Citoyen
  • Home
  • Countries
  • Categories
  • Comparison
  • Synthesis notes
  • Analyses
  • Reforms
  • Sources
  • About
  • Home
  • Countries
  • Search
  • Analyses
All synthesis notes
European Union·Pensions

Pensions take up one eighth of European GDP, older people are working longer and ageing continues; Brussels is pushing supplementary retirement saving while Member States adjust — or suspend — their reforms.

Citoyen·Data as of Nov 7, 2026·4 min read

Key points

Pension expenditure (% of GDP)12.6% GDP2024Employment rate, ages 60-6455%2025Relative median income of people aged 65+0.91ratio2025Share of people aged 65 and over22%2024
  • Pensions accounted for 12.6% of EU GDP in 2024, from 4.1% in Ireland to 15.9% in Italy (Eurostat).
  • The employment rate of people aged 60-64 gained four points in two years, to 55.0% in 2025.

Citoyen synthesis for the Pensions category in the European Union. Grounded in the bloc's data (Eurostat, OECD, European Commission, Council of the EU). ⚠️ Aggregate of 27 Member States with very different pension systems; pensions are a national competence, the Union only coordinates and frames supplementary saving. All values are the latest realized observation available — never a forecast. Data last updated: November 2026.

1. State of play — where pensions stand in the EU

One eighth of GDP. According to Eurostat (ESSPROS, all pensions, public and private), the Union devoted 12.6% of its GDP to pensions in 2024 (provisional), against 12.3% in 2023. The gap between Member States is considerable: 4.1% in Ireland (whose GDP is inflated by multinationals), 5.1% in Malta, but 15.0% in France, 15.4% in Austria and 15.9% in Italy. These gaps reflect demography, but also the choice between generous public pensions and the room left to occupational schemes.

Later exits. The average age of exit from the labour market reached 63.9 for men and 62.7 for women in 2024 (OECD), from 60.1 in Luxembourg to 66.9 in Estonia for men. The employment rate of people aged 55-64 rose to 66.4% in 2025 and that of 60-64-year-olds to 55.0%, against 50.9% in 2023 (Eurostat) — the effect of higher legal ages and the end of early-retirement schemes.

Living standards of older people close to those of the working-age population. The median income of people aged 65 and over was 0.91 times that of the under-65s in 2025 (Eurostat, EU-SILC). The ratio exceeds 1 in Italy and comes close to 1 in France (0.97), but falls to 0.83 in Germany.

Three pillars of unequal weight. The pay-as-you-go public pension dominates almost everywhere. Occupational schemes weigh heavily in a few countries: according to the OECD (SOCX, 2021), private benefits amounted to 3.1% of GDP in the Netherlands and 2.0% in Denmark and Sweden, against 0.2% in France. Individual saving remains marginal in most Member States.

A national competence, European coordination. Member States set ages, rates and funding. The Union monitors sustainability in the European Semester, publishes with Member States the Pension Adequacy Report and the Ageing Report, and regulates occupational pension funds (IORP II directive) and the pan-European personal pension product (PEPP).

Pensions

European Union — Pension expenditure (% of GDP)

Primary KPI: one of the key indicators tracked first for this country.Pension expenditure as % of GDP, Eurostat definition (ESSPROS): old-age, anticipated and partial old-age, survivors', disability and early-retirement pensions; public and private social-protection schemes. Comparable across European countries; a different scope from the COR's (French mandatory schemes). Outside Europe (United Kingdom, United States, Canada, Japan, Korea, Australia, Mexico): OECD SOCX, public old-age and survivors' pensions only — a narrower scope. The cross-country comparison uses the OECD for all.· Public and private spending, Eurostat definition· Not comparable with the COR projections for France (mandatory French schemes only, a different scope)
12.6% GDP
2024
Source: Eurostat — ESSPROS (spr_exp_pens)· Data 2024Source type : Official
Old age, survivors, disability, early retirement · Public and private spending, Eurostat definition · Upd. 10/2026 · Y-axis truncated
ExploreSource
Citoyen indicator — real data · EU · 2026-11-07
Pensions

European Union — Employment rate, ages 60-64

Share of people aged 60-64 in work (Labour Force Survey, Eurostat, annual average). It is the age group directly affected by a later retirement age.· Breaks in series flagged by Eurostat (2021: new LFS regulation)
55%
2025
Source: Eurostat — LFS (lfsa_ergan)· Data 2025Source type : Official
People aged 60-64 in work, % of the age group · Breaks in series flagged by Eurostat (2021: new LFS regulation) · Upd. 10/2026 · Y-axis truncated
ExploreSource
Citoyen indicator — real data · EU · 2026-11-07
Pensions

European Union — Relative median income of people aged 65+

Median equivalised disposable income of people aged 65+ divided by that of people under 65 (1 = parity). EU-SILC survey (Eurostat), survey year; incomes refer to the previous year.· Survey year: incomes of the previous year· Median income ratio — not the COR’s mean standard of living of retirees
0.91ratio
2025
Source: Eurostat — EU-SILC (ilc_pnp2)· Data 2025Source type : Official
Median income 65+ / under 65 (1 = parity) · Survey year: incomes of the previous year · Upd. 09/2026 · Y-axis truncated
ExploreSource
Citoyen indicator — real data · EU · 2026-11-07
Demography

European Union — Share of people aged 65 and over

People aged 65 and over as % of total population (World Bank, from the UN World Population Prospects). The main demographic driver of pension and health spending.
22%
2024
Source: World Bank· Data 2024· EU (World Bank aggregate)
Upd. 10/2026 · Y-axis truncated
ExploreSource
Citoyen indicator — real data · EU · 2026-11-07

2. Outlook — where pensions are heading

Ageing continues. The old-age dependency ratio (people aged 65 and over relative to those aged 15-64) reached 34.5% on 1 January 2025, against 33.4% two years earlier (Eurostat). Fertility was 1.34 children per woman in 2024. The Commission's 2024 Ageing Report projects public pension spending rising from 11.4% of GDP in 2022 to 12.1% in 2045, then 11.8% in 2070 — projections, on a scope (public pensions) different from ESSPROS.

The supplementary pensions package. Presented on 20 November 2025 as part of the Savings and Investments Union, it includes the review of the IORP II directive, that of the PEPP regulation and a recommendation inviting Member States to adopt automatic enrolment in supplementary schemes (with an opt-out), pension tracking tools and national dashboards. The Council adopted its position on both texts on 26 June 2026; negotiations with the European Parliament can only start once Parliament has adopted its own position.

An open debate. The European Economic and Social Committee supports the package while defending the central role of public pensions; the savers' association BETTER FINANCE welcomes the initiative but calls for more transparency on fees and returns; part of Nordic business opposes uniform rules for systems that are already mature.

National reforms on the move. In France, the 2026 social security financing act (promulgated on 30 December 2025) shifted the phase-in of the 2023 reform by one generation. Germany passed a "Rentenpaket" guaranteeing the level of pensions and introduced the "Aktivrente", which since January 2026 exempts from tax up to €2,000 a month of earnings beyond the statutory age.

Open questions. Three issues will shape the period: (1) funding public pensions with fewer workers per retiree; (2) lengthening careers without widening inequalities between occupations; (3) developing supplementary saving without weakening the first pillar.

3. International comparison — the EU among its peers

Compared with other large advanced economies, the Union combines an older population with rather early exits. ⚠️ For spending, non-EU countries publish no ESSPROS series: their OECD SOCX figures (public cash benefits) belong to another measure, given as context only — 7.3% of GDP in the United States (2023), 9.2% in Japan (2022).

Three lessons. (1) Exit age: earlier. At 63.9 for men, the Union is level with the United Kingdom (63.8) but clearly below the United States (67.3), Japan (66.9) and Korea (67.4).

(2) Demography: between the United States and Japan. With 22.4% of people aged 65 and over in 2025 (World Bank), the Union is older than the United States (18.4%) or the United Kingdom (19.7%), but far from Japan (30.0%).

(3) Fertility: low. At 1.34 children per woman in 2024, the Union stands below the United States (1.63) and the United Kingdom (1.55), but above Japan (1.15) and Korea (0.75).

Pensions

United States — Pension expenditure (% of GDP)

Primary KPI: one of the key indicators tracked first for this country.Pension expenditure as % of GDP, Eurostat definition (ESSPROS): old-age, anticipated and partial old-age, survivors', disability and early-retirement pensions; public and private social-protection schemes. Comparable across European countries; a different scope from the COR's (French mandatory schemes). Outside Europe (United Kingdom, United States, Canada, Japan, Korea, Australia, Mexico): OECD SOCX, public old-age and survivors' pensions only — a narrower scope. The cross-country comparison uses the OECD for all.· OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions)· Not comparable with the COR projections for France (mandatory French schemes only, a different scope)
7.3% GDP
2023
Source: BIS — Social Expenditure Database (SOCX)· Data 2023Source type : Official
Public old-age and survivors’ pensions · OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions) · Upd. 10/2025 · Y-axis truncated
ExploreSource
Pensions

Japan — Pension expenditure (% of GDP)

Primary KPI: one of the key indicators tracked first for this country.Pension expenditure as % of GDP, Eurostat definition (ESSPROS): old-age, anticipated and partial old-age, survivors', disability and early-retirement pensions; public and private social-protection schemes. Comparable across European countries; a different scope from the COR's (French mandatory schemes). Outside Europe (United Kingdom, United States, Canada, Japan, Korea, Australia, Mexico): OECD SOCX, public old-age and survivors' pensions only — a narrower scope. The cross-country comparison uses the OECD for all.· OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions)· Not comparable with the COR projections for France (mandatory French schemes only, a different scope)
9.2% GDP
2022
Source: BIS — Social Expenditure Database (SOCX)· Data 2022Source type : Official
Public old-age and survivors’ pensions · OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions) · Upd. 10/2025 · Y-axis truncated
ExploreSource
Pensions

South Korea — Pension expenditure (% of GDP)

Primary KPI: one of the key indicators tracked first for this country.Pension expenditure as % of GDP, Eurostat definition (ESSPROS): old-age, anticipated and partial old-age, survivors', disability and early-retirement pensions; public and private social-protection schemes. Comparable across European countries; a different scope from the COR's (French mandatory schemes). Outside Europe (United Kingdom, United States, Canada, Japan, Korea, Australia, Mexico): OECD SOCX, public old-age and survivors' pensions only — a narrower scope. The cross-country comparison uses the OECD for all.· OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions)· Not comparable with the COR projections for France (mandatory French schemes only, a different scope)
3.8% GDP
2022
Source: BIS — Social Expenditure Database (SOCX)· Data 2022Source type : Official
Public old-age and survivors’ pensions · OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions) · Upd. 10/2025 · Y-axis truncated
ExploreSource
Pensions

United Kingdom — Pension expenditure (% of GDP)

Primary KPI: one of the key indicators tracked first for this country.Pension expenditure as % of GDP, Eurostat definition (ESSPROS): old-age, anticipated and partial old-age, survivors', disability and early-retirement pensions; public and private social-protection schemes. Comparable across European countries; a different scope from the COR's (French mandatory schemes). Outside Europe (United Kingdom, United States, Canada, Japan, Korea, Australia, Mexico): OECD SOCX, public old-age and survivors' pensions only — a narrower scope. The cross-country comparison uses the OECD for all.· OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions)· Not comparable with the COR projections for France (mandatory French schemes only, a different scope)
7.1% GDP
2022
Source: BIS — Social Expenditure Database (SOCX)· Data 2022Source type : Official
Public old-age and survivors’ pensions · OECD definition, cash benefits: narrower than Eurostat’s (no disability or private pensions) · Upd. 10/2025 · Y-axis truncated
ExploreSource
Pensions

European Union — Pension expenditure (% of GDP)

Primary KPI: one of the key indicators tracked first for this country.Pension expenditure as % of GDP, Eurostat definition (ESSPROS): old-age, anticipated and partial old-age, survivors', disability and early-retirement pensions; public and private social-protection schemes. Comparable across European countries; a different scope from the COR's (French mandatory schemes). Outside Europe (United Kingdom, United States, Canada, Japan, Korea, Australia, Mexico): OECD SOCX, public old-age and survivors' pensions only — a narrower scope. The cross-country comparison uses the OECD for all.· Public and private spending, Eurostat definition· Not comparable with the COR projections for France (mandatory French schemes only, a different scope)
12.6% GDP
2024
Source: Eurostat — ESSPROS (spr_exp_pens)· Data 2024Source type : Official
Old age, survivors, disability, early retirement · Public and private spending, Eurostat definition · Upd. 10/2026 · Y-axis truncated
ExploreSource
International comparison — pension_spending_gdp · EU · 2026-11-07

International comparison — pensions

CountryExit age M / W (2024)Aged 65+ (2025)Fertility (2024)
United States67.3 / 66.718.4%1.63
Japan66.9 / 67.430.0%1.15
South Korea67.4 / 69.620.3%0.75
United Kingdom63.8 / 63.119.7%1.55
European Union63.9 / 62.722.4%1.34

Sources: average age of labour-market exit = OECD, Pensions at a Glance (2024); aged 65+ and fertility = World Bank (WDI). Pension spending is not shown in this table: Eurostat ESSPROS (EU) and OECD SOCX (other countries) are not comparable.

Data mobilized (data-journalism base)

Show the table
DataValueSource
Pension spending (ESSPROS)12.6% of GDP (2024, provisional)Eurostat — spr_exp_pens (Citoyen card)
Average age of labour-market exitMen 63.9 · women 62.7 (2024)OECD — Pensions at a Glance (Citoyen card)
Employment rate, aged 60-6455.0% (2025)Eurostat — lfsa_ergan (Citoyen card)
Employment rate, aged 55-6466.4% (2025)Eurostat — lfsi_emp_a
Relative median income, aged 65+0.91 (2025)Eurostat — ilc_pnp2 (Citoyen card)
Old-age dependency ratio34.5% (2025)Eurostat — demo_pjanind
Projected public pension spending12.1% of GDP in 2045 (projection)European Commission — 2024 Ageing Report

Sources (national analyses and references)

Eurostat (ESSPROS — spr_exp_pens; EU-SILC — ilc_pnp2; labour force survey — lfsa_ergan, lfsi_emp_a; demo_pjanind) · OECD (SOCX; Pensions at a Glance) · World Bank (WDI) · European Commission (2024 Ageing Report; supplementary pensions package of 20 November 2025: IORP II and PEPP review, recommendation on automatic enrolment) · Council of the EU (positions of 26 June 2026) · EESC · BETTER FINANCE · Légifrance (2026 social security financing act).

Methodological note — the synthesis distinguishes sourced facts from assessments, stays neutral, dates every figure, and does not extrapolate beyond its sources. ⚠️ The aggregate masks strong disparities between Member States. Eurostat ESSPROS spending, OECD SOCX spending and the Ageing Report projections have different scopes and are never compared as a single measure. All values are the latest realized observation available; projections are presented as such. Same safeguards as the rest of the observatory.