As my colleague Robin Hopkins has warned, the decision of the Upper Tribunal in Fish Legal looks like a pretty big beast: sixty pages on whether water companies are public authorities for the purposes of the Environmental Information Regulations, applying the CJEU’s lengthy ruling on the points of principle (for which, see this post by Chris Knight).
(If you just need the quick answer: yes they are, by virtue of having ‘special powers’, but not by virtue of being under the control of a public body. For the detail, read on.)
In fact, while it could never be classed as a minnow, on closer inspection Fish Legal is not the monster it first seems (see Part 1 of the judgment here and Part 2 here). Fifteen of those sixty pages are appendices. More importantly, the decision poses, but declines to answer, some wider questions. Although Mr Justice Charles was sympathetic to the Information Commissioner’s request for guidance on how to identify a public authority, he stopped short of laying down ‘broad general principles’ on the question (paras.95-97). He gave shorter shrift to the water companies’ request that he list all of their powers which fell within the definition of ‘special powers’– a request made, apparently, with a view to lobbying Parliament to rid the companies of those powers, and so save them from the burden of EIR (para.98).
And one would hardly have expected him to address the question with the widest ramifications: if water companies are public authorities by virtue of their “special powers”, what of the various other privatised industries? It is, of course, a very fact specific analysis. Anxious electricity chiefs, rail bosses, and telecoms honchos will just have to read the judgment and consider how the ‘special powers’ and ‘control’ tests would apply to their own particular circumstances (see para. 110).
The UT had two issues before it: (i) whether the Information Commissioner had jurisdiction to decide whether a body was a public authority for the purposes of the EIR or FOIA (as he had purported to do), and therefore whether the UT itself had jurisdiction to hear the case, and (ii) whether privatised water companies in England and Wales are public authorities for the purposes of EIR, applying the principles set down by the CJEU following a referral from the UT (blog post here).
By the time of this, the second outing before the UT, the cast list had expanded significantly, bringing in several 11KBW counsel to join Anya Proops, who acted for the Information Commissioner before the CJEU. The Secretary of State was joined as a party and was represented by Julian Milford. The parties in two related cases, Cross v IC and the Cabinet Office and Bruton v IC and Duchy of Cornwall, were also invited to make submissions on the nature of the tests. Karen Steyn QC and Joseph Barrett appeared for Mr Bruton; Amy Rogers appeared for the Duchy. (Those cases will now go forward to be decided applying the Fish Legal principles.)
The jurisdiction issue
The Secretary of State argued that under s.57 FOIA, the First-Tier Tribunal only has jurisdiction over a decision notice issued by the Commissioner under s.50(3)(b) FOIA, and that the Commissioner had no jurisdiction to serve a decision notice on the issue of whether a body is a public authority. Section 50 is based on the premise that a request has been made to a public authority; these elements are anterior to the Commissioner’s jurisdiction and he has no authority to decide them within the framework of FOIA.
The UT rejected these arguments. It took the view that jurisdictional provisions are routinely based on certain assumed conditions, but this does not deprive the body in question of the jurisdiction to decide whether those conditions have been met. So the UT’s jurisdiction to hear an appeal on any point of law arising from a decision made by the FTT assumes that a decision has been properly made by a properly-constituted tribunal, but it does not mean that the UT cannot rule on whether those conditions are met in a given appeal (para 31).
The UT applied this reasoning to both a positive decision by the Commissioner that a body is a public authority, and a negative decision that it is not, even though the latter is not a decision notice under s.50(3) FOIA. To hold otherwise would mean that while a body could appeal against a positive decision, a requester would face the more expensive route of judicial review of a negative decision (para.37). Furthermore, the Commissioner would have no power under the similarly structured s.51 FOIA to require the information he needed to reach an informed decision that a body was not a public authority (para. 41).
After scrutinising the decision of the House of Lords in BBC v Sugar  1 WLR 430, the UT decided that there was nothing in the case that disturbed its conclusions on the point (paras.43-54).
The Commissioner therefore has jurisdiction to decide the issue, the FTT to hear appeals against his decisions and the UT to hear appeals against the decisions of the FTT.
The public authority issue
Two of the limbs of the definition of ‘public authority’ under the EIR were in issue. A finding that the companies fell within either would suffice to make them public authorities. (A little care is needed with the numbers of the provisions in question: Article 2(2)(b) of the Environmental Information Directive is transposed as Reg. 2(2)(c) of the EIR, and Article 2(2)(c) of the Directive as Reg. 2(2)(d) of the EIR.)
Persons performing public administrative functions – the ‘special powers’ test
The CJEU expanded on Art. 2(2)(b) of the Directive by explaining that persons ‘performing public administrative functions’ are
52 […] entities, be they legal persons governed by public law or by private law, which are entrusted, under the legal regime which is applicable to them, with the performance of services of public interest, inter alia in the environmental field, and which are, for this purpose, vested with special powers beyond those which result from the normal rules applicable in relations between persons governed by private law.
Only the parts underlined were in dispute in the case of the water companies, which clearly meet the rest of the test.
The UT declined to draw any conclusion from the fact that the CJEU had not seen fit to apply the principles to the facts of this case (paras. 99-100). It also rejected a suggestion that it should ask itself whether the companies’ powers were in the nature of State powers (as the Advocate General had suggested). That was because the definition of ‘State powers’ was unclear and ever-changing, and also because the CJEU had not adopted that test. In the end, however, the UT did adopt the State powers test ‘as a check’ – leaving the status of the test somewhat unclear (paras.113-117).
The main analysis focussed on the following powers of water companies:
Compulsory Purchase (under s.155 of the Water Industry Act 1991, “WIA”): this looks like a quintessential government power unavailable to ordinary citizens, but in fact the water companies enjoy the power at one remove: before exercising it they require authorisation by the Secretary of State, which they can apply for via a process set out in Schedule 11 to the WIA. Nonetheless, the provisions conferred a real, practical advantage on the water companies. Firstly, the application process afforded them privileged access to those advising the Secretary of State on whether to authorise a compulsory purchase. Secondly, it conferred significant commercial leverage on the companies in any negotiation to purchase land, even if they rarely resorted to it in practice (para 107).
Power to make byelaws (s.157 WIA): such byelaws require confirmation from the Secretary of State, but as with compulsory purchase, the power still confers an advantage on the companies. The section confers power beyond that of any private landowner, since byelaws under s.157 can be backed by criminal sanctions, unlike a landowner’s licence. ‘Special powers’ extends to ‘special powers of enforcement’ (para. 110).
Power to lay pipes (s.159 WIA): this power was the subject of a detailed comparison with the powers ordinarily available under private law. The companies argued that the same powers could be acquired through a license or easement. While accepting that this was potentially so, the UT emphasised that private law typically requires consent of the parties to achieve such a result (through the law of contract or property). By contrast, the WIA gives the water companies effective power to compel this result (para.121).
Power to enter land (s.168 WIA): although there are powers within private law allowing entry to another’s land (eg self-help to abate a nuisance at common law), they are narrowly circumscribed (eg they require possession of neighbouring land). The water companies’ powers are both wider and deeper: they apply against any landowner in the company’s area of license, and they extend to surveying and even boring on the land (para.125).
Hosepipe bans (ss.76-76C WIA): these powers are unlike anything available at private law, and moreover are backed by criminal sanctions.
Since it was content that the companies enjoyed a cluster of special powers, the UT formally left open the question of whether one would have been enough (see para. 105). However its comment in conclusion that the powers mentioned were ‘sufficient, collectively in themselves and as examples of powers of the same type’ to meet the test (para. 130) suggests that some pattern of powers will probably be necessary before a body is considered a public authority.
Persons under the control of public authorities
The CJEU’s elaboration of Article 2(2)(c) set out the test for ‘control’ in the following terms:
68 […] this third, residual, category of public authorities covers any entity which does not determine in a genuinely autonomous manner the way in which it performs the functions in the environmental field which are vested in it, since a public authority covered by Article 2(2)(a) or (b) of the directive is in a position to exert decisive influence on the entity’s action in that field.
The test applies to the manner in which functions are performed, not the functions themselves: a body is not under control of the Government merely because its powers derive from statute (para. 133). There are two elements to the test: the body must (i) operate in fact in a non-autonomous manner, and (ii) do so because a public authority is in a position to control it (para. 134). In other words, although the public authority need not actually be exercising its powers of control, the existence of the powers must have a real constraining effect on the body in question (para.135). Furthermore, the test required the UT to look at the companies’ overall manner of performing their environmental services: it would not be enough to find control in ‘one or two marginal aspects of their business’ (para. 136).
As for prior authorities, Smartsource was simply no longer relevant: the task of the UT was to consider the issue afresh in the light of the CJEU’s ruling (para.137).
The UT was at pains to point out that ‘no legitimate business has complete freedom of action’: all operate in a framework of legal and commercial constraints. Something more is needed before one can say that they have lost their autonomy (paras 142-144).
The two counsel for the requesting parties sought to supply that ‘something more’ by advancing ‘macro’ and ‘micro’ examples of actual state intervention in the water industry: recommendations by the Secretary of State made on reviewing a Draft Water Resources Management Plan, and an enforcement notice issued by OFWAT in respect of the risk of sewer flooding in the Penketh area. These were dismissed as no more than ‘increased intensity of oversight at particular times and in respect of particular activities’ (para.148).
A list of potential interventions was also provided – an analysis of the Secretary of State’s powers under the WIA. Although these clearly put the Secretary of State ‘in a position to exert decisive influence’ over the water companies’ actions, they too were rejected as not demonstrating control. The UT’s first two reasons are a little puzzling. The first was that the powers are (to some extent) a substitute for the forces of competition, since water companies enjoy local monopolies (para. 151). This rather begs the question: the grant of a monopoly confers great power on a private entity, which needs to be limited in the public interest. Just because state control is a substitute for market forces, this does not stop it from being a form of control. The second reason was that ‘the risk of enforcement is at most only a marginal consideration for a reputable company’ (para.152). Really? Why then did OFWAT need to issue that enforcement notice for the Penketh sewers?
However, the UT went on to find that whatever the potential for intervention, the more basic aspect of the control test was not satisfied. Merely listing all the possible interventions distorted the picture of how the companies operate in fact, and only addressed the second element of the test. The UT’s overall conclusion on the control test was that ‘despite the extent to which there is scope to influence the companies’ decision-making on the way it delivers its services, the evidence does not show that that influence is actually exerted to such an extent that overall the companies lack genuine autonomy (para.153).
You might not have predicted this result if you had read the CJEU’s heavy hint on the question of control – see  of its judgment
The weight that the UT afforded to the companies’ status as ordinary private companies is one of the surprises of the judgment. And it will be one of the few points of comfort for other privatised utilities. Not many can be as heavily regulated as the water industry; perhaps they too can avoid at least this limb of the definition.
POSTSCRIPT: if you’ve come here from Robin’s post, you may be wondering what Lewis Carroll has to do with all this. Loyal reader (which you must be if you’ve got this far), your guess is as good as mine. Why is the section on the State powers test (paras. 113-117) headed ‘Hunting of the Snark’? Answers on an e-postcard please.