DR JOSEPH MULLEN – WRITTEN EVIDENCE (URA0010)

HOUSE OF LORDS INTERNATIONAL AGREEMENTS COMMITTEE: INQUIRY INTO THE AGREEMENT BETWEEN THE UNITED KINGDOM AND THE REPUBLIC OF RWANDA FOR THE PROVISION OF AN ASYLUM PARTNERSHIP AND SAFETY OF RWANDA (ASYLUM AND IMMIGRATION) BILL 2023-2024)[1]

A. Introduction

1. The Memorandum of Understanding (MOU) negotiated between the Governments of the UK and Rwanda was considered unlawful by a Decision of the Supreme Court, thereby ruling against the deportation of asylum seekers to Rwanda. In order to effectively maintain the policy which had been promoted into a signature political slogan of ‘stop the boats’, the format of an MOU was upgraded to a Treaty between the two countries, re-formulated into a Bill as entitled above and placed before Parliament on 06 December, with a second reading taking place on 12 December 2023 when the Bill was passed by a majority of 44 votes. Subsequent to the approval of Parliament, the House of Lords undertook an Inquiry into the aforementioned Bill and invited witness statements from a selected group of experts. This document constitutes the Statement of Evidence of the expert named above.

2. The Bill ‘Safety of Rwanda (Asylum and Immigration) Bill 2023-2024’, hereafter referred to as  ‘the Bill’ analyses key controversial issues contained  in the Asylum Partnership Bill. Central to its considerations is guidance issued relating to the safety status of Rwanda, ECHR compatibility, disapplying the HRA and its ramifications, individual challenges to removal and interpretation of interim measures of the ECtHR.  The Committee is seeking evidence on the questions outlined in Sections B to E below;

B. Whether the changes to the asylum partnership made by the Agreement are likely to meet the concerns raised by the Supreme Court.

3. The Supreme Court has ruled that ‘Rwanda is not safe for refugees because they might be sent away to face persecution and so the plan is unlawful’. The Supreme Court found that there were ‘substantial  grounds’ for believing that there was a real risk of people being returned to countries where they would be at risk (refoulement) due to deficiencies in Rwanda’s asylum system. Arguably, the risk of refoulement is extremely marginal; UNHCR could only cite three specific cases of ‘indirect’ refoulement out of some 134,000 asylum seekers throughout the whole of 2022[2]. In response to the Judgment, the UK Government has constructed a new treaty with Rwanda, addressing the aforementioned lacunae identified by the Supreme Court. The Government subsequently drew up a draft Bill to this effect; the second reading of which was presented to and passed by Parliament on 12 December 2023. The House of Lords Inquiry deals with the question of whether the Agreement as it is currently constituted, adequately responds to the objections of the Court and contains the robust legal assurances required in order to do so. As it is a partnership-based bilateral Agreement, it is also incumbent upon an expert witness to comment upon the counterpart Rwandan law into which the Agreement will be embedded.

4. While the principal objective of the MOU and the Agreement remains similar – providing a deterrence to irregular journeys of small boats into UK territorial waters- there is a considerable shift in how unsuccessful claimants will be finally processed. The Migration and Economic Development Partnership (MEDP) at para.1.1 asserts that ‘no one will be eligible for return to the UK’. By contrast the Agreement states that, in the event of a negative finding, they would be repatriated to the UK. (See Appeal system outlined below) This would remove the threat of refoulement, as the default position has shifted from being Rwanda or their home country to repatriation the UK. The threat of refoulement, as identified by the Supreme Court, originated in the Articles of the MOU but was not identified as being present in the Rwandan legal system. However, as the Rwandan legal system was cited generically by the Court as being challenging in relation to good human rights practice, I cite below the key domestic legal instrument currently utilised by the Government of Rwanda in determining refugee status. The additional assurances contained in the new Act should strengthen ramifications in the corresponding Rwandan law. The relevant Article in Rwandan Law is the following:

Refugee Governance in Rwanda: Refugee Status Determination Committee (RSDC): Law No. 13 ter/2014 of 21/05/2014 relating to refugees ‘sets out status determination procedures (RSD) and timelines with refugee rights and obligations[3]. This is central to the refugee determination/processing system in Rwanda to which all deportees will be subjected. Rwandan government officials have specified the timeline for key activities in the administrative process as follows:

•          Submission of application files by DGIE (Home Ministry): within 15 days

•          Decision by RSDC: within 45 calendar days (can be made sooner, sometimes 5 or 10 days)

•          The Appeal Notification of the decision: Within 10 days the committee to communicate to the applicant; provided by a letter in one of the 3 official languages of Rwanda – English, French or Kinyarwanda

•          Appeal: within 30 days

5. By contrast, in Annex B of the Act, which outlines the Claims Process in Part 2 –Process for Asylum and Humanitarian Protection Claims, it is confirmed that claims for Humanitarian Protection as well as asylum are feasible simultaneously . The communication of the rights of the relocated person, including the right to make a fresh claim, is clarified in paras. 2.1-2.4. Section 3 of Part 2 introduces the role of a new agency: the First Instance Body. The functions allocated to the First Instance Body resemble closely the UK Decision process  undertaken by the Home Office Caseworker in analysing the Statement of Evidence and making the initial decision on the status of the asylum seeker. This includes: the criteria for appointment to the Body, the ethical standards expected of the members and the inclusion of independent experts to support the First Instance Body in reaching a decision, preferably within a six month period but extendable. Section 3.5 enumerates the characteristics of good practice in interviewing the relocated person: their level of competence, same sex as the interviewee, familiarity with the cultural background of the interviewee, interpretation facilities and in the event of there being a child for interview, a responsible adult present (Act 3.5.2) A copy of the interview will be made available on request. Both individual and corporate/ NGO expert groups may also attend the proceedings to ensure transparency. A written report on the decision will be given to the relocated person. (Act 3.11) Procedures for determining Humanitarian Protection claims will be the same as determining asylum. This is truly the ‘engine room’ of the necessary fusion between the Act and Rwandan legislation. It is significant that UNHCR views the Government process in a positive light. In its ‘Refugee Response Plan’ UNHCR describes its close collaboration in its refugee response with the Government of Rwanda across as many as twenty programmes. This leads me to suggest that its Geneva HQ speaks with one voice while the field offices in Rwanda operate on a totally different plain, which leads one to speculate if its evidence to the Supreme Court challenges the operational realities of its field office in Rwanda and could be misleading for the judges.  

6. A key addition to the Act and superseding the MOU is the facility of Appeal against the Decision reached by the First Instance Body. (Act Section 4) . The Appeal body will be constituted of two co-presidents, one of which will be a non-Rwandan citizen and three judges, one of which will be a co-president, will hear Appeals referred from the First Instance body. The appeal body will have the jurisdiction to hear appeals de novo (4.2.6) . If a further Appeal is lodged, it will then re-enter the Rwandan legal system of the Refugee Status Determination Committee. However, if the second Appeal Court decision differs from the First Appeal Court, the case will be referred back to it to be heard de novo. This avoids the anomaly of a Rwandan Government Appeal Court overturning a Decision of the specially constituted Appeal Court to the First Instance body. Whether this is compatible with the constitutional jurisdiction of the President and Rwandan legislation may require an explicit legal opinion of the Rwandan Attorney General. These proposed amendments of the new Act, much of which duplicates the current Home Office’s refugee determination processes, taken as a whole, in my opinion meet the reservations expressed by the Supreme Court on refugee determination processes and ECHR compatibility. The issue of the incidence of refoulement remains opaque.

7. What appears to be further required for clarification is the appropriate thresholds of proof to the various levels of court; will the First Instance Body utilise the ‘reasonable degree of likelihood’ test currently used in the UK First Tier Tribunals. Should it be a higher level, e.g. balance of probability, this could be perceived as a disadvantage to the relocated person. Part 3 of Annex B of the Act specifies the legal and para-legal assistance that will be available to relocated persons which comes across as apposite to the asylum seekers needs

C. How strong and effective are the protections for persons relocated to Rwanda?

8. The afore-mentioned section 1 above, para. 4, outlines the legal formalities of refugee determination once the relocated person has arrived in Rwanda. The new system put forward in the Act incorporates key processes of the Home Office’s Determination and Decision processes and shadow Home Office Guidance on these matters. It must therefore be consistent with both the spirit and the letter of the findings of the Supreme Court. The issue of effectiveness, however, is difficult to judge ex-ante as this can only be ascertained by empirical evidential deduction, rather than inductive assumptions. The safeguard of independent expert opinion to the Appeal Body is only in force for 12 months which limits any potential corrective action on medium term effectiveness (Annex B 4.2.4). This provision should be extended to three years.

D. Precedents for requiring that claims must be for asylum in a third country

9. There are three specific precedents of asylum claims being processed in a third country: two of which involve Rwanda and the third refers to two South Pacific countries: Papua New Guinea and Nauru. The governments or organisations sponsoring these precedents are Australia, Israel and the United Nations. All three display different modus operandi and have only limited comparative applications to the current UK-Rwanda Agreement. There is, nonetheless, a particular advantage in assessing these examples, in that they provide valuable empirical ex-post experiences, rather than projected ex-ante aspirational outcomes. I shall commence with Australia.

The Australian off-shore deportation model.

10. The  offshoring and deportation of asylum seekers to overseas destinations was titled ‘Operation Sovereign Borders’,  described in Australian Government literature as ‘military-led border security operation, established in 2013, committed to protecting Australia’s borders; combatting people smuggling and preventing people from risking their lives at sea’. The operation essentially involved three elements: pushback of migrant boats in international waters away from Australia’s coastline, and forcibly returning the craft back to Indonesia or Malaysia, with which countries Australia had return agreements and detention.  Other boat passengers had their cases processed in two Pacific countries: Papua New Guinea and Nauru according to the laws of these countries. The Papua New Guineas arrivals were diverted to Manus Island, effectively a double offshoring. All claimants were subsequently interned at substantial cost on the Australian taxpayer[4] . The Papua New Guinean legal system is relatively young[5], particularly in migration law, and had great difficulty in coping with the numerous detainees awaiting resolution of their cases. This led to the Supreme Court declaring that the Operation Sovereign Borders agreement with Australia as unconstitutional and that Australia’s detention of asylum seekers as illegal. The Supreme Court ordered the PNG and Australian Governments to immediately take steps to end the detention of asylum seekers in PNG.

"Both the Australian and Papua New Guinea governments shall forthwith take all steps necessary to cease and prevent the continued unconstitutional and illegal detention of the asylum seekers or transferees at the relocation centre on Manus Island and the continued breach of the asylum seekers or transferees constitutional and human rights," the judges ordered.

11. In one of two lead judgments, Justice Terence Higgins said the detention also breached asylum seekers' fundamental human rights guaranteed by various conventions on human rights at international law and under the PNG constitution[6]. A stop was placed on receiving new asylum seekers and those remaining on Manus Island were repatriated to Port Moresby.  A small number of asylum seekers were granted leave to remain and integrated locally.

12. The rolling out of the Operation with Nauru, encountered strong resistance from the detainees who were subjected to prison-like conditions, resulting in riots, complaints against contractors and suicide attempts by internees. Under pressure from refugee advocacy groups, MPs from the Government party and doctors, the Government agreed to transfer all asylum seekers from Nauru to Australia by end 2018.

13. The Australia  case is widely quoted by UK legislators as a successful comparator for the reduction of asylum seekers, particularly by boat[7]. This has been reinforced within Government by the hiring of Alexander Dowden, a former Australian Foreign Minister, as an Adviser to an Independent Monitoring Committee. However, the Operation has been roundly criticised by international  agencies and discarded by the current government..

14. However, the Australian model should be viewed also in terms of its limitations and the policy volte face that has actually taken place since the Labour-controlled Albanese Government[8] came to office.  The ‘flag carrier’ of the policy was the turnaround action of the Australian Navy in international waters, for which they have been roundly criticised . In the Channel, territorial French and UK waters are contiguous to each other and no turnaround is possible without breaching the jurisdiction of one or other country and there appears to be differences of interpretation of marine safety clauses of the Law of the Sea[9]. Through its Union, UK Border Force officers have even brought a legal challenge against the use of force on the high seas. Hence the photogenic turnaround is a non-starter for the UK.

15. The overarching question is whether, with the benefit of hindsight, has the Australian Operation, with its detention and offshoring to two Pacific remote islands been effective in its           goal of decreasing illegal migration? In terms of raw statistics of arrivals of boat people on Australian shores, there appears to have been a substantial initial decrease. However, this has come at a cost both financial and moral and there are currently 73,000 asylum applicants awaiting determination. The mind boggling cost of over £10.7 mn per asylum seeker is unlikely to receive popular support in the UK. The Government of Papua New Guinea declared the scheme illegal and withdrew from it, after popular protests. And ironically, a new incoming Australian government considers migration to be an existential national asset and endeavouring to create safe and legal routes for asylum seekers. The numbers of irregular migrants arriving by plane has increased and there is also anecdotal evidence that cruise ships are being used as a modus of illegal entry. 

16. In conclusion, apart from the short lived euphoria of 2013-15 in the ‘turn around the boats’ operation, there is no empirical evidence that offshoring the processing of asylum seekers to Australia and Nauru has had any sustainable impact of deterrence on inward refugee flows. Evidence suggests that, with the passage of time and the raising of public consciousness, it has, counter-intuitively, made the Government more aware of the necessity of immigration.

The Israeli refugee deportation programme with Rwanda

17. Israel conducted an informal asylum seeker/refugee deportation scheme to Rwanda during the period 2013-2018.. During this five year period, some 4,000 asylum seekers were voluntarily moved from Israel to Rwanda on the promise of receiving $3500 in cash and a one way air ticket from Israel to Kigali. Those affected were mainly from Eritrea and Sudan, who probably chose Israel as a destination of choice because of religious and cultural links of Israel to the Horn of Africa[10]. It is reported that on arrival in Kigali, they were met by people traffickers who offered them taxi services to Uganda. Others are said to have used their money to proceed on to European destinations. There is no record of there have been any significant re-settlement or refugee processing of this group in Rwanda itself. Collaboration continues in the agriculture, education and cyber-security sectors[11] but broader considerations probably shaped this unsuccessful sortie into the migration field.

18. The Supreme Court found that the apparent failure of Rwanda to comply with a ‘processing agreement’ with Israel (See Bill paras. 58 ff.) is ‘relevant to the risk of refoulement’. Oral and observational evidence suggests that the incoming asylum seekers were deliberately avoiding refoulement and ‘legging’ it off to other destinations and certainly not back to their own country. The terms of the Agreement between Israel and Rwanda is not available for scrutiny so failure of compliance is an unproven assumption by the Court rather than a fact. The Israeli scheme was voluntary, open-ended and did not openly commit to a guaranteed acceptance or custodial role on the part of Rwanda, so bears little resemblance to the UK-Rwanda Treaty with little or no observable lessons to be learned. It is an unfortunate legal precedent to have been cited by the Court in the context of refoulement.

The United Nations Partnership with Rwanda

19. The United Nations High Commission for Refugees (UNHCR) also managed its own programme of refugee re-settlement in Rwanda, known as the Emergency Transit Mechanism. (ETM), evacuating asylum seekers from horrific conditions in Libyan refugee ‘warehouses’to Rwanda. The choice of Rwanda as a transit/processing country is consistent with para. 28 of the Act citing UNHCR to the effect that ‘the policy and legal frameworks of Rwanda continue to facilitate the inclusion of refugees within national systems in line with the Global Compact on Refugees’. This conclusion, validating Rwanda as a safe country, appears contrary to the observations on the MEDP submitted to the Supreme Court by UNHCR concerning ‘the adequacy of Rwanda’s asylum system and concerns about its ability to fulfill assurances under the MEDP/MOU (Treaty para.11) These remarks were given substantial weight by the Court but these could now be seen retrospectively as inconsistent. .

20. The intervention of UNHCR is conflated with the position of the whole United Nations system in Rwanda, of which UNHCR is only one of 24 agencies or entities. The different agencies are accountable to the Resident Coordinator of UNDP in Kigali and UN policy decisions relating to country policy are filtered through his office. There is no indication of the involvement of UNDP in the critical policy stance of UNHCR relating to the MEDP and I suggest that this is an outlier in overall country relations of the UN and the Government of Rwanda[12]. Given the current commitment of the Rwandan Government to the details of the MEDP and particularly to the expansion of its own refugee law Loi 13 ter to incorporate the MEDP conditionalities; on the balance of probabilities, this Agreement will be effectively implemented within the legal framework outlined in the Bill.

21. Another UN agency of significant interest in migration inside Rwanda is the International Organization of Migration (IOM), It spearheads the major resettlement programme of the UN and it has assisted the resettlement of more than 30,000 refugees from Rwanda since 2010. It describes itself (see Rwanda.un.org) as ‘the leading UN inter-governmental organization promoting humane and orderly integration for the benefit of all’. A showcase project it operates is the relocation of Afghan girls to Rwanda to continue their education. The IOM reports that it has assisted the migration and integration of 6,600 persons this year, 1,288 of which were carried out in collaboration with UNHCR as part of the Libyan Emergency Transit Mechanism. While the programmes of both UNHCR and IOM are different in scope from the Asylum Act, they are premised on the corporate commitment that Rwanda is a safe place from which to operate and that the necessary legal and administrative structure to provide the necessary protection and safeguarding of vulnerable groups are in place.  The recently-negotiated (20/12/2023) European Union migration and asylum pact, could hold lessons in the final stages of negotiation of the Act. This too, together with the role of the IOM, however, has recently been severely criticised by the Netherlands-based Transnational Institute[13].

E. Other Issues to bring to the notice of the Agreements Committee

22. The Safety of Rwanda (Asylum and Immigration) Bill declares that Rwanda is a safe country as an unquestionable given on the grounds that ‘the judgement of Parliament that the Republic of Rwanda is a safe country’. While one can argue any supposition by diktat, the safety of Rwanda for relocated Third World citizens (including children) is inevitably an issue to be tested by risk analysis. Significantly, its ramifications impinge upon ECHR compatibility and a derogation from the Human Rights Act. A principle of ‘the sovereignty of Parliament’ is unlikely to compensate for placing relocated persons in harms way. A gap in the Act is the lack of precision in the location of the accommodation facilities and the type (fixed or temporary). For security reasons these must obligatorily be confined to two urban locations: Kigali and Butare. Towns and villages along the borders with the DRC and Burundi have had military incursions from the FDLR or the remnants of Genocide perpetrator groups. Relations with the DRC are extremely febrile as Rwanda is accused of arming and supporting a Congolese dissident movement known as the M23[14]. To make a general statement that Rwanda is a safe country, when it is virtually at war with its neighbouring country,[15] without recognising that parts of Rwanda are extremely unsafe[16], could have reputational damage potential and accusations of naivete for Parliament itself. At time of writing, there are an estimated 1400 refugees per week crossing the border into Rwanda and being re-settled in Nkamira emergency transit camp[17]. It is difficult to reconcile the title of the Bill ‘Safety of Rwanda’ with existential reality, without acknowledging the specific locational threats and risks that affect its western border with the DRC. If ‘safety’ is uniquely predicated upon the ‘sovereign will of parliament’, this, per se, is unlikely to meet the test of consistency with the Decision of the Supreme Court.

STATEMENT OF TRUTH 

While having received the  invitation from the House of Lords International Agreements Committee to submit this expert report, I understand that my duty in writing this report is to help the Committee within my area of expertise. I understand that this duty overrides  any obligation to the parties from whom I have received instructions and I have complied with that duty in formulating this report. I confirm that insofar as the facts stated in this report are within my own knowledge, I have made clear which they are and I believe them to be true. The opinions I have expressed represent my true and complete professional opinion.

27 December 2023