681590

research-article2016
ISP0010.1177/0020764016681590International Journal of Social PsychiatryYoon et al.

E CAMDEN SCHIZOPH

Original Article

International Journal of

The effect of job loss on depressive Social Psychiatry

1­–6
© The Author(s) 2016
symptoms: The results from the Korean Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav

Welfare Panel Study (2007–2013) DOI: 10.1177/0020764016681590

isp.sagepub.com

Hyo Jung Yoon1,2, Jae Woo Choi1,2, Suk-Young Jang2,3,

Sang Ah Lee1,2 and Eun-Cheol Park2,3

Abstract
Background: Unemployment is closely associated with depressive symptoms. We conducted analysis to find whether
the job loss increased the risk of depressive symptoms according to job status, occupation and tenure.
Methods: Data were collected from Korean Welfare Panel Study (KOWEPS) from 2007 to 2013. To measure the
increase in depressive score, we selected respondents who answered for at least 2 years of continuous waves. We
performed a longitudinal analysis by generalized estimating equation (GEE) method with a total of 19,399 cases. Among
them, 608 cases (3.1%) experienced job loss.
Results: There was a significant rise in depressive score after job loss (β = 1.34, p = .000). In subgroup analysis, precarious
and low-tenure workers are considerably more depressed after job loss (precarious: β = 1.98, p < .0001, low-tenure
workers: β = 1.31, p = .001). Both white and blue collar workers showed a rise in depressive score significantly (white
collar: β = 1.16, p = .031; blue: β = 2.03, p = .001).
Conclusion: The results showed that precarious and low-tenure workers were relatively vulnerable in mental health
after experiencing job loss with low and negative expectation for re-employment. It implies that supports for encouraging
work skill and financial supports during the unemployment period should be needed.

Keywords
Unemployment, job loss, depression, low-tenure worker, precarious workers

Introduction
Unemployment is significantly associated with both poorer
physical functioning and mental health (Bartley, 1994; Deb,
Gallo, Ayyagari, Fletcher, & Sindelar, 2011; Falba, Teng,
Sindelar, & Gallo, 2005; Linn, Sandifer, & Stein, 1985;
McKee-Ryan, Song, Wanberg, & Kinicki, 2005; Singh &
Siahpush, 2016). Unemployment adversely affects health
outcomes through various pathways (Bartley, 1994; McKee-
Ryan et al., 2005): deterioration of economic conditions;
socio-psychological distress such as lowering social contact
or losing social status (Isaksson, 1989; Linn et al., 1985) and
increasing damaging behavior to health, such as smoking,
alcohol consumption and obesity (Deb et al., 2011; Falba
et al., 2005). Unemployment is linked with major depres-
sion, regardless of gender and age, generally with a greater
effect in men and in middle-aged persons (Andreeva,
Magnusson Hanson, Westerlund, Theorell, & Harvey
Brenner, 2015; Artazcoz, Benach, Borrell, & Cortès, 2004;
Backhans & Hemmingsson, 2012; S. N. Jang et al., 2009).
Greater socio-economic status, such as a higher level of
education or income, has been found to moderate depressive
symptoms, whereas higher occupation prestige has been
shown to degenerate depressive symptoms (Artazcoz et al.,
2004; Berchick, Gallo, Maralani, & Kasl, 2012; Gallo,
Bradley, Dubin, et al., 2006; S. Park, Cho, & Jang, 2012).
Unemployment can also potentially lead to negative health
outcomes for family members (Bunnings, Kleibrink, &
Wessling, forthcoming; H. Park & Kang, 2016). However,
as unemployment experiences recur, the effects on overall
depressive symptoms have been shown to weaken (Gallo,
Bradley, Teng, & Kasl, 2006).
1Department of Public Health, Graduate School, Yonsei University,
Seoul, Republic of Korea
2Institute of Health Services Research, Yonsei University, Seoul,
Republic of Korea
3Department of Preventive Medicine, Yonsei University College of
Medicine, Seoul, Republic of Korea
Corresponding author:
Eun-Cheol Park, Department of Preventive Medicine and Institute of
Health Services Research, Yonsei University College of Medicine,
50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea.
Email: ecpark@yuhs.ac

Downloaded from isp.sagepub.com at BOSTON UNIV on December 2, 2016

2 International Journal of Social Psychiatry
After an economic crisis, high unemployment rates
tend to increase worldwide. In Korea, the unemployment
rate skyrocketed in 2009 from 3.2% to 3.6%. It peaks at
3.7% in 2010. Although it decreased thereafter, it has now
increased again to 3.6% as of 2015. When it comes to the
real unemployment rate that includes part-time workers,
marginally attached workers (those who are interested in
working, but not actively seeking employment) and dis-
couraged workers (those who had given up finding a job
after long-term unemployment), the situation worsens; it is
nearly 3 times higher than the widely reported unemploy-
ment rate.
A high prevalence of depression has emerged as one of
the most significant issues in public health. Previous stud-
ies have demonstrated that depressive symptoms profiles
are similar regardless of region (Chee et al., 2015) and that
the burden of mental disorders is considerable (Ferrari
et al., 2014). A recent Korean study showed the prevalence
of depressive symptoms of 9%, 11% and 14%, respec-
tively, in age groups of 40–49, 50–59 and 60–69 years (Oh
et al., 2013). Furthermore, Korea has the highest suicide
rate among Organization for Economic Co-operation and
Development (OECD) countries, and it is well known that
depressive symptoms are the main cause of suicide.
In this study, we assessed changes in depressive scores
caused by a change in employment status using Korean
Welfare Panel Study (KOWEPS) data. First, we examined
depressive scores for people whose employment status
switched to being unemployed. Then, analysis was con-
ducted separately by previous job status, occupation and
tenure. We hypothesized that permanently employed and
skilled workers would more likely be re-employed; there-
fore, they would show less depression than low-skilled
workers.
Methods
Study population
Data were acquired from the KOWEPS waves from 2007
to 2013. The KOWEPS is conducted by the Korean
Institute of Social and Health Affairs in conjunction with
the Social Welfare Research Institute of Seoul National
University. Detailed information on families and individu-
als was collected from a nationally representative sample
of households. Although the KOWEPS started in 2006, we
used data from the second wave because questionnaires
were adjusted from the first wave. The initial 2007 base-
line data included 17,478 individuals from 6,580 house-
holds. In this study, we selected participants who were
working for pay, but not self-employed between 20 and
65 years of age. To measure the change in employment and
depressive scores, we limited respondents who reported at
least 2 years of continuous employment and depressive
scores. To avoid the healthy worker effect, we eliminated
respondents who indicated the reason for job loss as a
Downloaded from isp.sagepub.com at BOSTON UNIV on December 2, 2016
health problem. A total of 19,399 samples without any
missing variables remained in this study.
Outcome variable
In the KOWEPS, depressive symptoms were measured by
the Korean version of the Center for Epidemiological
Studies Depression (CES-D) scale. The scale measures the
1-week prevalence of 11 depressive symptoms on a 4-point
(0–3) scale (Cho & Kim, 1998). The total sum score was
multiplied by 20/11 to match the standard CES-D 20 score.
Primary variable
The KOWEPS consists of the study participants’ employ-
ment status at the last day (31 December) of each year
from 2007 to 2012. We classified full-time or precarious
workers as ‘employed’ and job seekers who actively had
applied for jobs during the previous 4 weeks as ‘unem-
ployed’. A precarious worker referred to individuals who
have signed a temporal employment contract with a low
certainty of continued employment. Respondents who
were not economically active, an unpaid family worker or
self-employed were excluded. People who reported unem-
ployment during a follow-up wave were categorized as
experiencing job loss.
Covariates
Demographic, socioeconomic, health-related and job-
related factors were selected in this study. Demographic fac-
tors included sex, age (19–34, 35–44 and 45–64 years),
marital status (married and single) and region (urban and
rural). Indicators of socioeconomic status included educa-
tion level (elementary school or below, middle school or
high school, college or above) and household income (low,
middle and high). Health-related factors included perceived
health (low, medium and high) and current smoking status
(yes, no and no answers). We made a category of ‘no answer’
for smoking status to include a number of respondents who
omitted this answer. We also included year to consider social
and economic circumstances. Job-related factors included
job status (permanent, self-employed and precarious), par-
ticipants’ occupation (white, blue collar, and sales and ser-
vice) and tenure (1–4 and ⩾5 years) of previous wave. To
categorize the job status, we used a previous study’s defini-
tion (S. Y. Jang, Jang, Bae, Shin, & Park, 2015).
Statistical analysis
To determine the study population, general characteristics,
t-tests and analysis of variance (ANOVA) were performed
between covariates categories. Then, we applied a general-
ized estimating equation (GEE) to analyze the longitudinal
relationship between job loss/gain event and depression
score. We then separately examined the relationships
Yoon et al. 3
stratified by age. All statistical analyses ware performed
using SAS version 9.4 (SAS Institute, Inc., Cary, NC,
USA).
Results
Table 1 shows the mean CES-D scores in relation to the
general characteristics of the study population. About 3%
of the study cases lost their job and the rest (97%) remained
employed. Their depression scores were 8.17 ± 9.61 and
5.16 ± 7.01, respectively, a statistically significant differ-
ence (p < .0001). Precarious workers reported considerably
greater depressive scores (precarious: 6.65 ± 6.47 and per-
manent: 4.34 ± 4.23). Low-tenure workers who had worked
less than 5 years showed higher depressive scores of about
1 unit (1–4 years: 5.91 ± 7.58 and ⩾5 years: 4.42 ± 6.4).
Table 2 shows the results from the GEE model reflect-
ing the linear effect of job loss on change in depression
score in regard to various variables. After controlling for
covariates, the results showed that job loss groups experi-
ence significant increases in depression, compared to indi-
viduals who remain employed (β = 1.34, p = .000).
Finally, the effect of unemployment on depressive
scores was conducted separately for job status, occupation
and tenure (Table 3). The depression score was signifi-
cantly increased for job losers who had been precarious
workers (β = 1.98, p  < 
.0001), compared to precarious
workers who remained employed. Regarding occupation,
both white collar and blue collar workers showed signifi-
cant differences in depression after being unemployed
(white collar β = 1.16, p = .031, blue collar β = 2.03,
p = .001), while sales and service workers did not quite
achieve significance (β = .46, p = .490). Low-tenure work-
ers showed considerable increases in depressive score (β:
1.31, p = .001) after job loss. Skilled workers who had
worked for more than 5 years of tenure failed to reach sta-
tistical significance (β = 1.38, p = .089).
Discussion
The aim of this study was to identify whether job loss is
closely associated with subsequent depressive scores. We
found that people who lost their job showed significantly
higher depressive scores than people who remained
employed. Additionally, our present findings revealed the
degree of increase therein differs according to previous job
status, occupation and tenure. Precarious workers and
lower-tenure workers showed significant increases in
depression scores after job loss. Depressive scores ele-
vated significantly after job loss for both white collar and
blue collar workers, while results for sales and service
workers did not achieve significance.
Our results are consistent with other studies. Comparing
unemployed individuals and to those with continued
employment, the odds ratio of an increased risk of depres-
sive symptoms was 2.08 in one panel study in the United
Downloaded from isp.sagepub.com at BOSTON UNIV on December 2, 2016
States (Dooley, Catalano, & Wilson, 1994). In a cross-sec-
tional study in Hong Kong, unemployed women showed a
higher prevalence rate of disturbance than employed
women (Lai, Chan, & Luk, 1997). Another Korean study
showed that unemployment was associated with new onset
of depressive symptoms, and precarious workers exhibited
a high odds ratio of depressive symptoms of 2.3 (Kim,
Subramanian, Sorensen, Perry, & Christiani, 2012).
To explain mental health problems after unemploy-
ment, we considered Warr’s theory (Warr, Jackson, &
Banks, 1988). In the theory, there are nine environmental
features that determine employment wellbeing, such as
opportunity for skill use (feature 2), environmental clarity
(feature 5) and availability of money (feature 6).
Unemployment deprives individuals of these features;
therefore, job losers can be easily exposed to bad mental
health (Warr et al., 1988).
Unemployed people typically lose financial certainty,
which is a determinant of wellbeing while under employ-
ment (feature 6). It has a greater effect on economically
vulnerable people (Gallo, Bradley, Dubin, et al., 2006).
Precarious and low-tenure jobs tend to pay less and lack
benefits. Therefore, they are more likely to be less pre-
pared with financial resources, such as savings and
investments, as a buffer from earning loss. Previous
research showed that semi-skilled and unskilled manual
workers report more financial problems and worry after
unemployment than professional and managerial work-
ers (Payne, Warr, & Hartley, 1984). Weakness of social
assistance also influenced depressive scores. In Korea,
unemployment benefits are provided only for the first
year of unemployment, with only a 40% net replacement
rate for the initial phase of unemployment, which is
below the OECD average of 58% (OECD, 2016).
Insufficient support might make job seekers feel pres-
sure from financial strain and it could be connected to
less satisfied re-employment that carries a high risk of
re-exposure to unemployment (Gallo, Bradley, Dubin,
et al., 2006).
Another factor that may have influenced the differences
in depressive scores is environmental unclarity with lower
expectations of re-employment (feature 6) and little poten-
tial for skill development (feature 2). The rate of re-
employment in 1 year after displacement for low-skilled
and low-educated people is relatively lower than that for
high-skilled and high-educated workers in most countries
(OECD, 2013). Precarious workers who are typically less
educated (Kim et al., 2012) and lower-tenure workers who
lack skill might feel restrictions with finding a new job.
Therefore, job loss for these workers could easily be con-
nected to long-term unemployment or involuntary retire-
ment. Moreover, the fact that about 60% of instances of
re-employment resulted in changes in occupation and skill
requirements (OECD, 2013) implies that there might be a
vicious cycle in which low-skilled workers remain low-
skilled permanently.
4 International Journal of Social Psychiatry

Table 1.  General characteristics of study population. Table 2.  Results of the GEE analyzing for the effect of job loss
on depressive score.
CES-D score
Variable CES-D score
  Cases (%) Mean ± SD p value
β SE p value
Employment status
  Job loss 608 (3.1) 8.17 ± 9.61 <.0001 Employment status
 Remaining 18,791 (96.9) 5.16 ± 7.01     Job loss 1.34 .35 .000
employed   Remaining employed Ref  
Job status (previous wave) Job status (previous wave)
 Permanent 11,796 (60.8) 4.34 ± 4.23 <.0001  Precarious .47 .11 <.0001
 Precarious 7,603 (39.2) 6.65 ± 6.47    Permanent Ref  
Occupation (previous wave) Occupation (previous wave)
  White collar 7,859 (40.5) 4.2 ± 6.17 <.0001   Sales and service .24 .14 .096
  Sales and service 3,395 (17.5) 6.12 ± 7.78     Blue collar .20 .13 .124
  Blue collar 8,145 (42.0) 5.9 ± 7.56     White collar Ref  
Tenure (previous wave), years Tenure (previous wave), years
 1–4 10,801 (55.7) 5.91 ± 7.58 <.0001  1–4 .33 .10 .001
  ⩾5 8,598 (44.3) 4.42 ± 6.4     ⩾5 Ref  
Sex Sex
 Female 11,193 (57.7) 4.51 ± 6.43 <.0001  Female 1.06 .13 <.0001
 Male 8,206 (42.3) 6.26 ± 7.86    Male Ref  
Age (years) Age (years)
 20–34 5,235 (27.0) 4.9 ± 6.91 <.0001  35–44 .28 .13 .028
 35–44 6,686 (34.5) 4.81 ± 6.74    45–65 .37 .14 .009
 45–65 7,478 (38.6) 5.88 ± 7.55  
 20–45 Ref  
Marital status
Marital status
 Married 13,287 (68.5) 4.69 ± 6.57 <.0001
 Single .60 .13 <.0001
 Single 6,112 (31.5) 6.46 ± 8.06  
 Married Ref  
Region
Region
 Urban 9,634 (49.7) 5.49 ± 7.29 <.0001
 Rural .15 .10 .130
 Rural 9,765 (50.3) 5.02 ± 6.95  
 Urban Ref  
Education level
Education level
 Elementary or 1,389 (7.2) 8.28 ± 8.99 <.0001
below   Elementary or below .26 .27 .332
 Middle/high 12,108 (62.4) 5.52 ± 7.29     Middle/high school .13 .12 .278
school   College or above Ref  
 College or above 5,902 (30.4) 3.98 ± 5.89   Household income
Household income  Low 1.78 .20 <.0001
 Low 2,456 (12.7) 8.91 ± 9.44 <.0001  Middle .54 .10 <.0001
 Middle 7,287 (37.6) 5.56 ± 7.27    High Ref  
 High 9,656 (49.8) 4.08 ± 5.88   Perceived health status
Perceived health status  Low 4.22 .20 <.0001
 Low 2,000 (10.3) 10.22 ± 10.13 <.0001  Middle 2.11 .10 <.0001
 Middle 4,266 (22.0) 7.07 ± 7.89    High Ref  
 High 13,133 (67.7) 3.9 ± 5.7   Smoking
Smoking  Yes .66 .13 <.0001
 No 6,610 (34.1) 4.29 ± 6.34 <.0001  Omitted −.02 .15 .875
 Yes 5,925 (30.5) 5.32 ± 7.26    No Ref  
 Omitted 6,864 (35.4) 6.12 ± 7.58   Year
Year  2008 1.32 .18 <.0001
 2008 3,254 (16.8) 6.96 ± 7.86 <.0001  2009 1.58 .19 <.0001
 2009 3,219 (16.6) 6.74 ± 8.05    2010 .12 .17 .465
 2010 3,209 (16.5) 4.99 ± 7.01    2011 .20 .16 .209
 2011 3,270 (16.9) 4.66 ± 6.48    2012 −.39 .14 .007
 2012 3,215 (16.6) 4 ± 6.18    2013 Ref  
 2013 3,232 (16.7) 4.13 ± 6.34   Depress score (previous wave) .27 .01 <.0001
CES-D: Center for Epidemiological Studies Depression; SD: standard GEE: generalized estimating equation; CES-D: Center for
deviation. Epidemiological Studies Depression; SE: standard error.

Downloaded from isp.sagepub.com at BOSTON UNIV on December 2, 2016

Yoon et al. 5
Table 3.  Results of the GEE analyzing for the effect of job loss
on depressive score.
Variable CES-D score
Job loss
β SE p value
Job status (previous wave)
 Permanent .69 .48 .156 Ref
 Precarious 1.98 .50 <.0001 Ref
Occupation (previous wave)
  White collar 1.16 .54 .031
  Sales and service .46 .67 .490
  Blue collar 2.03 .59 .001
Tenure (previous wave), years
 1–4 1.31 .39 .001
  ⩾5 1.38 .81 .089
GEE: generalized estimating equation; CES-D: Center for
Epidemiological Studies Depression; SE: standard error.
Adjusted for sex, age, marital status, region, education level, household
income, perceived health status, smoking and year.
Although precarious or low-tenure workers may be able
to get a new job, the job characteristic is likely to be poorer
than their previous work, with a lower wage, increasing
working hours and increasing insecurity (OECD, 2013).
After a financial crisis, the possibility of getting a perma-
nent job gets smaller due to increasing flexibility in the
labor market. Moreover, unemployment scarring has a neg-
ative effect on subsequent job quality (Dieckhoff, 2011).
Thus, precarious workers are at a greater risk of subsequent
temporary employment (Giesecke & Gross, 2003).
In terms of occupation, in this study, it seemed that poor
work quality among sales and service workers was associ-
ated with the result. Previous studies have shown that
employment is not always associated with better health
outcomes, but that they might be worse if work quality is
poor (Butterworth, Leach, McManus, & Stansfeld, 2013).
As sales and service workers might already show depres-
sive symptoms during their employment, job loss might
have little effect on depression scores.
We conducted this study to examine the relationship
between job loss and depressive scores. This analysis goes
further than previous researches by separating the analysis
according to previous job status, occupation and working
duration. Our findings support Warr’s theory (Warr, 1987)
and are consonant with other studies that have measured
the impact of unemployment on mental health according to
job status and occupation type (Warr et al., 1988; Little,
1976; Kim et al., 2012). We conducted analyses with more
recent data, and to the best our knowledge, this is the first
approach to report on a connection between depression
and job loss separated by previous work tenure.
There are some limitations to be considered. First, the
issue of a healthy worker effect cannot be excluded: work-
ers who were hired are more likely to be healthier both
physically and psychologically than unemployed people.
Downloaded from isp.sagepub.com at BOSTON UNIV on December 2, 2016
At first, we tried to control this issue using the question that
asked the reason of resignation. However, the data could be
biased because the answers depended on the perceptions of
the respondents, and persons may have no regard on their
Remaining health condition and instead may be much more concerned
employed with external factors like layoff as the resignation reason.
Therefore, we tried to control for the healthy worker effect
by including perceived health status and smoking con-
sumption as covariates. Second, depressive symptoms
related to employment change may be moderated or rein-
forced during the time between job loss event and the
Ref measured symptoms. Because KOWEPS respondents are
  asked every termination of year, there may be other unob-
Ref served life-threatening or moderating event that occurred
between waves. In the same way, employment status might
Ref
not be exact because we set the variable using employment
Ref
status at the end of year of each wave. Therefore, we might
have missed job-related events that could not be captured in
KOWEPS data. For example, the remaining employment
group might include people who had experienced job loss,
but were re-employed before the last day of the same year.
Third, we used self-reported symptoms and diagnoses of
mental health, rather than objective clinical assessment.
Nevertheless, the CES-D questionnaire is widely used to
assess the prevalence of depressive symptoms. Fourth, we
could not examine the reverse effect of unemployment.
Getting employed can help to enhance the mental health
and wellbeing of people who experience mental health
problems (Perkins & Daisy Choy, 1997). Therefore, further
studies are needed to measure whether employment could
decrease or mediate depressive symptoms.
In conclusion, we found that precarious, lower-tenure
workers experience more depression when they become
unemployed. They appear to be at higher risk of recurrent
unemployment and a continuous lack of work career. This
implies that support for encouraging work skills, such as
vocational training, may be needed to disconnect this vicious
cycle and to cross the job barrier. To do this, expanding unem-
ployment benefits should also be considered. Vulnerable
workers are more likely to suffer from earning loss and might
be in a hurry for re-employment. In turn, they might not have
enough time for vocational education. Financial support
would work as a buffer against earning loss and allow vulner-
able workers the time to receive education.
Conflict of interest
The author(s) declared no potential conflicts of interest with respect
to the research, authorship and/or publication of this article.
Funding
The author(s) received no financial support for the research,
authorship and/or publication of this article.
References
Andreeva, E., Magnusson Hanson, L. L., Westerlund, H.,
Theorell, T., & Harvey Brenner, M. (2015). Depressive
symptoms as a cause and effect of job loss in men and
6 International Journal of Social Psychiatry
women: Evidence in the context of organisational downsiz-
ing from the Swedish Longitudinal Occupational Survey of
Health. BMC Public Health, 15, Article 1045.
Artazcoz, L., Benach, J., Borrell, C., & Cortès, I. (2004).
Unemployment and mental health: Understanding the
interactions among gender, family roles, and social class.
American Journal of Public Health, 94, 82–88.
Backhans, M. C., & Hemmingsson, T. (2012). Unemployment
and mental health – Who is (not) affected? European
Journal of Public Health, 22, 429–433.
Bartley, M. (1994). Unemployment and ill health: Understanding
the relationship. Journal of Epidemiology & Community
Health, 48, 333–337.
Berchick, E. R., Gallo, W. T., Maralani, V., & Kasl, S. V. (2012).
Inequality and the association between involuntary job loss
and depressive symptoms. Social Science & Medicine, 75,
1891–1894.
Bunnings, C., Kleibrink, J., & Wessling, J. (forthcoming). Fear
of unemployment and its effect on the mental health of
spouses. Health Economics.doi:10.1002/hec.3279
Butterworth, P., Leach, L. S., McManus, S., & Stansfeld, S. A.
(2013). Common mental disorders, unemployment and psy-
chosocial job quality: Is a poor job better than no job at all?
Psychological Medicine, 43, 1763–1772.
Chee, K. Y., Tripathi, A., Avasthi, A., Chong, M. Y., Xiang, Y.
T., Sim, K., ... Sartorius, N. (2015). Country variations in
depressive symptoms profile in Asian countries: Findings
of the Research on Asia Psychotropic Prescription (REAP)
studies. Asia-Pacific Psychiatry, 7, 276–285.
Cho, M. J., & Kim, K. H. (1998). Use of the Center for
Epidemiologic Studies Depression (CES-D) Scale in Korea.
The Journal of Nervous and Mental Disease, 186, 304–310.
Deb, P., Gallo, W. T., Ayyagari, P., Fletcher, J. M., & Sindelar,
J. L. (2011). The effect of job loss on overweight and drink-
ing. Journal of Health Economics, 30, 317–327.
Dieckhoff, M. (2011). The effect of unemployment on subse-
quent job quality in Europe: A comparative study of four
countries. Acta Sociologica, 54, 233–249.
Dooley, D., Catalano, R., & Wilson, G. (1994). Depression and
unemployment: Panel findings from the Epidemiologic
Catchment Area study. American Journal of Community
Psychology, 22, 745–765.
Falba, T., Teng, H. M., Sindelar, J. L., & Gallo, W. T. (2005).
The effect of involuntary job loss on smoking intensity and
relapse. Addiction, 100, 1330–1339.
Ferrari, A. J., Norman, R. E., Freedman, G., Baxter, A. J., Pirkis,
J. E., Harris, M. G., ... Whiteford, H. A. (2014). The bur-
den attributable to mental and substance use disorders as
risk factors for suicide: Findings from the Global Burden of
Disease Study 2010. PLoS ONE, 9, e91936.
Gallo, W. T., Bradley, E. H., Dubin, J. A., Jones, R. N., Falba,
T. A., Teng, H. M., & Kasl, S. V. (2006). The persistence
of depressive symptoms in older workers who experi-
ence involuntary job loss: Results from the health and
retirement survey. The Journals of Gerontology, Series
B: Psychological Sciences & Social Sciences, 61, S221–
S228.
Gallo, W. T., Bradley, E. H., Teng, H. M., & Kasl, S. V. (2006).
The effect of recurrent involuntary job loss on the depres-
sive symptoms of older US workers. International Archives
of Occupational and Environmental Health, 80, 109–116.
Giesecke, J., & Gross, M. (2003). Temporary employment: Chance
or risk? European Sociological Review, 19, 161–177.
Downloaded from isp.sagepub.com at BOSTON UNIV on December 2, 2016
Isaksson, K. (1989). Unemployment, mental health and the psycho-
logical functions of work in male welfare clients in Stockholm.
Scandinavian Journal of Social Medicine, 17, 165–169.
Jang, S. N., Cho, S. I., Chang, J., Boo, K., Shin, H. G., Lee, H.,
& Berkman, L. F. (2009). Employment status and depres-
sive symptoms in Koreans: Results from a baseline survey
of the Korean Longitudinal Study of Aging. The Journals
of Gerontology, Series B: Psychological Sciences & Social
Sciences, 64, 677–683.
Jang, S. Y., Jang, S. I., Bae, H. C., Shin, J., & Park, E. C. (2015).
Precarious employment and new-onset severe depressive
symptoms: A population-based prospective study in South
Korea. Scandinavian Journal of Work, Environment &
Health, 41, 329–337.
Kim, S. S., Subramanian, S., Sorensen, G., Perry, M. J., &
Christiani, D. C. (2012). Association between change in
employment status and new-onset depressive symptoms in
South Korea – A gender analysis. Scandinavian Journal of
Work, Environment & Health, 38, 537–545.
Lai, J. C., Chan, R. K., & Luk, C.-L. (1997). Unemployment and
psychological health among Hong Kong Chinese women.
Psychological Reports, 81, 499–505.
Linn, M. W., Sandifer, R., & Stein, S. (1985). Effects of unem-
ployment on mental and physical health. American Journal
of Public Health, 75, 502–506.
Little, C. B. (1976). Technical-professional unemployment:
Middle-class adaptability to personal crisis. The Sociological
Quarterly, 17, 262–274.
McKee-Ryan, F., Song, Z., Wanberg, C. R., & Kinicki, A. J.
(2005). Psychological and physical well-being during
unemployment: A meta-analytic study. Journal of Applied
Psychology, 90, 53–76.
OECD. (2013). Back to work: Re-employment, earnings and
skill use after job displacement. Paris: OECD Employment
Outlook 2013, OECD Publishing.
OECD. (2016). Society at a Glance 2016. Paris: OECD Social
Indicators, OECD Publishing.
Oh, D. H., Kim, S. A., Lee, H. Y., Seo, J. Y., Choi, B. Y., & Nam, J.
H. (2013). Prevalence and correlates of depressive symptoms
in Korean adults: Results of a 2009 Korean community health
survey. Journal of Korean Medical Science, 28, 128–135.
Park, H., & Kang, M. Y. (2016). Effects of voluntary/involuntary
retirement on their own and spouses’ depressive symptoms.
Comprehensive Psychiatry, 66, 1–8.
Park, S., Cho, S. I., & Jang, S. N. (2012). Health conditions sensi-
tive to retirement and job loss among Korean middle-aged
and older adults. Journal of Preventive Medicine & Public
Health, 45, 188–195.
Payne, R., Warr, P., & Hartley, J. (1984). Social class and psy-
chological ill-health during unemployment. Sociology of
Health & Illness, 6, 152–174.
Perkins, R. B., & Daisy Choy, R. (1997). Access to employment:
A supported employment project to enable mental health
service users to obtain jobs within mental health teams.
Journal of Mental Health, 6, 307–318.
Singh, G. K., & Siahpush, M. (2016). Inequalities in US Life
Expectancy by Area Unemployment Level, 1990-2010.
Scientifica (Cairo), 2016, Article 8290435.
Warr, P. (1987). Work, unemployment, and mental health.
Oxford, UK: Oxford University Press.
Warr, P., Jackson, P., & Banks, M. (1988). Unemployment
and mental health: Some British studies. Journal of social
issues, 44, 47–68.