ARTICLE AD BOX
By Chris Mason
Political editor, BBC News
The cost of deporting migrants to safe countries is likely to be expensive in the short term and will only save money if it puts off around four in 10 of those attempting to get to the UK.
That is the conclusion of what is known as the government's own "economic impact assessment" of the policy idea they want to pursue.
Rwanda is the first of a number of countries ministers want to send migrants to.
No migrants have yet been sent - the Rwanda scheme is being challenged in the courts - and ministers have refused to say which other countries they are talking to.
So what should we make of the numbers government officials have come up with, and the response from ministers and their political opponents?
Privately, some on the Conservative side are happy to talk up that it is not known how much their plan will cost or the extent to which it will act as a deterrent.
Quite an admission, you might think.
But the work of the government officials is, in truth, a bit more specific than that. It estimates it will cost more until it is having a substantial deterrence effect and plenty are being deported.
It is also true, though, that all of the numbers in their document are worth approaching with caution: there is a fair dollop of guesswork going on and lots of different variables at play.
Given this estimate suggests the idea comes with a net financial cost in the short term and huge uncertainty too, what is the case for it if you're a government minister?
In short, and these are my words, not theirs: they think it's worth a try.
"Stop the boats" is one of the prime minister's five priorities.
The current situation, in the Home Secretary Suella Braverman's words last autumn, is "out of control".
Or, as the Home Office puts it now: "The costs of accommodating illegal migrants have increased dramatically since 2020. The asylum system currently costs the UK taxpayer £3.6bn a year and £6m a day to accommodate 51,000 asylum seekers in hotels."
Logically, they would argue, if the existing situation is expensive and out of control, radical alternatives should be considered.
Or, as Ms Braverman argues, "doing nothing is not an option".
But, at best, it is an argument grounded in an expensive situation probably becoming more expensive in the short term and only becoming cheaper if it works. And it might not.
Enter next, Labour - who can and do point out that this assessment implies the government is "totally clueless" about the cost and deterrent effect of their policy.
And, remember, there isn't legal approval for the deal with Rwanda.
And there aren't any deals with anywhere else, at least yet.
Which leaves the prospect, Labour argue, of "people held in indefinite detention" and a "huge further increase in the asylum backlog" because there are no "workable" agreements in place.
But how confident is it realistic to be that any incoming Labour government would be able to transform things either - particularly in the short term?
The current system is expensive, complex and tricky to defend.
And the proposed eye-catching change is, yes, you guessed it, expensive, complex and tricky to defend.
Dealing with illegal migration is - and will continue to be - incredibly difficult.