Transparency for public trust?

The text adopted today by the European Parliament is going to become the new EU General Food Law Regulation, with eight sectoral food regulations amended accordingly in the process (GMOs (cultivation and for Food/Feed uses), feed additives, smoke flavourings, food contact materials, food additives, food enzymes and flavourings, plant protection products, and novel foods). The outcome of this vote has been near certain since the final decision on the text occurred on late in the night of 19 February, at the European Parliament in Strasbourg.

And we are done! First ever citizens’ initiative gets an agreed legislation! It shows that with a political will and determination we can do wonders addressing citizens’ concerns. #transparency #gfl #food
Big thanks to @ayuso_pilar & co-rapporteurs & @ro2019eu ?? pic.twitter.com/mzYKi8W5RN

— Vytenis Andriukaitis (@V_Andriukaitis) 11 février 2019

Vitenys Andriukaitis, the Lithuanian EU Commissioner for food safety and health, in charge of the file of the European Commission, celebrated the deal he had just concluded by taking a Twitter selfie with the representatives of the European Parliament and the Council’s Presidency.  SidenoteRespectively, Conservative Spanish MEP Pilar Ayuso and, at that time, Romania

The result of a confidential negotiation between the three institutions – a trilogue – to reach an agreement on a common version, this new regulation revises the EU’s 2002 General Food Law Regulation SidenoteThis very important EU regulation, adopted in 2002 after the food scares of the late 1990s, created the European Food Safety Authority (EFSA). and, among other features, gives the public access to most of the previously confidential data provided by the food and agribusiness industries to obtain EU market authorisations for their products.

As the Commissioner makes clear in his tweet, the text is the EU’s official response to the 1.3 millions citizens who signed the European Citizens Initiative (ECI) ‘#StopGlyphosate’ (Corporate Europe Observatory was a co-organiser). Specifically, to the second of the three demands made by the ECI: “ensure that the scientific evaluation of pesticides for EU regulatory approval is based only on published studies, which are commissioned by competent public authorities instead of the pesticide industry.”

He knows from first-hand how important this is.

An EU Health and Food Safety Commissioner with a problem

Flash-back to March 2016. Back then, Vitenys Andriukaitis had a big problem on his hands.

A rare public row between two international public health bodies, the European Food Safety Authority (EFSA) and the World Health Organisation’s International Agency for Research on Cancer (IARC), had erupted over their conflicting evaluations of the most-used pesticide in the world, the total herbicide glyphosate (originally patented by Monsanto). The cancer specialists assembled by IARC had unanimously concluded that the substance was a probable carcinogen to humans, while EFSA didn’t classify glyphosate under any cancer category, not even Category 2 ("suspected human carcinogen" in the EU nomenclature). According to the EU’s pesticides regulations, recognising glyphosate as a probable carcinogen would have forced the Commission to propose a ban SidenoteOr to only re-authorise it through one of the loopholes in the regulation... of the substance to EU member states, who would take the final decision.

It is worth noting that EFSA partly acknowledged the wider question of the disastrous consequences for biodiversity of using this total herbicide at the scale it is currently used, pointing to direct harm to the environment for some uses as well as “concerns regarding the indirect effects (biodiversity) on non-target organisms via trophic interaction of extensively used herbicides such as glyphosate”. This could have been used as a basis for, if not a ban, at least a very serious reduction in authorised uses.

But the Commission, confronted with loud protests from the entire agribusiness industry and their many Brussels lobbyists, neither tried to propose a ban, nor serious restrictions. It feared the scale of the consequences of such a decision on farming practices in the EU and the global trade of food products. SidenoteThe European Commission was also fiercely lobbied by most major grain exporting countries not to ban glyphosate. See ‘Fears grow that EU glyphosate ban could disrupt global trade’, Politico Europe, 27 October 2017 So in early 2016 it suggested to propose to all EU member states to reauthorise the herbicide for fifteen years.

But, in February 2016, France announced that it would not support this, and several EU countries followed suit to the point that there was no longer a qualified majority to relicense the product. There was little hope the row between IARC and EFSA would go any time soon, because, and this is key to the whole affair, EFSA could not publish the studies underlying its assessment to defend itself (it could only publish the summaries of these studies as provided by industry, which was insufficient).

The reason is that the raw data of these studies was protected by copyright and belonged to the members of the Glyphosate Task Force, an industry consortium led by Monsanto and including most of the corporations producing glyphosate-based herbicides. As the EU food safety legislation requires, they had provided the studies – in confidentiality – to the EU as part of their application for glyphosate’s relicensing.

On 4 April 2016, Andriukaitis wrote to the Glyphosate Task Force (GTF) a probably slightly embarrassing letter, asking them to “publish proactively the full studies provided to EFSA” because “the glyphosate authorisation process has drawn an extraordinary degree of public attention and concerns, particularly in the European Parliament, about the transparency of the European assessment process”.

The GTF immediately deflected with a false solution, saying that in order to protect the alleged “intellectual property and confidential information” in the data they were creating a “reading room” in Brussels for whoever would be interested in looking at it. The conditions of access de facto prevented any meaningful use of it under the pretext of avoiding what the pesticides industry termed “unfair commercial use including misuse of copyright”.

These events brought into public light a little-known but fundamental scandal that scientists and NGOs have been denouncing for many years: the ‘scientific’ evaluation performed by EFSA on regulated food products in the EU is primarily based on data provided by these products’ manufacturers, who systematically oppose disclosure beyond EFSA - and EFSA itself, fearing industry lawsuits, has consistently interpreted the applicable legislation in a too conservative way, this lack of transparency leading to the agency being repeatedly condemned by the European Court of Justice.

This secrecy makes public and scientific scrutiny of EU food safety evaluations very difficult, if not impossible. Not a small irony when one considers how much industry repeats the ‘sound science’ mantra in the defence of its products.

As with many scientists and campaigners before him, SidenoteFor instance, it took Corporate Europe Observatory and Green MEPs about a year to obtain some of the raw glyphosate data from the European Food Safety Authority (EFSA) because the agency was worried it would get sued if it disclosed it. And what we obtained was partly redacted and in a format (images) that was very difficult to use, to put it mildly, even though this data could partly be used for finding important omissions in the pesticides industry scientists’ interpretation of the animal data. But up to today, nobody has been able to use it for a scientific publication. Andriukaitis was discovering this indefensible situation. When the #StopGlyphosate ECI petition was registered, in December 2016, he was given an opportunity to end it.

The European Commission answers EU Citizens

But the European Commission, which prepared the first version of the reform, decided to only answer the ECI’s data transparency demand: “ensure that the scientific evaluation of pesticides for EU regulatory approval is based only on published studies, which are commissioned by competent public authorities instead of the pesticide industry”.

The ECI’s first and third demands, namely “ban glyphosate-based herbicides” (given their environmental and human health impacts) and “set EU-wide mandatory reduction targets for pesticide use, with a view to achieving a pesticide-free future”, with “political and financial support” to alternative pests and weeds management, SidenoteTo avoid glyphosate-based herbicides simply being substituted with equivalent or more toxic pesticides. Given how prevalent the use of glyphosate is in farming, it was very clear that technical and trade support measures for farmers would have been needed for a ban of glyphosate to occur without worsening farmers’ health and economic situation. were ignored SidenoteEventually, glyphosate was renewed for five years because Germany’s Conservative Agriculture Minister Christian Schmidt breached his Government coalition’s agreement to obtain the necessary qualified majority among member states. or explained away as unnecessary. SidenoteThe European Commission explained its handling of the ECI demands in a Communication published in December 2017.

The European Commission is trapped in a dilemma here: acknowledging serious flaws in the EU food safety system risks undermining previous EU food safety decisions. So, understandably, they described their reform as about regaining the public’s trust rather than solving internal problems. SidenoteAs a matter of fact, a March 2019 ruling by the European Court of Justice again confirmed EU case law that existing legislation already warranted the full disclosure of pesticides toxicity tests data; from that perspective, the main improvement of the revised regulation as far as pesticides toxicity studies data are concerned (the ruling did not deal with the other food sectors) is that the ( disclosure system will move from reactive to proactive, saving everyone precious time and resources, and will also deliver data in a usable format.

To its credit, the European Commission broadened its revision of the text to all regulated products on the EU food market, not only pesticides: regulations on food contact materials, food additives, food enzymes and flavourings, novel foods, GMOs (cultivation and for Food/Feed uses), feed additives, and smoke flavourings were also reviewed along the same lines.

This large expansion of the scope of the reform angered many lobbyists from these other food and agribusiness industries, who argued during the consultation meeting organised in early February 2018 by the European Commission that they had nothing to do with the problems of glyphosate, and that this was a danger to their competitiveness.

To no avail: all these industries had also refused to let the EU publish the regulatory toxicology data they submitted to obtain market approvals, and this behaviour was just as indefensible as it was for glyphosate producers. The broadening of the reform’s scope to all regulated food products made perfect sense.

As a result of today’s vote, most of the data from the studies used by the EU to grant market approvals to products entering the human food chain is going to be made accessible to the public and, in particular, to scientists and researchers willing to test EU food safety evaluations (Article 38). The elements that industry is allowed to keep confidential (Article 39), as long as it justifies that this is deserved, are strictly limited and, it seems, correctly defined. Moreover, EFSA can still publish these elements if they are needed to understand the risks.

This proactive disclosure wasn’t possible before, and is a considerable improvement both for the public and not least for EFSA which will hopefully be able to use for its own purposes the work done by external scientists on its data.

Beyond its answer to the ECI, the European Commission proposed other changes in the revised legislation and two in particular are potentially very meaningful:

• For the EU to create a register for regulatory studies, where companies will have to notify (Article 32b) all the tests they commission the moment they start them. This is a good tool to prevent the practice of companies only sending studies that deliver ‘good’ results for them (the Monsanto Papers have shown examples of studies commissioned by Monsanto on glyphosate whose results were never disclosed because they threatened the firm’s interest).

• For the European Commission to propose that the European Food Safety Authority (EFSA) is given very substantial additional resources to make the new system work, including the possibility to commission (when asked by the European Commission) additional studies in (vaguely-defined) “exceptional circumstances”. Reminder: EFSA currently has hardly any resources to perform or commission research by itself. While this also depends on the outcome of the ongoing negotiations about the 2021-2027 EU budget among Member States (the so-called MFF), it is proposed (see p. 51) that the EFSA would receive an additional €25 millions in 2020, €44 millions in 2021 and €62,505 million from 2022 onwards, enabling the creation of 106 new staff positions. EFSA’s current budget is about €80 millions, so we’re talking about a 78 per cent budget increase here!

Of course, merely throwing money at a problem does no solve it, but internal expertise is a key condition for an administration to resist corporate lobbying and this needs to be paid for. The EFSA has been underfunded so far considering the size and importance of its task, SidenoteAs an element of comparison, the current annual budget of its US technical equivalent, the Food & Drug Administration (covering also medicines), is almost $6 billion. National agencies also contribute to EFSA’s work, but the total remains far off from the US counterpart. and, because it was designed that way, has largely relied since its creation on external expertise brought in by the food industry, in particular experts with financial interests (consultancy, research funding, management roles in captured organisations, etc) in companies whose products are regulated by the agency (in other words, experts in a conflict of interests situation). These new resources should help the agency defend its independence from the food industry; if it doesn't succeed, however, this large increase via taxpayers’ money becomes harder to justify.

Accessible data, great. But can it be used?

The core element of this reform is about increasing public access to data. But for this disclosure to be meaningful, the data has to be usable, and in particular, quotable in scientific publications. As the European Ombudswoman Emily O’Reilly recently said in a speech on conflicts of interests, “transparency is not an end in itself, rather a tool for accountability”.

The reform should allow EFSA to require companies to provide the data in a format that allows it to be “searched, copied and printed” (Article 39f), which is a technical but crucial improvement. So far, companies have mainly sent the data to regulatory agencies as hardly usable image PDFs… which worsens public food safety agencies’ limited ability to double-check what the food industry sends them.

As the ‘copy-paste’ scandal on glyphosate has indicated – where the German food safety national agency (BfR) was accused of having uncritically reused industry’s very dismissive evaluations of the academic literature on the substance – national regulatory agencies do not always check whether the summaries of studies provided by industry faithfully reflect the raw data of these studies’ experiments.

This was well demonstrated thanks to an investigation led by a group of German journalists, who compared the evaluation of 25 different pesticides by EU national food safety agencies (who perform the actual evaluation of industry’s application before EFSA reviews it) with the evaluations submitted by companies. They found that for 15 of them it was impossible to ascertain whether the national agency at stake had gone beyond uncritically copying industry’s own evaluations.

Re-using the data disclosed by EFSA mustn’t depend on having to obtain its copyright owner’s authorisation each time, because, as we’ve seen time and again, food and agribusiness companies have always opposed disclosure so far (with the exception of a recent initiative of the pesticides industry, see below).

The revised General Food Law is quite discouraging on the issue of data reuse. An important amendment (163) was adopted by the European Parliament explaining that, “For the avoidance of doubt, the information published may be used for the purpose of public and academic scrutiny of the results, including a better understanding of the potential adverse effects on health and the environment”; but this could not make it to the final text because it was unacceptable for the European Commission and the member states, for whom touching copyright legislation (also under heated review at the time) was a red line.

The final text therefore foresees very much the opposite (Article 38): “The disclosure to the public of the information mentioned in paragraph (1)(c) shall not be considered as an explicit or implicit permission or license for the relevant data and information and their content to be used, reproduced, or otherwise exploited in breach of any intellectual property right or data exclusivity rules and its use by third parties shall not engage the responsibility of the European Union.”

In other words, scientists can use the disclosed data but at their own peril. This peril has a name: EU copyright law.

Navigating the uncertain waters of copyright law

The pesticides industry’s own transparency initiative, initiated by Bayer in January 2018, before the European Commission proposed its draft reform (probably as a way to demonstrate good will and show that regulation was unnecessary), and transformed into a general commitment of the pesticides industry by its global lobby group CropLife, provides some insights into the legal restrictions the pesticides industry thinks it can impose.

According to the terms of use one must accept in order to access the data Bayer proposes, for instance, “Except as provided in the Quotation section below, the user is not allowed to distribute, reproduce and/or publish the Study Documents. The user is not allowed to use Study Documents for any regulatory or commercial purposes, which includes, without limitation, any use by the user and/or any third party that could reasonably be understood to be for or in support of the creation of profit.”

It continues: “The user may reproduce, distribute and communicate to the public reasonable parts of the Study Documents if these parts are included in an independent scientific work for the purpose of explaining the contents provided that the origin and Bayer’s copyrights are appropriately cited.”

It is not entirely clear whether “reasonable parts” and “regulatory purpose” would enable an analysis of the data before EFSA has finalised its evaluation. At least it seems that individual scientists could access and publish an analysis of the data provided they do not quote too much of it. But a systematic analysis of the data relevant to a given substance would, at least in theory, need the full dataset to be citable. Will this be the case?

EU copyright law on the issue of databases allows for free quotation by any EU citizen of “substantial parts” of a publicly accessible database, provided the use is non-commercial and that the author is always acknowledged. But how about analysing entire datasets?

According to Tristan Berger, a French law researcher currently finishing a PhD on “Access to information relating to sanitary and environmental expertise”, it seems possible but researchers from certain EU countries might have to pay for this:

“The Directive 96/9/EC on the legal protection of databases gives the possibility, but not the obligation SidenoteArticle 9 of the 96/9/EC Directive: “Member States may stipulate that lawful users of a database which is made available to the public in whatever manner may, without the authorization of its maker, extract or re-utilize a substantial part of its contents (...) in the case of extraction for the purposes of illustration for teaching or scientific research, as long as the source is indicated and to the extent justified by the non-commercial purpose to be achieved”., to Member States to allow for the non-commercial use of a database for research purposes. In certain countries such as in France, the Directive has been transposed in a way that forces researchers to negotiate a use fee with the database's copyrights holders.” SidenoteEmail communication with CEO, 15 April 2019

A directly relevant EU legislation to this issue is the much-debated and much-lobbied Copyright Directive, eventually adopted on Monday 15^th April by EU member states. This new Directive amends the EU Database Directive we already mentioned.

Its Article 3 creates an exception for “reproductions and extractions made by research organisations and cultural heritage institutions in order to carry out text and data mining of works or other subject-matter to which they have lawful access, for the purposes of scientific research”, which could precisely protect research institutions (so academics are covered, but not independent researchers, journalists, etc) using the full dataset disclosed by EFSA. What is more, no opt-out is available nor contractual override (art. 7).

Article 3a of the same Directive details what is available to non-academics: text and data mining is possible by everyone, but copyright owners can opt out and can impose a contract overriding this clause.

Thanks to the big fight put up by civil society organisations to secure strong exceptions when the European Parliament amended the file, the EU Trade Secrets Protection Directive does not enable industry to claim confidentiality for data they give to public administrations and that must disclosed according to EU law.

So, as things look now, and pending a more expert legal analysis, it looks like academics will be authorised to use and quote the entire data set disclosed by EFSA, but that industry could impose a fee, and a contractual override of this right to non-academics. Can implementing legislation by the European Commission and EFSA remedy these limitations?

The European Commission was in a hurry

The European Commission only focused on the first half of the ECI’s second demand: “ensure that the scientific evaluation of pesticides for EU regulatory approval is based only on published studies”.

But the regulatory data used by the EU for evaluating the toxicity of regulated products on the market is still going to be commissioned and provided by companies seeking market approvals. Whereas it could have been possible to take the commissioning of regulatory studies out of the hands of industry entirely (giving it instead to a public authority financed by a public fund collecting the costs of the tests from industry, for instance), the European Commission preferred to focus on this more limited approach.

It had little time to get this legislation adopted before the end of the Parliamentary term, and, even if it had wanted to, embarking on such an ambitious reform risked being too much to manage to convince the conservative forces in power in the European Parliament and among member states.

But it means that the fundamental problem of industry largely controlling the production of relevant regulatory data remains, despite this reform.

Beyond the European Commission not replying to the other ECI demands, there are other potential problems in the adopted regulation one needs to be aware of:

• With the exception of an important recommendation (Recital 14) that EFSA should prioritise the recruitment of external experts “who are also actively conducting research, and publishing their research findings in peer-reviewed scientific journals, provided that they comply with the strict criteria of excellence and independence”, nothing was added to close the two main existing flaws of EFSA’s independence policy, which enable EFSA to keep recruiting experts who still receive industry research funding. SidenoteAs long as the researchers at stake does not receive more than 25 per cent of their total research funding from the company at stake – which is so high a threshold that it still enables most researchers with industry research funding to stay on EFSA’s panels.

• A long-standing industry demand to EFSA, the possibility for companies requesting market approvals to have a meeting with the agency for guidance before submitting their application, has been accepted (Article 32a). While safeguards have been inserted (eg the staff delivering the advice cannot be the same as the same staff involved in the application assessment, minutes of the meetings will be published), they will need to be strictly enforced.

• Systematic public consultations on the completeness of application dossiers submitted by industry: this is already partly in place, but strong outreach to, for instance, academic scientists active on the issue will be important, otherwise the only contributions to these consultations will come from lobbyists, as they have a direct interest in the product and the resources to intervene. SidenoteEFSA has already started to organise expert panel meetings that are open to observers; in general, the overwhelming majority of these observers are industry lobbyists.

• The DNA sequence of a GMO can be kept confidential as a trade secret.

• There is a very strong insistence, throughout the regulation, on risk communication (ie to better explain the EU’s food safety decisions to EU citizens), including the necessity (from an EU perspective) to better coordinate what national food safety agencies and EFSA are saying. This is potentially problematic, as national food safety agencies have sometimes been earlier than EFSA at pointing at issues – for instance on Bisphenol A. What is the point of having different public agencies if they are forced to say the same things?

Crocodile cheers: industry ‘celebrates’ after having fought all the way

Actions speak louder than words, especially when it comes to lobbying. The two people who were the first to congratulate Commissioner Andriukaitis’ tweet about the EU's response to the Stop Glyphosate ECI were not ECI signatories but industry lobbyists: Jean-Pierre Azoulay and Graeme Taylor, respectively the Director General and the Public Affairs Director of the European Crop Protection Association (ECPA), the EU lobby association of the pesticides industry. “Improved transparency and confidence in our regulatory system is positive for all” celebrated Azoulay. But ECPA’s actions over the years belie these words of celebration.

We have good reasons to suspect that industry’s sudden congratulations to the Commissioner and transparency celebrations are hollow. The food and chemical industries have always opposed demands for transparency of their regulatory data, hiding behind claims of commercial confidentiality, investment ‘protection’, fearmongering about commercial espionage, and legal threats. Not only did they fight disclosure of their data before the reform was announced, for instance on glyphosate, but they lobbied to derail and hollow it out all along the way.

For instance, in October 2018, ECPA’s Director Azoulay – together with the directors of the lobby groups of the chemical, plastics, biotech, seeds and animal medicines industries – co-signed a discrete but angry letter to the official in charge of the GFL file in the Austrian Presidency of the Council to express their “considerable concerns” about the file, copying the letter to 120 other lobby groups directly or indirectly concerned by the reform.

This letter lists the main arguments pushed by industry throughout the discussions, from the European Commission’s draft to the debates in the European Parliament and among EU member states. It was a small miracle that most of these arguments were defeated two months later in the European Parliament, because the largest party in the European Parliament, the European People’s Party (EPP -Christian-Democrats/Conservatives), industry’s most faithful ally in general, lost most of the votes.

Industry’s friends lose in Parliament

The European Parliament had appointed as rapporteur SidenoteIn the European Parliament, a different ‘rapporteur’ is chosen for each piece of legislation the Parliament has to debate and vote. The MEP chosen is a central figure as (s)he will draft the Parliament’s first position on the file, and represent the Parliament later in trilogue negotiations. for the text Renate Sommer, a German CDU MEP of the European People’s Party (EPP). A very experienced MEP sitting in the Parliament since 1999, and trained as a food economist, Sommer is the current President of the European Parliament’s Beer club, a “forum for discussion and information exchange about issues that affect the brewing sector within the European Union”. It is quite popular among MEPs, with dozens of members, and the EU beer industry lobby Brewers of Europe provides generous “logistical support” to keep MEPs hydrated and duly informed of their priorities. Back in 2010, Sommer was the European Parliament’s rapporteur on food labelling, a regulation alcoholic beverages (except alcopops) had been excluded from.

It seems she did not like the European Commission’s reform proposal for the General Food Law very much.

In September-October 2018, she made clear her intentions not to let the Parliament finalise its work on the file before the end of this legislative term. In other words, this would have meant losing minimum two years, and leaving the decision to a new Commission and a new Parliament who wouldn’t have gone through the glyphosate crisis and to whom everything would have needed to be explained again. That's assuming they would even be willing to listen. Realistically, it is probable that such a move would have killed the reform for good.

Fortunately, the other political groups refused this (also thanks to NGOs lobbying them) and the Parliament’s Committee on Environment, Public Health and Food Safety (ENVI) went on to debate the file.

Both in ENVI then, after having lost there, and in plenary, she (and other MEPs, mainly from the EPP and the ECR group – the British Conservatives) tabled amendments reflecting rather closely the concerns formulated in industry’s letter to the Austrian Presidency mentioned above. None of these points made it to the final text, fortunately:

• Industry concern: “The reversal of the burden of proof, and the required level of proof necessary for confidentiality”.

Industry would have much preferred the responsibility to demonstrate the need for disclosure to fall on EFSA, rather than having to demonstrate themselves, as was proposed by the Commission, the need to keep certain items in their studies confidential. Renate Sommer tabled an amendment (141) proposing just that, which was defeated by a 10% margin (354 against vs. 295 in favour, see detailed vote analysis below).

• Industry concern: “The extremely limited scope of data that may be treated as confidential”.

An amendment (370) tabled by Renate Sommer in ENVI, added to the restricted list of items industry could claim confidentiality for a very vague, and therefore very broad, concept: “innovative product ideas”. This amendment passed.

• “The timing of release which is not in line with precedent in other sectors such as pharmaceuticals”.

This refers to one of the most intensely fought battles in the Parliament: the question of exactly when in the risk assessment process would EFSA make the data available to the public (at least in its non-confidential version). The Commission, here agreeing with the ECI demands, had proposed that this would be as soon as EFSA would get the data, but industry wanted this to be much later, so that no proper analysis could be performed before EFSA reaches its final opinion. There again, Renate Sommer (another MEP had done it before her in ENVI, Christopher Fjellner) tabled an amendment (142) that would have delayed the moment of disclosure to after EFSA had reached its first draft opinion, certainly too late to enable any meaningful scrutiny. It was rejected by a very small margin (332 against vs 312 in favour, see below for the detail).

Eventually, on 11 December 2018, Renate Sommer lost the plenary vote on most of her amendments, and all the really toxic ones (a detailed analysis of the main votes, MEP by MEP, is available here and below).

Very angry, she immediately resigned from her rapporteurship, denouncing a “populist vote”. The rapporteurship was given to another EPP MEP, Pilar Ayuso, who defended the Parliament's position in the next steps of the negotiations.

The text was further amended in trilogue negotiations, but its main features did not change much. Fortunately, the Council refused the “innovative ideas” amendment during the trilogue negotiations and a much better formulation was agreed upon: “the manufacturing or production process, including the method and innovative aspects thereof, as well as other technical and industrial specifications inherent to that process or method, except for information which is relevant to the assessment of the safety”.

Conclusion

While the EU’s answer to the #StopGlyphosate European Citizens Initiative did not, in the end, stop glyphosate, it promises more transparency on food toxicity data and this, potentially, is a considerable victory. The introduction of a register for regulatory studies to try to stop industry sending only favourable results to regulators, and more resources for the public administration in charge of delivering “trusted science for safe food” SidenoteThis is the new motto of EFSA, as well as the title of its new strategy. in the EU, are also important victories.

However, the main success criteria test for this new regulation should be whether or not it allows for more, and meaningful, scientific scrutiny on EFSA’s work. It seems – as far as we can tell for now – that copyright legislation should not prevent academic scientists from analysing the whole individual datasets published by EFSA, pending, of course, that they don’t have to pay (and if so not too much) for it and that EFSA discloses this data in a format that enables it. This will need to be watched closely, and solutions found for non-profit researchers not affiliated to academia.

Data transparency, then. But what about quality?

This reform does not change the fundamental problem of industry controlling the production of regulatory data.

It does not touch EFSA’s bad habit of giving more weight to regulatory than to academic studies. Industry would like the EU's risk assessment to keep focussed as much as possible on costly, standardised experiments it has to deliver legally, and only consider the peer-reviewed literature as second grade, whereas, as Monsanto already put it in 2001 in an internal email disclosed in the Monsanto Papers, “Data generated by academics has always been a major concern for us in the defense of our products”.

It does not challenge industry's often successful attempts to influence the very methods used by EFSA and regulatory agencies to evaluate regulated food products. According to a recent PAN Europe report, "in 92% (11 out of 12) of the EU methods for pesticide risk assessment examined, it was the industry that designed and/or promoted their regulatory use."

This reform also does not close the loopholes in EFSA’s independence policy.

And it does not bring transparency to the most crucial part of the whole decision-making process: the positions and votes of member states when they decide, in the end, whether or not to approve regulated products on the EU market (a technical process known as Comitology). The European Parliament itself voted to shed light on these food safety decisions. But Juncker’s attempt to reform the comitology process in this direction is currently sleeping in the drawers of József Szájer, a Conservative Polish member of the European Parliament... and it is obvious industry and member states both hate it.

Time will tell whether this reform, thanks to a better access to EFSA’s industry data, enables a better analysis of it by academia. But will it on its own, as industry puts it, deliver ‘improved confidence in our regulatory system’?