Flight Risk Capital

Who is leaving Britain and what it costs to stay

Do high earners leave when taxes rise? The evidence

Some high earners leave when taxes rise, but the response varies by group and reform. UK studies find increased departures after the 2017 non-dom reform, while most affected taxpayers stayed.

Last checked .

By Flight Risk Capital. Published: . Last updated: .

What did Advani, Burgherr and Summers find in the UK?

Arun Advani, David Burgherr and Andy Summers studied the 2017 reform that removed non-dom treatment from long-term UK residents. This lets them compare affected taxpayers with a group whose treatment did not change.

Their April 2025 paper finds that departures rose significantly, but temporarily. Their estimate is that the number of affected super-rich people in the UK fell by 0.26% for each 1% fall in the net-of-tax rate. The net-of-tax rate is the share of income left after tax. An elasticity describes how strongly one measure changes when another changes.

The September 2023 version also reports 0.26, framed as a 0.26 percentage point rise in the emigration rate for each 1% decline in the net-of-tax rate. The versions describe different measures, so their wording matters. The latest version’s finding concerns the number of affected residents, rather than a universal departure rate for high earners.

What did HMRC’s 2024 non-dom evaluation find?

HM Revenue and Customs published its evaluation of the 2017 deemed-domicile reform on 30 October 2024. It found that affected condition B taxpayers were 10% to 12% more likely to leave after the change, comparing the tax year ending 2018 with the year ending 2015.

Those are HMRC’s relative percentages, not percentage point increases. HMRC also found that most affected taxpayers stayed and paid more tax on average after becoming deemed domiciled. It described this as a limited reform affecting a small part of the non-dom population. It does not settle the effect of the later abolition of the regime.

The non-dom guide explains the replacement rules.

Does HMRC’s 50p evaluation measure emigration?

HMRC’s March 2012 evaluation estimated that the underlying annual yield from the 50p additional Income Tax rate was about £1 billion or less, against an original expectation of £2.5 billion. It said the yield could be negative.

HMRC estimated a taxable income elasticity of 0.48. That measures the response of reported taxable income, including shifting income between years and avoidance. It is not a migration elasticity. The evaluation does not give a departure estimate that can be applied to today’s high earners.

What does the OBR assume about non-dom departures?

In its 30 January 2025 costing, the Office for Budget Responsibility assumed a 12% reduction in 2025-26 in the affected population without trusts and a 25% reduction for those with trusts and deemed domiciles. These are forecasting assumptions, not observed departure counts.

The OBR’s 2 April 2026 update said the overall revenue yield remained broadly unchanged after other changes to the costing. It did not restate those departure shares or say they had changed. The costing remained highly uncertain. See the Budget tracker for the latest announcements.

What do Kleven and others show outside the UK?

Henrik Kleven, Camille Landais, Emmanuel Saez and Esben Schultz’s 2014 Danish study found that a preferential tax scheme doubled the number of highly paid foreigners relative to slightly lower-paid foreigners. Its migration elasticity was between 1.5 and 2. This measures attracting new arrivals under a special scheme, rather than established residents leaving Britain.

Kleven, Landais and Saez’s 2013 European football study found an elasticity of around 1 for foreign players and around 0.15 for domestic players. Professional football is a highly mobile labour market. Its estimates cannot stand for all high earners.

Kleven, Landais, Mathilde Muñoz and Stefanie Stantcheva’s 2020 review, with a working paper published in April 2019, found that some high-income groups respond strongly to tax. It cautioned that the response depends on the setting rather than being a fixed number.

Sam Friedman, Victoria Gronwald, Andy Summers and Emma Taylor’s January 2024 study interviewed 35 people in the top 1%. None was then planning to leave for tax reasons. These interviews help explain attachment to place, but are not a representative estimate of how many people would leave.

Can the tracker answer whether tax caused a move?

No. The leavers tracker records sourced reports of named departures. It has no comparison group and cannot measure the effect of a tax change. The Departures board brings together official population and tax figures, whose coverage and dates differ from the studies above.

Our view: tax needs stable rules

Opinion. This is commentary. It gives the view of Flight Risk Capital.

Why does predictability matter for high earners?

People who can work or hold investments in another country have options when the terms change. They still need to compare real jobs and living costs, but an uncertain future bill makes planning harder.

This is an argument for rules people can understand and plan around. A relief needs a clear purpose and a test of whether it works. A tax increase deserves the same discipline.

How much revenue is involved?

HMRC reported £13.6 billion in combined Income Tax, Capital Gains Tax and National Insurance liabilities for non-domiciled and deemed-domiciled taxpayers in 2024-25.

Those figures show why the response of a defined taxpayer group deserves attention. They describe liabilities before the replacement regime began. They cannot be counted as revenue already lost.

The non dom guide explains the new rules and the Departures board preserves the scope of the old statistics. Mixing a tax-population total with selected departure reports would exaggerate what the evidence says.

What did the OBR assume about departures?

A forecast must make assumptions before every outcome can be observed. Our view is that ministers should publish those assumptions clearly and judge later data against them. That creates a testable policy claim.

Do named departures explain why people leave?

A report establishes the case described by its source. It cannot establish a shared motive across high earners. This article attributes no motive to any named person.

The leavers tracker records status and reporting basis. The salary comparison shows one financial part of a possible move. Neither measures all the reasons to stay or leave.

What should a more stable tax policy require?

A reform should say what it aims to achieve and how its costs will be measured. A later review should test both receipts and the response of affected taxpayers. That standard should apply to a tax relief as readily as to a rate rise.

See the Budget tracker for announcements and effective dates. The practical test is whether the policy comes with a clear timetable and evidence that can be checked later.

People who cannot readily move also need dependable rules. A policy built around the loudest departure threat would give them little voice. In our view, predictable taxation has a wider case: it helps people plan work and investment across the income distribution.

What else do readers ask?

Do millionaires leave when taxes rise?

Some do. Studies find different responses across groups and tax changes. Evidence about foreign footballers or new arrivals cannot provide a departure rate for all UK millionaires.

Does the leavers tracker measure tax-driven migration?

No. It records sourced reports about named moves. It has no comparison group and cannot measure how many departures a tax change caused.

Is the OBR non-dom departure assumption an observed count?

No. The OBR’s January 2025 shares are assumptions used to estimate revenue. They are not measured departures after the reform.

Where can I explore this further?

Full sourced leavers tracker. Articles and tracker roundups.

Sources

Newsletter: sign-ups are closed.