The Generational Balance Sheet

This is an In Discussion working paper (September 2026). The underlying data — spending by age band, the historical trend, wealth by age, and the size of the debt-interest bill — is solid and drawn from primary sources (OBR, ONS, the Intergenerational Foundation). What is not yet settled is how to combine those layers into a single "generational burden" figure: the note's own methodology section below explains why that step is the hardest part, and deliberately stops short of picking one answer. See the note at the end for what's still open.

1. The Numbers

1.1 Current spending by age band

The OBR's July 2026 Fiscal Risks and Sustainability Report breaks per-capita primary spending and receipts into five age bands (these figures exclude debt interest):

Source: OBR, Fiscal Risks and Sustainability — July 2026, Chart 1.2; ICAEW "Chart of the week" summary, 16 July 2026.

The OBR's own underlying chart uses ten age bands, not five — this table is ICAEW's five-band simplification. The finer-grained series sits in the report's Annex C and the supporting chapter spreadsheets, not yet pulled into this note.

1.2 The historical trend, 1999-2019

The Intergenerational Foundation tracked per-capita spending on children, working-age adults, and pensioners over nineteen years, using PESA and DWP benefit data. The underlying figures are independently sourced and solid; IF's own interpretive framing (it is an intergenerational-fairness advocacy charity, and language like "unjustifiably large" pension costs is editorial) is a separate question from the numbers themselves.

Three things drove the widening gap: the introduction of the triple lock in 2010 explains 71% of the single-year jump in pensioner spend in 2010-11; Sure Start funding per child fell from £1,600 to £377 between 2010-11 and 2018-19; and a 2015-16 PESA methodology change first captured public-sector pension debt interest, adding roughly £1,520 per pensioner to the measured figure — meaning pre-2015-16 figures understate true historical spend on a like-for-like basis.

No clean bridge exists between this 2018-19 series (IF, three age bands) and the 2026 OBR series (five age bands) above — different methodologies, a seven-year gap. Splicing them into one continuous series would need methodological reconciliation this note hasn't done.

1.3 Wealth stock by age

The ONS Wealth and Assets Survey, Round 8 (April 2020-March 2022, published January 2025), gives median household total wealth by age of household head:

Composition across the GB-wide median (£293,700, or £181,700 excluding private pensions): net property wealth 40%, private pension wealth 35%, net financial wealth 14%, physical wealth 10%. These figures exclude state pension entitlement, which would raise the older-age-band totals further if counted as an asset.

One caveat on the data itself, not the interpretation: the Office for Statistics Regulation suspended the Wealth and Assets Survey's accredited-statistics status from Round 8 onwards (June 2025), stating the data is "no longer of sufficient value or quality to meet users' needs." Treat the table as directionally indicative, not statistically precise.

1.4 The debt carry

Debt interest spending reached £110 billion in 2025-26 — now the third-largest area of public spending after health and welfare, having more than doubled as a share of GDP since just before the pandemic. Attributing that £110bn to specific age cohorts requires a causal model, not a measurement. Section 4 below is about exactly why that step is contested.

2. International comparison

Placeholder — not yet built out. The OBR's own July 2026 report shows the UK's old-age dependency ratio has risen relatively little over the past 50 years compared with other G7 nations, thanks to sustained net inward migration, but is projected to converge toward the G7 pattern of substantial increase over the next 50 years. The stronger candidates for this section — the OECD/EU Ageing Report's triennial ageing-pressure projections, and the G7 old-age dependency comparison already sitting in OBR Box 2.1 — haven't been pulled in yet.

3. The Steel Man

The case that current patterns are not evidence of unfairness

Pay-as-you-go is a designed feature, not a bug: state pension and NHS spending on today's pensioners is substantially funded by taxes those same pensioners paid across their working lives, under the same reciprocal logic today's workers are meant to benefit from in future. Framing this as a "burden transferred to the young" presupposes the system is a one-way extraction rather than an ongoing exchange operating as intended. Older cohorts also paid higher marginal rates historically — under the fiscal regimes of the 1970s-80s, before index-linked thresholds, older cohorts contributed at rates well above what today's equivalent earners pay, a fact any lifetime-net-position calculation needs to incorporate rather than reading off a single year's snapshot.

Much of the wealth stock in Section 1.3 is also substantially unrealised and systemically driven rather than deliberately engineered: a large share of the 55-74 wealth bands is property appreciation caused by decades of planning restriction and low interest rates, a supply-and-monetary-policy failure rather than an age-targeted transfer pensioners arranged for themselves. Private transfers flow the other way, too, and are typically excluded from this kind of accounting — inheritances and lifetime gifts move wealth from old to young, partially offsetting the public-spending skew. McCarthy, Sefton, Lee and Sambt's Generational Wealth Accounts extends the standard generational-accounting method to include this private-sector transfer, and asks whether public and private flows offset each other around the financial crisis — a live academic question, not a settled one. And pro-elderly policy itself — the triple lock, universal benefits — can be read as the ordinary output of a functioning democracy in which older cohorts vote at higher rates and organise more effectively: an uncomfortable demographic-weighting implication, but not self-evidently illegitimate "capture."

The case that current patterns represent a genuine imbalance

Per-capita spending growth on pensioners between 1999 and 2019 outpaced the growth of the pensioner population itself, which cannot be explained by demographic change alone. Education spending per child fell in real terms after 2010-11 while pensioner benefit spending rose over the same period — a policy choice, not a demographic inevitability. The OBR's own baseline projects health spending rising from 8% to 13% of GDP and state pension spending from 5% to 9% of GDP by 2075-76 under unchanged policy, meaning today's cross-sectional snapshot understates the trajectory facing currently-young cohorts once they reach old age themselves. And the gaps compound within generations, not just between them: already-wealthy millennials are set to inherit roughly four times as much as those with no property-owning parents, meaning the "young lose out" framing on its own masks a starker asset-holders-versus-non-asset-holders divide that cuts across age.

4. Why This Is Harder Than Stacking Three Datasets

The three data layers above are not the same kind of thing. Spending flow by age and wealth stock by age are both point-in-time measurements — solid, mostly uncontested (the wealth figures carry the WAS data-quality caveat noted in 1.3). Debt-carry attribution requires a causal model, which makes it genuinely contested rather than merely measured. And any single combined "generational burden" figure requires a normative framing on top of that model — which is where advocacy enters, on any side of the argument.

Three defensible methods exist for attributing the debt-interest bill to cohorts, and they produce materially different conclusions from the same underlying debt stock:

1. Retrospective — allocate interest to whichever cohorts consumed the spending the debt originally financed: the 2008 bailout, COVID furlough, decades of deficit-financed current spending.
2. Prospective — allocate interest to whoever's working-age tax base actually services it going forward, which mechanically loads the cost onto the young and future cohorts regardless of who benefited from the original spending.
3. Lifetime cohort accounting (the Kotlikoff-style approach) — model each birth-year cohort's full lifetime taxes paid minus benefits received, including its share of debt service across its whole lifespan.

These are not interchangeable, and any published figure needs to state which one it used. There are also technical problems that apply regardless of which method is chosen. Present-value lifetime-burden estimates swing substantially with the discount rate selected, and that needs disclosing rather than burying in a footnote. A defined-benefit pension's Wealth and Assets Survey valuation already capitalises its expected future payouts — separately counting the future spending flow that funds those same payouts double-counts the same value. Comparing today's 20-39-year-olds to today's 80+-year-olds is not comparing the same people across time; it compares different birth cohorts at different life stages in a single moment, and today's net contributors become tomorrow's net recipients. The OBR states explicitly that its own 50-year scenarios hold policy and behaviour fixed — useful for illustrating pressure, not a forecast of what will actually happen. And what counts as a "tax" versus a "transfer" is, in Kotlikoff's own term, a labelling problem: an arbitrary fiscal convention, not an economic fact, and different labelling conventions produce different headline burden figures from identical underlying cash flows.

The moment a document picks one debt-attribution method, one discount rate, or one framing of pay-as-you-go systems — "burden" versus "exchange" — without disclosing the alternatives, it has moved from measurement into contested territory dressed as neutral analysis. That is the single largest reputational risk in building this note out further. The underlying data in Section 1 is solid and largely undisputed; the combined "impact" narrative is where a hostile reader gets real traction, and this note is deliberately not attempting that step yet.

What This Connects To

The debt-carry attribution methodology above directly feeds the tax-receipts asymmetry work drafted separately for the Economy pillar — whichever method this note eventually adopts, that work should reference it for consistency, rather than the two developing incompatible conventions independently. Any future triple-lock reform proposal in the Welfare pillar needs to engage with the pay-as-you-go steel-man in Section 3 directly, not merely assert unsustainability. The Sure Start and per-pupil funding decline documented in Section 1.2 overlaps with the Education pillar's own numbers and should be checked for consistency rather than sourced twice, independently. And the property-wealth component of Section 1.3 is substantially a housing-supply and planning-policy story, not a purely demographic one — worth flagging for whenever a dedicated Housing pillar revisits generational wealth.

Sourcing note: the data in Sections 1.1 to 1.4 is solid, drawn from the OBR's July 2026 Fiscal Risks and Sustainability Report, the Intergenerational Foundation's Age Bias (2021), and ONS Wealth and Assets Survey Round 8 — subject to the specific caveats noted in the text, particularly the WAS accreditation suspension and the lack of a clean bridge between the 1999-2019 and 2026 spending series. Section 2's international comparison is a placeholder, not yet built out. The debt-attribution method in Section 4 is deliberately left undecided rather than silently chosen; the McCarthy/Sefton/Lee/Sambt private-transfer offset and the Resolution Foundation's Intergenerational Audit are flagged as the next sources to pull in full before this note takes a firmer position. If this ever escalates from background material to an actual policy proposal, it will need the full five-stage falsification framework — predictions, magnitude, time horizon, counterfactual, and falsification conditions — that this project's numbered pillars require.

For public discussion. Not affiliated with any political party. | generationalreset.org

The Generational Reset | In Discussion: The Generational Balance Sheet | For public discussion. Not affiliated with any political party. | generationalreset.org