What The 70% Number Hides

On Friday 26 June 2026, UK Home Secretary Shabana Mahmood announced a refugee sponsorship route modelled on Canada's Private Sponsorship of Refugees Programme. Universities, community groups, churches, and employers will be able to sponsor refugees to come to the UK, with the first arrivals in 2027. The Home Office is selling the policy on a striking data point: in Canada, seventy per cent of privately sponsored refugees find work within a year, against forty per cent for government-sponsored — a thirty-point gap.

The number is real. The cohorts it was earned on are not the cohorts the UK will be resettling. Statistics Canada's own data on 830,000 refugees admitted between 1980 and 2009 shows that ten years after arrival, refugees from the former Yugoslavia, Poland, and Colombia earned roughly double what refugees from Somalia, Afghanistan, Pakistan, and China earned. The same paper concluded that "very little of the variation in earnings among refugee groups could be accounted for by differences in observable human capital characteristics, economic conditions or the program of entry to Canada." The cohort itself predicts the outcome.

The top ten asylum nationalities granted protection in the UK in 2023 were Pakistan, Eritrea, Iran, Afghanistan, Bangladesh, Sudan, Iraq, Syria, Albania, and India (Refugee Council, citing Home Office data). The 2024 spike in applications was driven by a 113 per cent rise in Vietnamese, 79 per cent in Pakistani, 70 per cent in Syrian, 42 per cent in Bangladeshi, 36 per cent in Sudanese applicants (Migration Observatory, Oxford). None of those cohorts dominate the top of the Canadian earnings distribution. The 70 per cent headline is an average. For the cohorts the UK is actually resettling, the average is materially lower. The Home Office is selling the Canadian number to a British audience because the British cohort data would never sell the scheme.

The Austrian Diagnosis

The Austrian school of economics, a tradition going back to Viennese thinkers of the late 1800s and most associated today with writers like Ludwig von Mises (1881–1973), an economist who spent his career showing that centrally planned economies cannot price what central planners cannot value, and Friedrich Hayek (1899–1992), a Nobel-prize-winning economist who argued that no central authority can possess the dispersed knowledge held by millions of individuals, would ask a single question first: is each new arrival a net contributor from day one, or a net recipient?

An Austrian is not sentimental about this. A society is a productive-exchange system. The people who join it must, on net, produce more than they consume. If they do, the productive-exchange filter works and the society grows. If they do not, the filter breaks and the bills fall on someone else. The question is not whether a refugee is "vulnerable" or "deserving" — it is whether the productive-exchange filter, applied honestly, will let them through.

Mahmood's scheme does not apply that filter. It applies a much weaker test: that a private sponsor is willing to fill out the forms. The seventy-per-cent Canadian number does not show that private sponsorship creates productive-exchange outcomes. It shows that private sponsorship works for the cohorts it works for. The UK is delivering the cohorts it does not work for. The filter is being bypassed.

The Income Threshold Problem

Employed is not the same as net contributor. A refugee earning £20,000 to £25,000 pays income tax and national insurance, then consumes healthcare, schooling, housing support, child benefit, and eventually a state pension whose present value, discounted back, exceeds the tax paid over a working life. The income threshold for net fiscal contribution in the UK is roughly £30,000 to £40,000 depending on household composition (Office for Budget Responsibility methodology; House of Commons Library CBP-8513). Refugees sponsored under Mahmood's scheme will, on the available cohort data, start below that line. Many will stay below it for years.

This is the second place the productive-exchange filter is bypassed. The scheme assumes that being employed is the same as being a contributor. The two are not the same. A refugee earning £22,000 and claiming housing support is a smaller net recipient than a refugee earning £0 claiming the same support, but is still a net recipient. The Home Office will be able to claim the scheme "works" because employment rose. It will not be able to claim the scheme added to fiscal capacity, because on the cohort data, it will not.

The Scale Problem

Between 2015 and 2024, the UK resettled around 33,000 refugees through non-Afghan schemes — roughly one sixteenth of the number granted asylum in the same period (Migration Observatory, Oxford). Even at full capacity, the Mahmood scheme is unlikely to handle more than 5,000 to 10,000 sponsored refugees a year against a UK asylum caseload of more than 100,000 outstanding cases. The fiscal impact is in the tens of millions of pounds. The UK welfare bill is over £250 billion a year. The new scheme is rounding error against the number that matters.

This is not a reason to oppose the scheme. It is a reason to refuse the rhetoric. The Home Office is presenting a rounding-error programme as if it solved a structural problem. It does not. It cannot. And the public money it consumes is the public money that is not available for the productive-exchange filter that does work.

The Sponsor Problem

The productive-exchange filter, where it has worked, has worked because private sponsors had skin in the game and selected refugees from their own communities. Diaspora networks carry the early-year costs — housing, language, introductions, employment leads — that the state cannot manufacture and the refugee cannot pay for upfront. The Canadian data shows this clearly: privately sponsored refugees outperform government-sponsored refugees not because the private sponsors are nicer, but because the diaspora network is the productive-exchange filter.

The UK scheme is structured to suppress that filter. The Home Office is reserving the right to vet sponsors, cap numbers, approve community groups of five or more, and decide which universities and employers qualify. The diaspora networks that drove the Canadian outcomes — the Polish, Ukrainian, Vietnamese, and Colombian communities — will have to apply through a state-approved channel. The networks that do not have the political or organisational muscle to clear that channel will be filtered out. The state is removing the variable that made the Canadian number work, then claiming the Canadian number as evidence.

Why This Matters for Sound Money

Rails to Freedom argues in Part 4 that the goal of a sound-money economy is to make the productive-exchange filter work — to ensure the resources a society consumes are matched by the value it produces. The same principle applies to refugee policy. The goal is not a more efficient asylum process. The goal is to ensure that every new arrival clears the productive-exchange filter honestly — produces more than they consume, from day one, with no fallback onto the public purse.

The Mahmood scheme fails that test on three independent grounds: it picks the wrong cohorts, it counts employment instead of net contribution, and it suppresses the very private sponsors whose presence would make the filter work. The seventy-per-cent Canadian number is being deployed as cover for a policy that, on the available data, will produce a smaller but persistent net fiscal deficit for years to come.

There is a clean Austrian position available. It is not that refugees should not be admitted. It is that admitting them through a process that does not enforce the productive-exchange filter is the same mistake the state makes in every other market — centralising a calculation the central authority cannot make honestly, then claiming the result as policy success.

What Markets Are Already Doing

Civil society is doing what the productive-exchange filter requires. The UNICEF CryptoFund (Ethereum-native, transparent) accepts ETH and USDC for refugee operations with on-chain reporting. Gitcoin Grants has funded refugee programmes through quadratic funding — small donors weighted by the number of contributors, not the size of donation. World ID has piloted proof-of-personhood for refugee aid distribution so that aid reaches the recipient and not the broker. All three are Ethereum rails. All three are doing the diaspora-network job the state cannot do at scale.

The cleanest Ethereum tie: an open ledger of the productive-exchange filter. Every pound committed, every refugee resettled, every employment outcome, every net-fiscal contribution or cost, recorded on chain so the public can see whether the system is producing productive-exchange outcomes or just processing applications. Not a solution to UK asylum costs, but a precondition for an honest conversation about which mechanisms work, for which cohorts, and at what scale.

Looking Ahead

By this time next year, the UK scheme will have approved its first sponsors. The numbers will be small — a few hundred, perhaps a few thousand. The Home Office will say the scheme is working. The seventy-per-cent Canadian number will be cited. The cohort breakdown will be missing from the press release. The fiscal-net position of the cohort will not be measured at all, because measuring it honestly would force a conclusion the Home Office has no interest in reaching.

Five years from now, the UK scheme will have underperformed the Canadian baseline, because the cohorts the UK is resettling are not the cohorts the Canadian data was earned on, and because the state has reserved the right to veto exactly the sponsors whose presence would have made the productive-exchange filter work. The hard cases will remain in the state system. The productive-exchange filter will remain suppressed. The Home Office will claim success on the basis of the Canadian number, in a country where the Canadian number was never the relevant number.

The Austrian lesson is not that private sponsorship is wrong. It is that private sponsorship only works when the productive-exchange filter is honestly applied — when the cohorts are right, the calculation is honest, and the sponsors have the autonomy to do the work the state cannot do. The Mahmood scheme fails on all three. The Canadian number is being used to dress that failure as reform.