econstor

A Service of

zbw
Leibniz-Informationszentrum
Wirtschaft

Make Your Publications Visible.

Leibniz Information Centre
for Economics

Piper, Alan; Jackson, Ian

Working Paper
She's leaving home: A large sample investigation of
the empty nest syndrome

SOEPpapers on Multidisciplinary Panel Data Research, No. 910

Provided in Cooperation with:

German Institute for Economic Research (DIW Berlin)

Suggested Citation: Piper, Alan; Jackson, Ian (2017) : She's leaving home: A large sample
investigation of the empty nest syndrome, SOEPpapers on Multidisciplinary Panel Data
Research, No. 910, Deutsches Institut für Wirtschaftsforschung (DIW), Berlin

This Version is available at:

http://hdl.handle.net/10419/162583

Standard-Nutzungsbedingungen: Terms of use:

Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Documents in EconStor may be saved and copied for your
Zwecken und zum Privatgebrauch gespeichert und kopiert werden. personal and scholarly purposes.

Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle You are not to copy documents for public or commercial
Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich purposes, to exhibit the documents publicly, to make them
machen, vertreiben oder anderweitig nutzen. publicly available on the internet, or to distribute or otherwise
use the documents in public.
Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen
(insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, If the documents have been made available under an Open
gelten abweichend von diesen Nutzungsbedingungen die in der dort Content Licence (especially Creative Commons Licences), you
genannten Lizenz gewährten Nutzungsrechte. may exercise further usage rights as specified in the indicated
licence.

www.econstor.eu
 The German
Socio-Economic
Panel study

910

2017
SOEPpapers
on Multidisciplinary Panel Data Research

SOEP — The German Socio-Economic Panel study at DIW Berlin 910-2017

She’s leaving home: a large

sample investigation of the
empty nest syndrome
Alan Piper and Ian Jackson
SOEPpapers on Multidisciplinary Panel Data Research at DIW Berlin

This series presents research findings based either directly on data from the German Socio-
Economic Panel study (SOEP) or using SOEP data as part of an internationally comparable
data set (e.g. CNEF, ECHP, LIS, LWS, CHER/PACO). SOEP is a truly multidisciplinary
household panel study covering a wide range of social and behavioral sciences: economics,
sociology, psychology, survey methodology, econometrics and applied statistics, educational
science, political science, public health, behavioral genetics, demography, geography, and
sport science.

The decision to publish a submission in SOEPpapers is made by a board of editors chosen

by the DIW Berlin to represent the wide range of disciplines covered by SOEP. There is no
external referee process and papers are either accepted or rejected without revision. Papers
appear in this series as works in progress and may also appear elsewhere. They often
represent preliminary studies and are circulated to encourage discussion. Citation of such a
paper should account for its provisional character. A revised version may be requested from
the author directly.

Any opinions expressed in this series are those of the author(s) and not those of DIW Berlin.
Research disseminated by DIW Berlin may include views on public policy issues, but the
institute itself takes no institutional policy positions.

The SOEPpapers are available at

http://www.diw.de/soeppapers

Editors:
Jan Goebel (Spatial Economics)
Martin Kroh (Political Science, Survey Methodology)
Carsten Schröder (Public Economics)
Jürgen Schupp (Sociology)

Conchita D’Ambrosio (Public Economics, DIW Research Fellow)

Denis Gerstorf (Psychology, DIW Research Director)
Elke Holst (Gender Studies, DIW Research Director)
Frauke Kreuter (Survey Methodology, DIW Research Fellow)
Frieder R. Lang (Psychology, DIW Research Fellow)
Jörg-Peter Schräpler (Survey Methodology, DIW Research Fellow)
Thomas Siedler (Empirical Economics, DIW Research Fellow)
C. Katharina Spieß (Education and Family Economics)
Gert G. Wagner (Social Sciences)
ISSN: 1864-6689 (online)
German Socio-Economic Panel (SOEP)
DIW Berlin
Mohrenstrasse 58
10117 Berlin, Germany

Contact: soeppapers@diw.de
She’s leaving home: a large sample investigation of the empty nest syndrome
Alan Piper* and Ian Jackson
* Corresponding author, Europa-Universität Flensburg, Germany

Abstract
This study considers life satisfaction in relation to the empty nest syndrome, which is a
situation where there are feelings of loss or loneliness for mothers and/or fathers following
the departure of the last child from the parental home. In particular, the investigation
considers the significance of Identity Economics when applied to parents experiencing a
reduction in well-being following an extended period of child-rearing. The origins of the
empty nest syndrome are first considered briefly before conducting an economic analysis of
life satisfaction using the German Socio-Economic Panel. Our particular focus is the change
in the subjective well-being of the individuals who become empty nesters, taking advantage
of the richness of this dataset. As a result, this is the first large sample economic analysis of
its kind to use identity to evaluate the effects of becoming “empty nest” parents in a
systematic way.
JEL Codes
D64, I31
Acknowledgments
We would like to thank Nick Adnett, Clemens Hetschko, Geoff Pugh and the participants of
the Internationales Institut für Management und ökonomische Bildung informal research
seminar series (Europa-Universität Flensburg) for their helpful comments on an earlier draft
of this paper. We also thank all of the people, from various walks of life, who have shared
their experience with us regarding this issue. The German Socioeconomic panel is a
wonderful source of data, and was made available by the German Socio-Economic Panel Study
(SOEP) at the German Institute for Economic Research (DIW), Berlin. The usual disclaimer applies.

Date
10 May 2017

Alan Piper (corresponding author): Internationales Institut für Management und ökonomische
Bildung, Europa-Universität Flensburg. Munketoft 3B, Flensburg, 24937, Deutschland Email:
alan.piper@uni-flensburg.de Ian Jackson: School of Business, Leadership and Economics,
Staffordshire University, Leek Road, Stoke-on-Trent, ST4 2DF, United Kingdom E-mail:
i.jackson@staffs.ac.uk

1
 1. Introduction
“Standing alone at the top of the stairs;
She breaks down and cries to her husband
Daddy, our baby’s gone.”
She’s Leaving Home, The Beatles (1967) 1

“Since she was born, I’ve wanted my daughter to have everything I didn’t have – a happy,
stable family, access to books and art, a good education, a beautiful home – but this quest
has become my whole world. I get upset if she’s had a bad day at school. Last week I broke
down and cried at the thought of her leaving home when she’s eighteen. That’s eleven years
away.”
Clothes, Clothes, Clothes. Music, Music, Music. Boys, Boys, Boys. Viv Albertine (2014: p319,
emphasis added)

Fifty years ago, The Beatles sang movingly about what has since been termed the “Empty
Nest Syndrome” (ENS); that is the phenomenon of a parent or parents who, having raised
children, find themselves bereft when the last or only child leaves home. The subject of
many newspaper articles, magazine features, blog posts as well as being discussed
elsewhere in popular culture (for example Albertine, 2014 and Godber, 2016), ENS was a
term first discussed in relation to families during the 1960s and thereafter it has entered
common usage in the English language (Dodd, 2011). Over the last generation or so it has
been applied loosely to the feelings of loss and loneliness experienced by mothers and/or
fathers when grown-up children leave the parental home, whether this is potentially
temporary, for instance to attend university or possibly permanently, to establish a separate
home.
To our knowledge, the notion of the ENS has not been studied previously in the specific
context of economic inquiry. If we accept Marshall’s broad definition of economics as being
about “the study of men as they live and move and think in the ordinary business of life”
(1890, p 773) then the ENS is a worthwhile area of economic study. This is because
becoming an empty-nester may well have profound effects on the lives of such individuals
both economically and emotionally. Indeed, this aspect of the parenting life-cycle has been
a focus of medical and sociological studies, where the difficulties of this transition from a
health and societal perspective have been discussed (Hiedemann et al. 1998). However, this
literature often relies on small-scale regional primary data samples (for example, an inland
mountainous area of China) (Liu and Guo 2008), or is qualitative with a commensurate small
sample size (Spence and Lonner 1971). In contrast, this investigation is, to our knowledge,

1
“She’s Leaving Home” (1967) by the Beatles is a poignant song made all the more powerful by its layered
narrative and complex harmonies that evoke loss and isolation, respectively for parent and child. The
underlying message of the song is that parenting brings both joy and sorrow but that the parental relationship
with a child is by definition both unequal and asymmetric that furthermore alters towards maturity. An equally
powerful poem was written a decade earlier by the Poet Laureate Cecil Day-Lewis (1956) about his son going
to school, “like a winged seed loosened from its parent stem.”

2
the first large sample investigation of the phenomena taking advantage of thirty-one
consecutive years of a nationally representative dataset, the German Socio-Economic Panel
(SOEP). Our particular focus is the change in the subjective well-being of the individuals who
become empty nesters. As a result, while ENS has no specific medical diagnosis, there can
be a meaningful economic analysis of empty-nesters in terms of the effects on life
satisfaction using parental identity as the basis for the investigation.

This paper is structured as follows. The current academic literature is discussed in section 2
taking in a consideration of the relevant economic theory, including the concept of identity
(Akerlof and Kranton 2010). A related brief discussion of the biological origins of ENS is
found in Appendix 1. Section 3 describes the data and methodology. Section 4 presents the
results. Section 5 contains a discussion of the results, as well as highlighting the limitations
of the investigation and presenting suggestions for future research. Finally, section 6 offers
concluding remarks.

2. The Empty Nest Syndrome literature and theoretical discussion

This section has two distinctive elements. Firstly, a general economics approach, which
explains why an analysis of ENS is valid within this sphere of enquiry. Secondly, a specific
focus on the economics of identity. This section is complemented by a discussion of the
biological origins of the empty nest syndrome, suggesting a “natural”, primal sadness
common to humans. In other words, there may be possible biological reasons for reduced
well-being following the departure of the last child from the parental home (see Appendix
1).

2.1 The Empty Nest Syndrome literature: extensions in economics

The antecedents of the ENS in a human context are rooted in the evolving organisational
structure of households as well as societal priorities in relation to the allocation of scarce
economic resources (see Zahidi, 2012 for a discussion on ENS in relation to women and
aging). In particular, as life expectancy increases and the birth rate declines, principally in
developed countries, then the incidence of ENS may become more prevalent since humans
can live significantly beyond reproductive age, especially post-menopausal women (see
Silverstein and Sayre, 2009).
At a macroeconomic level, aggregate economic activity rates of the workforce is viewed as
important, especially in China where the infamous “one child policy” means the parents can
return to work post-family much sooner than countries with larger family units containing
more children (China Development Research Foundation 2014) and hence there is a policy
imperative to study the labour force in early middle-age rather than at other times such as
closer to retirement. Nevertheless, more importantly for our well-being investigation is the
microeconomic context.

3
However, it should be emphasized that ENS 2 is not a formal medical condition and in the
social sciences it is often reduced to a convenient statistical label that is pre and/or post
empty nest (Cooper and Gutmann, 1987). Notwithstanding, it is symptomatic of informal
feelings and psychological fears ranging across a spectrum from sorrow to depression. On
the one hand, any reduction in well-being may be due to profound anxiety or a morbid
apprehension regarding a potential mid-life crisis. As a result, ENS forms an integral part of
parental well-being albeit at a specific stage in life and hence a rich source of investigation
through economic analysis. On the other hand, there may be an increase in well-being post-
empty nest due to greater freedoms for the parent psychologically, financially and
emotionally in full knowledge of a job well done. Thereby, becoming an empty-nester can
yield potentially contradictory outcomes, which may be dependent on the identity of the
parent.
The contemporary academic literature on ENS is sporadic with virtually no major research
specifically related directly to economics. However, a potentially fruitful area for economic
analysis is to link the whole notion tangentially to the economics of the family pioneered by
Gary Becker (1981). Although Becker does not refer to ENS directly, his approach is based
on the “productive complementarities” of the parents, which is why one parent may
specialise in earning income and the other parent may specialise in childrearing. This
methodology has been challenged more recently by Stevenson and Wolfers (2007) who
claim it is “consumption complementarities” that motivate family units (and not joint
production) and so households are organised to satisfy tastes and desires. Although Merrill
(2010) argues that the Stevenson and Wolfers approach does not wholly replace the original
Becker position, she does concede that this newer viewpoint is best applied to childless
couples as well as “empty nest” family units, where the children have become adults. This
outcome may imply that becoming an “empty nest” household could increase happiness,
because the same economic resources are being shared between fewer people in addition
to a sense of achievement once the children have grown-up all other things including
emotional feeling remaining equal. This possibility is returned to in the empirical analysis
below.
Additionally Merrill (2010) considers the economics of marriage using the families as
household firms, any economic analysis of ENS has to look further afield. One of the earliest
academic studies of the transition to the empty nest is by sociologists. Spence and Lonner
(1971) use intensive case studies of 27 women and find partial evidence of unhappiness as
these women are unprepared for life after children leave home. However, this study is not
only small-scale in nature; it is also unrepresentative with all the women from the same
homogenous grouping; namely a white, middle-class and metropolitan cohort. In

2
Nevertheless, ENS is a potentially strong indication of major change in the cycle of life. It is an indication of
the passing of time, which is especially significant for humans as the young take relatively long to nurture and
thereby each child consumes a substantial quantity of economic resources cumulatively. Hence, as with all
economic decisions and investment opportunities, “empty nest” parents can be subject to a cost-benefit
analysis. See Appendix 1 for some likely considerations.

4
psychology, Raup and Myers (1989, p 181) clarify the definition of ENS as “…a maladaptive
response to the post-parental transition, which is stimulated by reactions to loss...”.
Furthermore, using correlates of ENS over the post-war period, they found qualitative
evidence that full-time employed women are less susceptible to ENS and unemployed
women are the most at risk. More recently, Mitchell and Lovegreen (2009) use a mixed
methodological approach to assess a subsample of 316 parents from four ethnic
backgrounds and found only a minority reported ENS.
Fundamentally, while there is negligible analysis of the empty nest syndrome within the
wider economics literature, the effect of the change in circumstances on life satisfaction
such as becoming unemployed may offer close parallels. Becoming an empty nester can be
seen as similar to entering unemployment, because there is a profound change in life of the
affected person. For example Winkelmann and Winkelmann (1998), using the same dataset
as we do, famously demonstrated that non-pecuniary factors are far more important in
explaining the loss of well-being associated with becoming unemployed than any effects
from a reduction in income. Non-pecuniary effects matter, and Winkelmann and
Winkelmann consider the loss in well-being from unemployment being related to a loss of
self-esteem, the loss of social relationships and the change of identity within society. These
reasons put forward for the loss of well-being experienced by the newly unemployed are
likely to be similar to any losses experienced by the new empty nesters. The next subsection
in discusses the related change in identity of new empty nesters.

2.2 The Empty Nest Syndrome literature: identity economics

Given the broader mainly economic analysis discussed previously, ENS can be linked to
identity economics as developed by Akerlof and Kranton (2000). Being a full-time parent
(both mother and father) may be a choice of identity and an important economic decision
even though it may not always be made consciously (i.e. there is a natural, in-built instinct
to reproduce and care for children, Dawkins, 1978 p. 107).
The approach of identity can help to model the change from the pre and post-empty-nest,
which is a switch to a new status from a parent with dependent children to a parent of
grown-up independent children. To capture the well-being of the parental role, the utility
function should include the identity benefits gained from belonging to a group with similar
objectives. Hence, if ENS has any basis in life satisfaction then transitioning from a parent
with caring responsibilities for dependent children to an empty-nester will reduce a person’s
well-being. Following the pioneering work of Hetschko et al. (2014) on the identity of the
unemployed entering retirement, then the method involves labelling parents in a social
category before becoming empty nesters (Pre-EN) and afterwards (Post-EN). This means
there is a potentially strong identity with the parenting role and that any change viewed
from the perspective of identity means that the resultant sadness, melancholy or even
depression is logical and understandable. Thus any alternation in well-being is assumed to
reflect the change in identity. Hence, if a person identifies strongly with being a parent (Pre-
EN state) and subsequently the child or children leaves home (Post-EN state) then the

5
subjective well-being will be reduced irrespective of any financial gain that may accrue. One
potential reason for this reduction may come from the change in identity that affects social
relationships. In such a situation, the new empty nester may no longer have current
membership benefits previously derived from being a part of a network of other people
with children. In other words, a person may have a long-term and deeply-held affiliation
with the social category of being a parent of dependent children; an affiliation which is
suddenly no longer valid. As a result, identity-based behaviour and preferences will change
for the “empty-nester”. This is just one example highlighting how the change in identity
could have wide-ranging implications for an individual’s life and well-being once the identity
is internalized (Akerlof and Kranton 2000). Other closely related examples may include the
bereavement effects of becoming widowed and the life changing effects of being diagnosed
with illnesses such as diabetes or alcoholism.
In essence, the concept of identity directs individual behaviour and helps to explain
economic outcomes. In turn, this outcome determines the amount of resources a person is
prepared to invest in being a parent; and ultimately it can also adversely affect parental
identity when a person becomes an empty nester. In other words, for some people, the loss
of identity as a parent once becoming an “empty-nester” can have profound effects leading
to a reduction in well-being. Notwithstanding, for others it may be that the opposite effect
as newfound freedom (and more economic resources) may increase well-being. This
potentially contradictory outcome has implications for this research, not least as “empty-
nest” parents adjust to their revised economic situation, especially in the short term.
For most family units, there is little or no choice regarding the nest becoming empty; hence
the outcome can be viewed as binary and a function of time. That is, a family unit is formed
and children enter (classification one) and then they subsequently depart at a later date
(classification zero). Hence, becoming an “empty nester” is an example of low frequency
change (Akerlof and Kranton, 2010, p 126) as well as helping to understand the non-
pecuniary explanations of economic behaviour (Akerlof and Kranton 2000, p 749.
Hence, the extent of the ENS may depend on the determinants of identity such as the
commitment to parenting a child, although the outcome is largely self-identified (Fox and
Bruce, 2001, p 396). In general, those people that identify themselves most as a full-time
parent (or as a main carer of children) may be more susceptible to ENS within a group
identification; those people that individually identify themselves most as a working parent
(for instance where a career or a vocation is equally or more important than child-raising)
may be less susceptible to ENS or not susceptible at all. In terms of recent economic theory,
the latter category (i.e. individuals who identify themselves as working parents) is closer to
the ideas of Stevenson and Wolfers that family units are based more on “consumption
complementarities”; and the former category (i.e. self-identification as a full-time parent) is
closer to the Becker ideas of “production complementarities”. In other words, modern
parent-children relationships are developing to be more about maximising the benefits of
sharing services such as childcare through common interests; and not only about minimising
the costs of producing the childcare through the division of labour. For our subsequent

6
analysis this means that there might be differences in the effect of entering the empty-nest
on life satisfaction by labour force status. This view is examined below.
Furthermore, while traditional economic theory suggests a purely monetary-based
approach to the life-cycle of the family, the notion of identity can provide additional
insightful aspects because it focuses on the various outcomes created through social
differences in economic behaviour (Akerlof and Kranton 2000, p 748). Empirical analysis can
investigate whether mothers are affected more than fathers (i.e. role of gender); introduce
whether the parents are employed or unemployed (i.e. income levels); account for
technology such as lower cost communication (i.e. long term effects) and even test to see if
becoming an empty-nester gives the parent a new lease of life and therefore establish if the
last child leaving has a positive effect.
In summary, ENS can be developed primarily from the situation of the identity and the
status of the parent or parents. For example, the low-income and unemployed may be
differently affected by ENS than the employed and financially well-off, given the potential
reduction of a liquidity constraint and the actual and/or potential household income (i.e.
work-force status); widowed, divorced and separated parents may be more affected by ENS,
if they are the primary child-care provider, as there is a potential loss of purpose (marital
status); better health and well-being of the parent or parents could mean that they
withstand any adverse effects of children leaving home (health status); and finally much
improved and almost costless communication may well reduce ENS as parents can keep in
touch with their children, even if the offspring may have moved further away from the
parental home. All these areas contain potentially testable hypotheses with respect of
assessing ENS through life satisfaction. Thus, the empirical analysis below investigates the
representative sample as a whole, and then focuses on sub-samples of individuals.

3. Data and Methodology

This empirical investigation of the empty nest syndrome makes use of thirty-one
consecutive years of the German Socio Economic Panel survey, a rich longitudinal data set
replete with much individual socioeconomic information. A detailed description of this
survey is given by Wagner et al. (2007). The panel structure enables the identification of
empty nesters, which we define as individuals (either the head of the household, or the
partner of the head of the household) whose children have left the household within the
last year. Thus the investigated empty nesters are new empty-nesters, having become so
since the previous annual wave. As mentioned above, this enables the investigation of the
change in identity from a parent back to a partner (or, more generally, from a parent with
dependent children to a parent without dependent children) and may not have given empty
nesters the chance to get used to their new situation. In the dataset there are
approximately 1,806 such observations. As a result, the comparator group is the pre-empty
nesters who still have children in the household. Importantly, and in accordance with our
focus on identity change, individuals who have never had children or whose last child (or
children) has left the household more than a year previously are not in the sample.

7
Consequently, we have identified (new) empty nesters and pre-empty nesters Table 1
compares the two groups under comparison in the later empirical analysis.

Table 1 descriptive statistics: recent empty nesters compared with pre-empty nesters, SOEP 1985-
2014.
Empty Nesters Non Empty Nesters
Mean Std. Dev Mean Std. Dev.
Real Annual Income 25.81 27.29 24.47 28.20
Real Annual Household Income 49.05 37.13 49.47 36.96
Employed 0.59 0.49 0.60 0.49
Self-employed 0.10 0.30 0.07 0.26
Apprentice 0.00 0.06 0.00 0.07
Government employed 0.04 0.20 0.05 0.21
Unemployed 0.10 0.30 0.07 0.25
Not employed 0.07 0.26 0.15 0.36
Retired 0.07 0.26 0.02 0.12
In education 0.00 0.07 0.01 0.09
Married 0.46 0.50 0.84 0.37
Separated 0.28 0.45 0.02 0.13
Divorced 0.14 0.34 0.06 0.24
Widowed 0.03 0.18 0.01 0.10
Single 0.09 0.28 0.07 0.25
Male 0.62 0.49 0.46 0.50
Education: High School 0.62 0.48 0.62 0.49
Education: more than HS 0.16 0.37 0.20 0.40
Education: less than HS 0.21 0.41 0.19 0.39
Health: Very Good 0.08 0.27 0.10 0.30
Health: Good 0.41 0.49 0.49 0.50
Health: Satisfactory 0.32 0.47 0.29 0.46
Health: Less than Satisfactory 0.19 0.39 0.11 0.32
Age 45.63 9.75 39.71 7.92
Note: Apart from age (years) and the two income measures (thousands of euros, deflated by the CPI), all of the
variables are dummy variables. SOEP data used: Socio-Economic Panel (SOEP), data for years 1984-2014,
version 31, SOEP, 2016, doi: 10.5684/soepv31.

In many respects, and perhaps unsurprisingly, the recent empty nesters and pre-empty
nesters are somewhat alike. Both average real individual income and average real
household income, and many of the labour market status variables reveal similar patterns.
However, those not employed (but also neither unemployed nor retired) are more prevalent

8
in the pre-empty nest group, presumably reflecting the need to stay at home and care for
children. In contrast, the retired and unemployed are more prevalent in the new empty
nesters group. Our empty nesters are seemingly less healthy than the pre-empty nested.
Similarly, the age difference is noteworthy (and expected) with empty nesters being, on
average, five years older than those with children still in the household. The new empty
nesters are also more likely to report a marital status of separated or divorced than the
nested. This is potentially important for our empirical results and returned to in the next
section.
The SOEP survey offers another way of looking at differences between these two groups.
Since 1984, the survey has contained an open question asking individuals about their
worries and responses to question can also be used to highlight potential differences
between groups. A detailed discussion about using such questions for social science
investigations is provided by Rohrer et al. (2017). 3 With our two groups, it is notable that
the empty nesters worry considerably more about personal problems than non-empty
nesters. Further, there is evidence that empty nesters report more frequent worries about
unemployment (both in general, and own unemployment) but not youth unemployment.
Proportionally more pre-empty nesters report worrying about the health of relatives
(including children) than empty nesters.
As discussed above, we investigate the empty nest syndrome through the concept of life
satisfaction. Such a choice means that the impacts of becoming an empty nester can be
measured. The prism of life satisfaction or, more generally, subjective well-being is a
purposeful way for investigating Marshall’s idea of what economics is about (see
introduction), and is a currently popular area of enquiry within economics (and other
disciplines including psychology and sociology). The dependent variable comes from
individual responses in the SOEP to the following: ‘We would like to ask you about your
satisfaction with your life in general’, which is coded on a scale from 0 (completely
dissatisfied) to 10 (completely satisfied). Table 2 shows the distribution of responses for
both groups.

3
This specific SOEP question has been recently used by Wagner (2016), who presented information
demonstrating that voters for the right of centre political party Alternative for Deutschland consistently report
more worries than voters for other parties or no party; a result that substantially predates that party’s
formation in 2013, indicating that AfD supporters have, for a substantial amount of time, often had
considerably more worries than the rest of the population: an interesting finding.

9
Table 2 The distribution of life satisfaction for empty nesters and pre-empty nesters SOEP 1984-2014

Empty-Nest Pre-empty Nest

Life Satisfaction Count % Count %
0 19 1.05 558 0.37
1 23 1.27 485 0.32
2 54 2.99 1,587 1.04
3 108 5.98 3,687 2.42
4 100 5.54 5,011 3.29
5 297 16.45 16,985 11.16
6 261 14.45 16,424 10.79
7 376 20.82 34,304 22.54
8 374 20.71 47,129 30.97
9 121 6.70 19,018 12.50
10 73 4.04 6,992 4.59

Total 1,806 100.00 152,810 100,00

Note: Life satisfaction is positively coded with higher scores indicating higher life satisfaction. SOEP data used:
Socio-Economic Panel (SOEP), data for years 1984-2014, version 31, SOEP, 2016, doi: 10.5684/soepv31.

While the patterns are typical of most investigations of life satisfaction, Table 2 does seem
to show that empty-nesters report less satisfaction with life than those with a child or
children in the household supporting somewhat the hypothesis developed above. The mean
responses offer further support with empty nesters reporting on average 6.28 life
satisfaction, and pre-empty nesters 7.07. Regression analysis will investigate this
observation in more detail, controlling for socio-economic factors often found to be
important for life satisfaction (marriage, unemployment, health and so on). The descriptive
statistics above in table 1 highlight the importance of controlling for these variables, given
the differences between the two groups. The following section presents, and discusses,
results from multivariate regressions. In the regression analysis, the focus will be on the
whole sample, initially, and then subsamples where there are some interesting results. Our
main interest is on the coefficients obtained for the empty nest dummy variable, which
indicates (any) difference between the empty nesters and the pre-empty nesters.
These regressions are undertaken with both pooled ordinary least squares (OLS) and fixed
effects (FE) estimation techniques4. Here, particularly with the subsamples, there is
occasionally not enough ‘within’ variation for precise estimation with fixed effects
estimation and pooled OLS is often more informative with these smaller subsamples.
Generally, fixed effects estimation is preferred because of its well-known ability to control
for individual heterogeneity often important for life satisfaction. However, it is not ideal
because the coefficients obtained, which come from ‘within’ change of the particular

4
Random effects estimation, another possibility, is rarely supported in a well-being context; a Hausman test
confirms this with the particular equations estimated in this investigation. The fixed effects themselves are
statistically significant. A further possibility may have been difference and system GMM estimation, which can
both control for fixed effects and employ both between and within variation for estimation, though given our
main variable of interest and how these techniques’ generate internal instruments for coefficient estimation
purposes seems somewhat inappropriate.

10
individuals, should not be generalised to the wider population. Hence we report the
coefficients for both estimation techniques throughout.

4. Results

This section presents results from multivariate regression analysis, starting with the full
sample (table 3) before investigating the well-being effect of becoming an empty-nester on
specific subsamples (tables 4-6), in line with the above discussion of the literature and
theory. Thus, table 3 presents the overall full sample results. The first two columns include
both genders together, and are distinguished by the method used to obtain the coefficients
(i.e. pooled OLS and FE). The next two columns are coefficients for males only and the last
two for females only.

Table 3 Multivariate regression results for the life satisfaction of empty-nesters. SOEP 1985-2015
Dependent variable: Life Satisfaction (positively coded from 0 to 10)
(1) (2) (3) (4) (5) (6)
OLS FE OLS FE OLS FE
VARIABLES All All Male Male Female Female

Empty Nest -0.36*** -0.21*** -0.45*** -0.25*** -0.22*** -0.06

(0.041) (0.037) (0.054) (0.049) (0.066) (0.060)
Real Income (‘000s) 0.004*** 0.003*** 0.005*** 0.003*** 0.004*** 0.003***
(0.000) (0.000) (0.000) (0.000) (0.000) (0.001)
Self-employed -0.01 -0.03 -0.11*** -0.08** 0.12*** 0.00
(0.016) (0.026) (0.020) (0.037) (0.026) (0.037)
Government employed 0.18*** -0.05 0.20*** -0.04 0.16*** -0.05
(0.021) (0.053) (0.026) (0.082) (0.034) (0.070)
Not in labour Market 0.06*** -0.06*** -0.55*** -0.61*** 0.10*** -0.01
(0.014) (0.017) (0.059) (0.065) (0.016) (0.019)
In education 0.01 -0.07 -0.07 -0.20** 0.05 -0.01
(0.048) (0.049) (0.093) (0.094) (0.057) (0.059)
Unemployed -0.76*** -0.56*** -0.98*** -0.80*** -0.62*** -0.41***
(0.017) (0.019) (0.027) (0.029) (0.023) (0.025)
Retired -0.12*** -0.04 -0.12** -0.08 -0.15*** -0.07
(0.038) (0.055) (0.052) (0.071) (0.056) (0.084)
Military/Community -0.58 -0.65 0.15 -0.02 -1.09* -1.10
(0.471) (0.495) (0.723) (0.677) (0.620) (0.718)
Apprentice -0.25*** -0.05 -0.31*** -0.12 -0.21*** -0.01
(0.058) (0.058) (0.103) (0.102) (0.070) (0.071)
Male -0.21***
(0.010)
Age -0.05*** 0.005 -0.05*** 0.01 -0.05*** 0.001
(0.004) (0.007) (0.006) (0.011) (0.006) (0.010)
Age-squared 0.0005*** -0.0003*** 0.0005*** -0.0004*** 0.0005*** -0.0002**
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Married 0.31*** 0.16*** 0.10*** 0.02 0.42*** 0.25***
(0.017) (0.038) (0.028) (0.056) (0.021) (0.051)
Widowed -0.05 -0.25*** -0.25*** -0.40** 0.05 -0.13
(0.047) (0.092) (0.092) (0.165) (0.056) (0.114)

11
 Table 3 continued
(1) (2) (3) (4) (5) (6)
OLS FE OLS FE OLS FE
VARIABLES All All Male Male Female Female
Divorced -0.08*** 0.06 -0.18*** -0.12 -0.02 0.18***
(0.023) (0.047) (0.043) (0.079) (0.027) (0.060)
Separated -0.28*** -0.18*** -0.50*** -0.46*** -0.18*** -0.03
(0.032) (0.048) (0.063) (0.080) (0.038) (0.062)
Edu: High school 0.08*** -0.02 0.01 -0.16*** 0.11*** 0.08*
(0.013) (0.032) (0.020) (0.049) (0.016) (0.042)
Educ: Above high sch. 0.19*** 0.05 0.08*** -0.02 0.26*** 0.09*
(0.016) (0.042) (0.024) (0.068) (0.021) (0.054)
Very good health 2.34*** 1.32*** 2.37*** 1.32*** 2.30*** 1.32***
(0.019) (0.021) (0.028) (0.031) (0.025) (0.028)
Good health 1.72*** 1.03*** 1.78*** 1.06*** 1.65*** 1.00***
(0.014) (0.016) (0.021) (0.024) (0.019) (0.021)
Satisfactory health 0.96*** 0.63*** 1.01*** 0.67*** 0.90*** 0.60***
(0.015) (0.015) (0.022) (0.022) (0.020) (0.020)
Constant 6.44*** 6.20*** 6.42*** 5.81*** 6.50*** 6.97***
(0.103) (0.372) (0.152) (0.471) (0.136) (0.624)
Regional controls Yes Yes Yes Yes Yes Yes
Wave controls Yes Yes Yes Yes Yes Yes
Observations 123,050 123,050 55,523 55,523 67,527 67,527
R-squared 0.240 0.075 0.262 0.090 0.228 0.068
Number of people 24,988 11,268 13,720
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1. Reference categories: single, employed, poor
health, less than high school education. SOEP data used: Socio-Economic Panel (SOEP), data for years 1984-
2014, version 31, SOEP, 2016, doi: 10.5684/soepv31.

Table 3 demonstrates that becoming an empty-nester is associated with a considerable

reduction in life satisfaction, when compared to individuals who still have children in the
household. A finding that occurs after controlling for the standard variables employed by
most investigations within the ‘economics of life satisfaction’ area of inquiry: real individual
income, labour force status, marital status, age, health and education. 5 The region (one of
16 Länder) and year, not shown in the table, are also included as controls. Interestingly,
males appear to be affected more than females, the size of the coefficient is greater and
statistically different from zero with both OLS and FE. For females, only OLS estimation
results in a negative and statistically significant coefficient for the new empty-nesters. One
possibility for this finding for gender is that women have better innate coping skills (Tamres
et al. 2002).
Briefly, the other coefficients are in line with expectations formed by previous investigations
in the literature: unemployment is statistically significant, and negatively associated with life
satisfaction (Clark and Oswald 1994, Winkelmann and Winkelmann 1998)’ marriage is

5
Furthermore, all the results in the table are robust to using real household income, rather than individual
income.

12
positively associated with life satisfaction (Frey and Stutzer 2006, Qari 2014); and age and
age-squared generate the well-known U-shape between age and life satisfaction when OLS
is used but not fixed effects (not an unusual result when the ‘within’ movement of age and
wave is the same, see Piper (2015) for more information and a solution, and Blanchflower
and Oswald (2008) regarding age and life satisfaction more generally). Health is positively
and strongly associated with subjective well-being: a near regularity in the wider literature
(Dolan et al. 2008, Downward and Dawson 2016).
These overall coefficients demonstrate that becoming an empty-nester is a strongly
negative experience, with the overall coefficients about half that of being unemployed
(compared to the base category of being employed), which itself (as mentioned above) is a
well-known and understood negative phenomena in terms of life satisfaction. The results in
table 3 are general and, as suggested by the theoretical discussion previously, becoming an
empty-nester may be less, or more, impactful for certain groups. The precise reasons for
these subsample choices reflect the economic literature discussion above. However, a
specific discussion of these reasons will also briefly take place within the context of the
coefficients obtained for the empty nest variable for different subsamples. Standard
controls have been used in each case (including wave and region dummy variables), but are
not shown for brevity. Thus the only coefficients shown in the subsequent tables are those
for the new empty nesters. Table 4, the first subsample table, shows the coefficients for the
life satisfaction of empty nesters who are employed, unemployed and retired (in the first
three columns) and from empty nesters from two different time periods (last two columns).

Table 4 Multivariate regression results for the life satisfaction of empty-nesters, labour force status
and year. SOEP 1985-2015
Dependent variable: Life Satisfaction (positively coded from 0 to 10)
(1) (2) (3) (4) (5)
Empty Nest Employed Unemployed Retired Before 2000 From 2000

All, OLS -0.36*** -0.43*** -0.41** -0.33*** -0.38***

(0.050) (0.157) (0.183) (0.083) (0.046)
All, FE -0.21*** -0.31* -0.23 -0.16* -0.20***
(0.045) (0.174) (0.183) (0.083) (0.042)
Male, OLS -0.43*** -0.27 -0.21 -0.36*** -0.49***
(0.066) (0.225) (0.248) (0.109) (0.062)
Male, FE -0.23*** -0.11 -0.04 -0.16 -0.24***
(0.060) (0.257) (0.274) (0.109) (0.057)
Female, OLS -0.12 -0.83*** -0.35 -0.24* -0.21***
(0.085) (0.238) (0.278) (0.138) (0.076)
Female, FE -0.03 -0.60** -0.26 -0.03 -0.06
(0.079) (0.254) (0.257) (0.140) (0.068)
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1. Full controls from table 1, excepting labour
force (for models 1, 2, and 3). SOEP data used: Socio-Economic Panel (SOEP), data for years 1984-2014, version
31, SOEP, 2016, doi: 10.5684/soepv31.

The subsequent results discussion focuses initially on the results from the ‘all, OLS’ (both
genders together) estimation, and highlights interesting findings from both the FE analysis
and the individual genders estimations. With respect to the three labour force categories

13
there is a (slightly) bigger impact of becoming an empty nester in the past year for the
unemployed and retired than for the employed. This outcome is consistent with identity as
work could be a substitute for the time spent child-caring. Furthermore, there is some
evidence that females who are unemployed are particularly negatively affected by
becoming empty nesters. However, caution is necessary when considering the smaller
subsamples because of reduced numbers of new empty nesters (see Appendix 3 for details).
The last two columns demonstrate that becoming an empty nester seems to have a similar
impact on an individual’s life satisfaction before and from the year 2000. This is slightly
contrary to the customary expectation (and many blog posts on the subject) which reason
that, because the cost of communication has been severely reduced over time (and arguably
vanished in the later period of this sample), parents would have more contact with their
children helping to reduce somewhat any negative effects caused by becoming an empty
nester. This is not demonstrated by the results, and we speculate that it is even possible
that this decreased cost of communication change may have had a paradoxical result. When
the cost of communicating was higher, both in terms of time and money it was implicitly (or
explicitly) accepted that parent-child communication would not happen so often. However
when the costs have dropped or disappeared, this acceptance of limited communication,
may have been similarly reduced leading to a more keenly felt sense of disappointment
resulting from any lack of communication, and thus contributing to the decreased life
satisfaction of the new empty-nester. Similarly, the growth of social media enabling the
parent to see the child in photographs and videos (but not face to face) may be a further
cause for melancholia and unhappiness. 6
The next subsample investigation splits the sample based upon marital status, and table 5
displays the relevant new empty nest coefficients.

6
This speculation is enhanced when the year of the split is moved to a later one to better reflect the
development of communication software like Skype and social media: using before and after 2006 gives
approximately equivalent results though the size of the coefficients is generally larger for the more recent time
period (not shown, but available on request).

14
Table 5 Multivariate regression results for the life satisfaction of empty-nesters, marital status. SOEP
1985-2015
Dependent variable: Life Satisfaction (positively coded from 0 to 10)
(1) (2) (3) (4) (5)
Empty Nest Married Single Widowed Divorced Separated

All, OLS 0.00 -0.77*** -0.51** -0.65*** -0.58***

(0.059) (0.133) (0.144) (0.121) (0.114)
All, FE 0.01 0.34*** 0.24 -0.37*** -0.18
(0.053) (0.129) (0.289) (0.131) (0.237)
Male, OLS 0.00 -0.79*** 0.27 -0.57*** -0.88***
(0.082) (0.151) (0.425) (0.167) (0.147)
Male, FE -0.02 -0.29** 0.37 -0.36* 0.53
(0.073) (0.146) (0.592) (0.191) (0.378)
Female, OLS 0.03 -0.73*** -0.72** -0.66** -0.02
(0.084) (0.258) (0.306) (0.178) (0.228)
Female, FE 0.05 -0.47* 0.19 -0.38** -0.43
(0.076) (0.248) (0.332) (0.186) (0.380)
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1. SOEP data used: Socio-Economic Panel (SOEP),
data for years 1984-2014, version 31, SOEP, 2016, doi: 10.5684/soepv31.Controls are income, labour force
status, age, education, health, region and year.

When assessed by the all, OLS results (i.e. the first row), the coefficients are broadly in line
with expectations formed by the literature discussion above (in section 2). 7 The empty nest
syndrome has a substantial effect for those who are single, widowed, divorced and
separated, perhaps supporting Raup and Myers (1989) who state that women experience
becoming empty nesters differently if divorced since they have a different “identity and self-
esteem”. However, again it should be noted that some of the subsamples contain few
individuals. For example, in the sample generally there are few male widows, and only a
very small proportion of these enter the empty nest situation (see Appendix 3). In this case
– and, for similar reasons, in some others – a lack of statistically significant results is
unsurprising. Perhaps surprising is that married individuals do not seem to experience the
empty nest syndrome. We speculate that this is maybe because a married couple can take
solace from one another when they become empty-nesters and are consequently better
positioned to cope with the “new” normal such as caring for other (older) family members
and pursuing other interests previously deferred when raising children (see, for example,
Rubin 2008, p 51).
Finally, table 6 focuses on differences with respect to health status and income. For health
status, the split is straightforward: good health and above for the first subsample;
satisfactory health or worse for the second. For income, the subsamples are as follows: high
earners represented by the upper quartile of individual income; less well-off individuals
represented by having an income lower than half of the median income; and those with no

7
An additional test was undertaken, discussed at the end of this section, where individuals were restricted to
having the same marital status in both the year before and after entering the empty nest situation. The results
are qualitatively the same.

15
income. The less than half median income subsample was chosen rather than the more
symmetrical lowest quartile because many in the first (or lowest quartile) have no income.
The practical import of this raises the real income threshold from 3,000 euros to 10,000
euros for our less well-off subsample.

Table 6 Multivariate regression results for the life satisfaction of empty-nesters, health status and
real individual income. SOEP 1985-2015
Dependent variable: Life Satisfaction (positively coded from 0 to 10)
(1) (2) (3) (4) (5)
<half
Good Satisfactory Top quartile individual No
health or health or individual median individual
Empty Nest better worse income income income

All, OLS -0.37*** -0.39*** -0.41*** -0.32*** -0.29***

(0.052) (0.043) (0.078) (0.080) (0.105)
All, FE -0.16*** -0.20*** -0.31*** -0.01 -0.06
(0.050) (0.038) (0.064) (0.078) (0.109)
Male, OLS -0.35*** -0.43*** -0.39*** -0.39*** -0.19
(0.053) (0.050) (0.079) (0.153) (0.189)
Male, FE -0.14*** -0.22*** -0.26*** -0.11 -0.19
(0.051) (0.043) (0.072) (0.162) (0.214)
Female, OLS -0.32*** -0.38*** -0.01 -0.27*** -0.37***
(0.055) (0.053) (0.184) (0.101) (0.137)
Female, FE -0.13** -0.19*** 0.11 -0.01 -0.05
(0.053) (0.047) (0.175) (0.094) (0.135)
Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1. Controls for (1) and (2) are income, labour
force status, marital status, age, education, region and year. Controls for (3), (4) and (5) are labour force
status, marital status, health status, age, education, region and year. SOEP data used: Socio-Economic Panel
(SOEP), data for years 1984-2014, version 31, SOEP, 2016, doi: 10.5684/soepv31.

The health results appear to show little difference between the two groups. Given the size
of the coefficients and the prevalence of negative and statistically significant coefficients, it
is conceivable that the less healthy suffer slightly more than the healthy group though the
empty nest coefficients are not always statistically different. However, this is very cautious
evidence at best. 8
The results for income follow the trend of the previously discussed results by demonstrating
that becoming an empty nester is a near universal negative experience. Whether someone’s
income is high or low, or if they have no individual income a substantial loss of life
satisfaction is experienced by the new status of becoming empty nesters. For those who
have an individual income in the top quartile, the lack of significance for females is likely due
to a low sample size. Females are outnumbered in this category by nearly four to one, and
of those who are new empty nesters five to one. There are 135 females in this income
category who become empty nesters in our data set between 1984 and 2014. A very similar
number of males who have zero income become empty nesters in this time frame, perhaps

8
A subsequent check with a less healthy sample (those who report less than satisfactory health, as opposed to
those to satisfactory health or even worse as in column 2) results in very similar coefficients.

16
leading to the lack of precision with this subsample. Income gender inequality is an
interesting, and important issue but beyond the scope of this article.
The all, OLS, coefficients suggest a (slightly) lower loss of life satisfaction experienced by
those with less income. One possibility is that this result is due to income dilution. Perhaps
part of the sadness experienced by the less well-off is compensated by having somewhat
reduced financial constraints. This can be tested by controlling for household income,
meaning that household income has no direct impact on the subsequently obtained
coefficients for new empty nesters. The results containing household income as a control
are qualitatively the same as those in table 6, thus income dilution does not appear to
explain the results obtained for the different income subsamples. This outcome offers
support for the arguments that households are more organized around consumption
complementarities rather than production complementarities, because any reduction in
necessary expenditures has no impact on well-being (see section 2.1).
As the preceding discussion highlights, some alternative tests were undertaken (for time,
health, and income). Here a further alternative is discussed. The descriptive statistics of
table 1 demonstrates that it is important to thoroughly consider marital status. For this
reason estimations where undertaken equivalent to those of table 3 above, where an
individual’s marital status is the same in the year before becoming an empty nester and
when a new empty nester. The coefficients for the empty nest dummy variable are
sometimes a little lower, but support the above empirical analysis. Becoming an empty
nester seems to be near-universally negative for an individual’s well-being.

5. Discussion including limitations and suggestions for future research

This investigation, a large sample investigation of a phenomenon complementing the

studies that make use of qualitative data and small samples, found a substantial overall
negative effect of becoming an empty nester. However, a further result was that, for some
groups, the empty nest syndrome is an especially large problem. For example, unemployed
females appear particularly at risk from a large reduction in well-being following the
departure of the last child (or children). Similarly, individuals who are not married
experience becoming an empty-nester with a force approaching that of how people
experience unemployment (when judged by the obtained regression coefficients). Medical
research often discusses risk factors for various problems, for example heart disease, stroke,
suicide (Lutz et al. 2016; Lee et al. 2017). For these particular groups, the substantial drop in
subjective well-being suggests that becoming an empty nester may be a risk factor for
clinical depression, perhaps akin to the severe drop in well-being experienced by some
parents in the immediate post-natal phase. Post-natal depression has received much
attention from academics, healthcare professionals and other policy makers. Though the
potential impacts on others regarding ENS are likely lower, the size of the coefficients
suggest that the empty nest syndrome should also be of concern to a variety of academics
(from different fields), health professionals and policy makers.

17
One limitation is the possibility that these life satisfaction results, as substantial as they are,
may actually be undervaluing the negative effect of becoming an empty nester. As a more
cognitive measure of well-being, it may not capture the full sadness of the last child leaving
the parental home. Alternative use of more affect-based measures of well-being such as
happiness (or sadness) may result in a larger coefficient being obtained for the empty nest.
With the SOEP this possibility is not testable, but other datasets may enable such a
comparison: a worthy avenue for future research. Further information about comparisons
of different well-being measures is provided by Clark and Senik (2011) and Clark (2016).

A note about causality: even though we look at the difference in annual life satisfaction
before and after becoming an empty nester, we make no claims about causality. It is
plausible that other, non-included, factors might be driving the result of this study. It is
possible that unhappy parents per se might be causing the empty nest situation; and that
the departure of the last child simply serves to highlight a deteriorating relationship
between the parents. Further research can investigate the issue of causality. Further
research can also investigate the possibility of adaptation: do empty nesters get used to the
new situation? Equivalent adaptation regarding unemployment, divorce and marriage
(among other life changes) has been carried out (Clark and Georgellis 2013, Qari 2014), and
the anticipation of, and adaptation to, the empty nest situation is an issue worthy of
analysis.

A final recommendation focuses on the human aspects relating to the effects of the ENS.
Given the empirical results above, we suggest that if individuals know of someone who has
recently become, or is about to become an empty-nester in their neighbourhood or social
circle they should involve them in activities and try to make them feel less lonely. The
strong findings for the non-married in particular indicate that these individuals are
particularly in need of support and inclusion; and that these groups would benefit from
social interaction to overcome the feelings of isolation and loneliness; (not unlike elderly
people who become bereaved late in life). These acts of personal kindness, while not a
replacement for more systematic policy interventions, would be ultra-low cost and
potentially an effective way to dispel somewhat the adverse feelings experienced by empty-
nester.

6. Conclusions
The term Empty Nest Syndrome has existed for about 50 years and the widespread use of
the phrase has persisted in spite of no universally accepted medical diagnosis nor any
extensive proof in social science research. The evidence available on its existence is either
anecdotal or at best based on relatively small scale samples. In contrast this economic
investigation considers, with a large nationally representative sample, life satisfaction in
relation to the Empty Nest Syndrome. It draws upon an appropriately wide cross-section of
the relevant literature to complement an understanding of the theory of Identity
Economics.

18
Superficially, the empty nest syndrome may be viewed as melodramatic and/or an extreme
overreaction to the family life cycle that occurs naturally when children leave home. The
range of symptoms experienced by some empty nest parents from a loss of confidence
about the future to forms of depression are apparently irrational if seen in primarily
financial and economic terms because an absence of children at home could mean more
disposable income for the parents derived from the residual household budget (though
financial support may continue nevertheless). However, the results from this paper suggest
that ENS is more rational than considered previously especially when viewed through the
lens of Identity Economics which takes into account non-pecuniary aspects of parenting
such as situation and status.
The multivariate regression results show a substantial reduction in life satisfaction of empty-
nesters when compared to pre-empty nesters; even when controlling for the standard life
satisfaction variables of income, labour force status, marital status, age, health and
education. Furthermore, the size of the coefficients for the new empty nesters, particularly
for some subsamples, indicates that empty nest syndrome is a serious issue, being a
substantial and negative phenomenon for individual well-being.

19
Appendix 1: The Empty Nest Syndrome literature: origins in biology
The term empty nest has its origins in ornithology. The concept in its earliest application is
used to describe the end of the breeding season for any given pair of nesting birds; or the
time in the life-cycle when the young chicks have ceased being fledglings and departed as
independent adults. More recently, ENS has been applied to a human condition where
parents feel a sense of melancholy or deprivation and even mortality usually following an
extended period of child-rearing. Indeed, ENS has been defined primarily as a psychological
condition, but it does not have an actual clinical diagnosis unlike for instance the
menopause 9, which coincidentally often happens at a broadly similar time for women
(Keshishian et al., 2016).
In the biology setting, Richard Dawkins (1976) implicitly discusses the concept of the ENS in
the animal kingdom in terms of what he describes as the “Battle of the Generations”. In
particular, this notion explains the behaviour of animals as “machines programmed to do
everything in its power to propagate copies of the genes which ride inside it” (Dawkins,
1976, p 123). For example, this approach to evolutionary biology highlights the survival
instinct in nesting birds to deal with the cost of parental foraging and the risk of predation
by having a large clutch size and sometimes more than one nesting season per calendar
year, (Lima, 1987, p 1063). Relatedly, it may help to explain indirectly why some people
experience ENS because once the nest (that is, home) is emptied then the primary purpose
of the parent is partially diminished, (Fox and Bruce, 2001, p 396). Of course, this is
especially true in humans where the young take many years of intense upbringing.
Dawkins based his views on gene survival in part by using the work by Trivers (1972) who
developed the notion of Parental Investment (PI), which has its origins in social biology. In
the absence of a cost-benefit analysis for most animals, this concept considers the
advantages of any investment by a parent in an off-spring that increases the chance of
survival and thereby “reproductive success”. Hence, PI is used where monetary values are
difficult to quantify and in particular for non-human female parents:
”…represents the sum of all the food she can gather or manufacture in a lifetime of
work, all the risks she is prepared to take, and all the energy and effort that she is
able to put into the welfare of children.” (Dawkins, 1976 p 124)
In addition, this concept of PI can be extended to a novel study of human behaviour because
many monetary aspects of child-rearing are actually very difficult to value such as the joy
gained from seeing a child take its first steps or the satisfaction derived from attending its
wedding. However, these stages (and others) in the development of a child’s life are in

9
The menopause is an indicator of a life change as it is the sudden end of female reproduction often but not
always in mid-life. The phenomenon of the menopause is extremely rare in other animal species and a
potentially important way to view the ENS. From an economic perspective, the menopause may help to
explain the issues facing (female) middle-age parents. That is, given the non-trivial amount of economic
resources, energy and time invested in children plus the related matter of a relatively small number of human
births per female, the menopause may be a genetic tactic with an economic genesis.

20
reality indicators of continued existence; or in other words, these are signs that the
probability of gene survival are relatively high or increasing over time or across generations.
Hence, the amount of resources invested in an off-spring is an acceptable cost to make in
order to increase the probability of gene survival, even if this is not a conscious decision
emotionally or economically. Furthermore, this compulsion may be part of a trade-off
between current consumption by the parent and the longer term future of the child.
Therefore, not only is gene survival based on fundamental principles of economics such as
consumption and production but likewise the feeling of ENS may also be based in these
primeval instincts (see Wilson 1978 for a discussion on the origins of socio-biology).
However, given the non-clinical nature of the ENS, any meaningful analysis is perhaps best
understood in a much wider context beyond biology. As a result, Identity Economics (or a
sense of self-being) is well-placed and occupies a central role in this investigation as it
includes the related areas of sociology and psychology, (Akerlof and Kranton, 2000, p 748).
As a result, attention in the main body of the text is given to the economics literature, since
the concept of the “empty nester” is clearly applicable in this discipline, too due to
considerations of resource allocation.

21
Appendix 2: A Cost-Benefit Analysis Framework of ENS
Negative effects from ENS (COSTS), especially for full-time parents and careers
1. Loss of short-term purpose
2. Loneliness in the long-term
3. Reduction in social interaction e.g. other parents
4. No longer having influence in the child’s destiny

Positive effects from ENS (BENEFITS), especially for working parents and careers
1. Gain of independence from depend children
2. More financial autonomy
3. Sense of achievement and pride in the destinations of the children
4. No longer having to defer gratification and/or make self-sacrifice i.e. holidays, cars

ENS CBA is complicated by

1. Boomerang children; post-university/training and after divorce or relationship
breakdowns
2. Housing costs especially the relatively high cost house buying in certain areas of the
country
3. Grandparent responsibilities due to high child-care costs such as nursery, after-
school clubs
4. Women living sometimes half their lives after the menopause

The Correlates of ENS

1. Menopause (co-incidence i.e. function of age)
2. Divorce (correlated positively)
3. Historical setting
4. Age of mother at birth of last child
5. Career opportunities including female employment rates
6. The gender pay gap

22
Appendix 3: Empty-nesters and pre-empty nesters, observations in the
subsamples (tables 4-6)
Table A1: Both genders
(1) (2)
New Empty Nesters Pre-empty nesters Total
Labour Force Status
Employed 1090 94067 95157
Unemployed 180 10808 10988
Retired 129 2426 2555
Time
Before 2000 732 59336 60068
From 2000 1111 98687 99808
Marital Status
Married 819 127776 128595
Single 154 10551 10705
Widowed 62 1521 1583
Divorced 245 9370 9615
Separated 506 2822 3328
Health
Good health (and better) 730 73084 73814
Satisfactory and below 751 50543 51294
Income
Top quartile 488 39600 40088
Less than half median 563 57113 57676
Zero income 353 30781 31134

Table A2: Males

(1) (2)
New Empty Nesters Pre-empty nesters Total
Labour Force Status
Employed 715 50458 51173
Unemployed 104 4093 4197
Retired 70 1256 1326
Time
Before 2000 446 28343 28789
From 2000 690 44699 45389
Marital Status
Married 412 62993 63405
Single 111 3357 3468
Widowed 19 337 356
Divorced 132 1907 2039
Separated 427 497 924
Health
Good health (and better) 489 32982 33471
Satisfactory and below 440 22440 22880
Income
Top quartile 426 34128 34554
Less than half median 205 1964 8169
Zero income 137 3918 4055

23
Table A3: Females
(1) (2)
New Empty Nesters Pre-empty nesters Total
Labour Force Status
Employed 375 43609 43984
Unemployed 76 6715 6791
Retired 59 1170 1229
Time
Before 2000 286 30993 31279
From 2000 421 53998 54419
Marital Status
Married 407 64783 65190
Single 43 7194 7237
Widowed 43 1184 1227
Divorced 113 7463 7576
Separated 79 2325 2404
Health
Good health (and better) 241 40102 40343
Satisfactory and below 311 28103 28414
Income
Top quartile 62 5472 5534
Less than half median 358 49149 49507
Zero income 216 26863 27079

24
References
Albertine, V. (2014). Clothes, Clothes, Clothes. Music, Music, Music. Boys, Boys, Boys.
London: Faber and Faber.

Akerlof, G and Kranton, R (2000). ‘Economics and identity’, Quarterly Journal of Economics,
vol. 115 (3), pp715-53.

Akerlof, G. and Kranton, R. (2010). Identity Economics: how identities shape our work,
wages and well-being. Princeton, NJ: Princeton University Press.
Blanchflower, D. G. and Oswald, A. J. (2008). Is Well-Being U-Shaped over the Life Cycle?
Social Science and Medicine 66, 6: 1733-1749.
China Development Research Foundation (2014). Demographic Developments in China.
Routledge: Abingdon.
Clark, A. E. (2016). “SWB as a Measure of Individual Well-Being” Chapter 16 in the Oxford
Handbook of Well-Being and Public Policy, M. Adler and M. Fleurbaey (Eds.), Oxford: Oxford
University Press, pp 518-552.
Clark, A. E. and Senik, C. (2011). “Is Happiness Different From Flourishing? Cross-Country
Evidence from the ESS”, (Andrew Clark and Claudia Senik), Revue d’Economie Politique,
Vol.121, no.1, pp 17-34.
Cooper, K. L. and Gutmann, D. L. (1987). Gender identity and Ego Mastery Style in middle-
aged pre- and post-empty nest woman. The Gerontologist (1987) 27 (3) pp347-352.
Dawkins, R. (1976). The Selfish Gene. Oxford: Oxford University Press.
Day-Lewis, C. (1956). Walking away (poem).
Dodd, C. (2011). The Empty Nest: How to survive and stay close to your adult child. London:
Piatkus.
Dolan, P., Peasgood, T., and White M. (2008). Do we really know what makes us happy? A
review of the economic literature on the factors associated with subjective well-being.
Journal of Economic Psychology, 29, 94–122.
Downward, P. and Dawson, P. (2016). Is it pleasure or health from leisure that we benefit
from most? An analysis of well-being alternatives and implications for policy. Social
Indicators Research, Vol 126, No 1, 443-465.
Fox, G. L. and Bruce, C. (2001). Conditional fathering: Identity Theory and Parental
Investment Theory as alternative sources of Explanation of fathering. Journal of Marriage
and Family. Vol 63, No 2 pp394-403.
Godber, J. (2016). The Empty Nesters’ Club. The John Godber Company, Hull (play).
Hetschko, C., Knabe, A. and Schöb, R. (2014). Changing identity: retiring from
unemployment. The Economic Journal, Vol 124 pp 149-166.

25
Hiedemann, B., Suhomlinova, O. and O'Rand, A. M. (1998). “Economic independence,
economic status, and empty nest in mid-life marital disruption.” Journal of Marriage and the
Family 60: 219-31.
Keshishian, A., Wang, Y., Xie, L. and Baser, O. (2016). The economic impact of symptomatic
menopause among low-socioeconomic women in the United States. Expert Review of
Pharmacoeconomics and Outcomes Research. Volume 16 Issue 2.
Lee, P N., Thornton, A. J., Forey, B. A., and Hamling, J. S. (2017). Environmental Tobacco
Smoke Exposure and Risk of Stroke in Never Smokers: An Updated Review with Meta-
Analysis. Journal of Stroke and Cerebrovascular Diseases, 26(1), 204-216.
Lennon, J. and McCartney, P. (1967). She’s leaving home (song).
Lima, S. L. (1987). Clutch Size in birds: a predation perspective. Ecology Vol 68 No 4. Pp
1062-1070.
Liu, L. and Guo, Q. (2008). Life satisfaction in a sample of empty-nest elderly: a survey in the
rural area of a mountainous county in China. Quality of Life Research 17:823–830.

Lutz, J., Morton, K., Turiano, N. and Fiske, A. (2016) "Health conditions and passive suicidal
ideation in the Survey of Health, Ageing, and Retirement in Europe." The Journals of
Gerontology Series B: Psychological Sciences and Social Sciences 71, issue 5, 936-946.
Merrill, J. P. (2010). An Economist’s view of marriage. Society, Volume 47, number 3, pp
234-39.
Piper, A. T. (2015). Sliding down the U-shape? A dynamic panel investigation of the age-well-
being relationship, focusing on young adults, Social Science & Medicine, Vol. 143, pp 54-61.

Rubin, L. (2008). Sixty on the up: The truth about ageing in the twenty-first century. Beacon
Press, Boston, MA.

Qari, S. (2014). Marriage, adaptation and happiness: Are there long-lasting gains to
marriage? Journal of Behavioral and Experimental Economics, 50, pp29-39.
Rohrer, J.M., Brümmer, M., Schmukle, S.C., Goebel, J. and Wagner, G. G. (2017). “What Else
Are You Worried About?” – Integrating Textual Responses into Quantitative Social Science
Research [WWW Document]. doi:10.17605/OSF.IO/KSG7D
Silverstein, M. J. and Sayre, K. (2009). The female economy, Harvard Business Review.
September issue.
Spence, D. and Lonner, T. (1971). The “Empty Nest”: A Transition within Motherhood. The
Family Coordinator October pp 369-375.
Stevenson, B., and Wolfers, J. (2007). Marriage and divorce: Changes and their driving
forces. Journal of Economic Perspectives, 21 (2), pp 27-52.

26
Stutzer, A. and Frey, B.S. (2006). Does Marriage Make People Happy, Or Do Happy People
Get Married? Journal of Socio-Economics, vol. 35, pp 326-347.
Tamres, L. K., Janicki, D and Helgeson, V. S. (2002). Sex Differences in Coping Behaviour: A
Meta-Analytic Review and an Examination of Relative Coping. Personality and Social
Psychology Review, vol. 6 (1), pp 2-30.
Trivers, R. L. (1972). Parental investment and sexual selection. In Sexual Selection and the
Descent of Man (edited by B Campbell) Chicago: Aldine pp 136-79.

Wagner, G. G. (2016). “Expanding the toolbox for measuring well-being to include text
responses from quantitative social science survey data”, Presentation given at the London
School of Economics, 13 December 2016. (Available online:
http://cep.lse.ac.uk/conference_papers/12_12_2016/wagner.pdf).
Wagner, G. G., Frick, J. R. and Schupp, J. (2007). The German Socio-Economic Panel Study
(SOEP): scope evolution and enhancements. Schmollers Jahrbuch 127(1), 139-169.
Zahidi, S. (2012). Global population ageing: peril or promise? Chapter 3 in Women and
Ageing. World Health Organisation.

27