Beyond retirement: a closer look at the very old

The European Union should start to prepare for those over 85 making up 10 percent of the population in 2100

Publishing date
30 April 2024

Population ageing will have significant structural impacts on European economies. The gradual retirement of the large baby-boomer generation – those born between 1946 and 1964 – will widely impact European Union policy. The green and digital transition will not succeed without taking into account an ageing and changing workforce. While increasing life expectancy is welcome as a victory of past policy efforts, coupled with other factors, it does present a series of challenges.

The most obvious area that will be affected by an ageing population is pension policy. The old-age dependency ratio, measures the number of people aged 65 and more relative to the working-age population (aged 20-64). This ratio is increasing in the EU, projected to increase from 36 percent in 2022 to 55 percent in 2050 and 65 percent in 2100 1 Eurostat data. . While there were approximately 2.7 workers per elderly person in the EU in 2022, the number will fall to approximately 1.5 in 2100.

It makes sense to focus on those aged 65+ in the context of pension policy since retirement ages in most EU countries are around that threshold. However, it is important to look beyond retirement and show that the growth of the elderly population in the EU will be driven by those aged 85+ 2 This is not only true for the EU, but also for a wider set of developed countries (Gruber and McGarry, 2023). – the very old. Figure 1 shows the projected age composition of the EU population. The data clearly shows that 85+ is the fastest-growing age group in the EU. In 2022, 2.9 percent of people in the EU were 85 or older. This share is projected to more than double by 2050, and almost triple by 2100, reaching almost 10 percent of the total population. 

The very old are driving the increase in the old-age dependency ratio

Three factors explain the rising old-age dependency ratio. First, life expectancy is increasing. Second, the working-age population will shrink significantly between 2022 and 2070, as fertility rates will remain below the natural replacement rate. Third, projected net migration flows will not offset the ageing trend in the population (European Commission, 2021c). 

Figure 2 shows clearly that the projected increase in the old-age dependency ratio over the long-term will be driven mostly by the very old. The ratio of people aged 85 and older, relative to the working-age population, is projected to rise by almost 15 percentage points (from 5 percent to 19.7 percent) between 2022 and 2100. This projected increase significantly outpaces those of 65-74 and 75-84 age brackets. Figure 2 thus illustrates a significant shift from younger to older retirees. By the year 2100, each age category over 65 (65-74, 75-84 and 85+) will represent approximately one-third of the elderly, contrasting the dominance of those aged 65-74 today.


Differences between EU countries

In 2022, Finland had the highest old-age dependency ratio (41 percent), while Luxembourg had the lowest (23 percent). Italy, Portugal and Greece had some of the highest overall ratios, as well as the highest shares of people aged 85+ relative to working-age people, reflecting an older population profile in Southern Europe (Figure 3, Panel A). 

There is also substantial variation in how the dependency ratio and its composition are projected to evolve (Figure 3, Panel B). While some countries with relatively low dependency ratios will see those ratios increase the most, such as Ireland and Luxembourg, this pattern will not be seen in all countries. The dependency ratio is projected to increase by at least 25 percentage points between 2022 and 2100 in all countries except Portugal, Germany, Sweden and Czechia. Its increase is projected to be highest in Malta (43 percentage points). While Portugal and Italy had similar aggregate ratios in 2022, its growth in Italy is projected to be much higher.

Despite differences, there will be a common trend in EU countries: the 85+ population is projected to increase most relative to the working-age population in all EU countries between 2022 and 2100. While in 2022 over half of people over 64 were between 65-74 years old in all but three EU countries, this is projected to be the case in only four EU countries in 2050, and no member state by 2100.

Policy implications for social-security – and other – systems

The growing share of the very old will have consequences for social-security systems beyond pensions. There is a clear correlation between age and the likelihood of facing physical limitations, particularly for women (Figure 4). Therefore, the surge in the 85+ population will lead to a surge in people needing long-term care (LTC) 3 LTC is commonly defined as services needed by individuals experiencing diminished physical or cognitive abilities, leading to increased dependency on assistance for day-to-day activities including eating, washing, dressing and household chores such as shopping or laundry (OECD, 2023).  (Belmonte et al, 2023). While there has been a steady decrease in the share of people in the EU aged 65-74 and 75-84 with a severe activity limitation, the downward trend has stagnated for those aged 85+ since the onset of the COVID-19 pandemic in 2020. Moreover, demand for LTC already exceeds supply in many EU countries. This supply shortage is likely to worsen given the projected relative decline in working-age populations (Figure 1). Shortage of formal care workers is already evident in many EU countries, with, for example, nursing professionals ranking highest among the occupations experiencing labour shortages in the EU in 2020 (European Commission, 2020; 2021a).

Currently, a large share of people in need of care rely on informal care and care from relatives – this may represent up to 80 percent of total care time in the EU (European Commission, 2021b). In the future, such extensive reliance on informal care is likely to be unrealistic because of three trends. First, childlessness is becoming more prevalent, along with generally declining fertility rates. This means that there will be fewer potential informal carers in the future, and more elderly people will not have offspring who can provide care when needed. Second, informal care is predominantly provided by women. More women participate in the labour force than in the past, reducing the scope to provide care. Third, children live further away from their parents than in the past, making it more difficult to provide care. Hence, demand for formal care will likely increase in the future. 

Women leaving their jobs to care for elderly relatives is recognised as a significant factor behind the stagnation in female labour-force participation in the United States since the 2000s (US Department of Labor, 2023). But higher demand for long-term care does not necessarily result in less labour-market participation by women. In Japan, for instance, despite a significant proportion of the population being very old already in 2022 (5.4 percent compared to 2.9 percent in the EU), there is a considerably higher female labour supply than in the US or the EU (Figure 5). EU policymakers must plan for higher demand for LTC services in ways that do not disadvantage women.

More formal care will require an increase in public spending on age-related policies, notably pensions and long-term care. The Commission’s 2024 Ageing Report (European Commission, 2024) shows that age-related expenditures (including pensions, healthcare, long-term care and education expenditures) are a significant part of government spending. Aggregate EU age-related expenditure stood at 24.4 percent of GDP in 2022 and is projected to rise by 1.2 percentage points of GDP in the EU by 2070 4 In the Commission’s Ageing Report (European Commission, 2024), pensions expenditures are projected by member states directly taking into account the characteristics of the country-specific pension system. The projections for long-term care, healthcare and education expenditures are constructed by the Commission using a common projection model, taking country-specific age and cost profiles as a starting point. This risks underestimating some costs, notably for healthcare. .  

In many European countries, the primary pension system is pay-as-you-go, meaning that the pension contributions of current workers are used to finance the pensions of current retirees. The shifting demographics imply that a relatively smaller workforce will be tasked with supporting a larger population of retirees. The European Commission (2024) forecasts an increase in the EU average public pension spending from 11.4 percent of GDP in 2022 to 12.1 percent by 2045, followed by a decrease to 11.8 percent of GDP in 2070 (weighted by GDP) 5 After 2045, the Commission projects ageing trends to be counteracted by three factors. First, lower pension spending per pensioner. Second, the number of new pensioners increasing slower than the 65+ year old. Third, labour market trends. .

Healthcare and long-term care spending are projected to increase in almost all EU countries. Healthcare expenditure in the EU is set to increase from 6.9 percent of GDP in 2022 to 7.3 percent by 2070. Since 2017, public LTC expenditure has risen and is projected to increase by 0.8 percentage points of GDP, from 1.7 percent in 2022 to 2.6 percent by 2070. 

One direct implication of population ageing is that the average age of those in the labour force will increase. However, the labour market situation of older workers is different from younger workers. For instance, older workers are less likely to lose their jobs, but if they do, it takes them longer, on average, to find new employment (Martin, 2018). Additionally, older workers tend to participate less in on-the-job training compared to younger workers, making it less likely they will switch careers or industries (Fritzsche and Marcus, 2013). At the same time, with an ageing workforce, there is a higher risk that skills might become outdated relative to continuing innovations. Therefore, the EU labour market will increasingly rely on the up-skilling and/or re-skilling of the current workforce, rather than on people transitioning from the initial education system (Desjardins and Warnke, 2012). Moreover, recent pension reforms will lead to gradual increases in the statutory retirement age in many countries. But the average effective retirement age persistently remains below the statutory retirement age in most member states (European Commission, 2021c), indicating that raising the legal retirement age will not be enough to keep people in the labour market.

Policy recommendations

In readiness for a situation in which the very old (85+) make up almost 10 percent of the EU population in 2100, policymakers can look at three main areas.

First, policymakers must aim at better health in old age by promoting healthy ageing. This will be the most effective way to limit the demand for LTC. European funds should support research to cure cognitive diseases causing dementia, which is projected to be a significant driver of future LTC needs. 

Second, policymakers must prepare LTC systems for a significant increase in demand. This means investment in care facilities and equipment, and taking measures to increase the number of care workers, for example by raising wages or improving working conditions. Governments should also facilitate training in care occupations, for example by subsidising education and paying out stipends. Training programmes could target individuals in occupations currently at risk of becoming redundant and offer an alternative. 

Third, policymakers need to focus on women in the context of pensions and LTC. Care is predominantly provided by women, who are more likely to reduce working hours because they have care responsibilities. This not only results in lower income for women, but also lower social security contributions leading to lower pensions than men, on average. To support informal carers, governments should provide pension contribution credits for time spent away from work to care for others. Several countries have introduced such measures, including France, Belgium, Germany and Italy (European Commission, 2021d). The European Commission can play a role by identifying best-practices and coordinating policies. 


Belmonte, M., S. Grubanov-Boskovic, F. Natale, A. Conte, A. Belanger and P. Sabourin (2023) Demographic microsimulation of long-term care needs in the European Union, Publications Office of the European Union, Luxembourg

Desjardins, R. and A. Warnke (2012) ‘Ageing and Skills: A Review and Analysis of Skill Gain and Skill Loss Over the Lifespan and Over Time’, OECD Education Working Papers 72, Organisation for Economic Co-operation and Development, available at

European Commission (2020) Analysis of shortage and surplus occupations 2020, Publications Office of the European Union, available at

European Commission (2021a) Long-term care report : trends, challenges and opportunities in an ageing society, Volume II, Country profiles, Publications Office of the European Union, available at

European Commission (2021b) Study on exploring the incidence and costs of informal long-term care in the EU, Publications Office of the European Union, available at

European Commission (2021c) The 2021 Ageing Report. Economic and Budgetary Projections for the EU Member States (2019-2070), available at

European Commission (2021d) 2021 pension adequacy report: current and future income adequacy in old age in the EU, Volume 1, Publications Office of the European Union, available at

European Commission (2024) The 2024 Ageing Report. Economic and Budgetary Projections for the EU Member States (2022-2070), available at

Fritzsche, B. and J. Marcus (2013) ‘The senior discount: Biases against older career changers’, Journal of Applied Social Psychology 43(2), 350-362

Gruber, J. and K.M. McGarry (2023) ‘Long-term care in the United States’, NBER Working Paper w3188, National Bureau of Economic Researchavailable at

Martin, J.P. (2018) ‘Live Longer, Work Longer: The Changing Nature of the Labour Market for Older Workers in OECD Countries’, IZA Discussion Papers 11510, Institute of Labor Economicsavailable at

OECD (2023) ‘Ageing and Long-term Care’, chapter 10 in Health at a Glance 2023: OECD Indicators, Organisation for Economic Co-operation and Development, available at

US Department of Labor (2023) ‘Older women and unpaid caregiving in the U.S.’, Issue Brief, November, available at

About the authors

  • David Pinkus

    David Pinkus joined Bruegel as an Affiliate fellow in May 2023. He is an applied economist with a strong interest in social welfare policies, as well as the intersection of financial markets and the real economy.

    His work focuses on the challenges social security systems face due to an ageing population. He is also interested in the wider economic effects of funded pension systems and institutional investors. From 2014 to 2016, he worked as a consultant at the OECD’s Long-Term Investment Project, researching policies to enable institutional investors to finance infrastructure under a G20 mandate.

    David holds a PhD in Economics from Copenhagen Business School and is affiliated with the university’s Pension Research Centre (PeRCent). David also holds an M.Sc. in Economics from Bocconi University in Milan and a B.Sc. in Economics from Ludwig-Maximilians-University in Munich.

    David is fluent in German, French and English.

  • Nina Ruer

    Nina works at Bruegel as a research assistant. She holds a Master's of Research (MRes) in Analysis and Policy in Economics from the Paris School of Economics (PSE). Her master's thesis, titled "The Gender Pay Gap in Student Employment in France," was a comprehensive study that delved into income disparities among university students in France. Prior to that, she earned a B.Sc. in Economics with a final year in "Magistère" from Université Paris 1 Panthéon-Sorbonne.

    Prior to joining Bruegel, she was a research assistant on a series of projects funded by PSE where she gained hands-on experience in finding and cleaning replication datasets for Randomized Control Trials (RCTs). She also developed multiple surrogate index functions for long-term forecasting. Another set of projects focused on collecting subjective forecasts, where she assessed the calibration of various groups for forecast accuracy.

    Nina is a dual Dutch and French citizen and is a French native speaker, fluent in Dutch and English.

Related content


Labour market outlook dashboard

This dashboard offers a comprehensive overview of the evolution of key labour market outcomes across EU member states, from 2006 onwards

Giulia Gotti, Duygu Güner and Nina Ruer