Would Pensions Count as Wealth?
The worry behind this objection is a specific person: a teacher or a nurse retiring on a defined benefit pension, who has never thought of themselves as rich and is told that a wealth tax would count the pension. In the official statistics it would. Whether a tax would charge it is a different question, and the difference is where the whole answer lives.
The objection at full strength
The Office for National Statistics measures household total wealth as four components: net financial wealth, private pension wealth, physical wealth and net property wealth. Across all households in Great Britain in the April 2020 to March 2022 survey, net property wealth was the largest share at 40 per cent and private pension wealth the next at 35 per cent; net financial wealth was 14 per cent and physical wealth 10 per cent. For the wealthiest 10 per cent of households, property was 38 per cent and pensions 36 per cent. The same bulletin puts median household wealth at £293,700 and entry to the wealthiest 10 per cent at £1,200,500; entry to the wealthiest 1 per cent is £3,121,500. Pensions are about a third of what British households own, and the share does not fall on the way up into the top tenth.
A pension is also the least liquid thing most people own. The operational definition used in the UK wealth-tax literature, from Loutzenhiser and Mann, treats all pensions as illiquid alongside property, business assets and physical possessions, with only net financial assets counted as liquid; the authors describe it as a modelling assumption rather than an official classification. A charge on a pension usually cannot be paid out of the pension. On that reading the objection has its full weight: a wealth measure that includes pensions reaches the middle of the country, in an asset that cannot be sold.
Counting and charging
Wealth is a stock, measured as what a person owns at a point in time less what they owe, where income is a flow measured over a period. The ONS includes private pension wealth in that stock, beside houses, savings, businesses and possessions, because a pension is a claim on future money that has a present value, and the statisticians put a value on it for every household they survey. So a definition of net wealth can count pensions, and the national one does.
Whether a tax charges them is a separate decision, and the threshold makes it before any exemption does. The reference model this series argues for is a proposal Parliament has not enacted: an annual charge on individual net wealth above £10 million, assessed on each person separately, and marginal, so wealth below the threshold is never charged. Under that design the question for the retiring teacher is whether everything she owns, pension included and after her debts, exceeds £10 million. The objection is about whether a pension is wealth. The design turns on whose balance sheet crosses the line.
Big in the middle, small at the top
The distribution answers that for almost everybody. Advani, Bangham and Leslie, working from the earlier 2016 to 2018 round of the same survey and measuring per adult within the family, report that the median family has just over £100,000 in net wealth per adult, and that a family only just inside the top 1 per cent has £1.9 million per adult. Above £5 million per adult the composition changes shape. Their Figure 14 gives, for families above that line, business assets at 42 per cent, financial assets at 39 per cent, property at 14 per cent and pensions at 4 per cent. Business and financial holdings together are 81 per cent; property and pensions together are 18 per cent. Across the figure’s lower thresholds the business-and-financial share barely moves: £250,000 (18 per cent), £500,000 (17 per cent) and £1 million (18 per cent). At £2 million it is 27 per cent, and above £5 million it is 81 per cent.
So the asset that is about a third of wealth across all households is 4 per cent of it among families above £5 million per adult. The same authors draw the corollary for design in their own words: if the wealth tax threshold is set at a low level, the biggest sources of revenue would be property wealth and pension wealth, and excluding those asset types would reduce the tax base by 80 per cent. At a low threshold a pension carve-out guts the tax. At a high one the pension has already dropped out of the picture on its own.
Where the objection is right
Three limits on that evidence need stating. Each calls for care in reading it.
Get Rajeev Shah’s stories in your inbox
Join Medium for free to get updates from this writer.
The two sets of figures are not on the same footing. The ONS shares are per household and from the 2020 to 2022 round, on a four-way split of wealth; the Warwick authors’ shares are per adult within the family and from 2016 to 2018, on a five-way split that separates business assets from financial ones. A household figure and a per-adult figure can be read side by side. They cannot be subtracted from each other.
The survey behind both has lost its badge. On 13 June 2025 the ONS suspended the accreditation of the Wealth and Assets Survey from Round 8, the 2020 to 2022 period, onwards, citing concerns with the quality of estimates at a granular level and a further fall in response rates; that round was collected entirely by telephone because of the pandemic, and the ONS warns that its data can be particularly volatile at the extreme ends of the wealth distribution. Every figure in this article rests on that survey.
And the survey runs out before the threshold. Advani, Miller and Summers write that household surveys face well-documented difficulties with undercoverage and under-reporting at the top, and that the sample size at the very top, wealth above £10 million for the UK, is too small to draw reliable conclusions on the scale of wealth, let alone its composition. The 4 per cent is measured above £5 million per adult. What pensions are as a share of wealth above £10 million cannot be read from this data, and the direction of the trend is the most that can be claimed.
What the design has to contain
Value pension rights on a stated basis, and let the threshold decide who is charged at all. The stated basis already exists in official statistics: the ONS produces a value for private pension wealth for every household it surveys, so a tax would be choosing a method rather than inventing one, and the choice belongs in the statute where taxpayers, advisers, valuers and tribunals can all read it.
The threshold does the rest. The OECD’s 2018 review of net wealth taxes lists high exemption thresholds, a comprehensive base, consistent valuation, limited exemptions and liquidity provisions among its design guidance, and makes the first of them conditional: where a net wealth tax exists in addition to broad-based capital income taxes, exemption thresholds should be high to ensure that the tax is levied only on the very wealthy. The same report concludes that where those capital income taxes and well-designed inheritance and gift taxes already exist, the arguments for adding a net wealth tax are limited, because it finds such taxes tend to be more distortive and less equitable, being imposed irrespective of the actual returns that taxpayers earn on their assets. The UK has both taxes, and nothing here is offered as OECD endorsement of a UK wealth tax. What the list does carry is both a high threshold and a comprehensive base. Exempt pensions and the base narrows for everyone above the line; raise the threshold and the retiring teacher leaves the tax without the base changing at all.
A high threshold does more for an ordinary saver than a carve-out does. Under the reference model the charge reaches roughly 0.04 per cent of UK adults, and a person below the line pays nothing on any asset, pension included. Above it, a pension is a small part of a large balance sheet, and a rule that values it on a stated basis treats it like everything else on that sheet.
Liquidity is the question this article does not settle. A pension cannot be drawn down to pay a charge, and the answer lies in the payment rules, instalments, deferral, interest and security among them, which this series takes up separately. Start with the threshold, then the exemptions.
Sources and evidence status
Office for National Statistics, Household total wealth in Great Britain: April 2020 to March 2022 (released 24 January 2025), for the four-component definition and the household shares of wealth by component for all households and for the wealthiest 10 per cent, and Household total wealth in Great Britain quality and methods guide (revised 7 July 2026) for the definition. ONS notice of 13 June 2025 suspending accreditation of the Wealth and Assets Survey from Round 8. Advani, A., Bangham, G. and Leslie, J., The UK’s wealth distribution and characteristics of high-wealth households, Wealth Tax Commission Evidence Paper 1 (Warwick Economics Research Paper 1367), 2021, also in Fiscal Studies 42: the median and top-1 per cent thresholds and the tax-base corollary from the text, the composition above £5 million per adult and the trajectory across thresholds from Figure 14. Advani, A., Miller, H. and Summers, A., on the sample size of UK household surveys above £10 million. Loutzenhiser, G. and Mann, E., the liquid and illiquid asset definition used in UK wealth-tax analysis. OECD, The Role and Design of Net Wealth Taxes in the OECD, OECD Tax Policy Studies No. 26, 2018: the design guidance list and the overall conclusion at the Executive summary, pp.11 to 12. Threshold, tax unit, marginal structure and scope of the model this series references: a policy proposal, not enacted law.
The ONS definitions and shares, the Advani, Bangham and Leslie figures, the Advani, Miller and Summers statement on sample size and the accreditation notice are T1 Verified Facts drawn from the named documents, all of them resting on the Wealth and Assets Survey, whose accreditation has been suspended from Round 8. The ONS household shares and the per-adult family shares are on different units and periods and on different partitions, and are presented side by side, never combined. The Loutzenhiser and Mann definition is the authors’ modelling assumption. The OECD material is T5 Economic Interpretation, quoted as the report’s own wording, and nothing from it is presented as OECD endorsement of a UK net wealth tax. The reference model’s threshold, unit, marginal structure and scope are a T4 Policy Proposal; whether pension rights would enter its base, and on what valuation basis, is a design choice this article states as open. No figure for the share of pension wealth above £10 million is asserted here, and none should be inferred.
Nothing here is tax advice.
Originally published at taxwealthnot.work.
