International Population Conference / IUSSP Cape Town, 2017 - - PDF document

International Population Conference / IUSSP – Cape Town, 2017 Fatherhood after union breakup in Uruguay: A transitory or life-long commitment?

Ignacio Pardo (Universidad de la República, Uruguay) ignacio.pardo@cienciassociales.edu.uy Wanda Cabella (Universidad de la República, Uruguay) wanda.cabella@cienciassociales.edu.uy Teresa Martín-García (Spanish National Research Council, Spain), teresa.martin@cchs.csic.es Teresa Castro-Martín (Spanish National Research Council, Spain), teresa.castro@csic.es

• 1. Introduction

One of the more important consequences of parental separation is the weakening effect it has

• n father-child relationships. High rates of parental separation and the extent to which non-

resident fathers lose contact with their child(ren) have led indeed to concerns about the negative effects of father absence on children’s well-being and life chances (Härkönen et al. 2017, McLanahan and Percheski 2008, Sigle-Rushton and McLanahan 2004). In fact, the steady decline of father-child coresidence among men lends support to claims about shrinking fatherhood and father pulling out of family life (Cabrera et al. 2000). The prevalence

• f

father-absent families and the mechanisms that underlie the parental separation penalty are widely researched topics in more developed countries since the 1980s. Initially, research focused on showing that, after parental marital breakup, a significant proportion of children and adolescents lost contact with their fathers and, when they did not, the frequency of visits was highly variable (Seltzer and Bianchi 1988, Amato and Gilbreth 1999, Manning and Smock 1999), suggesting some men’s limited attachment to the paternal role after separation (Furstenberg and Cherlin 1991, Seltzer 1991). In recent years, the focus has shifted from the frequency of father-child contacts and child support payments to a more comprehensive analysis of the quality and content of nonresident fathers' involvement with their children –including contact, shared activities, communication, emotional closeness, and financial contributions– after the reconfiguration of residential arrangements caused by the breakup (Choi, Palmer and Pyun 2014). In advanced societies, the nature and meaning of the “distant” father –traditionally associated with the patriarcal model– has gradually weaken, leading to a new concept of paternity that entails a more active and emotionally “involved fatherhood”. Men have gone from being mere economic providers and protectors to adopting a larger and more significant role as caregivers of their offspring. This new type of father involves a growing emphasis on nurturing, practical care and co-parenting above the traditional role of the breadwinner (Machin 2015).

• 2. Background

2.1. Fathers’ contact and involvement with nonresident children: international evidence International evidence reveals a growing demand from nonresident fathers to spend more time with their children and to engage more actively in parenting. Various studies document that, over the past three decades, there has been a steady rise in the share of fathers who maintain contact with their children after divorce (Amato 2009, Nielsen 2011), and a

substantial increase in the proportion of divorced parents who voluntarily opt for shared residential and legal custody arrangements. In the US context, fatherhood has traditionally been viewed as part of a ‘package deal’ (Furstenberg and Cherlin 1991, Townsend 2004) in which the fulfillment of fatherhood roles is largely contingent on the relationship between the father and the child’s mother (Tach, Mincy and Edin 2010). More recent data challenge the package deal hypothesis, as the proportion of nonresident parents who want to maintain close affective bonds with their children and engage in cooperative co-parenting increases (Goldberg 2015). A recent study of father-child contact across divorce cohorts in the Netherlands found that contact between divorced fathers and their children has increased considerably over the past four decades, and that part of this increase can be explained by a rise in fathers’ involvement in child-rearing during the marriage (Westphal et al. 2014). If father’s involvement with their children regardless of living arrangements is globally on the rise, this would be indeed a welcome trend. Based on the existing literature, non-resident fathers’ involvement in child-rearing and continued father-child interaction after parental separation has a positive impact on children's well-being and education (Amato et al. 2009, Ryan et al. 2008). These positive effects are largely contingent on the quantity and quality of the time shared, and on the level of fathers’ engagement in childrearing decisions (Bernardi et

• al. 2013, Amato and Fowler, 2002).

In Latin America, these issues have not received the same attention as in developed countries, partly because of the scarcity of adequate data. With some exceptions (e.g. Bucheli 2003, Budowski and Rosero-Bixby 2003, Ishida 2010, Cuesta and Meyer 2014), there is very little research on the level of contact and financial transfers from non-resident parents to children in Latin American countries. This is highly problematic, as children born outside marriage currently outnumber those born within marriage in the region (Castro-Martín et al. 2011). Though the majority of nonmarital births are born to cohabiting parents, there is an increasing number of children born to women with no coresident partner –the overall proportion is estimated to have grown from 7 percent in 1970 to 15 percent in 2000 in the region (Laplante et al. 2015). The historical prevalence of cohabitation and its more recent rise (Esteve and Lesthaeghe 2016) has also contributed to the increase in union disruption (Cerruti and Binstock 2009, Ullmann et al. 2014). As a result, the proportion of families headed by women has become even more widespread over the past decades, and an increasing proportion of children are growing up in father-absent households, which are disproportionally poor (García and de Oliveira 2011). Cuesta and Meyer’s study (2014) shows that, in Colombia, although less than

• ne-third of custodial-mother families receive child support, these transfers are helping these

families to move out of poverty or get closer to the poverty line. Despite its high relevance to the region, there is very limited knowledge on the magnitude and consequences of father absence from the household as experienced by children as well as on the factors that predispose non-residential fathers to fulfill their paternal role from a distance. In this paper we focus on Uruguay, which provides a useful case study, since the country has a high level of conjugal disruptions (Cabella 1998, 2009), and child support and father-child contact after parental separation is well below what may be desirable.

2.2. Fathers’ contact and involvement with nonresident children in Uruguay In Uruguay, the proportion of non-resident fathers who do not pay child support is high and has remained stable over time. In 2001, 58% of separated or divorced men did not make financial transfers to their children under 22 years, as shown by a survey on family life (Encuesta de Situaciones Familiares, ESF 20011). In 2008, in the second wave of this survey, the proportion remained virtually unchanged (56%) (ESF 2008). Another survey conducted in 2004 (Encuesta Género y Generaciones) found similar figures: according to mother’s reports, 56% of nonresident fathers did not pay child support2. According to a more recent survey, nearly half of all fathers (48%) who did not coreside with their children under four years did not provide child support payments (Bucheli and Vigorito 2015, ENDIS 2015). This survey also showed that low-income fathers were more reluctant to pay child support. Furthermore, the more conflictive the relationship with the ex-wife/partner, the less likely a father was to transfer economic resources to his child/ren after union breakup. Father’s self-employment was also found to be a significant predictor of noncompliance with child support payment. On the other hand, the type of union (marriage or cohabitation) was not found to influence the realization of economic transfers to children after separation (Bucheli and Vigorito 2015). In line with international evidence, child support and the frequency of father-child contact after separation are correlated in Uruguay (Bucheli 2003). Children’s loss of contact or infrequent interaction with their father after parental rupture is commonplace. According to ENDIS wave 1, in 2013 half of children under 4 years of age who did not coreside with their father had at least weekly face-to-face contact, but nearly one-third (32%) did not maintain any contact. Fathers of higher social strata were more prone to maintain frequent contact with their offspring after separation or divorce (Bucheli and Vigorito 2015). As in many other countries, the “tale of two fathers” (Livingstone and Parker, 2011) is becoming salient in

• Uruguay. On the one hand, there is a growing proportion of coresident and non-resident

fathers who are heavily involved with their children and, on the other hand, there remains a significant portion of fathers (nearly one-third) who completely lose contact with their children after union breakdown. Numerous studies focused on more developed countries have documented that child visitation, father-child contacts and overnight stays tend to be higher, after separation, for nonresident parents with a higher socio-economic status (SES) (Grätz 2017). In this line, Kalmijn (2015) reported a higher level of involvement of nonresident parent in high-SES families than in low SES-families in Germany, UK, Sweden and the Netherlands. Since children’s loss of contact with their father is more common in disadvantaged strata, existing social differences in the rate of family disruption and in the relationships of fathers and children who live apart tend to exacerbate gender and social inequalities. The consequences of increased

1 The ESF coverage was the metropolitan area, which represents more than 50% of the Uruguayan

population.

2 In this survey, men with child(ren) under 22 living apart from them were also asked whether they paid

child support and their reported level of compliance was significantly higher than that of women’s reports: around 70% declared that they made financial transfers to their children (Bucheli and Cabella 2009).

family instability on children's well-being and family life as well as differentials in nonresident father involvement by social strata are poorly studied in Uruguay. The issue is particularly relevant insofar as the increasing rates of union rupture, coupled with the poor implementation of policies aimed at supporting families after disruption, may intensify existing social inequalities.

• 3. Research question

Parental separation usually entails the weakening of father-child relationships. Nevertheless, the literature has documented that there is considerable heterogeneity in post-separation father-child ties and the causes of this heterogeneity are not consistent (Kalmijn 2015; Grätz 2017). In many cases, the parent who leaves the household after separation –most often the father–, reduces or even loses contact with his child(ren), but in other cases, he remains highly involved in the child’s life. In this study, we will contribute to this research area by providing a novel analysis of the relationship between pre-separation fathering and post-separation child- father contact and involvement in Uruguay. We focus on union dissolution and fathers’ involvement in raising children. Mainly, we study if and how the nonresident father’s contact and involvement with the child differ according to his pre-separation fathering practices. We draw from the life course perspective (Elder 1998) that suggests that being involved early “affords a man the opportunity to develop a relationship with his born child, which may strengthen his commitment and engagement over time, even in the case of separation” (Cabrera et al 2008). Our main hypothesis is that even though there is a tendency –which may encompass even more implicated fathers– towards declining contact over time, fathers who were more involved before separation will be more engaged after separation. This is a new insight into the driving forces of post-separation fathers’ involvement (Haux et al. 2015). According to identity theory, a man’s identity is made up of a set of meanings, roles and expectations (e.g. provider, caregiver) that go together with statuses (e.g. being a father). Because individuals will simultaneously have different statuses and associated identities, the concept of centrality –which reflects the hierarchical importance that is conferred by an individual to a given identity compared to others (Rane & McBride 2000)– is key. A central identity will be linked to a greater motivation to enact the roles that are connected with it. Given the above, men who have internalized a non-traditional identity as fathers, demonstrate commitment to the latter and for whom it becomes central should be more invested in the father trajectory and this may affect his later parenting behavior. In this case, the father identity is not embedded in the basis of the “package deal” formula since they do not view fatherhood as closely dependent on the conjugal relationship; hence, even though parents separate, these men do not separate themselves from the father role and feel obliged to involve themselves in their children’s lives. Fathers’ relationship and commitment to the child during the conjugal union increases their preferences for contact and involvement and therefore can have a long reach. We would therefore expect that fathers with stronger bonds with their children will be most likely to want to stay involved over the child’s life course. We acknowledge that many other individual and contextual factors, such as fathers’ financial resources, geographical distance, subsequent partnerships or additional children, will shape

patterns of non-resident fathers’ level of involvement in their children’s lives (Cooksey and Craig 1998, Manning et al. 2003). However, we expect that, after controllling for relevant parent and child characteristics, the more involved the father was in child raising before conjugal rupture, the higher his level of involvement after separation.

• 4. Data and methods

High costs make longitudinal studies uncommon in Latin America. This article benefits from an

• ngoing longitudinal study on early childhood, the Encuesta de Nutrición, Desarrollo Infantil y

Salud (National Survey of Early Childhood Health, Nutrition, and Development), ENDIS. ENDIS is being carried out by academic and government institutions and headed by the National Statistics Institute (Instituto Nacional de Estadística, INE) in Uruguay. So far, the study comprises two waves. The first one was collected in 2013 and achieved a sample of 3,077 children aged 0 to 3. In 2016, the second wave collected data from 2,455 children aged 3 to 6 (sample attrition = 20.2%). This data source meets two crucial requirements. In the first place, 76.7% of the sampled children were living with both parents in wave 1. In the second wave, 177 of them ceased to live with their fathers because of union breakup, generating the focal subsample for this study.3 Secondly, the ENDIS database allows us to measure fathers’ engagement both pre- and post-separation. Data in Wave 1 describe fathers’ and mothers’ engagement in several home tasks with detail (cooking, buying food, housekeeping, paying the bills, changing diapers, feeding the children, scolding children when they misbehave, taking them to the doctor, playing with them, and so on) and measures the amount of time fathers spend on child care. Wave 2 focuses intensely on nonresident father-child involvement: at least 15 survey questions refer to frequency and intensity of contact between father and child, parental attitudes and economic and non-economic support. The analysis uses resident mothers’ report about paternal involvement after separation because ENDIS does not follow the parent leaving the household. In order to measure father involvement before separation (in Wave 1), we distinguish between fathers who dedicated 0 hours per week to care work when they were partnered (no involvement) , fathers who dedicated 1 to 39 hours (low involvement), and men who dedicated 40 or more hours (high involvement). In order to evaluate father involvement after separation we examine three dimensions of father-child relationships: visiting patterns, financial contributions to child support, and participation in childrearing decisions in several domains (education, health, habits and discipline). We also include a number of control variables that have been shown to influence father involvement after separation (Westphal et al 2014). Father’s educational attainment indicates whether the father was enrolled in education at least for nine years. Higher educated fathers may be more willing to stay involved after separation due to more modern and egalitarian parenting styles. We also include the variable father’s employment, which captures primarily

3 We excluded cases where parents were not coresiding in wave 1 as they provide us with no

information on the father’s pre-separation engagement in child care.

the level of employment precariousness throughout the measure of formal vs. informal jobs, i.e. with and without social security coverage. It may be difficult for a father who has little or no money to afford economic support to their nonresident children, and non-compliance with child support may condition frequency of contact. Fathers’ resources are reported to be more influential in reducing the negative effect of parental separation on father involvement than mothers’ (Grätz 2017). In addition, we include controls for the sex and age of the child. Previous studies on the effects of parental separation on parental involvement by child gender and age showed a higher involvement of separated fathers in their sons’ lives than in their daughters’ (Bastaits et al. 2015, Kalmijn 2015) and for younger children (Skevik 2006). Data are analyzed using descriptive statistics, and logit models are used to examine the association of pre-separation involvement with the three dimensions of post-separation involvement, controlling for relevant father and child characteristics. We have kept the number of covariates in the models to a minimum because of our relatively small sample size.

• 5. Results

The data show that there is a wide variation in the short-term patterns of paternal involvement after separation in Uruguay. The frequency of father’s face-to-face contact with their child ranges from every day (16.2% of separated fathers) to no contact at all (13.2%), but more than two-thirds of separated fathers are in contact with their child at least once a week (72.3%) (see Annex). With regard to nonresident fathers’ fulfillment of their economic responsibilities, the picture is mixed: 53.6% contribute child support regularly, 9.9%

• ccasionally, and 26.5% never. There is also notable diversity regarding the extent to which

nonresident fathers’ participate on childrearing decisions: about two-thirds participate on decisions concerning education, health, and discipline, and a lower percentage (53%) on decisions related to sleeping or eating habits. It should be noted that in our analytical sample all children are young (under age 6) and union breakdown is relatively recent (in the past 3 years), which probably affects favorably the level of father’s involvement, and that social, economic, and decision-making ties will probably diminish over time. In order to examine whether there is certain continuity in father’s roles before and after separation, we have classified fathers according to the amount of hours spent in childcare work while married or cohabiting. Our measure of pre-separation involvement divides fathers into thirds: fathers with no involvement (31.14%), low involvement (33.53%) and high involvement (35.33%) are roughly equal in size (table 1). In further versions of this article we will analyze involvement in each domestic task. Table 1. Pre-separation father’s involvement measure (Wave 1) No involvement 31.14% Low involvement (1 to 39 hours per week) 33.53% High involvement (40+ hours per week) 35.33% N=170

When we relate this measure with indicators from Wave 2, results confirm a strong association between pre-separation involvement and two key measures of post-separation involvement: frequency of contact and financial support. First, fathers with no and low pre-separation involvement mostly see their children 1-2 times a week or less (66% and 62% respectively). Meanwhile, only one-third of high involved pre-separation fathers see their children 1-2 times a week, and two-thirds see them more often (Figure 1). Second, only 42% of fathers with no pre-separation involvement provide regular child support, whereas this percentage is higher among father with low pre-separation involvement (58%) and high pre-separation involvement (54%) (Figure 2). Figure 1. Pre-separation involvement and current involvement: frequency of contact

N=167

Figure 2. Pre-separation involvement and current involvement: father's economic support (%)

N=167

65,74 62,55 33,83 34,26 37,45 66,17 10 20 30 40 50 60 70 No involvement Low involvement High involvement % Current involvement: How often does he/she sees his/her father? (%) 1-2 times a week or less More than 1-2 times a week 58,09 41,81 45,98 41,91 58,19 54,02 10 20 30 40 50 60 70 No involvement Low involvement High involvement % Current involvement: father's economic support (%) Occasional or no support Regular support

It is thought-provoking to notice how parents with greater pre-separation involvement participate to a greater extent in decisions about their children's education, health, and daily habits, while decisions related to discipline show almost no relation to pre-separation involvement (Figure 3). It is plausible that gender roles are related to these peculiar co- parenting practices. Next, logistic regression models will show if variables such as the sex of the child are linked to involvement in decisions concerning discipline, to explore this issue in more depth. Figure 3. Pre-separation involvement and current involvement: decisions about child rearing N=170 Finally, binary logistic regression models might show whether some factors are related to post- separation involvement. Dependent variables will be the same post-separation involvement dimensions: frequency of contact, financial support and decisions on child rearing in education, health, daily habits and discipline (Table 2). We use several specifications for the models (See Annex). The most complete specification include father’s educational attainment, father's employment (informal / formal), age and sex

• f the child, and pre-separation involvement. Alternative versions of the model show that

results might vary as they are sensitive to the specification of the model and challenged by a small sample size4. Thus, we decided not to get into great detail in the analysis of each coefficient (in our case, each odds ratio) but rather to interpret the most robust tendencies as a whole.

4 In same cases the whole model was statistically non significant.

54,95 55,65 43,85 66,93 64,56 60,32 55,58 67,53 81,14 82,99 63,18 68,34 10 20 30 40 50 60 70 80 90 Education Health Habits Discipline % Current involvement: involvement in decisions about child rearing No involvement Low involvement High involvement

Table 2. Logistic regression model: results (odds ratio) MODEL SUMMARY Dependent Variable: Frequency

• f contact

Financial support Decisions: education Decisions: health Decisions: habits Decisions: discipline Father's education level 9 years or less 1 1 1 1 1 1 > 9 years 1.32155 2.391*** 2.313 2.587** 2.722905** 4.172*** Father's job (social security) Informal 1 1 1 1 1 1 Formal 0.845 2.247 1.942 2.483* 1.349 0.471 Sex (child) Female 1 1 1 1 1 1 Male 1.122 0.780 2.781*** 3.704*** 2.062* 1.268 Age (child) <4 years 1 1 1 1 1 1 4 years or more 0.969 0.823 0.689 0.505 1.388 1.097 Pre-separation involvement No involvement 1 1 1 1 1 1 Low involvement 1.201 1.427 1.202 0.875 1.447 0.762 High involvement 3.904*** 1.170 2.453 2.652 1.640 0.869 Prob>chi2 0.1091 0.039 0.005 0.000 0.058 0.140 Gl 6 6 6 6 6 6 Log pseudolikelihood

• 7196.400
• 7045.878
• 6016.560
• 5624.745
• 6914.887
• 6354.5812

Pseudo R2 0.0673 0.083 0.141 0.204 0.091 0.074 N 153 153 153 153 153 153 *p<0.10; **p<0.05; ***p<0.01 Some of the most relevant results are as follows:  The frequency of contacts seems only linked to pre-separation involvement. More specifically, there is a statistically significant difference between the extreme categories of that measure. Fathers who had high involvement with their children are almost four times more likely to see them more than 1-2 times a week than parents with no pre-separation involvement5. The findings were robust to the inclusion of socio-economic factors and child characteristics that may influence level of involvement following separation.  Economic support, however, is linked to social stratification variables. Both a higher educational level and being employed in the formal sector increase the odds of granting regular economic support.

5 Some independent variables are correlated, even when they are not highly collinear and can remain in

the model, i.e., high involved fathers tend to have higher educational level and formal jobs (see Annex)

 The chance that the father takes part in childrearing decisions varies depending on which dimension we are considering as output variable, but in any case, a remarkable pattern arises. The effect of pre-separation involvement is not statistically significant. Instead, social stratification and the sex of the child are associated with the probability

• f being involved in childrearing decisions. Involvement in decision-making is greater if

the child is male and if the father has a high educational level.

• 6. Conclusions

Some preliminary conclusions can be drawn from the results: a) There is a growing emphasis on more intimate fathering practices and participation in childcare in parallel to increasing rates of parental break-ups, with more fathers living apart from their children. Although there is a prolific literature on father-child relationships after separation in more developed countries, research on these issues is rare in Latin America, partly because of the scarcity of adequate data. This study benefits from an ongoing longitudinal study on early childhood in Uruguay to shed some light on the patterns and driving forces influencing father-child relationships after union break-up. In particular, we have examined whether more involved fathers before separation maintain greater levels of involvement also after separation. b) Our findings confirm that father’s pre-separation involvement affects his post- separation parenting behavior, although this relation varies by the dimension of post- separation involvement considered. c) Pre-separation high involvement particularly increases the odds of a more frequent post-separation contact, whereas financial support and participation in child rearing decisions might be affected by other variables. Our results are in line with previous studies, but provide more detailed evidence on the different dimensions of post- separation involvement. d) Men can afford a more relaxed approach to childcare because they often rely on the mothers, as if everyday childcare were something optional for men, but not for

• women. While the exercise of motherhood continues to be regarded as a social duty,

paternity is seen as an individual choice (Miller 2011). e) From a policy point of view, the findings suggest that efforts to support the involvement of fathers in parenting early in the child’s life, for instance through paternity leave policies, may have payoffs in terms of subsequent contact, economic support and involvement in childrearing decisions, even in the event of parental

• separation. This is particularly relevant in the current context of increasing conjugal

instability and substantial changes in family organization. Some limitations need to be acknowledged: a) Small N. b) The measures of post-separation involvement of fathers are based on mother’s reports.

References

Amato, P. y Fowler, F. (2002). “Parenting Practices, Child Adjustment, and Family Diversity”. Journal of Marriage and Family, 64(3): 703-716. Amato, P.R. and Gilbreth, J.G. (1999). Nonresident fathers and children's well-being: A meta-analysis. Journal

• f Marriage and Family 61(3): 557-573.

Amato, P.R., Meyers, C. and Emery, R. (2009). Changes in nonresident father-child contact from 1976 to 2002. Family Relations 58(1): 41–53. Bernardi, F., Härkönen, J. & Boertien, D. (2013). Effects of family forms dynamics on children's well-being and life chances: literature review. Families and Societies Working Papers Series, No. 4. Bucheli, M. (2003). “Transferencias y visitas entre hijos y padres no corresidentes”. En Nuevas Formas de Familia, UNICEF, Udelar, Montevideo, 2003. Bucheli, M. & W. Cabella (2005), “El incumplimiento en el pago de las pensiones alimenticias, el bienestar de los hogares y el contexto legal vigente en Uruguay”. En Asignaciones familiares, pensiones alimenticias y bienestar de la infancia en Uruguay, Montevideo: UNICEF. Bucheli, M. & Vigorito, A. (2015) “Después de la ruptura: efectos de la separación en los contactos entre padres e hijos y en el bienestar de las mujeres”, en Cambio familiar y bienestar de las mujeres y los niños en Montevideo y área metropolitana. Un estudio longitudinal, Montevideo: Udelar-UNICEF. Cancian, M., Meyer D. R., Brown P. R., y Cook S. T. (2014). “Who Gets Custody Now? Dramatic Changes in Children's Living Arrangements after Divorce”. Demography. 51(4), 1381-96. Budowski, M. and Bixby, L. R. (2003). Fatherless Costa Rica: Child acknowledgment and support among lone

• mothers. Journal of Comparative Family Studies 34(2): 229–254.

ENDIS 2013 (2015) Salud, nutrición y desarrollo en la primera infancia en Uruguay. Primeros resultados de la ENDIS, UCC, Mides, Montevideo. Available at: http://www.ine.gub.uy/c/document_library/get_file?uuid=25fc124a-d876-4040-b784- 3b25a850b1fa&groupId=10181 Cabella, W. (2009). “Dos décadas de transformaciones de la nupcialidad uruguaya. La convergencia hacia la segunda transición demográfica”. Estudios demográficos y urbanos de El Colegio de México, 24, 2, pp. 389 – 427. Cabella, W. (1998). La evolución del divorcio en Uruguay (1950-1995). Notas de Población 26 (67/68), 209- 246, CEPAL/CELADE Cabrera, N.J., Tamis-LeMonda, C.S., Bradley, R.H., Hofferth, S. and Lamb, M.E. (2000). Fatherhood in the Twenty-First Century. Child Development 71(1): 127-136. Cabrera, N., Fagan, J., and Farrie, D. (2008) Explaining the Long Reach of Fathers’ Prenatal Involvement on Later Paternal Engagement. Journal of Marriage and Family 70: 1094-1107. Castro-Martín, T., Cortina, C., Martín-García, T. and Pardo, I. (2011). Maternidad sin matrimonio en América Latina: Un análisis comparativo a partir de datos censales. Notas de Población 93: 37-76. Cerruti, M. and Binstock, G. (2009). Familias latinoamericanas en transformación: Desafíos y demandas para la acción pública. CEPAL, Serie Políticas Sociales No. 147. Choi, J.-K., Palmer, R. J. and Pyun, H.-S. (2014). Three measures of non-resident fathers' involvement, maternal parenting and child development in low-income single-mother families. Child & Family Social Work 19: 282–291. doi:10.1111/cfs.12000 Cooksey, E.C. and Craig P.H. (1998). Parenting from a distance: The effects of paternal characteristics on contact between nonresidential fathers and their children. Demography 35(2): 187-200. Cuesta, L. and Meyer, D.R. (2014). The role of child support in the economic wellbeing of custodial-mother families in less developed countries: The case of Colombia. International Journal of Law, Policy and the Family 28(1): 60-76. Esteve, A. and Lesthaeghe, R. (Eds.) (2016). Cohabitation and Marriage in the Americas: Geo-historical Legacies and New Trends. Springer. Furstenberg, F.F. and Cherlin, A. J. (1991). Divided Families: What Happens to Children When Parents Part. Cambridge, MA: Harvard University Press. García, B. and de Oliveira, O. (2011). Family changes and public policies in Latin America. Annual Review of Sociology 37: 593-611. doi: 10.1146/annurev-soc-081309-150205

Goldberg, J.S. (2015). Coparenting and nonresident fathers’ monetary contributions to their children. Journal

• f Marriage and Family 77 (3): 612–627. doi:10.1111/jomf.12191

Grätz, M. (2017) Does Separation Really Lead Fathers and Mothers to be Less Involved in their Children’s Lives? European Sociological Review 33 (4): 551-562. Härkönen, J., Bernardi, F. and Boertien, D. (2017) Family Dynamics and Child Outcomes: An Overview of Research and Open Questions, European Journal of Population, 33 (2):163-184. Haux, T., Platt, L., and Rosenberg, R. (2015) Parenting and post-separation contact: What are the links? CASE Working paper 189. LSE, London. Ishida, K. (2010). The role of ethnicity in father absence and children’s school enrollment in Guatemala. Population Research and Policy Review 29: 569–591. doi:10.1007/s11113-009-9160-7 Kalmijn, M. (2015). Father-child relationships after divorce in four European countries: Patterns and

• determinants. Comparative Population Studies 40(3): 251-276.

Laplante, B., Castro-Martín, T., Cortina, C. and Martín-García, T. (2015). Childbearing within marriage and consensual union in Latin America, 1980–2010. Population and Development Review 41(1): 85-108. Livingston, G. and Parker, K. (2011). A tale of two fathers. More are cctive, but more are absent. Pew Social & Demographic Trends, Pew Research Center. Mc Lanahan, S. and Percheski, C. (2008). Family structure and the reproduction of inequalities. Annual Review

• f Sociology 34: 257-276.

Machin, A. J. (2015). Mind the gap: The expectation and reality of involved fatherhood. Fathering 13(1): 36- 59. Manning, W. D. and P. J Smock, (1999). New families and nonresident father-child visitation. Social Forces 78(1): 87-116. Manning, W.D., Stewart, S.D. and Smock, P.J. (2003). The complexity of fathers’ parenting responsibilities and involvement with nonresident children. Journal of Family Issues 24 (5): 645-667. Miller, T. (2011) Falling back into Gender? Men’s Narratives and Practices around First-time Fatherhood. Sociology 45 (6): 1094-1109. Nielsen, L. (2011). “Shared Parenting After Divorce”. Journal of Divorce & Remarriage, 52:586–609. Rane, T.R. and McBride, B.A. (2000) Identity Theory as a Guide to Understanding Fathers’ Involvement With Their Children. Journal of Family Issues, 21 (3): 347-366. Ryan, R.M., Kalil, A. and Ziol-Guest, K.M. (2008). Longitudinal Patterns of Nonresident Fathers’ Involvement: The Role of Resources and Relations. Journal of Marriage and Family, 70(4): 962–977 Seltzer, J. A. and Bianchi, S.M. (1988). Children's contact with absent parents. Journal of Marriage and Family 50(3): 663-667. Seltzer, J. A. (1991). Relationships between fathers and children who live apart: The father’s role after

• separation. Journal of Marriage and Family 53(1): 79-101.

Sigle-Rushton, W. and McLanahan, S. (2004). Father absence and child well-being: A critical review. In D. P. Moynihan, T. A. Smeeding and L. Rainwater (Eds), The Future of the Family, New York: Russel Sage Foundation, pp. 116-59. Skevik, A. (2006) “Absent fathers” or “reorganized” families? Variations in father-child contact after parental break-up in Norway. European Sociological Review 54 (1): 114-132. Tach, L., Mincy, R. and Edin, K. (2010). Parenting as a “package deal”: Relationships, fertility, and nonresident father involvement among unmarried parents. Demography 47(1): 181–204. doi:10.1353/dem.0.0096 Townsend, N.W. (2004). The Package Deal: Marriage, Work, and Fatherhood in Men’s Lives. Philadelphia PA: Temple University Press. Ullmann, H., Maldonado, C. y Rico, M.N. (2014). La evolución de las estructuras familiares en América Latina, 1990-2010: los retos de la pobreza, la vulnerabilidad y el cuidado. Serie Políticas Sociales, N°193 (LC/L.3819). Santiago de Chile: Comisión Económica para América Latina y el Caribe (CEPAL). Westphal, S. K., Poortman, A.-R. and van der Lippe, T. (2014). Non-resident father–child contact across divorce cohorts: The role of father involvement during marriage. European Sociological Review 30(4): 444–456.

ANNEX WAVE 1: Separated fathers Hours per week of care % None 32.35 1 to 39 35.94 40+ 34.71 N=170 Does the father take care of the children at home? Yes 70.11 No 29.89 N=174 When the child gets sick or he/she is off school, can he/she turn to his/her father? Yes 64.94 No 32.76 DK 2.3 Valid N=170 Pre-separation father’s involvement measure II No involvement 31.14 Low involvement (1 to 39 hours per week) 33.53 High involvement (40+ hours per week) 35.33 WAVE 2: Separated fathers Current involvement: How often does he/she sees his/her father? % Every day 16.24 5 - 6 days per week 7.38 3 - 4 days per week 20 1 - 2 days per week 28.72 Once per 15 days 8.53 Once a month 2.58 Some times per year 3.23 No contact 13.32 N=177 100

Current involvement: father's economic support % Yes, regularly 53.57 Yes, ocassionally 9.88 No 36.55 N=177 100 Decisions Education % Mainly mother 32.35 Mother and father (50/50) 65.01 Mainly father 1.22 Other 0.47 DK/DR 0.96 N=177 100 Health % Mainly mother 34.28 Mother and father (50/50) 65.22 Mainly father 0.5 N=177 100 Habits (eating, sleeping, etc) % Mainly mother 46.68 Mother and father (50/50) 51.27 Mainly father 1.72 Other 0.33 N=177 100 Discipline % Mainly mother 29.9 Mother and father (50/50) 56.28 Mainly father 13.82 N=177 100

Pre-separation involvement and current involvement in decisions about child rearing Father’s pre-separation involvement and father’s education level Pre-separation involvement Father's education No involvement Low involvement High involvement Total 9 years or less 60.2 40.2 47.67 48.01 > 9 years 39.8 59.8 52.33 51.99 Total 100 100 100 100 Father’s pre-separation involvement and father’s job (social security) Pre-separation involvement Father's job (social security) No involvement Low involvement High involvement Total Informal 48.51 34.54 29.65 36.42 Formal 51.49 65.46 70.35 63.58 Total 100 100 100 100 10 20 30 40 50 None On one dimension On two dimensions On three dimension On every dimension asked % Current involvement: involvement in decisions about child rearing No involvement Low involvement High involvement