The mental health of unemployed Brussels youth: the role of social and material resources

In the aftermath of the Great Recession, many European countries were confronted with an increase in their unemployment rates [1, 2]. Compared to adults, youth was continuously and disproportionately hit by this crisis [2‐4]. Ever since the start of the Great Recession in 2008 the average unemployment rate (until the second quarter of 2016) was higher for youth in the Brussels Capital Region (36.4%) than for youth in Flanders (14.3%) and Wallonia (29.3%) [5]. The BCR can be denoted as a region with a predominantly tertiary oriented labour market demand, related to its role as the capital of Belgium and Europe, and as the home town of many international institutions and headquarters. This labour market demand is contradictory to the high unemployment, which affects its inhabitants. An important explanation for this phenomenon is the educational mismatch between labour supply and demand within the region. Over 60% of the Brussels population is lower educated, which excludes them from a vast amount of the employment offered [6]. A large share of the employment is thus filled in by employees commuting from the Flemish and Walloon region [7]. This structural contradiction is known as the “Brussels paradox” [8].

Already in the classical works of Jahoda [9] and Warr¹ [10] it was shown that the association between unemployment and mental health can be explained through a wide range of material and immaterial factors, encompassing both selection and causation mechanisms [11‐18]. Parallel with the transition from education to employment, youth are confronted with several other important life changes (e.g., moving out of the parental house; first marriage or (pre-marital) break-up; having a child) that interact with entering the labour market [19]. In that context, unemployment experiences may be particularly detrimental for mental health, while at the same time mental health is an important factor in achieving labour market success [20].

Relations between (youth) unemployment and mental health are codetermined by important societal changes. The negative effects on mental health are more serious for unemployment occurring during an economic crisis [18, 21‐24]. Some studies argue that structural unemployment positively affects feelings of stigmatization, shame [25, 26] and stress [27]. Negative mental health effects can also be offset by welfare state protection and other policy measures [22, 28‐30]. In that regard, it should be noted that in Belgium, the right to unemployment benefits became a lot stricter from 2012 onwards, with less generous and accessible unemployment benefits for youth [31, 32]. Finally, since the 90ies also an evolution towards ‘more active labour market policies’ with an increasing focus on coaching, training and (subsidized) first work experiences has been seen [32]. Each of these macro-social and policy changes might have affected the mental health levels of unemployed youth over time. Therefore, a first objective of this study is to investigate the evolution of the ‘mental health gap’ between employed and unemployed youth in Belgium, and in Brussels in particular.

Research on the mental health effects of unemployment has often pointed towards the mediating effects of ‘contextual resources’. This concerns material resources in the first place: financial strain is found to be related to a higher risk of psychological distress and adverse wellbeing [33]. Reversely, experiencing a comfortable financial situation attenuates the mental health effects of unemployment [34]. Generous welfare state protection during unemployment can be seen as a very important source of ‘health-protecting material security’ [22, 28, 35]. However, also living arrangements are important: unemployed people living alone or with acquaintances have more adverse mental health outcomes compared to unemployed people living with a partner [36, 37]. Particularly for unemployed youth, living with one’s parents was found to be a protection against material deprivation [37]. The social resources are important as well. Low parental support increases the vulnerability for adverse mental health amongst the unemployed [37, 38], while high social support can reduce psychological distress [39]. A second objective of this study is to investigate the role of material and social resources in relation to mental health in a sample of unemployed young people at the moment of transition from education to the labour market.

Both the prevalence and type of mental health problems are different in men and women [40] and the same goes for mental health outcomes and coping styles during unemployment [41]. It is generally assumed that the effect of unemployment on mental health is less strong in women, given the different positions and roles they occupy in society and the family [42]. Moreover, while women tend to internalize social stressors (e.g., higher levels of psychological distress, higher rates of mood and anxiety disorders), men show externalising behaviour (e.g., use of medication, alcohol and drugs) [40, 43]. Research showed that unemployment correlated with a higher alcohol consumption [44] and a higher frequency of substance use [44, 45], especially amongst men [40]. Therefore, gender will be an important stratifying variable in our analyses.

Finally, the unemployment experience tends to vary with human capital [16, 46]. Recent research on the matter showed that low education is related to higher rates of psychological distress and mental disorders [40, 47]. In part this is related to the fact that the probability of exiting unemployment is higher for those with post-secondary education [48] and that - reversely - early-school-leavers have a higher likelihood of becoming unemployed [49].

In sum, this paper aims to explore the association between unemployment and mental health in Belgian youth, while paying particular attention to the BCR. This aim is translated into two sub-objectives. (1) To describe the evolution (1997–2013) of the difference in mental distress and possible mental disorder between unemployed and employed youth in the three Belgian Regions. This analysis is based on data from five waves of the Belgian Health Interview Survey (HIS). (2) To make an explanatory analysis of the determinants of mental health in a sample of unemployed Brussels youth entering the labour market. Hereby, specific attention will be paid to the effects of material and social resources. This analysis is based on a large sample of unemployed Brussels school leavers from 2015. Given the specificity of this sample, our results can provide more insight in the mechanisms linking unemployment to mental health in this specific transitional point in the life of youth.

This paper investigates the association of social and material resources with mental health amongst unemployed youth in their transition from education to employment. In the BCR-sample, we found a high prevalence of mental distress and a possible mental disorder. Compared with the prevalence found for both mental health outcomes in unemployed youth from the 2013 HIS-sample (see Table 3), the prevalence for poor mental health in the BCR-sample is very high. A potential explanation is the association between the duration of the unemployment spell and mental health, which follows a U-shaped pattern (a large initial impact, decreasing over time, but growing again with the duration of the unemployment spell) [10, 18, 53]. The youth in our sample have just entered the labour market and could be expected to experience a large initial impact. The HIS-sample, in contrast, can be assumed to be more heterogeneous with regard to the unemployment duration, since it is a mere random sample.

In line with existing literature, adverse mental health outcomes are systematically higher for unemployed compared to employed youth (objective 1). Moreover, our results suggest that the mental health gap between employed and unemployed youth increases over time. This is merely an indicative pattern, since a Cochrane’s Q test did not indicate that the prevalence rates of mental distress or possible mental disorder of unemployed versus employed were significantly different between the years. However, in some instances the absolute differences are huge: e.g., in 2001 the unemployed had a 6.01% higher prevalence of mental distress compared to the employed; in 2013 the same difference was 18.94% (data for Belgium). This makes us believe that the lack of a significant trend is mainly an artefact of low N -and that it is indeed appropriate to consider the fact that their might be an increasing mental health gap, both caused by improved mental health among the employed and deteriorated mental health among the unemployed. A first possible explanation for this evolution is related to the restrictive evolution of the welfare state protection from the 90ies onwards. We know that welfare state protection during unemployment can form a ‘health-protecting material security’ [22, 28‐30, 35]. A second explanation, for the two most recent waves, could be the negative effects of the economic crisis on mental health during unemployment [18, 21‐24].

Our study identified the importance of social and material resources in association with mental distress and a possible mental disorder (objective 2). The results show that social support can be considered a consistent protective factor for both adverse mental health outcomes among men and women. This corresponds with previous findings about the protective role of social support in the case of mental health [37‐39, 54]. We also found that, for both men and women, a poor financial situation was related to a higher likelihood of a possible mental disorder. This finding is supported by the literature, which found positive associations between mental health and a comfortable financial situation in youth [33, 34, 54]. We also found that living with a partner is consistently related to both adverse mental health outcomes for men, which is consistent with the literature as well [36]. Men in the BCR-sample who lived with a partner were on average 24 years old. A possible explanation for the association could be the presence of the male breadwinner perspective amongst this group.

Previous literature showed that escape-avoidance behaviour is more common amongst men [40, 44, 45]. Contrary to this, our results show that escape-avoidance behaviour is only significant in explaining mental health amongst women. Even though escape-avoidance behaviour is believed to be a predominantly male phenomenon, research found that the use of medication is more likely among women [55] and suggest that the “sex gap” for alcohol and drug use has been narrowing over time [55‐57]. Other literature suggests that men show escape-avoidance behaviour to increase positive feelings, whereas women will do this to decrease negative feelings [58].

This study has a number of limitations. Within this research field there is a clear on-going debate about the direction of the relation between unemployment and mental health (selection versus causation) [13‐17]. Based on this cross-sectional data no statements about causality can be made. As also stated by Leach et al. [59], cross-sectional data does not necessarily reflect the start of mental health problems. Readers should be aware of that when interpreting our results. A second limitation relates to potentially relevant information that was not available in the survey. Due to constraints of questionnaire length, we have limited information on some variables such as the specific types of medication used. More extensive research on these particularities is thus highly recommended. A third limitation that we need to take into account is linked to the retrospective nature of some of the questions in the BCR-questionnaire, which may have caused recall bias. To minimize recall bias, the presence of this type of questions was limited. To minimize bias related to the comprehension of the questionnaire at a semantic level, a pre-testing was organised to generate the questionnaire.

Despite these limitations, our findings can be considered as highly relevant with regard to the understanding of youth unemployment in urban regions. Our results highlight the association of unemployment and mental health, as well as the explanatory impact of material and immaterial resources in this association. These findings can contribute to the debate on how to guide this specific group of youth in their transition from education to employment. Youth can benefit from a multidisciplinary guidance approach, which helps them not only to find a job, but also to e.g., establish financial security and find affordable living arrangements. Our findings can also contribute to the policy debate on granted social benefits as recently done by Vahid Shahidi et al. [35]. This study found that the government can mitigate unemployment-related health inequalities by expanding the generosity and scope of social protection policies. Despite these strengths and contributions, more research on the topic is recommended.

The aim of this paper was to (1) describe the difference in mental health outcomes between unemployed and employed youth over time and region and (2) to explore the association of social and material resources with mental health amongst unemployed youth entering the labour market in the Brussels Capital Region.

This study found a high prevalence of mental distress and possible mental disorder among Brussels youth in the transition from education to employment. Secondly, when comparing unemployed with employed youth our results suggest that the mental health gap between both has been increasing. Thirdly, our study found that the role of social support is associated with both mental distress and a possible mental disorder for men and women. Living arrangements is an important factor for men and escape-avoidance behaviour is an important factor for women. The association of the financial situation is only important in understanding a possible mental disorder for both men and women.

In short, our results suggest the worsening of the mental health situation of unemployed youth compared with that of employed youth, as well as the importance of social and material resources in explaining the mental health situation of unemployed youth. By doing this, our study fills an existing gap in the research framed within the Great Recession.