Nuclear safety in France: the keys to understand the government's abrupt reform

 In the midst of the revival of the civil nuclear industry in France, Emmanuel Macron's government has just launched a reform of the nuclear safety expertise and control system that raises questions. By Michaël Mangeon, Université Paris Nanterre - Université Paris Lumières and Mathias Roger, IMT Atlantique - Institut Mines-Télécom

Faced with the climate emergency and the thorny issues about sovereignty and security in power supply, Emmanuel Macron's government has chosen to accelerate the revival and modernisation of the national nuclear fleet.

Since the announcement, in February 2022, of the "renaissance of French nuclear power", with the construction of six "new generation" reactors (EPR 2 type) from 2028, decisions have been flourishing. One example is the bill to speed up procedures for the construction of new nuclear facilities and the operation of existing facilities.

As part of this bill, a new government amendment proposes a reform of the nuclear safety expertise and control system. In order to understand the issues at stake, it is necessary to look back at the way this system has developed in France.

An expertise and control system to meet challenges arising from Messmer Plan

• It was during the 1970s, during the development of the French nuclear power plant programme, that a system of expertise and control of nuclear safety was set up around three actors: the industrialist; a small department of the Ministry of Industry created in 1973 to control nuclear safety; and the IPSN (Institut de Protection et de Sûreté Nucléaire - Institute for Nuclear Protection and Safety), an institute emanating from the Commissariat à l'Energie Atomique (Atomic Energy Commission - CEA) and in charge of expertise and research, created in 1976.

• Although the power plants were drawn fromAmerican technology, the model of expertise and control in force on the other side of the Atlantic is considered too dirigiste and regulatory to be transferred. It was therefore preferred to retain a historical, more flexible and less regulatory approach that allowed the convergence of positions between specialists from different organisations through direct exchange, which the protagonists called 'technical dialogue'. Economic and industrial issues are intertwined with technical and scientific aspects.

After Chernobyl, restoring confidence

Conversely, Chernobyl accident (1986), and in particular the highly publicised "radioactive cloud" affair, tarnished the image of nuclear energy and that of the French control and expertise system.

In response, the Parliamentary Office for the Evaluation of Scientific and Technological Choices (OPECST) advocated setting up a national nuclear safety and information agency, independent of the public authorities, to monitor and regulate installations and ensure communication with the public. This idea was not directly taken up, but nevertheless triggered a gradual change in the system.

In 1998, Jean-Yves le Déaut, MP and member of the OPECST, wrote a report at the request of the then Prime Minister Lionel Jospin entitled "The French system of radiation protection, control and nuclear safety: the long march towards independence and transparency". He recommended the creation of a public expert completely independent of the CEA and a strong safety authority independent of the government. The report also states that "creating too strong an organic link between the safety authority and the centre of expertise would limit the capacity of expression of the centre of expertise".

The creation in 2002 of a public expertise establishment (the Institute for Radiation Protection and Nuclear Safety, IRSN), comprising 1,800 experts and researchers in all fields related to nuclear safety and radiation protection and, in 2006, of an independent administrative authority in charge of control, with more than 500 agents, the Nuclear Safety Authority (ASN), completes this process aimed at restoring the public's confidence in nuclear control and expertise.

At the end of the 2000s, at a time when France was considering a revival of nuclear power, the system of expertise and control appeared to some to be inadequate. A report on the future of the nuclear industry commissioned by President Nicolas Sarkozy from François Roussely, former head of EDF, pointed out the "overzealousness" of the ASN. Similarly, the report recommends that the IRSN should henceforth ensure the dissemination and promotion of French safety rules and standards in order to favour French operators in export markets.

The Fukushima Daiichi accident, which occurred in March 2011, temporarily put an end to these efforts to bring nuclear safety and industrial issues a little closer together. The French system was then regularly held up as a model by international bodies in the face of the risks of collusion between the controller and the controlled, identified as a root cause of the accident in Japan.

The new reform, a historic break

Today, as the spectre of Fukushima recedes and the government has announced its intention to relaunch the nuclear programme, a proposal for reform of the expertise and control system has suddenly been put on the table, by means of a simple press release from the Ministry of Ecological Transition. In particular, it proposes the integration of the IRSN into a "super ASN", which would thus have the dual role of expert and decision-maker in safety matters.

One of the announced objectives of the reform is to "enshrine the independence and transparency of the French nuclear safety system" by transferring IRSN to ASN, an independent administrative authority, which is considered objectively more independent because it is not subject to ministerial supervision.

The government thus presents its project as a natural evolution of the existing system. However, a historical analysis shows that it is more of a break with the existing system, both in terms of form - the project was never discussed beforehand by bodies such as OPECST - and in terms of substance - the current system having been designed in response to a crisis of confidence in nuclear energy that has since diminished considerably.

More independence, but what independence?

Furthermore, the reform is based on a restrictive definition of the notion of independence, as resulting from a simple institutional status. Numerous studies by the International Atomic Energy Agency, the OECD and summaries of research studies have shown that the notion of independence has, on the contrary, multiple dimensions (functional, organic, factual, etc.).

As a 2004 report by the Parliamentary Office for the Evaluation of Legislation on Independent Administrative Authorities reminds us, "independence is a state of mind, and a state of mind cannot be decreed". In this sense, independence is never definitively acquired and there is always a risk of expertise and control being captured by political, industrial or economic interests. From this point of view, the increased proximity of expertise and decision-making within a "super ASN" will put the independence of expertise to the test.

Criticism and unspoken words as the basis of the reform?

Another announced objective of the reform, drawn from the comparison with other control systems that integrate expertise and decision in the same body, aims to "fluidify the decision-making process and gain in coordination" to "reinforce the competences and power of action of the ASN".

Although the government explains that the current system works and is not in question, this reform echoes criticisms of IRSN and ASN, often taboo subjects in the nuclear field, which have been publicly exposed in recent years.

The observation of a complicated power relationship between the two bodies, and even of media competition, was put forward by Yves Bréchet, former High Commissioner for Atomic Energy at the CEA, and Claude Le Bris, who also pointed out that the ASN's operations are too "legal" and not well adapted to industrial constraints.

In a much more direct way, the association Patrimoine nucléaire et climat (PNC) speaks openly of the excesses of the IRSN, which pollutes the "instruction-expertise-decision" process by making its opinions public before the ASN's decisions. IRSN's expertise would then constitute a kind of pre-decision that strongly restricts ASN's room for manoeuvre.

Assessing the risks posed by this reform

In the end, it seems clear to us that this reform transcribes a desire to better reconcile the organisation of expertise and safety control with the new industrial challenges (construction of new nuclear reactors and extension of the operating life of reactors in service).

In our opinion, this desire on the part of political and industrial circles for a more fluid nuclear safety system that is more in line with industrial issues should be better explained and, above all, assumed.

In a context of high industrial stakes and in a world in crisis, such a reform does not only represent an organisational rupture: within a system with interdependent components, organisational evolutions do not go without modifying the rules, practices, relations between the actors of nuclear safety and even the global philosophy of expertise and control. Moreover, from Three Mile Island to Fukushima, via Chernobyl, the functioning of the control and expertise system appears to be one of the causes of the major nuclear accidents.

A comprehensive assessment of the potential opportunities and risks seems to be an essential prerequisite for the launch of a reform potentially impacting the stability of the system, nuclear safety and, ultimately, the credibility of the new nuclear programme.