulation.

How did this complex procedure work out in the case of glyphosate’s renewal of approval? Did the Commission manage to build genuine consensus among Member State representatives in Comitology, and what role did the European Parliament play throughout the process? In March 2016, the Commission expressly proposed to renew the approval of glyphosate for 15 years: however, it failed to build a Qualified Majority in Comitology, and decided to cancel a planned vote. ¹⁰⁹ Just a few weeks later, in April 2016, a Resolution of the European Parliament called for the enactment of further restrictions on the use of glyphosate, ¹¹⁰ and proposed that the renewal of approval should not cover a period longer than 7 years. Notably, in this first Resolution, the European Parliament called on the Commission to submit a new draft Regulation to better address the sustainable use of herbicides containing glyphosate, ban any non-professional uses of glyphosate, and ban any other use in (or close to) public parks, gardens, and playgrounds.

In July 2017, after the release of the ECHA’s opinion on glyphosate’s carcinogenicity, the Commission submitted a new proposal for a 10-year renewal of approval of glyphosate – thus disregarding the European Parliament’s indication that glyphosate should not be reapproved for longer than 7 years. However, the Commission did incorporate part of the European Parliament’s recommendations: the Annexes to the proposed draft implementing Regulation ¹¹¹ included a set of provisions and risk mitigation measures to be complied with by Member States, when approving any glyphosate-based herbicides on their national territory. The Annexes provide for a ban on any glyphosate-based herbicides containing the polyexolyated tallow amine co-formulant. They also maintain that, in their overall assessment for the approval of glyphosate-based herbicides, Member States shall pay particular attention to the protection of groundwater, the protection of operators, the risk to territorial vertebrates and non-target territorial plants, and the compliance of any pre-harvest uses with good agricultural practices. For this purpose, in particular, the ‘conditions of use shall include risk mitigation measures, where appropriate’. ¹¹²

On 25 October 2017, a crucial meeting of the PAFF Committee was held: the discussion focussed on a set of allegations against the EFSA, the activities of the ECI campaign, and the European Parliament’s – second – Resolution on Glyphosate of 24 October 2017. ¹¹³ The latter Resolution ¹¹⁴ prompted the Commission and the Member States not to approve any non-professional uses of glyphosate, any uses of glyphosate in (or close to) public parks, gardens or playgrounds, any agricultural uses of glyphosate where integrated pest management systems are sufficient for the necessary weed control, and any use of glyphosate for pre-harvest desiccation after 15 December 2017. Moreover, and most importantly, it called on the Commission to adopt any necessary measures to phase out glyphosate in the EU no later than 15 December 2022, ensuring that no use of glyphosate is authorised after that date. In other words the European Parliament – with unprecedented clarity and firmness – exhorted the Commission to radically rethink its position on glyphosate, advocating a gradual but outright phasing out of this active substance and explicitly putting forward a request in this sense. The European Parliament’s position thus complemented and strengthened the ECI campaign’s request for a precautionary ban on glyphosate.

The opinions expressed by Member State representatives during the meeting of 25 October 2017 ¹¹⁵ unequivocally showed that the Commission would not be able to muster a Qualified Majority of the PAFF Committee in favour of its draft Regulation. ¹¹⁶ The Commission was thus indirectly forced to amend its draft Regulation again. Indeed, this meeting produced an important change: the draft implementing Regulation was amended to limit the renewal of approval of glyphosate to a period of 5 years. ¹¹⁷ On 9 November 2017, this last draft proposal was submitted to the PAFF Committee, which voted on it. A Qualified Majority of votes was not reached in the PAFF Committee, and the vote did neither result in an opinion in favour of the renewal of approval, nor against it. ¹¹⁸

At this point, as already explained, the Commission could either amend its draft implementing Regulation again, submitting it to the PAFF Committee for a new vote, or submit the same version of the draft Regulation to the Appeal Committee. The Commission chose to pursue the latter strategy, disregarding the clear indication that no EU-wide consensus exists on the determination that the uncertain risks posed by the use of glyphosate are socially acceptable. On 27 November 2017, the Appeal Committee voted on the 5-year renewal of approval of glyphosate. ¹¹⁹ After the last-minute swing of Germany, which decided to vote in favour, a Qualified Majority was reached and the vote resulted in a favourable opinion. ¹²⁰ On 12 December 2017, just a few days before the expiry of the extension of approval, ¹²¹ the Commission enacted Implementing Regulation (EU) No. 2324/2017, renewing the EU-wide approval of glyphosate as an active substance. ¹²²

Despite its final achievement, as this section has briefly explained, the Commission has for almost two years faced the adamant resistance of the European Parliament and of a majority of national representatives in Comitology. How are we to interpret the Commission’s approach, then, through the prism of a socially acceptable risk approach?

6. The legal grounds: The notion of ‘acceptable risk’ and the ‘intended level’ of health and environmental protection

Upon establishing that there is currently no unequivocal scientific evidence of a link between exposure to glyphosate, on the one hand, and any specific hazards, on the other, how are the uncertain health and environmental risks ensuing from its use to be managed? Shall glyphosate be banned, phased out, restricted or allowed tout court? Whilst it often occurs that different risk assessments reach diverging conclusions, the glyphosate case epitomises a deeper disagreement. At the core of the glyphosate saga is not only a dispute over the cognitive, technical–scientific elements of the case; but also, and crucially, a profound normative disagreement on the notions of ‘intended level’ of public health and environmental protection, ‘acceptable risk’ and ‘sustainable’ use of pesticides. ¹²³

Section 3 has argued that the Commission had sufficient scientific grounds to take a prudential stance on the uncertain risks posed by glyphosate, as emerging from conflicting technical assessments. This section explores whether the Commission had any legal grounds to follow a precautionary approach to glyphosate’s risk management. In other words, was the Commission in fact legally unable to enact a ban or a phasing-out plan?

First, an overview of the criteria enshrined in the PPP Regulation can provide a helpful basis to assess the standards of protection that active substances and plant protection products are expected to meet. Article 1(3) of the PPP Regulation maintains that the overarching aim of the regulatory framework is to ensure a high level of human and animal health as well as environmental protection; ¹²⁴ Article 1(4) further reads that, for this very purpose, the provisions of the PPP Regulation ‘are underpinned by the precautionary principle (…)’. ¹²⁵ Articles 4(2) and 4(3), as explained in Section 2, provide that the relevant PPPs and their residues shall not have any harmful – immediate or delayed – effects on human or animal health or any unacceptable effects on the environment and biodiversity. ¹²⁶ These requirements are then substantiated by a range of further technical ¹²⁷ specifications. These encompass, for instance, the requirement that the assessment of the impact on public health shall include a consideration of the effects on vulnerable groups or animal health, directly or through drinking water, food, feed or air, as well as any consequences in the workplace or other indirect effects, taking into account any known cumulative and synergistic impacts; ¹²⁸ moreover, the assessment of any residues with toxicological, ecotoxicological, environmental or drinking water relevance shall be taken into consideration. ¹²⁹ It is thus fair to conclude that the ‘benchmark’ standards of protection incorporated into the applicable regulatory framework are highly protective ones.

Secondly, it is worth getting back to the text of Article 13(2), mandating that when deciding on the approval of an active substance the Commission shall take into account the results of the risk assessment, other factors legitimate to the matter under consideration, and the precautionary principle – whenever scientific uncertainty persists as to the possibility of harmful effects that would not comply with the high level of protection chosen in the Union. ¹³⁰ The text of this article unequivocally reflects a socially acceptable risk approach to the governance of PPPs, whereby the risk manager is called upon to politically weigh and balance the relevant margins of risk, precautionary public health and environmental protection and a range of ‘other legitimate factors’ ¹³¹ – including public perception of risk, environmental sustainability goals and enhanced consumer protection.

In the light of these provisions, there can be little doubt as to the point that the Commission had sufficient legal grounds to enact – at least – a plan to gradually phase out glyphosate. If any such doubt persisted, the text of the second European Parliament’s Resolution would make it fade; in remarkably explicit language, the Resolution states that the Commission’s Regulation ‘fails to ensure a high level of protection of both human and animal health and the environment [and] fails to apply the precautionary principle’, ¹³² thereupon calling on the Commission to ‘withdraw the (…) Regulation and submit a new draft implementing Regulation in line with the requirements laid down by [the PPP Regulation], that is, including not only the EFSA’s opinion, but also other legitimate factors and the precautionary principle’. ¹³³

In the light of these considerations, the Commission’s renewal of approval of glyphosate appears to be nothing but a reflection of the latter’s reliance on a narrow evidence-based approach. The Commission’s decision ¹³⁴ has disregarded the ECI’s powerful argument that the uncertain risks posed by the use of glyphosate are publicly perceived as unacceptable; in turn, this has triggered a heated debate on the available risk management measures, which are regarded as insufficient to safeguard the intended – precautionary – EU level of health and environmental protection.

Further than that, by straightforwardly interpreting the lack of unequivocal scientific evidence as a legal obstacle to the implementation of precautionary measures, the Commission’s response to the ECI campaigners has signposted a highly debatable shift in the understanding of the boundaries and remit of risk management. Under a socially acceptable risk approach, as enshrined in EU risk regulation, setting out the acceptable level of risk for society is an ‘eminently political responsibility’; ¹³⁵ EU risk managers have the political responsibility to decide whether to act, when is ‘safe’ safe enough, ¹³⁶ and how safe is ‘safe’. ¹³⁷ As already explained, this enshrines the superiority of democratic legitimacy and politics over functional legitimacy and technical expertise: ¹³⁸ if the opposite were the case, the EFSA would be a European version of the American FDA. Nonetheless, the Commission has clearly taken a very different perspective throughout its handling of the glyphosate case.

The conclusion is that the Commission has fallen short of its political responsibility to consider whether glyphosate risks are socially acceptable, evaluating whether the use of this active substance complies with the intended level of EU health and environmental protection and enacting appropriate risk management measures. From a ‘socially acceptable risk’ perspective, this approach is both disputable and highly unsatisfactory.

7. Eluding the ECI’s third request: From EU-wide mandatory targets to the national implementation of the Sustainable Use of Pesticides Directive

Before any conclusions are drawn on the glyphosate saga, a few remarks on the Commission’s reaction to the ECI’s proposal ‘to set EU-wide mandatory reduction targets for pesticide use, with a view to achieving a pesticide-free future’ ¹³⁹ must be sketched.

How did the Commission answer this proposal? Specifically, did the Commission – upon rejecting the request for an EU ban or phasing out plan for glyphosate – consider the option of enacting EU-wide mandatory reduction targets for the use of glyphosate-based pesticides?

In its Communication on the glyphosate ECI, the Commission emphasized that EU policy does not aim at the total elimination of pesticides, but rather at achieving a sustainable use of PPPs. ¹⁴⁰ On the one hand, the Commission noted that ‘mandatory volume reduction targets alone do not necessarily reduce the risk from pesticide use’; ¹⁴¹ this occurs because pesticides obviously have different levels of toxicity, so that ‘the focus of the Member States and of the Commission’s work is on the reduction of risk from pesticide use, rather than a simple volume reduction of all pesticides’. ¹⁴² On the other hand, it remarked that many limitations on the use of pesticides – including the recommendation that non-chemical methods should always be preferred whenever they provide satisfactory pest control, and that the use of pesticides must be as specific as possible to target and have the least side effects – are enshrined in the Sustainable Use of Pesticides (SUD) Directive. ¹⁴³

Upon these preliminary clarifications, the Commission turned to the specific request of the ECI campaigners. First, whilst rightfully remarking that EU policy aims at reducing the risks posed by pesticide use, rather than the volume of any pesticides used, the Commission did neither clearly and unequivocally commit to propose EU-wide reduction targets on the grounds of pesticide risk, ¹⁴⁴ nor to enact any mandatory targets for the specific reduction of glyphosate-based herbicides. ¹⁴⁵ Despite vaguely mentioning that it would in the future establish harmonised risk indicators for different pesticides, ¹⁴⁶ the Commission fell short of proposing any measures; indeed, it concluded that at this stage it ‘does not envisage to submit a proposal to establish EU-level reduction targets for pesticide uses’. ¹⁴⁷

Secondly, building on the argument that all necessary conditions and limitations are already enshrined in the SUD Directive, ¹⁴⁸ the Commission turned the ECI’s request for EU-wide targets upside down, claiming that the responsibility to ensure a truly sustainable approach to the use of pesticides lies with the Member States. Notably, the Commission expressly mentioned the findings of its own 2017 Report to the European Parliament and the Council, ¹⁴⁹ acknowledging that the national implementation of the SUD Directive ‘remains patchy, (…) improvement is needed in particular as regards the use of all tools available under Integrated Pest Management [and] national plans are still very diverse in their completeness and coverage’. ¹⁵⁰ However, and despite this admission, the Commission merely committed to assist the Member States in the implementation of the eight principles of Integrated Pest Management set out in Annex II to the SUD Directive, facilitating their conversion and standardisation into specific benchmarks and monitoring national performance. ¹⁵¹

Is the Commission’s answer to the third request of the ECI campaigners politically appropriate and satisfactory? This does not appear to be the case. On the one hand, the Commission deliberately eluded the ECI’s request to take action at EU level; it did neither commit to enact EU-wide mandatory targets for the reduction of glyphosate-based herbicides, nor to put in place a harmonised system for the reduction of pesticide use on the grounds of their toxicity level. Thus, the Commission ultimately washed its hands of any potential reform of the SUD Directive system.

On the other hand, as already explained, it committed to assist the Member States in their implementation of the Directive; however, this is by no means an answer to the different question raised by the ECI. The ECI asked the Commission to propose the enactment of EU-wide mandatory reduction targets, whereas the Commission claimed that the regulatory system delegates the Member States to enact and implement national plans. Rather than answering the question, arguably, the Commission decided to adhere to the status quo.

The ECI campaigners perceive the faulty implementation of the SUD Directive at Member State level as a problem, and asked the Commission to solve it through an EU-wide response: namely, the enactment of mandatory targets. At the heart of this ECI’s proposal lies the acknowledgement that the availability, efficacy and effectiveness of risk management measures ¹⁵² directly impact on the risks which are run in practice; ¹⁵³ for this reason, the ECI advocated the enactment of different risk management measures – and specifically, EU-wide mandatory reduction targets – to safeguard the intended level of EU public health and environmental protection. The Commission, deploying a somehow circular argument, committed to assist the Member States in the enactment and implementation of their national plans: nonetheless, can the problem become part of the solution? Whilst this scenario cannot be aprioristically ruled out, it is certainly difficult to envisage any prospective margins of success. Further than that, and most importantly, by eluding the explicit ECI’s request for EU-wide targets, the Commission disavowed all political responsibility on the matter.

This section concludes the overview of the glyphosate saga. Against this overall backdrop, and pulling the strings of the whole analysis, the next and final section will argue that the Commission has – throughout its handling of the glyphosate case – failed to address the ECI’s powerful arguments and lost a significant political opportunity.

8. The political grounds: Re-legitimising EU risk regulation? A missed political opportunity

Shall uncertain health and environmental risks be run until science has unequivocally proven them, as the hegemonic – evidence-based – transnational legal narrative suggests? ¹⁵⁴ Section 1 has explained that reconciling functional legitimacy, as provided for by technical risk assessment and democratic legitimacy, as dialectically constructed throughout the phase of political risk management, lies at the very heart of EU risk regulation. The rationale for this regulatory choice was to enhance the political nature of risk governance and the democratic accountability of decision-makers, whilst safeguarding the independence of technical experts.

The analysis of the glyphosate saga through the prism of a socially acceptable risk approach triggers two questions : Where has the political and democratic component of EU risk regulation gone? Is it anywhere to be found?

Whilst consumer concerns and public aversion to glyphosate have certainly been magnified by the remarkable media coverage of the matter, the Commission has completely ignored the public view that the uncertain risks posed by glyphosate are not socially acceptable; in so doing, it has failed to consider how the existing risk management measures are deemed to be insufficient to safeguard the intended level of EU health and environmental protection. By eluding the ECI’s requests and by ignoring the European Parliament’s position the Commission has ultimately disregarded its own political role, underplaying the democratic component of EU risk regulation and altering the underlying balance of power in the field.

Moreover, by emphasising that risk management and risk mitigation measures are the sole responsibility of the Member State level, the Commission has reinforced the perception that EU decision-making has a mere ‘technical’ remit. On the one hand, in its proposal to renew the approval of glyphosate, the Commission deliberately relied on a technocratic, ‘sound’ science approach. On the other hand, by refusing to propose mandatory reduction targets for pesticide use, it left all ensuing risk management burdens to the national level; indeed, Member States are entrusted with the politically charged tasks of authorising glyphosate-based herbicides, enacting risk mitigation measures, adopting national plans under the SUD Directive and monitoring their implementation. Ultimately, and despite the ECI’s requests that the EU level shall take action, the regulation and monitoring of ‘sustainable’ use of pesticides is still delegated to the national level.

What lessons are taught by the glyphosate saga? In this case the EU procrastinated taking a decision, and finally mustered a Qualified Majority through compromising. However, this is not the time for compromises. As the answer of the ECI’s organizers suggests, the vague proposals of the Commission and its rhetorical emphasis on transparency and procedures are nothing but a partial and misleading response to the ECI’s campaign. ¹⁵⁵ The case of glyphosate in fact epitomizes a powerful clash and disagreement on the substantive issues at stake, rather than on any of the relevant procedures.

The time has long been ripe for a re-legitimisation of EU risk regulation. The Commission had compelling political grounds to listen to the ECI’s – transnational and EU-wide – message, and should have enacted stringent measures to comply with precautionary standards of public health and environmental protection. By refusing to acknowledge that the uncertain risks posed by glyphosate are publicly perceived as socially unacceptable, the Commission has lost an important opportunity to re-legitimise and re-democratise EU risk regulation, disregarding a set of – controversial – risk assessments and listening to widespread public demand for enhanced levels of protection.

At a time of resurging nationalism, technocracy and ‘output legitimacy’ ¹⁵⁶ will not save the EU; rather, a paradigm shift is urgently needed. Taming anti-EU, anti-globalisation and ‘sovereignty-enhancing’ movements will be impossible if the EU keeps on relying on a technocratic model, privileging regulatory convergence and trade liberalisation over democratic legitimacy and political values – such as consumer protection and environmental sustainability. ¹⁵⁷ Equally, it will be impossible as long as the engagement with political and socio-economic demands is left to the Member State level, away from the ivory tower of EU technicians. In a totally different vein, and well beyond the ‘output legitimacy’ paradigm, the EU needs to fight both technocracy and ‘sovereignty-enhancing’ movements through its own, transnational re-legitimisation and re-democratisation. ¹⁵⁸

The glyphosate case and the EU-wide, transnational ECI campaign offered a superb occasion for the EU to reshape its own transnational identity, getting back to the democratic and political foundations of EU risk regulation. Whilst it may not be too late for the EU to change its overall direction it is, unfortunately, too late for the Commission to look back at its decision in the glyphosate case.