Lone Mothers’ Repartnering Trajectories and Health: Does the Welfare Context Matter?

Abstract

We examined the relationship between lone mothers’ repartnering and health in three welfare contexts: the dual-earner, market-oriented, and general family policy model. Drawing on the resources and crisis models, we applied mixture modelling for spell data of the Harmonized Histories data set. We uncovered six distinct repartnering trajectories that varied regarding the timing, type, and stability of higher order unions for different cohorts of lone mothers. Unstable repartnering was more frequent in market-oriented contexts, while contexts with more comprehensive family support fostered more stable repartnering. Although repartnering trajectories were overall not associated with health, these associations differed by welfare context. Mothers experiencing repartnering, though unstable, reported to enjoy better health if living in market-oriented contexts rather than in general or dual-earner contexts. Altogether, our findings suggest that even if higher financial needs in less-generous welfare may translate in more unstable repartnering histories, this seems to have positive spillover effects on mothers’ health.

Keywords

lone mothers repartnering health disparities welfare states family policy

Family life has changed rapidly with the diffusion of living arrangements alternative to the predominant nuclear family model, including diverse family forms such as lone parents and stepfamilies (Sobotka & Toulemon, 2008). The incidence of lone parenthood is rapidly rising in many Western countries, and lone parents are becoming increasingly heterogeneous with regard to their sociodemographic characteristics (Bernardi, Mortelmans, & Larenza, 2018), even though the large majority are women and health penalties for lone compared with partnered mothers remain high (e.g., Avison & Davies, 2005; Wickrama et al., 2006). Despite variation across institutional welfare contexts (Burstrom et al., 2010; Pollmann-Schult, 2018), health disparities for lone mothers persist and are partly attributed to lone mothers’ higher levels of psychosocial and financial stress because they tend to work in low-paying jobs and be left alone to care and provide for their children (Dziak, Janzen, & Muhajarine, 2010).

Increases in lone parenthood ran parallel with more frequent and faster family transitions toward recomposed and blended families among recent cohorts (Bernardi et al., 2018; Bzostek, McLanahan, & Carlson, 2012). Because the presence of a new partner is positively associated with one’s well-being (Wang & Amato, 2000), and could bring another adult into the family who may contribute to the family income and care responsibilities, lone mothers have been thought to benefit from repartnering especially. Mixed empirical evidence on the link between repartnering and health may be due to the fact that prior studies often looked at repartnering as a way to exit lone motherhood with less consideration for the duration and stability of such unions. In addition to modeling repartnering choices as a single life event or over short time intervals—rather than a complex trajectory of multiple union statuses across the life course—previous studies rarely placed the association between repartnering and health within institutional contexts (Ivanova, Kalmijn, & Uunk, 2013; Perelli-Harris & Lyons-Amos, 2015).

However, repartnering choices also differ across institutional settings, like it has been shown for the choice of cohabitation over marriage in higher order unions (Gałęzewska, Perelli-Harris, & Berrington, 2017). Because the ripple effects of these emerging repartnering behaviors within different institutional contexts remain unclear, our study adds to the literature on lone mothers’ repartnering and health in three ways. First, we examine the association between repartnering and health by studying repartnering trajectories holistically rather than single events, which only show associations between health and a particular union type at a certain point in time. Modeling repartnering trajectories as a whole allows to account for timing, sequencing, and density of transitions in the total association between repartnering and health across the life course. Second, we expand the evidence on single event and country studies by examining whether repartnering trajectories differ across welfare contexts. Last, we contextualize repartnering-health-linkages by examining whether these associations vary systematically across welfare contexts to provide a more nuanced portrait of lone mothers’ repartnering and health.

Lone Mothers’ Repartnering

Over the past decades, the share of lone parent households increased in many Western countries due to the rise of union dissolution and childbirth outside of marriage (e.g., Amato, 2010; Sobotka & Toulemon, 2008). Lone parent families accounted for 16% of all families in the European Union in 2011 (Eurostat, 2015), yet 84% of lone parents are women defined as mothers who solely reside with and care for at least one minor. Up until the 1970s, lone mothers tended to be either widows or young, unmarried, and often of lower socioeconomic standing, whereas this group now comprises more divorced or separated parents (Bernardi et al., 2018). Consequently, lone mothers became more diverse with regard to age, education, and socioeconomic standing, which transformed the experience of lone motherhood and changed the pool of individuals who may look for a new partner (Bzostek et al., 2012).

The likelihood of experiencing an episode of lone motherhood increased over the past decades, but Bernardi et al. (2018) estimated that recent cohorts of lone mothers repartnered within 2 to 4 years compared with 8 to 10 years for older cohorts. Because both, the chances to repartner and of union dissolution, have increased simultaneously, we expect to find at least three types of repartnering trajectories for lone mothers: (1) mothers, primarily from older cohorts, who do not repartner (traditional lone mothers); (2) mothers who repartner and remain in that higher order union (stable repartnering); and (3) mothers who repartner and, at some point, split up again (unstable repartnering). Repartnering, in these cases, may not necessarily mean (re)marriage because lone mothers often choose alternative and more flexible family forms such as cohabitation or living apart together (Perelli-Harris & Lyons-Amos, 2015).

Prior research focused largely on examining the likelihood of repartnering and, to a lesser degree, on union trajectories after lone parenthood. Studies showed that mothers, regardless of their age, were generally less likely to repartner than fathers and childless women (e.g., Beaujouan, 2012; De Graaf & Kalmijn, 2003; Vanassche, Corijn, Matthijs, & Swicegood, 2015). Higher educated mothers and women, in general, were more attractive on the repartnering market than lower educated ones (Bastin, 2012). Across five European countries, children’s age was positively associated with mothers’ chances to repartner, which can be explained with mothers’ freed-up resources to engage in dating as children become more independent (Ivanova et al., 2013). Having multiple children and sole physical custody, however, reduced mothers’ chances to repartner (Schnor, Pasteels, & Van Bavel, 2017).

Lone Mothers’ Repartnering and Health

Despite recent demographic shifts in the experience of lone motherhood, being a lone mother is still a strong predictor of poverty, fragmented work histories, lower life satisfaction, and poorer health (e.g., Avison & Davies, 2005; Brady & Burroway, 2012; Cooper, McLanahan, Meadows, & Brooks-Gunn, 2009; Millar, 2010; Pollmann-Schult, 2018; Wickrama et al., 2006). Particularly, the exposure to chronic stressors in multiple life domains contributes to lone mothers’ adverse health trajectories (Kühn, 2018; Struffolino, Bernardi, & Voorpostel, 2016). Repartnering has therefore been interpreted as one strategy to improve lone mothers’ disadvantaged situation. However, there are arguments that the formation of new relationships could also represent a source of strain for lone mothers because the resource drain associated with unstable repartnering, due to possible changes in employment and housing, may bypass any benefit related to the additional resources brought in by a new partner (e.g., Williams & Umberson, 2004). We refer to two competing theoretical models—the resource model and the stress model—to generate contrasting hypotheses on the association of repartnering and health.

The resource model states that union formation is positively associated with health because couples share and pool resources such as income, social ties (Williams, Sassler, & Nicholson, 2008), and, particularly for lone mothers, potentially shared parental care responsibilities. Consequently, gaps in health between lone and repartnered mothers should increase over time. Osborne, Berger, and Magnuson (2012) showed that, among a sample of U.S. low-income mothers, coresidential repartnering was not or only marginally positively associated with mothers’ mental health within 5 years after childbirth. Using German panel data, Kühn (2018) found that the duration in lone motherhood was linked positively to well-being and only negatively to satisfaction with health, which could be attributed to the fact that mothers with worse health were less likely to repartner (Pevalin & Ermisch, 2004). Williams et al. (2008) suggested that health benefits of repartnering only apply to more lasting unions, even though they modeled union transitions as a single, isolated event. For more unstable patterns, the resource drain associated with exiting higher-order unions, for which lone mothers are at a heightened risk (Bastin, 2012), may cancel out any benefits gained by pooling resources.

In contrast, the crisis model assumes that changes in relationship statuses, such as transitions in and out of unions, represent stressful events with negative effects on mothers, which may fade over time (Amato, 2010; Williams & Umberson, 2004). Additionally, repartnering deteriorated prior arrangements in terms of decreased child support payments from the biological father and reduced father–child contact (Berger, Cancian, & Meyer, 2012). In previous studies, experiencing any kind of union transition, and particularly the exposure to multiple transitions, was linked to mental health issues (Meadows, McLanahan, & Brooks-Gunn, 2008). Union formation with a nonbiological father was also associated with increased parenting stress again within 5 years after childbirth among U.S. low-income mothers (Cooper et al., 2009). Repartnering may therefore be a costly endeavor that requires time and resources, an effect that could cumulate and magnify over time for unstable repartnering, where new partners drain more resources than they contribute.

Against this backdrop, we expect that, in line with the resource model, both stable and unstable patterns of repartnering are positively associated with lone mothers’ health (Hypothesis 1a). Based on the crisis model, however, we expect any repartnering to be negatively associated with mothers’ health (Hypothesis 1b). We further expect stronger health penalties for unstable repartnering trajectories (Hypothesis 1c), because these may either erode prior gains from pooled resources (resource model) or magnify drained resources over time (crisis model).

The Role of Welfare Contexts

Lone mothers’ repartnering trajectories and their potential ripple effects on health are not only a product of individual choices but rather of individuals’ embeddedness into institutional contexts (Mayer, 2004). Welfare states set legal regulations and provide support systems that either reward or discourage certain behaviors, such as mothers’ labor market participation (see Thévenon, 2011, for an overview across Organisation for Economic Co-operation and Development [OECD] countries). Cross-national comparisons suggested that more generous institutions and policies relevant for lone mothers, such as public child care, tax credits, or income supplements (Bernardi et al., 2018), buffered mothers from health and well-being penalties (Burstrom et al., 2010; Pollmann-Schult, 2018). More specifically, Burstrom et al. (2010) distinguished three welfare contexts based on countries’ level of gender equality and welfare support for dual-earner families to examine institutional influences on lone mothers’ health systematically. Based on this classification, our study focuses on six European countries as exemplary cases for each setting to contextualize repartnering-health-linkages, which have mostly been studied with U.S. data (e.g., Cooper et al., 2009).

First, the dual-earner model (here Sweden and Norway) highly encourages both parents to work by providing tax advantages for dual-earners, universal child care at a low net cost, and generous parental leave with compensations above OECD average (Thévenon, 2011). Levels of gender equality are high, which can partly be attributed to paternal leave policies that promote higher father involvement and universal welfare transfer to limit financial deterioration after union dissolution (e.g., in Sweden; Gałęzewska et al., 2017). Second, the general model (here Germany and France) actively discourages both parents to work through taxation, which leaves little fiscal benefit for dual-earners, promoting a more traditional two-parent norm and lower levels of gender equality (Thévenon, 2011). At the same time, financial support for all families regardless of financial need is above the OECD average, while leave entitlements with no strong incentives for paternal involvement and child care provision are average compared with other OECD countries. Third, the market-oriented model (here the United Kingdom and Switzerland) leaves family organization as private matters by providing only limited parental leave and a lower supply of public child care at higher costs compared with the OECD average, which may erode earnings from gainful employment (Thévenon, 2011). However, higher means-tested cash support for poor families, including lone mothers, are used as antipoverty policies within this context.

Family policies also shape lone mothers’ repartnering behaviors (Ivanova et al., 2013) by regulating mothers’ economic independence through the provision of child care that influences their labor market attachment (e.g., dual-earner model), setting means-tested welfare transfers (e.g., market-oriented model), or promoting father involvement rather than traditional gender norms (e.g., dual-earner vs. general model). Dewilde and Uunk (2008) showed that social welfare reliance delayed women’s entry into remarriage more in high- than in low-welfare countries, perhaps because generous welfare support may make women less economically dependent on a new partner (Griffiths, 2017). We therefore predict that in the market-oriented model, lone mothers are more likely to enter unstable repartnering patterns (Hypothesis 2), because limited state support regarding family-work-reconciliation increases women’s economic dependence on a partner, which could contribute to the formation of less attractive or beneficial unions. Last, we expect a stronger negative link between unstable repartnering and health in the market-oriented context (Hypothesis 3) because limited state support may aggravate the resource drain associated with union transitions, which in turn fosters existing disparities among lone mothers.

Method

Data

We drew a sample of lone mothers, who had lived alone with at least one biological child younger than the age of 18 years for one full year or longer between the ages of 15 and 55 years, from six countries representing three welfare contexts: the dual-earner (Sweden, Norway), market-oriented (the United Kingdom, Switzerland), and general model (Germany, France). Data for Sweden, Norway, Germany, and France stemmed from the first waves of the Generation and Gender Surveys (GGS; United Nations, 2005), which collect nationally representative samples of men and women between the ages of 18 and 79 years and contain individuals’ ratings on well-being and family dynamics. The British sample was drawn from the 2005/2006 Wave of the British Household Panel Survey, and we added a Swiss sample from the Families and Generations Survey (FGS) collected in 2013. For our pooled sample of 8,941 lone mothers, we also analyzed standardized information on retrospective union and fertility histories in spell format that is summarized in the Harmonized Histories data set based on information from national GGSs and the British Household Panel Survey (Perelli-Harris, Kreyenfeld, & Kubisch, 2011). We further supplemented Harmonized Histories with retrospective fertility and union history information from the 2013 FGS.

Measures

Mothers’ partnership histories were constructed based on the date of union formation, dissolution, and union type (cohabitation or marriage, and separation or divorce) for up to nine separate unions. Cohabitation referred to coresident relationships of at least 3 months. Unions with missing information on start, as well as start and end dates, were excluded. We categorized respondents’ reported unions into five categories that were established in prior research (Perelli-Harris & Lyons-Amos, 2015): (0) never in a union, (1) cohabiting, (2) marriage preceded by cohabitation, (3) direct marriage, and (4) single after a separation. This information was then expanded into person-years, which provides a sequence of union statuses for each respondent aged 15 to 55 years. We measured respondents’ overall evaluation of their health, which is a widely used and validated single-item health indicator (Idler & Benyamini, 1997), by asking: “In general, would you say your health is . . . ?” on a scale from 1 (very bad) to 5 (very good). Unfortunately, this health indicator was only assessed cross-sectionally at the time of the respective survey. Information about participants’ age at first lone mother spell (in full years), number of children (count), number of higher order unions after first lone mother spell (0 = 0 to 1; 1 = 2 or more), and total duration in lone motherhood (in full years) were derived from the retrospective fertility and partnership histories in Harmonized Histories and the FGS. Additionally, mothers’ cohort membership (born 1 = 1925-1945; 2 = 1946-1970; 3 = 1971-1994) and educational attainment (from 1 = low to 3 = high) were taken from the respective surveys.

Analytic Strategy

We employed latent class growth curve models (LCGCM) for event history data using Mplus 7.4 (Mikolai & Lyons-Amos, 2017) to extract latent trajectories of repartnering based on lone mothers’ categorized partnership histories. This holistic approach is particularly suited to capture the occurrence and path dependency of multiple life events, such as changing partnership statuses, because it identifies homogeneous, subgroup-specific patterns of status configurations within a larger heterogeneous population. Because our focus was on lone mothers’ repartnering trajectories, we restricted our analyses to unions that were reported after the first occurrence of a lone mother spell (i.e., higher order unions). Mothers’ age at the beginning of the partnership histories varied across the sample as a result.

We first fitted a series of LCGCM starting with a model with one and up to eight latent classes. The appropriate number of latent classes for the final model was determined by using several model fit indices, such as Akaike information criterion (AIC), Bayesian information criterion (BIC), and the Vuong-Lo-Mendell-Rubin and the Lo–Mendell–Rubin likelihood ratio test (LRT), the distribution of class counts across models, model parsimony, and the interpretability of model solutions. Information criteria, such as AIC and BIC, should be as low as possible, whereas LRTs compare the fit of a given model with the fit of a model with one class less.

In order to address our hypotheses, we then added distal variables to our final model. Distal variables are not part of the estimation process for deriving latent classes, but they describe and compare the composition of latent classes. We regressed the categorical latent class variable (i.e., class membership) on health, welfare context, and health × welfare context. Note that because health was only measured cross-sectionally at the time of the national surveys, regrettably we cannot examine changes in health due to repartnering. Our research focuses on the correlational link between long-term repartnering trajectories and subsequent mothers’ health.

Predictors were entered hierarchically into the model using robust standard errors to adjust for regional clustering. All models controlled for age at first lone mother spell, cohort, education, number of children, total duration in lone motherhood, and number of higher order unions. An overview of descriptive statistics and sample compositions by country is summarized in Table 1 and by latent class in the appendix Table A (supplementary material available online).

Table 1.

Descriptive Sample Statistics by Country (N_pooled = 8,941).

	General model		Dual-earner model		Market-oriented model
Indicators	GGS, Germany	GGS, France	GGS, Norway	GGS, Sweden	BHPS, The United Kingdom	FGS, Switzerland
n (%)	1,162 (13.0)	1,270 (14.2)	1,588 (17.8)	1,086 (12.1)	3,041 (34.0)	794 (8.9)
Survey year	2005	2005	2007-2008	2012-2013	2005-2006	2013
Age, M (SD)	51.36 (15.69)	52.03 (14.29)	50.72 (15.69)	55.45 (12.61)	49.36 (13.70)	55.19 (12.46)
Cohort, n (%)
1920/1930s	276 (23.8)	259 (20.4)	174 (11.0)	109 (10.0)	409 (13.5)	62 (7.8)
1940/1950s	415 (35.7)	552 (43.5)	736 (45.7)	499 (46.0)	1,301 (42.8)	365 (46.0)
1960/1970s	424 (36.5)	428 (33.7)	653 (41.1)	441 (40.6)	1,224 (40.2)	339 (42.7)
1980/1990s	47 (4.0)	31 (2.4)	35 (2.2)	37 (3.4)	107 (3.5)	28 (3.5)
Education, n (%)
Low	258 (22.7)	580 (45.7)	421 (26.8)	168 (15.5)	794 (28.0)	170 (21.5)
Medium	656 (57.8)	461 (36.3)	685 (43.5)	567 (52.5)	1,010 (35.7)	481 (60.8)
High	221 (19.5)	229 (18.0)	467 (29.7)	346 (32.0)	1,030 (36.3)	140 (17.7)
Number of higher order unions, M (SD)	0.49 (0.73)	0.63 (0.90)	0.85 (0.90)	0.95 (0.92)	0.96 (0.67)	0.86 (0.89)
Number of children, M (SD)	2.04 (1.15)	2.36 (1.47)	2.26 (1.03)	2.26 (1.01)	2.47 (1.28)	2.04 (0.96)
Age at first LM spell, years, M (SD)	26.73 (7.13)	30.77 (8.65)	30.66 (8.87)	32.12 (8.82)	27.71 (7.60)	28.52 (7.55)
Years as LM, M (SD)	11.93 (8.78)	9.60 (7.81)	7.64 (6.28)	8.07 (6.82)	12.38 (8.52)	8.94 (7.63)
Health, M (SD)	3.73 (0.88)	3.70 (0.91)	3.24 (1.21)	3.91 (0.93)	3.67 (1.00)	3.71 (0.89)

Note. LM = lone mother; GGS = Generation and Gender Surveys; BHPS = British Household Panel Survey; FGS = Families and Generations Survey. Self-rated health ranged from 1 = very bad to 5 = very good.

Results

Model Selection

The model fit indices AIC and (sample-size adjusted) BIC for the fitted LCGCMs in ascending order, from one latent class to up to eight classes, continued to decrease across all models. Yet the decline in information criteria seemed to level off sharply after the three-class solution. Insignificant p values on the .01 level for both LRTs for only the eight-class solution, however, suggested that the seven-class solution was a better fit than the eight-class solution. However, based on the final class counts and the interpretability of solutions, we chose the six-class solution as the final model. This model offered a more fine-grained picture of repartnering patterns compared with the three-class solution and reasonable class counts (i.e., at least over 5% of the sample in a given class) compared with the seven-class solution, while still performing well on the fit indices. A table of model fit indices is provided in the appendix Table B (supplementary material available online).

Latent Class Growth Curve Results

Figure 1 shows the six latent class trajectories that were extracted from the final model. The curves represent the probabilities of being in a given union status at any given year after entering the first lone parent spell until age 55. Our six classes fell into three broad categories of repartnering trajectories, namely, traditional lone mothers, unstable repartnering, and stable repartnering. These categories shared important characteristics, as well as distinct features, which we used to differentiate them further.

Figure 1.

Latent classes of repartnering trajectories for the six-class solution.

The traditional lone mother trajectories, Class 1 (n = 1,691) and 6 (n = 2,658), were both composed of lone mothers for whom lone parenthood seems to be an absorbing state; they were most likely to not repartner after having lone parenthood. However, the classes did differ with regard to their pathway into lone parenthood. Class 1 represented mothers who were never partnered, yet had a child relatively early, and remained without a partner subsequently. This class is mostly likely to be driven by older cohorts, for whom lone parenthood was a “quasi-absorbing state” due to a higher level of stigmatization for single, unmarried mothers and the limited pool of available partners both for lone mothers and widows. Class 6 comprised mothers who entered lone parenthood through union dissolution and remained single subsequently.

In both unstable repartnering trajectories, Class 3 (n = 836) and 4 (n = 739), mothers were relatively likely to repartner at some point, but the likelihood to be single after a separation was increasing rapidly toward the end of the observation period. The timing of when repartnering peaks, however, varied between the two classes. For Class 4, the likelihood of repartnering peaked later after having become a lone mother compared with Class 3.

Last, for the stable repartnering trajectories, Class 2 (n = 1,376) and 5 (n = 1,641), the chances to be single after break up were either decreasing (Class 2) or remained very low over the observation period (Class 5). Even though the likelihood of any given repartnering status (e.g., direct marriage or cohabitation) were rather stable over the observation period for both classes, the chance of entering marriage without prior cohabitation was slightly higher for Class 2. In contrast, Class 5 had a higher likelihood of cohabitation followed by marriage.

Multinomial Regression Results

Next, we employed stepwise multinomial regression models to test the associations between class membership and health status (Model 1), class membership and welfare context (Model 2), and the interaction between both (Model 3). Traditional lone mothers who were never partnered (Class 1) served as reference group for the regression models. To ease interpretation, we plotted the predicted probabilities of falling into each latent class by different levels of health status in Figure 2. Raw coefficients of the models are provided in the appendix Table C (supplementary material available online). Contrary to our first Hypotheses 1a and 1b on the potential benefits (resource model) or drawbacks (crisis model) of repartnering, belonging to either the stable or unstable repartnering trajectories was not associated with health. Figure 2 shows that health status varied only very little within each latent class, except for poorer health ratings for mothers who entered lone parenthood after a separation and never repartnered (Class 6; traditional lone mothers).

Figure 2.

Predicted probabilities of class membership by health status.

Figure 3 shows the predicted probabilities of falling into a given latent class by welfare context. It illustrates the results of the model testing of Hypothesis 2, which was confirmed. Lone mothers in market-oriented contexts were more likely to either belong to unstable repartnering trajectories (Class 3 and Class 4) or to the traditional lone mothers who were never partnered (Class 1). Mothers in the dual-earner context were either more likely to be in the stable repartnering groups, particularly Class 5 that had a higher likelihood of cohabitation before marriage, or traditional lone mothers who entered lone parenthood through union dissolution (Class 6). Participants in the general welfare context were more likely to be traditional lone mothers, particularly again the class of mothers that entered lone parenthood through separation (Class 6).

Figure 3.

Predicted probabilities of class membership by welfare context.

Last, to test whether the anticipated link between unstable repartnering and poor health was more pronounced within the market-oriented context (Hypothesis 3), we plotted the predicted probabilities of falling into a given latent class by welfare context and health status in Figure 4. Belonging to the unstable repartnering trajectories was related to better health among mothers in the market-oriented welfare context compared with the other two contexts. Yet among mothers in the market-oriented welfare context, belonging to the traditional lone mother trajectory that entered lone parenthood through separation (Class 6) was related to worse health compared with the general and dual-earner context. Belonging to the stable repartnering trajectories was related to better health for those mothers within the dual-earner context, particularly for Class 5.

Figure 4.

Predicted probabilities of class membership by welfare context and health status.

Discussion and Conclusion

The prevalence of higher order unions among lone mothers has increased substantially, and repartnering occurs at a much faster pace than ever before (Bernardi et al., 2018). While such developments could suggest that the existing health disparities between partnered and lone mothers shrink (e.g., Kühn, 2018; Wickrama et al., 2006), prior studies on the link between repartnering and health yielded mixed results (e.g., Osborne et al., 2012; Williams et al., 2008). However, these studies modeled repartnering mostly as a single event, or over a short time period, and primarily based on U.S. data only. Our study addressed this timely issue by examining long-term trajectories of repartnering for different cohorts of lone mothers from three distinct welfare contexts in Europe and linking the derived latent repartnering trajectories to health subsequently.

Repartnering-health-linkages can be probed from two competing theoretical angles. The resource model proposes that finding a new partner may liberate mothers’ added role strain of being the sole breadwinner and caregiver, if the new partner contributes to household expenses or care duties (Williams & Umberson, 2004). Mothers’ health may further benefit from the emotional support that a new partner could provide (Wang & Amato, 2000). On the contrary, experiencing family transitions—whether union formation or dissolution—could add additional stress to the family configuration, which harms mothers’ health (i.e., crisis model; Cooper et al., 2009). Lone mothers’ health could also suffer in newly blended families because mothers, who are more likely to repartner with another parent (Vanassche et al., 2015), reported more parenting stress and less favorable perceptions of their partners’ coparenting skills if the partners’ children lived with the couple (Guzzo, Hemez, Anderson, Manning, & Brown, 2019).

We found no systematic links between repartnering trajectories and health and therefore no support for either the crisis and resource model, or our hypothesis that particularly unstable repartnering is associated with poorer health. One potential explanation for these noneffects could be the selective nature of repartnering itself because healthier and more privileged mothers are more likely to repartner (e.g., Bastin, 2012; Pevalin & Ermisch, 2004). Nevertheless, if that were the case, we could have also expected poorer health ratings among traditional lone mothers with the lowest chances to repartner, which we did not find. Because mothers’ health was only measured once at the time of the respective national survey, it is possible that their health status at the time varied as a function of timing, duration, and density of transitions related to repartnering behaviors. For example, some mothers were long repartnered, or separated (e.g., the traditional lone mothers), compared with others who had just recently transitioned at data collection, which could in itself affect their health status. Even though the design of our study does not allow to make causal claims about the directionality of repartnering-health-linkages, we adjusted and controlled our models for the influence of timing (i.e., age at first lone mother spell), total duration (in lone motherhood), and density (i.e., number of higher order unions) of union transitions.

We also considered the role of welfare contexts in shaping repartnering trajectories and its links to health because both repartnering and health have shown to depend on the characteristics of institutional settings (e.g., Burstrom et al., 2010; De Graaf & Kalmijn, 2003; Gałęzewska et al., 2017). Three European welfare contexts were examined that differ considerably in their amount of support they offer for gender equality and families (Burstrom et al., 2010), which in turn affect lone mothers’ economic dependence on a new partner (Griffiths, 2017; Ivanova et al., 2013). Each welfare context was represented by two countries in our analyses: Sweden and Norway for the most generous and egalitarian dual-earner model, Germany and France for the more conservative general model, and Switzerland and the United Kingdom for the market-oriented model with the least public support across the three contexts. In line with our second hypothesis, we did observe that mothers in the least generous market-oriented welfare context were most likely to belong to the more unstable repartnering trajectories. This finding could indicate that lacking state support may threaten mothers’ economic independence and therefore push them into more fragile unions (Dewilde & Uunk, 2008). Relatedly, Pasteels and Mortelmans (2017) found that women in lower income quintiles were more likely to repartner compared with their more affluent counterparts, which seems to support the economic need hypothesis as one driver of repartnering.

Although there was no systematic link between repartnering and health, the pattern of results changed when we introduced welfare context as a moderator in this association. Contrary to our expectation, we found that mothers in the market-oriented welfare context, who were likely to belong to the unstable repartnering patterns, reported better health. Alternatively, one could speculate that unstable repartnering is not associated with poorer health within this context precisely because it seems more common in the market-oriented model and therefore presumably less selective or discriminatory compared with the other contexts with fewer mothers in this trajectory. It is also important to note, however, that not all mothers in the unstable trajectories were more likely to have denser union histories because only mothers with earlier repartnering (Class 3) were more likely to report multiple higher order unions. Mothers in the stable trajectories were also likely to having had multiple higher order unions (appendix Table C [supplementary material available online]). Thus, one could conclude that because lone mothers are not able to count on strong paternal support, some—rather than no—repartnering may still be a viable and adaptive coping strategy within contexts with only limited support for families in general and family-work reconciliation specifically (Dewilde & Uunk, 2008; Thévenon, 2011).

The finding that multiple repartnering of mothers in the stable and unstable trajectories were not related to overall or context-specific health detriments seems particularly striking because multiple repartnering has shown to harm mothers’ health and to reduce investments from the biological father (Bastin, 2012; Berger et al., 2012; Williams et al., 2008). Such results could indicate that mothers with denser repartnering patterns were not necessarily those in more economic need but rather those with more exposure to the marriage market because they may be employed or have more time and resources to invest in social leisure, which both foster health. Another plausible explanation for the positive effects of repeated repartnering on health could pertain to the potential expansion of lone mothers’ social network because of new partners (Keim, 2018). Additional meaningful ties such as step-grandparents, and close friends, may endure relationship breakups, and could, in turn, offset potential health risks.

This study has a number of limitations. First, welfare state classifications can only serve as a crude proxy of complex policy regimes that may overstate similarities between countries. For example, Germany and France were both grouped together despite considerable differences regarding the provision of child care. Further within-country differences, which can vary across cohorts, were not accounted for in the present classification either (e.g., between East and West Germany). Nevertheless, we ran sensitivity tests that used the country indicators rather than welfare context in our analyses, and the patterns of results remained largely unchanged.

Second, our small selection of countries representing European welfare contexts did not include Southern countries, whose culture tends to emphasize strong family ties rather than institutional trust, because we aimed to keep the welfare categories as homogeneous as possible. We therefore drew rather prototypical cases for each category (Thévenon, 2011), depending on data availability. Eastern, postsocialist countries were beyond the scope of our study as well because they would fall into a fourth category that was not part of the classification we mimicked (Burstrom et al., 2010). This context would be high on support for women’s attachment to the labor market and lower on gender equality because of women’s frequent “double shift” of gainful employment and household chores. Future studies will need to expand the scope of our first exploratory efforts by including more countries and welfare context in their analyses.

Third, we considered health to be an outcome of repartnering trajectories on a conceptual level, yet in our analyses, we incorporated it as a distal predictor of latent class membership. Since we cannot make causal claims about the link repartnering trajectories and health, we do acknowledge the possibility of self-selection into repartnering trajectories by health given that both modeling strategies seem feasible. Jointly shaped processes and reverse causality are pervasive in many studies on health (Adams, Hurd, McFadden, Merrill, & Ribeiro, 2003), and future studies will need to disentangle the directionality of these effects.

Fourth, survey data were collected at different time points across the countries. For example, the French GGS was conducted in 2005 compared with the Swedish GGS from 2012 to 2013. It is possible that timing difference in data collection may have contributed to overstating context variations in the likelihood of falling into a certain repartnering class or into repartnering-health-linkages. However, our regression models controlled for cohort membership as a proxy of historic time during which respondents negotiated repartnering behaviors to limit potential bias.

Despite such limitations, our paper contributes to the literature on repartnering and health by studying repartnering trajectories, which account for the timing, sequencing, and density of multiple transitions across the life course, holistically. We provide unique empirical evidence of variation in repartnering across welfare contexts and in repartnering-health-linkages across these contexts, showcasing the importance of contextualizing such associations in light of increasingly complex partnership histories. More comparative work on the dynamics between family policies and interwoven life course domains, such as union histories and health, are needed to understand the ways in which welfare operates on individual partnership choices and their consequences. Future research shall also consider accounting for possible mediators of such association, such as work histories or lone mothers’ networks.

Supplemental Material

Appendix_JFI_Lonemothers-2019_05_07 – Supplemental material for Lone Mothers’ Repartnering Trajectories and Health: Does the Welfare Context Matter?

Supplemental material, Appendix_JFI_Lonemothers-2019_05_07 for Lone Mothers’ Repartnering Trajectories and Health: Does the Welfare Context Matter? by Claudia Recksiedler and Laura Bernardi in Journal of Family Issues

Footnotes

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: Swiss National Centre of Competence in Research “LIVES—Overcoming vulnerability: Life course perspectives,” which is financed by the Swiss National Science Foundation (SNSF; 51NF40-160590). The authors are grateful to the SNSF for financial assistance.

Supplemental Material

Supplemental material for this article is available online.

ORCID iD

Claudia Recksiedler

References

Adams

Hurd

M. D.

McFadden

Merrill

Ribeiro

(2003). Healthy, wealthy, and wise? Tests for direct causal paths between health and socioeconomic status. Journal of Econometrics, 112, 3-56.

Amato

P. R.

(2010). Research on divorce: Continuing trends and new developments. Journal of Marriage and Family, 72, 650-666.

Avison

W. R.

Davies

(2005). Family structure, gender, and health in the context of the life course. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 60, 113-116.

Bastin

(2012). Dynamik alleinerziehender mutterschaft: Partnerschaftsverläufe in der frühen Elternschaftbiographie [Dynamics of lone motherhood: Partnership trajectories in early parental biographies]. Zeitschrift für Familienforschung, Sonderheft 2012, 201-228.

Beaujouan

(2012). Repartnering in France: The role of gender, age and past fertility. Advances in Life Course Research, 17, 69-80.

Berger

L. M.

Cancian

Meyer

D. R.

(2012). Maternal re-partnering and new-partner fertility: Associations with nonresident father investments in children. Children and Youth Services Review, 34, 426-436.

Bernardi

Mortelmans

Larenza

(2018). Changing lone parents, changing life courses. In Bernardi

Mortelmans

(Eds.), Lone parenthood in the life course (pp. 5-30). New York, NY: Springer.

Brady

Burroway

(2012). Targeting, universalism, and single-mother poverty: A multilevel analysis across 18 affluent democracies. Demography, 49, 719-746.

Burstrom

Whitehead

Clayton

Fritzell

Vannoni

Costa

(2010). Health inequalities between lone and couple mothers and policy under different welfare regimes: The example of Italy, Sweden and Britain. Social Science & Medicine, 70, 912-920.

10.

Bzostek

S. H.

McLanahan

S. S.

Carlson

M. J.

(2012). Mothers’ repartnering after a nonmarital birth. Social Forces, 90, 817-841.

11.

Cooper

C. E.

McLanahan

S. S.

Meadows

S. O.

Brooks-Gunn

(2009). Family structure transitions and maternal parenting stress. Journal of Marriage and Family, 71, 558-574.

12.

De Graaf

P. M.

Kalmijn

(2003). Alternative routes in the remarriage market: Competing-risk analyses of union formation after divorce. Social Forces, 81, 1459-1498.

13.

Dewilde

Uunk

(2008). Remarriage as a way to overcome the financial consequences of divorce: A test of the economic need hypothesis for European women. European Sociological Review, 24, 393-407.

14.

Dziak

Janzen

B. L.

Muhajarine

(2010). Inequalities in the psychological well-being of employed, single and partnered mothers: The role of psychosocial work quality and work-family conflict. International Journal for Equity in Health, 9, 1-8

15.

Eurostat. (2015). People in the EU: Statistics on household and family structures. Retrieved from http://ec.europa.eu/eurostat/statistics-explained/index.php?title=People_in_the_EU_-_statistics_on_household_and_family_structures

16.

Gałęzewska

Perelli-Harris

Berrington

(2017). Cross-national differences in women’s repartnering behaviour in Europe: The role of individual demographic characteristics. Demographic Research, 37, 189-228.

17.

Griffiths

(2017). No love on the dole: The influence of the UK means-tested welfare system on partnering and family structure. Journal of Social Policy, 46, 543-561.

18.

Guzzo

K. B.

Hemez

Anderson

Manning

W. D.

Brown

S. L.

(2019). Is variation in biological and residential ties to children linked to mothers’ parental stress and perceptions of coparenting? Journal of Family Issues, 40, 488-517.

19.

Idler

E. L.

Benyamini

(1997). Self-rated health and mortality: A review of twenty-seven community studies. Journal of Health and Social Behavior, 38, 21-37.

20.

Ivanova

Kalmijn

Uunk

(2013). The effect of children on men’s and women’s chances of re-partnering in a European context. European Journal of Population, 29, 417-444.

21.

Keim

(2018). Are lone mothers also lonely mothers? Social networks of unemployed lone mothers in Eastern Germany. In Bernardi

Mortelmans

(Eds.), Lone parenthood in the life course (pp. 139-169). New York, NY: Springer.

22.

Kühn

(2018). Changes in lone mothers’ health: A longitudinal analysis. In Bernardi

Mortelmans

(Eds.), Lone parenthood in the life course (pp. 357-373). New York, NY: Springer.

23.

Mayer

K. U.

(2004). Whose lives? How history, societies and institutions define and shape life courses. Research in Human Development, 1, 161-187.

24.

Meadows

S. O.

McLanahan

S. S.

Brooks-Gunn

(2008). Stability and change in family structure and maternal health trajectories. American Sociological Review, 73, 314-334.

25.

Mikolai

Lyons-Amos

(2017). Longitudinal methods for life course research: A comparison of sequence analysis, latent class growth models, and multi-state event history models for studying partnership transitions. Longitudinal and Life Course Studies, 8, 191-208.

26.

Millar

(2010). Lone mothers, poverty and paid work in the UK. In Chant

(Ed.), The international handbook of gender and poverty: Concepts, research and policy (pp. 147-152). Cheltenham, England: Edward Elgar.

27.

Osborne

Berger

L. M.

Magnuson

(2012). Family structure transitions and changes in maternal resources and well-being. Demography, 49, 23-47.

28.

Pasteels

Mortelmans

(2017). The socioeconomic determinants of repartnering after divorce and separation in Belgium. Demographic Research, 36, 1785-1812.

29.

Perelli-Harris

Kreyenfeld

Kubisch

(2011). Technical manual for the Harmonized Histories database (MPIDR Working Paper No. 011). Rostock, Germany: Max Planck Institute for Demographic Research.

30.

Perelli-Harris

Lyons-Amos

(2015). Changes in partnership patterns across the life course: An examination of 14 countries in Europe and the United States. Demographic Research, 33, 145-178.

31.

Pevalin

D. J.

Ermisch

(2004). Cohabiting unions, repartnering and mental health. Psychological Medicine, 34, 1553-1559.

32.

Pollmann-Schult

(2018). Single motherhood and life satisfaction in comparative perspective: Do institutional and cultural contexts explain the life satisfaction penalty for single mothers? Journal of Family Issues, 39, 2061-2084.

33.

Schnor

Pasteels

Van Bavel

(2017). Sole physical custody and mother’s repartnering after divorce. Journal of Marriage and Family, 79, 879-890.

34.

Sobotka

Toulemon

(2008). Changing family and partnership behaviour: Common trends and persistent diversity across Europe. Demographic Research, 19, 85-138.

35.

Struffolino

Bernardi

Voorpostel

(2016). Self-reported health among lone mothers in Switzerland: Do employment and education matter? Population, 71, 187-213.

36.

Thévenon

(2011). Family policies in OECD countries: A comparative analysis. Population and Development Review, 37, 57-87.

37.

United Nations. (2005). Generation and gender programme: Survey instruments. Geneva, Switzerland: United Nations.

38.

Vanassche

Corijn

Matthijs

Swicegood

(2015). Repartnering and childbearing after divorce: Differences according to parental status and custodial arrangements. Population Research and Policy Review, 34, 761-784.

39.

Wang

Amato

P. R.

(2000). Predictors of divorce adjustment: Stressors, resources, and definitions. Journal of Marriage and the Family, 62, 655-668.

40.

Wickrama

K. S.

Lorenz

F. O.

Conger

R. D.

Elder

G. H.

Jr. Todd Abraham

Fang

(2006). Changes in family financial circumstances and the physical health of married and recently divorced mothers. Social Science & Medicine, 63, 123-136.

41.

Williams

Sassler

Nicholson

L. M.

(2008). For better or for worse? The consequences of marriage and cohabitation for single mothers. Social Forces, 86, 1481-1511.

42.

Williams

Umberson

(2004). Marital status, marital transitions, and health: A gendered life course perspective. Journal of Health and Social Behavior, 45, 81-98.

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

0.15 MB