The impact of the parental division of paid labour on depressive symptoms: The moderating role of social policies

Abstract

This study examines the association between the parental division of paid labour and depressive symptoms in a comparative perspective. It investigates how this relationship varies across couples in countries with different social policies using data from European Social Survey, and multilevel models with cross-level interactions between the parental division of paid labour and macro-level indicators of social policies.

The results indicate that dual-earner couples report fewer depressive symptoms than parents in other types of families. This relative advantage of dual-earner couples varies across policy contexts. The benefits of a dual-earner model over a male breadwinner model are larger in countries where childcare services are easily available and do not disappear in countries with generous financial support from the state. Additional analyses reveal how these relationships differ across gender.

Keywords

Depression mental health family division of labour childcare social policy

Introduction

Depression is receiving increasing attention as a major factor in the deterioration of the quality of life in developed countries (Teghtsoonian, 2009). As the most common form of psychological distress, depression has been described as a barometer of life strains (Ross et al., 1983). Parenthood may be regarded as one of the risk factors for experiencing depressive symptoms (Evenson and Simon, 2005). While raising children is associated with emotional rewards and a sense of personal fulfilment, it also involves a substantial reorganization of both professional and private lives, which may have a negative impact on mental health. Couples often adjust to parenthood by adopting role specialization, with one of the partners focusing on paid work and the other on unpaid labour (Haas et al., 2006; Trappe et al., 2015). Role specialization may facilitate the management of family-related demands, but it also deprives one of the parents from career opportunities and non-monetary benefits related to employment such as time structure, social contacts and participation in a collective purpose (Jahoda, 1981), and it reduces household income, which makes a family vulnerable to the risk of poverty (Aassve et al., 2005). At the same time, financial strain affects mental health (Heflin and Iceland, 2009). Moreover, combining the role of a parent with the role of a worker provides opportunities to benefit from alternative sources of psychological gratification and social support (Gove, 1984). Against this background, this paper investigates how the parental division of paid labour is related to depressive symptoms in different contexts, with varying policies that support employment and financial situation of families.

The association between different configurations of parental involvement in paid work and depressive symptoms may be contingent on social policies (Aassve et al., 2005; Glass et al., 2016). Social policies determine the range of opportunities couples have to adopt specific forms of division of paid labour, and the barriers they face. Moreover, different dimensions of the welfare state may alter the balance of the advantages and the disadvantages associated with different models of parental division of paid labour. Depending on the institutional setting, the burden of financial and care-related responsibilities is allocated differently between mothers, fathers and the state (Lewis et al., 2008). While generous financial support may buffer families with only one earner (or with none) from economic difficulties, the availability of childcare helps dual-earner households reconcile their work and parenthood responsibilities.

This paper is the first to compare depressive symptoms among mothers and fathers across different constellations of their involvement in paid work. Previous research on parental mental health either focused on work-family conflict among working parents (see for instance Grönlund and Öun, 2010; Stier et al., 2012) or examined distress among families in which one or both parents are unemployed (Ström, 2003). Furthermore, the few studies that have compared the wellbeing of working and non-working parents examined mothers and fathers separately. This may be an important omission, because the impact of paid work on mental health may depend on how the mother's and the father's involvement in employment are combined within a couple. Applying a couple perspective offers a better understanding of how different constellations of parental labour market involvement affect parental mental health.

This study also adds to the literature by taking a comparative perspective on the mental health of parents, and revealing the moderating impact of social policies. Previous research on the effects of the division of paid labour in couples for depressive symptoms has lacked cross-country comparisons. Studies using more specific measures of wellbeing, such as happiness or life satisfaction, show mixed results¹. Some studies suggest that mothers who combine work with childrearing are worse off than non-working mothers unless they have access to public childcare services (Schober and Schmitt, 2017), and that childcare services alter the difference in wellbeing between employed women and homemakers (Treas et al., 2011). Others indicate that the cross-country variation in wellbeing gap between homemakers and women involved in paid work is not linked to the provision of childcare (Hamplová, 2019). However, we need more knowledge on the gaps in mental health between couples with different constellations of division of paid labour, and the way they are moderated by social policies.

This study focuses on two types of policies that correspond to the key dimensions of European welfare states. The analyses investigate to what degree financial support as well as availability of childcare alter the gap in the incidence of depressive symptoms between dual-earner couples and families with at least one stay-at-home parent. Hence, this study contributes to the growing research on the contextual factors that determine how different family constellations affect the mental health of couples (Carr and Springer, 2010). While the focus of this paper is on couples, mothers and fathers face different opportunities and challenges in relation to paid work and family life, and therefore additional analyses examine whether the policy benefits vary across gender.

Theoretical background

Research on mental health among couples has been inspired by two competing theoretical perspectives: the specialization hypothesis and social role theory of depression. The former highlights the benefits of adopting a strict division of paid and unpaid labour within a couple, while the latter stresses the benefits of combining the role of a parent with the role of a worker. In the following discussion, we present the key arguments and predictions based on these two perspectives. Neither of these theories considers the institutional context in which families are embedded. Therefore, in the next step, the discussion moves to the moderating role of the welfare state support for families.

Specialization versus combined roles

According to the economic theory of marriage, specialization requires one of the partners to invest all of his or her time in paid labour, whereas the other partner takes care of the domestic chores (Becker, 1991). By adopting a strict division of labour, both partners become more efficient in performing their tasks, which should improve their wellbeing. In principle, efficiency gains resulting from specialization may mean that breadwinners become more successful on the labour market, thus raising family incomes, and stay-at-home partners may become better at household duties and childcare, which may also bring some positive consequences for all the family members. Raised efficiency may also mean more time for leisure, with potential benefits for mental health. In principle, specialization functions as a mechanism for improving family welfare that operates independently of the gender of the parent who becomes the breadwinner. However, it has been argued that the structural and cultural conditions of modern labour markets offer men a comparative advantage in paid labour, whereas women enjoy an advantage in housework and childrearing (Becker, 1991). Based on this theoretical perspective, it may be expected that parents in male breadwinner families report fewer depressive symptoms than parents in dual-earner families (Hypothesis 1).

The specialization hypothesis has been influential in family research (Shockley and Shen, 2016). However, it has also received a lot of critique. Structural changes in labour markets, as well as the increasing social acceptance of women's employment, have reduced men's comparative advantage in paid labour. Moreover, a lack of fulfilment in professional life may not necessarily be compensated by full engagement in domestic life. While specialization hypothesis disregards the non-economic benefits of engaging in paid work, a large body of research has recognized that employment provides individuals with time structure, social contacts, participation in a collective purpose, a valued societal position, and regular activities (Jahoda, 1981). Stay-at-home parents are deprived of these benefits, which may contribute to feelings of inequity and isolation, and thus lead to the development of symptoms of depression (Bird, 1999). Finally, role specialization is related to reliance on a single income, which makes a family vulnerable to the risk of poverty (Aassve et al., 2005), with negative repercussions for mental health (Heflin and Iceland, 2009).

The non-monetary benefits from employment are given more attention in the social role theory (Gove, 1984; Gove and Geerken, 1977). This theory holds that combining the role of a parent with the role of a worker is favourable for mental health because these two roles function as alternative sources of psychological gratification and social support. Social support is in turn a crucial determinant of mental health (Thoits, 2011). Indeed, a number of empirical studies in this research field has shown benefits from combining work with parenthood, especially in countries where policies support work-family reconciliation (Grönlund and Öun, 2010; Nordenmark, 2004). Against this background, one can expect fewer depressive symptoms among parents in dual-earner families as compared to other types of families (Hypothesis 2).

The moderating role of the welfare state

The discussion on benefits and drawbacks of both specialization and sharing duties within parental couples needs to consider the moderating role of the institutional context. Previous research has distinguished two main dimensions of social policies: de-commodification and de-familization (Esping-Andersen, 1999). De-commodification, or the degree to which families have to rely on income from paid work, reduces the impact of employment status on wellbeing. De-familization corresponds to policies that make the wellbeing of family members less dependent on care provided within the family. Both de-commodification and de-familization are related to satisfying family needs, albeit in different ways (Grönlund and Öun, 2010). De-commodification policies, such as financial benefits for non-working population, provide income necessary to cover basic expenses. This may improve economic situation and hence mental health, especially in families with non-working parents. Financial support from the state may also decrease the social stigma of not doing paid work (Voßemer et al., 2017). De-familization policies, such as childcare services, increase levels of parental autonomy, as both parents have more opportunities to be involved in paid work. The benefits from better conditions for reconciliation of work with parenthood may go beyond the mere chances of having a job, and also include the opportunities of working full-time instead of part-time, having more responsibility, autonomy, and influence within organization (Strandh and Nordenmark, 2006). The opportunity to choose the way to organize professional and private lives contributes to parents’ feelings of agency and control over their own situations (Fryer, 1986). In addition, being more involved in paid work and having better career opportunities is related to higher incomes, with positive consequences for mental health. Childcare services also lessen the conflict between work and parenthood and reduce the stress related to role overload that parents tend to experience. Finally, childcare services increase the social acceptance of being involved in paid work for parents with young children (Sjöberg, 2004; Treas and Widmer, 2000). This policy may also potentially improve parents’ mental health by alleviating the guilt that parents – and mothers in particular – typically feel when not devoting all of their time to the parenthood role.

The benefits that families derive from specific social policies may differ depending on parental division of labour. Thus, social policies may alter the gap in the incidence of depressive symptoms between dual-earner couples and families with at least one non-working parent. When both parents are involved in paid work, they are more likely to need support in the form of childcare services and to benefit from the availability of such services. Hence, childcare services may alter the balance of costs and benefits between dual-earner families and other types of families, and affect the relative differences in their levels of depressive symptoms. One can expect therefore that in countries with higher childcare availability, the gap in mental health between dual-earner and other types of families is larger (Hypothesis 3).When one or both parents are not involved paid work, a couple is more likely to need financial support. Financial support may play a more important role in single-earner families, and becomes crucial if neither of the parents in a family is involved in paid work. Therefore, it is hypothesized that financial support reduces the relative mental health disadvantage of families in which one or both parents do not work (Hypothesis 4).

Due to unequal division of family duties among many couples, women are more likely to experience conflicting demands from the job and the family to a higher degree as compared to men. As a result, childcare services may play a relatively more important role for mothers’ than for fathers’ wellbeing. Therefore, the predictions from Hypothesis 3 may be particularly relevant for women. Since mothers are more likely to handle everyday expenses and assume responsibility of household budgets, financial strain may affect their depression levels more strongly. Hence, the gap in mental health between dual-earner couples and families with less than two earners, and the way it is moderated by financial support as outlined in Hypothesis 4, may be also relevant particularly for women.

Data and methods

This study draws on data from the European Social Survey (ESS 2018), specifically, the pooled waves of the ESS from 2006, 2012, and 2014, which included measures of mental health². The analyses focus on partnered men and women aged 20–50 with preschool children. The focus is on children aged six and under because younger children require more intense monitoring than children of school age. In addition, families with young children represent the group who are most affected by social policies, such those related to the provision of public childcare. Altogether, the sample includes 7779 individuals in 25 countries.

The data include a validated shortened version of the Centre for Epidemiologic Studies Depression scale (CES-D), which was constructed to identify populations at risk of developing depressive disorders (Radloff, 1977). This scale includes eight items. Respondents were asked to indicate how much of the time during the past week they had (i) felt depressed, (ii) felt that everything they did was an effort, (iii) had restless sleep, (iv) felt happy, (v) felt lonely, (vi) enjoyed life, (vii) felt sad, and (viii) could not get going. For each item, respondents could choose from the following categories: 1, none or almost none; 2, some of the time; 3, most of the time; and 4, all or almost all the time. All of the items were recoded into a scale from zero to three, with a high score indicating a high level of depressive symptoms. The recoded scale scores for the CES-D were then assessed using a summated rating that ranged from zero to 24. Using the same items in the ESS, Van de Velde et al. (2010) have shown that this shortened version of the CES-D is valid and reliable, and is cross-culturally equivalent for all of the countries in the survey.

The key explanatory variable is the parental division of paid labour. Based on information on involvement in paid work of both partners, the analyses distinguish between four categories of families: dual-earner families, male breadwinner families, female breadwinner families, and jobless families. Dual-earner families are defined as couples where both partners are involved in paid work. Male (female) breadwinner households are defined as those, where a male (female) partner does paid work and the other partner within a couple is not involved in paid work. Jobless families are a category where neither partner is involved in paid work. The data do not allow operationalizing the parental division of paid labour in a more detailed way, or assessing the degree to which these choices were voluntary or imposed.

The set of control variables considers factors that may confound the relationship between the parental division of paid labour and depressive symptoms, their distribution in the sample is summarized in Table A1 in the appendix. To avoid the so-called over-control bias (Elwert and Winship, 2014), factors that mediate the impact of division of paid labour on mental health are not controlled for in the regression models. The analyses control for age, sex, education attainment, and the immigration status. Age is included in linear form, and sex is a binary variable distinguishing between women and men. Educational attainment distinguishes between individuals with and without tertiary education. The immigration status distinguishes between people born in the country where they currently reside and those who came from other countries. The analyses distinguish between couples who are married, in a registered (or otherwise legally recognized) union, and in an informal union. The control variables include also the age and educational attainment of the partner, measured as a dummy that distinguishes between having and not having tertiary education. In addition, analyses control for the number of children of preschool age and for the age of the youngest child in the household. Differences across countries in macroeconomic conditions are accounted for by using variables measuring unemployment rate levels and gross domestic product (GDP) per capita, both indicators were derived from Eurostat and standardized. Finally, dummies for the survey rounds are controlled for to capture other potential time-specific factors affecting division of paid labour and depressive symptoms.

The microdata from the ESS are linked with indicators of social policies corresponding to de-commodification and de-familization. The measure of state financial support comes from the Out-of-Work Benefits Dataset (OUTWB) (Doctrinal et al., 2015) and captures the overall level of income replacement in the following cash benefits: unemployment benefits, social assistance, and family benefits for a two-parent family with earnings of 50 percent of an average wage or more. All sources of income are measured after tax and social security contributions. The database covers the period 2001–2011. The replacement rates for 2006 are linked to the 2006 ESS wave and values of the replacement rates for 2011 to the more recent ESS waves. In multivariate analyses, this variable was standardized so that the mean is equal to zero and standard deviation is equal to one, but the original values of this indicator are presented in Figure A1 in the appendix.

De-familization is operationalized with an indicator of the provision of childcare services based on Eurostat data. This indicator measures the percentage of preschool children participating in formal childcare at least 30 h per week. Formal childcare includes services provided by day-care centres, kindergartens, and preschools. It excludes childcare provided by child-minders or relatives. The indicators calculated by Eurostat correspond to two age groups: children aged 0–2 and children aged 3–6. These indicators are linked with ESS data on the parents of preschool children based on the age of the youngest child. Values of indicators for 2006 are linked to the 2006 ESS wave and values of indicators for 2012 and 2014 to the more recent ESS waves. Again, this country-level variable was standardized, but the original values of this indicator are presented on Figure A2 in the appendix.

Data from the ESS are hierarchically structured with the respondents nested in countries. This study employs multilevel linear regression models that allow for dependence of observations within countries and country-years (Schmidt-Catran and Fairbrother, 2016)³. Within a multilevel modelling framework, it is possible to assess the moderating impact of macro-level factors, such as the effects of social policies on outcomes of interest, while controlling for compositional differences in social categories across countries. To assess the moderating role of social policies, analyses include cross-level interactions between the parental division of paid labour and the indicators that operationalize welfare state support for parents. The models include random slopes for the parental division of paid labour, because the relationship between division of paid labour and parental depressive symptoms is expected to vary across countries. Heisig and Schaeffer (2019) show that models that do not include random slopes for the lower-level components of cross-level interactions often lead to anti-conservative inferences. The analyses use the whole sample of parents with preschool children in the first step, and in the next step models are estimated separately for men and women to test for any gender differences.

Empirical results

Parental division of labour and depressive symptoms

To get an overall impression of the association between the parental division of paid labour and mental health, the mean scores of depressive symptoms are compared across family types (Figure 1). The results indicate that, on average, the level of depressive symptoms among parents is lowest in dual-earner families, with a mean score of 4.21. The level of depressive symptoms among parents in male breadwinner families is higher, with a mean score of 4.7. Parents in female breadwinner families have an even higher level of depressive symptoms, with a mean score of 5.5. Couples in which neither the mother nor the father is involved in paid work have the highest level of depressive symptoms, with a mean score of 6.17.

Figure 1.

Mean levels of depressive symptoms according to the parental division of paid labour (with 95% confidence intervals). Source: ESS data.

In the next step, linear multilevel models are estimated controlling for the characteristics of the parents, to investigate how institutional settings moderate the relationship between the parental division of paid labour and depressive symptoms. Model 1 includes individual-level control variables as well as indicators that measure de-commodification and de-familization: the availability of childcare services and the replacement rates of cash benefits. The results confirm the conclusions from the descriptive analysis: namely, that parents in dual-earner families report fewer depressive symptoms than parents in families in which one (or both) of the partners is not doing paid work. This contradicts Hypothesis 1 stating that male breadwinner families should report fewer depressive symptoms as compared to parents in dual-earner families and confirms instead Hypothesis 2, which predicts that parents in dual-earner families experience fewer depressive symptoms than other types of families. Specifically, the results of Model 1 show that as compared to dual-earner families, parents in male breadwinner families have depression levels that are 0.62 points higher; while the parents in female breadwinner families have depression levels that are 1.14 points higher. For parents in jobless households the difference in depressive symptoms amounts to 1.71 points, indicating that families in this category are most vulnerable to depression.

Model 2 includes interactions between the availability of childcare and categories of the parental division of paid labour. As expected based on the theoretical discussion, the gap in depressive symptoms between parents in dual-earner and male breadwinner families varies according to the availability of childcare. Specifically, the benefits of a dual-earner family as compared to a male breadwinner family are larger when the availability of childcare services is higher, which supports Hypothesis 3. When availability of childcare increases by one standard deviation, the difference in the level of depressive symptoms between parents in male breadwinner and dual-earner families increases by 0.19 points. Interestingly, while on average, parents in female breadwinner families report more depression symptoms than parental in dual-earner couples, with a difference of 1.21 points, an increase in childcare availability by one standard deviation decreases this gap by 0.49 points. This contradicts the predictions from Hypothesis 3, and suggests that the benefits from availability of childcare are not necessarily restricted to dual-earner couples. Regarding the comparison between jobless families and dual-earner couples, again the gap in the levels of depressive symptoms between these two types of families is larger in countries where childcare services are more available, as predicted in Hypothesis 3.

Since the results from models with cross-level interactions may be difficult to interpret, to visualize the key findings, the predicted values of depressive symptoms for specific family types according to the availability of childcare are presented on Figure 2. Three exemplary levels of childcare availability were, for the purposes of visualization, arbitrarily selected at 5%, 20% and 70% levels⁴. Consistently with the results from Model 2, dual-earner couples have lowest levels of depressive symptoms. While the difference in levels of depressive symptoms of dual-earner couples and male breadwinner couples is not statistically significant in societal contexts where availability of childcare is most limited, this difference is larger and statistically significant in countries where availability of childcare is at the 70% level. Note that these results do not provide evidence that in countries with higher availability of childcare, the male breadwinner families are worse off as compared to the same family type in countries, where childcare services are limited. In such a comparison, there are no statistically significant differences, as the 95% confidence intervals overlap. What Figure 2 shows instead is the mental health advantage of dual-earner families in countries where childcare availability is restricted, and a lack of this advantage in settings with higher availability of childcare. Consistently with the results from cross-level interactions in Model 2, Figure 2 shows that the difference in depressive symptoms between dual-earner families and female breadwinner families is visible and also statistically significant when childcare services are restricted, whereas this difference is substantially reduced and not statistically significant when availability of childcare services is at the 70% level. In other words, this suggests that the mechanisms, which generally can be expected to benefit dual-earner couples, might also benefit female earner families. Regarding jobless couples, as Figure 2 shows, the difference between their levels of depressive symptoms and dual-earner couples is rather large and statistically significant both in countries where childcare is limited, and also in countries where availability of childcare services is at the 70% level.

Figure 2.

Predicted values of depressive symptoms according to parental division of labour and availability of childcare (with 95% confidence intervals). Source: ESS data. Predicted values of depressive symptoms based on Model 2 in Table 1.

Table 1.

The impact of the parental division of paid labour on depressive symptoms - results from multilevel models.

	Model 1		Model 2		Model 3
	Coef.	S.E.	Coef.	S.E.	Coef.	S.E.
Age	0.00	(0.01)	0.00	(0.01)	0.00	(0.01)
Higher education	−0.28**	(0.12)	−0.28**	(0.12)	−0.28**	(0.12)
Women	0.73***	(0.11)	0.72***	(0.11)	0.72***	(0.11)
Immigrant status	0.72***	(0.15)	0.72***	(0.15)	0.73***	(0.15)
Union type (ref. Married)
Registered partnership	0.44**	(0.18)	0.45**	(0.18)	0.44**	(0.18)
Informal partnership	0.36***	(0.13)	0.36***	(0.13)	0.36***	(0.13)
Partner's age	−0.01	(0.01)	−0.01	(0.01)	−0.01	(0.01)
Partner's higher education	−0.24**	(0.10)	−0.24**	(0.10)	−0.25**	(0.10)
No children (ref. One child)
2 children	0.06	(0.10)	0.06	(0.10)	0.06	(0.10)
3 children or more	0.23	(0.20)	0.24	(0.20)	0.23	(0.20)
Age of youngest child	0.07	(0.04)	0.07	(0.04)	0.07	(0.04)
Unemployment rate	0.08	(0.08)	0.08	(0.08)	0.08	(0.08)
GDP per capita	−0.22*	(0.13)	−0.23*	(0.13)	−0.22	(0.13)
Division of paid labour (ref. Dual earner)
Male breadwinner	0.62***	(0.08)	0.65***	(0.08)	0.61***	(0.07)
Female breadwinner	1.14***	(0.24)	1.21***	(0.25)	1.13***	(0.24)
Jobless couple	1.71***	(0.19)	1.73***	(0.19)	1.73***	(0.21)
Policy indicators
Childcare	0.11	(0.09)	0.07	(0.09)	0.11	(0.09)
Cash benefits	−0.05	(0.08)	−0.05	(0.08)	−0.05	(0.08)
Interaction: Parental division of paid labour # childcare
Male breadwinner # childcare			0.19**	(0.09)
Female breadwinner # childcare			−0.49**	(0.22)
Jobless # childcare			0.29**	(0.14)
Interaction: Division of paid labour # cash benefits
Male breadwinner # cash					−0.07*	(0.04)
Female breadwinner # cash					0.17	(0.24)
Jobless couple # cash					0.39	(0.41)
Year (ref. 2006)
2012	−0.36**	(0.15)	−0.36**	(0.15)	−0.37**	(0.16)
2014	−0.37**	(0.18)	−0.36**	(0.18)	−0.37**	(0.18)
Constant	4.13***	(0.45)	4.15***	(0.45)	4.14***	(0.45)
Male breadwinner (slope)	0.00	(0.00)	0.00	(0.00)	0.00	(0.00)
Female breadwinner (slope)	0.31	(0.39)	0.27	(0.31)	0.28	(0.38)
Jobless couple (slope)	0.33	(0.23)	0.36	(0.24)	0.43	(0.28)
Country level	0.19***	(0.09)	0.20***	(0.08)	0.20***	(0.08)
Country-year level	0.05*	(0.08)	0.04***	(0.04)	0.05***	(0.03)
Individual level	11.16***	(0.59)	11.14***	(0.66)	11.15***	(0.58)
N	7779		7779		7779

Source: ESS data. * p < 0.05, ** p < 0.01, *** p < 0.001. Standard errors in parentheses. Note: country-level variables standardized.

Returning to the results from Table 1, it was hypothesized that financial support from the state may buffer families from the negative consequences that are observed when one or both parents are not involved in paid work. Therefore, Model 3 includes interactions between categories of the parental division of paid labour and replacement rates of cash benefits. Cash benefits turn out to reduce the gap in depressive symptoms between dual-earner and male breadwinner families, as predicted in Hypothesis 4. However, these moderating effects are rather weak. An increase in financial support by one standard deviation reduces the gap in levels of depressive symptoms by 0.07 score. In addition, there are no moderating effect of financial support from the state when it comes to the difference in levels of depressive symptoms between dual-earner families and female breadwinner couples and jobless couples. Thus, Hypothesis 4 receives only partial support in the empirical evidence.

To visualize the results of cross-level interactions from Model 3, the predicted values of depressive symptoms for specific family types according to the replacement rates of cash benefits are presented on Figure 3. Three exemplary levels of replacement rates were selected at 42%, 62% and 77% levels⁵. Consistently with results from Model 3 in Table 1, the difference in levels of depressive symptoms for specific family types is neither attenuated nor magnified in countries with more generous financial support.

Figure 3.

Predicted values of depressive symptoms according to parental division of labour and replacement rates of cash benefits (with 95% confidence intervals). Source: ESS data. Predicted values of depressive symptoms based on Model 3 in Table 1.

The results from control variables are similar across specifications and show that higher education is negatively associated with depressive symptoms, while being a woman, an immigrant, or a non-married person is positively correlated with depressive symptoms. Partner's higher education is positively related to mental health, while other family-related characteristics such as the number or age of children seem to play no major role. Regarding contextual variables, higher GDP per capita is related to lower number of depression symptoms among parents. There are no statistically significant associations between unemployment rates and depressive symptoms.

Model 2 and Model 3 from Table 1 were re-estimated separately for men and women in order to test for gendered effects of the division of paid labour in diverging institutional settings (see Table 2). In countries with higher availability of childcare, living in a male breadwinner family is more strongly associated with depressive symptoms as compared to a dual-earner family among women, but not among men. An increase in availability of childcare by one standard deviation, widens this gap by 0.28 points. Men in male breadwinner families have a level of depressive symptoms that is 0.56 points higher as compared to men in dual-earner families when childcare services are at the average level, and this relationship is not altered by increases in availability of childcare services. Regarding female breadwinner families, the interaction effects are not statistically significant in the models for women. The gap in depressive symptoms is reduced by 0.61 points for men in female breadwinner households when childcare availability increases by one standard deviation. The association between living in a jobless household and depressive symptoms depends on the availability of childcare for women, but not for men. For women, living in a jobless household is associated with additional 0.65 points of depressive symptoms as compared to living in a dual-earner family, if childcare availability increases by one standard deviation. Among men, by contrast, this association is not moderated by the availability of childcare services, as the effects of cross-level interactions are not statistically significant.

Table 2.

The impact of the parental division of paid labour on depressive symptoms - results from multilevel models, by gender.

	Model 1, Women		Model 2, Women		Model 1, Men		Model 2, Men
	Coef.	S.E.	Coef.	S.E.	Coef.	S.E.	Coef.	S.E.
Age	−0.02	(0.02)	−0.01	(0.02)	0.04**	(0.02)	0.03**	(0.02)
Higher education	−0.39***	(0.15)	−0.40***	(0.15)	−0.08	(0.16)	−0.07	(0.16)
Immigrant status	0.56**	(0.24)	0.58**	(0.25)	0.91***	(0.20)	0.91***	(0.20)
Union type (ref. Married)
Registered partnership	0.64**	(0.26)	0.64**	(0.25)	0.29*	(0.18)	0.29*	(0.17)
Informal partnership	0.48***	(0.17)	0.47***	(0.18)	0.23	(0.20)	0.23	(0.20)
Partner's age	−0.00	(0.01)	−0.00	(0.01)	−0.02	(0.02)	−0.02	(0.02)
Partner's higher education	−0.28**	(0.13)	−0.28**	(0.13)	−0.16	(0.12)	−0.18	(0.12)
No children (ref. 1)
2 children	0.06	(0.14)	0.07	(0.14)	0.01	(0.14)	0.02	(0.14)
3 children or more	0.46*	(0.27)	0.43	(0.28)	−0.18	(0.43)	−0.16	(0.42)
Age of youngest child	0.10*	(0.06)	0.11*	(0.06)	0.04	(0.05)	0.03	(0.05)
Unemployment rate	0.06	(0.09)	0.07	(0.09)	0.08	(0.08)	0.08	(0.08)
GDP per capita	−0.31**	(0.14)	−0.29**	(0.14)	−0.16	(0.15)	−0.16	(0.15)
Division of paid labour (ref. Dual-earner)
Male breadwinner	0.74***	(0.13)	0.69***	(0.12)	0.56***	(0.11)	0.52***	(0.11)
Female breadwinner	0.80**	(0.38)	0.75**	(0.37)	1.60***	(0.36)	1.51***	(0.37)
Jobless couple	1.40***	(0.25)	1.38***	(0.27)	2.05***	(0.28)	2.12***	(0.29)
Policy indicators
Childcare	0.06	(0.09)	0.07	(0.09)	0.08	(0.08)	0.08	(0.08)
Cash benefits	−0.06	(0.11)	−0.05	(0.11)	0.02	(0.10)	0.01	(0.11)
Interaction: Parental division of paid labour # childcare
Male breadwinner # childcare	0.28*	(0.15)			0.12	(0.10)
Female breadwinner # childcare	−0.34	(0.28)			−0.61***	(0.23)
Jobless couple # childcare	0.65***	(0.21)			−0.17	(0.25)
Interaction: Division of paid labour × cash benefits
Male breadwinner			−0.12	(0.08)			−0.05	(0.05)
Female breadwinner			0.12	(0.32)			0.35	(0.57)
Jobless couple			0.26	(0.43)			0.42	(0.52)
Year (ref. 2006)
2012	−0.38*	(0.20)	−0.39**	(0.20)	−0.29*	(0.17)	−0.29*	(0.17)
2014	−0.20	(0.20)	−0.22	(0.20)	−0.56***	(0.20)	−0.56***	(0.21)
Constant	5.17***	(0.54)	5.13***	(0.56)	3.51***	(0.54)	3.55***	(0.54)
Random effects parameters
Male breadwinner (slope)	0.03	(0.16)	0.03	(0.21)	0.00	(0.00)	0.00	(0.00)
Female breadwinner (slope)	1.22	(0.90)	1.25	(1.15)	0.98	(0.79)	1.33	(0.90)
Jobless couple (slope)	0.47	(0.48)	0.52	(0.51)	0.67	(0.37)	0.86	(0.55)
Country level	0.21***	(0.11)	0.21***	(0.11)	0.23	(0.08)	0.19**	(0.14)
Country-year level	0.07***	(0.06)	0.07***	(0.06)	0.08	(0.07)	0.10	(0.24)
Individual level	12.13***	(0.66)	12.15***	(0.66)	9.56***	(0.59)	9.55***	(0.81)
N	4419		4419		3360		3360

Source: ESS data. * p < 0.05, ** p < 0.01, *** p < 0.001. Standard errors in parentheses. Note: country-level variables were standardized.

Regarding cash benefits, similarly as in the main analysis presented in Table 1, the difference in depressive symptoms among dual-earner and male breadwinner households is reduced when cash benefits are higher, but the corresponding effects are not statistically significant. The analyses stratified by gender do not reveal any moderating effects of the generosity of cash benefits for other family types.

Sensitivity analyses show that the correlation between the division of paid labour and depressive symptoms remains similar regardless of whether household income is controlled or not (Table S1 in the Online Supplement). Since the categories of parental division of paid labour in this study do not consider differences in working time, robustness check tested if the results do not change when hours of work are taken into account (Table S2). Some countries in ESS data have small sub-samples of individuals matching the criteria for inclusion in the analysis, this holds especially for countries that participated in only one of the ESS rounds. To see it this affects the results, analyses presented in Table S3 exclude such countries. The results of these sensitivity analyses corroborate the key results.

Discussion

Parents with young children face numerous pressures related to work-family conflict and financial stress, which may have consequences for their mental health (Evenson and Simon, 2005). Role specialization may be seen as a way to adjust to these pressures. There is a longstanding debate that compares the benefits of role specialization with the advantages of combining work with family duties for parents (Kalmijn and Monden, 2012; Ross et al., 1983; Treas et al., 2011). The results presented in this paper contradict the hypothesis suggesting that role specialization provides the most favourable conditions for the wellbeing of parents with young children. Parents in single-earner families tend to experience more depressive symptoms than parents in dual-earner families. Overall, this evidence suggests that the social role theory, which stresses the advantages of a dual-earner family model, fits the reality of families in contemporary European societies better than the economic theory of marriage.

This study also investigated whether the association between the parental division of paid labour and depressive symptoms depends on institutional arrangements. The results indicate that the gap in mental health between parents in dual-earner and male breadwinner families is observed in countries where childcare services are broadly available. This may be related to the impact of childcare policies on reducing the work-life conflict among dual-earner couples. Moreover, in countries with policies that support combining work with family duties, parents – especially mothers – have the possibility to get qualified jobs involving more hours of work (Strandh and Nordenmark, 2006), which means not only more responsibilities but also higher incomes. Better earning opportunities may in turn reduce the risk of financial hardship and consequently limit depressive symptoms in dual-earner families. Strengthening social pressure on parents may be an additional explanation. Mothers in male-breadwinner couples may experience more depressive symptoms in countries with high availability of childcare as compared to mothers in dual-earner couples because in such settings the social norm to be involved in paid work while raising children may be stronger. Under such circumstances, becoming a stay-at-home parent is not necessarily a socially acceptable alternative to employment (Strandh et al., 2013). These aforementioned explanations are plausible especially in the light of the by-gender analysis revealing that childcare availability increases the mental health benefits of a dual-earner model over the male breadwinner model among mothers, but not among fathers.

The findings that provision of childcare increases the mental health advantage of dual-earner couples as compared to male breadwinner couples is consistent with the idea that this policy fosters many goals that are central from societal perspective. Encouraging employment of both parents may have positive impacts in such areas as gender equality, increasing the aggregate employment levels and thus improving economic prosperity. While the availability of childcare offers dual-earner couples a stronger mental health advantage, this also means that especially in gender-egalitarian policy contexts, job losses may have relatively stronger negative consequences for couples (Baranowska-Rataj and Strandh, 2021). This underscores the role of policy responses that should reduce the risks and mitigate the consequences of unemployment for couples` wellbeing.

The results indicate that the availability of childcare diminishes the difference in depressive symptoms between female breadwinner families and dual-earner families. There are at least two (mutually non-excluding) explanations of this pattern. Previous research shows that in couples where men lose jobs, such a shift in the burden of paid labour is not fully compensated for by a corresponding re-allocation of unpaid labour (Gough and Killewald, 2011). If a lion`s share of household duties still remains women's responsibility in female breadwinner couples, availability of childcare may reduce the work-to-family spillover in such families, with benefits for both partners (Ruppanner and Pixley, 2012). Availability of childcare services may also offer men in female-breadwinner families better conditions for a job search, with better expected effects for future household incomes. The latter explanation seems convincing given the results from by-gender analysis showing that childcare availability is associated with relatively fewer depressive symptoms among fathers (but not among mothers) in female breadwinner families.

The results also indicate that parents in jobless households are most vulnerable to depressive symptoms. This finding adds to the understanding of the possible consequences of parental division of paid labour. Previous research has shown that families are most likely to become jobless when the male breadwinner loses his job (Härkönen, 2011). Thus, in addition to impeding parental wellbeing in the short run, role specialization involves a long-run risk of transitioning into dual joblessness, which is associated with even more severe mental health problems. The results do not confirm that the financial support for people who are out of work reduces the relative mental health disadvantage of families in which one or both parents are not employed. Additional analyses across gender show that this conclusion holds both for women and men. One possible explanation is that in most European countries, cash benefits do not fully replace earnings, and do not compensate for the non-monetary benefits of employment. One should keep in mind, however, that the indicators of this policy do not capture all the important aspects of financial support, such as the conditionality of some of the cash benefits. If non-working parents face restrictions in access to cash benefits, for example because of limited job tenure, a high level of these benefits will not reduce their distress.

This study has some limitations. It is based on cross-sectional data, which prevents causal interpretation. The number of observations for female breadwinner couples and jobless couples is small in some of the countries, which might limit the statistical power. The categories of parental division of paid labour do not consider the differences in parental working time, miss the distinction between short-term and long-term unemployment as well as inactivity. All these subcategories may be quite relevant for parents’ mental health. Still, due to sample restrictions, one cannot take all these detailed categories into account. In addition, the comparative analyses use data from 25 countries. This number constitutes the minimum level of observations needed for a robust inference of country-level effects (Bryan and Jenkins, 2015). Moreover, the indicators of policies do not capture all the important aspects. The indicator of availability of childcare misses the potential disparities between coverage and actual usage of childcare, and does not distinguish between private and public formal childcare. As argued above, the indicator of financial support does not consider that some cash benefits may be conditional on prior working experience, for instance. There may be also country-level factors other than the ones considered in this study, such as policies or labour market structures that are related to parental division of labour and affect parental mental health. Such omitted country-level variables may lead to bias in the analysis.

Despite these limitations, this study sheds light on how the parental division of paid labour is associated with depressive symptoms in European families. It provides evidence on mental health in male breadwinner and dual-earner couples, which used to be the most prevalent family types; and on mental health in female breadwinner and jobless couples, which have received less attention, but are becoming increasingly common in Europe (Härkönen, 2011; Kowalewska and Vitali, 2020). Moreover, while some prior studies show that combining work with childrearing reduces wellbeing among parents with limited access to childcare services (Schober and Schmitt, 2017), other studies find no evidence on such link (Hamplová, 2019). This study shows that childcare services alter the gap in depressive symptoms between couples in different ways, depending on the labour market status of parents’ spouses. Hence, this study highlights the need to take a couple- rather than individual-focused perspective in research on how the institutional context protects parents from the negative consequences of work-family conflict and economic hardship.

Supplemental Material

sj-docx-1-asj-10.1177_00016993211066261 - Supplemental material for The impact of the parental division of paid labour on depressive symptoms: The moderating role of social policies

Supplemental material, sj-docx-1-asj-10.1177_00016993211066261 for The impact of the parental division of paid labour on depressive symptoms: The moderating role of social policies by Anna Baranowska-Rataj in Acta Sociologica

Footnotes

Acknowledgements

I would like to thank the four anonymous reviewers and the editors for their helpful comments and suggestions. Final revisions of this article were undertaken during my research stay at the LIVES Centre, University of Lausanne, and I would like to express my gratitude for the opportunity to work in this welcoming and inspiring environment.

Declaration of conflicting interests

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Grant No. DEC-2011/03/D/HS4/04258 from the National Science Centre and funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme under grant agreement No. 802631.

ORCID iD

Anna Baranowska-Rataj

Supplemental material

Supplemental material for this article is available online.

Notes

Author biography

Anna-Baranowska-Rataj is an associate professor at the Department of Sociology and a member of the Centre for Demographic and Ageing Research (CEDAR) at Umeå University, Sweden. Her research interests concern processes at the intersection of labour market, family as well as health and well-being, especially from international comparative perspective.

Annex.

Table A1.

Means and proportions of explanatory variables according to family type.

	Dual-earner	Female breadwinner	Male breadwinner	Jobless couple
Age	33.35	32.57	31.76	29.86
Women	57%	56%	56%	57%
Men	43%	44%	44%	43%
Native	90%	83%	85%	80%
Non-native	10%	17%	15%	20%
University education	51%	36%	32%	13%
Lower education	49%	64%	68%	87%
Partner with university education	47%	31%	32%	12%
Partner with lower education	53%	69%	68%	88%
Marriage	76%	76%	67%	53%
Registered partnership	9%	8%	14%	14%
Informal union	15%	16%	20%	33%
Partner's age	34.28	33.87	32.58	30.43
One child	76%	77%	72%	71%
Two children	24%	23%	28%	29%
Three or more	1%	2%	3%	4%
Age of the youngest child	3.47	3.51	2.82	2.98
Sample size	4342	318	2666	453

Source: ESS data.

References

Aassve

Mazzuco

Mencarini

(2005) Childbearing and well-being: A comparative analysis of european welfare regimes. Journal of European Social Policy 15(4): 283–299.

Baranowska-Rataj

Strandh

(2021) When things go wrong with you, it hurts me too. The effects of partner's Employment status on health in comparative perspective. Journal of European Social Policy 31(2): 143–160.

Becker

(1991) A Treatise on the Family (Enlarged Edition). Cambridge, MA: Harvard University Press.

Bird

(1999) Gender, household labour, and psychological distress: The impact of the amount and division of housework. Journal of Health and Social Behavior 40(1): 32–45.

Bryan

Jenkins

(2015) Multilevel modelling of country effects: A cautionary tale. European Sociological Review 32(1): 3–22.

Carr

Springer

(2010) Advances in families and health research in the 21st century. Journal of Marriage and Family 72(3): 743–761.

Doctrinal

Nelson

Sirén

(2015) Comprehensive Indicators for the Analysis of out-of-Work Benefits: Introducing the Out-of-Work Benefits Dataset. Stockholm: Swedish Institute for Social Research.

Elwert

Winship

(2014) Endogenous selection bias: The problem of conditioning on a collider variable. Annual Review of Sociology 40: 31–53.

Esping-Andersen

(1999) Social Foundations of Postindustrial Economies. Oxford: Oxford University Press.

10.

European Social Survey (2018) ESS 1–9, European Social Survey Cumulative File, Study Description. Bergen: NSD - Norwegian Centre for Research Data for ESS ERIC. https://doi.org/10.21338/NSD-ESS-CUMULATIVE

11.

Evenson

Simon

(2005) Clarifying the relationship between parenthood and depression. Journal of Health and Social Behavior 46(4): 341–358.

12.

Fryer

(1986) Employment deprivation and personal agency during unemployment: A critical discussion of Jahoda's Explanation of the psychological effects of unemployment. Social Behaviour 1(1): 3–23.

13.

Glass

Simon

Andersson

(2016) Parenthood and happiness: Effects of work-family reconciliation policies in 22 OECD countries. American Journal of Sociology 122(3): 886–929.

14.

Gough

Killewald

(2011) Unemployment in families: The case of housework. Journal of Marriage and Family 73(5): 1085–1100.

15.

Gove

(1984) Gender differences in mental and physical illness: The effects of fixed roles and nurturant roles. Social Science and Medicine 19(2): 77–84.

16.

Gove

Geerken

(1977) The effect of children and employment on the mental health of married men and women. Social Forces 56(1): 66–76.

17.

Grönlund

Öun

(2010) Rethinking work-family conflict: Dual-earner policies, role conflict and role expansion in Western Europe. Journal of European Social Policy 20: 179–195.

18.

Haas

Steiber

Hartel

, et al. (2006) Household employment patterns in an enlarged European Union. Work, Employment and Society 20(4): 751–771.

19.

Hamplová

(2019) Does work make mothers happy? Journal of Happiness Studies 20(2): 471–497.

20.

Härkönen

(2011) Children and dual worklessness in Europe: A comparison of nine countries. European Journal of Population 27(2): 217–241.

21.

Heflin

Iceland

(2009) Poverty, material hardship, and depression. Social Science Quarterly 90(5): 1051–1071.

22.

Heisig

Schaeffer

(2019) Why you should always include a random slope for the lower-level variable involved in a cross-level interaction. European Sociological Review 35(2): 258–279.

23.

Jahoda

(1981) Work, employment, and unemployment: Values, theories, and approaches in social research. American psychologist 36(2): 184.

24.

Kalmijn

Monden

(2012) The division of labour and depressive symptoms at the couple level: Effects of equity or specialization? Journal of Social and Personal Relationships 29(3): 358–374.

25.

Kowalewska

Vitali

(2020) Breadwinning or on the breadline? Female breadwinners’ economic characteristics across 20 welfare states. Journal of European Social Policy 31(2): 125–142.

26.

Lewis

Campbell

Huerta

(2008) Patterns of paid and unpaid work in Western Europe: Gender, commodification, preferences and the implications for policy. Journal of European Social Policy 18(1): 21–37.

27.

Nordenmark

(2004) Multiple social roles and well-being: A longitudinal test of the role stress theory and the role expansion theory. Acta Sociologica 47(2): 115–126.

28.

Radloff

(1977) The CES-D scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1(3): 385–401.

29.

Ross

Mirowsky

Huber

(1983) Dividing work, sharing work, and in-between: Marriage patterns and depression. American Sociological Review 48(6): 809–823.

30.

Ruppanner

Pixley

(2012) Work-to-family and family-to-work spillover: The implications of childcare policy and maximum work-hour legislation. Journal of Family and Economic Issues 33: 283–297.

31.

Schmidt-Catran

Fairbrother

(2016) The random effects in multilevel models: Getting them wrong and getting them right. European Sociological Review 32(1): 23–38.

32.

Schober

Schmitt

(2017) Day-care availability, maternal employment and satisfaction of parents: Evidence from cultural and policy variations in Germany. Journal of European Social Policy 27(5): 433–446.

33.

Shockley

Shen

(2016) Couple dynamics: Division of labour. In: Allen

Eby

(eds) The Oxford Handbook of Work and Family. New York, NY: Oxford University Press, 125–139.

34.

Sjöberg

(2004) The role of family policy institutions in explaining gender-role attitudes: A comparative multilevel analysis of thirteen industrialized countries. Journal of European Social Policy 14(2): 107–123.

35.

Stier

Lewin-Epstein

Braun

(2012) Work-family conflict in comparative perspective: The role of social policies. Research in Social Stratification and Mobility 30(3): 265–279.

36.

Strandh

Hammarström

Nilsson

, et al. (2013) Unemployment, gender and mental health: The role of the gender regime. Sociology of Health and Illness 35(5): 649–665.

37.

Strandh

Nordenmark

(2006) The interference of paid work with household demands in different social policy contexts: Perceived work–household conflict in Sweden, the UK, the Netherlands, Hungary, and the Czech Republic. The British Journal of Sociology 57(4): 597–617.

38.

Ström

(2003) Unemployment and families: A review of research. Social Service Review 77(3): 399–430.

39.

Teghtsoonian

(2009) Depression and mental health in neoliberal times: A critical analysis of policy and discourse. Social Science and Medicine 69(1): 28–35.

40.

Thoits

(2011) Mechanisms linking social ties and support to physical and mental health. Journal of Health and Social Behavior 52(2): 145–161.

41.

Trappe

Pollmann-Schult

Schmitt

(2015) The rise and decline of the male breadwinner model: Institutional underpinnings and future expectations. European Sociological Review 31(2): 230–242.

42.

Treas

van der Lippe

Tai

(2011) The happy homemaker? Married Women's Well-Being in cross-national perspective. Social Forces 90(1): 111–132.

43.

Treas

Widmer

(2000) Married women's Employment over the life course: Attitudes in cross-national perspective. Social Forces 78(4): 1409–1436.

44.

Van de Velde

Bracke

Levecque

(2010) Gender differences in depression in 23 european countries. Cross-national variation in the gender gap in depression. Social Science and Medicine 71(2): 305–313.

45.

Voßemer

Gebel

Täht

, et al. (2017) The effects of unemployment and insecure jobs on well-being and health: The moderating role of labour market policies. Social Indicators Research 138(3): 1229–1257.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB