The government yesterday published its economic impact assessment for the Illegal Migration Bill and its Rwanda plan. The assessment reveals that Rwanda will be paid approximately £105,000 per refugee received on top of the £120 million already paid and any other undisclosed payments. At least, this is the figure used in the impact assessment; the government also says that the precise amount is ‘commercially sensitive’ but it seems reasonable to assume the cost must be something like that, otherwise the impact assessment is entirely pointless.

The subsequent headlines have suggested a net cost to the United Kingdom of £63,000 per refugee. This net cost is the difference in the impact assessment between all the estimated costs of of implementing the Illegal Migration Bill (detention, flights, etc) and the cost of processing and supporting a refugee. As we’ll see, though, the support costs look grossly exaggerated. The real net cost is likely to be a lot higher.

Financial cost is not the main issue, though, really. Supporters of the Rwanda plan consider it worth any amount of money to ship refugees to somewhere else. Opponents object to the scheme on moral and ethical grounds rather than cost grounds. There are also other costs; the lives ruined by removal to Rwanda; the integrity of the international refugee protection regime and the UK’s international reputation, to name but a few.

We did not weigh the costs of offering protection to Ukrainian refugees; that’s not the morally right thing to do. We didn’t prevent their entry because they could have claimed asylum elsewhere. We showed solidarity with neighbouring and other European countries so they did not bear the burden without us. The same ought to be true for other refugees.

But, on any rational level, financial cost is relevant. The government has a limited amount of money and spending a great deal of that money on a political vanity project that is unlikely to have the desired effects in any event is utterly indefensible by anyone.

Does deterrence work?

The government admits that it has no idea how far the Illegal Migration Bill will deter arrivals or lead to alternative clandestine entry. But, for the first time, we can see an attempt to engage with the issue of deterrence. The document acknowledges that

the academic consensus is that there is little to no evidence suggesting changes in a destination country’s policies have an impact on deterring people from leaving their countries of origin or travelling without valid permission, whether in search of refuge or for other reasons.

It then goes on to suggest that looking at real world examples might suggest the academic consensus is wrong.

Australia is cited as an example of a country where deterrent policies were implemented. The impact assessment acknowledges, though, that deterrent policies were accompanied by actual interception at sea, after which the refugees were

either returned to Indonesia, removed to third countries in the Pacific, or sent to Australia’s immigration facilities at Christmas Island (which had been made an overseas territory for this purpose).

The number actually detained at Christmas island or removed to a third country compared to being returned to Indonesia is not stated. It is “difficult to disentangle the impacts of the individual measures” as the assessment itself states, and there was also evidence of displacement onto alternative routes.

Almost all the other examples cited in the impact assessment — Spain/Morocco, Italy/Libya, EU/Turkey, Scandinavia and UK/France — involve simple security measures, in the sense of physic ally blocking arrivals. The one possible example of deterrence cited is the UK/Albania deal, which seems to have facilitated swift removals from the UK to Albania, which may have had a deterrent effect. In any event, arrivals from Albania certainly fell sharply after the deal was reached.

The assessment goes on to recognise that

Any deterrence impact relies on the policy working as intended, with sufficient capacity to detain and remove an appreciable proportion of individuals in scope to a safe third country.

Legal challenges might prevent implementation, which might reduce the deterrent effect. It’s a bit like sailors complaining about the weather; of course individuals facing removal to Rwanda will want to challenge their removal. If the policy depends on their not doing so, it is bound to fail. The document recognises that changes in behaviour, such as switching more clandestine routes, might also be one unintended outcome of the Bill (something I’ve written on before: How will refugees react to the Illegal Migration Bill?).

What are the financial costs?

The impact assessment attempts to count the per person costs and savings of the proposed scheme. Estimates are always going to be wrong but there’s reason to think these estimates are VERY wrong. The costs are summarised as follows:

The overall cost saving the Home Office claimed to be able to quantify was £106,000 per person. The net cost is therefore £63,000 per person, according to these figures.

The estimates do not include the additional cost of building or converting immigration detention facilities. These are likely to be considerable given existing capacity is something like 2,500 places and the government is committing to detaining and removing thousands of people.

Bizarrely, the impact assessment assumes that a person would have to be accommodated in the United Kingdom for four years if it weren’t for the Illegal Migration Bill. It is perfectly possible to make most asylum decisions within six months, as the government was managing to achieve until 2017. The government is now itself making strenuous efforts to reduce the asylum backlog by streamlining processes and recruiting more staff. 75% of initial decisions are grants of asylum, meaning the Home Office is no longer liable to support the person concerned and they are allowed to work.

The protracted waiting times at the moment are likely to have a dampening effect on future refugee earning potential. Time out of the workforce does that to anyone. But a lot of refugees will presumably want to work. Nick ‘Windrush’ Timothy, former adviser to Theresa May, does a Twitter drive-by on Afghans, Syrians and Eritreans, asserting that “on average not one of these nationalities pays more in income tax or NICs than they cost in tax credits and child benefit”.

The link he provides proves no such thing as far as I can see. On the face of it, this looks like gratuitous racism from Timothy, who is thought to be one of the architects of the Bill. I’m not really into attaching a financial cost or benefit to human beings but it seems unlikely Timothy is himself a net contributor given that we know that £62.7m has been paid out in the Windrush compensation scheme so far.

Anyway, the impact assessment itself says:

The Home Office do not know the net impact on public services of this cohort – the money they contribute (whether this be directly or indirectly) versus the services they consume – and therefore this has not been monetised in this appraisal.

So there is no reason at all to think that support would normally be needed for a period of four years; this is obviously a very considerable exaggeration of the alleged cost savings of the scheme.

Weirdly, there’s also no reference so far as I can see to some of the other costs of the asylum system which presumably the Home Office would normally claim would be saved. The cost of assessing asylum applications would presumably be saved, as would the costs of appeals. The equivalents under the Illegal Migration Bill are severely curtailed, so presumably cheaper than at present.

If the Bill does have a deterrent effect — the impact assessment does not actually anywhere claim that it will — then there would also be savings in border security and search and rescue.

What’s not in the impact assessment?

The impact assessment does not attempt to quantify the extent to which the Illegal Migration Bill might actually deter refugees from coming to the United Kingdom. The calculations assume that every single refugee will be removed in accordance with the legal duty imposed on the Home Secretary to remove all illegal entrants who arrived since 7 March 2023.

Even now, that is already thousands of refugees. The government does not know how many will continue to come here. That would be a lot of people to force onto flights to Rwanda.

It seems extremely unlikely that the government will be able to remove all of these two groups. Indeed, the impact assessment tacitly acknowledges this reality in the section on deterrence. For there to be any deterrent effect, an “appreciable proportion of individuals in scope” would have to be removed. What happens to the others and what costs are associated with them?

There will be some sort of remainder of refugees who will never be granted immigration status and never given the right to work but also have to be supported. The impact assessment is silent on the cost of this new perma-backlog.

It also looks like the impact assessment underestimates the scale of the challenge facing the government. Repeated references are made to the 45,000 refugees who arrived in small boats in 2022. Reference is also made to a further 3,400 refugees who were detected at port or within the UK within 72 hours of their arrival. But in fact 89,398 people claimed asylum in 2022, including dependents. The vast majority of those would fall within the scope of the Illegal Migration Bill, as far as I can see. So it is not clear why the impact assessment implies that fewer than 50,000 would have faced removal in 2022 had the legislation been in place.

Finally, as I was writing this I saw that Migration Observatory put out a very good thread on Twitter making many of the same points I was making here plus some more. You can check it out for yourself:

Interested in refugee law? You might like Colin's book, imaginatively called "Refugee Law" and published by Bristol University Press.

Communicating important legal concepts in an approachable way, this is an essential guide for students, lawyers and non-specialists alike.

Buy now