An Taisce (The National Trust for Ireland), R (on the Application of) v The Secretary of State for Energy and Climate Change & Anor [2014] EWCA Civ 1111

Case No C1/2013/3763

[2014] EWCA Civ 1111

IN THE COURT OF APPEAL (CIVIL DIVISION)

ON APPEAL FROM THE HIGH COURT OF JUSTICE

QUEEN’S BENCH DIVISION (ADMINISTRATIVE COURT)

MRS JUSTICE PATTERSON

CO/5020/2013

Venue Royal Courts of Justice, Strand, London, WC2A 2LL

Date 01/08/2014

Before

LORD JUSTICE LONGMORE

LORD JUSTICE SULLIVAN

LADY JUSTICE GLOSTER

Between

The Queen on the Application of An Taisce (The National Trust for Ireland)Claimant

The Secretary of State for Energy and Climate ChangeDefendant

NNB Generation Company LimitedInterested Party

David Wolfe QC, John Kenny and Blinne Ni Ghralaigh (instructed by Leigh Day Solicitors) for Appellants

Jonathan Swift QC, Rupert Warren QC and Jonathan Moffett (instructed by the Treasury Solicitor) for Respondent

Nathalie Lieven QC and Hereward Phillpot (instructed by Herbert Smith Freehills LLP) for Interested Party

Hearing Hearing dates: 15 th& 16 th July 2014

Judgment

Introduction

[1]In this claim for judicial review the Claimant challenges the decision dated 19th March 2013 of the Defendant to make an Order (“the Order”) granting development consent for the construction of a European pressurised reactor (“EPR”) nuclear power station at Hinkley Point in Somerset (“HPC”).

Background

[2]The background to the claim is explained in considerable detail in the judgment of Patterson J

[2013] EWHC 4161 (Admin)

dismissing the Claimant’s application for permission to apply for judicial review following a “rolled up” hearing. On the 27th March 2014 I granted the Claimant permission to apply for judicial review and ordered that the application should be retained in the Court of Appeal.

[3]The judge set out the factual background in paragraphs 5-62 of her judgment. There was no challenge to this aspect of her judgment, and I gratefully adopt, and will not repeat, all of the detail that is contained in those paragraphs.

[4]There is no dispute as to the legal framework, which the judge set out in paragraphs 63-79 of her judgment. Article 7 of Directive 2011/92/EU (“the EIA Directive”) is of central importance in this claim, and for convenience I set out the material paragraphs:

“1. Where a Member State is aware that a project is likely to have significant effects on the environment in another Member State or where a Member State likely to be significantly affected so requests, the Member State in whose territory the project is intended to be carried out shall send to the affected Member State as soon as possible and no later than when informing its own public, inter alia: (a) a description of the project, together with any available information on its possible transboundary impact; (b) information on the nature of the decision which may be taken. The Member State in whose territory the project is intended to be carried out shall give the other Member State a reasonable time in which to indicate whether it wishes to participate in the environmental decision-making procedures referred to in Article 2(2), and may include the information referred to in paragraph 2 of this Article. 2. If a Member State which receives information pursuant to paragraph 1 indicates that it intends to participate in the environmental decision-making procedures referred to in Article 2(2), the Member State in whose territory the project is intended to be carried out shall, if it has not already done so, send to the affected Member State the information required to be given pursuant to Article 6(2) and made available pursuant to points (a) and (b) of Article 6(3). 3. The Member States concerned, each insofar as it is concerned, shall also: (a) Arrange for the information referred to in paragraphs 1 and 2 to be made available, within a reasonable time, to the authorities referred to in Article 6(1) and the public concerned in the territory of the Member State likely to be significantly affected; and (b) ensure that the authorities referred to in Article 6(1) and the public concerned are given an opportunity, before development consent for the project is granted, to forward their opinion within a reasonable time on the information supplied to the competent authority in the Member State in whose territory the project is intended to be carried out.”

[5]It is common ground that the construction of HPC is a project which falls within Annex I to the EIA Directive. An environmental impact assessment was required and was carried out, and the necessary public consultation was undertaken within the United Kingdom, in accordance with Articles 4-6 of the Directive.

[6]The Defendant did not carry out transboundary consultation in accordance with Article 7 because he did not consider that the HPC project was “likely to have significant effects on the environment in another Member State.” A transboundary screening assessment carried out by the Planning Inspectorate (“PINS”) on the Defendant’s behalf, having referred to Appendix 7E to Volume 1 of the Interested Party’s Environmental Statement, which contained an assessment of potential transboundary effects, said:

“On the basis that licensing and monitoring conditions are effective, impacts will not be significant.”

The screening assessment also said, when dealing the “Probability”:

“The probability of a radiological impact is considered to be low on the basis of the regulatory regimes in place. There could be direct impacts related to the discharge of water during normal operational conditions. However, the discharge of water is expected to be controlled by appropriate licensing conditions and regular monitoring, and hence the probability of any adverse impacts is likely to be low. The Developer has indicated that information is included in the Government’s submission to the European Commission under Article 37 of the Euratom Treaty to show that transboundary impacts from accidents during operation or decommissioning will be so low as to be exempt from regulatory control.”

[7]The Austrian Government wrote to the Department of Energy and Climate Change indicating that it wished to participate in the process of considering the application for the Order. It was sent a copy of the application, and its response included an expert report. The decision letter dated 19th March 2013 summarised the expert report, and the Defendant’s response thereto, in paragraphs 6.6.2(ii) and (iii):

“6.6.2(ii) The expert report focuses on nuclear safety issues and as such has been reviewed by the Office of Nuclear Regulation (“ONR”). It draws heavily on documents published by the ONR during the Generic Design Assessment of the EPR. Although broadly technically sound, it tends to overemphasise the significance of those areas where ONR has in any event determined that more work needs to be done during any subsequent construction and commissioning of a power station based on the EPR (i.e. such as at Hinkley Point) as part of its own regulatory processes. 6.6.2(iii) The Austrian expert contends that in assessing the likely environmental effects of HPC project, I should take into account the effects of very low probability, extreme (or severe) accidents. Effectively the report says that unless it can be demonstrated that a severe accident (involving significant radiological release) cannot occur, then no matter how unlikely it is, I must consider its consequences as part of the development consent process, having regard, in particular, to the possible deleterious effects on Austria. However, in my view such accidents are so unlikely to occur that it would not be reasonable to “scope in” such an issue for environmental impact assessment purposes.”

[8]The Claimant contends that there was a failure to comply with Article 7 of the Directive. The Defendant failed to consult the public in the Republic of Ireland in accordance with Article 7 because:(1) He misdirected himself as to the meaning of “likely” within Article 7 by “scoping out” severe nuclear accidents on the basis that they were very unlikely (Ground 1 “likelihood”); and

(2) Even if he was correct as to the meaning of “likely”, the Defendant erred in relying on the existence of the UK nuclear regulatory regime to fill gaps in current knowledge when reaching his conclusion as to the likelihood of nuclear accidents (Ground 2 “regulatory regime”).

[9]Before considering these two grounds, it is necessary to understand the reference in the decision letter to “very low probability” severe accidents. The Austrian Expert Report had said that severe accidents with high releases of caesium-37 cannot be excluded, and there would be a need for official intervention in Austria after such an accident, but the report recognised that the calculated probability of such an accident is below 1E-7/a, which means that such an accident would not be expected to occur more frequently than once in every 10 million years of reactor operation: see paragraph 53 of Patterson J’s judgment. Discussion Ground 1 Likelihood

Discussion

[10]The words “likely to have significant effects on the environment” occur in a number of places in the EIA Directive: in recitals (7) and (9), in Articles 1(1), 2(1) and 7(1), and in a slightly different formulation – “likely significant effects of the proposed project on the environment” – in Annex IV. In similar vein, an Environmental Statement must include “the data required to identify and assess the main effects which the project is likely to have on the environment”: see Article 5(3).

[11]Two points should be made at the outset of any consideration of what is meant by “likely” in Article 7(1). It is now common ground that:(1) The words “likely to have significant effects on the environment” must have the same meaning throughout the EIA Directive (not least because the environmental information to be supplied to the authorities and the public in the other Member State under Article 7 is the information that must be provided under Article 6 to the public in the Member State in which the project is located); and

(2) Whatever that meaning might be, in the context of the EIA Directive the word “likely” does not mean, as an English lawyer might suppose, more probable than not.

[12]The CJEU has not ruled on the meaning of “likely to have significant effects on the environment” in the EIA Directive. The Domestic authorities were considered by Patterson J in paragraphs 123-126 of her judgment. None of those authorities is binding on this Court. In R (Morge) v Hampshire County Council

[2010] EWCA Civ 608

[2010] PTSR 1882

, Ward LJ recorded the parties’ agreement that “likely” connotes real risk and not probability (paragraph 80). In R(Bateman) v South Cambridgshire District Council

[2011] EWCA Civ 157

Moore-Bick LJ expressed the view in paragraph 17 that “something more than a bare possibility is probably required, though any serious possibility would suffice”, but he did not find it necessary to reach a final decision on the question (paragraph 19).

[13]The Claimant’s submission that a project is “likely to have significant effects on the environment” if such effects “cannot be excluded on the basis of objective evidence” is founded on the decision of the Grand Chamber of the CJEU in

Case C-127/02

Landelijke Vereniging tot Behoud van de Waddenzee and Nederlandse Vereniging tot Bescherming van Bogels v Staatssecretaris van Landbouw, Natuurbeheer en Visserij (“Waddenzee”). Waddenzee was concerned with the Habitats Directive, Article 6 of which materially provides:

“1. For special areas of conservation, Member States shall establish the necessary conservation measures involving, if need be, appropriate management plans specifically designed for the sites or integrated into other development plans, and appropriate statutory, administrative or contractual measures which correspond to the ecological requirements of the natural habitat types in Annex 1 and the species in Annex II present on the sites. 2. Member States shall take appropriate steps to avoid, in the special areas of conservation, the deterioration of natural habitats and the habitats of species as well as disturbance of the species for which the areas have been designated, in so far as such disturbance could be significant in relation to the objectives of this Directive. 3. Any plan or project not directly connected with or necessary to the management of the site but likely to have a significant effect theron, either individually or in combination with other plans or projects, shall be subject to appropriate assessment of its implications for the site in view of the site’s conservation objectives. In the light of the conclusions of the assessment of the implications for the site and subject to the provisions of paragraph 4, the competent national authorities shall agree to the plan or project only after having ascertained that it will not adversely affect the integrity of the site concerned and, if appropriate, after having obtained the opinion of the general public.”

(emphasis added)

[14]In paragraphs 42-44 of its judgment the Grand Chamber said:

“42. As regards Article 2(1) of Directive 85/337 [now Article 2(1) of the EIA Directive], the text of which, essentially similar to Article 6(3) of the Habitats Directive, provides that “Member States shall adopt all measures necessary to ensure that, before consent is given, projects likely to have significant effects on the environment … are made subject to an assessment with regard to their effects’, the Court has held that these are projects which are likely to have significant effects on the environment (see to that effect Case C-117/02 Commission v Portugal [2004] ECR I-5517, paragraph 85). 43. It follows that the first sentence of Article 6(3) of the Habitats Directive subordinates the requirement for an appropriate assessment of the implications of a plan or project to the condition that there be a probability or a risk that the latter will have significant effects on the site concerned. 44. In the light, in particular, of the precautionary principle, which is one of the foundations of the high level of protection pursued by Community policy on the environment, in accordance with the first subparagraph of Article 174(2) EC, and by reference to which the Habitats Directive must be interpreted, such a risk exists if it cannot be excluded on the basis of objective information that the plan or project will have significant effects on the site concerned (see, by analogy, inter alia Case C-180/96 United Kingdom v Commission [1998] ECR I-2265, paragraphs 50, 105 and 107). Such an interpretation of the condition to which the assessment of the implications of a plan or project for a specific site is subject, which implies that in case of doubt as to the absence of significant effects such an assessment must be carried out, makes it possible to ensure effectively that plans or projects which adversely affect the integrity of the site concerned are not authorised, and thereby contributes to achieving, in accordance with the third recital in the preamble to the Habitats Directive and Article 2(1) thereof, its main aim, namely, ensuring biodiversity through the conservation of natural habitats and of wild fauna and flora. 45. In the light of the foregoing, the answer to Question 3(a) must be that the first sentence of Article 6(3) of the Habitats Directive must be interpreted as meaning that any plan or project not directly connected with or necessary to the management of the site is to be subject to an appropriate assessment of its implications for the site in view of the site’s conservation objectives if it cannot be excluded, on the basis of objective information, that it will have a significant effect on that site, either individually or in combination with other plans or projects.”

(emphasis added)

[15]On behalf of the Claimant, Mr. Wolfe QC, understandably, placed much emphasis upon the Grand Chamber’s interpretation of the “essentially similar” text of Article 6(3) of the Habitats Directive; and the fact that the Grand Chamber had drawn an analogy with the judgment in the United Kingdom case in which the Court was considering the meaning of likelihood in a very different context: the United Kingdom’s response to the BSE crisis, and a Directive which required notification of “any zoonoses, diseases or other cause likely to constitute a serious hazard to animals or to human health.” This demonstrated, he submitted, that the Grand Chamber’s approach to the likelihood of significant harm in any context where environmental concerns, including the protection of human health, were in issue was based on first principles, and was not confined to the specific characteristics of the Habitats Directive.

[16]While the text of Article 2(1) of the EIA Directive and Article 6(3) of the Habitats Directive is essentially similar, and both Directives are concerned with environmental protection, there is in my view a clear distinction between the two Directives. The scope of the EIA Directive is wide ranging, it ensures that any project which is likely to have significant effects on the environment is subject to a process of environmental impact assessment. The EIA Directive does not prescribe what decision must be taken by the competent authority – to permit or to refuse – if the environmental impact assessment concludes that the proposal is likely to have significant effects on the environment. The Habitats Directive is more focussed, it protects particular areas of Community importance, which have been defined as “special areas of conservation”, and which must be maintained at, or restored to, “favourable conservation status”: see Articles 2 and 3. In order to achieve this aim Article 6(3) provides that, subject only to “imperative reasons of overriding public interest” (see Article 6(4)), where there has been an “appropriate assessment”:

“the competent authorities shall agree to the plan or project only after having ascertained that it will not adversely affect theintegrity of the site concerned.”

(emphasis added)

[17]Thus, where there has been an “appropriate assessment” Article 6(3) imposes a very strict test for approval. The Grand Chamber said that competent authorities may approve a plan or project:

“55…… only after having made sure that it will not adversely affect the integrity of the site. 56 It is therefore apparent that the plan or project in question may be granted authorisation only on the condition that the competent national authorities are convinced that it will not adversely affect the integrity of the site concerned. 57 So, where doubt remains as to the absence of adverse effects on the integrity of the site linked to the plan or project being considered, the competent authority will have to refuse authorisation. 58 In this respect it is clear that the authorisation criterion laid down in the second sentence of Article 6(3) of the Habitats Directive integrates the precautionary principle (see case C-157/96 National Farmers’ Union and Others [1998] ECR I-2211. paragraph 63) and makes it possible effectively to prevent adverse effects on the integrity of protected sites as the result of the plans or projects being considered. A less stringent authorisation criterion than that in question could not as effectively ensure the fulfilment of the objective of site protection intended under that provision. 59 Therefore, pursuant to Article 6(3) of the Habitats Directive, the competent national authorities, taking account of the conclusions of the appropriate assessment of the implications of mechanical cockle fishing for the site concerned, in the light of the site’s conservation objectives, are to authorise such activity only if they have made certain that it will not adversely affect the integrity of that site. That is the case where no reasonable scientific doubt remains as to the absence of such effects (see, by analogy, Case C-236/01 Monsanto Agricoltura Italia and Others [2003] ECR I-8105, paragraphs 106 and 113).”

(emphasis added)

[18]In order to achieve this very high level of protection for special areas of conservation an equally stringent approach is required at the screening stage when the competent authority is deciding whether an “appropriate assessment” is required: see paragraph 70 of the Opinion of Advocate General Kokott [2004] ECR 1-7405. It is for this reason that in a case falling within the Habitats Directive an “appropriate assessment” must be carried out unless the risk of significant effects on the site concerned can be “excluded on the basis of objective information.” Reading the Waddenzee judgment as a whole, it is clear that significant effects can be excluded on the basis of objective evidence if “no reasonable scientific doubt remains as to the absence of such effects.”

[19]Standing back from a detailed analysis of the text of the two Directives, there is no obvious reason why such a strict approach should apply to the screening stage in the EIA Directive, which merely seeks to ensure that any likely significant effects on the environment are identified and properly taken into account in the decision making process. Even if significant environmental effects are identified, and are not merely likely, but are certain to occur, the EIA Directive does not require that approval for an EIA project within either Annex I or II of the EIA Directive must be refused in the absence of some overriding public interest. The Grand Chamber referred to the precautionary principle in Waddenzee (see paragraph 44), but it was applying that principle in the context of the Habitats Directive, where the objective is the protection of the integrity of particular sites designated for their conservation importance. In the wider context of environmental protection a “real risk” test embodies the precautionary principle: see Evans v Secretary of State for Communities and Local Government

[2013] EWCA Civ 114

, per Beatson LJ at paragraph 21.

[20]I have already mentioned the fact that, by contrast with the Habitats Directive, the EIA Directive has a broad scope: it applies to all “projects which are likely to have significant effects on the environment” (Article 1); and the Environmental Statements prepared for all such projects must include information about all of the likely significant effects (Article 5), and must be subject to public consultation (Article 6). While the claimant stresses the need for any likely environmental effect to be “significant”, it seems to me that adopting the Claimant’s approach to the meaning of likelihood – that a significant environmental effect is “likely” if it cannot be excluded on the basis of objective evidence – would inevitably have the effect of both(a) materially increasing the number of projects within Annex II which would have to be the subject of an EIA; and

(b) increasing the number of “likely” significant effects that would have to be included in all Environmental Statements, and consulted upon.

[21]Many Environmental Statements for major projects which are now prepared on a “real risk” basis are already very lengthy. If, in addition to being required for more Annex II projects, Environmental Statements had to deal with every possible significant environmental effect, however unlikely, unless it could be excluded on the basis of objective evidence, there is a real danger that both the public when consulted and decision takers would “lose the wood for the trees”, thereby causing the EIA process to become less effective as an aid to good environmental decision making: see R (Loader) v Secretary of State for Communities and Local Government

[2012] EWCA Civ 869

[2012] 3 CMLR 29

, per Pill LJ at paragraph 46; and Bateman per Moore-Bick LJ at paragraph 19.

[22]In addition to these wider policy considerations, it is necessary to consider the text of the EIA Directive as a whole. I accept the submission of Mr. Swift QC on behalf of the Defendant that the Claimant’s approach to likelihood is inconsistent with the selection criteria that are set out in Annex III, which must be taken into account when a decision is being taken as to whether an Annex II project shall be made subject to an environmental impact assessment, ie. whether it is likely to have significant effects on the environment. The selection criteria include “Characteristics of the Potential Impact”. The potential significant effects of projects must be considered in relation to the criteria set out in points 1 and 2 [the characteristics and the location of projects] and having regard in particular to: “(a) the extent of the impact (geographical area and size of the affected population);” (b) the transfrontier nature of the impact: (c) the magnitude and complexity of the impact; (d) the probability of the impact; (e) the duration, frequency and reversibility of the impact”. (emphasis added) Mr Swift submits, rightly in my view, that the need to have regard to “the probability of the impact” would be redundant if the test of likelihood was whether the risk of any impact, however improbable, could be excluded on the basis of objective evidence.

[23]For these reasons, I consider that the differences between the scope, purpose and text of the two environmental Directives are such that it is unduly simplistic to say that, because one part of the text in both Directives is “essentially similar”, the meaning of that part of the text in the context of Article 6(3) of the Habitats Directive as determined by the Grand Chamber in Waddenzee can simply be carried over into the EIA Directive. The “real risk” test adopted in the domestic authorities (above) incorporates the protective principle in the context of the EIA Directive.

[24]Mr Wolfe submitted that even if we were minded to conclude that the Defendant had not erred in his approach to likelihood for the purposes of Article 7, a reference to the CJEU was required because this Court could not be convinced that applying the “real risk” test in the context of the EIA Directive would be correct as a matter of EU law: see CILFIT (Srl) v Ministry of Health [1982] ECR 1-3415 at paragraphs 16-20. In support of that submission he relied, in addition to the Grand Chamber’s judgment in Waddenzee (above), upon five considerations, as follows:(a) the German text of Article 7(1);

(b) the Russian text of the Convention on Environmental Impact Assessment in a Transboundary context, (“the Espoo Convention”);

(d) the Aarhus Convention; and

(e) Directive 2001/42/EC on the assessment of the effects of certain plans and programmes on the environment (“the SEA Directive”).

[25]While both (a) and (b) support the proposition that “likely” in Article 7(1) has a broader meaning than “more likely than not”, they do not support the Claimant’s proposition that “likely” in Article 7(1) means “cannot be excluded no matter how unlikely.” In Waddenzee [2004] ECR 1-7405 Advocate General Kokott explained in paragraph 69 of her opinion:

“As regards the degree of probability of significant adverse effect, the wording of various language versions is not unequivocal. The German version appears to be the broadest since it uses the subjunctive “könntė (could). This indicates that the relevant criterion is the mere possibility of an adverse effect. On the other hand, the English version uses what is probably the narrowest term, namely “likely”, which would suggest a strong possibility. The other language versions appear to lie somewhere between these two poles. Therefore, according to the wording it is not necessary that an adverse effect will certainly occur but that the necessary degree of probability remains unclear.”

[26]There is no dispute that Article 7 of the EIA Directive gives effect to the Espoo Convention: see recital (15) to the EIA Directive. The English language version of the Convention uses the word “likely”. The Claimant obtained a translation of the Russian version of the Espoo Convention (of which there are three authentic texts, English, French and Russian). The translator states that the word “may” in the expression “may cause a significant adverse transboundary impact”, “fails to convey the meaning of likelihood and expresses a mere possibility which can be either high or low.” In a further statement, the translator explains that the Russian word for “may” “includes something which cannot be excluded or ruled out.” It seems that the Russian word for “may” conveys a flexible concept of possibility which ranges from a high possibility at one end of the spectrum to a possibility which cannot be excluded. As with the German text of the EIA Directive, the Russian text would not constrain the CJEU to adopt the lowest level of possibility inherent in the Russian version of the Espoo Convention. I will deal with the view expressed by the Implementation Committee after I have considered whether any assistance can be obtained from the Aarhus Convention and the SEA Directive.

[27]There is no dispute that the EIA Directive must be construed so as to give effect to the Aarhus Convention. Recital (20) to the EIA Directive records the fact that:

“Article 6 of the Aarhus Convention provides for public consultation in decisions on the specific activities listed in Annex I thereto and on activities not so listed which may have a significant effect on the environment.”

(emphasis added) In broad terms, Annex I to Aarhus lists the kinds of projects that are listed in Annex I to the EIA Directive, while Annex II projects in the EIA Directive may fall within the second part of Article 6(1) of Aarhus. While the word “may” indicates a lower threshold than “likely” (used in the sense of more likely than not), it does not indicate that the test for public consultation across the board – for all activities which may have a significant effect on the environment – is so low as to include any activity where a significant effect on the environment, however unlikely, cannot be excluded.

[28]Article 3(2) of the SEA Directive requires an environmental assessment for all plans and programmes(a) which are prepared for certain purposes and which set the framework for future development consent of projects listed in Annexes I and II to the EIA Directive; and

(b) “which in view of the likely effect on sites [special areas of conservation] have been determined to require an [appropriate] assessment pursuant to Article 6 or 7 of [the Habitats Directive].” In the latter case, the CJEU has held that an environmental assessment is required if a significant effect on the site cannot be excluded: see

Case C-177/11

Syllogos Ellinon Poleodomon kai Khorotakton v Y pourgos Perivallontos, Khorotaxias & Dimosion Ergon and Others. This decision of the CJEU merely applies the Waddenzee approach to plans or programmes which are likely to have a significant effect on sites of Community importance, which have been designated as special areas of conservation by the Member States: see paragraphs 19-23 of the judgment. It does not address the issue in the present case: whether the Waddenzee approach to likelihood should be carried over into the EIA Directive.

[29]For these reasons, I am not persuaded that any of these considerations assists the Claimant’s case. Against this background, I turn to the views expressed by the Implementation Committee (“the Committee”). The judge dealt with this issue in paragraphs 132-142 of her judgment. In summary, the Claimant had relied upon the endorsement by the Parties to the Espoo Convention at their Fourth Meeting of the findings of the Committee in Annex I that Ukraine had not complied with the Convention in, what for convenience I will call the “Danube Black Sea” case. In paragraph 54 in Part III of the Committee’s report “Consideration and Evaluation”, preceding its “Findings” in Part IV, the Committee said:

“Article 3, paragraph 1. of the Convention stipulates that Parties shall notify any Party of a proposed activity listed in Appendix I that is likely to cause a significant adverse transboundary impact. The Committee is of the opinion that, while the Convention’s primary aim, as stipulated in Article 2, paragraph 1, is to “prevent, reduce and control significant adverse transboundary environmental impact from proposed activities”, even a low likelihood of such an impact should trigger the obligation to notify affected Parties in accordance with Article 3. This would be in accordance with the Guidance on the Practical Application of the Espoo Convention, paragraph 28, as endorsed by decision III/4 (ECE/MP.EIA/6 annex IV). This means that notification is always necessary, unless significant adverse transboundary impact can be excluded with certainty. This interpretation is based on the precautionary and prevention principles.”

(emphasis added)”

[30]The judge concluded that the Meeting of the Parties was not purporting to determine the legal position under the Convention, but was setting out a pragmatic approach for the parties to follow, and also said that the Committee had no status to give a legal ruling: see paragraph 135 of the judgment. At the Fourth Meeting, the Parties also asked the Committee “To promote and support compliance with the Convention including to provide assistance in this respect, as necessary.” In response to that request the Committee published its Opinions, as expressed in the reports of its sessions, from 2001 to 2010. Those Opinions included its views expressed in paragraph 54 of Annex 1 to decision IV/2 (above).

[31]In 2013 the European Commission published “Guidance on the Application of the Environmental Impact Assessment Procedure for large-scale Transboundary Projects.” Under the heading “Need for notification” the Commission’s guidance says:

“The Espoo Convention requires that the Party of origin notifies affected Parties about projects listed in Appendix 1 and likely to cause a significant adverse transboundary impact (Article 3(2)). The notification triggers the transboundary EIA procedure. The Espoo Convention’s primary aim is to ‘prevent reduce and control significant adverse transboundary environmental impact from proposed activities’ (Article 2(1), but in fact the Party of origin is obliged to notify affected Parties (in accordance with Article 3 of the Espoo Convention) even if there is only a low likelihood of such impact. This means that notification is always necessary, unless significantadverse transboundary impact can be excluded with certainty.17 This interpretation is based on the precautionary and prevention principles.”

(emphasis added) Footnote 17 cross-refers to paragraph 54 of decision IV/2 (above).

[32]As I explained when granting permission to appeal,

[2014] EWCA Civ 666

, the Chair of the Committee wrote a letter dated 14th March 2004 to the United Kingdom Government. The Committee had requested a copy of Patterson J’s judgment, and had considered the matter between 25th and 27th February 2014 at its 30th session held in Geneva. The Committee’s letter dated 14th March 2014 expressly endorsed the view that it had expressed in the Danube Black Sea case, as to the circumstances in which transboundary consultation was required by the Convention:

“This means that notification is necessary unless a significant adverse transboundary impact can be excluded (decision IV/2, annex I paragraph 54)” 33. The letter continued: “On the above grounds, the Committee found that there was a profound suspicion on non-compliance and decided to begin a Committee initiative further to paragraph 6 of the Committee’s structure and functions. In line with paragraph 9 of the Committee’s structure and functions, the Committee decided that the United Kingdom should be invited to the Committee’s thirty-second session (9-11 December 2014) to participate in the discussion an to present information and opinions on the matter under consideration.”

[33]Having read the Committee’s letter, I was satisfied that there was a compelling reason for granting permission to appeal. There was a need for this Court to decide whether it was possible to give a definitive ruling as to the approach to likelihood in the EIA Directive, or whether there should be a reference of that question to the CJEU. I have explained in paragraphs 16-23 (above) why I consider that the Defendant was not required to apply the Waddenzee approach to the likelihood of significant transboundary environmental effects under Article 7 of the EIA Directive. This is not a court of final appeal. If we had to apply CILFIT I could not say that I was convinced that the other Member States and the CJEU would necessarily conclude that the “real risk” approach is the correct approach to the likelihood of significant effects on the environment for the purposes of the EIA Directive. Does this mean that a reference to the CJEU is necessary for the purpose of deciding this claim?

[34]Mr. Swift acknowledged that the threshold for the likelihood of significant effects on the environment for the purposes of the EIA Directive is a very important issue, with EU-wide implications. However, both he and Miss Lieven QC on behalf of the Interested Party submitted that a reference to the CJEU was not necessary for the purpose of determining this claim for judicial review, because no matter how low the threshold for a likely significant effect on the environment might be set by the CJEU, the Defendant’s decision dated 19th March 2013 would still be lawful.

[35]I accept that submission. There is an artificiality in the Claimant’s claim. The Defendant was not writing an academic dissertation on the concept of likelihood in the EIA Directive, he was deciding whether to grant development consent for a particular project: the construction of an EPR nuclear power station, HPC. In its submissions, the Claimant posited a stark contrast between the “real risk” and the “cannot be excluded on the basis of objective information”, approaches, to the issue of likelihood in the EIA Directive. The distinction between these two approaches to likelihood is clear as a matter of abstract legal analysis, but the Defendant, unsurprisingly in the context of a proposal for the construction of a nuclear power station, did not purport to apply a “real risk” approach. The disagreement between the approach adopted by the Defendant and the approach advocated in the Austrian expert report was not a disagreement as to whether the “real risk” approach or the “cannot be excluded on the basis of objective evidence” approach should be applied to the risk of a serious nuclear accident. It was a disagreement as to the point at which the significant environmental effects of a severe nuclear accident could properly be “excluded on the basis of objective evidence.” Was that point reached only when it had been demonstrated that the probability of such a severe accident was zero; or was the Defendant entitled to conclude that that point had been reached in this case because the probability of a severe accident was very remote indeed – in circumstances where the Austrian expert report had calculated the probability of such an accident to be as low as 1 in 10 million years of reactor operation?

[36]The true nature of the dispute in this case – whether the exclusion of a significant environmental effect from the EIA process is permissible only if it has been demonstrated that there is no risk whatsoever of it occurring, or if exclusion is permissible where it has been demonstrated that the risk is extremely remote – emerges most clearly from the response of the Department of Energy and Climate to the letter dated 14th March 2014 from the Espoo Implementation Committee (paragraph 32 above). In its letter dated 19th June 2014 the Department maintained that the present case was very different from the Danube Black Sea case in which there was no doubt that the Convention was engaged:

“On any analysis, the risk of an accident occurring from the proposed new nuclear development at Hinkley Point C is extremely low. Given the very remote nature of the risk, it is difficult to quantify, and the estimates produced will depend to some extent on the accident scenarios considered. However, the literature on this issue is summarised in the European Commission’s 2005 Report ‘Externe – The Externalities of Energy, Methodology 2005 Update’, which points to a probability of major accidents (core meltdown plus containment failure) in the UK of 4x10-9 . This suggests that the potential for a major accident in the UK – the meltdown of the reactor’s core along with failure of the containment structure – is one in 2.4 billion per reactor year; by comparison, it is thought that the risks of a meteorite over a kilometre hitting the earth, which could have significant global environmental impacts, could be one in 0.5 million per year. The Austrian Government also commissioned its own expert analysis of the risks of an accident from a new nuclear development at Hinkley Point C, which expressed the risk of an accident as being not expected to occur more frequently than once in every 10 million years of reactor operation. On no natural understanding of the term could such a remote risk be considered be constitute a ‘likely significant effect’.”

[37]The Claimant’s challenge to the Defendant’s decision in this case does not simply depend upon the proposition that the Grand Chamber’s approach in Waddenzee to the meaning of “likely to have a significant effect” in the Habitats Directive should be carried over into the EIA Directive, it also depends upon a very literal meaning being given to the Grand Chamber’s words “cannot be excluded on the basis of objective information” in its judgment in Waddenzee. If a remote risk can properly be excluded, the Claimant does not challenge the Defendant’s assessment that the remoteness of the risk in this case was such that it could be excluded. In order to succeed in this claim the Claimant has to establish that any risk, no matter how remote, cannot be excluded unless it has been demonstrated that there is no possibility of its occurring. It is, in effect a “zero risk” approach to the likelihood of significant environmental effects.

[38]It would be surprising if the Grand Chamber had intended to impose such a high and inflexible threshold for “appropriate assessment”, even in the context of the Habitats Directive. However purposive the interpretation of the Habitats Directive, its text cannot be ignored. The word “likely”, and the concept of likelihood, implies at least some degree of flexibility. There comes a point when the probability (to use the word in Annex III to the EIA Directive) of a significant effect is so remote that it ceases to be “likely”, however broad the concept of likelihood. In Waddenzee the Grand Chamber said that, following an appropriate assessment, a project could be authorised only if the competent authority “have made certain that it will not adversely affect the integrity of that site. That is the case where no reasonable scientific doubt remains as to the absence of such effects….” (see paragraph 17 above). Thus, certainty was equated with the absence of reasonable scientific doubt.

[39]Even if the Waddenzee approach to likelihood is carried over into the EIA Directive, it must be open to a competent authority to conclude that the risk of a significant adverse effect on the environment is so remote (eg if it is more remote than the risk of a meteorite of over a kilometre hitting the earth) that there is “no reasonable scientific doubt” as to the absence of that adverse effect for the purpose of the EIA Directive. The competent authority does not have to be satisfied that there is no risk, however remote, that a severe nuclear accident will occur in order to be satisfied that there is “no reasonable scientific doubt” that such an accident will not occur. This approach is consistent with the guidance that is contained in the Planning Inspectorate’s Advice note 12: Development with significant transboundary impacts consultation.

[40]I do not accept Mr. Wolfe’s submission that the Defendant failed to follow this advice from the Planning Inspectorate. When dealing with “Screening”, and with those cases in which it is necessary for the Secretary of State to determine whether or not a proposed development is likely to have significant effects on the environment in another EEA State, the Advice note say this:

“In reaching a view, the precautionary approach will be applied and following the court’s reasoning in the Waddenzee case such that ‘likely to have significant effects’ will be taken as meaning that there is a probability or risk that the development will have an effect, and not that a development will definitely have an effect…” 42. Mr. Wolfe emphasised the reference to the CJEU’s reasoning in Waddenzee; but the Advice note continues: “As a rule of thumb (taking the precautionary approach), unless there is compelling evidence to suggest otherwise, it is likely that the Planning Inspectorate may consider the following [Nationally Significant Infrastructure Projects] as likely to have significant transboundary impacts: • nuclear power stations; and • off-shore generating stations in a Renewable Energy Zone.”

I accept Mr. Swift’s submission that evidence that the risk of a severe nuclear accident is not merely unlikely, but extremely remote, is capable of being “compelling evidence” that a proposed nuclear power station is not likely to have significant transboundary effects, since it is common ground that such effects would be likely to occur only if there was such an accident.

[41]The contrast between the evidential basis for the low level of risk in the present case and the extent of the scientific uncertainty in the United Kingdom case to which the CJEU referred by way of analogy in its judgment in Waddenzee (see paragraph 14 above) is instructive. In the United Kingdom case the Spongiform Encephalopathy Advisory Committee (“SEAC”) had said that “it was not in a position to confirm whether or not there was a causal link between BSE and the recently discovered variant of Creutzfeldt-Jacob disease, a question which required further scientific research” (paragraph 14). A similar position had been adopted by the Scientific Veterinary Committee of the European Union: while it was not possible on the available data to prove that BSE was transmissible to humans, in view of the possibility of such transmission, which the committee had always considered, it had recommended certain precautionary measures and that research on the question of transmissibility of BSE to humans be continued (paragraph 13). The recitals to the Directive that was challenged by the United Kingdom reflected the extent of the scientific uncertainty:

“Whereas under current circumstances, a definitive stance on the transmissibility of BSE to humans is not possible; whereas a risk of transmission cannot be excluded; whereas the resulting uncertainty has created serious concern among consumers; ... ”

[42]In the present case, it is common ground that the probability of a severe nuclear accident is very low indeed. There may be an issue as to just how low that probability is (see the correspondence with the Implementation Committee, paragraph 36 above) but there is no doubt that the Defendant was entitled to describe it in his decision as a “very low probability”. The issue, therefore, is whether the risk of a significant effect on the environment can properly be excluded on the basis of a very low probability, or only upon the basis of a zero probability. In this case we are concerned with a proposal for a nuclear power station, and the environmental consequences of a severe nuclear accident. In that context, for obvious reasons, “very low probability” means very low probability indeed, far below the levels of probability (or “risk”) that might be regarded as acceptable in the context of other developments. Although Annex I to the EIA Directive includes other inherently dangerous projects, eg chemical installations for the production of explosives, where only the remotest of risks will be acceptable, the Directive covers a very wide range of projects in Annexes I and II. In the context of very many, if not most, of the projects listed in the Directive, it is difficult to see how it could seriously be contended that a significant effect on the environment which would not be expected to occur more frequently than once in every 10 million years could not properly be excluded from environmental impact assessment on the basis of objective information.

[43]Annex III requires the Member States to consider both the magnitude and complexity of an environmental impact and the probability of such an impact when deciding whether an Annex II project is likely to have significant effect on the environment (see paragraph 22 above). As a matter of common sense, the greater the potential impact, the lower will be the level of probability at which the competent authority will decide that it should be subjected to the environmental impact assessment process: see Miller v North Yorkshire County Council,

[2009] EWHC 2172 (Admin)

per Hickinbottom J at paragraphs 31 and 32. This leaves an area of judgment for the competent authority – balancing the severity of any potential environmental harm against the probability of it occurring. It recognises the fact that some significant effects on the environment, eg a significant radiological impact, are much more significant than others. Given the wide range of projects covered by the EIA Directive and the express requirement to consider the probability of any impact, I am satisfied that, even if it is appropriate to apply the “cannot be excluded on the basis of objective evidence” approach to the likelihood of significant effects on the environment in the EIA Directive, there is no realistic prospect of the Claimant’s “zero risk” approach being adopted by the CJEU. I would add that our attention was not drawn to any decision of a Court in which the Claimant’s approach to exclusion has been adopted. However purposive the interpretation of the EIA Directive, a “zero risk” approach to likelihood would be an interpretative step too far and would frustrate, rather than further the purpose of the Directive.

[44]In reaching that conclusion, I have not ignored the views expressed by the Committee in its letter dated 14th March 2014. They provide the only possible support for a “zero risk” approach to the point at which a serious environmental impact may be excluded from the EIA process. While I respect the Committee’s view, it is not the function of the Committee to give an authoritative legal interpretation of the Convention. The correspondence with the Committee makes it clear that there is a dispute as to the proper interpretation of the Convention. Article 15 makes provision for the settlement of such disputes. If the dispute cannot be resolved by negotiation between the Parties it may be either submitted to the International Court of Justice, or referred to arbitration in accordance with the procedure set out in Appendix VII to the Convention.

[45]The Committee does have an important role in promoting best practice under the Convention, and it is noteworthy that its conclusion in paragraph 54 of Annex I to decision IV/2 - that even a low likelihood of a significant adverse transboundary environmental impact would trigger the obligation to notify affected parties in accordance with

Article 3 of the Convention

[Article 7 of the EIA Directive] - is expressly based upon its “Guidance on the Practical Application of the Espoo Convention”, as endorsed by decision III/4. Thus, it would appear that the views expressed by the Committee are based upon a combination of its advice as to what would be best practice, and its view as to what is the legal position, under the Convention. I intend no criticism of the Committee when I say that, insofar as its decision in paragraph 54 of Annex I to decision IV/2 moves from advice as to what would be best practice to a statement of what the legal position is, it is not based upon any legal analysis (that is not surprising, the Committee is not a legally qualified body). Even if a “low likelihood” of a significant transboundary effect not merely should (as a matter of good practice), but does (as a matter of law) trigger the obligation to notify any affected party, the Committee will still have to consider the issue raised in this case: whether a “likelihood” may be so very low that it can be excluded for the purpose of transboundary consultation, or whether exclusion is permissible only when all risk has been eliminated. Of critical importance for present purposes, the Committee understandably focuses simply upon the terms of the Espoo Convention, and does not consider the need for the words “likely to have significant effects on the environment” to have a consistent meaning throughout the EIA Directive. For these reasons, the views expressed by the Committee in its letter dated 14th March 2014 do not persuade me that it is necessary for this Court to make a reference to the CJEU in order to determine this claim.

Ground 2

[46]The judge dealt with this issue in paragraphs 177-193 of her judgment. She concluded in paragraph 193:

“In my judgment there is no reason that precludes the Secretary of State from being able to have regard to, and rely upon, the existence of a stringently operated regulatory regime for future control. Because of its existence, he was satisfied, on a reasonable basis, that he had sufficient information to enable him to come to a final decision on the development consent application. In short, the Secretary of State had sufficient information at the time of making his decision to amount to a comprehensive assessment for the purposes of the Directive. The fact that there were some matters still to be determined by other regulatory bodies does not affect that finding. Those matters outstanding were within the expertise and jurisdiction of the relevant regulatory bodies which the defendant was entitled to rely upon.”

I agree with the judge. Had this ground of challenge stood alone I would not have granted the Claimant permission to apply for judicial review.

[47]There is no dispute that the Defendant was in principle entitled to have regard to the UK nuclear regulatory regime when reaching a conclusion as to the likelihood of nuclear accidents: see Gateshead Metropolitan Council v Secretary of State for the Environment [1995] Env LR 37.

[48]Many major developments, particularly the kind of projects that are listed in Annex I to the EIA Directive, are not designed to the last detail at the environmental impact assessment stage. There will, almost inevitably in any major project, be gaps and uncertainties as to the detail, and the competent authority will have to form a judgement as to whether those gaps and uncertainties mean that there is a likelihood of significant environmental effects, or whether there is no such likelihood because it can be confident that the remaining details will be addressed in the relevant regulatory regime. In paragraph 38 of his judgment in R (Jones) v Mansfield District Council

[2004] 2 P & CR 14

, Dyson LJ (as he then was) adopted paragraphs 51 and 52 of the judgment of Richards J (as he then was) which included the following passage:

“It is for the authority to judge whether a development would be likely to have significant effects. The authority must make an informed judgment, on the basis of the information available to it and having regard to any gaps in that information and to any uncertainties that may exist, as to the likelihood of significant environmental effects. Everything depends on the circumstances of the individual case.”

[49]This is precisely what happened on the facts of the present case. The elaborate regulatory regime for nuclear power stations is described in the Witness Statements filed on behalf of the Defendant and the Interested Party. For present purposes, it is sufficient to note that by the time the Defendant made his decision dated 19th March 2013 the Office for Nuclear Regulation (“ONR”) had issued a nuclear site licence, and both the ONR and the Environment Agency had completed the Generic Design Assessment (GDA) process, including a severe accident analysis, for the EPR, the type of reactor to be used at HPC. All of the GDA issues had been addressed, and the ONR had issued a Design Acceptance Confirmation (“DAC”). The ONR had said that it was confident that the design was “capable of being built and operated in the UK, on a site bounded by the generic site envelope, in a way that is safe and secure”. Site specific matters not covered by the GDA process would still need to be considered, but the ONR was confident that they could, and would, be addressed under the site licence conditions. As the ONR explained:

“Whilst the GDA process, leading to the issue of a DAC, is not part of the licensing assessment, the successful completion of GDA does provide confidence that ONR will be able to give permission for the construction, commissioning and operation of a nuclear power station based on that generic design.”

[50]In view of this factual background, it might be thought that this case was the paradigm of a case in which a planning decision-taker could reasonably conclude that there was no likelihood of significant environmental effects because any remaining gaps in the details of the project would be addressed by the relevant regulatory regime. Undaunted, Mr. Wolfe submitted that there was a distinction between reliance upon a pollution regulator applying controls “which it has already identified in the light of assessments which it has already undertaken on the basis of a scheme which has already been designed”, which he said was permissible, and reliance upon “current” gaps in knowledge “being filled by the fact of the existence of the pollution regulator [who] will make future assessments… on elements of the project still subject to design changes….”, which was not.

[51]There is no basis for this distinction, which is both unrealistic and unsupported by any authority. The distinction is unrealistic because elements of many major development projects, particularly the kind of projects within Annex I to the EIA Directive, will still be subject to design changes, and applying Mr. Wolfe’s approach those projects will not have “already been designed” at the time when an environmental impact has to be carried out. The detailed design of many Annex I projects, in particular nuclear power stations, is an immensely complex, lengthy and expensive process. To require the elimination of the prospect of all design changes before the environmental assessment of major projects could proceed would be self-defeating. The promoters of such projects would be unlikely to incur the, in some cases, very considerable expense, not to mention delay, in resolving all the outstanding design issues, without the assurance of a planning permission. If the environmental impact assessment process is not to be an obstacle to major developments, the planning authority (in this case the Defendant) must be able to grant planning permission so as to give the necessary assurance if it is satisfied that the outstanding design issues – which may include detailed design changes – can and will be addressed by the regulatory process.

[52]In support of his submission Mr. Wolfe relied on the decision of the CJEU in

Case C-435/97

World Wildlife Fund (WWF) and Others v Autonome Provinz Bozen and Others [1999] ECR 1-5613. Bozen was concerned with whether there was a power under Article 4(2) of the EIA Directive to exclude from the environmental impact assessment process, from the outset and in their entirety, certain classes of projects falling within Annex II (paragraph 35). Unsurprisingly, the CJEU decided that it was not permissible to exempt whole classes of projects in advance from the obligation to carry out a screening exercise. The criteria and/or the thresholds mentioned in Article 4(2) must “facilitate examination of the actual characteristics of any given project” (paragraph 37 emphasis added). No project should be exempt from environmental assessment “unless the specific project excluded could, on the basis of a comprehensive assessment be regarded as not being likely to have [significant effects on the environment].” (paragraph 45 emphasis added)

[53]Bozen was not concerned with the level of detail that is required about a project if, as in the present case, an environmental assessment is carried out. The CJEU was not asked to, and did not address the issue raised by Ground 2 in the present case: at what point may the competent planning authority conclude that it has sufficient information about the “actual characteristics” of a project, and/or that the environmental assessment is sufficiently “comprehensive”, to enable it to decide that a significant environmental effect is not likely because any outstanding details will be satisfactorily addressed by the relevant pollution regulator.

[54]I have considered Ground 2 upon the basis that, as submitted by the Claimant, it has a life of its own even if Ground 1 is rejected. In the abstract, the Claimant’s submission is correct – the circumstances in which a planning authority may rely upon a pollution regulator is a separate issue – but on the facts of this case Ground 2 has no substance if Ground 1 is rejected. The Claimant does not contend that the Defendant’s decision that severe nuclear accidents were very unlikely to occur was unreasonable. There has been no suggestion by any Member State, or any recognised scientific body, that such accidents are anything other than very unlikely. If Ground 1 is rejected, and it is concluded that the Claimant’s “zero risk” approach is not well founded, there is nothing to suggest that the Defendant’s assessment of the degree of unlikelihood of the risk of such accidents was erroneous. The views expressed by the ONR, the European Commission, the Austrian expert report and the Radiological Protection Institute of Ireland, were all to the same effect: that the risk of a severe nuclear accident is very low indeed. If the Defendant was not required to adopt a “zero risk” approach there is no basis for a submission that he should not have concluded that the risk was so unlikely that the environmental effects of such an accident should not be “scoped in” (ie should be excluded) for environmental impact assessment purposes.

Conclusion

[55]A reference to the CJEU is not necessary. I would dismiss this application.

[56]I agree.

[57]I also agree.

Cited in 12 later judgments

and 3 more of the most senior