Annex B. Selected cases involving precautionary approaches

This annex presents a selection of cases where precautionary approaches have been at play. While it does not intend to be comprehensive, a variety of sectors and interpretations of the PP are included in this selection. Particular attention is paid to energy transition issues, as well as to innovative technology solutions.

There is evidence on the causes and effects of climate change, but certain issues (particularly those related to future impacts and the deployment of geoengineering solutions) remain uncertain (Pinto-Bazurco, 2020[1]). Article 3 of the United Nations Framework Convention on Climate Change (UNFCCC) establishes that “parties should take precautionary measures to anticipate, prevent, or minimise the causes of climate change and mitigate its adverse effects”, and states that a lack of full scientific certainty should not be used as a reason for postponing measures to prevent serious or irreversible damage.

Nuclear energy represents an important component in the energy mix of 13 of the 27 EU Member States, accounting for about one-quarter of the electricity produced in the EU. However, it remains controversial, in spite of its strong benefits in terms of carbon and other emissions. Among the reasons for this are catastrophic accidents such as Chernobyl (1986) and Fukushima (2011)1, as well as intrinsic factors that tend to increase the risk perception level of some types of technologies irrespective of their actual, scientifically assessed level of risk (Slovic, 1987[2]) (Slovic and Peters, 2006[3]) Nuclear energy in the EU is governed to a large extent by the Euratom Treaty. All EU member states are party to it by default. The EU Directive 2009/71/EURATOM, amended through Directive 2014/87/EURATOM2 provides a community framework for the nuclear safety of nuclear installations. Article 2(b) of the Directive calls for the establishment of uniform safety standards to protect the health of workers, as well as the public. According to the Directive, Member States must ensure that the national framework in place requires license holders to establish and implement management systems which give priority to nuclear safety. These must be regularly verified by the competent regulatory authority.3 However, the Directive does not provide guidance on specific impact estimates that should be used by the Member States to that end.

In the wake of the 2011 Fukushima Daiichi nuclear disaster, Japanese authorities shut down most of the country’s nuclear plants (a number of which have since then reopened). Many lessons were learned from the disaster in terms of improving regulatory and management effectiveness to strengthen safety (Ostendorff, 2013[4]). Critics noted that, as a result of parting with an important source of electricity, Japan had to meet the resulting energy demand by importing fossil fuels, which led to higher energy prices and an increase of greenhouse gas emissions (Pinto-Bazurco, 2020[1]). A similar debate has taken place in Germany, ultimately leading to the country exiting nuclear power. The recent energy crisis following the invasion of Ukraine, as well as the growing urgency of climate change and its consequences, have led to renewed discussions around nuclear power, and its revival in a number of countries, illustrating the need for a more holistic vision of what “precaution” actually means, as suggested in this report – i.e. that all aspects (climate risk, security of supply, etc.) of risk should be taken into account, and not only one category of safety or environmental risk.

The topic of nuclear energy, its risks and their adequate regulation has been the subject of decades of science, analysis, and debate, which would go far beyond the scope of this report to properly consider or summarize. The readers can refer to a number of important works on the topic (Wellock, 2021[5]) (NEA, 2011[6]). From the perspective of this report, what matters is that understanding what actually constitutes real precaution in the case of nuclear energy and its use requires a consideration (a) of scientifically assessed risks rather than risk perceptions, (b) of climate, economic, social and other risks that may be increased if other energy sources are used, and not only of narrowly defined safety issues.

The PP has found some wind power applications in biodiversity management. According to the European Commission, The Capercaillies (Black Forest) case study on wind-power developments illustrates a “gradated” approach to precaution (European Commission, 2017[7]). Box A B.1 provides further detail.

The Dutch Ministry of Infrastructure and Water Management has issued a series of guidelines4 on risk, safety and the application of the PP within the Dutch legal order. These guidelines are based on the Communication of the European Commission on the PP and the general risk management principles. They also mention a set of additional precautionary criteria that Dutch courts use, deriving from case law research in the area of health and safety (which can be of interest for the Dutch government in most areas):

  • If scientific data indicates that a new technology is harmful to health or safety, both market players and the government must conduct further research to determine whether precautionary measures are necessary.

  • The higher the estimated risk to health and safety, the greater the duty of care for both market players and government to prevent damage as far as possible.

  • A duty to warn and inform is part of the duty to act.

  • This responsibility translates into an ascending scale: first, the government must provide information and require the market to provide information on the dangers, including safety measures. Then a regulatory framework must be put in place. Finally, this framework must be enforced.

  • As insights grow, the government must keep a hand on the tap to see whether increasing stringency of regulation is warranted. Scaling down is also allowed, if objective scientific research shows the dangers are less than expected.

These guidelines also highlight a strong correlation between the application of the PP and liability in Dutch practice. Even though the application of precaution is not solely the responsibility of the national government, government ministers are seen as liable and accountable, as they must be able to prove at any time that adequate measures are taken and that other parties also deal with risks in a careful manner.

The Memorandum “Consciously dealing with safety; red threads” of the Dutch Ministry of Infrastructure and the Environment (182014) states that “measures can be taken in the form of further research to reduce uncertainty, information to offer action perspectives and possibly exposure-reducing measures, a provisional ban or (adapted) regulation. Stakeholders should be involved at the earliest possible stage". The government will also have to establish that they monitor and benefit from any new development, and that new risks are being handled responsibly and carefully.

The EU REACH Regulation (EC) No. 1907/20065 on chemicals states that its provisions “are underpinned by the precautionary principle” (Article 1(3). Burden of proof lies with the supplier or manufacturer, requiring companies to demonstrate to the European Chemicals Agency how the substances can be safely used, and to communicate health and safety information to the other users in the supply chain (European Commission, 2017[7]).

A 2011 report on the topic highlighted that the application of the PP in the chemicals sector requires taking decisions on what is considered an "acceptable" level of risk for society, identifying gaps in knowledge that result in uncertainty concerning the nature of a potentially unacceptable risk, and managing that risk in the face of uncertainty (Milieu Ltd, 2011[10]). The report also notes that the PP should only be applied in the event of a potential risk that cannot be fully ascertained or quantified, or its effects cannot be determined because of the insufficiency or inconclusive nature of the scientific data. It stresses that a scientific evaluation of the potential adverse effects (risk assessment) should always be undertaken based on the available data (hazard identification, hazard characterisation, estimation of exposure and risk characterisation). This should lead to a conclusion on the possibility of the occurrence and the severity of a hazard's impact on the environment or the health of a given population (including the extent of possible damage, persistency, reversibility and delayed effect). It should also lead to a description of the remaining uncertainties in order to help decision makers in the risk management phase.

The report provides a framework for determining whether the information available indicates the potential for harmful effects. The seven steps are as follows:

  • Has a potential negative effect been identified?

  • Has a scientific evaluation of the substance been carried out?

  • What are the uncertainties concerning the nature or probability of the possible harm?

  • What are the options available for controlling the risk of possible harm?

  • Do the options for risk management measures meet the five elements of the precautionary principle?

  • What is the process followed for reaching implementation decisions?

  • Is there a plan for reviewing the actions taken if new scientific knowledge emerges?

For each step, the report provides specific guidance on the various elements that should be considered. This includes an excerpt from the 2000 Communication, information on developments in smart regulation, and summaries of appropriate ECJ case law.

Since the late 1940s, antibiotics have been added to animal feed to accelerate the livestock growth and productivity. Antibiotic resistance in bacteria was observed in the 1950s and the possibility of transferring resistance to other species of bacteria was documented in the 1960s. In 1985, Sweden banned this use of antibiotics because of its uncertain long-term effects. In 1998, the European Union took the precaution of prohibiting the use of four antibiotics for this purpose (European Parliamentary Research Service, 2015[11]).

By November 1986, important amounts of meat from cattle infected with bovine spongiform encephalopathy (BSE) had been consumed in the United Kingdom. As soon as the first cases of BSE had been diagnosed, senior officials realised that BSE posed a possible risk to human health. However, UK policymakers’ chose to avoid taking any regulatory actions and decided to keep information about BSE from the public. By 1988, the British media began devoting more attention to the cattle disease and, for the first time, a small expert advisory committee was set up to provide advice on BSE. This only occurred at the insistence of the Chief Medical Officer at the Department of Health, who was not informed about the new disease until March 1988. Senior officials from the Ministry of Agriculture, Fisheries and Food introduced a slaughter and compensation policy for diseased cattle which, at the time, were being sold for human food. The delay in implementing the regulations also meant there had already been repeated human exposures to BSE (European Environment Agency, 2011[12]).

(Millstone and Van Zwanenberg, 2007[13]) find that the UK government claimed to be sensibly protecting public health, while in practice it “covertly subordinated the protection of public health to the support of agricultural sales, with a view to minimising state intervention and public expenditure.” According to the authors, if the UK government had adopted a “genuinely precautionary approach”, it would have necessitated, firstly, reforming policy institutions to separate responsibilities for regulation from those of sponsorship. Secondly, a precautionary approach would also have required acknowledging how little substantiated scientific information was available and engaging in open and accountable discussions about the possible costs and benefits of a wide range of different possible courses of action. The authors argue that separating the institutions between those who are responsible for providing the scientific advice, risk assessments and research to policymakers, would encourage more open discussion of possible risks. Lastly, the authors find that research conducted by a wide range of interdisciplinary groups, with open access to the evidence and data, would make it harder to conceal uncertainties (Millstone and Van Zwanenberg, 2007[13]).

These substances include CFCs, HFCs and PFCs. Their use as refrigerants became widespread in the 1930s and subsequently in aerosol sprays. In 1974, scientists suggested that they could destroy the ozone layer. Some countries began to restrict or ban their use in 1977. In 1987, the Montreal Protocol provided for the phase-out of ozone depleting substances (European Environment Agency, 2013[14]).

Following the first indications of risks to human health (skin and bone conditions, liver cancer) in the 1950s and 1960s, and after having denied the risks for some time, the chemical industry financed carcinogenic tests and then significantly lowered the exposure limits. The International Agency for Research on Cancer has recognised vinyl chloride as a “human carcinogen” since 1979 (European Environment Agency, 2013[14]).

The EC’s Recommendation on a code of conduct for responsible nanosciences and nanotechnology research (European Commission, 2009[15]) was designed to help foster collaboration and communication between relevant parties, including policymakers, researchers, industry and civil society. On the basis of a precautionary approach, France has introduced a mandatory nanotechnology reporting scheme, which requires companies to file a declaration for each nanomaterial they produce, import or distribute. It is argued (European Environment Agency, 2013[16]) that “political decision-makers have yet to address many of the shortcomings in legislation, research and development, and limitations in risk assessment, management and governance of nanotechnologies and other emerging technologies. If left unresolved, this could hamper society's ability to ensure responsible development of nanotechnologies”.

A number of countries have established a moratorium on GMOs purportedly based on the uncertainty concerning their effects on public health and ecosystems. Critics contend that this moratorium could jeopardise food availability, especially in developing countries.

In an example of precautionary approach, under the EU Directive on the Deliberate Release of Genetically Modified Organisms, suppliers of GMOs have to demonstrate the safety of the organism before it is placed on the EU market (see Box A B.2 for an overview of selected examples of burden-of-proof provisions). There has been criticism that the principle is being applied selectively, with organisms that pose similar risks but have not been produced using genetic engineering not being subject to the same precautionary approach (European Commission, 2017[7]).

The UK’s experience with GMO regulation has also been discussed in academic literature in connection with the PP. The country’s government initially saw GM crops as a potential new market, as no hazards had been shown by early research. Given spreading concerns, organisations such as the Royal Society for the Protection of Birds and Friends of the Earth called for GM crop development to be stopped. These concerns included whether GM crops, such as plants resistant to insect pests, might escape into wild populations and impact negatively on biodiversity, as well as potential impacts on human health. As a result, the UK Government delayed development of GM food crops until field trials had been completed. The government also formed a Cabinet Committee on Biotechnology and initiated public debate on genetic modification. The scientific review of GM crops found gaps in the knowledge base and it was therefore decided that approval of these products should be assessed on a case-by-case basis (European Environment Agency, 2013[14]).

In 2011 the World Health Organization's International Agency for Research on Cancer (IARC) categorised the radiation fields from mobile phones and other devices that emit similar non-ionizing electromagnetic fields (EMFs) as a “possible” human carcinogen. Nine years earlier, the IARC had given the same classification to the magnetic fields from overhead electric power lines. The decision on mobile phones was principally based on two sets of case-control studies of possible links between mobile phone use and brain tumours: the IARC Interphone study and the Hardell group studies from Sweden. Both provided complementary and generally mutually supportive results.

There are by now several meta-analyses and reviews on mobile phones and brain tumours that describe the methodological limitations of the major studies published so far and the difficulties of interpreting their results. The fact that the mechanism for radiofrequency electromagnetic fields carcinogenesis is unclear seems to have supported the view that descriptive data on brain tumour incidence is of limited value. It is believed that precautionary actions to reduce head exposures would limit the size and seriousness of any brain tumour risk that may exist. Evidence is increasing that workers with heavy long-term use of wireless phones who develop glioma or acoustic neuroma as a result should be compensated. The first case in the world was established in October 2012. The Italian Supreme Court affirmed a previous ruling that the Insurance Body for Work (INAIL) must grant workers’ compensation to a businessman who had used wireless phones for 12 years and developed a neuroma in the brain. It is acknowledged that the benefits of mobile telecommunications are many, but such benefits need to be accompanied by consideration of the possibility of widespread harms (European Environment Agency, 2013[14]).


In a dispute concerning a French ban on asbestos and products containing asbestos, the WTO’s Appellate Body confirmed that a country may take measures to protect human health from serious risks on the basis of a divergent opinion coming from qualified and respected sources. See below for further details.


[9] Braunisch, V. et al. (2015), “Underpinning the precautionary principle with evidence: A spatial concept for guiding wind power development in endangered species’ habitats”, Journal for Nature Conservation, Vol. 24, pp. 31-40,

[8] European Commission (2021), Guidance document on wind energy developments and EU nature legislation,

[7] European Commission (2017), Future brief: The precautionary principle: decision-making under uncertainty.

[15] European Commission (2009), Commission recommendation on A code of conduct for responsible nanosciences and nanotechnologies research, (accessed on 4 May 2023).

[16] European Environment Agency (2013), “Late lessons from early warnings: science, precaution, innovation”, (accessed on 4 May 2023).

[14] European Environment Agency (2013), Late lessons from early warnings: science, precaution, innovation,

[12] European Environment Agency (2011), Late lessons from early warnings:,

[11] European Parliamentary Research Service (2015), The Precautionary Principle. Definitions, applications and governance,

[10] Milieu Ltd, T. (2011), Considerations on the application of the Precautionary Principle in the chemicals sector. Report prepared for DG Environment of the European Commission,

[13] Millstone, E. and P. Van Zwanenberg (2007), “Mad Cow Disease — Painting Policy‐Making into a Corner”, Journal of Risk Research, Vol. 10/5, pp. 661-691,

[6] NEA (2011), Improving Nuclear Regulation, OECD Publishing,

[4] Ostendorff, W. (2013), Adequate protection after the Fukushima Daiichi accident: A constant in a world of change, OECD Publishing, pp. 23-41,

[1] Pinto-Bazurco, J. (2020), The Precautionary Principle,

[2] Slovic, P. (1987), “Perception of Risk”, Science, Vol. 236/4799, pp. 280-285,

[3] Slovic, P. and E. Peters (2006), “Risk Perception and Affect”, Current Directions in Psychological Science, Vol. 15/6, pp. 322-325,

[17] Smetana, M., M. Onderco and T. Etienne (2021), “Do Germany and the Netherlands want to say goodbye to US nuclear weapons?”, Bulletin of the Atomic Scientists 77:4, pp. 215-221,

[18] Taleb, N. N. et al (2014), The Precautionary Principle (with Application to the Genetic Modification of Organisms),

[5] Wellock, T. (2021), Safe Enough?, University of California Press,

[19] Yin, M. and Zou, K. T (2021), The Implementation of the Precautionary Principle in Nuclear Safety Regulation: Challenges and Prospects,


← 1. Nuclear energy | Fact Sheets on the European Union | European Parliament (2022). Retrieved 9 February 2022, from

← 2. Cf., retrieved 9 June 2022.

← 3. Council Directive 2009/71/Euratom of 25 June 2009 establishing a Community framework for the nuclear safety of nuclear installations OJ L 172, 2.7.2009, p. 18–22.

← 4. Beleidsgerei handreikingen voor risico-en veiligheidsvraagstukken Voorzorg Wat komt er allemaal bij voorzorg kijken? and Voorzorg Adequate voorzorg, verantwoorde en zorgvuldige omgang met nieuwe risico’s (Policy tool handouts for risk and safety issues Precaution What is involved in precaution? and Precaution Adequate precaution, responsible and careful handling of new risks), Ministry of Infrastructure and Water Management.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2023

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at