An Taisce loses Hinkley Point case at Court of Appeal but may fight on

Angus Walker picture-13This entry reports on the outcome of the judicial review challenge in the Court of Appeal to the granting of consent for the Hinkley Point C nuclear power station.

Background

The Irish National Trust, An Taisce ('an tashka') challenged the grant of a development consent order (DCO) for the proposed Hinkley Point C nuclear power station on the ground that the Government should have consulted the Irish government on the transboundary effects of the project but did not do so.

Mrs Justice Patterson heard the case in December and ruled that consultation had not been necessary. An Taisce appealed to the Court of Appeal and this was heard last month by Lord Justices Sullivan and Longmore and Lady Justice Gloster (who said it was fine to address them as 'my Lords' rather than 'my Lords and my Lady' each time).

Judgment was issued on Friday and can be found here. The case centred around the meaning of 'likely' in this context, given that Ireland should have been consulted if environmental effects from the project were considered 'likely'.

Summary of judgment

It was common ground that there was a low probability that there would be an accident at a new nuclear power station in the UK - something like a one in 10 million chance, or a one in 2.4 billion chance in any particular year, which the Department for Energy and Climate Change calculated was about five thousand times less likely than a meteorite of at least 1km in diameter hitting the earth (I'm not sure if I find that reassuring). I would add that winning the lottery jackpot is also more than a thousand times more likely.

Before you start wondering why anyone would think that was 'likely', given that the word's ordinary meaning is 'more than a 50% chance', Lord Justice Sullivan in giving the leading judgment said that in this context it did not mean that. It would have been rather a short judgment otherwise.

The claimants argued that 'likely' meant the same for the purposes of the Habitats Regulations (which is quite strict) and the Environmental Impact Assessment regulations (which was being considered here), but the judges disagreed.

While the judges accepted that the greater the environmental impacts that would occur, the less the chance of them occurring would have to be for them to require consultation, taking a 'zero risk' approach was going too far. This does mean that the judges disagree with the views of the Aarhus Convention Implementation Committee, but the latter have admitted that their views are not legal ones.

The second of two grounds of challenge was that relying on the existence of a regulatory (safety) regime was not relevant, or at least had suffient gaps in it not to be of any help, to deciding whether environmental effects were likely.

The judges concluded that neither ground had merit and dismissed the application, also deciding that no reference should be made to the Court of Justice of the European Union (CJEU, formerly the ECJ). I understand that An Taisce are, however, considering whether to appeal to the Supreme Court.

Analysis

It may seem obvious that if something might only occur once every 10 million years, then it could not be described as 'likely', but in fact the legal arguments are more subtle than that. The judges had to distinguish the term's use in two different environmental directives, and also disagree with the Aarhus Convention Implementation Committee's precautionary approach (and the committee has still summoned the UK Government to appear before it and explain itself in December). Nevertheless looking at the case from a common sense point of view, it seems right that Ireland need not have been consulted.

From my mathematical background, perhaps their Lordships (and her Ladyship) were grappling with the idea of 'expectation' without realising it. In maths, the expected gain from a 1 in ten chance of winning £100 is the same as a 1 in 100 chance of winning £1000, namely £10. The consultation threshold should be treated similarly - include low-probability high-impact events, but not where the probability is much smaller than the magnitude of the impact. Anyway.  It was a technical judgment - I had to look up 'zoonoses', one of the words used in it.

The legal challenge to this decision has taken 509 days so far since the decision was taken, and may not be over. The application took 496 days to be decided from when it was made, and so the challenge has more than doubled the time from application to implementation. The Court of Appeal stage on its own has taken more than seven months.

I have said this before, but the reforms to judicial review such as a Planning Court and target timescales are welcome, but only affect the High Court and not the Court of Appeal or Supreme Court, and thus there remains a significant risk of lengthy delays from court proceedings should there be determined challengers.