Executive Summary

The Safety of Rwanda (Asylum and Migration) Bill is a legitimate and broadly effective response to the Supreme Court’s recent Rwanda judgment.  It should be supported by the House of Commons when it receives its second reading on Tuesday 12 December.  The Bill is necessary to make it lawful for asylum-seekers to be removed to Rwanda.

The Bill builds on the UK’s new treaty with Rwanda, a Treaty which meets the Supreme Court’s concerns about the risk that Rwandan officials will make mistakes in processing asylum claims and may wrongly send a genuine refugee back to an unsafe country.  Parliament is entitled to give legislative effect to its judgement that, in view of the change in facts since the Supreme Court’s judgment, asylum-seekers may be removed to Rwanda without putting them in danger.

While the Bill is an intelligent response to the Supreme Court’s judgment, it does not fully address the risk of further litigation, which may challenge the foundations of the Rwanda plan or may frustrate its operation in practice. However, these are soluble problems and this paper sets out a series of careful amendments that will make the Bill an effective means to secure the intended end.

The Bill disapplies some but not all provisions of the Human Rights Act 1998 (HRA).  The fundamental premise of the Bill – that it is necessary and appropriate to give legislative effect to Parliament’s judgement that removals to Rwanda do not put asylum-seekers in danger – is put in jeopardy by the Bill’s failure to disapply section 4 of the HRA, which authorises courts to declare legislation incompatible with Convention rights.  The Bill should be amended to disapply section 4 of the HRA, or at least to require courts to treat Rwanda as a safe country in proceedings seeking a declaration of incompatibility.  If it is not amended in this way, the courts will be able to question Parliament’s judgement and denounce its legislative choice.  This would place immense pressure on the Government to stop removals to Rwanda and would weaken the UK’s capacity to defend the policy in litigation before the European Court of Human Rights in Strasbourg.

The Bill should also disapply section 10 of the HRA, which empowers a Minister by order to amend legislation when a UK court has made a declaration of incompatibility or when it appears to the Minister, in view of a judgment of the Strasbourg Court, that the legislation is incompatible with the UK’s obligations under the European Convention on Human Rights.  Parliament should not permit ministers to change, or effectively to abandon, the policy of this legislation by order.  Any such change, or abandonment, should require new primary legislation.

Clause 4 of the Bill makes it possible for persons facing removal to Rwanda to continue to argue that even if Rwanda is safe in general, it will not be safe for them.  The clause distinguishes between general arguments about the safety of Rwanda and the particular circumstances of an individual person.  The risk is that in practice this distinction may prove difficult to identify and to maintain and that many asylum-seekers will be able to frustrate their removal from the UK by raising objections to removal that tacitly question whether Rwanda can be trusted to honour its assurances.  The solution is to amend clause 4 to require the individual circumstances in question to involve or to relate to a prior connection to Rwanda and/or to make it clear that removals from the UK can only be blocked when there is a real risk that the Rwandan government itself will persecute the person.  Note that quite apart from these grounds to limit removals to Rwanda, no person should be removed from the UK, whether to Rwanda or any other country, unless and until he or she is fit to fly.  This of course concerns fitness to fly rather than whether Rwanda is a safe country (the means of removal, not the destination).

These amendments are consistent with the policy of the Bill and would not place the UK in breach of its international obligations.  In any case, it is for Government and Parliament to decide when, or whether, to enact legislation that would place the UK in breach of its international obligations – or would risk being understood to breach those obligations.  The Bill should be amended in the ways that this paper recommends.  The amended Bill would not provoke the Supreme Court to question the validity of the legislation, as some commentators have wrongly suggested.  Neither would the amendments we propose provide any reasonable basis for Rwanda to withdraw from the Treaty.

Introduction

The Safety of Rwanda (Asylum and Migration) Bill was published on Wednesday 6 December and has its second reading in the House of Commons on Tuesday 12 December.  This research note considers the main provisions of the Bill and evaluates the likelihood that the Bill will enable prompt removal of asylum-seekers to Rwanda without the policy being frustrated, directly or indirectly, in the courts.

We argue that the Bill is a legitimate and, in general and in principle, effective response to the Supreme Court’s recent judgment, which held that Rwanda was not a safe state because of the risk that some asylum-seekers would wrongly be removed from Rwanda and sent to another country in which they would be in danger.  The Bill builds on the new Rwanda Treaty,¹The “Rwanda Treaty” is defined in clause 7 of the Bill as the agreement between the Government of the United Kingdom of Great Britain and Northern Ireland and the Government of the Republic of Rwanda for the provision of an asylum partnership to strengthen shared international commitments on the protection of refugees and migrants, signed at Kigali on 5 December 2023. https://www.gov.uk/government/publications/uk-rwanda-treaty-provision-of-an-asylum-partnership. which responds to the Supreme Court’s judgment by providing categorical assurances, inter alia, that no asylum-seeker will be removed from Rwanda to any other country apart from the UK.  That guarantee against removal to an unsafe country entirely changes the and supersedes the factual foundation for the Supreme Court’s judgment (and the Court itself stressed that its judgment related only to the state of facts existing at the time of the litigation’s first phase).  In no way whatever does the Bill contradict, challenge or even put in question the Supreme Court’s judgment.  In enacting the Bill, Parliament will give legislative effect to its own judgement that once the Treaty comes into force Rwanda is a safe country and that asylum-seekers may be removed there without placing them in danger and, in particular, without any real risk that they will be sent on to an unsafe third country.  The Bill requires decision-makers, including courts, conclusively to treat Rwanda as a safe country, and specifically not to consider whether Rwanda will send asylum-seekers to an unsafe third country.

In giving effect to Parliament’s judgement that Rwanda is a safe country, the Bill disapplies certain provisions of the Human Rights Act 1998 (HRA).  However, the Bill does not disapply section 4 of the HRA and does not require courts conclusively to treat Rwanda as a safe country in the course of litigation seeking a judicial declaration that the Bill is incompatible with Convention rights.  Thus, the Bill leaves it open to the UK courts to review and contradict Parliament’s judgement and to declare that removing asylum-seekers to Rwanda breaches Convention rights and so would place the UK in breach of its international obligations under the European Convention on Human Rights (ECHR).  The Bill should be amended to prevent courts from second-guessing Parliament’s legislative choices.  The proper solution is for the Bill to disapply section 4 of the HRA altogether; a distinctly second-best amendment would be to require the courts conclusively to treat Rwanda as safe for the purposes of an application for a declaration of incompatibility.

Taken together with the Treaty, the Bill is an effective answer to the Supreme Court’s judgment that Rwanda is unsafe because of a risk that asylum-seekers will be sent on from Rwanda to unsafe third countries.  However, the Bill leaves it open to each asylum-seeker facing removal to Rwanda to argue that Rwanda is an unsafe country for him or her in particular.  The Bill aims to impose a high threshold in relation to such claims and to require the claimants to avoid arguments that Rwanda is generally an unsafe country.  Nonetheless, there is a real risk that the courts will allow such claims in many individual cases, accepting arguments that in effect reopen the question about whether Rwanda is a safe country.  If courts act in this way that would undercut Parliament’s judgement and the policy to which the legislation aims to give effect.  The Bill should be amended to set out more precisely the grounds on which courts may consider risks relating specifically to a person’s individual circumstances.

The legitimacy of Parliament’s judgement that Rwanda is a safe country

The Supreme Court’s judgment held that Rwanda was not a safe country because there was a real risk that the Rwandan asylum system would mishandle asylum claims and would wrongly send some genuine refugees back to their country of origin where they would face mistreatment.²R (AAA) v Secretary for the Home Depart­ment [2023] UKSC 42.   The Court accepted that it was lawful, and consistent with the Refugee Convention 1950, to remove asylum-seekers from the UK and to send them to a safe third country.  However, the Court was sceptical about the accuracy and integrity of the Rwandan asylum system and, while the Court said that it was not questioning Rwanda’s good faith, it also doubted whether Rwanda would honour the assurances that it had made to the UK,³Id. and cast doubt on the independence of the Rwandan judiciary.⁴Id., paras 82-83. The Supreme Court relied heavily on the evidence of the United Nation High Commissioner for Refugees (UNHCR) in concluding that there was a real risk that the Rwandan asylum system would misfire.

The contestable nature of the Supreme Court’s conclusion is partly illustrated by the fact that the two judges of the Divisional Court and the Lord Chief Justice of England and Wales sitting in the Court of Appeal reached the opposite conclusion on the same evidence. The former concluded that the Home Secretary’s conclusion that Rwanda is a safe country, and the Home Secretary’s decision to give significant weight to assurances given by the Rwandan government in reaching this conclusion, was lawful.⁵R (AAA) v Secretary for the Home Depart­ment [2022] EWHC 3230 (Admin) paras 62-71. In contrast to the Supreme Court, the Divisional Court was sceptical about affording special weight to the UNHCR’s evidence about the deficiencies of the Rwandan asylum system. The judges of the Divisional Court, Lewis LJ and Swift J, pointed to published statements made by the UNHCR itself in July 2020 in respect of Rwanda’s asylum system that described its legal framework as “fully compliant with international standards” and with no suggestion of any “protection gap”.⁶Id., para 55. Lewis LJ and Swift J noted that while this 2020 statement did contain some criticisms of the Rwandan government, there was “no hint in that document of any concern of the order that might prompt the conclusion that Rwanda could not be relied on to comply with its obligations under the Refugee Convention.”⁷Paras 55, 70. The Divisional Court were satisfied that they were given no good explanation to account for the UNHCR’s marked and swift change in the critical nature of their assessment of the compliance of Rwanda’s asylum system with international law.⁸Id., para 70. In the Court of Appeal, the Lord Chief Justice of England and Wales dissented from the majority on the grounds that there were good reasons to expect Rwanda to comply with the terms of its agreement with the UK, not least because of the incentives the agreement provided and the system for monitoring its operation in practice.  He was not willing to conclude that this new administrative scheme, the success of which rested in part on diplomatic considerations, was unsafe.⁹R (AAA) v Secretary for the Home Depart­ment [2023] EWCA Civ 745, paras 517, 527.

While the Government has said that it disagrees with the Supreme Court’s conclusion, it is important to stress that the new Treaty agreed between the UK and Rwanda does not ignore the Supreme Court’s conclusions nor its treatment of the evidence. Indeed, the Government has acted precisely as if it agreed with the Supreme Court’s conclusion.  It has acted to eliminate the risk on which the Supreme Court passed judgment.  The Treaty directly addresses the risks that the Supreme Court identified by providing a categorical assurance that no asylum-seekers will be removed from Rwanda to any other country than the UK.  That is, even if their asylum claims fail, they will not be removed to an unsafe third country, including their country of origin.  (In addition, the Treaty provides for changes to the Rwandan system for processing asylum claims, which provide greater procedural protections and similar.)  Thus, any risk of mistakes in processing resulting in unintended refoulement has been directly addressed.  The Treaty imposes an obligation on Rwanda that otherwise does not exist in international law, including under the terms of the Refugee Convention 1950, namely to refrain from removing failed asylum-seekers to any country other than the UK.  Monitoring compliance with this obligation is much more straightforward than determining whether Rwandan officials were wrong to reject a particular asylum claim, although the Treaty also makes provision for reducing the risk of refugee status wrongly being refused, as well as virtually eliminating the risk that a wrong decision could have the Convention-breaching consequence which was the Supreme Court judgment’s concern.

The evidential foundation for the risk assessment made by the Supreme Court has been superseded as a result of a period of diplomatic negotiation and the agreement of a legally binding international agreement.  It is thus perfectly appropriate for Parliament to legislate on the basis of the new situation and to settle authoritatively that Rwanda is a safe country and that the Home Secretary may lawfully remove asylum-seekers to Rwanda for their claims to be processed and for settlement.  Legislating in this way displays no disrespect, unconstitutional or otherwise, for the jurisdiction of the Supreme Court or for the findings the Court reached in its recent judgment.

The significance of the Bill is that it does not leave to the courts the function of determining whether the new Treaty is sufficient to make Rwanda a safe country.  That is, it does not make the lawfulness of removals to Rwanda conditional on the outcome of future litigation in which the Supreme Court would be invited by claimants, with the support of interested parties such as the UNHCR, to cast doubt on whether Rwanda will honour the Treaty.  It is perfectly proper for Parliament to agree with the Government that there are good reasons to expect Rwanda to honour its treaty obligation (which will be monitored and supported by UK officials), so that persons sent to Rwanda will not be at any real risk of being removed to an unsafe third country.  It would be inappropriate, in constitutional and practical terms, to continue to leave to the courts the ultimate responsibility for making the final and conclusive decision, in the course of judicial proceedings, about whether Rwanda is to be trusted or whether the Treaty is likely to prove to be a sufficient guarantee.

In clause 1, the Bill recites the various commitments that Rwanda has made in the Treaty and defines a safe country for the purposes of this legislation to be a country to which a person may be removed in compliance with the UK’s international obligations and, more specifically, a country from which a person will not be moved on in breach of the Refugee Convention.  Clause 2 requires decision-makers to treat Rwanda as a safe country.  This applies to the Secretary of State and to immigration officers when making a decision to remove a person to Rwanda and to a court or tribunal considering any such decision.  The clause makes clear that a court or tribunal may not consider a review of, or an appeal against, a decision of the Secretary of State or immigration officers on the grounds that Rwanda is not a safe country.

Clause 4(1) of the Bill carves out an exception to clause 2 by providing that a decision maker may conclude that Rwanda is not a safe country for the removal of a particular person, provided that this conclusion is based on compelling evidence relating specifically to the person’s particular individual circumstances, rather than on the grounds that Rwanda is not a safe country in general.  Clause 4(2) says that clause 4(1) does not permit a decision maker to consider whether Rwanda will send a person to an unsafe third country in breach of its international obligations.  Thus, the legislation makes quite clear that no court or tribunal will be able to question whether Rwanda is likely to send asylum-seekers, removed to Rwanda in accordance with the Treaty, to an unsafe third country.  Parliament will have given clear legislative effect to its judgement that Rwanda can be relied upon not to act in this way.

For many critics of the Bill, this is a legislative usurpation of the judicial function.  The point is a serious one in principle, but in relation to the present issue is entirely misconceived.  Parliamentary sovereignty entails that the King-in-Parliament may enact any law, but of course there are many laws that it should not enact, including for reasons of constitutional principle and/or comity to other institutions – or other states.  However, Parliament is entitled to legislate based on its own understanding of the relevant facts and its evaluation of their significance and the legal effects which should flow from them.¹⁰It has long been accepted that it is an integral element of the doctrine of parliamentary sovereignty that the factual premise on which Parliament chooses to legislate should not be capable of being questioned in judicial proceedings; British Railways Board v Pickin [1974] AC 765.

When the facts in question concern not a particular past event, but rather an assessment of how a new administrative scheme is likely to operate in the future, it is reasonable for Parliament to act on its own different assessment of the inferences about risk that should be drawn from the evidence and, as always, to prescribe the evidence that should be admissible for that purpose and the weight that should be given to it.  In any event, in developing this particular legislative proposal, the Government has not disagreed with the Supreme Court.  On the contrary, it has taken the Supreme Court’s judgment seriously and has met the Court’s concerns – satisfied its implicit requirements – by agreeing the Treaty.  In legislating in reliance on the Treaty, Parliament takes responsibility for deciding that asylum-seekers may reasonably be sent to Rwanda for processing and settlement – a decision the continuing force of which will be something for which Ministers will remain accountable to Parliament.  In enacting legislation to give effect to this policy, Parliament is not required to, and should not, make the policy’s implementation conditional on the outcome of future litigation, in which judges will again speculate about the likelihood that another friendly state, and Commonwealth country, may not be capable of being trusted to honour its assurances, or about the independence of its judiciary.

The rule of law and international law

Because a person’s removal to Rwanda will not result in wrongful return to a refugee’s country of origin –refoulement – the Government is on firm ground in saying that the Bill is compatible with international law, in particular with the Refugee Convention 1950.  However, it is important to note that even if the Government did take the view that provisions of the Bill were clearly contrary to propositions of international law, including the ECHR, it would not follow that the legislation was for that reason alone illegitimate.  There is no constitutional rule that the Government or Parliament must never act in ways that place the UK in breach of international law, or, relatedly, that the risks to the UK inherent in such breach or potential (or possible) breach must never be run. As we noted in a recent paper,¹¹Richard Ekins KC (Hon), Sir Stephen Laws KC (Hon), Conor Casey, Government Lawyers, the Civil Service Code, and the Rule of Law (Policy Exchange, December 2023), paras 7 and 10. the constitutional and legal position in the UK, so far as international law is concerned, is quite clear. It has been repeatedly reaffirmed by our highest courts (previously the Appellate Committee of the House of Lords and more recently the Supreme Court),¹²JH Rayner (Mincing Lane) Ltd v Department of Trade and Industry [1990] 2 AC 418, 499; R (Miller) v Secretary of State for Exiting the European Union (Birnie intervening) [2017] UKSC 5; [2018] AC 61, paras 56, 167 and 244, para 78. and indeed is also reinforced by what is said in the Supreme Court’s judgment in the Rwanda case itself.¹³In R (AAA) v Secretary for the Home Department [2023] UKSC 42 paras 140-144 clearly reaffirm the supremacy of Parliament and its capacity, with clear and unambiguous words, to change the law. The UK is a dualist state.  Treaty obligations have effect in domestic law only so far as they are expressly incorporated into domestic law by or under an Act of Parliament; and, once incorporated, the relevant law is just as capable of being modified as any other domestic law, and in the same way.

The core of the rule of law in our constitution is that “Ministers can neither claim any immunity, by virtue simply of their office, from the rules of common law, nor by any decree or order impose a legal duty (or relieve anyone of a legal duty), except to the extent that an Act of Parliament authorizes them to do so.”¹⁴John Finnis, “Ministers, International Law, and the Rule of Law” (Policy Exchange, 2 November 2015).   The integrity of our constitution is put in jeopardy when civil servants and Ministers assume that legislation that is, or risks being found to be, incompatible with the UK’s international obligations cannot legitimately be put before Parliament or enacted.

The risks of the courts questioning Parliament’s judgement

The Bill disapplies some key provisions of the Human Rights Act 1998, namely section 2 (interpretation of convention rights), section 3 (interpretation of legislation) and sections 6-9 (acts of public authorities).  However, the Bill does not disapply section 4 of the HRA (declaration of incompatibility), nor does it provide that Rwanda must be treated as a safe country for the purposes of an application for a declaration of incompatibility under section 4.  What this means is that opponents of the Rwanda plan will be able to apply to the High Court – and thence to the Court of Appeal and Supreme Court – for a declaration that the Safety of Rwanda (Asylum and Migration) Act 2024, as it will be then, is incompatible with Convention rights.  The grounds for the challenge, as the Bill stands, would be first and foremost that Rwanda is not in fact a safe state, that Rwanda’s assurances cannot be relied upon, and that accordingly removals to Rwanda by virtue of this legislation would constitute breaches of Article 3 of the ECHR, and perhaps other Convention rights (including Article 2 and Article 8).  A secondary ground would be that the question of whether Rwanda is a safe country needs to be decided by a court on facts placed before the court, rather than by Parliament.

The failure to disapply section 4 of the HRA means that the courts will be left entirely free to reconsider and contradict Parliament’s judgement that Rwanda is safe, as well as the facts on which that judgement was made, and, on that basis, to denounce the Bill and other legislation providing for removals to Rwanda as incompatible with Convention rights.  The fundamental premise of the Bill is that Parliament is entitled to give legislative effect to its judgement that, in view of the Treaty, Rwanda is a safe country in the defined sense and for the relevant purposes.  This premise is both necessary and appropriate.  Litigation under section 4 of the HRA, however, will enable opponents of the Rwanda plan outside of Parliament to challenge in court not only the constitutional legitimacy of Parliament’s judgement, but also, in particular, its factual foundation and the assessment of risk made on those facts.  Litigation to this effect will begin immediately after the Bill receives royal assent.  It remains to be seen what the courts would decide, but at a minimum they would have an opportunity to reconsider Parliament’s decision making and to assert the superiority of their own evaluation of the Treaty and the future prospects for removals of asylum-seekers to Rwanda.  The Bill thus contains an important internal contradiction, which significantly weakens its force.

This is not a minor point.  It is true and important that a declaration of incompatibility under section 4 of the HRA has no effect on the “validity, continuing operation or enforcement” of the legislation that is declared to be incompatible with Convention rights.  In theory, the Home Secretary and immigration officers would be able to continue to remove persons to Rwanda in reliance on the new legislation.  But in practice, it would be very difficult to maintain the policy in the face of a declaration of incompatibility – especially one issued by the Supreme Court.  In such a case, the Court will have had an opportunity to consider evidence, to question Parliament’s judgement, and (if it disagrees) to hand down a judgment that declares Parliament to have been wrong.

By necessary implication such a judgment would amount to a denunciation of all past and future removals to Rwanda as breaches of Convention rights. The political pressure on the Home Secretary to discontinue removals would be very significant, not least because the courts’ different risk assessment would be misrepresented to be a simple question of fact finding, rather than a contested interpretation of abstract rights – or a predictive assessment of how a new administrative scheme underpinned by diplomatic assurances is likely to operate in the future with respect to those rights.  The matter is more complex than ordinary fact-finding, for the reasons given above, but in terms of how any declaration of incompatibility would be received, it is highly likely that it would be presented, and wrongly seen by many, as firmly within the judicial competence and not a matter for discussion.

If the Government intends to continue removals to Rwanda even in a scenario in which the Supreme Court declares this new legislation incompatible with Convention rights under section 4 of the HRA, it should give an assurance to that effect to Parliament now.  But if it is willing to give such an assurance, then it should also disapply section 4 of the HRA so as to prevent the litigation – the outcome of which would be of purely academic interest in the light of the assurance – from being initiated in the first place, and thus from wasting time and money.  Any such litigation would only serve to provide (a) an opportunity for opponents of the legislation to secure judicial support for their political campaign against it and (b) a temptation for the courts to vindicate their original findings, or at least to substitute their judgement for the judgement of Parliament on the significance of the new situation, without having to accept that Parliament’s judgement on that matter ought to be and is authoritative.  It is reasonable for the Government to invite Parliament to give legislative effect to its policy and not to make its implementation conditional on judicial approval.  While a declaration of incompatibility strictly has no legal force (subject to its relationship to section 10 of the HRA, on which more below), its political significance would be very grave indeed.

It is inevitable that this legislation will be challenged in the European Court of Human Rights.  When the Government comes to defend the legislation in that forum, its position would be gravely compromised if the domestic courts have declared the legislation incompatible with Convention rights.  Any declaration would be presented as an exercise in judicial fact-finding – factual findings which the Strasbourg Court would then feel able simply to adopt.

The Government should attempt to argue, in proceedings before the Strasbourg Court, that the legislation gives effect to Parliament’s reasonable judgement that Rwanda is a safe country and that it is reasonable, in light of the traditional demarcation of responsibilities in the UK constitution, for Parliament to make that judgement.  It involves an evaluation of another friendly country’s assurances, trustworthiness and legal system, including the independence of that country’s judiciary.  The Government should maintain that these are questions which it is not appropriate for courts to answer, particularly UK domestic courts.  They are not well-placed to do so, not least because the issues are interwoven with the conduct of foreign relations and are likely to impact on those relations adversely, and because of the inherent unfairness of judging persons in a forum to whose jurisdiction they cannot reasonably be expected to submit themselves and where they may regard it as inappropriate to defend themselves. But it will be very difficult for the Government to make this argument if it has left section 4 of the HRA in play, for this will undercut the argument that, as a matter of principle, questions about diplomatic relations, about the reliability of another sovereign state to comply with its international agreements or about the independence of its judiciary are not properly within the competence of the courts, particularly UK domestic courts.

Why does the Bill not disapply section 4 of the HRA?  Two explanations are possible.  The first is that government lawyers have persuaded ministers that the power to issue a declaration under section 4 of the HRA forms part of how the UK meets its obligations under Article 13 of the ECHR.¹⁵Article 13 provides that “Everyone whose rights and freedoms as set forth in this Convention are violated shall have an effective remedy before a national authority notwithstanding that the violation has been committed by persons acting in an official capacity.”  This line of reasoning assumes that legislation disapplying section 4 of the HRA would be incompatible with Article 13.  This would not be a good reason to fail to disapply the provision.  Despite having felt unable to make a statement of compatibility under section 19(1)(a) of the HRA, and instead having made a section 19(1)(b) statement,¹⁶A section 19(1)(b) statement is “a statement to the effect that although [the Minister] is unable to make a statement of compatibility the government nevertheless wishes the House to proceed with the Bill.” the Government’s position is and should continue to be that the Bill is compatible with Convention rights, precisely because removal to Rwanda will not subject any person to inhuman treatment.¹⁷A political practice has emerged since the enactment of the HRA that a Minister will not issue a section 19(1)(a) statement of compatibility unless the Minister is satisfied there is a greater than 50% probability that it will be found to be Convention-compliant by the courts. This principle cannot be found in the HRA itself but was set out by Jack Straw, the then Minister in charge of the Bill’s passage. During the passage of the Illegal Migration Act 2023, the Minister, Lord Murray of Blidworth, confirmed that the current Government also adhered to this principle and added that:

“The effect is that a Section 19(1)(a) statement is that you are satisfied that the measures are absolutely compliant, and a Section 19(1)(b) statement is that you are less than absolutely sure. Therefore, by placing a declaration of this kind on the front of the Bill, it is not a statement that the Government believe that the measures in it are not compatible; it is clearly the case that there is a strong—in my submission—legal basis for contending that these measures are compatible. However, applying the principles enunciated by Jack Straw following the passage of the Human Rights Act, the Home Secretary has quite properly appended her name to the statement on the front of this Bill.” https://hansard.parliament.uk/lords/2023-03-08/debates/44FF4E27-404E-454A-A77A-B2E4BC6D6AC9/IllegalMigrationBill

While we think this application of section 19 of the HRA is a dubious constitutional practice that should be abandoned by the Government, it provides important context for understanding the Government’s position that there are strong grounds to say that the Safety of Rwanda Bill is compatible with the ECHR, notwithstanding the Home Secretary’s section 19(1)(b) statement that a statement of compatibility cannot be made.  Parliament does not need to make provision for our courts to denounce the legislation as incompatible in order to support that argument.  In any case, a declaration of incompatibility is only an effective remedy if it is very likely to result in change to the legislation in question. That claim is in tension with any proposition that the Government would or could simply ignore the declaration and continue with flights.  To be clear, we do not accept that there is a constitutional convention that Parliament must amend or repeal legislation that a court declares incompatible or that the Government is constitutionally bound to ask it to do so.¹⁸It is worth noting, though, that the UK has, in the past, unwisely and unsuccessfully argued in Strasbourg, principally in the case of Burden v UK [2008] ECHR 357, that something to that effect is indeed the consequence of a section 4 declaration – a fact which certainly reinforces the case for disapplying section 4 of the HRA and the need to take the argument referred to in the next footnote seriously.  But neither do we accept that Parliament must empower the courts to second guess its legislative choices in order to avoid the risk of a secondary breach of the ECHR.

The second possible explanation is that section 4 of the HRA has been left in play in order to soften the Bill’s appearance – mollifying some parliamentarians who wrongly conflate the application of provisions of the HRA with respect for human rights and the rule of law, and providing a relatively harmless outlet for judicial opposition to (scepticism about) the Bill.  We doubt whether many parliamentarians or others are likely to have a more favourable impression of the Bill because it leaves open the prospect of litigation challenging the Bill’s central purpose.  The Bill is already being attacked in the strongest possible language, with section 4 of the HRA being identified as one main way in which political opposition to the Bill may be carried on after enactment.¹⁹Tom Hickman KC has said that the Bill’s failure to exclude section 4 of the HRA leaves open a potentially significant route for a legal challenge, noting that:

“While declarations of incompatibility do not affect the continuing operation of legislation, in practice legislation is always amended to comply with such declarations and the government relies upon section 4 of the HRA as satisfying its obligation under Article 13 where primary legislation is said to contravene the ECHR. The preservation of the ability of domestic courts to consider whether the Safety of Rwanda Bill, if passed, is compatible with the ECHR is welcome. It will allow the courts to consider the treaty and the argument that it moves the dial.”

Tom Hickman KC, ‘What is in the prime minister’s ‘emergency’ asylum legislation?’ Institute for Government (December 2023), https://www.instituteforgovernment.org.uk/sites/default/files/2023-12/examining-rwanda-safety-bill_0.pdf.  There is no reason to believe that letting section 4 of the HRA run will somehow discourage other litigation that attempts more directly to frustrate the operation of the Rwanda plan.  What it will do, instead, is provide a serious risk that opposition to the plan, and perhaps also to Parliament having chosen to legislate in response to the Supreme Court’s judgment, will result in a very high-profile politically significant defeat for the Government.  In salience and significance, this would be comparable to, the House of Lords’s judgment in the Belmarsh case in 2004,²⁰A and others v Home Secretary [2004] UKHL 56 which quashed an important element of the then Government’s counter-terrorism policy.²¹For incisive criticism, see John Finnis, “Judicial Power: Past, Present and Future” in Richard Ekins (ed.), Judicial Power and the Balance of Our Constitution (Policy Exchange, 2018), 26, 41-47 It may be even more significant than that.  Perhaps the present Government would be much more likely to ignore a declaration of incompatibility than was the government that introduced the HRA.  But it seems to us a mistake to think that litigation under section 4 of the HRA would be a sideshow, harmlessly absorbing litigious energy, and judicial restlessness, that might otherwise have a more damaging outlet.

The risk arising from the power of Ministers to amend the legislation by order

It is important too to recognise that section 4 of the HRA does have legal consequences.  It is the principal trigger for the application of section 10 of the HRA (power to take remedial action), which empowers Ministers to amend legislation by order to the extent necessary to remedy the incompatibility identified by the court.  (Section 10 of the HRA is a far-reaching Henry VIII clause, which empowers Ministers to change primary legislation by order.²²For discussion and criticism of the misuse of section 10 of the HRA in another context, see Richard Ekins, Against Executive Amendment of the Human Rights Act 1998 (Policy Exchange, 2020).)

The Bill does not disapply either section 4 or section 10 of the HRA.  That means that if a UK court makes a declaration of incompatibility, the Government will have power to amend the Safety of Rwanda (Asylum and Migration) Act 2024, as it then will be, without needing to persuade Parliament to enact further primary legislation.  That would be wholly inappropriate given the significance and political salience of the issues with which the Bill deals and the close scrutiny to which it is likely to have been subjected before being enacted.  In order to indicate its commitment to the policy of the Bill, the Government should make clear that it will not respond to any declaration of incompatibility by using the section 10 power to make a remedial order amending the legislation but will ensure that any proposed changes to the Bill after its enactment will be proposed to Parliament in the form of primary legislation.

Disapplying section 4 of the HRA would mean that a British court could not trigger section 10 by making a declaration of incompatibility.  However, section 10 also applies if it appears to a Minister, in light of a decision of the European Court of Human Rights made after the HRA came into force on 2 October 2000, that any legislation is incompatible with the UK’s obligations under the ECHR.  It follows that unless section 10 of the HRA is expressly disapplied, Ministers will retain a legal power to amend (and in practice effectively to repeal) this new legislation if or when they form the view that a decision of the Strasbourg Court indicates that the legislation is incompatible with the ECHR.  The Bill should be amended to remove this ministerial power to reverse Parliament’s judgement, a power that might be used by a Minister in the wake of, or even in anticipation of, a decision of the Strasbourg Court.  Indeed, in view of the Government’s section 19(1)(b) statement, which indicates that there is a serious risk that the Bill may be found to be incompatible with Convention rights, it is arguable that section 10 of the HRA would otherwise authorise a Minister to amend the legislation at any time after its enactment.

It may be that the present government would never exercise section 10 of the HRA in this way – to amend the Bill and undermine the provision it makes for asylum-seekers to be removed to Rwanda.  If that is the case, the Government should make this clear, giving an assurance to Parliament to that effect or, better yet, setting out this assurance on the face of the Bill by disapplying section 10 of the HRA.  But whatever the present government’s intentions, if section 10 of the HRA continues to apply to this Bill then a future government will have a legal power to reverse Parliament’s judgement without having to persuade Parliament to enact new primary legislation.  This is constitutionally objectionable, and the Bill should be amended to address the point.

The risks that decisions based on particular individual circumstances will frustrate the policy

The Bill is an effective response to the Supreme Court’s judgment insofar as it rules out further argument that removal of persons to Rwanda is unlawful because of a real risk that Rwanda may send them on to an unsafe third country.  The Bill aims to rule out other general challenges to the safety of Rwanda and thus to minimise the risks of systemic challenge to the policy.  In deliberating about the Bill, parliamentarians will need to consider the breadth of clause 4 of the Bill and to decide whether it draws a stable distinction between general and particular grounds.

Clause 4(1) provides that despite clause 2 a decision maker (the Secretary of State, the court) may conclude that Rwanda is not a safe country for the person in question.  This conclusion requires “compelling evidence relating specifically to the person’s particular individual circumstances (rather than on the grounds that the Republic of Rwanda is not a safe country in general)”.  This provision needs to be read alongside clause 4(4), which provides that a court or tribunal may grant an interim remedy that prevents or delays removal to Rwanda “only if the court or tribunal is satisfied that the person would, before the review or appeal is determined, face a real, imminent and foreseeable risk of serious and irreversible harm if removed to the Republic of Rwanda.”

The immediate question that arises is what types of case the Government envisages as falling within clause 4(1) and how particular individual circumstances are to be distinguished from grounds that Rwanda is not a safe country in general.  Clause 1 says that a “safe country”:

• means a country to which persons may be removed from the United Kingdom in compliance with all of the United Kingdom’s obligations under international law that are relevant to the treatment in that country of persons who are removed there, and
• includes, in particular, a country—
• from which a person removed to that country will not be removed or sent to another country in contravention of any international law, and
• in which any person who is seeking asylum or who has had an asylum determination will both have their claim determined and be treated in accordance with that country’s obligations under international law.

Between the Treaty and its other international obligations, Rwanda is committed to treating asylum-seekers in ways that meet the standards required of a safe country.  But what that suggests is that whenever a person argues under clause 4(1) of the Bill that Rwanda is an unsafe state for that person individually, that person is very likely to be raising questions about the extent to which Rwanda can be relied on to honour its international obligations.  The terms of clause 4(1) imply that there is a discernible difference between particular circumstances and general arguments.  The risk is that this difference may break down in practice.

This risk is possibly heightened by the otherwise very sensible provision made in clause 4(2), which specifies that clause 4(1) does not permit a decision maker to consider whether Rwanda might remove the person to an unsafe third country.  This caveat risks implying that such considerations would otherwise fall within the scope of clause 4(1) despite seeming to raise general points and that the reliability of Rwanda to comply with its international obligations, and perhaps the independence of its judiciary, will again be in issue when those issues do not relate to the specific risk that a person will be removed from Rwanda and sent on to an unsafe country. Allowing those questions to be raised in this way would again undermine the main premise for clause 2 of the Bill, which is that it is necessary and appropriate for those issues to be determined by Parliament, rather than by UK courts.

Parliamentarians should ask the Government to make clear the types of cases that are envisaged to fall within clause 4(1) and to consider whether the difference between particular individual circumstances and general arguments might be spelt out more precisely in the Bill.  The risk is that if this is not cleared up then in practice persons facing removal to Rwanda will routinely claim that Rwanda may be a safe country in general but will not be safe for them, because of their particular circumstances, with arguments which in substance might be quite general repackaged to be specific in relation to the individual.  Much turns on how courts and tribunals would respond to such arguments, which might not accord with the spirit of the legislation but are possibly left open (or at least not firmly closed down) by its letter.

The risk is that courts and tribunals continue to think, per the Supreme Court’s judgment, that whatever Parliament may say Rwanda is not safe and that while this is not a conclusion they may reach in general, they are free (according to the statute) to reach it in individual cases.  That is, the evidence raised about why Rwanda is not a safe country for this or that person might be received in the context of generalised suspicion about Rwanda’s capacities and commitments, suspicion that the legislation may attempt to contradict, but fails to extinguish.

It seems to us that clause 4 of the Bill should be tightened to specify that the particular individual circumstances in question must involve, or relate to, a pre-existing connection to Rwanda.  Such a change would serve to focus the exception on cases where there really is a risk to a particular individual, rather than an individual attempting to deploy what are in substance general objections to Rwanda in order to frustrate his or her removal.  Alternatively, clause 4 might be amended to specify that removals to Rwanda should proceed unless there is a real risk that the Rwandan government itself will persecute the person in question.  Either change would serve to narrow, specify and clarify the exception, screening out types of challenge that seem likely to involve an attempt to deploy general objections to Rwanda as grounds for why a particular individual should not be removed.

These suggested amendments would appear to cover the majority of cases that the Government has in mind for clause 4(1), while at the same qualifying the more generalised propositions that would otherwise tend to overlap with the factors relevant to the safety of Rwanda in general.  The Government may envisage other types of case falling within clause 4(1), including perhaps a person whose medical condition means that their life expectancy would be significantly reduced by removal to Rwanda because of the climate or absence of the necessary medical care. Work needs to be done to set out all the plausible types of case that are intended to fall within clause 4(1) and then to provide for them more precisely. The Government must have considerable experience of the sorts of issues that can arise and so should be able to move quickly to work out all the possibilities that need to be covered and to set them out more clearly on the face of the Bill.

Our point is not that judges will deliberately set out to obstruct removals to Rwanda by interpreting and applying clause 4 in this way.  We very much hope this is not the case.  However, claimants will certainly be motivated to develop ingenious arguments that attempt to exploit the imprecision, instability, or inherent tension in the wording of clause 4, building on the Supreme Court’s scepticism about Rwandan official capacity and motivation and taking advantage of the sentiment shared by many lawyers that Parliament is acting improperly in legislating to provide that Rwanda is a safe country.  Without crystal clear drafting, these background assumptions and dispositions may prove very influential.  It may be that judges will dispose of individual challenges effectively and efficiently.  But it is also possible that argument about the relevant facts proves time-consuming and/or that legal argument about how to interpret and apply clause 4 provides a ready means to delay effective implementation of the scheme.  These are serious risks.  They may not eventuate.  But Parliament should consider them and should consider amendments that would minimise them.

Note that clause 4 addresses the question of whether Rwanda is a safe country for the removal of a particular person.  Whatever answer one gives to that question, which is for Parliament to decide in accepting or amending clause 4, no person should be sent to Rwanda, or any other country, unless he or she is fit to fly.²³In Policy Exchange’s work on how to address the Channel crisis, we have recommended processing asylum claims outside the UK, in a British Overseas Territory, with genuine refugees to be resettled in safe third countries.  In each report, we have made clear that ascertaining fitness to fly is an essential first step before removing someone from the UK and that unfitness to fly is and must be a bar on removing someone from the UK.  See Stopping the Small Boats: a “Plan B” (Policy Exchange, 2022), 8, 21 and 44 and Richard Ekins and Stephen Laws, How to legislate about small boats (Policy Exchange, 2023), 11, 17 and 22.  This concerns fitness to fly rather than the safety of Rwanda (the means of removal, not the destination).  The amendments that we propose to clause 4 are confined (like the rest of the Bill) to specifying under what circumstances Rwanda is an unsafe country for the removal of a particular person.  They would not require any person who is unfit to fly to be removed from the UK.

The purported risks that would arise if Parliament were to enact more tightly drawn legislation

The amendments we suggest above would strengthen the Bill.  In our view, they would not place the UK in breach of its ECHR obligations, although there is of course a risk that the Strasbourg Court will conclude that removals carried out under the legislation (or possibly even the very act of enacting the legislation in response to the Supreme Court’s judgment) is incompatible with Convention rights.  And well in advance of any such final judgment, one or more judges on the European Court of Human Rights might well indicate interim measures under Rule 39 of the Rules of Court,²⁴See further Richard Ekins, Rule 39 and the Rule of Law (Policy Exchange, 2023). purporting to restrain the UK from removing a person to Rwanda.

The Bill anticipates the risk of interim measures being issued by judges of the Strasbourg Court.  The Bill’s answer, in common with section 55 of the Illegal Migration Act 2023, is to provide that the Minister, acting personally, has authority to decide whether the UK will or will not comply with an interim measure, a choice which will bind civil servants.  Further, the legislation provides that no court or tribunal may have regard to such an interim measure in hearing an application or an appeal that relates to a decision to remove a person to Rwanda.  (There remains a risk that the Minister’s decision not to comply with a Rule 39 interim measure will be challenged by way of judicial review proceedings.)  This is a welcome provision, so far as it goes, but parliamentarians should be well aware that the Rwanda plan is likely to collapse if the Government is unwilling in practice to remove asylum-seekers to Rwanda if or when Rule 39 interim measures have been made.

Some legal critics of the Bill have suggested that it is such an outrageous measure that the courts may be tempted to revive and to act on dicta from the Jackson²⁵R (Jackson) v Attorney General [2005] UKHL 56. and Privacy International²⁶R (Privacy International) v Investigatory Powers Tribunal [2019] UKSC 22. cases, where a minority of Law Lords and then Supreme Court justices suggested that the courts might be able to appeal to the constitutional principle of the rule of law in order to deny the validity of an Act of Parliament that limited or excluded judicial review proceedings.²⁷Professor Mark Elliott (University of Cambridge), for instance, has queried whether the Bill might be:

“challenged on the ground that it is simply unconstitutional? It is, after all, an affront to the separation of powers and the rule of law, in that it effectively reverses a Supreme Court judgment, undermines the judicial function and attempts to remove from the courts’ jurisdiction questions about the legality of Government decisions. In orthodoxy, the principle of parliamentary sovereignty — which makes whatever Parliament enacts lawful — would be a complete answer to these charges. But in Privacy International, Lord Carnwath said ‘it is ultimately for the courts, not the legislature, to determine the limits set by the rule of law to the power to exclude review’. For a court to take the step implied in this comment, by holding, in effect, that Parliament had exceeded its authority by seeking to limits the courts’ constitutional role, would be fraught with risk for the judiciary. It is, however, conceivable that the Rwanda Bill might transform what has largely remained a hypothetical question about the fundamental relationship between Parliament and the courts into a live one.”

See Mark Elliot, ‘The Rwanda Bill and its constitutional implications’, Public Law for Everyone (6 December 2023), https://publiclawforeveryone.com/2023/12/06/the-rwanda-bill-and-its-constitutional-implications/.

Another academic lawyer, Dr Adam Tucker (Liverpool University), writing a few weeks before the Bill was published, has said that:

“it is not possible for the UK to simply legislate Rwanda to safety, or to make it safe by negotiating and concluding a Treaty, or both. Any legislation which Parliament enacts in the foreseeable future which provides that Rwanda is safe, or is to be considered or regarded as safe for certain purposes (and so on) would not be a truthful statement of the situation on the ground in Rwanda. It would be enacting a falsehood into law. It would be a legal fiction.”

Tucker goes on to say that if the Parliament proceeded in this way that we should not forget that:

“famous minority dicta in Jackson are a subtle but pervasive background feature to the contemporary constitution. Those dicta reflected a willingness to contemplate refusing to apply legislation which removes executive action from judicial supervision. Whilst the court has not since exercised that power to strike down or disapply legislation which offends the rule of law, those dicta remind us that in the right circumstances, senior judges may still feel the temptation to do just that. And the enactment of the Rwanda fiction may yet provide those circumstances.”

Adam Tucker, ‘The Rwanda Policy, Legal Fiction(s), and Parliament’s Legislative Authority’, U.K. Const. L. Blog (22nd November 2023) (available at https://ukconstitutionallaw.org/).   It is startling to see these dicta once again being raised in public argument.  Our constitutional law is crystal clear on this point.  Whatever the King-in-Parliament enacts is law and British courts have no authority to question the validity of an Act of Parliament.  These propositions have been stated and restated repeatedly, not least by eleven Supreme Court judges in Miller (No 1) and Miller (No 2),²⁸R (Miller) v Secretary of State for Exiting the EU [2017] UKSC 5 and R (Miller) v Prime Minister; Cherry v Advocate General for Scotland [2019] UKSC 41. in 2017 and 2019.  Lord Bingham, the greatest judge of his generation, denounced his colleagues’ constitutional heresy in the Jackson case²⁹Tom Bingham, The Rule of Law (Penguin, 2011), chapter 12. and Lord Reed and Lord Sales, the two leading members of the Supreme Court, both rejected the argument that led to the Privacy International dicta.

It is wildly irresponsible for lawyers to invite the Supreme Court to overthrow the fundamental rule of our constitution – or to suggest that such an unconstitutional act, which would be tantamount to a constitutional revolution, would be a justified response to Parliament enacting legislation that makes it lawful to remove persons to Rwanda.  We doubt very much that the Supreme Court would entertain this course of action, which would invite and warrant intense political criticism of the Court, and would and should lead to far-reaching parliamentary action to reverse the Court’s attack on the fundamentals of the constitution.

Nothing in the Bill, or in our proposed amendments, seems to us contrary to the UK’s ECHR obligations, still less to the Refugee Convention.  (We note that the repeated refrain on the part of critics of the Rwanda plan, that the Refugee Convention prohibits removal of asylum-seekers to a third country for processing and settlement, is simply false and has not been accepted by any of the courts who have considered the recent challenge to the Rwanda plan.³⁰See further Richard Ekins, John Finnis and Simon Murray, The Nationality and Borders Bill and the Refugee Convention 1951: Where the Joint Committee on Human Rights goes wrong (Policy Exchange, 2022).)

However, the Prime Minister has said that Rwanda will not accept the Treaty if the UK acts in breach of its international obligations, implying that the Bill simply cannot go any further.³¹https://www.politicshome.com/news/article/government-publishes-new-rwanda-legislation-declaring-country-safe.   The amendments we propose in this paper, disapplying section 4 of the HRA and narrowing clause 4 of the Bill, would not place the UK in breach of its obligations and would not somehow implicate Rwanda in any course of action that would involve the breach of any person’s human rights.  In view of the repeated denunciation of Rwanda in the press, Parliament and courts, it is readily understandable that the Rwandan government might insist that it does in fact comply with its international obligations and might understandably require that, if it is to cooperate with the British government, we should do likewise.  Still, the objection is hard to understand in the light of the fact that the main ground on which the arguments that removals to Rwanda are in breach of the UK’s international obligations are likely to be based is that the Rwandan government will not honour its assurances and will mistreat or allow the mistreatment of asylum-seekers who are sent to that country.  That is an argument that Rwanda of course rejects and which the Treaty, and related arrangements, effectively addresses.