GBTT Investigation · Tax · Wealth
Twelve OECD countries had an annual wealth tax in 1990. Three still do. Britain already raises over £45 billion a year from taxes on accumulated wealth, none of which are called wealth taxes. The case for a new one has not gone untested. It has been tested, and the results are public.
The current British wealth tax debate has a centre of gravity. Modelling from the Wealth Tax Commission and the IFS suggests an annual levy of 0.17 per cent on net wealth above £500,000 would raise around £10 billion in gross revenue. The Institute of Economic Affairs and the Centre for Policy Studies have argued the proposal underweights administrative costs and ignores the international evidence; supporters argue it is a small price to ask of the very wealthy. The figure that makes the proposal sound modest, £10 billion, is gross. The net is unknown, because every published estimate omits the same three things: the cost of building a national asset register, the behavioural response of the people the tax targets, and the recurring annual cost of valuing assets that have no market price.
The British debate has been conducted as if the country were considering its first tax on accumulated wealth. It is not. The UK already operates one of the broadest stacks of de facto wealth taxes in the developed world. Most are not labelled as such, which is precisely why the proposal for a new one rarely engages with them.
Stamp Duty Land Tax raised £15.2 billion in 2025-26, levied at up to 12 per cent on residential purchases above £1.5 million plus a 5 per cent surcharge on second homes. Stamp Duty Reserve Tax adds 0.5 per cent to every UK share purchase regardless of whether the trade is profitable, and raised around £3 billion. Inheritance Tax, Britain's most unpopular levy, hit a record £8.5 billion in 2025-26. Capital Gains Tax leapt to a record too, roughly £20 billion in the twelve months to January 2026, up 44 per cent on the year after the October 2024 rate rises. Since the indexation allowance for individuals was abolished in April 2008, CGT has been charged on nominal rather than real gains, so the holder of a long-held asset can pay tax on what is mathematically pure inflation. Council tax, banded against 1991 valuations and never revalued in England, functions as a regressive recurring tax on property wealth at around £46 billion a year. The Vehicle Excise Duty Expensive Car Supplement adds £425 a year for five years to any vehicle with a list price above £40,000. The Air Passenger Duty premium-cabin rate reaches £253 per long-haul departure.
None of these taxes is called a wealth tax. All of them function as one. Their combined annual yield, excluding council tax, is now over £45 billion, more than four times the gross figure modelled for the proposed annual levy. Including council tax the figure exceeds £90 billion. The UK does not have a missing wealth tax. It has a portfolio of them, badly coordinated, with thresholds frozen at different historic dates and reliefs that interact in ways no Treasury document has set out in full.
Methodology: SDLT, IHT and CGT figures are HMRC receipts for 2025-26 (CGT stated as the twelve months to January 2026, its most recent published window). Council tax is a 2025-26 figure. The CGT total reflects a one-off leap after the October 2024 rate increases and the carried-interest reform landing from 2026-27. The category "wealth-related" excludes income tax on dividends and rents (taxes on flows from wealth, not on the stock).
In 1990, twelve members of the OECD operated a recurrent tax on individual net wealth: Austria, Denmark, Finland, France, Germany, Iceland, Luxembourg, the Netherlands, Norway, Spain, Sweden and Switzerland. The OECD's 2018 review counted four. Today the working definition typically lists three: Norway, Spain and Switzerland.
The pattern of abolition is itself the most informative dataset in the debate. Austria abolished its wealth tax in 1994. Denmark and Germany followed in 1997. The Netherlands replaced its version in 2001. Finland, Iceland and Luxembourg abolished theirs in 2006. Sweden, the country most often cited as a model of high-tax democratic governance, repealed its wealth tax in 2007 and its inheritance tax in 2005, both by parliamentary consensus. France abolished its general impôt de solidarité sur la fortune in 2018 after evidence that it was driving wealth abroad, and replaced it with a much narrower property-only levy. Switzerland's surviving wealth tax operates at the cantonal level alongside income tax rates roughly half those in the UK; it is a substitute for a heavier income tax base, not an additional layer on top of one.
The OECD's own 2018 study, conducted by tax economists rather than political advocates, concluded that wealth taxes had been abandoned principally because of "high administrative and compliance costs, in particular compared to their limited revenues", and the consistent finding that exemptions and reliefs introduced to make the tax workable also undermined its redistributive purpose.
In 2022, Norway's Labour-led coalition raised the top wealth tax rate from 0.85 per cent to 1.1 per cent on net assets above NOK 20 million. The Ministry of Finance projected an additional NOK 1.5 billion in annual revenue, equivalent to roughly $146 million.
What happened next is, in the words of the Norwegian financial press, the most expensive single tax change of the post-war era. Within twelve months, more than thirty of the country's wealthiest individuals had emigrated, more than the previous thirteen years combined. By 2024 the cumulative figure had passed three hundred. The departures included Kjell Inge Røkke, the country's largest single taxpayer for several years, who relocated to Lugano and removed an estimated NOK 175 million a year (~$16 million) from the Norwegian tax base on his own. Independent analysis by Norwegian economists estimated that the individuals who left between 2022 and 2024 would have paid around NOK 6.5 billion (~$594 million) in wealth and dividend tax over the period had they stayed. The intended uplift was $146 million. The estimated result was a net hole roughly four times the projected gain, in the wrong direction.
The exodus produced a second, less obvious effect. Norway's largest companies are still substantially family-owned. The wealth tax falls on shares whether or not those shares produce dividends. Family business owners describe the tax as creating an annual demand for cash from a non-cash asset, in many cases forcing the partial sale of equity to outside investors merely to fund the levy. Several of Sweden's documented reasons for abolition in 2007, accelerating sales of family firms to foreign capital, are now repeating across the border in real time.
Any annual wealth tax requires, at its base, a register of what is taxed. The Wealth Tax Commission's own background paper on administration is unusually candid: HMRC does not currently have records on the value of individual properties, let alone of the underlying private companies, art collections, agricultural land, pension assets, intellectual property, foreign trusts or unlisted equity stakes that constitute the bulk of upper-end wealth. The Commission estimated the up-front build cost of a UK wealth tax at £600 million, more than 10 per cent of HMRC's full annual operating budget, before a single return has been processed.
Even if the register existed, every individual contribution to it generates its own market. The Commission's same paper records that valuing an unlisted private company can cost £2,500 to £25,000 per company per year. Disputes routinely reach £10,000 in legal costs and beyond £20,000 in protracted cases. These are professional fees paid by the taxpayer, not the state, and they recur every twelve months. The same applies to art, intellectual property, agricultural land, racehorses, classic cars, jewellery, and any business asset whose value is opinion rather than tape. None of this is hypothetical: France's ISF generated thousands of cases on art valuation alone before its abolition.
The argument that a wealth tax could be quietly introduced without changing behaviour collides with what Britain has just watched happen. In its 2025 report, Henley & Partners forecast a net outflow of 16,500 millionaires from the UK, a figure reported worldwide as if it were a headcount. In its 2026 edition the firm dropped precise counts altogether, saying the data did not support them, and described current millionaire migration as “modest” — a retreat that followed sustained criticism of its method. The exact number is contested, and GBTT will not assert one that its own source has withdrawn.
The hard data complicates the exodus story rather than confirming it. HMRC's own statistics show the number of non-doms fell by around 1,200 in 2024-25, the last year before the regime was abolished, but that was driven as much by fewer arrivals as by departures: the roughly 9,000 non-doms who left was actually down on the 11,200 who left the year before, and the group that remained paid a record £9.7 billion in UK tax. The full behavioural response to abolition, which only took effect in April 2025, will not show in the figures until later returns.
What the episode establishes is narrower, but real. This is a mobile population; the Treasury judged the risk serious enough to soften the non-dom reform once the scale became visible; and that mobility is exactly what a recurring wealth tax would test. Norway and France documented the same behavioural response before their own reversals. The lesson is not that Britain has already lost the wealthy, it plainly has not, but that the state has watched how they respond to a far smaller change than a wealth tax, and blinked.
Britain has not failed to tax wealth. It has taxed it through seven different levies, in seven different ways, with the thresholds frozen across seven different decades.
The IFS's institutional position, set out across a sequence of papers since 2020, is not that wealth is undertaxed in Britain. It is that wealth is undertaxed in the wrong places. The Institute argues that the existing British tax base mistreats the flows from wealth, the dividends, capital gains, rents and inherited transfers, far more than it underweights the stocks themselves. Closing the gap between earned income, dividend income and capital gains; reforming council tax bands frozen at 1991 valuations; tightening pension-IHT interactions; and removing distortive reliefs in agricultural and business property would, on IFS modelling, generate revenue comparable to a wealth tax at a fraction of the administrative cost and without the recurring valuation problem.
This is the position the IEA and CPS converge on from the other direction. Both reject a recurring wealth tax outright but agree the reform agenda is real: the British system is full of frozen thresholds, distortive reliefs and base inconsistencies that produce exactly the inequalities the wealth tax is intended to address, without the international evidence of failure attached to the wealth tax solution. The points of agreement between the IFS, IEA and CPS are themselves a data point. None of them recommends an annual wealth tax.
Spain's solidarity tax raised €632 million in 2024, less than 1 per cent of Spanish tax revenue. France's general wealth tax, in its final years, raised below 0.4 per cent. Norway's wealth tax revenue is around 2 per cent of total receipts and falling against the cost of departed taxpayers. The proposed UK wealth tax, at the modelled £10 billion gross, would represent under 1 per cent of the roughly £1.1 trillion the British state currently collects.
The same dynamic that hollows out IHT applies to a wealth tax with extra force. The political heat is generated by the visibility of the very rich; the revenue is generated by the threshold catching the merely comfortable. The marginal rate paid by the family with a £700,000 home, a £200,000 SIPP and £50,000 in savings will, in every functioning model, be higher than the marginal rate paid by the diversified holder of unlisted equity, agricultural land and family trusts who can afford the advice to manage them.
None of this is the fault of the people working inside the system. It is the cumulative output of a tax regime that has spent thirty years adding levies to wealth without measuring or reforming the levies it already had. Twelve OECD countries had an explicit annual wealth tax in 1990. Three still do. The countries that ran the experiment kept the lessons. They did not keep the tax.