In the summer I wrote on this blog about the Administrative Court’s decision in R (Oceana) v Upper Tribunal [2023] EWHC 791 (Admin). Oceana was the first time the new ‘super ouster clause’ contained in section 11A of the Tribunals, Courts and Enforcement Act 2007 (‘TCEA 2007’), introduced by section 2 of the Judicial Review and Courts Act 2022, fell for judicial consideration, and it was informative to observe how judges responded to the clause in judicial review. But while Oceana raised an important point of constitutional law, namely, Parliament’s ability to exclude judicial review with sufficiently express statutory language, the underlying case in Oceana was a weak one. It was little surprise, therefore, that Oceana did not fall for reconsideration on appeal. However, Saini J’s judgment in Oceana that Parliament is institutionally competent to exclude judicial review in certain circumstances has now come under further judicial scrutiny by the Court of Appeal in R (LA (Albania)) v Upper Tribunal [2023] EWCA Civ 1337. And so it is time for another blog post, addressing again the important question on the effectiveness of ouster clauses and the potential limits of Parliament’s legislative sovereignty.
The applicant in the present case was a citizen of Albania. She had entered the UK by clandestine means in November 2018 before seeking asylum on the basis that, as a lesbian Muslim, she feared persecution in her home country. Her claim for asylum was denied by the Home Secretary and she was eventually detained and issued with removal directions. The applicant’s attempt to challenge this decision failed in the First-tier Tribunal (‘FTT’). When the Upper Tribunal refused permission to appeal the FTT’s decision, the applicant applied for judicial review arguing that as the FTT’s decision was marred by an error of law, the Upper Tribunal was wrong in law to deny her permission to appeal. But the High Court denied leave on the papers, the judge holding it would be ‘hopeless’ to argue that section 11A TCEA 2007 did not exclude judicial review. None of the (very limited) exceptions to the ouster clause—‘threshold’ jurisdictional error of law in the pre-Anisminic sense, improper constitution, bad faith or a ‘fundamental breach of the principles of natural justice’—could be made out in the applicant’s case, and so the court had to give effect to the clause. The applicant’s appealed that refusal to the Court of Appeal, thus requiring the court to decide, in the face of section 11A, ‘whether it had jurisdiction to hear the appeal and whether Oceana was rightly decided’ (para 10).
It is worth reminding ourselves how express the ouster clause in section 11A TCEA 2007 really is. The standard reasoning employed in Anisminic is that decisions made without jurisdiction are nullities; they are merely purported decisions, and so not covered by any statutory ouster of judicial review. This reasoning was not novel to Anisminic. It went back, at the latest, to 19th-century certiorari proceedings in the Court of King’s Bench, and made perfect sense in a world where judicial review was available not only to check whether a decision-maker had acted within its jurisdiction, but also to review errors of law revealed on the face of the record itself: these latter errors would not lead to a decision being considered a nullity, and so the ouster clause would still be effective in excluding review in such a case (for more on this history, see my earlier article in the CLJ). But section 11A is drafted expressly to exclude ‘jurisdictional’ review too. Hence ‘the Upper Tribunal is not to be regarded as having exceeded its powers by reason of any error made in reaching the decision’ (s 11A(3)(a)); ‘the supervisory jurisdiction [of the High Court] does not extend to, and no application or petition for judicial review may be made or brought in relation to, the decision’ (s 11A(3)(b)); and, for the avoidance of any doubt, ‘“decision” includes any purported decision’, ie, one affected by jurisdictional error (s 11A(7)). In Oceana, Saini J rejected the bold assertion by counsel that section 11A was an impermissible ouster of judicial review that the courts could, at common law, ignore. It was this finding that fell for reconsideration in the present case.
In seeking to overturn Oceana, counsel for the applicant relied on the judgment of Lord Carnwath (Lady Hale and Lord Kerr agreeing) in R (Privacy International) v Investigatory Powers Tribunal [2019] UKSC 22. In that case Lord Carnwath had said, obiter, that even where Parliament used sufficiently clear statutory language to overcome the common law’s fundamental presumption against the ouster of judicial review, the courts might not give effect to an ouster clause which tried wholly to exclude the High Court’s supervisory jurisdiction. Lord Carnwath’s opinion on this point was a minority, obiter opinion: Lord Lloyd-Jones did not express his views, and Lord Reed, Lord Wilson and Lord Sumption expressly rejected such an approach. But it was an important obiter opinion nonetheless, echoing earlier comments by Lord Steyn and Lady Hale in the Jackson case as to common-law limits on Parliament’s legislative authority. Already Lord Carnwath’s views in Privacy International have served as grist to the mill of many a constitutional law essay, and it is good that their significance for our public law has now been judicially considered in greater depth.
In the present case, Dingemans LJ, like Saini J in Oceana, accepted that ‘the wording of section 11A of the 2007 Act is sufficiently clear to change the scope of judicial review’. He also held, importantly, that the court would be ‘bound to apply the wording of section 11A’ (para 31). As such, he rejected the approach suggested in Privacy International that even where Parliament has been sufficiently clear in ousting judicial review, such an ouster may be declared ineffective.
But it is worth paying close attention to Dingeman LJ’s reasons for concluding the court was bound by section 11A. Five reasons were given:
- It was ‘essential to note’ that section 11A did not oust the supervisory jurisdiction of the High Court entirely. Judicial review was still available for jurisdictional errors of law in the pre-Anisminic sense as well as fundamental breaches of natural justice, improper constitution of the tribunal and bad faith. The effect of section 11A was to reduce the scope of judicial review of the Upper Tribunal, whose expert judicial nature was noted (para 31).
- The effect of section 11A was to restore the approach to review of Upper Tribunal permission-to-appeal decisions preferred by the Divisional Court and Court of Appeal in R (Cart) v Upper Tribunal, according to which review should be limited to pre-Anisminic excesses of jurisdiction and fundamental denials of procedural justice. The Supreme Court’s preference for a scope of review delimited by the second-tier appeals criteria had been adopted,per Lord Dyson in Cart, in the absence of any clear Parliamentary delimitation of the scope of review. But as Saini J held in Oceana, section 11A was Parliament’s own determination of the matter (para 32).
- Preferring the lower courts’ approach in Cart was not constitutionally problematic. There was no suggestion by the Supreme Court that the Divisional Court’s and Court of Appeal’s judgments had failed to attribute sufficient constitutional importance to the High Court’s supervisory jurisdiction (para 33). Indeed, Lady Hale in Cart described Laws LJ’s judgment in the Divisional Court as ‘subtle and erudite’.
- Both the Supreme Court’s approach in Cart and section 11A TCEA 2007 countenance some errors of law continuing in the tribunals system without correction. While these errors might be considered as making the decision a nullity on traditional administrative law grounds, it can at least be noted that section 11A purports to modify this by declaring that the Upper Tribunal should not be regarded as having exceeded its jurisdiction where it makes an error of law in reaching its decision (para 34). (The same cannot be said, it appears to be suggested by Dingemans LJ, for the Supreme Court’s approach, which countenances the non-reviewability of jurisdictionalerror.)
- Section 11A(7) tackles the issue of nullity ‘head on’ by making clear that references to an Upper Tribunal ‘decision’ in section 11A includes a purported decision (para 35).
In light of these reasons, Dingemans LJ held that section 11A was effective in ousting judicial review, and that the court had no power to constructively reinterpret the clause otherwise or directly ignore it. As Dingemans LJ held at the end of his consideration of section 11A’s bindingness, ‘[i]t is the duty of the Courts to give effect to the clear words used by Parliament, because no one, including a Court, is above the law. The decision by Saini J in Oceana was right.’ (Para 36.)
Because the applicant could not convincingly argue that one of the exceptions in section 11A applied to her case, her appeal failed. Lewis and Underhill LJJ agreed with Dingemans LJ in short concurring judgments. Of particular note are Underhill LJ’s comments, at para 51, that ‘there is nothing constitutionally improper’ in Parliament limiting the judicial review jurisdiction of the High Court in the way it did in section 11A TCEA 2007.
A few brief comments can be made on the significance of the judgment. The first echoes in part my earlier comments on Oceana. Ouster clauses are often presented as raising an intractable conflict between the rule of law one the one hand and Parliamentary sovereignty on the other. But cases like Oceana and now LA reveal that this conflict is not as clear-cut as it is sometimes presented. For a start, we need to remember that ouster clauses are not to be interpreted in a vacuum; the broader legislative and constitutional context will be decisive in determining how courts might respond to Parliament’s attempts at ouster. As my colleague Mark Elliott has suggested in his perceptive comment on the Oceana case, the fact that section 11A allows for some judicial review, and that it merely restricts review of decisions made by an expert judicial body (which, indeed, exercises its own judicial review powers in some circumstances) means that the threat to the rule of law is not as great here as it might otherwise be. If Parliament had sought to exclude review in toto, or if it excluded review in the same way but as to a more squarely ‘executive’ decision-maker, such as a government minister or a local authority, the outcome may have been different. But the Upper Tribunal has its own role to play in upholding the rule of law, and making its decisions final in most (if not all) circumstances seems no greater threat to the rule of law than limiting the availability of review or appeal in the ordinary courts. Indeed, Dinegmans LJ made this precise point in his judgment (see especially paragraph 31).
At the same time, Elliott was clear in his reflection on Oceana that there is a tension between Saini J’s wider comments in that case about the courts’ duty to give effect to sufficiently clear statutory ousters of judicial review and Lord Carnwath’s dicta in Privacy International. As Elliott said,
It is one thing for a clearly worded (albeit partial) ouster clause to tip the balance against (some) judicial reviews in such circumstances. It is another thing to conclude that a court would or should be equally sanguine if, in different circumstances, it was presented with an ouster clause that was prima facie an affront to basic constitutional values.
This point is revisited (obiter?) in LA. And it is notable that the judges of the Court of Appeal expressly question the significance of Lord Carnwath’s dicta, observing that they represented an opinion shared by only three of the seven Justices of the Supreme Court in that case. Not only this, both Dingemans and Lewis LJJ call the correctness of such an approach directly into question. Following Saini J in Oceana, the Court of Appeal questions the very possibility, as a matter of constitutional law, of the courts choosing not to apply an express provision of an Act of Parliament. Indeed, the courts seem to be clear that the rule of law requires, at least in part, fidelity to clear law promulgated by the sovereign legislature. As Dingemans LJ said, ‘no one, including a Court, is above the law.’
Above all, then, LA shows the constitutional issues raised by ouster clauses ought not to be reduced to a game of constitutional-law Top Trumps, parliamentary sovereignty being pitched against the rule of law with only one potential winner. For while the courts undoubtedly have a special responsibility in upholding the rule of law, they too are bound by that constitutional principle. Judicial interpretation that is sensitive to the limits imposed by statutory language, properly constructed, and common law adjudication respectful of long-established precedents, is all part and parcel of this. Respect for the rule of law requires respect for formally promulgated law, sometimes even in cases where that leads to a tension with other rule-of-law requirements like access to the courts and judicial review of executive action. And while it certainly requires adjudication informed by, per Lord Wilson in the Evans case, ‘precious constitutional principles’, among the most precious of these ‘is that of parliamentary sovereignty, emblematic of our democracy’ (Evans, paragraph 168).
It will be interesting to see whether LA or another case finally reaches the Supreme Court where these issues can be determined conclusively. Until then, it seems that the bolder assertions of some of the Supreme Court Justices in Privacy International cannot be taken as reflective of established constitutional law. This is not, of course, only of academic interest. Since I wrote on Oceana, the Illegal Migration Act 2023 (‘IMA 2023’) has appeared on the statute book. That Act provides for the deportation of persons who have travelled to the United Kingdom from a ‘safe third country’ and who do not have the necessary leave to enter or remain in the UK. The Act imposes a duty on the Home Secretary to deport such persons to the country of their nationality where it is safe to do so, or otherwise to one of 57 specified safe third countries. Under section 5, the Home Secretary is required to disregard any application a person makes for judicial review (not to mention protection, human rights and modern slavery claims). Immigration officers are given the power to detain such persons for up to 28 days prior to their deportation.
Important to our purposes, amendments are made by section 13 IMA 2023 to Schedule 10 of the Immigration Act 2016 on immigration bail. A new paragraph 3A(2) that an immigration officer’s decision to detain someone ‘is final and is not liable to be questioned or set aside in any court or tribunal’. Sub-paragraph (3) is drafted, it seems, expressly to exclude judicial review on ordinary jurisdiction-based grounds:
(a) the powers of the immigration officer or the Secretary of State (as the case may be) are not to be regarded as having been exceeded by reason of any error made in reaching the decision;
(b) the supervisory jurisdiction does not extend to, and no application or petition for judicial review may be made or brought in relation to, the decision.
Sub-paragraph (6) makes clear that the reference to a ‘decision’ in the ouster clauses is to be read as including any purported decision. While it should be noted that sub-paragraph (4) preserves the possibility of judicial review for bad faith or fundamental breaches of the principles of natural justice, and sub-paragraph (5) preserves judicial review by means of the ancient remedy of habeas corpus, it must be questioned whether this will be found to be sufficient to satisfy the rule of law’s demands for independent judicial oversight. This is particularly questionable in light of the fact that immigration officers and the Home Secretary are not independent, expert, judicial decision-makers like the Upper Tribunal, and so the imperative to read down or otherwise refuse to apply the ouster clause might be felt more powerfully than in Oceana and LA. Section 51 and section 53 IMA 2023 also include ouster clauses substantively similar to the clause in LA, seeking to make Upper Tribunal and Special Immigration Appeals Commission decisions immune from judicial review. Clearly different considerations will apply to these provisions when determining what effect should be attributed to them.
Does the IMA 2023 constitute a more direct conflict between the common law’s requirements of the rule of law on the one hand, and Parliament’s express will on the other, such that a choice between the two constitutional values will finally have to be made? Calls by the previous Home Secretary to amend the IMA 2023 following the Supreme Court’s Rwanda judgment to allow for even swifter deportations make this question yet more pertinent. As things stand, it still appears possible for Parliament to oust judicial review, Anisminic, Cart and Privacy International notwithstanding. Indeed, respect not only for parliamentary sovereignty but also the rule of law might be said to require as much. But provisions in the IMA 2023 might put this constitutional orthodoxy under new strain. Time will tell whether the Supreme Court maintains such a view, or whether Lord Carnwath’s apparent preference for common-law supremacy will finally win out. Perhaps, though, it will take an ouster clause even more brazen than that found in section 11A of the 2007 Tribunals, Courts and Enforcement Act—perhaps even more brazen than that in the Illegal Migration Act 2023—to get us to that point.
Dr Philip Murray, Fellow, College Lecturer and Director of Studies in Law, Robinson College, Cambridge.
(Suggested citation: P. Murray, ‘Ouster Clause Redux: The Court of Appeal’s Decision in LA (Albania)’, U.K. Const. L. Blog (21st November 2023) (available at https://ukconstitutionallaw.org/))
UK Constitutional Law Association 