UKR0008

Written evidence submitted by The Migration Observatory

University of Oxford

This document presents evidence on the costs of the UK-Rwanda Partnership (the “Rwanda policy”).

The government has indicated that saving money is an important goal of the Rwanda policy. Government statements about the policy have typically pointed to the costs of the UK asylum system and suggested that a policy that deters people from coming to the UK and claiming asylum would make the system more ‘sustainable’. However, the costs of the Rwanda policy are not known. This is partly because the government has not provided sufficient information about its plans. It is also because some of the things that will affect the costs are hard to predict.

This document outlines four unknowns that will substantially influence the cost of the Rwanda policy and hence an assessment of its value for money:

• The deterrent effect. The policy's financial impact hinges on whether it dissuades people from coming to the UK through irregular routes. The number of people likely to be deterred cannot be known in advance, but evidence from other countries suggests deterrent effects are often small.
• The numbers relocated. If more people are relocated, this will push up the scheme's costs in the short run, but may also increase any deterrent effect.
• The departure rate from Rwanda. If a person relocated to Rwanda decides to leave, the UK would stop its per-person payments to Rwanda and pay £10,000 to facilitate their voluntary departure. As such, the more relocated people decide to leave Rwanda, the lower the cost of the policy.
• The costs of the status quo. If the government succeeds in making the asylum system more efficient, the economic rationale for the Rwanda policy is weakened. 

We do not consider the wider costs of the Illegal Migration Act (IMA). That includes the cost of supporting people who are not deterred by UK government policies but also are not removed to Rwanda or another country. These costs are potentially considerable, as we explain in the Migration Observatory analysis, Why the government’s economic impact assessment of the Illegal Migration Impact tells us little about the Act’s economic impact.[1]

We also do not consider costs to individual asylum seekers, which a full social cost-benefit analysis would include.

Unknown 1: The deterrent effect of the Rwanda plan

The total financial costs of the Rwanda policy depend fundamentally upon how many people it deters from entering the UK irregularly. This is because sending asylum seekers to Rwanda will be more expensive than keeping them in the UK, according to the Home Office. However, if enough people are deterred from coming to the UK to claim asylum, the government saves the resources it would otherwise have spent on them.

It is not possible to know what the Rwanda policy’s deterrent effect will be, especially without any information about the number of people who would be removed. Neither the Impact Assessment for the Illegal Migration Act nor the National Audit Office’s investigation attempt to estimate the deterrent effect. Studies on asylum policies in other countries have typically found that the deterrent effects are small, because:

• Policy is not the most important factor driving changes in the number of people claiming asylum. Global developments, such as conflicts in countries of origin, appear statistically to be more important.
• Prospective asylum seekers do not always know what policies will face them when they arrive. The information they have may be inaccurate or misleading and not particularly detailed.
• Decisions about where and how to move depend not just on policy. Many other factors, such as the presence of family members or the language in the countries of destination, also play a role.

While there are examples of countries that have seen much larger reductions in maritime arrivals, such as Australia, those cases usually involved physical enforcement activities, such as boat interceptions and pushbacks, in addition to policies designed to affect asylum seekers' decisions. Evidence on the impacts of asylum deterrence policy is summarised in the Migration Observatory’s report, UK policies to deter people from claiming asylum.[2]

In practice, the deterrent effect of the Rwanda policy may well depend on how many people are removed there. If the number is relatively small (up to a few per cent of newly arriving asylum seekers), the evidence from other countries suggests the deterrent effect may also be small. If the number is very large (i.e., a substantial majority of all asylum seekers), it is reasonable to assume a bigger deterrent effect. The international evidence showing small deterrent effects is largely based on less radical policies, such as restricting access to benefits for asylum seekers, and would be less directly relevant to a policy that sent a large majority of asylum seekers to another country.

Unknown 2: The number of people who will be relocated to Rwanda

The main known cost of the Rwanda policy is the cost of removing people to Rwanda. This includes the payments made to the government of Rwanda, plus other associated costs such as logistics and administrative and legal costs. The NAO estimates that the marginal per-person cost of sending someone to Rwanda is around £182,000. The average cost per person will be higher, because the scheme has fixed costs, including lump-sum payments to Rwanda and possibly the expansion of detention facilities in the UK.

Given that the number of people sent to Rwanda crucially affects the finances of the programme and its likely impacts, it is surprising that the government has been so vague on this issue. Available information suggests the number could be anywhere between a few hundred and tens of thousands of people. If the Home Office can provide at least some basic working assumptions about the number of people it will relocate under different scenarios, this would make it easier to assess the likely fiscal costs of the scheme.

The current uncertainty about the number of people relocated means there are radically different scenarios for the scheme's finances. For example, in one plausible scenario, the UK spends over £500 million to remove 350 people,[3] with limited to no deterrent effect. This would be a cost of over £1m per relocated person.

In another scenario, the UK initially sends tens of thousands of people to Rwanda at a much higher cost. For example, one estimate by IPPR suggested it would cost £3.8 billion to relocate 20,000 asylum seekers[4], plus the additional costs of expanding detention facilities. In this scenario, however, the UK may spend less in subsequent years if fewer asylum seekers come to the UK due to a larger deterrent effect. Some people would still be removed to Rwanda in the subsequent years, but not as many. It is not clear whether this latter scenario is at all plausible, because it requires such a large initial outlay, high capacity in Rwanda, and the legal and operational ability to remove asylum seekers. If it did happen, the overall financial impact of the scheme would depend on how many years it lasted. This is because the costs would be front-loaded. A key factor in the long-term impact would thus be whether Rwanda was willing to negotiate an extension of the scheme, and at what cost.

In practice, the number of people who will be relocated to Rwanda will depend on several factors, including:

• The capacity of Rwanda to process asylum applications.
• Accommodation capacity in Rwanda. This may depend in part on how many people stay in Rwanda as opposed to leaving shortly after arrival.
• Legal challenges delaying or preventing removal, potentially including appeals made to the European Court of Human Rights.

All of these factors are highly uncertain.

The total number of people who could potentially be sent to Rwanda is also not known. This number is relevant as it affects the potential maximum size of the scheme (and thus its costs), as well as the share of people who are potentially subject to the policy who are removed. The maximum number of people who are potentially liable to be removed to Rwanda will depend on factors such as:

• The number of people detected arriving in the UK through irregular routes. When the Illegal Migration Act is fully implemented, the Home Secretary will be obliged to remove all people who arrived irregularly in the UK on or after 20 July 2023. Of these, some can be removed to their country of nationality because the Home Office views these countries as safe. These currently include all EEA countries, Switzerland, and Albania, with India and Georgia due to be added. Currently, irregular entrants of all other nationalities could be removed only to Rwanda. There are currently no precise estimates of how big this group is. Home Office statistics on irregular arrivals suggest around 33,000 people were detected having arrived irregularly in 2023 (either by small boat or other irregular routes), excluding the nationals of the ‘safe countries’ listed above (including India and Georgia). The number who will arrive in future years is highly uncertain (see above).
• The number of people who accept voluntary return to their own country, or move on to another country. In 2023, over 19,000 people who were liable for removal left the UK voluntarily, with over half going to Albania (~3,000), India (~2,800), Brazil (~2,200), China (~1,000), and Turkey (~900). There were few voluntary returns to many of the top countries of nationality for asylum seekers, such as Afghanistan or Iran.
• Implementation decisions. For example, the government can remove irregular migrants who have not claimed asylum to Rwanda, but may choose to focus on those who have claimed asylum.

Beyond the fixed and per-person costs, the Home Office will face certain additional costs if the number of people removed to Rwanda is relatively large. In particular, the IMA extends powers to detain asylum seekers. Current policy enables the government to detain people if there is reason to believe they will abscond, although there are not sufficient data to establish the likely prevalence of absconding among people eligible for removal to Rwanda.

At present, the UK has fewer than 1,000 detention spaces. The government is planning to increase this number, although it is not clear what total capacity would be required under current government plans. The Home Office has not provided enough detail about how it plans to implement detention policies under the IMA and Rwanda scheme to assess the likely costs of detention resulting from either policy measure.

Building new detention capacity has substantial up-front costs, and holding people in detention is more expensive than housing them in regular accommodation. The IMA Impact Assessment assumes a cost of £85 per day to support someone in regular asylum accommodation, while in Q4 2023, the cost of holding someone in immigration detention was estimated at around £116 per day.[5]

Unknown 3: The departure rate of people relocated to Rwanda – and how long they stay in Rwanda

The cost of relocating a person under the Partnership depends on how long the person remains in Rwanda. If a person leaves Rwanda within five years, the UK government will stop its quarterly operational payments (up to around £151,000 over five years, made quarterly), and pay Rwanda £10,000 to assist in the person’s voluntary departure.

This means the payments to Rwanda will depend heavily on how many relocated people remain in Rwanda, and for how long.

If a relocated person leaves immediately, the UK government would pay around £41,000.[6] If they stay at least five years, the UK would pay around £171,000 in per-person payments – around four times more. The longer a person stays in Rwanda, the greater the per-person payments. Figure 1 shows how the per-person payment changes according to when a person decides to leave Rwanda.

Figure 1

The NAO report states that the Home Office’s “working assumption” on the departure rate is that 10% of people who are relocated to Rwanda will leave voluntarily (presumably within five years of arrival).

However, this ‘departure rate’ is highly uncertain. Evidence[7],[8] on Israel’s similar policy of sending Eritrean and Sudanese people to Rwanda suggested a much higher departure rate, which would reduce the total cost of the scheme.

Unknown 4: the costs of the status quo

The fiscal impact of the Rwanda scheme depends on the cost of the system the scheme is replacing. The more it costs to process and support asylum seekers in the UK, the greater the savings from removing those people to Rwanda instead. By contrast, if the government makes the asylum system more efficient, the economic rationale for the Rwanda policy is weakened.

The financial cost of the UK’s asylum system has increased sharply in recent years, from £733 million in the financial year 2018/19 to around £4 billion in 2022/23.[9] This increase is primarily due to the large backlog of unprocessed claims. As the backlog falls, costs can be expected to decline.

However, the future costs of the backlog and the UK asylum system more broadly are uncertain. Some of this uncertainty results from the unpredictable nature of asylum migration, which is not possible to forecast accurately.

Additional uncertainty comes from the IMA. If the IMA creates a large cohort of people who cannot be granted or refused asylum and who also cannot be removed from the UK (see above), the government could incur high costs. Some of these costs would fall to central government, which is expected to provide accommodation and support to destitute asylum seekers who do not have a decision on their claim. Other costs would fall to local government—for example, if people leave the asylum support system and become irregular migrants in the UK. Either way, it is possible that the IMA improves the fiscal effect of the Rwanda policy, by pushing up the cost of the UK asylum system.

Conclusion

This submission has highlighted four factors that could substantially influence the cost of the Rwanda policy: its uncertain deterrent effect, the number of people relocated, how long people remain in Rwanda, and the costs of the UK’s asylum system without the Rwanda scheme.

The influence of these factors upon the cost of the Rwanda policy are unknown. In large part, this is because they are highly uncertain and cannot be forecast with precision. But the government could provide more information about its plans, which would enable broad estimates of costs to be calculated. For example, the government could provide its working assumptions for the number of people it will send to Rwanda under different scenarios, as well as the number of people it envisages detaining before their removal, and for how long.

April 2024