Keywords

1 Introduction and Study Aims

The landscape of family forms and living arrangements with and without children is changing rapidly in Switzerland similar to the trends in many other Western nations (Bernardi et al. 2018; Sánchez Gassen and Perelli-Harris 2015; Goldschneider et al. 2015). Steady increases in divorce rates and non-marital childbirth across many European countries have, among other reasons, led to growing numbers of lone parents and blended families. For example, crude divorce rates have more than doubled in Switzerland and Belgium from 0.9 and 0.5 per 1,000 inhabitants in 1960 to 2.0 and 2.1 per 1,000 inhabitants in 2016, respectively (Eurostat 2018). Due to the growing numbers of lone parents and blended families, minors are at a higher risk of growing up without the presence of both biological parents in one household (Sobotka and Toulemon 2008; Smyth and Moloney 2008). Even though children are still more likely to stay with their mothers after a separation, changes in legal and cultural practices across Europe have led to a higher share of parents who opt for shared physical custody (hereafter, ‘SPC’; Cancian et al. 2014; Juby et al. 2005; Kitterød and Lyngstad 2012). SPC is hereby defined as a post-separation custody arrangement where children spend time at two alternating places of residence—one at each of the parents’ homes—and parents are able share daily responsibilities and routines compared to other non-residential parents with sparse contact to their children (i.e., no physical custody; hereafter, ‘NC’). How SPC parents decide to split children’s time between the two households, however, can vary widely from equal amounts of time (50%:50%) to up to at least one-quarter of time (75%:25%) across families and legislative contexts (Baude et al. 2016; Meyer et al. 2017).

Prior studies documented that the low prevalence of SPC families meant higher levels of selectivity. Children in SPC families tended to have parents with higher levels of education, higher income, and lower reported levels of relationship conflict compared to children in more traditional sole (physical) custody (hereafter, ‘SC’) arrangements (Bauserman 2012; Nielsen 2018; Steinbach 2019). In countries where SPC was legally regulated, either as default or as the priority custody arrangement, such as many of the Nordic countries, the share of SPC increased. Consequently, profiles of SPC families became less selective, at least in terms of parental education and income (Cancian et al. 2014). Far from these developments, in Switzerland, clearly defined legal regulations for SPC were lacking until 2017. Although separation and divorce among Swiss parents is rising at a similar rate to other Western countries, no prior study has yet examined the spread and profiles of families practicing novel post-separation custody arrangements, such as SPC, in the Swiss context to our knowledge. In addition, while the majority of research on correlates and consequences of SPC has so far focused on children’s adjustment to SPC, little research has devoted attention to the correlation of SPC arrangements with parental health. This is all the more surprising given that previous studies established a reciprocal dependence of union separation, family structure, parenting, and parental health across the life course more generally (Amato 2010; Osborne et al. 2012; Umberson et al. 2010). To fill these research gaps, the aim of this chapter is twofold. First, we aim at providing a comprehensive overview of the prevalence and characteristics of SPC families in Switzerland before the legislative change of 2017. Second, we examine whether SPC is associated with Swiss parents’ physical health and well-being.

1.1 Spread and Characteristics of SPC Families

More involved and active fathering has been on the political agenda of many Western countries with the introduction of more gender-neutral family polices—particularly in the Scandinavian countries—and a societal re-definition of manhood that encourages men to engage in multiple social roles besides the traditional breadwinner role (Goldschneider et al. 2015). For example, legal changes facilitating the establishment of longer parental and paternal leave were introduced in many European countries (Thévenon 2011), which may have contributed to the substantial daily time increase coupled fathers reported to spend with their children over the last decade (e.g., 57% increase in time spent with children from 2001/2002 to 2012/2013 among German men; Klünder and Meier-Gräwe 2018). The majority of fathers also wish to continue their engagement with their biological (or step-) children upon the separation from a former partner or spouse indicating shifting social norms related to parenting and fatherhood that may consequently impact changes on the social policy-level (Bengtson and Allen 2009). However, the implementation of post-separation legislation that fosters father involvement in custody arrangements, such as SPC, varies considerably across Europe. Note that, throughout the chapter, SPC and other custody arrangements are referring to decisions concerning children’s living arrangements only (e.g. SPC, SC) and not to legal custody, which defines which parent can make decisions about a child’s life such as schooling, religious upbringing, or medical care.

In countries where SPC was introduced as the legally preferred solution for married couples seeking divorce or separation early-on (e.g., 1998 in Sweden and 2006 in Belgium), incident rates of parents utilizing shared-time custody models have risen to nearly 50% in Sweden (Bergström et al. 2013) and 37% in Belgium (Vanassche et al. 2017). In contrast, countries without a clear legal framework for SPC, such as in Germany, only about 5% of post-separation families reported to practice SPC (Walper 2016). Changes can also occur rather rapidly. In Spain, where similar legal changes happened in 2010, SPC almost tripled in just the space of a few years but with considerable regional variation in rates of SPC (ranging from 13% to up to 32% in some regions in 2013; Steinbach 2019). Studies have shown that at the initial stage of SPC diffusion, separated parents opting for this custody arrangement, tended to be more affluent, highly educated, practiced greater gender role equality, and reported lower levels of marital or post-separation conflict compared to ex-couples with SC arrangements (e.g., Bauserman 2012; Juby et al. 2005; Kitterød and Lyngstad 2012; Nielsen 2018; Schier and Hubert 2015). Since SPC arrangements became more common after strengthening the legal pathways toward SPC, however, profiles of SPC parents pluralized as well (Cancian et al. 2014; Sodermans et al. 2013). Even though the share of high-conflict and average to low educated SPC couples increased in response to the legal changes, the prevalence among the lowest educational groups—particularly families with two lower-educated parents—remained relatively low. Reasons for this development are most likely the financial means necessary for practicing SPC, such as maintaining two rooms for a child, or coordinating commutes between parental residences, which limit the feasibility of SPC particularly for less resourceful parents (Melli and Brown 2008; Steinbach 2019). Furthermore, parents’ ability to find non-conflictual custody agreements that are beneficial and acceptable for all parties involved, or to seek help to negotiate respective solutions (e.g., with a counsellor), may also be strongly correlated with one’s educational attainment.

In Switzerland, similarly to Germany, no explicit legal framework for SPC was available before 2017 and SPC arrangements could only be formalized if both parents demanded and agreed to it. Only since 2014, the Swiss Family law no longer differentiates between children born to married and non-marital unions with regard to parental responsibilities and duties, which particularly strengthened the rights of post-separation fathers to a child born out of wedlock upon separation. Nevertheless, judges were only advised to investigate the possibility of SPC routinely regardless of parents’ demands or disagreements since 2017. Additionally, gender-biased employment practices in the labor market and the lack of widespread, affordable childcare, despite some institutional efforts to promote gender equality at the policy level, encourage Swiss women to reduce their time in paid labor once they became mothers (Le Goff et al. 2009). It is important to note, however, that rates of childcare coverage vary across Switzerland’s multilingual cantons with higher coverage in French-speaking cantons and urban centers (Bonoli 2008). Working part-time, and often in lower-paying, less career-oriented jobs has, in turn, shown to contribute to the emergence of more traditional role distributions among Swiss couples prior and after a separation (Bernardi et al. 2013; Bühlmann et al. 2009; Le Goff et al. 2009). The lack of explicit family policies to support women’s full-time engagement in the workforce may further reinforce the establishment of (female-headed) SC arrangements upon separation (Davis and Greenstein 2009; Juby et al. 2005), despite recent advances in fostering legal pathways toward SPC.

Examining the Swiss case more closely—a late and only partial adopter of gender-neutral parenting policies—our study is the first attempt to provide a comprehensive overview of the prevalence of families who opted into SPC before the 2014 and 2017 legislative change, as well as to describe the sociodemographic make-up of those families. We expect SPC arrangements to represent a small proportion of custody arrangements among separated parents because of Switzerland’s lack of legal guidance at the time of data collection. Relatedly, we expect that SPC parents are rather selective in terms of education and socio-economic standing (SES) compared to parents with NC, SC, or visitation rights only (hereafter, ‘VR’; Hypothesis 1).

1.2 SPC and Parents’ Health

Prior research on SPC has largely focused on child adjustment, such as the perceived stress or emotional well-being of SPC children compared to their peers in SC arrangements or those growing up with both parents in one household (e.g., Bergström et al. 2013; Baude et al. 2016; Nielsen 2018; Turunen 2017). Going through a divorce or separation has shown to have—at least temporarily—detrimental effects on the physical health and mental well-being of former spouses or partners due to the expose to multiple stressors, such as shouldering sole parenting responsibilities, losing of emotional support, being exposed to continuing relational conflict, or experiencing economic decline (Amato 2010; Cooper et al. 2009; Osborne et al. 2012). Changing social roles of and relationships between the linked lives of family members need to be re-assigned and re-negotiated in response to such an impactful life event, which may have lasting ripple effects on parents’ subsequent life course (Bengtson and Allen 2009). Even though parenting stress has generally shown to weigh on parents’ well-being—at least when children are young—separated families may be particularly prone to the negative impact of parenting stress because they encounter more life strains related to parenting duties (e.g., through potentially stressful, yet constant consultation and negotiation with the former partner or spouse; Umberson et al. 2010). Findings further suggest that well-being and health penalties for parents having gone through union dissolution and potentially re-formation, such as lone parents, were substantially larger in countries with less generous policy support for families and lower levels of gender equality (Burstrom et al. 2010; Hübgen 2018; Pollmann-Schult 2018).

Despite the well-established linkages between family structure, parenting, and health, the associations between SPC and parental outcomes have rarely been examined. Apart from potential selectivity effects into SPC after a separation, splitting parenting duties may enable both parents, and particularly fathers, to enjoy time with their children and maintain supportive and positives ties to them—particularly father-child contact—which, in turn, could foster parents’ health and well-being (Baranowska-Rataj et al. 2014; Steinbach 2019; Vogt Yuan 2016). SPC parents may further benefit from having more time for activities unrelated to parenting, such as work, dating, or leisure activities, which may decrease parenting stress for both mothers and fathers compared to lone parents. On the flip side, SPC may also increase parenting stress and therefore curb parental health and well-being—especially for those with sparse resources to fall back on or for conflictual couples—because practicing SPC is rather costly compared to traditional SC arrangements (e.g., paying for commutes between parental homes and maintaining the child’s rooms in each of these home or duplicate sets of cloth, school supplies, etc.), and requires constant communication with the ex-spouse or partner. These additional burdens may outweigh potential benefits from reduced parenting duties. For example, the challenges of negotiating parenting roles and responsibilities in complex family configurations with residential stepchildren, whether they reside in the household full- or part-time, have shown to weigh particularly on stepmothers’ perceived parenting stress (Guzzo et al. 2019). Yet the role strain of separated fathers has also shown to be higher compared to married fathers (Umberson and Williams 1993), even though fathers’ psychological distress generally seemed to be diminished by higher-quality father-child contact after a separation (Vogt Yuan 2016).

In a review of 50 empirical studies primarily based on data from the 1980s to 1990s, Bauserman (2012) found that only SPC fathers were more satisfied with custody arrangements compared to SC parents. SPC mothers experienced less parenting burden and stress and both, SPC fathers and mothers, reported more emotional support and less conflict in their relationship with their former partners or spouses. Recent studies also found that SPC mothers had a more active social life, more time for leisure activities, and better employment opportunities compared to lone mothers (Bonnet et al. 2018; van der Heijden et al. 2015; Schnor et al. 2017; Steinbach 2019; Vanassche et al. 2017), presumably because of the freed time and resources due to sharing parenting responsibilities. Whereas reduced time pressure that allows for participation in gainful employment and networking seems to be the main benefit of SPC for mothers, SPC fathers experienced higher levels of time pressure compared to nonresidential fathers among a sample of Dutch divorced or separated parents (Van der Heijden et al. 2016). Findings on the overall psychological adjustment and health of SPC parents compared to SC parents, however, are inconsistent. Melli and Brown (2008) reported that, among a sample of U.S. divorced mothers and fathers, SPC parents were better off in terms of their physical and psychological health compared to SC parents. Other studies using data from Belgium and the Netherlands, however, did not find a direct association between SPC and parents’ well-being (Sodermans et al. 2015; Spruijt and Duindam 2009). Potential reasons for these surprising null effects—particularly given the highly selective nature of SPC parents’ characteristics—could be related to the gap between official arrangements (legal and declared) and daily practices. Some research showed that rather unequal time splits are often common practice in SPC arrangements, which still leave one parent shouldering the majority of childcare costs and responsibilities.

To address inconsistencies in prior findings, we further examine the association between SPC and parents’ well-being and physical health, as well as potential variations in these links by gender and educational attainment. Overall, we expect that SPC parents report better health and well-being than lone parents, who generally tend to suffer from poorer health and well-being outcomes compared to their partnered peers and particularly in less generous and more gender-biased family policy regimes such as Switzerland (Hypothesis 2). However, we also expect mothers’ health and well-being to benefit more from SPC compared to fathers because mothers, as traditional caregivers, stand more to gain from stronger father involvement in care duties in terms of freed up resources for employment and leisure time (Hypothesis 2a). Lastly, we expect SPC parents with higher levels of education attainment to report better health and well-being compared to their lower-educated peers because, in addition to being more likely to opt for SPC, higher-educated parents may have the sufficient resources and problem-solving skills to implement SPC successfully, particularly with limited institutional support for SPC, and can benefit from it subsequently (Hypothesis 2b).

2 Method

2.1 Data

Data were drawn from the cross-sectional Families and Generations Survey (FGS) conducted by the Swiss Federal Office of Statistics in 2013 (N = 17,298; 53% females), which collected information on respondents’ current household composition, employment and partnership characteristics, fertility, and well-being. In addition, the FGS includes retrospective information on the respondents’ union and fertility histories. The survey randomly sampled permanent Swiss residents aged 15 to 80 years. Survey interviews were conducted in one of the three official languages, depending on the respondents’ preferences (German, French, and Italian), during computer assisted telephone interviews and via complementary online or paper questionnaires. Retention rate was 50% and 82% of the sample were Swiss nationals.

For the purpose of this study, we drew a subsample of respondents with at least one biological child aged 18 years or younger (N = 5,002), where both biological parents were not living in the same household at the time of data collection. Note that the parent that took part in the survey also provided basic socio-demographic information (e.g., sex and educational attainment) about the other biological parent of their child, but that this parent was not contacted to participate in the survey as well. Thus, the sample consisted of full-time residential parents with children in their household (i.e., lone parents), non-residential parents with children outside of the household providing information about their children and the parent taking care of them, and parents splitting parenting duties with children alternating between the parental residences. The final analytic sample consisted of 875 post-separation parents (59.8% females; M age(SD) = 43.90 (7.69)) of 1,269 minors (49.2% females; M age(SD) = 12.12 (4.59)).

2.2 Measures

SPC, again referring to children’s living arrangements and not legal custody, was assessed with two questions asking whether another parent takes care of the child on a regular basis (0 = no, 1 = yes), and if so, for how many days per month (M (SD) = 8.52 (5.44)). Based on this information, we differentiated four custody arrangements: no custody (i.e., non-residential parents; ‘NC’; 0 days), sole custody (i.e., full-time residential parents; ‘SC’; 0 days), visitation rights (‘VR’; 1–9 days per month), and shared physical custody (‘SPC’; for at least 30% of the time, i.e., 10 days or more per months). We chose a 70%:30% time split, rather than equal time splits, as the cut-off criteria to be classified as SPC because of the left-skewed distribution of the time spent in the other parents’ household.

We measured respondents’ own overall evaluation of their physical health, which is a commonly used and validated single-item health indicator (Idler and Benyamini 1997), by asking: “In general, would you say your health is …?” on a scale from 1 (very bad) to 5 (very good). Emotional well-being was measured with a shortened version of the Positive and Negative Affect Schedule (Watson et al. 1988). Participants were asked to rate how often they generally felt the following emotions on a scale from 1 (never) to 6 (always): happy, tired, lonely, energetic, sad, and nervous. Items were recoded so that higher values indicate higher levels of emotional well-being and a mean composite score was formed (Cronbach’s α = 0.73).

Lastly, information about respondents’ sex (0 = female; 1 = male), age (in full years), mother’s and father’s educational attainment (1 = primary [basic schooling]; 2 = secondary [vocational training]; 3 = tertiary [advanced vocational training or university degree]), children’s age (in full years), respondents’ employment status (0 = unemployed; 1 = employed), perceived economic difficulties (0 = no; 1 = yes), current partnership status (0 = single; 1 = partnered), time since the separation or divorce (in full years), whether child support payments were received or made (0 = no; 1 = yes), and the linguistic region of residence (1 = German-speaking cantons; 2 = French-speaking cantons; 3 = other [Italian or Romansh]) were available.

3 Results

3.1 Descriptive Results

Table 5.1 displays summary statistics of the sample and key study variables for parents and children. Descriptive results showed that, as expected, SPC was only reported for 10.9% of the children in post-separation families (vs. 32.6% of children outside of the parental household (NC), 35.9% of children in SC households, and 20.6% of children in VR arrangements). When a 50:50% time split is used, the share of SPC children even dropped to 6.5% (vs. 25.0% of children in VR arrangements). Not surprisingly, the majority of post- separation households with children were female-headed. For example, 87.5% of the SC parents were female compared to 10.5% of non-residential (NC) mothers with minors outside of their household. Furthermore, 92.6% of residential mothers granted another parent VR and 59.8% of SPC parents were female. SPC parents were overrepresented in the French-speaking Swiss cantons (55.8% vs. 28.4% of NC parents, 37.3% of SC parents, and 37.8% of parents with VR arrangements).

Table 5.1 Descriptive statistics of the key study variables by custody arrangements
Full size table

SPC parents, both mother and father, reported higher levels of education (41.7–48.2% with the highest level of education, respectively) compared to respondents in the NC, SC, and VR groups (27.6–35.8%, 23.1–34.0%, and 24.5–36.8% with the highest levels of education, respectively). Yet at the same time, a larger share of SPC parents reported to have financial difficulties (62.0% vs. 48.5% of NC parents, 56.1% of SC parents, and 56.8% of parents with VR). Most post-separation parents reported to be employed across all custody arrangements, even though the share of unemployed parents was highest among SPC parents (21.1% vs. 16.6% of NC parents, 20.0% of SC parents, and 15.4% of parents with VR). Note that within the SC and VR groups, most parents reported to work part-time (47.1% part-time vs. 32.9% full-time for SC parents and 61.7% part-time vs. 22.9% full-time for VR parents) compared to NC parents who worked predominantly full-time (73.9% vs. 9.5% part-time).

Lastly, SPC was most frequent in families with high-school aged children (33.3% for children aged 9–12 years and 39.2% for children aged 13–18 years) compared to other families. The share of SPC actually increases with children’s age: being the least in families with infants aged 0–2 years (2.9%) and growing gradually for children aged 3–5 years (8.7%) and those aged 6–8 years (15.9%).

3.2 Regression Results

In order to address the first hypothesis on the relatively privileged positon of SPC parents compared to NC, SC parents, or parents with VR, we employed multinomial regression models predicting the likelihood of belonging to each custody arrangement. Our data is hierarchical with potentially multiple children (Level 2) nested within families (Level 1), and yet the number of siblings did not allow the use of multilevel modeling (Mchildren in families(SD) = 1.80 (0.69)). Therefore, we estimated stepwise regression models with cluster robust standard errors that allow for intragroup correlation. SC served as reference category in all regression models. Model 1 included basic socio-demographic characteristics of the surveyed parent (e.g., age), as well as mothers’ and fathers’ education, and children’s age. Model 2 added partnership characteristics, such as time since the separation of the parents, and whether surveyed parents were currently partnered. In Model 3, we introduce employment status and the financial situation of the surveyed parent.

To ease interpretation and allow comparability across nested models, we report discrete differences in average marginal effects (AME) of the multinomial regression models in Table 5.2 (Long 2015). AME represent the average impact of the independent variable on the likelihood of each outcome category (i.e., belonging to each respective custody arrangement in our case). For continuous variables, the table shows average discrete change in the predicted probabilities for a one-unit increase in the predictor and, for categorical variables, it represents average differences in predicted probabilities for pairs of levels of the predictor.

Table 5.2 Discrete differences in average marginal effects for multinomial regression models predicting custody arrangements
Full size table

Results revealed some gender differences in the likelihood of belonging to different custody arrangements. Compared to women, men were more likely to belong the NC group (56–69%) and less likely to belong to the SC and VR groups (36–38% and 25–36%, respectively) across all of the models. Higher paternal education, both secondary and tertiary, compared to primary levels of educational attainment, were related to lower chances to belong to the SC group (13–21% and 14–23%, respectively) and higher chances to belong to the VR group (14–17% and 14–18%, respectively). Yet, there was no significant association between SPC and education. Children’s age was positively related to the likelihood of belonging to the SC group and negatively to the VR group, but again there was no significant link between children’s age and SPC across the models. Model 2 revealed that the more time had passed since the parental separation, chances was greater to belong to the SC group and lower to belong to the SPC group. However, the latter effect faded when employment and financial characteristics were introduced in the final model. Respondents in Italian- or Romansh-speaking cantons were less likely to belong to the SPC group and more likely to belong to the NC group in Model 3.

A second set of regression models predicting self-rated health and emotional well-being examined Hypothesis 2 on the positive link between physical and mental health and SPC compared to more traditional custody arrangements. We estimated stepwise regression models with basic socio-demographic characteristics and custody arrangements in Model 1, partnership characteristics in Model 2, and employment and financial characteristics in Model 3. Again, SC served as the reference category for the custody arrangements in all models. To explore the gender and educational differences in the link between physical or mental health and custody arrangements, we also examined interaction terms between custody arrangements and gender (Hypothesis 2a; Model 4) and custody arrangements and parental education (Hypothesis 2b; Model 5). In order to address potential selectivity effects into custody arrangements based on parents’ characteristics, we used the predicted probabilities (i.e., 1 – P(custody arrangement)) from the fully-adjusted multinomial model as regression weights for the these models. Again, a cluster robust standard error estimator was employed to account for the hierarchical structure of the data.

Table 5.3 shows that, contrary to our expectations, custody arrangements were neither significantly associated with self-rated health nor with emotional well-being across all models. Most of parents’ socio-demographic characteristics, such as age, gender, and educational attainment, were also not associated with self-rated health and emotional well-being across all models (expect for a negative link between parents’ age and well-being in Model 1, and a positive link between fathers’ tertiary vs. primary education and health in Model 1 and 2). For emotional well-being only, being partnered was related to higher ratings (Model 2–5) and living in a French-speaking canton (compared to German-speaking cantons; Model 1–2) to lower ratings of emotional well-being. The employment status and financial situation of the surveyed parent entered in Model 3, however, were significantly related to both health and well-being. More specifically, part- and full-time employed parents reported better health and higher ratings of emotional well-being compared to their unemployed peers (Model 3–5). Having financial difficulties was also linked to lower ratings of self-rated health and well-being (Model 3–5). Lastly, a significant SPC by gender interaction term for health only (Model 4) suggests that, contrary to our expectations, SPC fathers reported significantly poorer health compared to SPC mothers. In Model 5, a significant SPC by secondary (vs. primary) paternal education interaction term emerged for health only, which indicates that SPC parents where fathers had secondary schooling reported better health compared to SPC parents where fathers had primary schooling. Yet there was no difference in reported health between SPC fathers with secondary vs. tertiary schooling.

Table 5.3 Regression results predicting self-rated health and emotional well-being
Full size table

4 Discussion

Non-traditional custody arrangements, such as SPC, are on the rise among separated or divorced parents due to changes in cultural and legal practices across many Western nations (e.g., Bauserman 2012; Cancian et al. 2014; Kitterød and Lyngstad 2012). While the profile of SPC parents has pluralized in terms socio-demographic characteristics in countries where SPC has become a legal default or is institutionally-encouraged, SPC parents tend to be more educated and affluent compared to lone parents or those with VR arrangements in countries where SPC is less prevalent (Vanassche et al. 2017; Sodermans et al. 2013). Our study is, to our knowledge, the first to examine the socio-demographic make-up of SPC parents in Switzerland, which is a rather late and only very partial adopter of gender-neutral parenting policies, and adds to the sparse and contradicting findings on SPC and parents’ health and well-being.

In line with our expectations, SPC was practiced by a rather small share of post-separation parents with about 11%, even though a relative low cut-off point was used to define SPC (i.e., at least 30% of the time at the other parents’ home). When we used an equal-time sharing model, the percentage of SPC parents was closer to the incidence rate of about 5% reported from German survey data, where clear-cut legal pathways to SPC are equally sparse as in Switzerland (Walper 2016). We therefore conclude that opting for SPC upon separation still represents the exception rather than the rule in Switzerland, which could be attributed to the lack of legal and institutional support for SPC at the time of data collection in 2013. Gender-biased cultural practices in the workforce and, in turn, the private lives of young couples—particularly the traditional role distributions in the division of household and childcare tasks upon entry into parenthood (e.g., Bernardi et al. 2013; Bühlmann et al. 2009)—may further increase the likelihood of establishing more traditional (female-headed SC) post-separation custody arrangements upon separation (Juby et al. 2005). For example, we did observe descriptively that the household head in SC and VR arrangements was pre-dominantly female, which highlights persisting gender disparities in parental involvement and the shouldering of childcare duties in post-separation families (Bernardi et al. 2018; Bjarnason and Arnarsson 2011).

We also anticipated that Swiss SPC parents would represent a more privileged group in terms of educational attainment and financial assets compared to lone parents and those with VR arrangement because they would need to voluntarily seek out SPC given the lack of legal guidance and have the available resources to implement it (e.g., maintaining the child’s room at both parental homes or having duplicate sets of cloths, school supplies, etc.; Nielsen 2018; Schier and Hubert 2015; Steinbach 2019). Again descriptively, SPC parents tended to be more highly educated compared to the parents with other childcare arrangements, which was in line with our expectation. Yet, the share of parents reporting financial difficulties and unemployed parents was also highest among SPC parents in our descriptive statistics. The discrepancy between parents’ higher educational attainment on the one hand, and higher perceived economic strain on the other hand, may be related to the still more unevenly distributed time that children alternate between the parental households and the higher costs of SPC compared to other custody arrangements, as we outlined before. First, the 70%:30% time split could indicate that a bulk of care duties and child-related expenses still needs to be shouldered by one parent (Steinbach 2019), which, in addition to potentially reduced institutional support due to the (formally) higher contribution to parenting duties from both parents, could contribute to the higher perceived economic strain of SPC parents. The high costs of childcare in Switzerland, which could subsequently lead to more precarious work situations for parents unable to afford childcare expenses (Struffolino and Bernardi 2017), on top of the costs related to SPC, could also erode any gains from SPC, such as freed up time and resources to invest in other life domains such as work, dating, or leisure (e.g., Van der Heijden et al. 2015). Thus, SPC could represent more of an economic resource drain than a relief for the more involved parent, when the time is not truly shared equally.

Lastly, we predicted that overall SPC parents would be healthier and happier parents compared to NC, SC, and VR parents because of their more privileged position in terms of SES (Bauserman 2012; Vanassche et al. 2017) and because SPC parents may experience less parental role strain and related stressors (Umberson et al. 2010; Van der Heijden et al. 2016). This expectation was not confirmed in our models. It could be the case that experiencing such fundamental changes in social roles and the reconfiguration of relationships between family members after a divorce or separation (Amato 2010; Bengtson and Allen 2009) may have the effect of ‘leveling’ or reducing social differentiation between parents (Leopold and Leopold 2016). Or, in other words, union dissolution itself may leave a more permanent mark on or represent a “turning point” for all families, regardless of prior SES or subsequently chosen custody arrangement, which would explain the lack of differences between the custody arrangements. Additionally, we also anticipated that SPC mothers may experience more health and well-being benefits compared to SPC fathers because shared parenting would alleviate mothers from their more traditional role as full-time custodial caretaker and may free up time and energy to invest into other life domains (Bonnet et al. 2018; Bauserman 2012; Schnor et al. 2017), which could in turn foster mothers’ health and well-being. Our models partially supported that notion because SPC fathers tended to report poorer health compared to SPC mothers. It could be the case that SPC fathers may perceive health impairments because, in the Swiss context, SPC was still relatively rare and a rather non-traditional post-separation custody arrangement at the time of data collection for which fathers may receive little institutional support or social acceptance (Bjork 2013; Haas and Hwang 2019).

Because SPC fathers may be likely and expected to continue their engagement in full-time employment compared to mothers who often work part-time only, family-work-conciliation issues and parental role strain may weigh harder on their health (e.g., Umberson and Williams 1993). At the same time, we found that only SPC parents where fathers have secondary levels of schooling reported better health compared to SPC parents where fathers had primary levels of schooling only. It could therefore be the case that the positive link between higher levels of paternal education and parental health may buffer with the previously outlined gendered effects that may contribute to SPC fathers’ poorer health ratings. For example, SPC families where fathers have higher levels of education may simply have more resources to outsource some care responsibilities (e.g., paying for costly childcare), which may facilitate fathers’ family-work-conciliation, or to seek professional help in light of conflict (e.g., seeing a counselor to establish mutually beneficial SPC routines), while still enjoying the health benefits of maintaining father-child bonds (Vogt Yuan 2016). There may not have been a further gain from fathers’ tertiary vs. secondary schooling because of ceiling effects or because fathers’ potential outsourcing of care duties may in turn contribute financial difficulties, which we observed descriptively.

Nevertheless, involvement in the labor market—whether part- or full-time—seemed to be overall a driving force of the physical health and emotional well-being of post-separation parents, which is in line with previous research on lone mothers (Struffolino et al. 2016). Because the presence of a new partner was related to higher levels of well-being as well, one could argue that both being able to repartner and to engage in gainful employment may indirectly be facilitated by SPC because these parents simply have more time to be engaged in life domains other than parenting (van der Heijden et al. 2015; Vanassche et al. 2017). However, in a plausible inversed causal relation, more well-adjusted parents may also simply be more likely to take up employment and find a new partner.

4.1 Limitations and Future Research

Our study has several limitations. First, we used a rather low cut-off point to define SPC for our sample due to low case numbers for SPC and a left-skewed distribution of the amount of shared time between households. With our criteria, children needed to spend at 30% of the time alternating between parental homes to be grouped into SPC. Even though definitions and criteria of SPC vary widely between studies (i.e., ranging from equal amounts of time up to at least one-quarter of time; Bauserman 2012), unequal time sharing still implies that one parent—often the mother (Bjarnason and Arnarsson 2011)—is likely shouldering the majority of daily childcare task and responsibilities. It could therefore be the case that some SPC parents and those with VR are more similar than we anticipated in terms of sharing parental roles and duties, which could explain the lack of significant differences in health and well-being by custody arrangements. Results may have looked somewhat different if we could have used a stricter criteria for SPC with a larger sample of post-separation and SPC parents specifically.

Second, custody arrangements among post-separation families are likely to evolve and change over time (Poortman and van Gaalen 2017; Smyth and Moloney 2008). Schedules and routines that parents agreed on upon separation or divorce—even if court mandated—may be revised if deemed not feasible in practice, altered to the child’s needs or wishes, or adapted to new realities such as the emergence of a new partner in a parents’ romantic life or even new-partner fertility. Our cross-sectional measure of custody arrangements, however, only provided a single snapshot into the routines of separated parents and was therefore not able to capture any dynamic changes over time. Targeted longitudinal data collection will be needed to explore temporal dynamics in the establishment and development of SPC and other custody arrangements among post-separation families. Our dataset also captures parents at different stages after their separation or divorce because the time since separation or divorce varied considerably between former couples. Future research based on these longitudinal follow-ups will need to examine the stability of SPC because prior research has suggested that particularly fathers’ investment in non-residential children has shown to fade over time (Berger et al. 2012). It would also have been interesting to look at SPC parents’ division of childcare-related task or their ratings of satisfaction with SPC, yet this information was not available in the secondary dataset that was not specifically designed with a focus on studying post-separation custody arrangements. Unfortunately, using existing panel data for Switzerland was still not an option because of the relatively small number of observations available in each wave for this subpopulation, which makes it difficult to apply time-lagged models. As SPC diffuses further among the Swiss population and the panel progresses, future research will be able to exploit bigger sample sizes and should be able to address and disentangle causal paths.

Despite these limitations we are confident that our study provides some first evidence of the characteristics of SPC parents as well as correlates of SPC (or the lack thereof) for separated parents in Switzerland, which was still a rather gender-biased and traditional family policy context that promoted more traditional childcare arrangements at the time of data collection in 2013. We therefore argued that the lack of institutional support for SPC and the rather traditional role distribution among Swiss couples hindered the spread of SPC and reinforced gendered SC models. Furthermore, because of the high costs of childcare in Switzerland, SPC may represent more of a resource drain than a relief for the parent shouldering the majority of the care responsibilities, when time is not shared equally between caregivers. Nevertheless, SPC did neither seem to diminish nor foster parents’ health and well-being. Future data collection and research will need to monitor the spread of SPC and its’ potential ripple effects of the 2017 legislative change in Switzerland that established SPC as a routinely investigated option for post-separation custody arrangements.