Content uploaded by Kene Henkens
Author content
All content in this area was uploaded by Kene Henkens on Mar 11, 2019
Content may be subject to copyright.
Empirical Article
Working After
Retirement and Life
Satisfaction: Cross-
National Comparative
Research in Europe
Ellen Dingemans
1,2
and Ke
`ne Henkens
1,2,3
Abstract
This study examines differences in life satisfaction between full retirees and
working retirees in Europe. We hypothesize that these differences depend
on the financial resources of retirees and the resources available in the
household and country context. We selected retirees from the “Survey of
Health, Ageing and Retirement in Europe” project and estimated country
fixed effects models to explain their life satisfaction. The results indicate a
positive relationship between working after retirement and life satisfaction
for retirees with low pension income without a partner. Additionally,
working after retirement seems to be most important for life satisfaction in
relatively poor countries.
1
Department of Work & Retirement, Netherlands Interdisciplinary Demographic Institute
(NIDI-KNAW), The Hague, the Netherlands
2
University Medical Center Groningen, University of Groningen, Groningen, the Netherlands
3
Department of Sociology, University of Amsterdam, Amsterdam, the Netherlands
Corresponding Author:
Ke
`ne Henkens, Netherlands Interdisciplinary Demographic Institute (NIDI-KNAW), P.O. Box
11650, NL-2502 AR, The Hague, the Netherlands.
Email: henkens@nidi.nl
Research on Aging
1–22
ªThe Author(s) 2019
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0164027519830610
journals.sagepub.com/home/roa
Keywords
working retirees, late life well-being, cross-national comparison, country
fixed effects analysis
Participation in paid work after retirement is increasingly common in various
European countries and the United States (Alcover, Topa, Parry, Fraccaroli,
& Depolo, 2014; Giandrea, Cahill, & Quinn, 2009). Retirement is no longer
necessarily an abrupt withdrawal from the labor force but increasingly takes
the form of a gradual process in which older adults make several transitions
in and out of paid work (Beehr & Bennett, 2015). Previous research is
fragmented across various single-country studies and mainly addresses
research questions concerning the determinants of working after retirement.
Studies on the consequences of postretirement work for life satisfaction, and
how these consequences are shaped by country-specific contexts, are rather
limited. The aim of this study is to examine the relationship between working
after retirement and life satisfaction in European contexts.
The question of how working after retirement affects the life satisfaction
of older adults has received some attention in the literature. Generally,
positive outcomes of working after retirement have been reported. For
instance, positive outcomes were found for retirement satisfaction (Dorf-
man & Rubenstein, 1993; Kim & Feldman, 2000) and for well-being and
life satisfaction (Kim & Feldman, 2000; Zhan, Wang, Liu, & Shultz, 2009),
particularly when retirees continue to work for enjoyment reasons (Dinge-
mans & Henkens, 2014). Furthermore, research among older Americans
(mean age 76) has shown that working after retirement has a protective
effect on physical health (Luoh & Herzog, 2002). However, working after
retirement could also threaten life satisfaction when older adults feel forced
to work after retirement because of financial reasons (Dingemans &
Henkens, 2014).
These studies have provided important insights into the consequences of
postretirement employment transitions. However, they largely neglect that
such transitions occur in very diverse country-specific contexts, which may
affect the relationship between postretirement work and life satisfaction. For
instance, Hershey, Henkens, and Van Dalen (2010) have pointed to the large
degree of cross-national variation in the worries people have with regard to
their retirement income. Also, European statistics show that levels of poverty
in old age differ across countries (Organisation for Economic Cooperation
and Development (OECD), 2017). The importance of work after retirement
for life satisfaction may depend upon these country-specific factors. For
2Research on Aging XX(X)
instance, the access to additional income that working after retirement can
provide may be particularly important for life satisfaction if the availability
of resources at the country level is limited.
Our investigation of the relationship between working after retirement
and life satisfaction in European contexts aims to contribute to the existing
literature in three ways. First, our research is the first to adopt a European
comparative perspective on working retirees’ life satisfaction. The large
majority of previous studies on working after retirement have taken a
country-specific design (e.g., Dingemans & Henkens, 2014; Kim & Feld-
man, 2000). However, it is difficult to compare the results from single-
country studies because of their different operationalizations of working after
retirement. In this study, we provide information about cross-national varia-
bility in working after retirement and life satisfaction levels using harmo-
nized data from 16 European countries.
Second, this research contributes to a better theoretical understanding of
the relationship between working after retirement and life satisfaction by
focusing on the role of financial resources. Scholars agree that it cannot be
determined a priori whether retirement has positive or negative effects on life
satisfaction. Retirement is linked to both positive and negative experiences
(Quine, Wells, De Vaus, & Kendig, 2007), partly depending on whether the
decision to retire is voluntary or involuntary (Shultz, Morton, & Weckerle,
1998). Similarly, it is hard to determine a priori how working after retirement
relates to older adults’ life satisfaction. Using conservation of resources
(COR) theory (Hobfoll, 1989, 2002), we explicitly focus on individual pen-
sion income as a boundary condition for the effect of working after retire-
ment on life satisfaction.
Third, we integrate both individual-level factors and cross-national factors
into the theoretical framework. Building on COR theory (Hobfoll, 2002), we
argue that it is not only individuals’ financial resources but also resources in
the context that may affect how working after retirement relates to life
satisfaction. More specifically, we investigate whether the relationship
between working after retirement, individual pension income, and life satis-
faction is moderated by conditions in the household and the country.
The data for this study are derived from the “Survey of Health, Ageing
and Retirement in Europe” (SHARE) project, which includes information on
work and retirement issues and life satisfaction in several European coun-
tries. We use the cross-sectional information of 54,156 retirees in 16 Eur-
opean countries. Working after retirement is thereby defined as participation
in paid work while also receiving pension income (Dingemans & Henkens,
2014; Parry & Bown Wilson, 2014).
Dingemans and Henkens 3
Theoretical Framework
Individual Pension Income
The central tenet of the COR theory is “that people strive to retain, protect,
and build resources and that what is threatening to them is the potential or
actual loss of these valued resources” (Hobfoll, 1989, p. 516). Retirement, in
this theory, could be seen as a stressful event for some older adults, since it is
defined by changes in income sources that can be experienced as threatening.
Traditionally, the transition to retirement is marked by a change from receiv-
ing income from paid work to receiving (different forms of) pension income.
Generally, income is only mildly related to life satisfaction (Easterlin, 2005).
Nevertheless, as with the income from paid work (Kahneman & Deaton,
2010; Markides & Martin, 1979), it can be expected that a higher pension
income is associated with a high level of life satisfaction, while a low pension
income may be threatening for life satisfaction. A coping strategy in case of
low pension income is to participate in paid work after retirement. In COR
theory, working after retirement could be seen as a way to supplement the
pension income with the income from paid work. For instance, Yang (2011)
showed that, among male Korean retirees, working after retirement substan-
tially lowered the risk of poverty because these working retirees were able to
supplement their pension incomes with income from paid work. It follows
from this literature that retirees may benefit from working after retirement by
supplementing their pension income with income from paid work, thereby
increasing their life satisfaction.
The question remains of whether all retirees benefit equally from this
additional income from paid work. COR theory states that resources gains
in themselves have only modest impact on emotional outcomes (Hobfoll,
2002). However, the income from paid work may become increasingly
important when the resources gain is crucial to sustain goal pursuit, such
as the sufficiency of resources to make ends meet in retirement. At the lower
end of the pension income continuum, the amount of pension income retirees
receive may not be enough to allow a decent standard of living and therefore
life satisfaction may be threatened. For low pension income groups, paid
work may thus be important to supplement their income and obtain higher
levels of life satisfaction. In contrast, among well-off individuals, further
accumulation of financial resources may no longer add to a sense of well-
being and happiness (Muffels, Skugor, & Dingemans, 2014). On the basis of
the foregoing, it is therefore hypothesized that paid work after retirement is
more important for life satisfaction for retirees with low levels of pension
income than for retirees with high levels of pension income (Hypothesis 1).
4Research on Aging XX(X)
Contextual Resources
The relationship between working after retirement and life satisfaction may
be more complex. COR theory refers to conditions that may be seen as a kind
of resources (Hobfoll, 1989), for instance, whether you live with a partner (a
condition) or in which country you live (another condition). The role of
working after retirement in supplementing pension income may be particu-
larly important in conditions where access to other resources is limited.
Previous research has investigated the relationship between unemployment
and life satisfaction among older adults and found that workers without
access to financial resources in their broader contexts were more likely to
experience depressive moods after being laid off than workers with financial
alternatives (Kessler, Turner, & House, 1988). Similarly, when older adults
make the decision to work postretirement, safety nets may be available to
them at the level of their household and/or at the national level.
Households. Living with a partner has been shown to be an important social
resource (Damman, Henkens, & Kalmijn, 2015). Spouses can provide the
other spouse with structured time, affection, and guidance. Moreover, one’s
spouse has a strong influence on various aspects of life (Settersten, 2003) and
can affect satisfaction in retirement (Szinovacz & Davey, 2005). Addition-
ally, partners could benefit from each other’s financial resources in retire-
ment. Unitary models of households consider the household “as if it were a
single individual,” thereby presuming that income sources within households
are pooled (Eurostat, 2013, p. 11). Retirees with low levels of pension
income may (partly) rely upon support from their partner. As a result, the
link between their own pension income level and life satisfaction may be
weaker. We therefore hypothesize that paid work after retirement is more
important for life satisfaction for retirees with low levels of pension income
who are single, as compared to retirees with low levels of pension income
who have a partner (Hypothesis 2).
Countries. The same mechanism of resource substitution may apply in rela-
tion to the material wealth of the broader country context. In richer coun-
tries, that is, those with higher per capita levels of material wealth, living
standards and population well-being areassumedtobebetterthaninpoorer
countries (United Nations Statistics Division, 2016). It can be assumed that
more resources are available in rich countries, for instance, to support
citizens in need of financial or medical support or to provide tax reductions
or practical assistance. The availability of resources in the country may
Dingemans and Henkens 5
weaken the link between individual-level pension income and life satisfac-
tion. For instance, previous research has shown that financial satisfaction is
less strongly associated with life satisfaction in richer countries (Diener &
Diener, 2009; Oishi, Diener, Lucas, & Suh, 1999). We hypothesize that
paid work after retirement is more important for life satisfaction for retirees
with low pension incomes in countries with low material wealth than for
those with low pension incomes in countries with high material wealth
(Hypothesis 3).
Research Design
Data
The data for this study were obtained from the SHARE project. SHARE is a
longitudinal and cross-national data collection project that focuses on adults
aged 50 and older (Borsch-Supan et al., 2013). Compared to other cross-
national European data collection projects that focus on the total adult pop-
ulation, a clear advantage of SHARE is the large sample of older adults. As a
result, the sample sizes per country are large enough to investigate less
common behavior such as working after retirement, which is relatively com-
mon in some countries, but exceptional in others.
We used data from the second, fourth, and fifth waves, collected in 2006,
2011, and 2013, respectively. The first wave (2004) was excluded because
life satisfaction was not measured in this wave. Wave 3 (2008) was excluded
because of its different format, focusing on life histories. We included data
on 16 European countries: Austria, Belgium, Czech Republic, Denmark,
Estonia, France, Germany, Hungary, Italy, the Netherlands, Poland, Portu-
gal, Slovenia, Spain, Sweden, and Switzerland. The analytical sample was
restricted to respondents aged 60–75 because pensions can be received as
early as age 60 in most European countries and working after retirement
often begins shortly after career exit (Maestas, 2010). The analytical sample
was further restricted to retirees. Retirement in this case is defined as the
receipt of pension income—being a public pension, an occupational pension,
or a combination of both (Dingemans, Henkens, & Van Solinge, 2017).
Those without any form of pension income, that is, those still working in
their main careers, were excluded from the analyses (11%). Item nonre-
sponse was low (2%), and missing values were imputed using a multiple
imputation procedure (mi in STATA, version 15). The final sample included
54,156 individuals, with 53%women and a mean age of 68.
6Research on Aging XX(X)
Measures
The dependent variable, life satisfaction, was measured by asking respon-
dents: “On a scale from 0 to 10, where 0 means completely dissatisfied and 10
means completely satisfied, how satisfied are you with your life?” The main
independent variables were working after retirement and pension income. All
retirees (i.e., those receiving a pension income) were asked whether they still
participated in a paid job. Retirees who reported that they received income
from paid work were classified as working retirees. The amount of pension
income respondents received was measured via various questions on differ-
ent forms of pension income. The SHARE team imputed each of the income
variables to deal with missing cases and constructed cross-national compa-
rable measures (De Luca, 2018; Malter & Borsch-Supan, 2015). We used
information on partner status and material wealth of the country to explore
contextual effects at the household and country level. With regard to partner
status, those in marriages or in registered partnerships were contrasted with
retirees who reported they had never been married, had experienced a
divorce, or had been widowed. At the country level, we accounted for gross
domestic product (GDP) per capita, which reflects the material wealth of
populations (Be´renger & Verdier-Chouchane, 2007; United Nations Statis-
tics Division, 2016).
In the explanatory models, we controlled for factors that have been shown
to impact both decisions on retirement and well-being in late life, such as
whether one has children, gender, age, educational level, and health. Health
status is measured by the number of chronic diseases and feeling limited in
the execution of daily activities. A full description of the wording of the
questions is given in the Online Appendix, Table A1. Means, standard devia-
tions, and ranges of all variables are presented in Table 1.
Analytic Strategy
To examine the relationship between working after retirement and life satis-
faction, we have been stacking the data of the three selected waves (2006,
2011, and 2013) as to make optimal use of the SHARE data. The advantage
of stacking the data is that we can include both between-individual and
within-individual variation, even though it is not possible to distinguish
between these forms of variation due to the relatively low number of transi-
tions to working after retirement in the data (within-individual variation). We
used the cluster option in our regression models to control for the multiple
measurements among some of the respondents. In addition, we include
Dingemans and Henkens 7
country dummies in the regression analyses to control for the clustering of
individuals within countries, which is referred to as country fixed effects
modeling (Mo¨hring, 2012). As a result, it is not possible to test the direct
effect of GDP on life satisfaction. However, the country fixed effects
approach is preferred over more advanced multilevel techniques because
of the relatively low number of cases at the country level, which could bias
the results (Bryan & Jenkins, 2013). We checked our estimations using
multilevel techniques (i.e., mixed models) but did not find any substantively
different results.
The analyses are divided into two parts. First, we concentrate on the
individual level to investigate the moderating impact of individual pension
income on the relationship between working after retirement and life satis-
faction (two-way interaction). Second, we challenge the individual-level
findings by incorporating contextual factors at the household level (partner
status) and the country level (material wealth in the country). We therefore
use two three-way interactions to explain life satisfaction, interacting indi-
vidual pension income and working after retirement with both partner status,
and material wealth in the country. The latter three-way interaction reflects a
cross-level interaction—crossing the individual level (working status and
individual pension income) with the country level (material wealth in the
country)—that can beneficially be explored in a country fixed effects
approach (Bennett & Mo¨hring, 2015; Mo¨hring, 2012).
Table 1. Means, Standard Deviations (SD), and Ranges of Variables.
Range
Variable Mean SD Minimum Maximum
Life satisfaction 7.61 1.80 0 10
Working after retirement 0.11 0 1
Pension income (/1,000) 13.77 18.76 0.02 545.68
No partner 0.27 0 1
Child(ren) 0.91 0 1
Female 0.53 0 1
Age 67.98 4.21 60 75
Education 2.80 1.43 0 6
Chronic diseases 1.84 1.52 0 14
Limited daily activities 0.09 0 1
GDP per capita 35.09 8.05 20.24 50.88
Source. Survey of Health, Ageing and Retirement in Europe (www.share-project.org).
Note. GDP ¼gross domestic product.
8Research on Aging XX(X)
Our dependent variable, life satisfaction, is standardized, so that the
coefficients of the dummy variables in our analyses can be interpreted as
Cohen’s deffect sizes. Following Cohen, small effects have a score of
around .2, medium effects around .5, and large effects approach .8 (Rice
& Harris, 2005).
Results
Descriptive Results
On average, about 1 in 10 of the retirees in the sample was working after
retirement (i.e., combining the receipt of a pension income with participation
in paid work), but the variation across the investigated countries was sub-
stantial, as shown in Figure 1. In countries such as Slovenia, Spain, and
Poland, only 2–4%of the retirees in the age range between 60 and 75 worked
past retirement. In countries such as Switzerland, Sweden, and Estonia,
working after retirement was much more common; about one in five retirees
in these countries was working after retirement. Country variations are fur-
ther described in Online Appendix, Table A2 (Online Appendix). For
instance, life satisfaction was lowest in countries such as Poland, Hungary,
and Estonia, while the highest levels of life satisfaction were found in Den-
mark, Switzerland, and Sweden.
0 5 10 15 20 25
Estonia
Sweden
Switzerland
Denmark
Czech Republic
Germany
Netherlands
Portugal
Austria
Belgium
Hungary
France
Italy
Poland
Spain
Slovenia
Figure 1. Percentage of working retirees across countries.
Source: Survey of Health, Ageing and Retirement in Europe, Waves 2, 4, and 5.
Dingemans and Henkens 9
Multivariate Results
Individual pension income. Table 2 presents the estimates from the country
fixed effects models explaining life satisfaction on the basis of individual-
level factors. The main relationships with life satisfaction are presented in
Model A. The results show that, overall, working retirees reported higher
levels of life satisfaction than full-time retirees. We found a positive rela-
tionship with pension income meaning that life satisfaction was higher
among retirees with higher pension incomes. Additionally, retirees without
a partner were significantly less satisfied with life than retirees with a partner.
The model also contained control variables and the results provide support
for some well-known relationships with life satisfaction. In particular, health
status was important for life satisfaction. The negative coefficients for the
health indicators (b¼.13 and b¼.46 for number of chronic diseases and
limited daily activities, respectively) suggest that life satisfaction was lower
among retirees with health problems. We conducted sensitivity analyses to
test the relationship with other health measures as well, such as subjective
Table 2. Country Fixed Effects Models Regressing Life Satisfaction (std) on Work
Status and Pension Income.
Model A Model B
Working .135 (.013)** .132 (.014)**
Working Pension Income (ln) .039 (.014)**
Working GDP .008 (.002)**
Working Level of Education .002 (.013)
Pension income (ln) .060 (.005)** .065 (.006)**
No partner .275 (.009)** .275 (.009)**
Child(ren) .054 (.014)** .055 (.014)**
Female .033 (.008)** .032 (.008)
Age (std) .028 (.004)** .028 (.004)**
Level of education (std) .063 (.004)** .063 (.004)**
Chronic diseases (std) .134 (.004)** .133 (.004)**
Limited daily activities .463 (.014)** .460 (.014)**
Constant .268 (.021)** .215 (.024)**
F481.15 433.94
df 26 29
Source. Survey of Health, Ageing and Retirement in Europe, Waves 2, 4, and 5.
Note. N ¼54,156. Standard errors are in parentheses. Pension income and GDP per capita are
centered. All models are controlled for country and year of measurement. The models account
for 19% of the variance in life satisfaction. GDP ¼gross domestic product.
*p< .05. **p< .01.
10 Research on Aging XX(X)
health and the number of symptoms people reported, and this did not change
our substantive conclusions. Educational level was positively related to life
satisfaction. We checked whether the relationship between work and life
satisfaction was dependent upon educational level, which was not found to
be the case. Further, older retirees, retirees with children, and women were
significantly but only slightly more satisfied with their lives than their
younger counterparts, those without children and men.
The two-way interaction between work status and pension income was
added to the explanatory model in Model B of Table 2 (because the level of
pension income is likely to be correlated with the per capita GDP of a
country, we also controlled for the interaction between work after retirement
and GDP). Similarly to Model A, in Model B, the coefficient of working
status implies that working retirees were more satisfied with life than full
retirees. The coefficient of the interaction term is negative. Without inter-
preting this causally, this suggests that the difference in life satisfaction
among working retirees and full retirees decreases with pension income.
Thus, when pension income was higher, differences in life satisfaction
between working retirees and full retirees were smaller.
Contextual resources. Table 3 shows the results of the analyses investigating
the individual-level results in the broader contexts of households and coun-
tries. At the household level, it was expected that for retirees with low
individual pension income, participationinpaidworkwouldbemore
strongly related to the life satisfaction of those who were single than for
those with a partner (Hypothesis 2). The results in Model A of Table 3 point
into this direction. The three-way interaction between working status, indi-
vidual pension income, and partner status was negative and significant. The
interaction effect of individual pension income and working thus decreases
more for retirees without a partner than for retirees with a partner. Put
differently, it means that working in retirement is more strongly related to
life satisfaction for retirees who do not have a partner. Moreover, the two-
way interaction between working status and partner status is significant in
Model A of Table 3, suggesting that working after retirement is more ben-
eficial for life satisfaction of retirees who do not have a partner independent
of the financial context. For instance, for those without a partner work might
be important as a means of social integration (Damman et al., 2015).
To fully understand the interpretation of the results in Model A, the
relationship between working after retirement and life satisfaction is illu-
strated in Figure 2. Working after retirement is associated with higher life
satisfaction than being in full retirement, and this difference decreases with
Dingemans and Henkens 11
pension income. The difference between the two lines in Figure 2 illustrates
the different relationships of working after retirement and individual pension
income for retirees with and without a partner. Working after retirement
seems to be particularly important for retirees with low pension income who
do not have a partner. For them, the Cohen’s deffect size is around .8,
illustrating a strong relationship. For retirees with a partner and a low pension
income, the effect size is reduced to .4.
At the country level, we hypothesized that paid work after retirement was
more important for the life satisfaction of retirees with low pension income in
countries with low material wealth compared with those with a low pension
Table 3. Country Fixed Effects Models Regressing Life Satisfaction (std) on Work
Status, Pension Income, and Contextual Factors.
Model A Model B Model C
Working .101 (.016)** .114 (.016)** .091 (.018)**
Working Pension Income (ln) .054 (.013) .043 (.015)** .025 (.016)
Working No Partner .079 (.028)** .075 (.028)**
No Partner Pension Income (ln) .045 (.010)** .044 (.010)**
Working No Partner Pension
Income (ln)
.080 (.026)** .072 (.026)**
Working GDP .007 (.002)** .006 (.002)**
Pension Income (ln) GDP .001 (.001) .001 (.001)
Working Pension Income
(ln) GDP
.003 (.001)* .003 (.001)*
Working Level of Education .001 (.013) .004 (.013) .004 (.013)
Pension income (ln) .060 (.006)** .065 (.006)** .055 (.006)**
No partner .207 (.026)** .275 (.010)** .209 (.026)**
Child(ren) .056 (.014)** .054 (.014)** .055 (.014)**
Female .031 (.008)** .031 (.008)** .029 (.008)**
Age (std) .029 (.004)** .028 (.004)** .029 (.004)**
Level of education (std) .062 (.004)** .063 (.004)** .063 (.004)**
Chronic diseases (std) .133 (.004)** .133 (.004)** .133 (.004)**
Limited daily activities .461 (.014)** .460 (.014)** .460 (.014)**
.267 (.021)** .226 (.024)** .230 (.024)**
F406.06 406.19 371.13
df 31 31 34
Source. Survey of Health, Ageing and Retirement in Europe, Waves 2, 4, and 5.
Note. N ¼54,156. Standard errors are in parentheses. Pension income and GDP per capita are
centered. All models are controlled for country and year of measurement. The models account
for 19% of the variance in life satisfaction. GDP ¼gross domestic product.
*p< .05. **p< .01.
12 Research on Aging XX(X)
income in high material wealth countries (Hypothesis 3). This hypothesis is
supported by the results in Model B of Table 3. The three-way interaction
between working status, individual pension income, and GDP is positive and
significant. This means that the decrease in the interaction effect of working
status with individual pension income is weaker in countries with high GDP.
It seems that the increase in life satisfaction is particularly high for working
retirees with low individual pension income and living in countries with
relatively low levels of material wealth.
Again, in order to fully understand the complexity of the three-way inter-
action, the results are illustrated in Figure 3. The three lines in Figure 3
illustrate the relationships for countries with low, medium, and high levels
of GDP (relative to the full sample of the countries in this study). The size of
the effect for individuals in the lowest pension income group in low GDP
countries is around .6, meaning a medium-sized effect. In contrast, the size of
the effect in high GDP countries approaches 0 irrespective of the individual
pension income level.
In our final model in Table 3 (Model C), we included both three-way
interactions. The results reveal that both three-way interactions remain sig-
nificant. Figure 4A and B shows the relationship between working after
retirement and life satisfaction by individual pension income. The three lines
illustrate the relationships for low, middle, and high levels of GDP per capita.
Figure 4A shows the results for retirees without a partner, while Figure 4B
-0.2
0
0.2
0.4
0.6
0.8
1
45678910
Relaonship (Cohen' d)
individual pension income (ln)
partner
no partner
Figure 2. The impact of work in retirement on life satisfaction by the level of
individual pension income and partner status (micro-context; based on Table 3,
Model A).
Dingemans and Henkens 13
shows the results for retirees with a partner. Among retirees without a partner
(Figure 4A), the relationship between working after retirement and life satis-
faction was strongest for those with low individual pension income in coun-
tries with relatively low levels of GDP. The Cohen’s dfor these retirees is
around .9, which illustrates a strong relationship. For single retirees with low
individual pension incomes in mean GDP countries, the Cohen’s dis around
.6, which represents a medium-sized effect. Finally, in high GDP countries,
the Cohen’s dis further reduced to around .3 but still reflects a small effect
size. Figure 4B shows that among retirees with a partner, the relationship
between working after retirement and life satisfaction is also strongest for
those with a low individual pension income in countries with a low GDP.
However, the Cohen’s deffect size for these retirees is much smaller than for
retirees without a partner. The Cohen’s dis close to .5, which illustrates a
medium-sized effect. The relationship for retirees in mean and high GDP
countries is low (Cohen’s d¼.2) to 0.
Discussion
With the increasing prevalence of working after retirement in European
countries, questions have been raised about the impact it has on older adults’
life satisfaction. In this study, we investigated the relationship between work-
ing after retirement and life satisfaction from a European perspective. Using
-0.2
0
0.2
0.4
0.6
0.8
1
45678910
Relaonship (Cohen' d)
individual pension income (ln)
low gdp
mean gdp
high gdp
Figure 3. The impact of work in retirement on life satisfaction by the level of
individual pension income and gross domestic product per capita (macro-context;
based on Table 3, Model B).
14 Research on Aging XX(X)
data from 16 European countries, we found that the relationship between life
satisfaction and working after retirement, defined as the combination of
pension income with the participation in paid work, is highly dependent on
individual pension income and the resources available in the broader context.
We elaborate on the three main findings below.
First, this study showed a considerable variety in the rates of paid work
after retirement across the 16 European countries under study. Based on the
harmonized cross-national data, we conclude that working after retirement is
exceptional in some countries, while it is rather common in other countries.
Particularly these latter countries confirm the suggestion made in recent
retirement research that retirement is no longer an abrupt and all-absorbing
state (Beehr & Bennett, 2015).
F
igure A
F
igure B
-0.2
0
0.2
0.4
0.6
0.8
1
45678910
Relaonship (Cohen's d)
Individual pension income (ln)
Rerees with a partner
low gdp
mean gdp
high gdp
-0.2
0
0.2
0.4
0.6
0.8
1
45678910
Relaonship (Cohen's d)
Individual pension income (ln)
Rerees without a partner
low gdp
mean gdp
high gdp
Figure 4. The impact of work in retirement on life satisfaction by the level of
individual pension income and gross domestic product per capita (based on Table 3,
Model C).
Dingemans and Henkens 15
Second, this study has revealed one of the boundary conditions that may
strengthen or weaken the relationship between working after retirement and
life satisfaction. Following our expectations derived from a resources per-
spective, we found that among retirees with low pension income, those still
working in a paid job report higher levels of life satisfaction than full-time
retirees. This difference is smaller among retirees with higher pension
incomes. Among retirees with the highest pension incomes paid work did
not seem to make any difference to life satisfaction. It seems that for retirees
with a low pension income, working after retirement is a way to supplement
their (financial) resources and, as such, to help make ends meet (Cahill,
Giandrea, & Quinn, 2017; Yang, 2011). This does not necessarily mean that
retirees in paid employment have a higher life satisfaction after retirement in
comparison with their preretirement life satisfaction. Panel research in the
Netherlands has shown that a financial motive to work after retirement is
associated with a decline in life satisfaction over the retirement transition
(Dingemans & Henkens, 2014). The current study adds to this knowledge
and finds that despite the possibility of declining life satisfaction levels over
the retirement transition, retirees with low levels of individual pension
income with a paid job are better off—financially and psychologically—than
retirees in the same position without a paid job.
Third, in the examination of the relationship between working after retire-
ment and life satisfaction, we not only investigated the role of individual
financial resources but also accounted for resources that could be available in
the broader context. At the household level, we found that the presence of a
partner was important. The relationship between paid work and life satisfac-
tion was stronger for low pension income groups who did not have a partner
than for retirees in low pension income groups who did have a partner. In
addition, at the country level, the material wealth of the country (indicated by
GDP) was important in explaining the relationship between paid work and
life satisfaction. We found that working after retirement was strongly related
to life satisfaction among the low pension income groups, particularly in
relatively poor countries. In contrast, paid employment is less important for
the life satisfaction of retirees with low individual pension income in rich
countries. Seemingly, rich countries may have stronger labor markets, better
health-care systems, and more robust social safety nets to deal with lower
levels of individual financial resources. Further research may aim to look at
the impact of these different macro-level factors in understanding the rela-
tionship between paid work and satisfaction with life among retirees.
Theoretically, our study strongly indicates that working after retirement
seems to have different meanings for various groups of retirees and across
16 Research on Aging XX(X)
countries (Alcover et al., 2014). To date, the view on working after retire-
ment in the literature is mainly based on samples in rich Western countries
and has been rather positive (e.g., Zhan et al., 2009). Postretirement work is
often framed as a voluntary choice—a choice on how to spend your leisure
time in retirement. Our study results suggest that work may have a different
connotation when less resources are available to the individual and in the
context as a whole. In these cases, work after retirement is necessary to
satisfy basic needs. However, only about one in six retirees in the lowest
percentile of pension income in our sample participated in paid work, which
suggests that the others may suffer from poor financial conditions in retire-
ment. Even though working after retirement could be a way to improve their
material and psychological welfare, recent research shows that many do not
have access to paid employment (Dingemans, Henkens, & Van Solinge,
2016) partly because many employers are reluctant to recruit them (Oude
Mulders, Henkens, & Schippers, 2017).
The current study is not without limitations. First, although the SHARE
project has a longitudinal study design, the number of transitions to working
after retirement between the waves of data collection was too low. We did
not have information on preretirement life satisfaction for a large share of the
respondents and, therefore, we were not able to investigate changes in life
satisfaction over time. Second, while our results are based on a substantial
number of European countries, from a statistical perspective, the number of
countries is low (Bryan & Jenkins, 2013). As a result, we were limited in the
macro-level indicators that we could include in our models. Finally, the use
of existing data has its limitations. For instance, due to the specific answer
formats and routing in the questionnaire, we were unable to account for
individuals self-report on their work and retirement status in operationalizing
working after retirement. Additionally, we did not have information on
motivations to work after retirement, while previous research has shown that
it matters for life satisfaction whether retirees work for enjoyment or finan-
cial reasons (Dingemans & Henkens, 2014). Apart from motivations, future
research could focus on working conditions in the postretirement job itself.
Unfavorable working conditions and participation in large full-time jobs may
not be a matter of preference but be a forced choice, having negative con-
sequences for late life health and well-being (Lain & Vickerstaff, 2014).
The numbers of retirees in the Western world will increase substantially in
the coming decades. A key challenge of governments is to keep their pension
systems financially sustainable and at the same time provide retirees with
sufficient pension incomes. Expectations are, however, that for many older
workers, financial prospects in retirement will be poor. Public policy might,
Dingemans and Henkens 17
on the one hand, be aimed at increasing pension coverage and individual
savings over the life course. On the other hand, this study clearly shows that
for low pension income retirees, access to the labor force is crucial for their
life satisfaction. This implies that policy may put more effort in removing
barriers for employment among financially vulnerable groups of retirees. As
Lain and Vickerstaff (2014) argue, finding paid work may be difficult for
certain groups of retirees because many employers are reluctant to hire them
(Oude Mulders et al., 2017). It might be difficult to find a right balance in
policies aimed at increasing pension coverage and opening the labor force for
retirees. Moreover, there may be also negative individual health implications
to continued working. As such, working after retirement may thus not be seen
as the simple solution to prevent poverty after retirement.
Authors’ Note
This article uses data from Survey of Health, Ageing and Retirement in Europe
(SHARE), Waves 2, 4, and 5 (DOIs: 10.6103/SHARE.w2.500, 10.6103/
SHARE.w4.500, 10.6103/SHARE.w5.500). See Borsch-Supan et al. (2013) for meth-
odological details.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research,
authorship, and/or publication of this article: This work was supported by a VICI
Innovational Research Grant awarded to Ke`ne Henkens by the Netherlands Organiza-
tion for Scientific Research (NWO) and an Individual Research Grant awarded to Ellen
Dingemans by the Network for Studies on Pensions, Aging and Retirement (Netspar).
The SHARE data collection has been primarily funded by the European Commission
through FP5 (QLK6-CT-2001-00360), FP6 (SHARE-I3: RII-CT-2006-062193, COM-
PARE: CIT5-CT-2005-028857, SHARELIFE: CIT4-CT-2006-028812), and FP7
(SHARE-PREP: N211909, SHARE-LEAP: N227822, SHARE M4: N261982).
Additional funding from the German Ministry of Education and Research, the U.S.
National Institute on Aging (U01_AG09740-13S2, P01_AG005842, P01_AG08291,
P30_AG12815, R21_AG025169, Y1-AG-4553-01, IAG_BSR06-11, OGHA_04-064),
and from various national funding sources is gratefully acknowledged (see www.share-
project.org).
Supplemental Material
Supplemental material for this article is available online.
18 Research on Aging XX(X)
References
Alcover, C. M., Topa, G., Parry, E., Fraccaroli, F., & Depolo, M. (2014). Bridge
employment. A research handbook. New York, NY: Routledge.
Beehr, T. A., & Bennett, M. M. (2015). Working after retirement: Features of bridge
employment and research directions. Work, Aging and Retirement,1, 112–128.
doi:10.1093/workar/wau007
Bennett, J., & Mo¨hring, K. (2015). Cumulative (dis)advantage? The impact of labour
market policies on late career employment from a life course perspective. Journal
of Social Policy,44, 213–233. doi:10.1017/S0047279414000816
Be´renger, V., & Verdier-Chouchane, A. (2007). Multidimensional measures of well-
being: Standard of living and quality of life across countries. World Development,
35, 1259–1276. doi:10.1016/j.worlddev.2006.10.011
Borsch-Supan, A., Brandt, M., Hunkler, C., Kneip, T., Korbmacher, J., Malter, F., ...
Zuber, S. (2013). Data resource profile: The Survey of Health, Ageing and Retire-
ment in Europe (SHARE). International Journal of Epidemiology,42, 992–1001.
doi:10.1093/ije/dyt088
Bryan, M. L., & Jenkins, S. P. (2013). Regression analysis of country effects using
multilevel data: A cautionary tale. University of Essex, Colchester, United King-
dom: Institute for Social and Economic Research.
Cahill, K. E., Giandrea, M. D., & Quinn, J. F. (2017). To what extent is gradual
retirement a product of financial necessity? Work, Aging and Retirement,3,
25–54. doi:10.1093/workar/waw027
Damman, M., Henkens, K., & Kalmijn, M. (2015). Missing work after retirement:
The role of life histories in the retirement adjustment process. The Gerontologist,
55, 802–813. doi:10.1093/geront/gnt169
De Luca, G. (2018). 14.3 Imputations: SHARE release guide 6.1.1. (pp. 42–47).
Diener, E., & Diener, M. (2009). Cross-cultural correlates of life satisfaction and self-
esteem. In E. Diener (Ed.), Culture and well-being. The collected works of Ed
Diener (pp. 71–97). Dordrecht, the Netherlands: Springer Netherlands.
Dingemans, E., & Henkens, K. (2014). Involuntary retirement, bridge employment,
and satisfaction with life: A longitudinal investigation. Journal of Organizational
Behavior,35, 575–591. doi:10.1002/job.1914
Dingemans, E., Henkens, K., & Van Solinge, H. (2016). Access to bridge employ-
ment: Who finds and who does not find work after retirement? The Gerontologist,
56, 630–640. doi:10.1093/geront/gnu182
Dingemans, E., Henkens, K., & Van Solinge, H. (2017). Working retirees in Europe:
Individual and societal determinants. Work, Employment and Society,31,
972–991. doi:10.1177/0950017016664677
Dingemans and Henkens 19
Dorfman, L. T., & Rubenstein, L. M. (1993). Paid and unpaid activities and retire-
ment satisfaction among rural seniors. Physical & Occupational Therapy in Ger-
iatrics,12, 45–63.
Easterlin, R. E. (2005). Building a better theory of well-being. In L. Bruni & P. L.
Porta (Eds.), Economics and happiness: Framing the analysis. New York, NY:
Oxford University Press.
Eurostat. (2013). Income pooling and equal sharing within the household—What can
we learn from the 2010 EU-SILC module? Luxembourg, Europe: Publications
Office of the European Union.
Giandrea, M. D., Cahill, K. E., & Quinn, J. F. (2009). Bridge jobs: A comparison
across cohorts. Research on Aging,31, 549–576. doi:10.1177/0164027509337195
Hershey, D. A., Henkens, K., & Van Dalen, H. P. (2010). What drives retirement
income worries in Europe? A multilevel analysis. European Journal of Ageing,7,
301–311. doi:10.1007/s10433-010-0167-z
Hobfoll, S. E. (1989). Conservation of resources: A new attempt at conceptualizing
stress. American Psychologist,44, 513–524. doi:10.1037/0003-066X.44.3.513
Hobfoll, S. E. (2002). Social and psychological resources and adaptation. Review of
General Psychology,6, 307–324. doi:10.1037/1089-2680.6.4.307
Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not
emotional well-being. Proceedings of the National Academy of Sciences of the
United States of America,107, 16489–16493. doi:10.1073/pnas.1011492107
Kessler, R. C., Turner, J. B., & House, J. S. (1988). Effects of unemployment on
health in a community survey: Main, modifying, and mediating effects. Journal of
Social Issues,44, 69–85. doi:10.1111/j.1540-4560.1988.tb02092.x
Kim, S., & Feldman, D. C. (2000). Working in retirement: The antecedents of bridge
employment and its consequences for quality of life in retirement. The Academy of
Management Journal,43, 1195–1210.
Lain, D., & Vickerstaff, S. (2014). Working beyond retirement age: Lessons for
policy. In S. Harper & K. Hamblin (Eds.), International handbook on ageing and
public policy (pp. 242–255). Cheltenham, England: Edward Elgar.
Luoh, M. C., & Herzog, A. R. (2002). Individual consequences of volunteer and paid
work in old age: Health and mortality. Journal of Health and Social Behavior,43,
490–509.
Maestas, N. (2010). Back to work. Expectations and realizations of work after retire-
ment. The Journal of Human Resources,45, 718–748.
Malter, F., & Borsch-Supan, A. (2015). SHARE Wave 5: Innovations & metho-
dology. Munich, Germany: MEA, Max Planck Institute for Social Law and
Social Policy.
Markides, K. S., & Martin, H. W. (1979). A causal model of life satisfaction among
the elderly. Journal of Gerontology,34, 86–93. doi:10.1093/geronj/34.1.86
20 Research on Aging XX(X)
Mo¨hring, K. (2012). The fixed effect as alternative to multilevel analysis for cross-
national analyses (GK SOCLIFE Working Papers Series) (pp. 1–19). Cologne,
Germany: University of Cologne.
Muffels, R., Skugor, D., & Dingemans, E. (2014). If money does not buy much
happiness, what does? A multilevel analysis on the impact of absolute and relative
income, social values, and modernization on subjective well-being in Europe. In
L. Halman & W. Arts (Eds.), Value contrasts and consensus in present-day
Europe. Painting Europe’s moral landscapes (pp. 375–401). Leiden, the Nether-
lands: Koninklijke Brill N.V.
Organisation for Economic Cooperation and Development (OECD). (2017). Pensions
at a glance 2017: OECD and G20 indicators. Paris, France: Author.
Oishi, S., Diener, E., Lucas, R. E., & Suh, E. M. (1999). Cross-cultural variations in
predictors of life satisfaction, perspectives from needs and values. Personality and
Social Psychology Bulletin,25, 980–990. doi:10.1177/01461672992511006
Oude Mulders, J., Henkens, K., & Schippers, J. (2017). European top managers’ age-
related workplace norms and their organizations’ recruitment and retention prac-
tices regarding older workers. The Gerontologist,57, 857–866.
Parry, E., & Bown Wilson, D. (2014). Career transitions at retirement in the United
Kingdom: Bridge employment or continued progression? In C. M. Alcover, G.
Topa, E. Parry, F. Fraccaroli, & M. Depolo (Eds.), Bridge employment. A reserach
handbook (pp. 138–153). New York, NY: Routledge.
Quine, S., Wells, Y., De Vaus, D., & Kendig, H. (2007). When choice in retirement
decisions is missing: Qualitative and quantitative findings of impact on well-
being. Australasian Journal of Ageing,26, 173–179. doi:10.1111/j.1741-6612.
2007.00251.x
Rice, M. E., & Harris, G. T. (2005). Comparing effect sizes in follow-up studies: ROC
area, Cohen’s d, and r.Law and Human Behavior,29, 615–620. doi:10.1007/
s10979-005-6832-7
Settersten, R. A. (2003). Propositions and controversies in life-course scholarship. In
R. A. Settersten (Ed.), Invitation to the life course. Toward new understanding of
later life (pp. 15–48). New York, NY: Baywood.
Shultz, K. S., Morton, K. R., & Weckerle, J. R. (1998). The influence of push and pull
factors on voluntary and involuntary early retirees’ retirement decision and adjust-
ment. Journal of Vocational Behavior,53, 45–57. doi: 10.1006/jvbe.1997.1610
Szinovacz, M. E., & Davey, A. (2005). Retirement and marital decision making:
Effects on retirement satisfaction. Journal of Marriage and Family,62,
387–398. doi:10.1111/j.0022-2445.2005.00123.x
United Nations Statistics Division. (2016). Gross domestic product per capita.
Retrieved from http://www.un.org/esa/sustdev/natlinfo/indicators/methodology_
sheets/econ_development/gdp_percapita.pdf
Dingemans and Henkens 21
Yang, Y. (2011). No way out but working? Income dynamics of young retirees in
Korea. Ageing & Society,31, 265–287. doi: 10.1017/S0144686X1000084X
Zhan, Y., Wang, M., Liu, S., & Shultz, K. S. (2009). Bridge employment and retirees’
health: A longitudinal investigation. Journal of Occupational Health Psychology,
14, 374–389. doi:10.1037/a0015285
Author Biographies
Ellen Dingemans is a postdoctoral researcher at the Netherlands Interdisciplinary
Demographic Institute (NIDI-KNAW) and is affiliated to the University Medical
Center Groningen (UMCG-RUG). Her dissertation research focused on the determi-
nants and consequences of bridge employment in the Netherlands and Europe. Her
main research interests include retirement, postretirement employment, and well-being.
Ke
`ne Henkens is a head of the Work & Retirement Department of the Netherlands
Interdisciplinary Demographic Institute (NIDI-KNAW) and is a professor of Sociol-
ogy of Retirement at the University of Amsterdam (UvA) and the University Medical
Center Groningen (UMCG-RUG). He has published extensively on issues regarding
retirement and the aging workforce.
22 Research on Aging XX(X)
