Skip to main content Accessibility help
Hostname: page-component-848d4c4894-75dct Total loading time: 0 Render date: 2024-05-26T16:38:48.031Z Has data issue: false hasContentIssue false

34 - Legitimate Protection against Slow-Onset Disasters and the Prohibition against Trade-Restricting Measures

from Part VI - The Future of the Internal Market

Published online by Cambridge University Press:  04 May 2019

Fabian Amtenbrink
Erasmus Universiteit Rotterdam
Gareth Davies
Vrije Universiteit, Amsterdam
Dimitry Kochenov
Rijksuniversiteit Groningen, The Netherlands
Justin Lindeboom
Rijksuniversiteit Groningen, The Netherlands


The Internal Market and the Future of European Integration
Essays in Honour of Laurence W. Gormley
, pp. 590 - 602
Publisher: Cambridge University Press
Print publication year: 2019

I Disaster Protection vs Free Movement

Not all disasters strike out of the blue. Some disasters develop over a long time – sometimes over several years. This allows societies to adopt measures to ward off – or at least reduce – the onslaught. Indeed, if a society can predict a future disaster with any reasonable certainty, most would probably agree that the authorities must be under a duty to adopt measures to ward it off.

Perhaps climate change is the best example of a slow-onset disaster: science is able not only to establish that climate change takes place but also to point out its likely consequences. Thus, we do not know precisely where and when climate change will manifest itself, but in practice there is consensus that the world will experience more severe hurricanes and heatwaves, increased wildfires, more prolonged droughts, rising sea levels, etc.Footnote 1 In other words, we have fairly clear and convincing predictions about what is in store for us – and for our children. The development of a number of diseases is also likely to be a slow-onset disaster. For example, we are witnessing an increased spread of multi-resistant pathogenic bacteria, i.e. bacteria that have become resistant to antibiotics. If we are unable to address this challenge, in the future we may expect to see people dying from infections that today we consider minor and easy to treat.Footnote 2

Thus, it is clear that we are faced with a number of predictable disasters – and it therefore seems only natural and well founded that we seek to protect ourselves against these threats. Where the authorities of a Member State introduce measures aimed at protecting that state against these predictable, future disasters there might well be instances where the measures, actually or potentially, directly or indirectly are capable of hindering the free movement of goods and/or services within the European Union – giving rise to the question whether the measures are lawful under EU law. This schism between the protection against slow-onset disasters and the EU’s free movement rules is the focal point of the present contribution in honour of Professor Laurence Gormley.

Three observations must be made before we embark upon our examination. First, we presuppose that the Member State measures constitute a barrier to trade covered by the provisions on free movement (goods and services)Footnote 3 laid down in the Treaty on the Functioning of the European Union (TFEU). It follows that we set out to clarify when a Member State measure is lawful irrespective of the fact that it impedes trade between Member States. Second, we presuppose that no secondary EU regulation applies in the field in question having the same objective as the one pursued by the Member State measure.Footnote 4 Third, we only consider ‘future disasters’ – meaning that, by definition, there will always be some uncertainty as to whether the disaster actually occurs – and if so how (i.e. when and with what consequences).

In what follows we first turn to consider the notion of ‘risk’ in an EU law context (section II). Next, we discuss how Member States may justify introducing measures against slow-onset disasters (section III). We thereupon consider how the (scientific) certainty and specificity regarding the different manifestations of the slow-onset disasters impact upon the legal assessment (section IV). We also discuss the situation where a Member State adopts measures addressing disasters occurring outside the Member State’s own territory (section V). Finally, we sum up our findings (section VI).

II Risk and EU Law

Before considering how to approach slow-onset disasters today, it may be useful to take a small step back in time; back to the 1980s when first of all the late Ulrich Beck published his seminal work Risikogesellschaft. Auf dem Weg in eine andere Moderne (English title: Risk Society: Towards a New Modernity).Footnote 5 In this work Beck argued that the world was witnessing a new and systematic way of dealing with hazards and insecurities that were induced and introduced by modernisation itself. For our purposes, what is essential to keep in mind with regard to this ‘risk approach’ is, first, that in modern society human agency has come to play a central role both as an important cause of these risks and with regard to mitigating them.Footnote 6 And, second, that contemporary natural sciences play a key role in this context both because to a considerable extent ‘modern risks’ may be traced back to innovations based on the natural sciences, and because we need the natural sciences to identify the risk as such, to predict the consequences of the risk where it materialises, to estimate the likelihood that it materialises and to find ways of countering it.

Below we will attach particular importance to two aspects inherent in slow-onset disasters, namely (1) the likelihood that the risk materialises and (2) the consequences it produces if it materialises.

Unsurprisingly, we must clearly distinguish the situation where there is a high likelihood that a risk will materialise producing major (adverse) consequences from the situation where there is a low likelihood that a risk will materialise – and that this will only have minor consequences.

Figure 34.1 The four categories of slow-onset disasters

Taking a risk approach is in no way alien to EU lawyers. Indeed, such an approach often forms an integral part of the EU legislative process. For example, if the authorities set out to regulate the use of a food additive, they may ask the producer of the additive to provide verifiable scientific data about the likely positive and negative effects of using the additive as well as the likelihood that the different negative effects will materialise. The legislator must also determine what risk level they consider acceptable – and they may draft the legislation accordingly. Similarly, we find the risk approach reflected in the proportionality principle as developed under EU law; thus, the authorities have a wider array of measures to choose between when they want to protect society against a risk that is likely to materialise and which will produce very adverse consequences as compared to a risk that is less likely to materialise and which will produce less adverse consequences if, nevertheless, it does materialise.

III Justifying Measures Addressing Slow-Onset Disasters

Any EU lawyer will know that where an EU Member State introduces measures that, directly or indirectly, actually or potentially, are capable of hindering intra-EU trade, this constitutes a Treaty infringement unless the measure pursues a legitimate objective and is proportionate. The Treaty itself lists some particularly weighty objectives that may justify the introduction of measures that constitute restrictions or are indirectly discriminatory or (in principle) even directly discriminatory.Footnote 7 Moreover, the Court of Justice of the European Union has established that measures that restrict intra-EU trade without being directly discriminatory can be lawful if they pursue so-called ‘mandatory requirements’Footnote 8 and are proportionate.

Slow-onset disasters may manifest themselves in many different ways. Still, where Member States adopt measures to prevent the disasters from materialising, in practice they are likely to only pursue a rather narrow range of objectives. Thus, a disaster will normally imply a risk to the health and life of humans, animals or plants, a risk to property and infrastructure and a risk to the environment more broadly. Moreover, there may be situations where a Member State fears that a disaster will affect public policy or public security – although these situations are likely to be less common. Measures introduced to ward off such risks are, as a clear rule, justifiable.Footnote 9

In other words, where a Member State adopts measures to prevent the negative impact of a slow-onset disaster while at the same time restricting trade between the Member States, it seems rather unlikely that the measures will not be found to pursue some legitimate interests under EU law.Footnote 10

IV Predicting and Countering Slow-Onset Disasters: Certainty and Specificity

A key characteristic of slow-onset disasters is that they develop over time, thereby allowing us to foresee them – at least to some extent. For example, drought is a classic slow-onset disaster that often develops over a long period of time. Adopting measures to address this type of slow-onset disaster does not differ from other Member State measures; thus, if the measure pursues a legitimate interest in a non-discriminatory way and if it is proportionate with this objective, it will be lawful under EU law.Footnote 11

The above concerns situations where it is possible to foresee the slow-onset disaster with a high degree of certainty meaning that scientific data permit a full evaluation of the inherent risk. However, not all slow-onset disasters can be foreseen with any appreciable degree of certainty. For example, considerable efforts are put into predicting the consequences of climate change, but still there are appreciable uncertainties associated with these predictions. And the further we look into the future, the more uncertain and imprecise are the predictions.

In other words, we are faced with a situation where, on the basis of ‘the most reliable scientific data available and the most recent results of international research’,Footnote 12 it is possible to identify specific climate change-induced phenomena that may pose serious threats to human societies – such as more powerful and more frequent hurricanes or more frequent and more prolonged water shortages – identified by a scientific and objective evaluation, but where it is impossible to determine with certainty the existence or extent of the alleged risk because of the insufficiency, inconclusiveness or imprecision of the results of studies conducted.Footnote 13 In this situation, under EU law, we must revert to the precautionary principle.Footnote 14

Thus, the precautionary principle applies to cases where scientific evidence is insufficient, inconclusive or uncertain and where a preliminary scientific evaluation indicates that there are reasonable grounds for concern that the potentially dangerous effects on the environment or on human, animal or plant health may be inconsistent with the desired level of protection that has been laid down by the authorities. Where the authorities can establish that such situation exists, under the precautionary principle they may adopt measures aimed at providing protection against this presumed risk in order to ensure the level of protection laid down by the authorities.Footnote 15 These measures may be in force pending further scientific information that allows for a more comprehensive risk assessment, they must be proportionate and they may not be more restrictive of trade than what is required to achieve the level of protection laid down by the authorities.Footnote 16 Moreover, it is incumbent on the authorities to regularly review the measures in order to ensure that they continue to be justifiable.

When establishing whether a given measure, adopted on the basis of the precautionary principle, is proportionate we must take into account not only (1) the likelihood that the risk materialises and (2) the consequences it produces if it materialises, but also (3) the certainty with which we can establish the likelihood and the consequences. In other words, if our scientific data with a very high degree of certainty show that there is a high likelihood that climate change will cause some adverse consequences the authorities will have a strong basis for adopting measures targeting these consequences even if the consequences are not particularly far-reaching. In contrast, if scientific data indicate that there is a high likelihood that some very adverse consequences will materialise sometime in the future, but these data involve a high level of uncertainty, the authorities have a much weaker basis for adopting the measures.

In the opinion of the present author, the precautionary principle in combination with the principle of proportionality, as first of all developed in the case law of the European Court of Justice, provide a sound legal framework for when and how Member States may (or may not) introduce measures addressing slow-onset disasters. Still, as convincingly pointed out by Cass Sunstein in Laws of Fear,Footnote 17 whereas the European Union’s approach to the precautionary principleFootnote 18 provides a plausible start, it nevertheless leaves many open questions as well as a number of doubts.Footnote 19

V Member State Pursuing (Legitimate) Interests beyond Its Own Territory

Slow-onset disasters do not recognise borders. Thus, multi-resistant pathogenic bacteria may travel from one continent to another – in a very short time. And climate change may manifest itself in many different ways and places. That slow-onset disasters are a matter for the international society is reflected in a number of international agreements. For example, in 2015 WHO, the UN World Health Organization, laid down a Global Action Plan on Antimicrobial Resistance,Footnote 20 also in 2015, 196 parties entered into the Paris Climate Agreement,Footnote 21 and in the same year the so-called Sustainable Development Goals (SDGs) were adopted by the UN Member States and global civil society actors.Footnote 22 Many more international arrangements exist where the participating parties undertake to cooperate in order to address what essentially are slow-onset disasters.

The EU Member States and/or the European Union take a very active part both in the international legislative work and in the ensuing arrangements. These activities only very rarely give rise to questions regarding possible infringements of the Treaty rules on free movement, but nevertheless it may still be useful to briefly examine this possible conflict. To this end, we shall distinguish between, on the one hand, situations where a Member State adopts measures aimed at promoting its own interests, and, on the other hand, situations where the measure is intended to promote the interests of the state where the slow-onset disaster takes place. Let us illustrate this through an example where a slow-onset disaster begins to materialise.

VI Example

In Iceland a contagious and deadly bacteria slowly starts to spread. The bacteria is resistant to all but one type of antibiotics – and if at some point it turns into an epidemic, it is feared that we will have a new ‘Spanish flu epidemic’. In order to protect its citizens against this danger, Member State A prohibits all air carriers that continue to service Iceland from servicing Member State A airports.

Member State A’s measure may be contrasted with that of Member State B. Thus, Member State B prohibits all Member State B holders of the only type of antibiotics that is known to defeat the bacteria from selling (including exporting) any part of their stock. Instead they are required to surrender their full stock at cost price to Member State B in order that this antibiotic may be sent to Iceland.

Member State A’s measure excludes air carriers in other Member States from servicing Member State A, and Member State B’s measure prevents the holders of a specific antibiotic from exporting their stock of this antibiotic to other Member States.Footnote 23 Both measures therefore constitute a trade restriction as defined in the European Court of Justice’s Dassonville ruling (capable of directly or indirectly, actually or potentially hindering trade).Footnote 24 But whereas the measure adopted by Member State A is intended to protect the interests of this Member State, the measure adopted by Member State B is first of all adopted in the interest of Iceland.

In either of the two situations the state on whose territory a slow-onset disaster manifests itself must (as a rule) be considered to be the principal responsible for addressing the disaster. Therefore, in order to introduce a measure that is capable of restricting intra-EU trade and which suitably addresses slow-onset disasters that occur outside the Member State in question, the Member State must demonstrate that the measure is ‘necessary’.Footnote 25 If the slow-onset disaster takes place outside the Member State’s own territory, but may be expected to produce adverse effects within the Member State’s territory, this will be very similar to the situation where the slow-onset disaster manifests itself directly within the Member State.Footnote 26 Thus, whether a slow-onset disaster directly or indirectly produces adverse consequences within the territory of a Member State should not have any bearing on the legality assessment of any measures adopted to counter the slow-onset disaster.Footnote 27

The situation is rather different where a Member State adopts measures that are capable of restricting intra-EU trade in order to protect interests exclusively outside its own territory. If these interests are found in another EU Member State and if the measures are not supported by this other Member State, it seems to be very difficult to establish an argument that the measures are necessary (and lawful).Footnote 28 In contrast, if the measures are introduced with the acceptance of this other Member State (perhaps even at its instigation) the measures may be presumed lawful (presupposing that they are proportionate and comply with the principle of proportionality).

The situation is likely to be the same where an EU Member State adopts measures addressing slow-onset disasters in third countries, unless the third country is incapable of or unwilling to protect its own citizens (etc.) against slow-onset disasters. For example, during the Arab Spring, governments in several Middle Eastern and North African states were paralysed while at the same time the countries suffered from widespread droughts. In this situation assistance from EU Member States would seem warranted.

Moreover, under international public law – for example the Paris Climate Agreement – Member States may be under a legal obligation to provide assistance to third countries. If a Member State, in full conformity with EU law, has assumed a legal obligation to provide assistance to a third country and if such assistance can only be provided in a way that is capable of hindering intra-EU trade, we may expect the European Court of Justice to accept this.Footnote 29 Admittedly, it is rather difficult to imagine such situation in practice.

VII Conclusions

In this contribution we have first seen that applying a risk approach has become an integral part of EU law. Slow-onset disasters really are about risks – and may manifest themselves in many different ways. Such manifestations are particularly likely to pose a risk to the health and life of humans, animals or plants, a risk to property and infrastructure and a risk to the environment more broadly. Proportionate Member State measures introduced to ward off such risks are, as a clear rule, justifiable. We have also found that the precautionary principle in combination with the principle of proportionality provide a sound framework for the lawful introduction of measures addressing slow-onset disasters – while acknowledging that the European Union’s construction of the precautionary principle leaves many open questions as well as a number of doubts. Turning to the question of how a Member State may address slow-onset disasters that take place outside the Member State’s own territory, we find that there is no reason to distinguish situations where a slow-onset disaster may be expected to produce adverse effects directly within the territory of the Member State as compared to the situation where it only indirectly produces such (appreciable) effects. Somewhat more hesitantly, we also find that a Member State may adopt measures addressing slow-onset disasters that are unlikely to produce effects in the Member State’s own territory provided that the measures are supported by the state where the effects will occur. We also find that, exceptionally, such measures may be lawfully adopted even where the state where the slow-onset disaster occurs does not support (or even opposes) the measures. Finally, we consider to what extent a Member State may invoke public international law in support of the introduction of measures that are aimed at slow-onset disasters beyond that Member State’s territory. We find that in theory it may be possible to make such legal argument, but we also find that in practice it is difficult to imagine a situation where this will be relevant.


1 Among scientists there is consensus that the world experiences climate change. To the extent that there is disagreement within the scientific community, this disagreement concerns whether climate change is caused by humans or not (where the vast majority of scientists agree that human agency plays a crucial role in climate change).

2 See e.g. D. B. Wernli et al., ‘The European Approach to Antimicrobial Resistance: Success Stories and Challenges’, Global Health Europe Issue Brief (February 2011).

3 The arguments – and findings – that follow below arguably apply to all four freedoms (i.e. goods, services, persons and capital). However, to keep the text as simple as possible, only goods and services are considered explicitly.

4 Or, if secondary EU regulation applies in the field, we presuppose that the Member States adopt measures to prevent slow-onset disasters from occurring because they consider the existing EU measures addressing the same risk to be insufficient

5 U. Beck, Risikogesellschaft. Auf dem Weg in eine andere Moderne (Suhrkamp 1986). The English edition was published in 1992, U. Beck, Risk Society: Towards a New Modernity (Sage 1992).

6 Anthony Giddens uses the term ‘manufactured risk’ (A. Giddens, ‘Risk and Responsibility’ (1999) MLR 1, 4). In the words of Giddens:

Manufactured risk is risk created by the very progression of human development, especially by the progression of science and technology. Manufactured risk refers to new risk environments for which history provides us with very little previous experience. We often don’t really know what the risks are, let alone how to calculate them accurately in terms of probability tables.

Manufactured risk is expanding in most dimensions of human life. It is associated with a side of science and technology which the early theorists of industrial society by and large did not foresee. Science and technology create as many uncertainties as they dispel – and these uncertainties cannot be ‘solved’ in any simple way by yet further scientific advance. Manufactured uncertainty intrudes directly into personal and social life – it isn’t confined to more collective settings of risk. In a world where one can no longer simply rely on tradition to establish what to do in a given range of contexts, people have to take a more active and risk-infused orientation to their relationships and involvements.

7 In Art. 36 Consolidated version of the Treaty on the Functioning of the European Union [2016] OJ C 202/42, the following objectives are listed: ‘public morality, public policy or public security; the protection of health and life of humans, animals or plants; the protection of national treasures possessing artistic, historic or archæological value; or the protection of industrial and commercial property’. This list does not include the environment, but the Court of Justice of the European Union has ruled that this objective is so essential that it must be included among the weightiest ones (M. Broberg and N. H. Christensen, Free Movement in the European Union (Djøf 2016), 30, but cf. J. Nowag, Environmental Integration in Competition and Free-Movement Laws (Oxford 2016), ch. 10, who takes a significantly more cautious approach).

8 e.g. a non-exhaustive list of protected interests in the framework of the Treaty provisions of free movement, first laid down in 120/78 Rewe-Zentrale v Bundesmonopolverwaltung für Branntwein (Cassis de Dijon), EU:C:1979:42.

9 In order to be lawful, the measures must also be proportionate, however. See further below section IV.

10 Here we are exclusively concerned with measures that restrict intra-EU trade. However, as noted by S. Bogojević, ‘Climate Change Law and Policy in the European Union’, in K. R. Gray et al. (eds.), The Oxford Handbook of International Climate Change Law (Oxford University Press 2016): measures in the form of ambitious climate laws sometimes with extraterritorial effects ‘are motivated not solely by environmental ambition but also by economic concerns to stimulate the internal market and ensure that the EU (green) economy is at the international forefront’.

11 The proportionality principle under EU law has primarily been developed through the case law of the Court of Justice of the European Union. In short, it consists of four cumulative conditions that are known as: (1) suitability, (2) consistency, (3) necessity, (4) proportionality stricto sensu.

12 See e.g. C-282/15 Queisser Pharma, EU:C:2017:26, para. 56, and C-333/08 Commission v France, EU:C:2010:44, para. 92.

13 Two situations may be distinguished here. The first is where there is general agreement among scientist about a given conclusion, while the scientists also generally agree that the risk cannot (yet) be determined with sufficient certainty. The second is where there is real disagreement among recognised scientists regarding the risk. From a legal point of view the latter situation is much more challenging than the former.

14 Originally the precautionary principle was only applied in the environment field. However, today it is clear that it also covers the protection of human, animal and plant health. In contrast, it is less clear whether the precautionary principle can be applied to the protection of property more broadly speaking.

15 As a main rule, the burden of proof weighs on the authorities.

16 Regard being had to legitimate factors such as technical and economic feasibility.

17 C. R. Sunstein, The Laws of Fear: Beyond the Precautionary Principle (Cambridge University Press 2005).

18 Sunstein’s work appeared in 2005 and it is therefore not surprising that he primarily focuses on European Commission, ‘Communication on the Precautionary Principle’ COM(2000)1 final. However, subsequently the EU precautionary principle has undergone important developments, first of all through the European Court of Justice’s case law.

19 Sunstein, Laws of Fear, 122.

20 See further World Health Organization, ‘Global Action Plan on Antimicrobial Resistance’, [accessed 23 August 2017].

21 See further United Nations Framework Convention on Climate Change ‘The Paris Agreement’ (entered into force 4 November 2016).

22 See further Sustainable Development Knowledge Platform, [accessed 23 August 2017].

23 Also, Member State B’s measure constitutes a strong deterrent from importing the specific antibiotics into Member State B.

24 8/74 Procureur du Roi v. Dassonville, EU:C:1974:82, para. 5.

25 This also follows from the proportionality principle.

26 C-180/96 UK v. Commission, EU:C:1998:192 concerned measures aimed at protecting EU citizens against British BSE-infected meat – since it was feared that consuming such meat could lead to the deadly Creutzfeldt–Jakob disease. Since there was a presumption that British beef posed a threat to the consumers, the European Commission prohibited all beef exports from the UK. The UK government challenged this export ban – inter alia on the basis that even if the ban were lawful, it could only apply to exports to other EU Member States, not also to exports to third countries. The Commission argued that a complete export ban was required as otherwise reimports of BSE-infected meat could take place (para. 90). The Council of the European Union (intervening in support of the Commission) added that the European Union could not apply one food safety standard for EU citizens and another (lower) standard for citizens in third countries (para. 44). The European Court of Justice ruled that the total ban was lawful and in justifying this pointed to the risk of reimportation (para. 109), but (arguably) did not reject the argument put forward by the Council.

27 In particular, the measures adopted by the Member State must fulfil both the precautionary principle requirements and the principle of proportionality.

28 See in support of this, C-1/96 World Farming, EU:C:1998:113.

29 If international public law does not impose a duty on the Member State in question, but merely provides ‘persuasive’ support in favour of the measure adopted by the Member State, this does not appear to provide sufficient support for such measure (see C-1/96 World Farming, where the UK government invoked a ‘1988 Recommendation concerning Cattle’, drawn up to apply the principles of the European Convention on the Protection of Animals kept for Farming Purposes in support of restrictions on the export of live calves with a view to preventing those calves from being reared in the veal crate systems used in other Member States that did not comply with the recommendation. Despite the UK government’s being able to back its claim on this international law measure, the European Court of Justice found against the UK government).

Figure 0

Figure 34.1 The four categories of slow-onset disasters

Save book to Kindle

To save this book to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the or variations. ‘’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats

Save book to Dropbox

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Dropbox.

Available formats

Save book to Google Drive

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Google Drive.

Available formats