The active ageing policy supports several types of activities, including labour force participation, caregiving, social participation, and physical activity. The paper illustrates the prevalence of supported activities across individual characteristics and four supra-national European regions to assess how these activities are available for specific groups of older people. The analysis draws on wave 6 from the Survey of Health, Ageing and Retirement in Europe held in 2015. A set of figures describes the availability of activities sorted by gender, age, health status, and the level of education in 17 European countries divided into four regions, and thus, presents the unavailable descriptive data important for researchers and policymakers. The results most of all show that the majority of the 50+ population engages in vigorous physical activity, whilst labour force participation and caregiving concern about one-third of it, and other activities much less. The findings show the inadequacy of the active ageing as a uniform context-insensitive EU policy and detect its potential for raising inequalities in later life, whilst the theoretical implications are discussed.
- Active ageing policy
- older age
- European regions
- descriptive data
The active ageing approach has shaped European policies and normative expectations towards ageing in recent decades (Foster & Walker, 2015; Hasmanová Marhánková, 2017; Timonen, 2016). The central thesis of this approach is that activities in later life improve not only the image of older adults, but also their health and quality of life (Walker & Maltby, 2012; Zaidi & Howse, 2017). This assumption legitimises the support of certain types of activities in the older population, which spans from initiatives targeted to social clubs and educational courses adjusted to older people, towards an increase of the retirement age. Thereafter, the activities containing labour force participation, caregiving, social participation, and physical activities are monitored and evaluated on various European geographical levels (European Commission, 2013; Zaidi & UNECE, 2015). Still, the previous research doubt that the promoted activities are beneficial under majority of conditions (Di Gessa & Grundy, 2013; Kim & Moen, 2001; Lakomý, 2019; Neuberger & Haberkern, 2014). Moreover, strongly under-researched is the question how the activities are distributed across European regions and specific groups of their populations.
Employment, social participation, and independent living comprise three main pillars of active ageing in the EU (EU Council, 2012; European Commission, 2013; European Union, 2012; Zaidi et al., 2017; Zaidi, Harper, Howse, Lamura, & Perek-Bialas, 2018). Each of these pillars supports some activity or group of activities in later life. The pillar of employment focuses on the labour market participation via increasing the retirement age, fighting age discrimination, and providing education and training. The pillar of social participation includes organised activities performed in the presence of other people, such as volunteering, lifelong learning, caregiving, political participation, and networking in general. The pillar of independent living consists of one type of activity – physical exercise – and more general concepts of health, autonomy, housing, and so on (EU Council, 2012; Eurostat, 2012; Zaidi et al., 2018). The EU supports these three dimensions of active ageing by various means at different geographical levels of the country members (AGE platform Europe, 2011; European Union, 2012; Zaidi & Howse, 2017).
One of the most ambitious efforts in the EU of ageing policies is the development of the Active Ageing Index (AAI), which is a multidimensional operationalisation measuring active ageing across countries (European Commission, 2013; Sidorenko & Zaidi, 2013; Zaidi et al., 2018). The AAI is used extensively in areas of research and social policy (Hess, Nauman, & Steinkopf, 2017; Perek-Białas, Zwierzchowski, Antczak, & Panek, 2017; Zaidi & Howse, 2017; Zaidi & UNECE, 2015) for the regular evaluation of the fulfilment of the active ageing goals and the recent progress in its member countries. The dimensions and indicators of AAI strongly overlap with the three pillars of active ageing. The employment rate of older adults and the provision of informal care are represented by more (and quite highly weighted) indicators, whilst the social participation and physical activities are less emphasised by this index. The documents formulating active ageing policy and the AAI operationalisation suggest that the uniform set of supported activities should a) uniformly increase as much as possible across all EU contexts and b) spread over time amongst older adults irrespective their preferences, opportunities and barriers.
This paper considers the support of the same activities and expectation of the same outcomes regardless of macro- and micro factors as highly problematic. Many authors showed that the availability of these activities, their outcomes, and their meaning for older people differ substantially (Conkova, Fokkema, & Dykstra, 2017; Hank, 2011; Lakomý, 2019; Timonen, 2016; Walker, 2009). These differences can be explained by many factors structured in several theories. Pierre Bourdieu (1984) describes differences in perceptions of reality, tendencies to act, and outcomes of the same behaviour by the term ‘habitus’, which is defined as ‘the systems of dispositions characteristic of different classes and class fractions’ (Bourdieu, 1984: 6). This definition means that the position in a social field – dependent on the magnitude of economic and cultural capital, gender, age, and geographical location – shapes specific social practices that ‘fulfil a social function of legitimating social differences’ (Bourdieu, 1984: 7).
These assumptions would make active ageing a product of habitus (
Several studies present vast differences in the prevalence of activities based on the EU country or region. The labour force participation amongst older adults is higher in Nordic countries and lower in Southern Europe with other regions in between (Zaidi et al., 2017). Furthermore, the retirement is associated with increasing obesity in Southern Europe, and to some point in Western Europe (Pertold, 2020). The occasional caregiving is more prevalent in Nordic countries and less in Southern Europe with the opposite being right for the intensive provision of care. Western Europe is close to Nordic countries and post-communist countries to Southern Europe in this regard (Albertini & Kohli, 2013; Hank & Buber, 2009; Igel & Szydlik, 2011; Neuberger & Haberkern, 2014). Finally, volunteering and social leisure as the indicators of social participation are performed by older adults more often in Nordic countries followed by Western Europe and Southern Europe with no findings for post-communist countries; this information is still incomplete (Galenkamp et al., 2016; Siegrist & Wahrendorf, 2009). The four European regions used in this paragraph are quite homogeneous and structure the data comprehensibly.
Moreover, the chance of devoting oneself to particular activities in later life is shaped by individual and relational characteristics, such as class, gender, age, geographical location, education, and social capital (Galenkamp et al., 2016; Holstein & Minkler, 2003; Lakomý, 2019; Rozanova et al., 2012; Serrat, Villar, & Celdrán, 2015; Siegrist & Wahrendorf, 2009). This empirical research shows a higher prevalence of all activities except intensive caregiving amongst younger, healthier, wealthier, and more educated older adults. Gender differences are more complex – women seem less active on the labour market (Hofäcker & Naumann, 2015; Lakomý, 2019), less involved in various types of social participation (Galenkamp et al., 2016; Serrat et al., 2015; Siegrist & Wahrendorf, 2009), and providing informal care more often (Hank & Buber, 2009; Igel & Szydlik, 2011; Rozanova et al., 2012). In sum, many studies indicated macro- or micro-contextual differences in the performance of activities supported by active ageing policy, but none of them describes their prevalence clearly and comprehensively.
Majority of studies examine the causes, covariates, and consequences of activities of older adults in a multivariate perspective (Adams, Leibbrandt, & Moon, 2011; Cattan, Hogg, & Hardill, 2011; Hofäcker & Naumann, 2015; Lakomý, 2020), which is a fruitful approach to the topic. Then, particular research provides descriptive statistics presented in univariate or bivariate form as a framework for a more extensive multivariate analysis (Arpino & Bordone, 2018; Hofäcker & Naumann, 2015; Hubatková, 2018; Igel & Szydlik, 2011). However, the existing literature lacks a systematic and comprehensive description of the overall prevalence of activities and prevalence sorted by the most important individual and contextual characteristics. This gap is not filled – at least for a data source used by this paper – by any available publication, including data documentation. This omission is even more important due to severe theoretical and practical implications of descriptive statistics itself, as illustrated by the presented study. Moreover, statistics presented in this form can guide further research and policies in the heterogeneous European context.
The paper focuses on activities promoted by active ageing in the form in which they are most often defined and measured, and describes their prevalence and its differences based on the key individual characteristics in four European regions. The description of the prevalence of activities amongst older adults in a cross-national perspective provides an important outlook on the topic. How often are activities supported by active ageing performed within the target population? How large are the differences in prevalence of the supposedly beneficial activities amongst the subgroups of older adults? This text presents an overview of these questions, based on the crucial data source for the study of ageing processes in Europe. Therefore, the resulting descriptive statistics can be used by researchers and policymakers not only for reconsideration of the active ageing policy, but also for getting the outlook about prevalence of these activities. Moreover, some conclusions are drawn regarding inequalities in the prevalence of desirable activities and their outcomes, which may focus further attention on this both academic and societal issue. The paper argues that a description of the prevalence of crucial activities in later life across the European regions and individual characteristics is essential for the understanding of this problem and states this as its main goal.
This paper expects the activities supported by the active ageing policy to vary strongly both across the European regions and individual characteristics. Specifically, the following research questions derived from the previous research are examined. It is important to stress that the main aim of the paper is not to elaborate the following relationships, but to describe the magnitude of differences on the high-quality comparable data, which was not done so far. The research questions are:
RQ1: Is the labour force participation the highest in Nordic countries and the lowest in Southern Europe? RQ2: Is the occasional caregiving more prevalent in Nordic countries and Western Europe and less prevalent in Southern Europe and post-communist countries? RQ3: Is the intensive caregiving more prevalent in Southern Europe and post-communist countries and less prevalent in Nordic countries and Western Europe? RQ4: Is the level of social participation the highest in Nordic countries, followed by Western Europe, Southern Europe, and post-communist countries? RQ5: Are all activities except intensive caregiving more prevalent amongst younger, healthier, and more educated older adults?
RQ1: Is the labour force participation the highest in Nordic countries and the lowest in Southern Europe?
RQ2: Is the occasional caregiving more prevalent in Nordic countries and Western Europe and less prevalent in Southern Europe and post-communist countries?
RQ3: Is the intensive caregiving more prevalent in Southern Europe and post-communist countries and less prevalent in Nordic countries and Western Europe?
RQ4: Is the level of social participation the highest in Nordic countries, followed by Western Europe, Southern Europe, and post-communist countries?
RQ5: Are all activities except intensive caregiving more prevalent amongst younger, healthier, and more educated older adults?
The paper uses data from wave 6 of the Survey of Health, Ageing and Retirement in Europe (SHARE) collected in 2015 (Börsch-Supan, 2017; Börsch-Supan et al., 2013). The sample from wave 6 consists of 64,585 respondents from 17 countries over 50 years of age, with the mean age 68.1 and median age 67 years and with women prevailing in a ratio of 56:44. The mean age varies moderately across countries from 65.4 (Croatia) and 70.6 (Sweden) due to differences in demographic structure and sampling methods. All the subsequent analyses use sample weights correcting for both nonresponse factor and sampling procedure, and thus, the results can be generalised to regional and national populations.
Countries in this study are sorted into four supra-national European regions grouping countries according to several criteria at the same time. These regions are the Nordic countries (Denmark and Sweden), Western Europe (Austria, Belgium, France, Germany, Luxembourg, and Switzerland), Southern Europe (Greece, Italy, Portugal, and Spain), and post-communist countries (Croatia, Czechia, Estonia, Poland, and Slovenia). The regions are defined mainly geographically, but also based on a typology of European welfare regimes (Ebbinghaus, 2012b, 2012a; Esping-Andersen, 1990; Fenger, 2007), empirical similarities and differences in contextual factors, and individual behaviours of older adults. This way, four groups of countries, which are relatively homogenous geographically, institutionally, economically, and culturally (Albertini & Kohli, 2013; Borges Neves, Barbosa, Matos, Rodrigues, & Machado, 2013; De Jong Gierveld & Tesch-Römer, 2012; Di Novi, Jacobs, & Migheli, 2015; Hank, 2011; Hofäcker, 2015; Igel & Szydlik, 2011), were created. The Nordic countries are represented by 7,486 respondents, Western Europe by 21,358, Southern Europe by 17,091, and post-communist countries by 18,650 respondents.
The dependent variables – activities – were recoded in order to follow their definition by the active ageing approach as closely as possible and to simplify the analysis by their binarization at the same time. The labour force participation takes the value 0 for retired, homemakers, pensioners, and permanently disabled respondents, and the value 1 for those employed, self-employed, and unemployed. Care outside the household indicates at least monthly care or help provided to any recipient outside the household. Care within the household indicates daily personal care as measured in SHARE. These two types of care should be distinguished to address the large differences in their context and consequences (Broese van Groenou, de Boer, & Iedema, 2013; Colombo, Llena-Nozal, Mercier, & Tjadens, 2011; Lakomý, 2020). Other activities take the value 1 if they are performed at least monthly. The original formulation of these activities in the questionnaire is ‘volunteer or charity work’, ‘educational or training course’, ‘sport, social, or other kind of club’, ‘participation in a political or community-related organisation’, and ‘vigorous physical activity, such as sports, heavy housework, or a job that involves physical labour’.
The individual characteristics used for sorting the prevalence of activities are gender, age, health status, and the level of education. Some of these variables were categorised to provide more readily interpretable results. Gender gets values 0 = male and 1 = female; age is grouped in categories 51–59, 60–69, 70–79, and 80+ years; health status in poor or fair, good, and very good or excellent; and the level of education in primary (ISCED 0, 1), secondary (ISCED 2–4), and tertiary (ISCED 5, 6).
This study graphically displays the prevalence of activities supported by active ageing in the European population 50+ in the nationally-representative samples available in the data. Then, the analysis shows the differences in the availability of the promoted activities based on gender, age group, health status, and the level of education. Finally, all the graphics are also presented separately for each of the European regions in order to achieve more context-sensitive results. The large sample provides sufficient number of cases for subsequent analyses and a distribution fitting to population parameters is ensured by the utilised weights. The overall aim of the analysis is to describe the prevalence of crucial activities in later life across the European regions and individual characteristics, which is essential for the understanding of this problem.
The paper does not use any statistical tests for several reasons. It is important to note that all the differences commented by the paper are statistically significant with this sample size. Moreover, testing for statistical significance assumes that a small fraction of cases were randomly selected from the target population, which does not work in the case of this paper. Instead, I argue that published descriptive statistics in a form substantive differences between groups/regions are very scarce, even though this information is usually widely cited and invaluable to researchers and policymakers, who don’t have high-quality data ready for the analysis. Finally, the decision not to use statistical significance is in line with a recommendation to use it only in necessary cases and do not overestimate its importance (Eisenhart et al., 2006; Nuzzo, 2014; Wasserstein & Lazar, 2016).
Figure 1 shows the share of respondents performing specific types of activities. Overall, the most prevalent in the 50+ population is physical activity at 52 per cent, followed by labour force participation (32 per cent), care provided outside the household (28 per cent), involvement in sport or social club (21 per cent), volunteering (12 per cent), care provided within the household (6 per cent), and participating in educational and political organisation at the same value of 4 per cent. Furthermore, the regional differences are profound. All activities, except volunteering and caregiving within the household, are most common in the Nordic countries – followed by Western Europe – with participation in a sport or social club, educational course, and volunteering being multiple times more prevalent than in Southern Europe and post-communist countries. The only activity slightly more prevalent in Southern Europe and post-communist countries than in Nordic countries and Western Europe is caregiver within the household. The regional differences in the prevalence of activities are in line with previous research. The most important of these differences are interpreted in the conclusions.
The findings in Figure 2 are based on the same data, but show the prevalence of activities across gender as the first examined individual characteristic. The upper part of Figure 2 presents the results for the pooled sample, whilst the lower part focuses on regional differences with the effect of gender. The largest differences between gender are in labour force participation (11 percentage points more for men), physical activity (9 percentage points more for men), and caregiving outside the household (5 percentage points more for women), which reflect the traditional gendered division of labour. The gender differences are similar across regions, but the Nordic countries evince a large prevalence of sport and social clubs for both men and women, together with a higher percentage of women in educational courses and volunteering activities, whilst men participate in sport or social clubs about two times more often than women in post-communist countries.
Figure 3 follows the same logic as Figure 2 and displays the differences in the prevalence of activities amongst age groups. Some activities are substantially differentiated here, with labour force participation and physical activity as the most significant examples. Whilst these two activities peak in the youngest group of 51–59 years, it is the group of 60–69 year olds for care provided outside the household, volunteering, and sport or social club and the group of 80+ years for care provided within the household. The results for regions are similar, with a very high prevalence of physical activities and sport or social club for all age groups in the Nordic countries and a low prevalence of people working over 60 years in post-communist countries (13 per cent in the 60–69 group) as the most notable region-specific findings.
The differences between activities based on health status are displayed in Figure 4. The differences amongst groups based on health status are similar to those based on age. Nevertheless, health status is associated with activities in a more linear and often stronger way than age. All activities are substantially most prevalent for respondents with very good or excellent health, and least prevalent for those with poor or fair health, with the only exception being in caregiving provided within the household. The association of this type of care provision is reversed, as the respondents with poor or fair health provide care within the household two times more often than those with very good or excellent health. Furthermore, some differences amongst groups based on health are lower in Nordic countries (caregiving within the household, volunteering, and political organisation) and Southern Europe (caregiving outside the household) than in other European regions.
Figure 5 focuses on the effect of the level of education on the activities performed in later life. The differences amongst educational groups represent differences based on the ascribed social status much more than the previous findings, and their volume is again substantial, with multiplicative differences between primary and tertiary education for some activities. The respondents with primary education participate the least often and those with tertiary education the most often in all but one activities, with caregiving within the household is structured reversely. The large differences are present in all types of social participation, but also in labour force participation, which is performed by 45 per cent of respondents with tertiary education and by 13 per cent of respondents with primary education over 50 years of age. This difference could be ascribed to the structure of subpopulations, which is partially addressed in the region-specific analysis. The educational difference reappeared in all regions, with the effect of education on labour force participation being higher in Nordic and post-communist countries and many other effects being most pronounced in post-communist countries as well.
This paper uses data from SHARE to describe the prevalence of activities supported by active ageing across European regions and individual characteristics. Such descriptive information is needed for researchers and policymakers in the area amidst their professional activities. The main findings show large differences in the prevalence of activities supported by active ageing policy both across macro- and micro-context, which are not properly reflected by the EU policy initiatives. This means that the uniform support of the predefined set of activities favours some countries and older adults with higher inclination to them. Supposedly, a group of actors with certain habitus formulated policies fitting their opportunities and preferences to legitimise the inequalities in later life both across and within the EU countries (Bourdieu, 1984). Hence, the uniform and context-insensitive active ageing policy is not fitting to the various needs and barriers of older adults living in the EU and should be reconsidered. Moreover, this main conclusion supports calls of Timonen (2016) and Phillips et al. (2010) or for replacing active ageing approach by more reflexive and flexible approaches.
The specific prevalence of all activities brings important contribution and should be taken into account as well, as this comprehensive set of data has not been published in this way. The differences in current working status across European regions could be attributed to pension systems, economic situation, the health status of older adults, values connected to work, or ageism on the labour market (Hofäcker, 2015; Neuberger & Haberkern, 2014; Petrová Kafková & Rabušic, 2010). The patterns of care provision across four European regions are congruent with the previous research, usually comparing Northern and Southern Europe (Albertini & Kohli, 2013; Hank, 2011; Hank & Buber, 2009; Igel, Brandt, Haberkern, & Szydlik, 2009; Igel & Szydlik, 2011).
On the one hand, Southern Europe has a low prevalence of occasional care and a high prevalence of regular care due to strong familial norms and intergenerational transfers based on co-residence, which lead to intensive interactions based on functional dependence without motivating towards more casual and voluntary interactions (Albertini & Kohli, 2013; Di Novi et al., 2015; Neuberger & Haberkern, 2014). These tendencies also apply to post-communist countries, according to the presented findings. On the other hand, strong welfare states and weak familial norms in Nordic countries reduce the burden of caregiving and create an environment enabling occasional interactions, including the provision of care as well as social participation (Igel & Szydlik, 2011; Neuberger & Haberkern, 2014). Western Europe is placed somewhere between the two types (Albertini & Kohli, 2013; Igel & Szydlik, 2011), although this paper locates it as being closer to Nordic countries. In summary, the results indicate: a) substantial regional differences across Europe, and b) similarities between two pairs of regions, which may form some North-West/South-East axis.
The examined individual characteristics are connected to activities in later life too. All activities apart from caregiving within the household are much (and often multiple times) more common amongst younger, healthier, and more educated older adults. The vast differences raise the question of who has access to supposedly beneficial activities, who is able and motivated to participate, and which groups of older individuals benefit from the active ageing approach. According to some literature, those beneficiaries come from already privileged groups including governments and groups of older adults already inclining towards supported activities (de São José, Timonen, Amado, & Santos, 2017; Golant, 2014; Rozanova et al., 2012). Hence, the unintended consequence of the support of some beneficial activities can be a relocation of responsibility for the quality of life in older ages to individuals (Timonen, 2016), legitimisation of inequalities, and their further increase.
In contrast, intensive care within the household is slightly more often provided by older, less healthy, and less educated individuals, and women, which suggests that it more concerns less privileged groups omitted by the welfare system (Fokkema, ter Bekke, & Dykstra, 2008; Igel & Szydlik, 2011). In addition, women are less active in social participation and the labour market, whilst they provide both types of care more often. The gender differences are not strong in the data despite their salience in the research focusing on the content of caregiving, which stresses the importance of qualitative research for presenting different contents and meanings of activities in later life and generally the gendered nature of ageing (Adams et al., 2011; Kim & Moen, 2002; Rozanova et al., 2012; Souralová, 2014). To sum up, this paper provides the prevalence of activities supported by active ageing and its differences across four European regions, but also shows the potential of the active ageing approach to raise inequalities in later life, which needs to be investigated more by future research.
The findings presented in this paper have three main limitations. Firstly, the results are based on data from 17 countries, which are divided into four European regions. Hence, we can neither make conclusions about all European countries nor observe a heterogeneity within the defined regions. Still, the results fulfil the goal to provide context-specific and generalizable knowledge at the same time. Secondly, the presented graphics are far from explaining the differences in the prevalence of activities in the complexity of social reality. All the included individual characteristics interact amongst themselves and a lot more micro- and macro-factors shape the activities. Thirdly, the findings cannot be directly used for the formulation of policy action, as no social policy addresses groups of EU countries and more research is needed, both for an explanation of these differences and formulation of specific social policy. However, the presented findings can lead to a reconsideration of the context-insensitive EU policy and help to identify the European regions and groups of older adults that struggle most to fulfil the policy goal. Finally, it seems crucial to provide a still unpublished basic description of the unequal prevalence of activities supported by the most prominent social policy approach in Europe and the limitations rather show the future directions of the research field.