Who Cares

Essay · society · September 2026

The number of Portuguese aged 80 and over will double in the next thirty-five years. They live long, but with fewer years of good health than almost anyone else in Europe. Most of their care is given by families, mostly by women, and by care homes that families largely pay for; public spending is under a third of the European average. The daughters who did the caring are now at work.

Reading time ≈ 10 min Need → family → homes → cost

The question

This series has counted time: the hours of paid work, and in The Unpaid Ledger the hours of unpaid work that the national accounts leave out. Care for the old is the largest of the unpaid tasks still to come. This essay asks who cares for old people in Portugal, at what cost in time and money, borne by whom, and how other countries have organised it. It sets out the case for families as the backbone of care as fairly as the case for more public care.

0.76m → 1.6mPortuguese aged 80 and over, 2025 and projected for 2060

7.8 of 22.7years left at 65 for a Portuguese woman that are free of limitation, 2023

about halfof spending on long-term care paid by households, against 5–25% in the other countries compared

2,807hospital beds held by patients discharged but with nowhere to go, March 2026

The need

Long lives, short health

The number of Portuguese aged 80 and over more than doubled between 2000 and 2025, from 340,000 to 760,000, and in Eurostat’s latest projection it doubles again by 2060, to about 1.6 million (Figure 1). The number aged 85 and over, who need care most, rises from under 400,000 to about a million by 2070. INE expects the number of people aged 65 and over to peak early in the 2050s; the very old keep growing after that.[1][2]

Fig. 1 — Portugal: people aged 80 and over and 85 and over, observed 2000–2025 and projected to 2070, millions. Data: Eurostat (demo_pjan; proj_25np, EUROPOP2025 baseline).

The Portuguese live long but not well. A Portuguese woman aged 65 in 2023 could expect to live another 22.7 years, more than the EU average of 21.8, but only 7.8 of them without a limitation in daily activities, the fewest among the countries compared and almost two years fewer than the EU average (Figure 2). That leaves about fifteen years with a limitation, against twelve in the EU and eight in Sweden.[3] The Long Retirement looked at what this means for the pension age.

Fig. 2 — Life expectancy at 65 for women, split into healthy life years (without activity limitation) and years with limitation, 2023. Data: Eurostat (hlth_hlye). The 2024 Portuguese figures show a jump of 1.6 healthy years in one year, probably a change in the survey, and are not used.

In the European health survey of 2019, a third of Portuguese aged 65 and over, and 42 per cent of the women, had a severe difficulty with personal care, such as washing or dressing, or with household tasks, such as shopping or cooking. In the EU it was 27 per cent.[4] Applied to today’s population, that is more than 800,000 people.

Who helps

Families first, then the care home

Portugal’s model is sometimes described as having no care. It is better described as family care plus care homes, with little in between (Figure 3). Of the Portuguese aged 65 and over with a severe difficulty in personal care, 29 per cent used professional help at home in 2019, against 50 per cent in the EU. Yet fewer said they lacked help: 39 per cent of those with any severe difficulty, against 47 per cent in the EU and 54 per cent in Sweden. Families fill the gap. Fewer old people live alone: 28 per cent of those aged 80 and over, against 40 per cent in the EU.[4][5]

Fig. 3 — Need for care and who provides it, Portugal and the EU. “65+ with severe difficulty”: severe difficulty with personal care or household tasks. *“Use home care”: professional home care, among those with severe difficulty in personal care. “Lack help”: among those with any severe difficulty. Informal carers: people aged 15 and over giving care at least weekly. Data: Eurostat, European Health Interview Survey 2019 (hlth_ehis_tadle, tadlh, am7sa, ic1e); Census 2021 (cens_21he_r2).

Fewer Portuguese give informal care than other Europeans, 12 per cent of those aged 15 and over each week against 17 per cent, about 1.07 million people, 65 per cent of them women. But their care is heavier. In the national time-use survey of 2015, women who lived with a dependent adult spent about three to three and a half hours a day caring for them, and the survey found that “in almost all cases, no support services of any kind, formal or informal, are used”.[6][7]

At the other end, more very old Portuguese live in institutions than in Spain or Italy: 11.7 per cent of those aged 80 and over in 2021, against 3.9 and 2.3 per cent, and close to France and the Netherlands.[5] Care homes, known in Portugal as lares, had about 107,500 places in 2024, 93 per cent of them occupied; most of the organisations that provide social care are non-profit, and many places are funded under agreements with Social Security.[8] Waiting is long: a consumer survey found an average wait of five months.[9] And some of the gap is filled illegally: in 2023 Social Security inspections closed 150 establishments, all private and 97 per cent unlicensed, most of them homes for old people.[10]

International statistics miss most of this. The figures Portugal reports to the OECD, Eurostat and the European Commission on care beds, recipients and workers appear to cover only the health-side care network, about 10,000 beds, and not the lares. Comparisons that show Portugal with almost no residential care are wrong.[11]

Who pays

Households pay half

Portugal spends about 1.1 per cent of GDP on long-term care, and households pay about half of it themselves, roughly €1.6 billion in 2024, more than the government. In the other countries compared, households pay between 5 and 25 per cent (Figure 4). Public spending, at about 0.5 per cent of GDP, is under a third of the EU average of 1.7 per cent.[12][13]

Fig. 4 — Spending on long-term care (health and social), per cent of GDP, split into public and compulsory insurance spending and payments by households, latest year 2023–2025. The household share is of the two together. *Germany and Italy report only health-related long-term care. Data: OECD, System of Health Accounts.

The reference monthly cost of a place in a care home under the agreements with Social Security was €1,400 in 2024; the average old-age pension in the general scheme was €674 a month in 2025, paid fourteen times a year. The State pays the home a fixed amount per resident, €730 a month in 2026, and the resident and the family pay the rest according to their income.[14][15] Where care is missing, the hospitals pay. In March 2026, 2,807 hospital beds, 13.9 per cent of inpatients, were held by people who had been discharged but had no care place to go to, at an estimated cost of €351 million a year, 19 per cent more beds than a year before.[16]

The carers

The cost to carers

In 2019 Portugal created a legal status for informal carers, with a monthly allowance for the main carer. By May 2025 about 17,000 people had the status and 5,816 received the allowance, which averaged €415.50 a month; the estimated number of weekly carers is about a million.[17][6] The five days a year of carers’ leave required by EU law were written into the Labour Code in 2023, unpaid, and only for workers who hold the status of secondary carer, 6,878 people.[18]

Among Portuguese women aged 50 to 64 who cared for a sick or disabled relative in 2018, 47 per cent were outside the labour force, against 34 per cent of all women that age.[19] That association runs both ways: women who are not working are more available to care. But the international evidence finds real costs. In the United States, women carers who keep working lose three to ten hours of work a week and about 3 per cent of their wages; in Germany, caring harms women’s mental health in the short run; across Europe, the harm is larger for daughters caring for parents with heavy needs.[20] With a pension calculated on the whole working career, years out of work in one’s fifties also mean a lower pension later.

The shift

Fewer daughters at home

The supply of family care is shrinking. The employment rate of Portuguese women aged 60 to 64 rose from 33 per cent in 2015 to 57 per cent in 2025, above the EU’s 49 per cent, as the pension age rose (Figure 5).[21] The women who in earlier generations would have cared for their parents at home are now more often at work until their mid-sixties.

Fig. 5 — Employment rate of women aged 55–64 and 60–64, Portugal and the EU, 2005, 2015 and 2025, per cent. Data: Eurostat, Labour Force Survey (lfsa_ergan).

Preferences also matter, and on them the Portuguese have been distinct. Asked in 2007 what would be best for an elderly parent who could no longer live alone, 44 per cent of the Portuguese said living with one of their children, against 30 per cent in the EU and 4 per cent in Sweden; 10 per cent said a care home, the same as the EU average.[22] No newer Portuguese survey was found. Whether those preferences survive when the daughters work is an open question.

Elsewhere

How other countries do it

Germany created a compulsory long-term care insurance in 1995, now financed by a contribution of 3.6 per cent of earnings. It pays a fixed amount by grade of need, in cash or in services; most people are cared for at home, and most of those receive only the cash, used largely to support care by relatives. The insurance pays for 1.2 million family carers’ pension contributions. But it covers only part of the cost: in 2026 a resident in a care home still paid on average €3,245 a month in the first year, and a third of residents need social assistance.[23]

Japan followed in 2000 with an insurance that provides only services, not cash, financed half by premiums from everyone over 40 and half by taxes; 20 per cent of people aged 65 and over are certified as needing care, and the average premium of the over-65s has more than doubled since the start.[24] Sweden and Denmark provide care through municipalities, paid from local taxes and mostly free or capped for users, at a cost of 3.3 to 3.7 per cent of GDP; about a quarter of people aged 80 and over receive care at home.[12][25]

Spain is the closest comparison and a warning. Its Dependency Law of 2006 gave everyone a legal entitlement to care, but it was cut after 2012 and is slow: in 2025 the average application took 341 days, against a legal limit of six months, and 32,704 people died while on the waiting list. Almost a third of its benefits are cash allowances for family carers.[26]

The research on these designs points two ways. Cash for family carers keeps more care in the family: in Spain, the allowance raised informal care by 20 to 22 percentage points and reduced hospital use. But it also keeps carers out of work: in German data, cash benefits reduce carers’ paid work, while services slightly increase it.[27][28]

The cost

What more public care would cost

The European Commission’s projections put Portugal’s public spending on long-term care at 0.9 per cent of GDP in 2070 if nothing changes. If Portugal’s coverage converged on the European average, it would reach 9 per cent of GDP.[13] That scenario overstates the gap, because it is built on the figures that leave out the care homes. Care is also a sector where productivity barely rises, so its cost rises relative to everything else as wages grow; this is true whether the carer is a daughter or a paid worker.

Family care is not free either; its cost is simply not counted. As an illustration, if each of the million weekly carers gave ten hours a week, valued at the minimum wage, the total would be about €3.4 billion a year, 1.2 per cent of GDP, more than all current spending on long-term care; the true hours are not known.[29] Formal care does not remove that work. It turns unpaid hours, mostly of women in their fifties and sixties, into paid hours, mostly of other women: in Portugal’s charitable sector the lowest wage in the care agreement is the national minimum wage, and across the EU about one care worker in five was born abroad.[30][31]

The balance

Whose time?

The case for Portugal’s family model is real. Most Portuguese have said they prefer it, fewer old people report going without help than in richer countries, and it is cheap for the State. The case against it is also real. It rests on women’s unpaid time and their pensions, it is paid for by households out of pensions that do not cover a care home, it spills into hospital beds, and the women it relies on are now at work.

For the question this series set out to answer, care is where more time and an ageing population collide. Hours freed from paid work may be absorbed by caring for parents, and the missing institutions that The Three Goals described, a general right to reduce working hours and paid leave, matter most here. The options range from paying and protecting family carers, with pension credits as in Germany, to more home care, to a social insurance. All cost more public money than Portugal spends now; the alternative is that the cost stays where it is, in women’s time and families’ savings.

On method and tools

This piece was written collaboratively with Claude Opus 5.5 (Anthropic): human specification, editorial direction and critical review; machine data research, analysis and drafting. The figures and derived numbers are computed by scripts/care.py from Eurostat’s population, health survey, census and labour force data and the OECD’s health accounts. Portugal’s international returns on long-term care beds, recipients and workers appear to leave out the care homes run under agreements with Social Security, so they are not used; this is inferred from their size, not stated by the sources. The number of care-home places is read from a chart in the Carta Social report, which prints totals but not this figure; the reconstruction matches the printed totals to within 0.01 per cent. The number of informal carers applies survey rates to the population. The value of family care is an illustration with assumed hours, not an estimate. The wait for a care home and some of the figures on Germany’s contribution history come from secondary sources. Results are in docs/care-results.json and the figures in docs/care-figures.html; the downloaded sources, with the dataset code or page for every number, are kept with the script’s data.

The cover photograph is Monsaraz January 2020-1 by Alvesgaspar; CC BY-SA 4.0, via Wikimedia Commons, cropped.

Authored by: Luis Matos Ferreira — Physicist, Developer, Writer

Related essays on this blog

1. The Unpaid Ledger — paid work, unpaid work and leisure over a lifetime.
2. The Long Retirement — pensions, exit ages and healthy years.
3. The Sustainability Story — how Portugal’s pensions are financed.
4. The Arrivals — immigration and Portugal’s workforce.
5. The Three Goals — time, the planet and satisfaction together.
6. Work, Time and Money — the reading guide to the whole series.
7. What We Found — the conclusions of the whole series in ten points.

Sources

1. Eurostat, population on 1 January by age (demo_pjan) and EUROPOP2025 baseline projections (proj_25np), retrieved 30 September 2026; author’s sums of single ages.
2. INE, Projeções da População Residente 2025–2100, Destaque, 30 September 2025, pp. 1 and 6.
3. Eurostat, healthy life years and life expectancy at 65 (hlth_hlye), updated 17 July 2026.
4. Eurostat, European Health Interview Survey 2019: severe difficulty in personal care or household activities (hlth_ehis_tadle), lack of assistance (hlth_ehis_tadlh), use of home-care services (hlth_ehis_am7sa); author’s totals.
5. Eurostat, Census 2021 hub, population by household status and age (cens_21he_r2); author’s shares.
6. Eurostat, European Health Interview Survey 2019, persons providing informal care at least weekly (hlth_ehis_ic1e), and population (demo_pjanbroad); author’s estimate of the number of carers.
7. H. Perista et al., Os Usos do Tempo de Homens e de Mulheres em Portugal, CESIS and CITE, 2016, pp. 88–91.
8. DGCP-MTSSS, Carta Social: Rede de Serviços e Equipamentos, Relatório 2024, June 2026, pp. 13–16 and 46–51; care-home places reconstructed from Fig. 49.
9. DECO PROteste, “Lares de idosos: reformas não cobrem custos”, 27 March 2026 (survey of 694 relatives).
10. ISS, Relatório de Atividades 2023, pp. 111–112.
11. European Commission and Economic Policy Committee, 2024 Ageing Report, Institutional Paper 279, Tables II.1.127–130; OECD, Long-term care resources and utilisation; Eurostat (hlth_rs_bdltc); author’s comparison with source 8.
12. OECD, System of Health Accounts, long-term care (health HC.3 and social HCR.1) by financing scheme, retrieved 30 September 2026; Eurostat (hlth_sha11_hchf, nama_10_gdp); author’s shares.
13. European Commission and Economic Policy Committee, 2024 Ageing Report, Institutional Paper 279, April 2024, Tables I.3.4, I.3.7 and II.1.114 (PDF pp. 128, 131 and 239).
14. Government of Portugal and social-sector confederations, Compromisso de Cooperação para o Setor Social e Solidário, Biénio 2023–2024, December 2023, pp. 6–7; Governo de Portugal, “Governo aumenta comparticipações para o Setor Social e Solidário”, 15 April 2026.
15. GEP-MTSSS, Síntese de Informação Estatística da Segurança Social, May 2025, pp. 2 and 5.
16. APAH and EY, Barómetro de Internamentos Sociais, 10th edition, press release, 23 April 2026.
17. Lei 100/2019 (Estatuto do Cuidador Informal); ISS, Guia Prático: Estatuto do Cuidador Informal, v1.25, 30 June 2026, pp. 15 and 19–20; source 15, p. 2.
18. Directive (EU) 2019/1158, art. 6; Lei 13/2023, Labour Code arts. 101-A and 101-B, pp. 60 and 64.
19. Eurostat, Labour Force Survey 2018 ad hoc module on reconciliation of work and family life (lfso_18cresls); author’s shares.
20. C. H. Van Houtven, N. B. Coe and M. M. Skira, “The effect of informal care on work and wages”, Journal of Health Economics 32(1), 2013; H. Schmitz and M. Westphal, “Short- and medium-term effects of informal care provision on female caregivers’ health”, Journal of Health Economics 42, 2015; D. Heger, “The Mental Health of Children Providing Care to their Elderly Parent”, Health Economics 26(12), 2017; abstracts.
21. Eurostat, employment rates by sex and age (lfsa_ergan), 2005–2025.
22. European Commission, Special Eurobarometer 283, Health and long-term care in the European Union, December 2007, p. 67.
23. Bundesministerium für Gesundheit, Zahlen und Fakten zur Pflegeversicherung, July 2026, pp. 9, 17, 18 and 23; vdek, press release, 22 January 2026, pp. 1–2; Destatis, press release 478, 18 December 2024.
24. Ministry of Health, Labour and Welfare (Japan), long-term care insurance annual report summary FY2024, pp. 1–2, and “Structure and finances of long-term care insurance”, p. 1; N. Tamiya et al., The Lancet 378, 2011, abstract.
25. OECD, Long-term care recipients (at home, aged 80 and over), 2024–2025; Social Protection Committee and European Commission, 2021 Long-Term Care Report, vol. I, p. 24.
26. Asociación Estatal de Directoras y Gerentes de Servicios Sociales, XXVI Dictamen del Observatorio Estatal de la Dependencia, 2026, pp. 3–5; IMSERSO, SAAD monthly report, 31 July 2026, p. 11.
27. J. Costa-Font, S. Jiménez-Martín and C. Vilaplana-Prieto, Journal of Health Economics 84, 2022, 102639, and 58, 2018, pp. 43–66; abstracts.
28. J. Geyer and T. Korfhage, “Long-term Care Insurance and Carers’ Labor Supply: A Structural Model”, Health Economics 24(9), 2015, abstract.
29. Author’s illustration from source 6 and the 2026 minimum wage (Decreto-Lei 139/2025), with an assumed ten hours a week per carer.
30. Contrato coletivo CNIS–FNSTFPS, revision published in Boletim do Trabalho e Emprego 21, 8 June 2024, pp. 2–3.
31. Social Protection Committee and European Commission, 2021 Long-Term Care Report, vol. I, pp. 56–57.