Judicial Review and Courts Bill

Supplementary written evidence submitted by Dr Jonathan Morgan, Reader in English Law, University of Cambridge (JRCB12)

Judicial Review and Courts Bill: Public Bill Committee

Following my oral evidence to the Public Bill Committee on 2 November 2021 on Part I of the Judicial Review and Courts Bill, I hope it might be useful to provide a follow-up memorandum substantiating my points at greater length, with references.

Clause 1 (temporal effect of quashing orders)

1. I have written a pair of articles about Clause 1 that are freely available on the UK Constitutional Law Blog. These are (including hyperlinks):

"IRAL's missing remedy: Compensation for unlawfulness" U.K. Const. L. Blog (12th October 2021) https://ukconstitutionallaw.org/2021/10/12/jonathan-morgan-irals-missing-remedy-compensation-for-unlawfulness/

"In Praise of Flexibility: Clause 1 of the Judicial Review and Courts Bill 2021"

U.K. Const. L. Blog (23rd Sept. 2021)

https://ukconstitutionallaw.org/2021/09/23/jonathan-morgan-in-praise-of-flexibility-clause-1-of-the-judicial-review-and-courts-bill-2021/

2. Prior to IRAL being set up, I published an article in the Law Quarterly Review in 2019 (volume 135 at p.585): ‘"O Lord make me pure-But not yet": Granting time for the amendment of unlawful legislation’. I can supply copies of the article on request.

3. Since I have written at length elsewhere on the general question and specifically the drafting of Clause 1 as introduced, I can be relatively brief.

4. Clause 1 is a most desirable reform, overall. It will confer a beneficial discretion on courts to limit the temporal effect of quashing orders. (While some argue that the courts already enjoy this discretion, that claim is hard to reconcile with the Supreme Court’s decision in Ahmed v HM Treasury (No 2) [2010] UKSC 5; one can also contrast the explicit statutory power to make time-limited orders when impugning devolved legislation (e.g. in s.102, Scotland Act 1998)).

5. While the new remedial discretion will sometimes be highly beneficial, this will be so only in rare cases. Mostly courts should, and will, continue to quash with immediate retrospective effect. This is the ordinary concomitant of finding that a government body acted outside the limits of its powers (ultra vires).

6. It is difficult, perhaps impossible, exhaustively to define the rare or exceptional cases where postponed (or prospective-only) quashing orders would be appropriate. Broadly speaking, the court would need to be convinced that the benefits of postponement (e.g. for orderly public administration) clearly outweigh the inevitable hardship to the successful applicant for judicial review. Ahmed v HM Treasury itself provides a powerful example: had quashing been postponed, Parliament could have debated its response to the Supreme Court’s judgment in a considered fashion. As it was, with quashing being immediate, the government had little choice but to seek-and Parliament little choice but to grant-emergency retrospective approval for the asset-freezing orders that the Supreme Court had quashed. It is no criticism of Parliament to point out that the legislation received imperfect, hasty scrutiny. This was an unfortunate necessity given the urgency of the situation. It was the Supreme Court’s refusal to postpone quashing that created the unfortunate necessity. In cases of that kind in the future, Clause 1 could appropriately be used to permit an orderly legislative response rather than rushed emergency legislation.

7. Other examples include R (Hurley and Moore) v Secretary of State for BIS [2012] EWHC 201 (Admin) (a case where, in the absence of a power to quash in a time-limited fashion, the court declined to issue any remedy at all). There are cases considering the power to postpone quashing in the context of devolved legislatures-e.g. Salvesen v Riddell [2013] UKSC 22. In other legal jurisdictions too, such powers exist and are sparingly employed (for example in the jurisprudence of the Court of Justice of the European Union, to which UK law was of course subject until the recent withdrawal from the EU).

8. To give examples is not to define. Thus my first concern is with the proposed subsection (8) in the new s.29A Senior Courts Act 1981 (which Clause 1 would insert). Subs (8) issues a list of factors to guide the court. Despite the final ‘round up’ provision (including anything else of relevance), there is a danger that important factors may be omitted and controversy could arise in litigation about whether or not the court should take them into account. One could suggest many more relevant factors to list here (as other commentators have suggested). In my view however, subs (8) should be deleted entirely. The Administrative Court has always had a discretion over which remedy to grant, or whether to grant a remedy at all. In no other area is the court’s remedial discretion "guided" by legislation. It has not been suggested that the remedial discretion (in general) should be codified. I see no need to do so for the specific discretion to time-limit quashing orders. To the extent that guidance is needed, it would be more appropriately laid down in due course in the reasoning of courts when considering whether to make use of the power. Fully reasoned judgments provide much better guidance than a compressed statutory list.

9. (On this last point I align myself with J-E-M Portalis, a Commissioner for the drafting of the French Code Civil (Code Napoleon) and in turn with Lord Hailsham of St Marylebone. As Lord Chancellor, Lord Hailsham expressed the problem thus (on the particular subject of regulating ‘unreasonable contract terms’):

one must either give the court a fairly wide carte blanche in order to decide what is reasonable, or one is going to try to foresee in advance every kind of factor which may arise in a dispute of this kind, and see whether it can be put into the Bill. This, I think, is indeed an impossible task, and one which, in the event, would lead the court into an impossible position if one failed in a given case to specify all the possible considerations which might enter into the infinitely varied circumstances… [Monsieur Portalis said of drafting the Code Civil:] "We have guarded against the dangerous ambition of wishing to regulate and to foresee everything. The wants of society are so varied that it is impossible for the legislator to provide for every case or every emergency. … [The] office of the law is to fix by enlarged rules the general maxims of right and wrong, to establish firm principles fruitful in consequences, and not to descend to the detail of all questions which may arise upon each particular topic…" [1]

(Lord Hailsham’s / Portalis’ warnings apply equally to Clause 1 (subs (8)) of the present Judicial Review and Courts Bill).

10. A more serious objection arises about subs (9). As stated above, it will only rarely be appropriate for the court to issue a time-limited quashing order. Given that occasions for using the power will be infrequent and unusual, it seems quite inappropriate to set up a presumption in favour of postponed (or prospective-only) quashing in subs (9).

11. As it stands, subs (9) embodies such a weak presumption that it achieves little of substance, and its main function would be to complicate litigation (thus increasing expense to litigants and wasting court time). Courts would rarely find that a postponed order would provide ‘adequate redress’. It would usually NOT be ‘adequate redress’ to hold that a successful applicant for judicial review continued to be governed by a decision held to be ultra vires and unlawful. So the presumption would rarely be triggered. And if it were triggered, the court would usually find there was a ‘good reason’ not to time-limit the quashing order. Namely the principle that public bodies must act within the limits of their legal powers, and that decisions outside those powers are invalid from the outset, and it is the court’s constitutional function to give effect to that invalidity (i.e. the ultra vires rule).

12. If I am wrong and subs (9) has teeth, then it becomes pernicious. It could constrain the court to postpone or make prospective a remedy which the court would otherwise have made retrospective and immediate. Such fettering of discretion would be incongruous when Clause 1’s rationale is to enlarge the discretion of the Administrative Court. By granting a discretion but then seeking to fetter it, the present draft bill approaches self-contradiction.

13. For these reasons I strongly recommend that subs (9) should be deleted.

14. Thirdly I strongly recommend an express power for the court to make payment of compensation a condition for the postponement or limited retrospection of a quashing order. This would help address concerns about the potential injustice of such orders.

15. It must be recognised that preserving in force an unlawful decision (in advance of its eventual and/or prospective quashing) could frequently perpetuate unlawfully-imposed hardship on the successful applicant for judicial review, and others in a similar position. (For example those deprived of a licence necessary for their livelihood by unlawful decision-making; those refused asylum and ordered to be deported under an unlawful government policy). The court could decide that such hardship to individuals was justified to avoid the chaos of immediate retrospective quashing (see discussion of Ahmed above, and the facts of Hurley & Moore). But only rarely would this be appropriate. Court could be considerably more willing to use the new discretion about quashing if they were able to order payment of compensation as a condition for postponement or limited retrospection.

16. This is quite familiar in other branches of the law. For example in property disputes, courts sometimes suspend the enforcement of an injunction (e.g. in tort claims in respect of trespass or nuisance). Suspension of the injunction may give the defendant time to relocate the offending activity. But it is usual for the court to order that the defendant pay compensation to the claimant for losses attributable to the ongoing interference with the claimant’s rights, during the period of suspension. (The power for courts of equity to grant damages in lieu of an injunction is longstanding, having been introduced by the Chancery Amendment Act 1858)

17. Courts of equity are then more willing to refuse or postpone ‘specific relief’ (e.g. injunctions) given that they can award compensation in lieu of such orders.

18. If the Government wishes to ‘encourage’ courts to consider making use of the power of time-limiting quashing orders, a power to award compensation in lieu would be a more effective way to achieve this than the current presumption in subs (8)-the deletion of which I have recommended above.

19. In some cases, compensation could ‘square the circle’. The court might postpone quashing for pressing reasons of public interest (for example preserving in force a wide-ranging policy, such as the tuition fees scheme in Hurley & Moore). Potential injustice to the successful applicant for judicial review could be reduced, perhaps removed entirely, by an order that the applicant be compensated for losses sustained as a result of the unlawful decision’s preservation in force.

20. I accept however (as Professor Feldman commented in his oral evidence to the Committee on 2 November 2021) that not all situations of hardship can be adequately remedied by compensation. It would seem a poor remedy, for example, to offer compensation to an asylum-seeker who had actually been deported from the United Kingdom pursuant to an unlawful policy which the court had declined immediately to quash. In other situations however (such as a trader deprived of a necessary licence under an unlawful policy), where the loss would be economic in nature, it could aptly be compensated by monetary payment.

21. Clause 1 of the Bill currently empowers the court to make time-limitation of a quashing order ‘subject to conditions’: subs (2). The Bill’s Explanatory Notes (para 87) state: ‘The clause provides no limit or prescription on the type or nature of the conditions, leaving this determination to the court’. While that is correct, I doubt that the court would be able to order compensation as a condition for the preservation in force of an unlawful act. It is an axiom of administrative law that compensation is NOT available for losses arising through an unlawful act per se (i.e. unless a tort, breach of contract etc can additionally be shown). Given that axiom (which I personally regret, without doubting its binding force), it is at the very least uncertain whether a court would, or even could, make payment of compensation one of the ‘conditions’ to which an order under subs (1) could be made subject.

22. For these reasons, I recommend that subs (2) should be amended to read: (2) ‘Provision included in a quashing order under subsection (1) may be made subject to conditions, including the payment of compensation to the applicant, or other affected persons, for losses sustained by the preservation in force of the impugned act’ (words in bold added).

Clause 2 (Cart judicial review)

23. Clause 2 is a legitimate response to a longstanding problem of the how much court time should be allocated to appeals on immigration and asylum matters (in particular). Legislation to overrule the Cart decision is justified on at least two grounds. First, that the Supreme Court in Cart misunderstood Parliament’s intention in the 2007 Act, in part because of deficiencies in the 2007 Act’s drafting. Parliament should now restate its intention to insulate decision of the Upper Tribunal in unmistakable terms. A second justification is that insofar as the Supreme Court in Cart tried to come up with a pragmatic compromise for the proportional deployment of judicial resources to immigration and asylum appeals, this was always a matter more appropriate for parliamentary resolution. It raises questions of resource allocation which are properly governmental in nature. The Supreme Court made a good faith effort to reach an acceptable compromise, but a decade’s experience with Cart judicial review indicates that the compromise has not worked as the Supreme Court hoped. Again, it is legitimate for Parliament now to intervene.

24. It is important to stress that Cart, and its proposed reversal, concerns reviews and appeals within the judiciary. Thus the ‘ouster’ provisions in Clause 2 do not provide either a helpful ‘template’ (pace the Government’s claims) or a worrying precedent (pace the Government’s critics’ claims) for future ‘ouster clauses’ in quite different areas. The effectiveness of an ouster clause cannot depend solely on the words used. It also depends on the context of the legislative scheme as a whole-what kind of body is being immunised from challenge with what alternative safeguards (i.e. as an alternative to judicial review by the High Court).

What did Parliament intend in the Tribunals, Courts and Enforcement Act 2007?

25. It is likely that in the 2007 legislation setting up the present system of tribunals, Parliament intended that there should not be judicial review (JR) of (or further appeal from) the Upper Tribunal’s decision not to grant permission to appeal from decisions of the First Instance Tribunal. This is offered less as a criticism of the Supreme Court’s decision in Cart [2011] UKSC 28, more a reflection on the ambiguous drafting of the Tribunals, Courts and Enforcement Act 2007.

26. The 2007 Act essentially implemented the recommendations of the committee chaired by Sir Andrew Leggatt, Tribunals for users-One System, One Service (2001) (‘Leggatt’). Pertinently for Cart JR, Leggatt had proposed that the Upper Tribunal be immune from JR altogether. To adopt the account of Lord Hope in Cart’s Scottish twin, Eba v Advocate General: [2]

In para 6.31 [the Leggatt Report] offered two options for the removal of judicial review from the Upper Tribunal. One was to constitute all the appeal tribunals a superior court of record... The other was to exclude judicial review by express statutory provision.

In the event the 2007 Act does not contain any provision which excludes judicial review of decisions of either the First-tier or the Upper Tribunal. It provides instead that the Upper Tribunal is to be a superior court of record: section 3(5).

27. In other words, Parliament in 2007 chose one of the Leggatt ‘options for the removal of judicial review’. But, with the advantage of hindsight, it chose wrong. As Lord Phillips (President of the Supreme Court) commented in Cart, the Leggatt Report ’commented, erroneously, that this goal could be achieved by making the [Upper Tribunal] a superior court of record’. [3]

28. The suggestion, then, is that Parliament had intended to protect Upper Tribunal decisions from JR by using Leggatt’s recommended ‘superior court of record’ formula. But as ouster clauses go, this one was distinctly ambiguous. In the Divisional Court in Cart, Lord Justice Laws observed that only ‘the most clear and explicit words’ could oust JR; designation as a ‘superior court of record’ was not sufficiently clear and explicit (indeed it was a ‘constitutional solecism’ to suggest otherwise); ‘it cannot be supposed that judicial review may be ousted by an implication, far less one contained in a formula which amounts in effect to a deeming provision’ (i.e. by deeming the Upper Tribunal to be a superior court of record). [4] The government accepted Laws LJ’s reasoning as unanswerable when the Cart case reached the Supreme Court.

29. It seems that Parliament in 2007 acted on the misapprehension that by designating the Upper Tribunal a ‘superior court of record’ it would preclude JR of its decisions by the High Court-a misapprehension shared by (or generated by) the Leggatt Report on Tribunals. In which case, the amendment to the 2007 Act proposed in Clause 2 of the present JR and Courts Bill would simply restore what Parliament (and Leggatt) had initially intended: that the tribunal system should be ‘wholly self-sufficient’, [5] and not under the JR jurisdiction of the High Court.

Achieving ‘Proportionate Dispute Resolution’

30. As is well known, Cart JR does not proceed on full, ordinary JR principles (i.e. JR for any error of law) but only when the more restrictive ‘second-tier appeal’ principles are met.

31. The Supreme Court’s justification for this was realistic and pragmatic. The judges recognised the extremely high caseload generated by challenges in asylum and immigration cases (in particular). Baroness Hale explained the reason for the volume of challenges:

In most tribunal cases, a claimant will have little to gain by pressing ahead with a well-nigh hopeless case. ... But in immigration and asylum cases, the claimant may well have to leave the country if he comes to the end of the road. There is every incentive to make the road as long as possible, to take every possible point, and to make every possible application. This is not a criticism. People who perceive their situation to be desperate are scarcely to be blamed for taking full advantage of the legal claims available to them. But the courts’ resources are not unlimited and it is well known that the High Court and Court of Appeal were overwhelmed with judicial review applications in immigration and asylum cases until the introduction of statutory reviews. [6]

32. Lady Hale repeated the point about limited judicial resources:

There must be a limit to the resources which the legal system can devote to the task of trying to get the decision right in any individual case. There must be a limit to the number of times a party can ask a judge to look at a question. [7]

33. It is a point accepted by the courts in many contexts. For example in the tort case Sutradhar v NERC (striking out a negligence claim at a preliminary stage rather than permitting it to go to trial), Lord Hoffmann emphasised that ‘compromises’ were necessary:

It is not the case that the administration of justice, alone among the services provided by the state, is exempt from any considerations of cost … however unpalatable it may be to those who think that justice is priceless. [8]

34. The point was echoed by other members of the Supreme Court in Cart. Perhaps most memorably in Lord Brown’s observation:

The rule of law is weakened, not strengthened, if a disproportionate part of the courts' resources is devoted to finding a very occasional grain of wheat on a threshing floor full of chaff. [9]

35. Thus the Supreme Court squarely acknowledged the undesirability of an overly wide JR jurisdiction. This would be ‘disproportionate’ given the inevitable constraints of limited judicial resources (i.e. court time and judicial ‘manpower’).

36. Once the question has been framed this way, it is in my view obvious that Parliament and the government are at least as competent to address the question as was the Supreme Court. Indeed Parliament ultimately controls the power of the purse (taxation and the allocation of tax revenues), and matters about the level and allocation of public spending are better resolved in Parliament than in the courts. Secondly, Parliament is certainly entitled to revisit the pragmatic balance struck by the Supreme Court in Cart, and to conclude that despite the court’s declared intention Cart JR has ultimately led to disproportionate review of Upper Tribunal appeal decisions. The Supreme Court made a good faith attempt to impose proportionate dispute resolution, but in the light of experience its scheme based on ‘second-tier appeal criteria’ has failed.

37. One advantage that Parliament and Government have over the courts is the evidence at their disposal. A court is the prisoner of the arguments, and evidence, put before it by the parties. The judges cannot commission, let alone undertake, research of their own. On the present point, both IRAL and latterly the Government have put forward figures on the success rate of Cart JRs. These figures differ from each other, and even the Government’s later study (which is manifestly preferable to the statistically flawed IRAL figure) has been challenged. The Second Reading Debate on the Bill also saw the cost of Cart JR framed in quite different (but not necessarily inconsistent) terms. The Lord Chancellor noted that Cart JR annually consumes 180 judge-days. His shadow, David Lammy MP, quantified the annual cost as £400,000 (within a total Government Legal Department budget of £226.7 million). It is, I hope, uncontroversial to suggest that if Parliament is to fulfil its superior potential to resolve cost-benefit questions such as ‘proportionate dispute resolution’ it needs to act upon full, accurate and robust data.

38. Apart from the evident difficulty of quantifying and costing Cart ‘success’, there is a qualitative point. As the Government accepts (and as the Government’s critics emphasise), some Cart JRs DO succeed. However small in absolute numbers, or as a proportion, each successful case must have revealed an error of law of general public importance, or a seriously flawed initial decision by the Upper Tribunal in the particular case (or both). (This is necessary for the High Court to grant a Cart JR using the second-tier appeal criteria.) In such sensitive areas as asylum, immigration, or social security entitlements, it is no exaggeration to say that every successful Cart JR corrects a life-changing mistake by the tribunal system. Parliament must squarely confront this point: the price of economising on (admittedly scarce) judicial resources is that in such cases in future, life-changing mistakes will go uncorrected. Ultimately this requires a very difficult political judgement. It is better taken by Parliament than by the judges.

Clause 2: A ‘Model’ Ouster Clause?

39. The Government has suggested the ouster clause in Clause 2 could be a template or ‘framework’ for ouster clauses in other situations. This has attracted fierce criticisms (in the Second Reading Debate, Mr Lammy thought this ‘truly chilling’: that in proffering this ‘murkier’ justification for Clause 2, the government had ‘let their mask slip’).

40. I consider the Government is wrong to see Clause 2 on Cart JR as any sort of template for future ouster clauses. If I am correct, criticisms of the Government for preparing a wider assault on the rule of law (i.e. to remove JR) fall away.

41. Crucially, Cart concerns a matter within the judiciary. The post-2007 tribunals form one branch of the judicial system. The debate about Cart concerns the correct location and number of rights to appeal within the judicial system. The ouster in Clause 2, if enacted, would not be immunising ministerial, governmental or administrative decisions from judicial scrutiny. It would be preventing JR of the decision of one judicial body (the Upper Tribunal) by another judicial body (the Administrative Court).

42. For this reason, I predict that the ouster in Clause 2 would achieve its aim if enacted (although predicting the success of ouster clauses is a perilous business given the unpredictability of the judicial treatment of them). My prediction rests on two factors. The wording and scope of the Clause but also, crucially, its context.

43. The wording of Clause 2: it is not an absolute ouster of judicial review on all grounds, but would still permit challenge to Upper Tribunal decisions flawed by ‘truly jurisdictional errors’ (my paraphrase of the grounds of challenge in subs (4) of new s.11A). This contrasts with the absolute exclusion of judicial review once proposed in the original Clause 11 of the Asylum and Immigration Bill 2004. Lord Donaldson (a former Master of the Rolls) criticised that clause as ‘a constitutional outrage and an affront to the rule of law’. [10] Lord Donaldson spoke for many critics, and in the face of such opposition the clause was dropped from the Bill.

44. The context of Clause 2 in the present Bill is no less crucial. As noted above, it is a qualified ouster of JR of judicial decisions. Even if Clause 2 abolishes Cart JR, a party disappointed by a decision of the First Instance Tribunal still has two opportunities to challenge it. They may request permission to appeal from the First Instance Tribunal itself. Failing which they may apply to the Upper Tribunal for permission to appeal to it.

45. In their Cambridge Law Journal commentary on the Cart decision, Mark Elliott and Robert Thomas reason about ouster clauses (and judicial suspicion of them) as follows:

the true objection to bald privative clauses is not that they oust the [High Court’s] supervisory jurisdiction itself, but that they threaten the rule of law by excluding from the purview of any independent judicial body of appropriate constitutional stature the interpretation of the relevant legislation and the supervision of the statutory body concerned. Consequently, when exclusion of judicial review is accompanied by some alternative arrangement for oversight that meets the minimum requirements of the rule of law, there is no constitutional need, or justification, for judicial intransigence, whether by means of outright disobedience or strained interpretation [of the ouster clause]. [11]

46. It is strongly arguable that the ouster in Clause 2 is justifiable on these grounds (thus making ‘judicial intransigence’ unjustifiable). Viz that the two opportunities to apply for permission to appeal, to both First Instance and Upper Tribunals, constitute sufficient ‘alternative arrangement for oversight’ by an ‘independent judicial body of appropriate constitutional stature’.

47. Yet if precisely the same wording as Clause 2 were used in a different context to preclude JR of (say) ministerial decisions, the test laid down by Elliott and Thomas would manifestly not be satisfied. Such a ministerial ouster would, clearly, engage all the concerns which (on those authors’ view) WOULD justify an ouster’s ‘intransigent’ treatment by courts.

48. One can see why the government wishes to develop ‘common, fixed and stable’ language and categories by which Parliament can clearly communicate its intentions about ouster clauses. However, in my view, an ouster clause’s compatibility with the rule of law can never depend solely on its wording. It must also depend on its context. (A point that Sir Stephen Laws emphasised in his oral evidence on 2 November 2021). Thus there cannot be a linguistic template for ouster clauses, a particular form of words that can simply be copied on every occasion when Parliament wishes to protect decisions from judicial review.

49. Ultimately, Clause 2 does not set any far-reaching precedent for future ouster clauses. This conclusion will be to the Government’s disappointment and to the relief of the Government’s critics.

Conclusion: Other measures?

50. I have indicated above that Clause 1 is strongly welcome (although capable of significant improvement by amendments to the drafting), and Clause 2 is a legitimate response to a longstanding and difficult problem.

51. Both clauses reverse a decision of the UK Supreme Court. Respectively Ahmed (No 2) and Cart. I wish to say in conclusion that it is clearly legitimate for Parliament to reverse judicial decisions with which it disagrees. Particularly in a situation like Cart, where Parliament would overrule a judicial decision about the meaning of a statute. Where Parliament considers that the court has not, in fact, correctly interpreted a statutory provision consistently with what Parliament intended, it is a corollary of Parliament’s legislative authority that it can intervene to put right the judicial (mis-) interpretation.

52. I do not wish to extend a lengthy consideration of the provisions currently in Part I of the Bill by considering other measures that it does not at present contain. I will simply say that I have read the various additional proposals made by Professor Richard Ekins in his recent paper for Policy Exchange. [12] I see considerable merit in all these proposals. Many of them fall squarely within the rationale identified in the previous paragraph: legitimate parliamentary correction of a judicial misinterpretation of statutory intent. Each of the Ekins proposals would of course need careful scrutiny on its merits: I cannot embark on that here. If they were thought meritorious, there can be no doubt about their constitutional legitimacy.

Dr Jonathan Morgan

Reader in English Law, University of Cambridge

Corpus Christi College, Cambridge CB2 1RH

10 November 2021