Unlock the Editor’s Digest for free
Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.
The “mansion tax” was arguably the standout policy in the last Budget. At the time, the feeling among many in the property industry was that it could have been worse — an annual charge starting at £2,500 on homes above £2mn hardly felt like it had the market-distorting potential of some of the other ideas that were being mooted.
That didn’t stop it being met with howls of outrage, of course. It has been branded by some as “anti-aspirational”, “spiteful” and a “burglar’s charter”.
My problem with it is somewhat different: I think it’s very poorly designed.
The high‑value council tax surcharge (HVCTS), to give it its proper name, will be paid to central government, but collected by local authorities. Starting in April 2028, the additional charge is expected to raise about £400mn.
Last week, the chief executive of the Valuation Office tried to shed some light on how the process would work during a Treasury Select committee hearing. It has not changed my view.
Jonathan Russell told MPs that between 150,000 and 200,000 properties were “within scope” of the new tax — though he did not explain how many more would need to be valued to find those (saying only that they would be looking at properties worth more than £1.5mn). As local authorities have no experience in collecting taxes from owners — rather than occupiers — as was pointed out at the hearing, it remains to be seen whether HM Revenue & Customs will also need to be involved.
The charge is structured in fixed “cliffs”: £2,500 for properties worth £2mn to £2.5mn; £3,500 for £2.5mn to £3.5mn; £5, 000 for £3.5mn to £5mn; and £7,500 for anything above that.
The current council tax system has been routinely criticised for disproportionately affecting those in poorer parts of the country, who end up paying far more than those who live in luxury homes in the capital. Will the new HVCTS help? From 2028, a nurse in Nottingham with a modest band C house will still pay roughly 1 per cent of her home’s value in tax, whereas a hedge fund manager in Kensington or Wandsworth in a £20mn mansion will pay about 0.05 per cent of theirs. Policymakers appear to regard this disparity as acceptable, even desirable.
The most obvious flaw is the abrupt jumps at each band: a homeowner whose property is valued at just over £2mn could suddenly owe an extra £2,500 per home, though tax deferral is on the cards. One rationale for broad valuation bands, as embedded in the whole council tax system, is that they supposedly generate fewer appeals than point valuations.
In practice, this claim doesn’t hold up. Homeowners whose properties sit near the bottom of a band are often motivated to appeal, because being placed in too high a band can impose a substantial financial burden — especially when the bands are wide.
In contrast, when taxes are calculated directly from a property’s assessed value, a 5 to 10 per cent over‑assessment typically has only modest financial impact. Moreover, in the UK a higher official valuation facilitates equity withdrawals and improves sale values. Given the usual uncertainty around assessments, owners are unlikely to appeal an over‑valuation of this size.
Apart from Ireland, the rest of the world rejects the use of valuation bands for property tax. And though the Irish system is needlessly complex, bands are based on recent market values unlike the UK’s retention of outdated 1991 values. Moreover, the Irish use of many more bands greatly reduces cliff edges and at the top, the Irish tax is proportional, rather than capped at the UK’s low level.
A proportional design to the mansion tax would be fairer and raise more revenue. If the surcharge were 0.5 per cent on the amount above £2mn, a £2.1mn home would incur a £500 charge instead of £2,500. This approach would likely lift total revenue to more than £1bn.
Crucially, the opportunity was missed for joint reform with stamp duty land tax. High marginal SDLT rates currently offset the regressivity of council tax for high‑value properties, but it still creates economic inefficiencies — and is deeply unpopular, especially among wealthy, internationally mobile buyers.
Halving the top marginal rate of stamp duty would more than offset the cost to an owner of a 0.5 per cent mansion tax on a £20mn home held for 15 years. While a lower SDLT rate would reduce the tax paid per transaction, it would stimulate many more property sales and boost corporation‑tax receipts from high‑end estate agents. Thus, a co-ordinated reform would not diminish total tax intake; it would improve fairness and economic efficiency.
A percentage‑based surcharge combined with a reform of SDLT would reduce sharp tax jumps and align fairness with economic efficiency and growth, showing that well‑designed policy can achieve both goals. With the supposed fiscal “black hole” less of an issue, the economic quality of policy design now needs to come more to the fore. It is unfortunate that the Scottish government has just chosen a version of this flawed design when sensible options were available.
The writer is professor of economics at Oxford university