Content uploaded by Jørgen Svalund
Author content
All content in this area was uploaded by Jørgen Svalund on Nov 13, 2015
Content may be subject to copyright.
33
Nordic journal of working life studies Volume 5 ❚ Number 2 ❚ June 2015
Job Insecurity and Mental Well-Being in Finland,
Norway, and Sweden
Consequences of Flexicurity in a Nordic Welfare Setting
Patrik Vulkan1
Doctoral Candidate, Department of Sociology and Work Science, University of Gothenburg, Sweden
Antti Saloniemi
Professor, School of Social Sciences and Humanities, University of Tampere, Finland
Jørgen Svalund
Researcher, Fafo Institute for Labour and Social Research, Norway
Anna Väisänen
Master of Social Science, School of Social Sciences and Humanities, University of Tampere, Finland
ABSTRACT
This article describes how the flexicurity arrangement of low job security, high employment security,
and good income security advocated by various authors affects the mental well-being of employees.
Data are derived from a survey carried out in 2010–2011 among employees in Finland, Norway, and
Sweden. The main findings are that all three forms of cognitive security (the perceived risk) have an
independent effect on mental well-being and that the worry of insecurity (the affective component)
mediates the relationship with mental well-being. The interaction effects show that high levels of
employment security can alleviate the detrimental effects of job insecurity on mental well-being. No
similar interaction effect was found with job insecurity and income security. The results are discussed
in relation to the institutional arrangements of the Nordic countries’ welfare states, concluding that the
high employment security needed for a successful flexicurity arrangement requires either low levels of
unemployment or effective and extensive active labor market programs. Flexicurity is thus susceptible
to economic turmoil and requires further labor market investments, even in the Nordic countries.
KEY WORDS
Flexicurity / job insecurity / mental well-being / employment security / income security /
cognitive and affective insecurity
Introduction
Researchers are increasingly examining job insecurity and its consequences for
employees in exible labor markets (Cheng and Chan 2008). The present study
contributes to the eld by analyzing how a potential increase in job insecurity
1 Patrik Vulkan, Department of Sociology and Work Science, Box 720, 405 30 Göteborg, Sweden.
Email: patrik.vulkan@socav.gu.se
34 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
relates to mental well-being among employees and whether any negative consequences
can be reduced by the presence or interaction of employment or income security. We
differentiate between cognitive and affective insecurity in our analysis, allowing us to
better understand whether the perception of high insecurity by an employee in a given
situation actually relates to poor mental well-being (Anderson and Pontusson 2007;
Greenhalgh and Rosenblatt 1984; Huang et al. 2012). By taking into account three
forms of insecurity and their interaction, in conjunction with both cognitive and affec-
tive components, we contribute by providing a new multidimensional model of job inse-
curity.
This study is a systematic comparison of Finland, Norway, and Sweden with regard
to the different forms of insecurity in the labor market, using data collected in 2010–
2011. We consider the institutional arrangements relevant to the discussion of exicurity
in the three welfare states as important contextualization of employees’ experience of
insecurity and security (Chung and Van Oorschot 2010; Mau et al. 2012). Several stud-
ies identify the Nordic countries as having employment regimes that combine exibility
and security, in which the risk of increased job insecurity in exible labor markets is
compensated by employment and income security, thereby reducing detrimental results
for employees (cf. Muffels and Luijkx 2008; Muffels et al. 2014). As such, Finland,
Norway, and Sweden are relevant cases for examining the potential outcomes of a com-
pensatory arrangement.1
The next section outlines the implementation of exicurity policies in labor mar-
kets and their consequences for employees’ well-being. Theories regarding relationships
between mental well-being and insecurity, and how employment and income security
provide means to alleviate the consequences of job insecurity, are then reviewed. This
review is followed by a description of institutional welfare arrangements relevant to the
discussion and presentations of the acquired data, analytical methods, and analysis of
the results. The article ends with a concluding discussion.
Flexicurity and job insecurity
The concept of exicurity, as advocated by Wilthagen and Tros (2004) and others
(Madsen 2004, 2006; Muffels et al. 2008) is a well-known proposal for achieving ex-
ibility in labor markets without detrimental insecurity. It suggests that we need to view
security in the labor market as composed of job, employment, and income security,
rather than simply job security, in order to achieve an optimal balance between security
for employees and exibility for employers. Excessive employment protection is con-
sidered a hindrance to numerical exibility by employers and a crucial area for reform
by exicurity proponents. Weak employment protection is likely to cause job insecurity
among employees. However, adverse consequences of this can be avoided, according
to exicurity advocates, through high levels of employment security (opportunities for
nding a new job) and income security (abilities to avoid nancial hardship during job-
less periods). The presence of these two forms of security should supposedly compen-
sate for negative consequences stemming from job insecurity such as mental ill-being.
Critics of the exicurity approach argue that the exibility and security mentioned are
incompatible goals and that the arrangement is simply another form of deregulation of
employment protection, which reduces welfare states’ protective functions and results
Nordic journal of working life studies Volume 5 Number 2 June 2015 35
in insecure jobs and lower salaries (Burroni and Keune 2011; Howell et al. 2007; Ozaki
1999; Tangian 2008). Accordingly, Burchell (2009) nds no evidence that exicurity
policies protect the mental well-being of employees, although the only form of insecu-
rity the cited author specically analyzed is job insecurity.
From this brief summary of relevant literature, a key question can be formulated:
What are the full consequences of job insecurity for the mental well-being of employees?
Wilthagen and Tros (2004) suggested a way to address this question in detail by using
a more multifaceted concept of security, but this raises another question: ‘How does a
multidimensional notion of insecurity function and how does it translate into the risk
of ill-being?’
The analytical starting points of the present study are the components of job inse-
curity. Job insecurity is usually considered to be either objective or subjective, the former
referring to the actual risk of dismissals and layoffs, and the latter (the focus of this
article) to individuals’ evaluation of the likelihood of losing their current jobs in the
near future (Ellonen and Nätti 2013). Subjective job insecurity is generally dened as
being involuntary, in the sense that employees do not actively seek it (De Witte 2005). It
should be noted that the same job threat can lead to different experiences of insecurity
among employees with different characteristics (De Witte 1999; Sverke and Hellgren
2002). However, the uncertainty of job insecurity may compromise employees’ ability
to deal with the ongoing situation, since it is unclear what action should be taken. They
may thus perceive themselves as powerless to resist the threat (Anderson and Pontusson
2007; Hellgren and Sverke 2003; Lazarus and Folkman 1984; Sjöberg 2010).
The study also addresses job insecurity as a multidimensional phenomenon, with
both cognitive and affective aspects (Greenhalgh and Rosenblatt 1984; Huang et al.
2012). The cognitive aspect refers to the employee’s estimation of the likelihood of a
job loss in the future. While this estimation may be incorrect, there is usually a positive
correlation between individuals’ assessments of risks and ensuing job losses (Chung and
Mau 2014; Dickerson and Green 2012; Klandermans et al. 2010). The affective aspect is
the emotional reaction to the estimation of job loss likelihood. Research linking the two
aspects has also found that cognitive job insecurity is strongly associated with worry
about losing the job (Berglund et al. 2014). Anderson and Pontusson (2007) argued
that cognitive job insecurity should be considered an inuential determinant of affective
job insecurity, but the affective experience also depends on the individual’s capacity to
handle the risk of job loss or deal with the event (cf. Huang et al. 2010).
Mental well-being
Job insecurity among employees is negatively related to subjective well-being (Sjöberg
2010; Vulkan 2012). This can be understood as individuals’ subjective stress reactions
to a state of unpredictability and lack of control. Job insecurity has been shown to be
associated with psychological distress, anxiety, and depression as well as mental, emo-
tional, and physical exhaustion. This state has been shown to be as potentially harmful
as actual unemployment (Burchell 2011; De Witte 1999; Sjöberg 2010).
Two theories provide a deeper theoretical explanation of the mechanism whereby
insecurity affects mental well-being. The rst is the classical latent deprivation theory
presented by Marie Jahoda (1982), which holds that employment fullls a manifest
36 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
function, by generating an income, and ve latent functions: time structure, social con-
tact, sharing of common goals, status, and activity. In a situation with great risk of
unemployment the latent functions are held to be the main factors explaining the nega-
tive effect on mental well-being, with the risk of unemployment implying the obstruction
and deprivation of these current needs (Creed and Bartrum 2006; Ervasti and Venetok-
lis 2010). However, none of these authors offer indications of the relative explanatory
importance of the latent functions. The second is agency restriction theory, developed by
David Fryer, which is complementary to Jahoda’s theory. It treats individuals as:
socially embedded agents who are actively striving for purposeful determination, attempt-
ing to make sense of, initiate, inuence and cope with events in line with personal values,
goals, expectations of the future in a context of cultural norm, traditions and past experi-
ence. (Fryer 1995)
The disruption of plans and strategies caused by insecurity severs individuals from a
meaningful future if they lack, or are obstructed from, effective strategies to deal with
insecurity, leading to mental ill-being. Consequently, it is not work in itself that causes
this but hindrance of the ability to plan and control one’s own life that work can fulll.
This theory stresses the manifest function as the explaining mechanism, with the poten-
tial loss of income and threat of poverty rendering the employees unable to deal with
insecurity. Each theory suggests a main aspect of insecurity affecting mental well-being,
while recent research on insecurity among Swedish employees indicates that manifest
and latent functions are of approximately equal explanatory value (Vulkan 2012).
Employment and income security as means to alleviate job insecurity
Job insecurity in itself is considered detrimental for well-being, but can be countered by
the presence of employment security and income security according to exicurity theory
(Wilthagen and Tros 2004). Employment security is the possibility of nding a new
job; knowing that you will be able to nd a new job if the need arises should lessen the
insecurity associated with losing a specic job. This concept of employment security is
similar to the strongly related concept of perceived employability, but it usually stresses
a structural or institutional, rather than individualistic, view of the labor market (cf.
Berglund et al. 2014; Berntson 2008; Garsten and Jacobsson 2004; Silla et al. 2009).
Employment security is believed to be enhanced by institutional arrangements such as
lifelong learning institutions and active labor market policies, as well as a dynamic econ-
omy in general (Berglund and Furåker 2011). Important antecedents to employment
security are age, education, and tenure. Generally, young people believe they have better
employment opportunities than old people and those with higher education generally
have better chances of nding a new job of equal or better value, which also translates
into a more positive outlook (Furåker 2010b). Regarding tenure, there seems to be a
‘lock-in’ effect, i.e., length of tenure and employees’ beliefs that they could nd a new
job seem to be negatively correlated (Berglund et al. 2014).
Employees can also attain income security, usually through unemployment insur-
ance or other institutional arrangements for social security intended to ensure that nan-
cial needs are met during a period of unemployment. Income security is assumed to ease
Nordic journal of working life studies Volume 5 Number 2 June 2015 37
the burden of unemployment and facilitate a smooth process of return to work, thus
acting as a compensating mechanism for job insecurity in the exicurity arrangement
(Berglund et al. 2014; Muffels et al. 2014). Income security has also been found by
Sjöberg (2010) to enhance employees’ mental well-being directly by reducing the uncer-
tainty connected to becoming unemployed and providing insurance against unforeseen
economic hardships (cf. Carr and Chung 2014).
These two forms of security can be understood as means for mitigating conse-
quences of job insecurity and should, according to proponents of exicurity, negate
the adverse effect on mental well-being. Employment security and, to a lesser extent,
income security are both associated with a lower degree of anxiety (affective job inse-
curity) among employees and we expect the presence of these two forms of security to
be related to improved mental well-being (Berntson and Marklund 2007; Ervasti and
Venetoklis 2010; Green 2011; Nordenmark et al. 2006; Vulkan 2012). Employment
security and income security are also treated as multidimensional phenomena, with both
cognitive and affective aspects (cf. Vulkan 2012), in the present study.
We thus anticipate an interaction effect between cognitive job insecurity and employ-
ment security, reducing the impact of cognitive job insecurity on mental well-being.
Earlier research (cf. Berglund et al. 2014) has shown that employees are less worried if
cognitive job insecurity is combined with employment security than the additive effects
of these factors would suggest. Thus, high levels of employment security mitigate worry
associated with the anticipated risk of a job loss. A similar interaction effect between
cognitive job insecurity and cognitive income security could be possible, but previous
research has not found signicant effects on worries to support this (Berglund et al.
2014). Moreover, we predict that affective job insecurity should decrease the interaction
effect between cognitive job insecurity and employment security by acting as a mediator
between cognitive job insecurity and mental well-being (Vulkan 2012).
Institutional arrangements in the three Nordic countries
To further contextualize the results, we examine the institutional frameworks that
research has found to affect the different forms of security and insecurity at play. In line
with the key components of exicurity, we describe the focal countries in terms of indi-
cators of the institutional arrangements most likely to affect security (Chung and Mau
2014): employment protection legislation (EPL), active labor market programs (ALMP),
and unemployment benets (UBs). The unemployment level is also presented since it
affects the performance of the aforementioned factors in the labor market. Contextual-
ization by focusing on the institutional arrangements can help us understand variations
in the results. Moreover, ndings that the results remain the same despite institutional
variations between the three countries would corroborate our key argument that there
are strong relationships between the considered forms of security and well-being. The
data presented in this section are based on information reported by ofcial bodies of
the three countries to the OECD, as shown in Table 1 and associated endnotes. The
presented indicators show the institutional arrangements in 2010 or as close as possible
to this year.
To evaluate the three countries’ EPL, which is likely to inuence job insecurity
(Anderson and Pontusson 2007; Chung and Van Oorschot 2010), we use the measure of
38 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
EPL strictness developed and applied by the OECD concerning legislation for employ-
ees in regular employment and temporary employment. Collective agreements between
employers and employees are also taken into account (OECD 2013; Venn 2009). The
measure for both forms of employment is a scale from 0 to 6, with a high number
indicating strict EPL. Regarding EPL for regular employees, all countries have similar
strictness scores, although the Swedish score is slightly higher than the Norwegian and
Finnish scores. These gures have been constant for some years. EPL strictness for tem-
porary employment shows clear differences, being high for Norway, intermediate for
Finland, and very low for Sweden. In the Swedish case, new legislation was enacted in
2007 that allowed a greater degree of xed-term contracts (Nordic Council of Ministers
2010). Consequently, EPL strictness was reduced to almost half its previous value. Such
institutional changes are notable, since in themselves they can affect the well-being of
employees (Lübke and Erlinghagen 2014). In summary, while the protection for per-
manent employees is quite similar in all three countries, EPL strictness for temporary
employees differs greatly across the Nordic countries.
ALMP is likely to affect employment security, especially if unemployment is high
(Chung and Van Oorschot 2010). Thus the proportion of Gross Domestic Product
(GDP) spent on ALMP per percentage point of unemployment is used as an indicator of
the ambitiousness and prioritization of ALMP (Berglund and Furåker 2011). Levels of
ALMP investment relative to the unemployment level are quite similar in all three coun-
tries. The absolute amount of GDP invested in Norway is smaller, but due to a low level
of unemployment, the relative amount is higher. Unemployment fell in all three countries
from 2005 until 2009 then rose again. Over the same time, the amount of GDP spent
on ALMP has been constant in Finland and declined somewhat in Norway and Sweden,
although it is still well above the OECD mean.
UB, which is likely to inuence income security (Chung and Mau 2014), is measured
here using the net replacement rates of an average wage for a single household with no
children during the initial phase of unemployment. Reported rates are lowest in Sweden,
intermediate in Finland, and most generous in Norway. The Norwegian replacement rate
has been stable over the last decade. The Finnish replacement rate decreased from 61.5% in
2002 to 50.7% in 2008, but has subsequently increased. The Swedish replacement rate has
diminished since 2002 when it was 67.5%, drastically so since 2006 when the new Con-
servative government reduced the generosity of the UB system (Bengtsson and Berglund
2012). It should also be noted that unemployment insurance is mandatory in Norway, but
following the Ghent system, voluntary in Finland and Sweden (cf. Scruggs 2002).
Table I Summary of employment-related measures for the three countries
EPL (2010)2
ALMP (2009)3UB (2010)4Unemployment
(2010)5
Regular Temporary
Finland 2.38 1.88 0.11 0.53 8.5
Norway 2.23 3.04 0.14 0.64 3.7
Sweden 2.52 0.79 0.13 0.47 8.7
OECD mean 2.13 2.06 0.06 0.57 8.5
Nordic journal of working life studies Volume 5 Number 2 June 2015 39
To summarize the institutional setting in the three countries, Finland, Norway, and
Sweden emerge as similar with respect to ALMP. EPL for regular employees and unemploy-
ment levels are similar in Finland and Sweden, although Sweden has a lower UB replace-
ment rate and weaker EPL for temporary employees. In comparison to both Finland and
Sweden, Norway has slightly weaker EPL for regular employees, but much stricter EPL for
temporary workers, and a more generous UB system. Furthermore, Norway has a signi-
cantly lower unemployment level than the other two countries. All these factors are likely
to result in higher levels of mental well-being among Norwegian employees.
Survey data and methodology
The data concerning the various forms of security and insecurity used in this article were
collected using three separate, but similar, questionnaires designed specically for the
study and sent to employees in Sweden (autumn 2010), Finland (winter 2010–2011),
and Norway (spring/summer 2011). The ordinary Labor Force Survey (LFS) was used
as a sampling framework. For comparative reasons, we used the age group 19–64 years
in the analysis. The response rates were 40% in Norway (1,634 responses), 53% in Fin-
land (2,252 responses), and 54% in Sweden (2,023 responses). Young people under the
age of 24, the temporarily employed, and, in Finland, men are overrepresented among
the non-respondents. These groups have been further analyzed with regard to potential
selection bias, but no deviations of note have been found. The further drop in response
rate that can be observed in the regression analyses is mainly due to removing non-
responses from the security and insecurity variables.
The dependent variable is mental well-being, which has been operationalized using a
modied version of General Health Questionnaire 12 (GHQ-12), a commonly used mea-
sure of individual psychological well-being. The instrument targets two areas – the inability
to carry out normal functions and the appearance of distress – to assess a person’s well-
being (Goldberg and Williams 1988). Each respondent is instructed to answer how they
have been feeling during the past few weeks, by agreeing or disagreeing with statements
about their mental well-being. We used Finnish, Norwegian, and Swedish translations of
GHQ-12, and modernized the wording of some questions. For all 12 items there was a
four-point Likert response and scoring system (0–3) in the Finnish and Swedish versions,
while there was also a fth, middle alternative in the Norwegian version. GHQ-12 is widely
used as a unidimensional instrument and a factor analysis shows that all 12 questions work
well as single scales (α = 0.898 in Finland, α = 0.888 in Norway, and α = 0.866 in Sweden).
The scale ranges from 0 to 36 (the Norwegian range was rescaled from 0–48 to 0–36), with
higher values indicating better mental well-being. The unidimensionality and high alpha
value of the Norwegian GHQ-scale are strong indications that the difference in scale points
does not result in a qualitatively different construct compared to Finland and Sweden.
Table II Cognitive and affective dimensions included in the article
Job insecurity Employment security Income security
Cognitive dimension X X X
Affective dimension X X
40 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
The dimensions of security and insecurity comprising the focal independent variables
are shown in Table 2. Questions that could effectively gauge affective employment secu-
rity were not included in the surveys at the time of construction.
Cognitive job insecurity is measured by the question ‘How do you assess the risk
that, in the coming 12 months, you are going to lose your job?’ The rst option includes
‘Very large’ and ‘quite large,’ the second consists solely of ‘neither large nor small,’ the
third includes ‘quite small’ and ‘very small.’ Respondents who chose the fourth option,
‘Don’t know,’ were excluded from subsequent analyses.
Affective job insecurity is measured by the question ‘To what extent do you worry
about losing your present job?’ ‘I worry a great deal’ and ‘I worry to a certain extent’ are
grouped as the rst response option, while ‘I worry a little bit’ and ‘I do not worry at all’
are treated as separate options. Norwegian respondents could also answer ‘no opinion’;
they were excluded from subsequent analysis.
Cognitive employment security is measured by the question ‘In general, what do you
think of your current opportunities for nding another job that is equal or better than
your current job?’ with the answers ‘good’ and ‘quite good’ grouped as the rst response
option, ‘neither good nor poor’ as the second, and ‘quite poor’ and ‘very poor’ as the third.
Cognitive income security is measured by the question ‘How would you/your
household manage economically if you became unemployed and had to rely on unem-
ployment benets for between 3 and 6 months?’ with the answers ‘Very well’ and
‘Quite well’ making up the rst response option, ‘Neither well nor poorly’ the second,
and ‘Quite poorly’ and ‘Very poorly’ the third. In Norway, the respondents could also
answer ‘I don’t know.’ These respondents were excluded from subsequent analysis. Note
that the respondent might share their nancial responsibilities with other people. Previ-
ous research has shown that manifest and latent functions of work affect mental health
within 6 months, thus our chosen time span seems reasonable for the question (Selenko
and Batinic 2013).
Affective income security is measured by the question ‘In general, do you worry
about your/your household’s economy?’ Due to events beyond our control certain
response options differ between the countries. The answers ‘I worry a great deal’ and ‘I
worry to some extent’ make up the rst option. The second option consists of ‘I do not
worry much’ for the Finnish and Swedish respondents and ‘Neither a lot nor a little’
for the Norwegian respondents. The third response option for the Finnish and Swedish
respondents is ‘I do not worry at all,’ while it consists of two options for the Norwegian
respondents: ‘I worry to a quite small degree’ and ‘I worry to a very small degree.’ There
was also a ‘No opinion’ category in all countries. Respondents choosing this option were
excluded from subsequent analysis.
A central question addressed in the article is whether employment and income secu-
rity can compensate for job insecurity. This is tested using ordinary least square (OLS)
regression analysis, for which the variables are coded in different directions, with job
insecurity (both cognitive and affective) measuring insecurity (from low to high), while
employment and income security (both cognitive and affective) measure security (from
low to high). Thus, the relationships between mental well-being and job insecurity fac-
tors are expressed as negative coefcients (as job insecurity is inversely related to men-
tal well-being) while the relationships between mental well-being and employment or
income security factors are expressed as positive coefcients (as security is positively
related to mental well-being).
Nordic journal of working life studies Volume 5 Number 2 June 2015 41
Control variables with a known relationship to either job insecurity or mental well-
being are also included in the analysis. They include demographic (age and civil status),
human capital (occupation and income), and psychosocial (job demand, social support,
and job satisfaction) variables (Berntson and Marklund 2007; Chung and Van Oorschot
2010; De Witte 1999, 2005; Ellonen and Nätti 2013; Karasek and Theorell 1990; Keim
et al. 2014; Sverke and Hellgren 2002). Job satisfaction reportedly mediates part of the
effect of job insecurity on mental well-being, but also has a unique effect on mental well-
being (Chirumbolo and Hellgren 2003). In order to control for the latter it is necessary
to include job satisfaction as a control variable, but we do not expect this to result in
an overadjustment risk. For parsimonious presentation, control variables showing no
signicant relationship with mental well-being are not included in the results displayed
here. These variables are gender, educational level, working hours, employment contract,
number of employees at workplace, tenure, and sector.
The analysis starts by describing distributions of the focal independent variables
(Table 3), allowing a country comparison. Three OLS regression models are then used
to estimate effects of the independent variables (Table 4). The rst focuses on how the
cognitive variables relate to the subjective mental well-being of employees, the second
includes the affective variables, while the control variables are also included in the third
model.
Finally, interaction effects of job insecurity and employment security on mental
well-being are estimated (Table 5). The rst model here shows interaction effects of the
relevant cognitive variables. While not shown in the table, the control variables are also
included at this point. In Model 2, the relevant affective variable is included, to measure
to what extent the cognitive interaction effect on mental well-being is mediated through
it. The interaction between cognitive job insecurity and cognitive income insecurity was
also tested, but yielded no signicant results, which are therefore not shown. All ndings
mentioned below are signicant at a probability level of at least 0.05.
Results
As shown in Table 3, employees in Finland report relatively high levels of cognitive job
insecurity. Unemployment rates are similar in Finland and Sweden, and although EPL
for regular employees does not differ much between these countries, a higher proportion
of Finnish employees perceive a great risk of losing their job. The same pattern occurs
with regard to affective job insecurity, with Finnish employees reporting to a larger
extent than Swedish employees that they worry about losing their job. The lowest levels
of affective job insecurity are found in Norway, as expected considering its low level of
unemployment and relatively strict EPL.
Concerning cognitive employment security, employees in Norway perceive a smaller
risk of not nding a new job than employees in Finland and Sweden. Swedish and Finn-
ish respondents exhibit greater levels of polarity, with large proportions of employees
perceiving the opportunities as either very good or very poor.
With regard to cognitive income security, a higher proportion of Swedish employ-
ees than either Finnish or Norwegian employees stated that they would manage well
economically during unemployment, which is surprising considering that Swedish
employees have the lowest replacement rate. In sharp contrast, in terms of affective
42 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
income security, 62.6% of the Norwegian respondents reported that they do not worry
at all about their, or their household’s, economy, whereas corresponding percentages for
Swedish and Finnish respondents were 30.0% and just 18.3%, respectively. A larger
percentage of Finnish employees also reported that they worry a great deal or to some
extent. The Norwegian results should be interpreted with some caution, since the dif-
ference in wording of the response options could explain some of the results. Affective
income security refers to worry about ‘your/your household’s economy,’ and may reect
economic aspects beyond the scope of this article, for instance economic responsibilities
encompassing varying numbers of individuals, the presence and scale of household debt,
Table III Frequency table of insecurity variables (%)
Finland
(n = min 2,099,
max 2,196)
Norway
(n = min 1,489,
max 1,573)
Sweden
(n = min 1,872,
max 1,993)
Cognitive Job Insecurity
Ver y or quite large risk of job loss 10.6 3.4 5.4
Neither large nor small risk 16.3 6.6 9.9
Ver y or quite small risk 73.1 90.1 84.7
Affective Job Insecurity
Worr y a great deal or to some
extent about job loss
10.0 4.8 9.0
Worry a little bit 27.0 18.7 17.3
Do not worr y at all 63.0 76.6 73.7
Cognitive Employment Security
Very or quite good opportunities
of finding a new job
35.7 37.5 35.9
Neither good nor poor
opportunities
25.2 33.1 27.1
Ver y or quite poor oppor tunities 39.1 29.4 37.0
Cognitive Income Security
Would manage very or quite well
economically during unemployment
41.8 43.5 49.7
Neither well nor poorly 26.2 27.2 23.0
Ver y or quite poorly 32.0 29.4 27.3
Affective Income Security
Do not worry at all about my/my
household’s economy
18.3 62.6 30.0
Worry a little bit 44.3 23.8 49.9
Do worry a great deal or to some
extent
37.4 13.5 20.1
Nordic journal of working life studies Volume 5 Number 2 June 2015 43
and possibilities that people may receive external nancial support, for example from
parents.
OLS regression (Table 4) is used to estimate specic signicant effects of indepen-
dent variables. In Model 1, cognitive job insecurity has the expected effect on mental
well-being in all three countries, with poorer mental well-being reported among employ-
ees perceiving a large risk of losing their job. Among Finnish respondents, unexpectedly,
the group perceiving neither a large nor small risk of job loss report poorer mental
well-being than the group reporting a large risk of job loss. Employment security has
the expected relationship in Finland and Sweden: those with good perceived opportuni-
ties of nding a new job report better mental well-being than those with poor perceived
opportunities. In Norway, there is no signicant effect of employment security on mental
well-being according to this model. Income security shows the expected relationship in
all three countries, with those perceiving that they would manage well having better
mental well-being.
In Model 2, we include the affective variables, and expect to nd evidence that
they act as mediating variables between the cognitive variables and mental well-being.
Generally, the results seem to conrm this expectation. In Norway and Sweden, affective
job insecurity is negatively related to mental well-being and the distinct effect of cogni-
tive job insecurity is noticeably weakened in all three countries by the large covariation
between the two variables. This indicates that affective job insecurity acts as a mediator.
The same pattern also holds true for income security. Affective income security is
positively related to mental well-being, while the effect of cognitive income security is
much weaker than in Model 1 and hardly signicant. The effect of affective income
security may potentially relate to more than UBs (the variable used to estimate cognitive
income security here). Nevertheless, affective income security seems to act as a mediat-
ing variable, not only for cognitive income security but also for cognitive job insecu-
rity (separate analysis not shown), indicating that the risk of unemployment is directly
related to economic worries. Cognitive employment security still has a unique effect on
mental well-being in Finland and Sweden, showing that including all three forms (job,
employment, and income) is relevant when researching links between security and men-
tal well-being.
The results for cognitive income security among Finnish employees seem to be con-
trary to expectations in Model 2, as negative coefcients indicate that those stating they
would manage well during a period of unemployment report poorer mental well-being
than those stating that they would not manage well. This occurs when we introduce
affective income insecurity, and indicates a suppressed relationship (cf. Aneshensel 2002).
Thus, there are two tendencies in the Finnish data regarding cognitive income security,
with a strong effect on mental well-being in the expected direction and a weaker effect
in the opposite direction. In Model 1 this weaker tendency is suppressed by the stronger
tendency, but it emerges when we introduce affective income insecurity (for reasons that
are currently unclear).
The general pattern still holds after introducing the control variables in Model 3.
Both affective variables still exhibit a unique (but weaker) effect on mental well-being.
This now also includes affective job insecurity among Finnish employees. Most of these
effects do not signicantly differ between the countries, suggesting that the processes
involved are not specic to one country. Examining the cognitive variables, job inse-
curity is not signicant in any of the countries except in the aforementioned category
44 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
Table IV Results of OLS regression analysis: unstandardized b-coefficients for subjective mental well-being
Model 1 Model 2 Model 3
Variables Finland Norway Sweden Finland Norway Sweden Finland Norway Sweden
Cognitive Job Insecurity
(very small or quite small
risk=ref.)
Neither large nor small risk −2.90*** −2.30*** −2.79*** −2.10*** −0.38 −1.16* 1.13** 0.15 −0.38
Very or quite large risk −2.60***
(N)
−5.05***
(F) −4.15*** −1.50** −1.87 −1.31 −0.70 −0.87 −.88
Cognitive Employment Security
(very or quite poor=ref.)
Neither good nor poor 1.08** 0.07 0.68 0.94** −0.15 0.48 0.61 0.59 0.48
Very or quite good 1.80*** (N) 0.57 (F. S) 1.65*** (N) 1.40*** (N) 0.19 (F. S) 1.30***(N) 1.31*** 1.07* 1.34***
Cognitive Income Security
(very or quite poorly=ref.)
Neither well nor poorly 0.09 0.89 1.31** −0.74 0.14 0.32 −0.56 −0.03 0.31
Very or quite well 0.81* 1.07** 2.20*** −0.88* (S) 0.11 0.53 (F) −0.68* (S) −0.14 0.35 (F)
Affective Job Insecurity
(do not worry at all=ref.)
Worry a little bit −0.62 −1.13* −0.56 −0.62 −0.69 −0.42
Worry a great deal or to
some extent −0.80 (N) −4.10***
(F) −2.32** −1.09* (N) −3.46*** (F) −1.80**
Affective Income Security
(worry great deal or to some
extent=ref.)
A little bit 2.47*** 1.53** (S) 3.07*** (N) 1.86*** 1.31** 2.09***
Nordic journal of working life studies Volume 5 Number 2 June 2015 45
Do not worry at all 3.84*** 2.79*** (S) 4.59*** (N) 2.46*** 2.20*** 3.26***
Job Satisfaction (0=satisfied) 3.21*** 3.58** 3.08***
Demand −0.51*** −0.39*** (S) −0.57*** (N)
Social support 0.34*** 0.32*** 0.27***
Civil status (0=married/co-
habitant) −0.39 −0.67 −0.75*
Age 0.03* 0.05** 0.02
Income (high income (6)=ref.)
5 −0.41 3.42 −0.48
4 −0.16 2.85 −0.39
3 −0.20 2.24 −0.54
2 0.07 2.81 −0.26
Low (1) −1.83* 2.57 −0.88
Occupational Category (manual
worker =ref.)
Service workers −0.64 −0.35 −0.17
Semi-professionals −0.94* −0.65 −0.29
Professionals −1.06* −1.10 −0.74
Managers −1.18 −1.24 0.70
Intercept 25.66 (S) 25.49 24.63 (F) 24.97 (S) 24.18 22.92 (F) 23.11 (N) 15.38 (F. S) 22.84 (N)
N 1681 1142 1465 1681 1142 1465 1681 1142 1465
R2adj. 0.066 0.048 0.087 0.112 0.084 0.152 0.305 0.257 0.307
Note: Coefficients significantly different (p < 0.05) in the three countries in bold types, and marked by country (F=Finland, N= Norway, S=Sweden). Levels of significance: *p< 0.05;
**p<0.01; ***p<0.001.
46 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
in Finland. The remainder of the effect after controlling for the affective variable now
co-varies with the control variables. Cognitive employment security now shows a sig-
nicant effect on mental well-being in all three countries. This indicates another sup-
pressed relationship, in Norway, which emerges when the control variables are included.
Employment security per se has an important relationship with mental well-being, but
we should allow for the possibility that this may be due to reversed causality, i.e., that
employees who report poor mental well-being may (consequently) see smaller chances
of nding a new job (Furåker 2010a). Cognitive income security shows no signicant
effect except in Finland, due to the weakening of its effect by the suppressed relation-
ship. Regarding control variables, Model 3 shows that job satisfaction, job demand and
social support are all strongly related to mental well-being. These relationships were
expected and act to further isolate effects of the security variables. Most of the other
control variables have minor or insignicant effects on mental well-being.
To investigate the detected interaction effects (shown in Table 5), we start by exam-
ining Model 1, which was used to study effects of the cognitive variables and interaction
effects. Included, but not shown, are all of the other variables from Table 4 to further
isolate effects of the security variables on mental well-being. The main results from
Model 1 are also presented in Figure 1.
The reference category consists of the interaction effect between low job insecurity
and low employment security. The aim is to emulate the combination of securities usu-
ally accredited to the Nordic model; high job security but also low employment security
(resulting in low exibility), and its relationship to mental well-being. This allows us to
compare how interactions between high job insecurity and various levels of employment
security fare in relation to the reference category, thereby testing how the combination
of securities suggested in the exicurity arrangement relates to mental well-being.
For Norwegian employees, perceiving a large risk of losing the current job while
also perceiving a small chance of nding a new job is related to a substantial reduction
in mental well-being score (more than 8 points). In comparison, those who also perceive
a large chance of nding a new job of equal or better value report a small (1.2 point)
increase in mental well-being. For those who perceive a large risk of losing their job, but
neither a large nor small chance of nding a new job, mental well-being decreases by
almost 2 points. The results are consistent with previous ndings (Berglund et al. 2014),
illustrating that high levels of employment security can provide the means to deal with
the detrimental effects of job insecurity on mental well-being. These ndings are interest-
ing with regards to the claim by exicurity proponents that employment security under
the right conditions should offset the negative relationship between job insecurity and
mental well-being, successfully relating exicurity with well-being.
Looking at the results for Finnish employees, the interaction between high job inse-
curity and low employment security is related to a 2-point reduction in mental well-
being score, compared to the reference interaction. However, high employment security
in combination with high job insecurity is related to an increase of 1.4 points. Among
Swedish employees, perceiving a large risk of job loss and a small chance of nding a
new job is related to a 4.5-point reduction in mental well-being score, compared to
the reference interaction. The interaction with high employment security is related to a
mental well-being score of 0, indicating that the combination of securities suggested in
the exicurity arrangements does not differ in its relationship to mental well-being from
the interaction of low job insecurity and low employment security.
Nordic journal of working life studies Volume 5 Number 2 June 2015 47
Table V Results of OLS regression of interaction effects between job insecurity and employment security
Finland Norway Sweden
Variables Model 1 Model 2 Model 1 Model 2 Model 1 Model 2
Cognitive Interaction effects (Low Job Insecurity
*Low Employment Security = ref.)
High Job Insecurity * Low Employment Security −2.11** (77) −0.99 −8.37*** (14) −5.71*** −4.48*** (36) −2.20*
* Medium Employment Security −1.16 (41) −0.61 −1.88 (13) 0.17 −0.35 (20) 1.42
* High Employment Security 1.37 (43) 1.79* 1.21 (16) 2.49 0.03 (21) 0.72
Medium Job Insecurity * Low Employment Security −1.26* (116) −0.58 −2.61** (22) −1.27 −1.25 (55) −0.21
* Medium Employment Security −1.00 (77) −0.47 1.36 (27) 0.12 −0.99 (43) −0.16
* High Employment Security −0.60 (68) −0.57 2.30 (19) 3.13* −0.01 (46) 0.73
Low Job Insecurity * Medium Employment Security 0.83* (298) 0.74 0.43 (340) 0.36 0.49 (334) 0.32
* High Employment Security 1.66*** (517) 1.37*** 0.79 (405) 0.57 1.52*** (476) 1.27**
Affective Job Insecurity (Do not worry at all = ref.)
Worry a little bit −0.63* −0.83 −0.41
Worry a great deal or to some extent −1.08* −2.91** −1.76**
Intercept 23.66 23.09 17.87 16.34 23.63 22.83
N 1 681 1 681 1142 1142 1465 1465
R2 adj. 0.283 0.306 0.249 0.269 0.279 0.308
Note: The n of the combination of the variables reported in parentheses. Significance levels: *p<0.05; **p<0.01; ***p<0.001.
48 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
When affective job insecurity is introduced in Model 2, the signicant effect of the inter-
action is still present in most of the cases, showing that the cognitive interaction remains,
even when controlling for the affective dimension.
Figure 1 illustrates the value of including the interaction effects in a multiplicative
model rather than in a purely additive model. The horizontal reference line at 0 shows
the mental well-being score associated with the reference category of low job insecurity
and low employment security, while the plotted points show scores (relative to the refer-
ence) associated with combinations of high job insecurity and the three levels of employ-
ment security in each country (and condence intervals of the scores).
The apparent relationship between job insecurity and employment security is more
modest if the interaction effects are excluded. Including the interaction effects reveals a
more nuanced and divergent pattern, showing that the relationship with mental well-
being is not only more negative when employment security is low, but also more positive
when employment security is high. Several of the interactions are based on data from
quite small groups of respondents, as reported in Table 5, but the condence intervals
show that the interaction effects between high and low employment security differ sig-
nicantly in all three countries.
The interaction effects are clearly relevant when trying to discern how insecurity
relates to mental well-being. They also indicate that exicurity arrangements could com-
promise employees’ mental well-being. Under favorable conditions such arrangements
could have positive effects on mental well-being, but a lack of appropriate employment
security in a exicurity arrangement is clearly related to much lower levels of well-being.
Concluding discussion
The presented results show that the tested components of security (job, employment,
and income) are related to mental well-being in Finland, Norway, and Sweden. While
Figure 1: Effects of the interaction between high job insecurity and employment security (catego-
rized as low, medium, and high) compared to the reference category 0 (low job insecurity and low
employment security) on mental well-being.
Nordic journal of working life studies Volume 5 Number 2 June 2015 49
the three countries’ institutional arrangements share some similarities, there are also
differences among them, and this suggests that the processes involved are not specic to
one country, and highly important for understanding how the well-being of employees
relates to insecurity.
All of the three main forms of security and insecurity are independently related to
well-being and all three are of analytical importance. These results underline the need
to view job insecurity as a multidimensional phenomenon. Employment and income
security provide means for mitigating job insecurity, since they are related to better
well-being. Each offers a distinct strategy to deal with an insecure situation. Income
security corresponds to Fryer’s theory of agency restriction, which holds that the main
source of poor well-being is the lack of control and planning resulting from insufcient
nancial resources. Employment security corresponds more closely to Jahoda’s theory of
the latent functions of employment. This suggests that future employment, and associ-
ated prospects of fullling the general latent functions, in itself can act as a viable way
to mitigate the consequences of job insecurity. We come to this conclusion because we
control for economic aspects using income security and the particular combination of
latent functions found in the current job through job insecurity. Both ways should thus
be considered when discussing means to deal with job insecurity.
This leads us to the question as to whether these forms of security can fully negate the
impact of job insecurity on mental well-being, as suggested in exicurity theory. Burchell
(2009) argues that there is no empirical evidence to support this. The results obtained
from our multidimensional model indicate that the exicurity strategy could function
effectively as proposed, provided that appropriate forms of institutional arrangements
are in place. High perceived levels of employment security seem to counter even high lev-
els of job insecurity. Thus, exicurity arrangements could work, but only if employees at
risk of unemployment perceive relatively high levels of employment security. Less than
40% of our respondents fell into this class. This should be compared to the relatively
few employees who experience job insecurity. The success of the exicurity strategy rests
on few employees perceiving employment insecurity in the labor market, which is likely
to require both low levels of unemployment and successful ALMPs. High demand for
labor will probably result in high employment security. Likewise, ALMPs could lead to
high employment security even if there is low demand for labor. However, this rests on
the premise that the ALMPs include effective measures. Employment security could be
achieved by improving the matching of skills to jobs through extensive ALMP measures,
but simply relying, for instance, on matching can be problematic during an economic
downturn, when there are few available jobs, and the exicurity arrangement offers few
other options for raising employment security during economic decline. The exicurity
arrangement is clearly dependent on successful ALMP measures during periods of low
labor demand, corresponding with a previous conclusion that exicurity does not work
well in ‘bad weather’ (Tangian 2010). Furthermore, extensive ALMP measures are costly
and it could prove challenging to nd the political support to implement them fully.
Finally, this study clearly illustrates the relationship between the cognitive and affec-
tive dimensions. In all three countries both cognitive job insecurity and cognitive income
security show a signicant relationship with mental well-being that is strongly reduced
when the relevant affective aspect is controlled for. This behavior is expected and illus-
trates the theoretical understanding presented earlier, that the cognitive perception of
risk generally translates into the affective experience of worry.
50 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
Our study setting has limitations that need to be overcome with new approaches
and further research. Using cross-sectional data, we cannot denitively elucidate the
causal relationships between the variables, for this more research in the eld using panel
data is required. Research on security in general, and exicurity in particular, in the
Nordic welfare states would also benet from including Denmark, as it provides a prime
example of a labor market characterized by exicurity. It would also be useful to include
the level of unemployment as a control variable, to see how it relates to insecurity in the
three countries. It should also be mentioned that a more restricted denition of affective
income security would be useful, since the one used in this article may potentially relate
to economic factors beyond the intended individual or household. Further questions
that warrant attention include the generalizability of the results beyond the Nordic wel-
fare states, and potential differences in the focal relationships elsewhere.
Acknowledgments
The authors thank the following research funding bodies for nancing the study: The
Finnish Work Environment Fund, The Norwegian Research Council’s research program
on welfare, working life and migration (VAM) and the Swedish government agency
VINNOVA.
References
Anderson CJ and Pontusson J (2007) Workers, worries and welfare states: Social protection
and job insecurity in 15 OECD countries. European Journal of Political Research 46:
211–235.
Aneshensel CS (2002) Theory-Based Data Analysis for the Social Sciences. Thousand Oaks:
Pine Forge.
Bengtsson M and Berglund T (2012) Labour market policies in transition: From social engi-
neering to standby-ability. In: Larson B, Letell M and Thörn H (eds) Transformations of
the Swedish Welfare State: From Social Engineering to Governance? Houndmills, Basing-
stoke: Palgrave Macmillan, pp. 86–103.
Berglund T and Furåker B (2011) Flexicurity institutions and labour market mobility. The
International Journal of Comparative Labour Law and Industrial Relations 27: 111–127.
Berglund T, Furåker F and Vulkan P (2014) Is job insecurity compensated for by employment
and income security? Economic and Industrial Democracy. 35(1): 165–148.
Berntson E (2008) Employability Perceptions: Nature, Determinants, and Implications for
Health and Well-Being. Stockholm: Stockholm University.
Berntson E and Marklund S (2007) The relationship between perceived employability and
subsequent health. Work & Stress 21: 279–292.
Burchell B (2009) Flexicurity as a moderator of the relationship between job insecurity
and psychological well-being. Cambridge Journal of Regions, Economy and Society 2:
365–378.
Burchell B (2011) A temporal comparison of the effects of unemployment and job insecurity
on wellbeing. Sociological Research Online. 16(9). Available at: www.socresonline.org.
uk/16/1/9.html (accessed October 10, 2014).
Burroni L and Keune M (2011) Flexicurity: A conceptual critique. European Journal of
Industrial Relations 17: 75–91.
Nordic journal of working life studies Volume 5 Number 2 June 2015 51
Carr E and Chung H (2014) Employment insecurity and life satisfaction: The moderating
inuence of labour market policies across Europe. Journal of European Social Policy
24(4): 383–399.
Cheng G and Chan D (2008) Who suffers more from job insecurity? A meta-analytic review.
Applied Psychology: An International Review 57: 272–303.
Chirumbolo A and Hellgren J (2003) Individual and organizational consequences of job inse-
curity: A European study. Economic and Industrial Democracy 24(2): 217–240.
Chung H and Mau S (2014) Subjective insecurity and the role of institutions. Journal of Eu-
ropean Social Policy 24(4): 303–318.
Chung H and Van Oorschot W (2010) Employment insecurity of European individuals dur-
ing the nancial crisis: A multi-level approach. Working Papers on the Reconciliation of
Work and Welfare in Europe (REC-WP) 14.
Creed PA and Bartrum D (2006) Explanations for deteriorating wellbeing in unemployed
people: Specic unemployment theories and beyond. In Kieselbach T, Wineeld AH, Boyd
C and Anderson S (eds.) Unemployment and Health. International and Interdisciplinary
Perspectives. Brisbane: Australian Academic Press.
De Witte H (1999) Job insecurity and psychological well-being: Review of the literature and
exploration of some unresolved issues. European Journal of Work and Organizational
Psychology. 8: 155–177.
De Witte H (2005) Job insecurity: Review of the international literature on denitions, prev-
alence, antecendents and consequences. Journal of Industrial Psychology 31: 217–240.
Dickerson A and Green F (2012) Fears and realisations of employment insecurity. Labour
Economics 19(2): 198–210.
Ellonen N and Nätti J (2015) Job insecurity and the unemployment rate: Micro- and mac-
ro-level predictors of perceived job insecurity among Finnish employees 1984-2008. Eco-
nomic and Industrial Democracy. 36(1): 57–71.
Ervasti H and Venetoklis T (2010). Unemployment and subjective well-being. An empirical
test of deprivation theory, incentive paradigm and nancial strain approach. Acta Socio-
logica. 53(2): 119–138.
Fryer D (1995) Benet agency? Labour market disadvantage, deprivation and mental health.
The Psychologist. June: 265–272.
Furåker B (2010a) Inlåsning på den svenska arbetsmarknaden [Lock-in effects in the Swedish
labour market]. Arbetsmarknad & Arbetsliv 16(4): 55–70.
Furåker B (2010b) Job insecurity and job change prospects. Views among Employees in Swe-
den with a focus on youth. In: Blanpain R, Bromwich W, Rymkevich O and Spattini S
(eds) Labour Productivity, Investment in Human Capital and Youth Employment. The
Netherlands: Kluwer Law International BV, pp. 3–21.
Garsten C and Jacobsson K (2004) Learning to be employable: An introduction. In:
Garsten C and Jacobsson K (eds.) Learning to be Employable: New Agendas on Work,
Responsibility and Learning in a Globalizing World. Basingstoke: Palgrave Macmillan,
pp. 1–22.
Goldberg D and Williams P (1988) A User’s Guide to the General Health Questionnaire.
Windsor: NFER-Nelson.
Green F (2011) Unpacking the misery multiplier: How employability modies the impacts
of unemployment and job insecurity on life satisfaction and mental health. Journal of
Health Economics 30(2): 265–276.
Greenhalgh L and Rosenblatt Z (1984) Job insecurity: Toward conceptual clarity. Academy
of Management Review 9(3): 438–448.
Hellgren J and Sverke M (2003) Does job insecurity lead to impaired well-being or vice
versa? Estimation of cross-lagged effects using latent variable modeling. Journal of Orga-
nizational Behaviour 24: 215–236.
52 Job Insecurity and Mental Well-Being in Finland, Norway, and Sweden Patrik Vulkan et al.
Howell DR, Baker D, Glyn A and Schmitt J (2007) Are protective labor market institutions
at the root of unemployment? A critical review of the evidence. Capitalism and Society
2: 1–71.
Huang G, Lee C, Ashford S, Chen Z and Ren X (2010) Affective job insecurity: A mediator
of cognitive job insecurity and employee outcomes relationships. International Studies of
Management and Organization 40(1): 20–39.
Huang G, Niu X, Lee C and Ashford S (2012) Differentiating cognitive and affective job in-
security: Antecedents and outcomes. Journal of Organizational Behaviour 33: 752–769.
Jahoda M (1982) Employment and unemployment: A social-psychological analysis. Cam-
bridge: Cambridge University Press.
Karasek R and Theorell T (1990) Healthy Work: Stress, Productivity, and the Reconstruction
of Working Life. New York: Basic Books.
Keim A, Landis R, Pierce C and Earnest D (2014) Why do employees worry about their jobs?
A meta-analytic review of predictors of job insecurity. Journal of Occupational Health
Psychology. 19(3): 269–290.
Klandermans B, Hesselink JK and Van Vuuren T (2010) Employment status and job insecu-
rity: On the subjective appraisal of an objective status. Economic and Industrial Demo-
cracy. 31(4): 557–577.
Lazarus RS and Folkman S (1984) Stress, Appraisal, and Coping. New York: Springer.
Lübke C and Erlinghagen M (2014) Self-perceived job insecurity across Europe over time:
Does changing context matter? Journal of European Social Policy 24(4): 319–337.
Madsen PK (2004) The Danish model of exicurity: Experiences and lessons. Transfer:
European Review of Labour and Research 10: 187–207.
Madsen PK (2006) How can it possibly y? The paradox of a dynamic labour market in a
Scandinavian welfare state. In: Campbell J, Hall J and Pedersen O (eds) National Iden-
tity and the Varieties of Capitalism: The Danish Experience. Montreal: McGill-Queen’s
University Press, pp. 321–355.
Mau S, Mewes J and Schönek N (2012) What determines subjective socio-economic insecuri-
ty? Context and class in comparative perspective. Socio-Economic Review 10: 655–682.
Muffels R and Luijkx (2008) Labour market mobility and employment security of male
employees in Europe: ‘trade-off’ or ‘exicurity’? Work, Employment and Society 22(2):
221–242.
Muffels R, Chung H, Fouarge D, Klammer U, Luijkx R, Manzoni A, Thiel A and Wilthagen T
(2008) Flexibility and Security over the Life Course. Report for the European Foundation
for the Improvement of Working and Living Conditions. 8 February. Dublin.
Muffels R, Crouch C and Wilthagen T (2014) Flexiblity and security: National social models
in transitional labour markets. Transfer 20(1): 99–114.
Nordenmark M, Strand M and Layte R (2006) The impact of unemployment benet system
on the mental well-being of the unemployed in Sweden, Ireland and Great Britain. Euro-
pean Societies 8(1): 83–110.
Nordic Council of Ministers (2010). Labour Market Mobility in Nordic Welfare States. Nor-
dic Council of Ministers, TemaNord 2010: 515.
OECD (2013) Protecting jobs, enhancing exibility: A new look at employment protection
legislation. In Keese M (ed.) OECD Employment Outlook 2013. Paris: OECD Publish-
ing, pp. 65–126.
Ozaki M (1999) Negotiating Flexibility: The Role of the Social Partners and the State.
Geneva: ILO.
Scruggs L (2002) The Ghent system and union membership in Europe, 1970-1996. Political
Research Quarterly 55(2): 275–297.
Selenko E and Batinic B (2013) Job insecurity and the benets of work. European Journal of
Work and Organizational Psychology 22(6): 725–736.
Nordic journal of working life studies Volume 5 Number 2 June 2015 53
Silla I, De Cuyper N, Gracia FJ, Peir JM and De Witte H (2009) Job insecurity and well-being:
Moderation by employability. Journal of Happiness Studies 10(6): 739–751.
Sjöberg O (2010) Social insurance as a collective resource: Unemployment benets, job
insecurity and subjective well-being in a comparative perspective. Social Forces 88:
1281–1304.
Sverke M and Hellgren J (2002) The nature of job insecurity: Understanding employment
insecurity on the brink of a new millennium. Applied Psychology: An International
Review 51: 23–42.
Tangian A (2008) Is Europe ready for exicurity? Empirical evidence, critical remarks and a
reform proposal. Intereconomics 43: 99–111.
Tangian A (2010) Not for bad weather: Flexicurity challenged by the crisis. ETUI Policy
Brief, Issue 3, Brussels.
Venn D. (2009) Legislation, collective bargaining and enforcement: Updating the OECD
employment protection indicators. Available at: www.oecd.org/els/workingpapers.
Vulkan P (2012) Labour market insecurity: The effects of job, employment and income inse-
curity on the mental well-being of employees. Revista Internacional de Organizaciones
9:169–188.
Wilthagen T and Tros F (2004) The concept of ‘exicurity’: A new approach to regulating
employment and labour markets. Transfer: European Review of Labour and Research
10: 166–186.
End note
1 Danish employees and arrangements were also covered in the original study design, but
lack of funding for a Danish survey prevented their inclusion.
2 OECD (2013) Employment protection legislation: Strictness of employment protection
legislation: overall. Index 3, OECD Employment and Labour Market Statistics (data-
base). doi: 10.1787/lfs-data-en (accessed on October 14, 2014).
3 OECD (2013) Social expenditure: Aggregated data. OECD Social Expenditure Statistics
(database). doi: 10.1787/data-00166-en (accessed on October 14, 2014).
4 OECD (2013) Tax-benet models, http://www.oecd.org/els/soc/benetsandwagestax-ben-
etcalculator.htm (accessed on October 14, 2014).
5 OECD (2013) Unemployment rate. Employment and Labour Markets: Key Tables from
OECD. doi:10.1787/20752342 (accessed on October 14, 2014).
