Retirement and Loneliness: Most Adapt, but Those Pushed Out of Work Fare Worse
Published 2026-11-15
Leaving work does not, on average, leave people lonelier. Large studies that followed the same people for years in Australia, England and across Europe found no rise in loneliness after retirement, and in Europe a fall over the longer run, although one US analysis found somewhat more loneliness among retired people. How people retired mattered: in Australia and the United States, those who had not chosen to stop work were more likely to feel lonely than those who had.
For many people, work supplies a daily rhythm and a cast of familiar faces: colleagues, customers, the conversation over a cup of tea. Retirement takes some of that away and hands back time, which many people fill with family, friends, volunteering, learning and clubs. This study sets out what research from several countries shows about how social life changes when people stop work, whether loneliness rises or falls, and who is most likely to struggle.
Work is a source of everyday contact, and retirement mostly trims the outer circle of acquaintances
A systematic review of programmes for people around retirement put the starting point plainly: “After retiring from work, people may lose roles that provide purpose and social contacts”[3]. The question is how much goes, and which part.
The clearest measurement comes from the Finnish Retirement and Aging study, which surveyed 2,319 people (84% of them women, average age 63.2) every year before and after they retired. Before retirement they named an average of 21.6 people in their social network: 5.6 in the inner circle of closest relationships, 6.9 in the middle circle and 9.1 in the outer circle of more peripheral ties. During the retirement transition the outer circle shrank by an average of 0.67 people, and then stopped shrinking. The inner and middle circles did not fall at all, and the pattern was the same for women and men and across occupations and marital status[4]. The authors conclude that this loss is more likely linked to retirement than to aging: the people who drop away tend to be acquaintances, not confidants.
Working lives also shape the networks people take into retirement. Among retired Europeans in the SHARE survey, those who had worked in more jobs had larger networks that more often included children and friends, though with less emotional closeness, and women who had spent longer in paid work had larger networks and more often counted friends among their confidants[5].
In the strongest long-running studies, retiring did not make people lonelier on average
Comparing retired people with workers can mislead, because early retirees often differ in health, wealth and job. Stronger studies follow the same people through the change, and some go further, using official retirement-age rules as a natural trigger, so that the timing is set by the rules rather than by a person’s circumstances. Three recent studies that follow the same people, two of them using retirement-age rules, point the same way.
- Australia. Using about two decades of the HILDA survey, which has followed roughly 14,000 people, and looking at those aged 55 to 75, with the official eligibility age for the state Age Pension as the trigger, a 2026 study found “no statistically significant effects of retirement on loneliness, a result consistent across various demographic groups and individual circumstances”. Loneliness was measured each year by how strongly people agreed, on a scale of 1 to 7, with the statement “I often feel very lonely”[6].
- England. In the English Longitudinal Study of Ageing, 766 of 3,758 people aged 50 and over retired between two survey waves in the period 2008 to 2017. Loneliness was measured with the UCLA 3-item scale, with a score of 6 or more counted as lonely, which applied to almost a fifth of the sample at the start. Retiring made no measurable difference to loneliness in the short term (adjusted odds ratio 0.96, 95% confidence interval 0.77 to 1.21) and was linked to lower odds of social isolation (adjusted odds ratio 0.76, 0.58 to 0.98), measured by less than monthly contact with friends and family and living alone without belonging to any clubs or organisations. Over the longer term there was no association with either[7].
- Europe. A 2024 analysis of the Survey of Health, Ageing and Retirement in Europe (SHARE), covering 19,699 people in 13 European countries and Israel in waves from 2013 to early 2020 and using differences in national retirement rules, found no clear effect in the short run and a fall in loneliness in the long run (four to six years after retiring), of about a quarter of a standard deviation on the 3-item UCLA scale. The fall came mainly from people being less likely to feel isolated and to lack companionship[8].
Each of these designs has limits. A study that uses official retirement ages as its trigger describes people who retire around those ages, not those who stop work much earlier through illness or redundancy. And all three surveys sample people living in private households. But taken together, they do not support the idea that retirement in itself is a common route into loneliness.
The picture differs by country, and by whether people chose to retire
A 2024 study compared adults over 50 in three national surveys that follow the same people over time: HILDA in Australia (2008–2019), the Health and Retirement Study in the United States (2008–2018) and the China Health and Retirement Longitudinal Study (2011–2015). Each used a single loneliness question, scored as lonely or not, worded differently by country. In Australia and the US, retired people had a somewhat higher predicted prevalence of loneliness than people who had not retired (a group that included the partly retired), though the gap was statistically significant only in the US (19.3% against 15.7%). In China the pattern reversed: 10.0% of retired people against 17.1% of those not retired[1]. The authors conclude that the effect of retirement should be considered “within a cultural context”.
The sharpest difference in the same study was among retired people themselves. In both Australia and the US, those who said they had retired voluntarily had the lowest loneliness of any group, and those who had retired involuntarily the highest: 9.6% against 16.9% in Australia, and 10.5% against 18.1% in the US[1]. These figures come from a separate model and cannot be set beside the chart above. In that model, people who had not retired stood at about 16% in Australia and 14% in the US, so involuntary retirees were close to that level in Australia and above it in the US, while voluntary retirees were well below it in both. Involuntary retirement can follow ill health, caring or job loss, each of which can bring loneliness of its own, so this is an association, not proof of cause. Social engagement did not change the link between retirement and loneliness in this study.
Evidence from Taiwan also points to circumstances rather than retirement as such. Among 2,370 retirees aged 50 to 74 followed between 2015/16 and 2018/19, and asked whether they had felt lonely in the past two weeks (any degree counted as lonely), about two-thirds remained not lonely at both surveys and fewer than 10% were lonely at both. Those who had experienced conflict between work and family before retiring, or stressful life events after it, had higher odds of remaining lonely, while an increase in social engagement, especially social contact, was linked with lower odds of becoming or remaining lonely[2].
Women, men and couples move through retirement differently
Averages can hide different experiences. In the European SHARE analysis, retirement was linked to a short-term rise in loneliness for women (a result at the margin of statistical significance), and the authors traced this to women lacking companionship when their partner was still working. A couple analysis showed the short-term rise and later fall in lack of companionship most clearly for women whose partner had not yet retired. Over the longer run, overall loneliness fell for women; for men the change was not statistically significant, though both women and men became less likely to feel isolated[8]. The Australian study found no statistically significant effect in any demographic group it examined[6].
Education mattered too. In SHARE, the longer-term fall in loneliness was particularly clear among people with more education, and the authors link the benefit to people taking up more activities after retirement[8]. Education often stands in for wider advantages such as income and health, so this says as much about access as about individuals.
Men’s later-life friendships and health are covered in our study of older men, and what evaluations of community workshops for men have found is set out in our study of men’s sheds.
People who were already lonely, and those whose networks shrink, may find the change hardest
One US study using two waves of the Health and Retirement Study noted that retirement “requires more self-directed efforts in order to stay socially engaged”. It found that loneliness among older workers was more strongly linked to later depressive symptoms in those who retired between the waves than in those who stayed in work. The authors stress that many manage easily, but “lonely older workers are at increased risk for maladjustment and the experience of depressive symptoms following retirement”[9].
What happens to a person’s wider network matters as well. In SHARE data from 2011 and 2015 covering 9,249 people, those whose close network grew in size, contact or closeness as they retired had a more favourable change in quality of life, the outcome that study measured[10]. In the Taiwanese study, increased social contact after retirement went with lower odds of becoming or staying lonely[2]. These are associations, not proof of cause.
New roles are one way people rebuild their week. The 2013 review of programmes to give people meaningful social roles in retirement screened 9,062 publications and found eleven evaluations of acceptable quality, covering seven programmes, only two of them outside North America and only three studies with random allocation. Programmes that gave people an explicit role and a supportive group were “somewhat effective” in improving outcomes such as life satisfaction, social support and activity, and physical health; loneliness itself was not among the outcomes the review listed[3]. Volunteering is one of the commonest of these roles, and the evidence on it, including who gets to volunteer, is in our study of volunteering in later life. Learning is another; our study of lifelong learning in later life sets out what research on u3a groups and other courses has found.
What the evidence asks of us
The research does not show retirement to be a common cause of loneliness, and people leaving work are not a problem to be managed. It does point to ways of helping the minority for whom the change is hard.
- Pay attention to people who leave work involuntarily. In Australia and the US, people who had not chosen to retire were more likely to feel lonely than those who had[1]. Employers, health services and welfare agencies that see people at redundancy, ill-health retirement or giving up work to care can make sure they hear about local groups.
- Support social engagement throughout later life, not only at the moment of retirement. The Australian study concluded that efforts to reduce loneliness “might therefore consider factors beyond retirement transition, focusing instead on promoting sustained social engagement throughout later life”[6]. That means steady support for the clubs, groups, learning and volunteering that older people keep up long after retiring.
- Make roles easy to take up for everyone. The benefits of activity after retirement in Europe were clearest among people with more education[8]. Funders and organisers can lower barriers of cost, transport and confidence.
- Build the evidence. Few programmes for people around retirement have been tested with control groups, and most were in North America[3]. The WHO Commission on Social Connection’s 2025 roadmap calls for action on policy, research, interventions, better measurement, including a global Social Connection Index, and public engagement[11]; retirement studies that measure loneliness directly, in more countries, belong in it.
For the wider picture of loneliness in later life, see the Loneliness in Later Life research library.
Language: this piece says "older people" and treats loneliness as a common human experience that anyone can have, not a personal failing or an illness. It keeps apart three things that are often mixed up: loneliness (how people feel about their relationships), social isolation (how few social contacts they have) and living alone (a household arrangement). Where a study uses another term, such as "seniors" or "older adults", we keep it when quoting that study.
Why we're publishing this
Retirement is often assumed to bring loneliness. The groups that many people join after leaving work, from u3a groups and men’s sheds to volunteering projects and lunch clubs, are mostly run by older people themselves, and they deserve evidence that is accurate rather than gloomy. We publish this so that organisers, funders and campaigners can see what long-running studies found: most people adapt well, and loneliness was more common among people who had no choice about leaving work than among those who chose to retire.
How to read this data
Measures. The studies here measured loneliness in different ways. The English and European studies used the 3-item UCLA scale, which asks how often people feel they lack companionship, feel left out and feel isolated from others; the English study counted a score of 6 or more as lonely, while the European study analysed the score itself. The Australian studies used agreement with one statement, “I often feel very lonely”. The US and Chinese figures used the loneliness question in a depression scale about the past week. The Taiwanese study asked whether people had felt lonely in the past two weeks. Figures from different measures are not directly comparable. Social isolation (few contacts) and network size were measured separately, and are reported here as such.
Designs. All the studies follow the same people over time. Only the Australian (2026) and European studies used official retirement ages to get closer to cause and effect; the comparisons by country, sex and type of retirement are associations. Predicted prevalence figures are model estimates, not raw survey shares.
Coverage and gaps. The surveys sample people living in private households and cover mostly high- and middle-income countries in Europe, North America, Australia and East Asia; several of the studies deliberately stop before the COVID-19 pandemic. We found no retirement and loneliness studies from Africa, South Asia or Latin America in our search.
Use this data
Free to cite with attribution to Understanding Aging. u3a groups, men's sheds, senior centres, lunch clubs, befriending services, charities, campaigners and journalists are welcome to reuse these figures and charts. For the underlying figures as a spreadsheet, get in touch.
Important: This article summarises published research and statistics for general information. It is not medical or care advice and cannot be used to diagnose, treat or predict anything about an individual. If loneliness, low mood or isolation is affecting your health or someone else’s, a doctor or another health professional can help, and local services for older people can often put you in touch with groups and support nearby.
Sources
- Hagani N, Clare PJ, Luo M, Merom D, Smith BJ, Ding D. Effect of retirement on loneliness: a longitudinal comparative analysis across Australia, China and the USA. Journal of Epidemiology and Community Health, 2024. europepmc.org/article/MED/38834231
- Chiao C, Kuo PH, Li DC, Hu SC. The changes in feeling of loneliness after retirement among baby boomers and pre-boomers in Taiwan. SSM – Population Health, 2022. europepmc.org/article/MED/36281245
- Heaven B, Brown LJ, White M, Errington L, Mathers JC, Moffatt S. Supporting well-being in retirement through meaningful social roles: systematic review of intervention studies. The Milbank Quarterly, 2013. europepmc.org/article/MED/23758511
- Kauppi M, Virtanen M, Pentti J, Aalto V, Kivimäki M, Vahtera J, et al. Social network ties before and after retirement: a cohort study. European Journal of Ageing, 2021. europepmc.org/article/MED/34786012
- Cohn-Schwartz E, Naegele L. Employment over the life course and post-retirement social networks: a gendered perspective. International Psychogeriatrics, 2024. europepmc.org/article/MED/37462123
- Kunz JS, Trinh TA, Berk M, et al. The impact of retirement on loneliness: evidence from Australia. Scientific Reports, 2026. europepmc.org/article/MED/43019198
- Kenny L, Doherty N, Doherty AS, Lawlor B, O’Sullivan R, Leavey G. The relationship between retirement, social isolation and loneliness: a longitudinal analysis using the English Longitudinal Study of Ageing. BMC Public Health, 2025. europepmc.org/article/MED/41184890
- Guthmuller S, Heger D, Hollenbach J, Werbeck A. The impact of retirement on loneliness in Europe. Scientific Reports, 2024. europepmc.org/article/MED/39528537
- Segel-Karpas D, Ayalon L, Lachman ME. Loneliness and depressive symptoms: the moderating role of the transition into retirement. Aging & Mental Health, 2018. europepmc.org/article/MED/27624519
- Settels J, Böckerman P. Impact of retirement on quality of life: role of changes in social network involvement. Research on Aging, 2025. europepmc.org/article/MED/40073282
- World Health Organization. Social connection linked to improved health and reduced risk of early death (news release on the WHO Commission on Social Connection report). WHO, 2025. who.int/news/item/30-06-2025-social-connection-linked-to-improved-heath-and-reduced-risk-of-early-death