T-Mobile (UK) Ltd. & Anor v Office of Communications

The Office of Communications (“Ofcom”) is the UK’s independent regulatory authority for electronic communications services and networks. One of its functions is to license the use of the electromagnetic spectrum for telecommunications purposes. On 4^th April 2008, following substantial consultation it published a document entitled “Award of available spectrum: 2500-2690 MHz, 2010-2025MHz” (“the Award”) The subheading was “This document sets out Ofcom’s decisions for the award of wireless telegraphy licences in these spectrum bands.”

The Award contains a number of decisions made by Ofcom in principle – policy decisions as to how and when it intended to go about licensing the spectrum concerned. The one that matters for present purposes is the decision to proceed forthwith by way of auction for the two wavebands concerned. With the Award were draft regulations for the implementation of this auction, published for consultation.

Both of the appellants, O2 (whose case was argued by Lord Pannick QC) and T-Mobile (whose case was argued by Michael Fordham QC) object to this manner of proceeding. The basis of their objection is essentially the same: that they already hold licences for other parts of the spectrum, that there is a possibility that they may lose some of the licences they now have for other parts of the spectrum pursuant to a decision to be made by Ofcom at some time in the future (called “refarming”) so that it would be unfair to make them bid now for more licences when they will not know what they will need until the refarming decision has been made. We were told that Ofcom are not in a position to make a decision about refarming now because the problems involved are complex.

On the other hand Ofcom takes the view that it is in the public interest that the new spectrum licences be allocated now in order to enable new players to enter the market – particularly for broadband access to mobile devices. Ofcom also observes that it is in the interests of existing operators such as O2 and T-Mobile that new operators be kept out of the market for as long as possible. Not only will that obviously avoid any competition from new operators during the period of delay, but also it will enable the existing operators to garner the market for this new technology before the newcomers can get going.

T-Mobile object to the whole Award and say it should be set aside entirely. O2 take a less radical position: that only some of the new wavelengths should be auctioned now, the rest to be auctioned or otherwise marketed after the refarming decision is made known.

We are not concerned with the merits of these objections. The narrow point before us is simply this: are these matters to be raised by way of an appeal to the Competition Appeal Tribunal (“CAT”) or must they go by way of judicial review (“JR”).

O2 and T-Mobile launched appeals from the Award to the CAT. T-Mobile has also commenced JR proceedings in which O2 (and another party) have intervened. These are in abeyance pending a final decision on the question of jurisdiction of the CAT.

By its decision of 10^th July this year, the CAT (Vivien Rose, Dr Arthur Pryor CB and Adam Scott TD) decided it did not have jurisdiction to hear the appeals. This is the appeal from that decision.

So the relevant parts of the legislation can be seen as a whole, I have set these out separately in an Appendix. Where it has been amended it is set out as amended, though nothing turns on the amendments. The relevant UK legislation is contained in the Communications Act 2003 (CA 2003) and the Wireless Telegraphy Act 2006 (WTA 2006). The relevant EU legislation is the “Framework Directive”, Directive 2002/21/EC “on a common regulatory framework for electronic communications networks and services.”

Lord Pannick and Mr Fordham divided up the arguments for jurisdiction in the Court of Appeal between them, each also espousing the arguments of the other. I have to say that one always has a suspicion that if a great number of alternative arguments are deployed, there is no really good argument. And all the more so if some of them involve a turgid trawl through a mass of travaux préparatoires. So it proved here, for I think the CAT was clearly right. It has no jurisdiction to hear these appeals; the Administrative Court is the place where the appellants' complaints can be heard and decided in compliance with the requirements of the Directive.

Lord Pannick went first, working on the assumption Mr Fordham’s arguments on the construction of the UK legislation would be rejected. I will consider the arguments in the same order. But before I do so it is essential to lay down some important foundation stones.

First, it is common ground that the challenge to the Award which the appellants wish to make is an “appeal” within the meaning of Art. 4(1) of the Framework Directive. This requires that “effective mechanisms .. under which any user … has a right of appeal be provided by Member States.”

Art. 4 goes on to require that the “appeal body” be independent, and that:

It is also common ground that Art. 4 confers on affected parties such as O2 and T-Mobile a directly applicable right of appeal. So, if no effective appeal mechanism is created by UK law, part of that law must be disapplied so such a right of appeal is conferred. I suppose an alternative to disapplication might perhaps be a Francovitch claim against the State, but no one suggested this and the point does not arise. However that may be one would obviously strive to avoid the conclusion that the UK, in seeking to implement the Framework Directive, had failed to achieve that.

But, and this lies at the heart of this case, it is not now suggested that the UK is in breach of Art. 4. For it is now common ground (it was not below) that if the route of challenge to the Award must be by way of JR rather than appeal to the CAT, such a route would be an “effective appeal mechanism” within the meaning of Art. 4.

Although, as I say, this point is now common ground, it is appropriate to spell out in more detail why I think Lord Pannick was right so to concede.

Section 31 of the Supreme Court Act 1981 provides that an application for JR relief (mandatory, prohibiting or quashing orders, or certain forms of declaration and injunction) can be made to the High Court. But statute does not seek to spell out or limit what the court can consider on an application for JR. All the Act says about jurisdiction is this:

s.29(1A) The High Court shall have jurisdiction to make mandatory, prohibiting and quashing orders in those classes of case in which, immediately before 1^st May 2004, it had jurisdiction to make orders of mandamus, prohibition and certiorari respectively.

Those limits are set by the inherent jurisdiction of the court, themselves governed by the rules of precedent. Traditionally those limits indeed confined the courts to considering things like procedural unfairness or Wednesbury unreasonableness – various forms of error of law. JR did not allow an attack purely on the merits of the impugned decision. And that is still broadly so, as the cases cited by Lord Pannick demonstrate. He took us to R (Daly) v Secretary of State for the Home Department [2001] UKHL 26,[2001] 2 AC 532 and R (SB) v Governors of Denbigh High School [2006] UKHL 15, [2007] 1 AC 100.Both were concerned with the impact of the ECHR in JR cases. It is sufficient for present purposes to go to what Lord Bingham said in the latter case:

[34] Secondly, it is clear that the court’s approach to an issue of proportionality under the Convention must go beyond that traditionally adopted to judicial review in a domestic setting. The inadequacy of that approach was exposed in Smith and Grady v United Kingdom (1999) 29 EHRR 493, para 138, and the new approach required under the 1998 Act was described by Lord Steyn in R (Daly) v Secretary of State for the Home Department [2001] 2 AC 532, paras 25-28, in terms which have never to my knowledge been questioned. There is no shift to a merits review, but the intensity of review is greater than was previously appropriate, and greater even than the heightened scrutiny test adopted by the Court of Appeal in R v Ministry of Defence, Ex p Smith [1996] QB 517, 554.

It was Lord Pannick’s initial submission that JR would not comply with Art. 4. The submission was that the emboldened passage showed that a JR court could not duly “take into account” the merits of the impugned decision as required by Art. 4. But as matters developed during the course of argument he rightly and fairly accepted that the common law in the area of JR is adaptable so that the rules as to JR jurisdiction are flexible enough to accommodate whatever standard is required by Art. 4.

Despite Lord Pannick’s concession, Miss Dinah Rose QC for Ofcom rightly, firmly and forcefully went on to demonstrate that it was correctly made, and that the JR standard of review can and does mould itself to any requirement imposed by other rules of law. She did so to establish that the appellants would be in no way disadvantaged if they had to go via the JR route as compared with the CAT.

She tackled it two ways, both leading to the same result, one via EU law and the other via the cases on JR which show that it can adapt its intensity of review where that is necessary.

For the EU route she relied on Unibet v Justitiekanslern, Case C-432/05 [2007] ECR I-2271 where the ECJ said:

For the domestic route she showed that our courts have recognised that JR can be suitably adapted where necessary. She took us to the judgment of Carnwath LJ (with whom Mance LJ agreed) in IBA Healthcare v OFT [2004] ICR 1364, [2004] EWCA Civ 142. Under the heading “Principles of JR” Carnwath reviewed the cases on intensity of JR, concluding:

[100] … Those principles, whether applied by a court or a specialised tribunal, are flexible enough to be adapted to the particular statutory context.

And, in the context of human rights, Miss Rose showed us cases where it was held that it was necessary to go into the merits on a JR application. Thus in Wilkinson v Broadmoor Special Authority [2002] 1 WLR 419, [2001] EWCA Civ 1545, this was held necessary in the context of a case concerning the human rights of a compulsorily detained convicted mental patient. The issue was whether the patient was mentally capable or not to consent to a treatment regime. The Court held that this issue could be investigated in JR proceedings, even, if necessary, by the calling of medical witnesses in those very proceedings.

Simon Brown LJ ended his judgment saying this:

[36] That, however, is essentially by the way. For the reasons earlier given I have concluded that what would be required on a substantive challenge here would be a full merits review of the propriety of the treatment proposed and, for that purpose, cross-examination of the specialists. I would allow this appeal.

Brooke LJ referred to the availability of a “full merits review” (see [44]) and Hale LJ said:

[83] … Super-Wednesbury is not enough. The claimant is entitled to a proper hearing, on the merits, of whether the statutory grounds for imposing this treatment upon him against his will are made out …

It is true that the Court referred to other machinery by which the point could be considered, e.g. an action in tort. But that was not the reason for holding that there can be, if necessary, a full merits investigation on a JR application if that is what the Human Rights Act requires.

Wilkinson does not stand alone as direct authority for the proposition that JR can provide a full merits investigation where that is necessary - see also R v Haddock [2006] HRLR 40, [2006] EWCA Civ 961 at [63-65] per Auld LJ with whom the other members of the court agreed.

It may be additionally noted that the present case is not as strong as Wilkinson or Haddock. For this case is concerned only with the merits being duly taken into account in the JR proceedings, not with a full investigation ab initio of facts and evidence for the first time in the JR application itself.

Accordingly I think there can be no doubt that just as JR was adapted because the Human Rights Act so required, so it can and must be adapted to comply with EU law and in particular Art. 4 of the Directive. If the CAT did not exist JR would have to and could do the job. The CAT’s existence does not mean that JR is incapable of doing it.

I would add this: it seems to me to be evident that whether the “appeal” went to the CAT or by way of JR, the same standard for success would have to be shown. In either case it would not be enough to invite the tribunal to consider the matter afresh – as though the Award had never been made. Lord Pannick suggested the standard of investigation on an Art. 4 appeal would be intensive, relying on Mobistar v IBPT, Case C-438, [2006] ECR I-6675. He relied on the Court’s ruling:

[43] Accordingly, the answer to the second question must be that Article 4 of the Framework Directive must be interpreted as meaning that the body responsible for hearing an appeal against a decision of the national regulatory authority must have at its disposal all the information necessary in order to decide on the merits of the appeal, including, if necessary, confidential information which that authority has taken into account in reaching the decision which is the subject of the appeal. However, that body must guarantee the confidentiality of the information in question whilst complying with the requirements of effective legal protection and ensuring protection of the rights of defence of the parties to the dispute.

I do not accept that Mobistar is authority for the suggested proposition. The case was about something quite different from the standard of review to be applied on appeal. It was about whether, on an appeal, confidential information used by the regulator must be disclosed to the appellant. That is a question of basic fairness – whatever the standard of review, the appellant cannot get going if he cannot see the basis of the impugned decision.

After all it is inconceivable that Art. 4, in requiring an appeal which can duly take into account the merits, requires Member States to have in effect a fully equipped duplicate regulatory body waiting in the wings just for appeals. What is called for is an appeal body and no more, a body which can look into whether the regulator had got something material wrong. That may be very difficult if all that is impugned is an overall value judgment based upon competing commercial considerations in the context of a public policy decision.

Since JR does comply with Art. 4, many of the points initially made by Lord Pannick in support of his contention that the JR standard of review was non-Art. 4 compliant fall away. I will discuss them briefly therefore.

I have already referred to Lord Pannick’s “intensive review” point based on Mobistar. He also sought to support the point by a linguistic contrast. Art. 4 uses “effective appeal mechanism” in 4(1) and “subject to review by a court” in Art. 4(2). Lord Pannick suggested that the difference in language pointed to a higher standard for an Art. 4(1) appeal. I think that is much too semantic an approach. The point of Art. 4(2) is to ensure that if the appeal body is not a court it gives proper reasons. The reference to “review” by a court is so that it is possible for a reference to be made to the ECJ under Art. 234 of the treaty – it is only courts which can make such a reference. It is inconceivable that those who drafted Art. 4 had in mind the sort of distinction English law draws between “review” and “appeal”.

Lord Pannick also took us on a trawl through the travaux préparatoires to Art. 4 to suggest that an intensive appeal on the merits was intended. I do not think it a worthwhile exercise to set this out (going through pages of travaux seldom is). None of the material goes so far as to show that anyone expected the appeal body to start de novo or that it should in effect be a duplicate of the regulator waiting to hear appeals.

Lord Pannick referred to the “appropriate expertise” requirement, using it to suggest a JR court would not comply with it. But since the Chairman of the CAT is a Chancery Judge and all Chancery Judges can sit in the CAT and, by arrangement, in the Administrative Court there was nothing in this. Lord Pannick emphasised that the CAT has lay members whereas the JR court does not. But the CAT’s lay members may well not have any experience of the particular subject-matter, so that point goes nowhere. And besides the JR court itself, if it felt expertise were necessary (which would be a rare case) could sit with assessors – see s.70 of the Supreme Court Act 1981 and CPR Part 35 r.15. Also, Lord Pannick did not identify with particularity any expertise which would be unavailable to a JR court so far as this case is concerned

He also developed a point based on Connect Austria v Telecom-Control-Kommission [2003] ECR I-5197, Case C-462/99. The predecessor provision to Art. 4, Art. 5 of Directive 90/387/EEC, provided that a party affected by a decision should have a right of appeal. Appeal from Austria’s telecom regulatory authority was to the constitutional court. But, by national law, this had very limited jurisdiction and could not take into account the merits of a case. So there was no proper appeal provided for in national law as required by the Directive. The ECJ said:

[40] Where application of national law in accordance with the requirements of Article 5a(3) of Directive 90/387 is not possible, the national court must fully apply Community law and protect the rights conferred thereunder on individuals, if necessary disapplying any provision in the measure the application of which would, in the circumstances of the case, lead to a result contrary to that directive, whereas national law would comply with the directive if that provision was not applied (see, to that effect, Engelbrecht, paragraph 40).

[41] It follows that a national court or tribunal which satisfies the requirements of Article 5a(3) of Directive 90/387 and which would be competent to hear appeals against the decisions of the national regulating authorities if it was not prevented from doing so by a provision of national law which explicitly excluded its competence, such as Article 133(4) of the B-VG, has the obligation to disapply that provision.

[42] Therefore, the answer to the first question referred for a preliminary ruling must be that in order to ensure that national law is interpreted in compliance with Directive 90/387 and that the rights of individuals are effectively protected, national courts must determine whether the relevant provisions of their national law provide individuals with a right of appeal against decisions of the national regulatory authority which satisfies the criteria laid down in Article 5a(3) of Directive 90/387. If national law cannot be applied so as to comply with the requirements of Article 5a(3) of Directive 90/387, a national court or tribunal which satisfies those requirements and which would be competent to hear appeals against decisions of the national regulatory authority if it was not prevented from doing so by a provision of national law which explicitly excluded its competence, such as that at issue in the main proceedings, has the obligation to disapply that provision.

Lord Pannick submitted, on the hypothesis that s.192(a) of the CA 2003 precluded jurisdiction in the CAT, it should, as in Connect-Austria, be disapplied. But that is to ignore the reason why the ECJ said the Austrian provision limiting jurisdiction should be disapplied. It was because unless that were done, there would be no appeal remedy as required by EU law – no effective jurisdiction. So EU law had to trump national law. That is not this case – national law does provide a remedy in the shape of JR. There is no national law which needs to be trumped.

Moreover, as Miss Rose pointed out, if it were a case where disapplication of an English rule of law was required by EU law, the appropriate disapplication would not be of a statute specifically limiting the jurisdiction of the CAT; it would be disapplication of whatever restriction within the rules of JR precluded effective jurisdiction, as in Unibet.

Lord Pannick also relied on some decisions about benefits which he submitted showed that a restriction on jurisdiction contrary to EU law should be disapplied. They were ex parte EOC [1995] 1 AC 1, Staffordshire CC v Barber [1996] ICR 379 and R (Manson) v MoD [2006] ICR 355. They were all about claims where there was a domestic rule, contrary to EU law for one reason or another, excluding certain claimants. The courts held that those rules, so far as inconsistent with EU law, should be disapplied.

I do not consider these cases assist at all for two reasons. Firstly there was nothing like the express exclusion of jurisdiction we have here. And more fundamentally they were not cases about disapplication of “jurisdiction.” They were about the statutory criteria to be applied by the employment tribunal – exclusionary criteria invalid by EU law did not count. “Jurisdiction” is a slippery word. In one sense no court or tribunal has any “jurisdiction” other than to find in favour of the party who is right. But that is a different meaning from “jurisdiction” to decide who is right. The employment tribunals had overall jurisdiction to decide on claims of the kind in question. What the employment tribunals had to do was decide whether a claim was justified – the cases merely held that in so doing they were to ignore preclusions bad under EU law.

Lord Pannick also suggested that these cases showed that where a provision excluding jurisdiction was to be disapplied, it was to be disapplied so as to favour jurisdiction in a specialist tribunal. I do not accept that either. Firstly it is clear from Recital 12 of the Framework Directive that it is up to each Member State how it is to be implemented. Indeed the reference to “division of competences” expressly contemplates that different subject-matters may be within the competence of different tribunals.

Secondly there is no such rule of EU law as he suggests. If there had been such a rule, the reasoning in Impact v Minister for Agriculture and Food, Case C-268/06 (not yet reported), would have been quite different. A Directive required the State to provide certain benefits. Ireland had set up a tribunal to deal with such claims, but had done so late. Under Irish law the tribunal only had jurisdiction to deal with claims covering the period from implementation of the Directive. Claims based on the period during which the Directive should have been but was not, implemented, would be entertained by the ordinary courts. The ECJ said this:

[43]      In that regard, it is important to note that the principle of effective judicial protection is a general principle of Community law (see, to that effect, Case C-432/05 Unibet [2007] ECR I-2271, paragraph 37 and the case-law cited).

[44]      The Court has consistently held that, in the absence of Community rules governing the matter, it is for the domestic legal system of each Member State to designate the courts and tribunals having jurisdiction and to lay down the detailed procedural rules governing actions for safeguarding rights which individuals derive from Community law (citations omitted).

[45]      The Member States, however, are responsible for ensuring that those rights are effectively protected in each case (citations omitted)

So it is up to Member States to decide how to implement – they can have one or more tribunals. But there is a secondary rule which may come into play, where by splitting jurisdictions up, it is made too difficult for claimants. This is because in that case there would be a breach of the principle of effectiveness. The court expressed it this way:

[51] In those circumstances, where the national legislature has chosen to confer on specialised courts jurisdiction to hear and determine actions based on the legislation transposing Directive 1999/70, the obligation which would be placed on individuals in the situation of the complainants – who sought to bring a claim based on an infringement of that legislation before such a specialised court – to bring at the same time a separate action before an ordinary court to assert the rights which they can derive directly from that directive in respect of the period between the deadline for transposing it and the date on which the transposing legislation entered into force, would be contrary to the principle of effectiveness if – which is for the referring court to ascertain – it would result in procedural disadvantages for those individuals, in terms, inter alia, of cost, duration and the rules of representation, such as to render excessively difficult the exercise of rights deriving from that directive.

That is miles from saying that where there is a specialist tribunal all claims must be put in that box.

The upshot of all this is that EU law is completely neutral about the question we have to decide. It points neither to the CAT nor JR. Given that one route or the other complies with Art. 4, EU law stands aside because its requirements are met either way. It follows also that any reference to the Marleasing principle in construing our domestic legislation is misplaced.

I therefore turn to Mr Fordham’s argument: that as a matter of construction, the CAT does have jurisdiction over the proposed appeals.

I start with the essence of the statutory provisions. s.192(2) of the CA 2003 says that a person affected by a decision “to which this section applies may appeal to [the CAT].” s. 192(1)(a) tells us that the section applies to a decision by Ofcom … that is not a decision specified in Schedule 8.” Schedule 8 (headed “Decisions not Subject to Appeal”) identifies a mass of provisions of the CA 2003 and the WTA 2006, all of which are not to be subject to appeal. The relevant provision is, here, rule 40. This specifies inter alia, a decision given effect to by regulations under, inter alia, s.14 of the WTA 2006. s.14 of the WTA 2006 sets out Ofcom’s power to make regulations which specify a procedure for bidding for licences.

So the question is, is the Award, or part of it, a decision given effect to by regulations under s.14? In one sense it is not – because as it stands no regulations have yet been made. Mr Fordham made that point at the outset. But it is unrealistic – to say that regulations cannot be challenged by appeal but the decision to make them can be so challenged is absurd and cannot have been intended by the statutory exclusion of jurisdiction. Taken to its illogical conclusion one could have a challenge to the decision to make regulations running side by side with a JR challenging the regulations so made. Parliament must have intended that the exclusion from the CAT’s jurisdiction covered both regulations made under s.14 and the decision to make such regulations.

Mr Fordham’s main attack was directed at what he called the “sequencing decision” – the decision to go ahead with the auction now and not wait for re-farming. He submitted that that was an independent decision from the decision to have a regulated auction: it was not a decision to be implemented by regulations to be made under s.14. Accordingly it was not excluded by Schedule 8.

Miss Rose answered that in a number of ways. First she submitted that was simply the wrong way to look at things. The decision was to get on with an auction now in the public interest. That is all. There was no actual decision not to wait until refarming could be decided, that is merely the consequence of getting on with s.14 regulations now.

Next she submitted that if that were regarded as a decision, it would not be a decision under the WTA as such and so no appeal could arise. This is because there was no proper application or request to exercise a power or duty for the purposes of s.192(7)(b) CA 2003 (a letter from T-Mobile of 7^th May 2008 was not good enough – Mr Fordham did not challenge that).

Thirdly she submitted the whole appeal scheme is directed at appeals against decisions taken under identifiable provisions (see s.192(5) saying that a notice of appeal must identify the provision under which the decision under challenge was made). Miss Rose said the parties had been repeatedly asked to do that. Lord Pannick and Mr Fordham each produced overnight a document seeking to identify the provisions concerned. They differed, suggesting as Miss Rose had contended, that the so-called “sequencing” decision was not in reality a decision at all. Lord Pannick openly admitted that “it is difficult to identify the particular provision”. He opted for s.14 of the WTA. But that is just a complaint that Ofcom is going to make s.14 regulations now –and s.14 regulations are excluded from the CAT’s appeal jurisdiction. Mr Fordham went back to submitting that there were two decisions, one to proceed by auction now and the other to do so before refarming. As regards the latter he suggested the decision was made under s.9 of the WTA 2006. This provides the general power for Ofcom to grant licences. Miss Rose observed with force that this is the first time anyone has suggested that the “sequencing” decision was made under that section. It obviously was not.

I accept all of Miss Rose’s contentions I have so far mentioned. But above all I accept her overriding contention about the nature of the matters excluded from CAT appeal jurisdiction. It is that all the legislative powers of Ofcom to make regulations have been systematically included in Schedule 8. In her words “Parliament did not intend legislative and quasi-legislative decisions to go to the CAT.” It is not the function of a statutory tribunal to impugn statutory instruments or regulations made pursuant to statutory powers. Challenges to these are classically matters for JR and that is so in the case of the Award.

Accordingly I would dismiss this appeal. It follows that the appellants’ route of redress, if any, is via JR. I would hope that arrangements can be made to bring this on as soon as possible, preferably with a Chancery Judge sitting in the Administrative Court, that Judge being, if possible the President of the CAT, Barling J.