It is well documented that unemployment is associated with health deterioration [1]. Mental health is affected by unemployment [2,3,4], leading to, for example, depression and anxiety [5]. Unemployed people report poorer self-assessed health [6, 7] and utilize healthcare services more often [8,9,10]. Excess mortality also tends to be high among the unemployed [11, 12]. Furthermore, cross-national comparative studies have revealed that unemployment is harmful for health in all countries covered [13, 14].
Financial strain appears to play a role in explaining why mental health is affected by unemployment [15, 16], whereas elevated levels of inflammatory markers (e.g., C-reactive protein and interleukin 6) represent one path from unemployment to somatic health conditions [17,18,19]. Health-related social mobility – that is, the impact of poor health status on, for example, educational attainment, occupational careers, and job loss likelihood – is also of importance for the strength of the unemployment-health association [20, 21].
A handful of previous studies have found no evidence of a negative causal health effect of unemployment within a counterfactual framework [22,23,24]. However, there is broad agreement overall in the existing literature that unemployment is harmful for various aspects of health [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19]. Whether the health consequences of unemployment differ between males and females is more disputed, though. Some studies have found that men are more prone to health consequences than women post-unemployment [13, 25,26,27,28], whereas other studies have found that women are equally or more affected by unemployment [9, 29,30,31,32,33].
Inconsistencies in the existing literature can probably be explained partly by differing data materials, health outcomes, and length of follow-up. The ‘gender health paradox’ could be one potential explanation. Females report more health problems and utilise healthcare services more often, whereas males die earlier, on average[34, 35]. This paradox may, to some extent, explain why existing studies disagree on how gendered the health consequences of unemployment really are. For example, two studies from Sweden published in 2011–12 showed diverging findings [27, 31]. On the one hand, a register-based study reported that the effects of unemployment on all-cause mortality are more pronounced among men than women [27]. A survey-based study, on the other hand, indicated more self-reported mental health problems among unemployed females compared to their male counterparts [31]. Comparing results across differing health outcomes could therefore reveal important insights.
The presence or absence of strong cultural expectations for the man to be the main financial provider for his family (i.e., the male breadwinner model) might matter as well [36, 37]. According to social role theory, women have multiple roles to alternate between (i.e., mother, wife, friend, worker), whereas the worker role tends to be more crucial for men [38, 39]. Since women can alternate between, and gain recognition from, various social roles, it may be easier for women to deal with the experience of unemployment. Men’s social identity, by comparison, is more tightly connected to work and employment, and job loss could therefore prove to be more harmful. However, gender differences in the importance of the worker role have probably become smaller over time, as, for example, women have increasingly entered (previously) male-dominated occupations, female rates of part-time work have decreased, men take more responsibility for childrearing and housework, et cetera. Accordingly, gender differences in the health consequences of unemployment should decrease over time as society grows gradually more gender equal.
Social role theory also predicts that health consequences will be particularly gendered (i.e., men experience graver consequences) in countries and regions where the male breadwinner model prevails. Conversely, we should observe rather small gender differences in countries and regions where gender norms are comparatively egalitarian. Two recent papers [40, 41] analysing the German Socio-Economic Panel found some support for this theoretical model, showing that males are hurt more by unemployment than females, but only in the former West Germany. The differences between men and women were negligible for respondents who grew up in East Germany, where there is a longer tradition of female labour force participation and gender egalitarianism due to its socialist past.
The current study attempts to move these discussions forward by illuminating the gendered health consequences of unemployment after the turn of the century in one of the most gender-egalitarian countries in the world: Norway [42, 43]. Numerous linked administrative register data sources were analysed in order to answer the following overarching research question: How gendered are the health consequences of unemployment in Norway from 2000 to 2017?
Four health outcomes were examined longitudinally, which correspond with the aetiology of mental and somatic health conditions that may arise due to unemployment and associated stress and worries. The first outcome is hospital admissions due to mental and behavioural disorders; diseases of the nervous, circulatory, and respiratory systems; and injuries, poisoning, and other external causes. Second, we looked at receiving sick pay (i.e., a temporary health-related benefit). Third, we analysed disability benefit utilisation (i.e., a permanent health-related benefit). Fourth and finally, we examined the 10-year mortality likelihood. We analysed, by means of linear probability models and logistic regression, health trajectories over time among people that received unemployment benefits in three different exposure years: 2000, 2006, and 2011. The three exposure years were chosen because the economic conditions were very similar, which should ease cross-cohort comparisons. Based on the literature review and theoretical reflections above, two main hypotheses can be derived:
-
H1: Gender differences in the health consequences of unemployment are greater for mortality than for the other health outcomes, in accordance with the ‘gender health paradox’.
-
H2: Gender differences in the health consequences of unemployment have decreased over time, as the Norwegian society has gradually become more gender equal.
The current study aims to add to the existing literature in three domains. First, by analysing four register-based outcomes (hospitalisation, receiving sick pay, disability benefit utilisation, and mortality), which together will hopefully paint a comprehensive picture of the gendered health consequences of unemployment in Norway. Second, by following unemployed cohorts longitudinally, covering an 18-year period, we can examine the medium-to-long-term consequences of unemployment, which represents a gap in previous research, according to Norström et al. [33]. Third, we can examine potential time trends by reporting empirical findings for three unemployed cohorts that experienced unemployment during different times (2000; 2006; 2011) yet with similar economic conditions.