The figure borrows the Migration Advisory Committee’s lifetime estimates for the 2022/23 Skilled Worker cohort. MAC’s work is careful and openly caveated. IPPR has kept the parts that flatter and dropped the parts that don’t.
IPPR’s own methodology table gives the game away. Of the £115bn, £113bn comes from a single group: 164,000 skilled worker main applicants outside health and care. Every other group put together nets out at around £2bn.
The losses are there, but buried. IPPR’s own table shows health and care adult dependants at minus £9bn. The government’s estimate for care workers and their adult dependants settling in the same period is a net cost of £9.5bn – yet IPPR folds care workers into a single health and care line alongside doctors and nurses, so the loss disappears into an average.
Even that one profitable group is not what it seems. MAC finds that 72% of its fiscal gain comes from the top 30% of earners. The top tenth alone – paid at least £131,000 on arrival – supply 39%, averaging £2.7m each over their lives. The bottom tenth supply 1%.
The typical skilled worker looks very different. The Home Office puts median earnings for a Skilled Worker main applicant at £56,600 in 2023/24, well below the £75,700 mean behind MAC’s figure. And MAC finds there is “almost no age at which the median person is fiscally positive” over the rest of their life.
IPPR counts main applicants and adult dependants – and stops there. Child dependants are left out entirely, even though MAC models them and they made up a quarter of all visas in the very cohort IPPR borrows from: 82,300 children. MAC puts their first-year cost at around £13,100 each, mostly schooling, and finds child dependants a net cost even before counting those who stay.
Skilled workers do not come alone. In MAC’s cohort, every 100 workers outside health and care brought 72 dependants; every 100 health and care workers brought 108. Home Office figures for the year to March 2026 show 1,029 Pakistani skilled workers accompanied by 6,155 dependants, and 230 Nigerians by 1,114.
IPPR applies per-person estimates for people who arrived in 2022/23 to whoever happens to settle in 2026-30, and admits in its own notes that the two groups will differ. Those settling are the ones who stayed. MAC warns that if higher earners are the likelier to leave, its lifetime estimates are too high.
Most of those settling arrived when the general salary threshold was around £26,000 and care workers could come in below it. Today’s £41,700 threshold does not apply to them.
IPPR’s sums rest on a specific assumption: that the people who settle will look like the average skilled worker who arrived. The MAC’s latest evidence shows otherwise.
Its May 2026 study followed skilled workers who arrived between 2014 and 2016. Eight years later, 77% of those paid under £40,000 were still here. Among those paid £125,000 or more, just 45% remained. People who left within five years had started on an average salary of £94,400, against £70,900 for those who stayed.
Nationality tells the same story. Americans, with median pay of £112,100, had a five-year stay rate of 66%. Nigerians and Filipinos, whose Health and Care visa holders earn medians of £27,400 and £35,500, stayed at 93% and 92%.
This matters because MAC finds that 72% of the fiscal gain from skilled workers outside health and care comes from the top 30% of earners. Those are the people most likely to leave before they ever settle. By applying cohort-wide averages to those settling in 2026-30, IPPR keeps the biggest contributors in its sums long after many have gone home. The real settling cohort will be lower paid, and costlier, than its £115bn assumes.
Skilled Worker visa holders cannot claim most benefits until they settle, and MAC builds that in. The costs come after settlement – precisely what IPPR wants to bring forward. MAC itself names post-settlement benefit take-up as perhaps the most uncertain part of its estimates, uncertain enough to turn groups near zero from positive to negative. IPPR’s headline carries none of that doubt.
The direction of travel is clear. Households with at least one foreign national received £941m in Universal Credit in March 2025, double the £461m of March 2022 and 15.5% of all payments. Most of that goes to people with settled status of one kind or another, not current visa holders. Settlement is the gateway, and IPPR wants it opened sooner.
The weaker groups are already visible in MAC’s own numbers. Care workers cost an estimated £36,000 each over their lifetimes; their adult dependants £67,000, or £92,000 if their employment does not improve. Count the children, and the whole health and care route barely breaks even, at +£0.2bn.
Nor does IPPR mention that MAC’s figures exclude debt interest. MAC asks readers to bear that cost in mind; shared across adults, it would cut each person’s contribution by about £1,800 a year.
IPPR argues that the government’s £10bn figure for care workers is not a real saving, because earned settlement delays settlement rather than preventing it. Apply the same logic to IPPR’s own £115bn and the argument collapses.
A skilled worker on a visa pays exactly the same tax as one with settled status. MAC assumes every main applicant works, and claims nothing, until settlement. Making them wait forfeits none of that revenue. What settlement changes is the right to stop working and draw on benefits.
IPPR’s last line of defence is that high earners might be deterred. But someone paid £130,000 is here for the job, and pays full tax on it whether or not they hold settled status. If a career in Britain is not draw enough without a guaranteed path to permanent residence, that is no reason to hand one out.
IPPR’s case rests on a premise that should be rejected outright: that settlement is something you qualify for by paying enough tax. Indefinite leave to remain is the gateway to citizenship. Being British is not something that can be bought – not with a care worker’s five years of service, and not with a banker’s tax bill.
Even on IPPR’s own narrow terms, MAC concludes that for lower-earning skilled workers, temporary migration is better for the public finances than permanent settlement. But who joins the national community should never have been reduced to a balance sheet in the first place. Ministers should:
1. Reject any exemption for care workers and any “interim status” that bypasses the qualifying period.
2. Apply the ten-year default, and fifteen years for jobs below graduate level, to everyone already here.
3. Scrap the income-based reductions that would let high earners buy a shorter wait.
4. Link visa records to DWP benefit data before anyone in the 2026-30 cohort settles.
A work visa is a contract for work, not a down payment on citizenship. Settlement should be slow, rare and grounded in long residence and lasting commitment to this country – not priced up by a think tank’s spreadsheet.
(UKR)