Claiming Fatherhood: Race and the Dynamics of Paternal Involvement among Unmarried Men

Abstract

In 1965, Daniel Patrick Moynihan argued that the black family was nearing “complete breakdown” due to high rates of out-of-wedlock childbearing. In subsequent decades, nonmarital childbearing rose dramatically for all racial groups and unwed fathers were often portrayed as being absent from their children’s lives. The authors examine contemporary nonmarital father involvement using quantitative evidence from the Fragile Families and Child Wellbeing Study and qualitative evidence from in-depth interviews with 150 unmarried fathers. The authors find that father involvement drops sharply after parents’ relationships end, especially when they enter subsequent relationships and have children with new partners. These declines are less dramatic for African American fathers, suggesting that fathers’ roles outside of conjugal relationships may be more strongly institutionalized in the black community. The challenges Moynihan described among black families some forty years ago now extend to a significant minority of all American children.

In 1965 Daniel Patrick Moynihan, then assistant secretary of labor for President Lyndon Johnson, penned the now-infamous report The Negro Family: A Case for National Action. In this document, Moynihan claimed that owing to the sharp increase in out-of-wedlock childbearing—a condition affecting only a small fraction of white children but one in five African Americans at the time—the black family was nearing what he called “complete breakdown,” particularly in America’s inner cities. Over the next two decades, the rate of unwed childbearing tripled for the nation as a whole. Its prevalence among whites is now as high as it was for African Americans when Moynihan released his report.

In the wake of this dramatic increase in so-called fatherless families, public outrage grew and policy makers responded. Liberals wanted to help supplement the incomes of single mothers, who were disproportionately poor, yet conservatives believed this would only reward their behavior and lead to more female-headed families. Meanwhile, taxpayers increasingly demanded to know why their hard-earned dollars were going to support what many saw as an unfortunate lifestyle choice, not unavoidable hardship.

Scholars responded by devoting a huge amount of attention to studying single mothers and children, detailing the struggles of the parents and documenting the deleterious effects on the children. These studies offered the American public a wealth of knowledge about the lives of the mothers and their progeny, yet told us next to nothing about the fathers of these children. Part of the problem was that surveys provided very little systematic information from which to draw any kind of representative picture. Unwed fathers’ often tenuous connections to households made them hard to find, and many refused to admit to survey researchers that they had fathered children. Thus, vast numbers of fathers have been invisible to even the largest, most carefully conducted studies (Hofferth et al. 2002).

From this mix of scholarly ignorance and public indignation, a compelling yet distorted image of unwed inner-city fathers emerged and captured the American public’s imagination: the “hit-and-run” father. According to this portrayal, men who father children outside of a marital bond are interested only in sex, not fatherhood. When their female conquests come up pregnant, they quickly flee the scene, leaving the expectant mother holding the diaper bag. “[Unwed fathers] never signed on to anything,” wrote marriage movement founder David Blankenhorn in 1995 in Fatherless America. “They never agreed to abide by any fatherhood code. They do not have—they have never had—any explicit obligation to either their children or to the mothers of their children” (pp. 134–35).

In the same spirit, well-known conservative William Bennett, in his 2001 book The Broken Hearth, raged, “It is unmarried fathers who are missing in record numbers, who impregnate women and selfishly flee. … Abandoning alike those who they have taken as sexual partners, and whose lives they have created, they … traduce generations yet to come, and disgrace their very manhood” (pp. 93–94). In 2004, an infuriated Bill Cosby publicly indicted unwed fathers for merely “inserting the sperm cell” while blithely eschewing the responsibilities of fatherhood. “No longer is a boy considered an embarrassment if he tries to run away from being the father of the unmarried child,” Cosby declared.¹ Two days before Father’s Day in 2007, democratic presidential candidate Barack Obama admonished the congregants of Mt. Moriah Baptist Church in Spartanburg, South Carolina: “There are a lot of men out there who need to stop acting like boys, who need to realize that responsibility does not end at conception, who need to know that what makes you a man is not the ability to have a child but the courage to raise a child.”²

This image of unwed fatherhood as a hit-and-run encounter plays a dominant role in the public discourse about poverty, family structure, and race. However, a growing body of evidence from the social sciences suggests that unwed fathers who “hit and run” are much rarer than the public assumes. We evaluate the accuracy of this view using evidence from a recent nationally representative longitudinal survey of nonmarital births in large cities—the Fragile Families and Child Wellbeing Study. We find that even by a child’s fifth birthday, rates of father involvement are quite high, especially when compared to public perceptions.

These relatively high levels of involvement mask considerable variation in involvement based on the parents’ relationship status, however. After relationships between parents end, father involvement drops sharply. We also find that declines are particularly dramatic when the father and mother enter subsequent relationships and have children with new partners. These declines are less dramatic for African American fathers, suggesting that the role of the father outside the context of a conjugal relationship may be more strongly institutionalized in the black community.

We then draw on qualitative interviews with 150 low-income white, black, and Hispanic fathers from high-poverty neighborhoods in the Philadelphia metropolitan area, all of whom have had a child outside of marriage, to understand the complex social processes that may underlie these statistical associations. We find that contrary to public perceptions, for the large majority of unwed fathers, the father role has high salience, and most strive to be highly engaged with their progeny. Ironically, though we find that the majority of fathers strongly reject traditional notions that the father–child bond should be conditioned upon the state of the relationship between the father and mother, they nonetheless typically enact it. Due to a variety of challenges, especially those stemming from transitions into new partner and parenting roles, children born to unmarried parents only rarely have a stably involved father throughout childhood and adolescence.

Theory and Research on Father Involvement

In the American context, fatherhood has traditionally been viewed as part of a “package deal” (Furstenberg and Cherlin 1991; Townsend 2004). Fatherhood is a relationship that is not independent of, but largely flows through and is contingent upon, the relationship between the father and the child’s mother. This explanation is often used to account for the surprisingly low levels of father–child contact and child support payment following a divorce (Furstenberg and Cherlin 1991). To the extent that notions of “the package deal” are still strongly institutionalized within American society, men attempting to father outside of the context of a marriage, a coresidential union, or a romantic relationship will have more difficulty staying involved with their children.

Beyond the additional transaction costs fathers must pay to retain contact with children after a coresidential or romantic partnership ends, the package deal hypothesis holds that there are normative barriers to investment as well. As Cherlin has repeatedly reminded family scholars (1978, 2004), much of family behavior is “automatic”—it relies on ready-made solutions to daily problems based on widely shared normative expectations. These normative expectations not only guide and constrain the behavior of the father but also of the mother, who, as the custodial parent, must cooperate for father–child contact to occur. Following Furstenberg (1995), we extend the application of the package deal hypothesis, arguing that it not only predicts declines in involvement after breakup, but also that subsequent transitions into new partner and parenting roles may pose significant added barriers to involvement. As the father and mother of a child enter into new “family-like” relationships, they may feel considerable pressure to enact the cultural norm of the package deal with the new family without the interference of prior partners or children from past partnerships.

Not all subgroups in American society are equally influenced by this overarching cultural ideal, however. Ronald Mincy and Hillard Pouncy (2007) draw on data from the Caribbean (Brown, Anderson, and Chevannes 1993; Clarke 1957; Senior 1991) and evidence from surveys and focus group interviews with unwed African American fathers and mothers in Louisiana to argue that the role of the father among men who are no longer in a conjugal union with the mother—the “baby father” role—may have achieved a higher degree of institutionalization among African Americans than among other U.S. racial and ethnic groups.

In the Caribbean, for example, where rates of nonmarital fertility and multiple-partnered fertility have been high for decades, the institutionalization of the roles and relationships between parents and children is indicated through the existence of a system of kinship terms that are used to describe the roles and relationships between parents and children in these situations (Senior 1991; Brown et al. 1997; Brown, Anderson, and Chevannes 1993; Roopnarine, et al. 1995, 2004). Mincy and Pouncy’s (2007) survey and focus group data drawn from African American residents of Louisiana likewise suggest that a similar process of institutionalization may be occurring in the United States (see also Anderson 1996; Hamer 1998, 2001; Waller 2002). If ongoing father involvement outside of a romantic relationship with the mother is more normative among African Americans than other groups, father involvement should decline less for black fathers than for other fathers after the parents’ romantic involvement ends, as subsequent relationships are taken on, and after subsequent childbirth with new partners.

Despite the dominant image of unwed fatherhood as a hit-and-run encounter, the wealth of the evidence, whether qualitative or quantitative, suggests otherwise. Over the past three decades, a number of in-depth qualitative studies have focused on behaviors and attitudes among young unmarried fathers, and with only a few exceptions (Anderson 1993; Bell Kaplan 1997), they have found that the salience of the father role and engagement in fathering activities is high (for example, see Anderson 1996; Hamer 2001; Sullivan 1993; Waller 2002; Young 2003). Similarly, two panel studies—the National Longitudinal Survey of Youth, which began to follow a sample of youth aged fourteen to nineteen in 1979; and the National Survey of Families and Households, a national probability sample of all U.S. households that was launched in 1981—offered much the same portrait using nationally representative samples (Lerman 1993; Seltzer 1991; Mott 1990).

Specifically, by the mid-1980s, we learned that, consistent with qualitative studies that focused mainly on men who were new fathers, unmarried fathers with very young children were usually quite involved. These statistical portraits demonstrated, however, that involvement declined quite dramatically as the children got older (Lerman 1993; Seltzer 2000). Additional surveys conducted in the 1990s showed consistent evidence of a downward trend in involvement as the children aged, although the rates differed considerably across the studies (Argys et al. 2007). By the time nonmarital children reach adolescence, their chances of having a regularly involved father appear to be very low (Argys and Peters 2001).

This decline is somewhat puzzling, given recent evidence from the Fragile Families and Child Wellbeing survey, launched in twenty large U.S. cities in between 1998 and 2000, which showed that the vast majority of nonmarital fathers were present at the time of the birth and said they wished to remain involved in the child’s life. This survey of a nationally representative cohort of nonmarital births in large cities found that nine in ten children born to an unwed father were not subject to a hit-and-run father’s indifference, but welcomed into the arms of someone who said he was committed to “being there” for his child. When the surveyors interviewed the mothers of these children just after the birth, eight in ten said the father had been supportive during pregnancy (McLanahan et al. 2003). Furthermore, nearly all—99.8 percent—of the fathers interviewed said they intended to stay involved (Center for Research on Child Wellbeing 2000).

Studies that consider all nonresidential children find lower rates of involvement for nonmarital than marital children. Father involvement also varies by race and ethnicity, with rates for Africans American being higher and those for Hispanics lower than the average American father (Danziger and Radin 1990; Huang 2006; King 1994; King, Harris, and Heard 2004; Mott 1990; Seltzer 1991; but see Seltzer and Bianchi 1988). Father involvement is also predicted by parental education (Argys and Peters 2001; Huang 2006; King, Harris, and Heard 2004), fathers’ age (Lerman and Sorensen 2000) and earnings (Lerman and Sorensen 2000; Seltzer 1991), work status (Danziger and Radin 1990); child gender (King, Harris, and Heard 2004; Manning and Smock 1999; but see Cooksey and Craig 1998), the presence of additional children, father’s current marital status, the number of years since the father left the home (Argys and Peters 2001), payment of child support (Seltzer 1991), and the quality of the coparenting relationship (Sobolewski and King 2005; but see Amato and Rezac 1994). Waller and Swisher (2006) focus solely on unmarried fathers and find that a wide array of risk behaviors, such as physical abuse, drug and alcohol use, and incarceration are associated with lower odds of father–child contact.

The literature has shown that for fathers with noncustodial children, living with a new biological child is related to lower support payments (Manning and Smock 2000; Manning, Stewart, and Smock 2003), but findings on involvement are inconsistent (Manning, Stewart, and Smock 2003; Juby et al. 2007). The effect of subsequent partnerships is also examined by Stephens (1996), who looks at nonresident fathers’ behavior following divorce and finds that a father’s remarriage reduces both payment and visitation. Stewart (1999), who considers nonresident fathers’ transitions into both cohabiting and marital relationships, finds that the dampening effect of a father’s new partnerships is stronger for cohabitation than for marriage. Other analyses find that when mothers remarry and children acquire a new stepfather, they see less of their biological fathers (Juby et al. 2007; Seltzer and Bianchi 1988), although the effect of mother’s remarriage may be weak (Stephens 1996).

These studies, though closely related to ours, have several limitations. First, the data sources they use all suffer from significant underrepresentation of nonmarital fathers, a problem that is minimized in the data set we use. Second, none of the studies considers all possible subsequent partner and parental transitions, including those of fathers as well as mothers. Third, no study focuses specifically on fathers of nonmarital children, and many studies do not even include them at all.

The lack of attention to the effect of subsequent relationship transitions among unmarried fathers is somewhat surprising, as levels of multipartner fertility are dramatically higher among them. Recent data suggest that about 40 percent of all fathers of nonmarital children born between 1998 and 2000 already had at least one child by another partner at the time of this child’s birth (Carlson and Furstenberg 2006). Given that the average father surveyed was only in his midtwenties at the time, the proportion of unmarried fathers who will eventually split up, repartner, or have subsequent children with a new partner is likely to be very high.

For this large and growing subset of parents, we expect that transitions into subsequent relationships, and subsequent fertility within those relationships, are the key mechanisms though which father involvement declines over time. First, it is likely that as fathers move on to subsequent partners and parental roles, the demands inherent in working to enact the cultural model of the package deal in these new relationships could supersede obligations to children from prior relationships. Second, it is equally likely that as mothers enter into subsequent partnerships and parent roles, they might respond similarly and seek to exclude the biological father in favor of the new father figure in the home. We also expect that white and Hispanic fathers are somewhat more vulnerable to the threat of new relationships than African American fathers are, due to the greater degree of institutionalization of the “baby-father” role in black communities.

Quantitative Data

In the analyses that follow, we use four waves of the Fragile Families and Child Wellbeing Study to examine levels and changes in father involvement by race among fathers who had a nonmarital birth, focusing on how subsequent partnerships and new parental roles of both mothers and fathers affect this involvement. The Fragile Families and Child Wellbeing Study followed a cohort of nearly five thousand children born in twenty U.S. cities between 1998 and 2000. The study interviewed mothers and fathers at the time of the child’s birth and again after one year, three years, and five years. The survey oversampled nonmarital births and, when weighted, is representative of births in all U.S. cities with populations larger than two hundred thousand. Both the mother and father were interviewed at each follow-up, regardless of their relationship status. These data are ideal for the study of father involvement not only because of the large sample of unmarried and nonresidential fathers, but also because they contain detailed longitudinal economic, attitudinal, and behavioral information collected independently from both the mother and the father.

At each survey wave, our analyses are based on the subsample of children in the Fragile Families Study who were born outside of marriage, who lived with their biological mother, whose mother responded to the survey, and for whom we have nonmissing data on parents’ relationship status and father involvement from mothers’ surveys. This results in sample sizes of 3,243 at the one-year survey, 3,123 at the three-year survey, and 3,050 at the five-year wave of the study. Nonresponse and attrition were higher for unmarried mothers and fathers than for married parents. At baseline, 87 percent of eligible mothers and 75 percent of the fathers agreed to participate in the survey. In subsequent surveys, response rates for unmarried mothers were 90 percent at wave 2, 88 percent at wave 3, and 87 percent at wave 4. The mothers who dropped out of the study tended to be white or Latino, were less likely to be married to the father when the child was born, and had lower average socioeconomic status (Cooper et al. 2007). Fathers had higher attrition rates, at 70 percent, 68 percent, and 66 percent for waves 2, 3, and 4, respectively. Fathers who dropped out of the study were less likely to be involved with their children and were less likely to be residing with the mother of the focal child.

Because fathers’ attrition is nonrandom and correlated with our outcome of interest, we use mothers’ reports of father involvement. For fathers’ independent variables, we use fathers’ reports if they are available, mothers’ reports if fathers’ reports are unavailable, and single imputation if neither mothers’ nor fathers’ reports are available.³ Item nonresponse for our analysis variables was generally low, in most cases less than 5 percent. The items for which nonresponse was higher include whether the father repartnered (9 percent), whether the father was employed (8 percent), and whether the father had subsequent children (10 percent).

The main dependent variable in our study is father involvement. We use mothers’ reports of fathers’ involvement because fathers have higher rates of attrition, which are systematically related to their level of involvement. Fathers are coded as seeing child several times weekly if they saw their child for at least eight out of the past thirty days.⁴ All fathers who were living with the mother (either married or cohabiting) were coded as seeing child daily. Fathers were coded as seeing child monthly if they saw their child at least once in the past thirty days, as seeing the child yearly if they had seen the child since the previous interview, and as having no contact if they had not seen the child since the previous interview.

A father’s race and ethnicity was determined using his own report if available and the mother’s report if his own was not available. Fathers were classified into four mutually exclusive categories: non-Hispanic white, non-Hispanic black, non-Hispanic other, and Hispanic. Based on mothers’ reports at each wave, their relationship status was categorized as married, cohabiting, romantically involved, or no relationship with the child’s father. Couples were defined as cohabiting if they were romantically involved and living together all or most of the time, as living together if mothers reported they lived together all or most of the time, and as not living together otherwise.

Several demographic and background characteristics of fathers are also included as control variables in some analyses. Father’s earnings were measured by a dummy variable for whether the father earned more than $15,000 in the past year, and father’s age was measured in years at the time the child was born. Father’s education was coded as a series of dummy variables indicating less than high school, high school or GED, some college, and college plus. Fathers were coded as father employed if they reported doing any regular work for pay during the week prior to the interview. For each of these variables, we relied on the fathers’ reports if they were available and the mothers’ reports if they were not available from the fathers.

Several other relationship characteristics were also used in the following analyses. We measured the time since parents stopped coresiding as an ordinal index of the number of survey waves the parents had not lived together. For example, in the fourth survey wave, parents were coded as 0 if they still lived together, 1 if they were living together at the third wave but were not living together at the fourth, 2 if they were living together at the second wave but not in the third or fourth wave, and 3 if they were living together at the first wave but not any of the subsequent ones. Parents who never lived together during the study period were coded as 4. This indexing was repeated for each of the survey waves. Time since parents stopped romantic involvement was indexed in the same way. We also measured whether the father had a new partner, the mother had a new partner, the father had subsequent children with different partner, and the mother had subsequent children with different partner. Again, fathers’ relationship and fertility measures were taken from their own reports if available and from mothers’ reports if they were unavailable.

Table 1 shows descriptive statistics for black, white, and Hispanic fathers who had a nonmarital birth and who were not living with their child by the five-year follow-up, weighted by national sampling weights. The majority of fathers who fell into the nonresidential category were black. White fathers were more likely to have a college degree, earn more than $15,000, and be employed. On average, black fathers had spent the longest time not living with their child by the child’s fifth birthday, but there were few racial differences in the amount of time since parents were romantically involved. This discrepancy reflects the fact that black fathers were more likely to be romantically involved with mothers when they were not living together than white or Hispanic fathers were. By the time the child was five years old, repartnering and subsequent fertility were common among both mothers and fathers of all races. About half of mothers and fathers had a new romantic partner, and more than a quarter had a subsequent child with a new partner. Men were more likely than women to have children with new partners.

TABLE 1.

DESCRIPTIVE STATISTICS FOR FATHERS WHO HAD A NONMARITAL BIRTH AND ARE NONRESIDENT BY WAVE 4

	Overall	White	Black	Hispanic
Baseline characteristics
Age (in years)	26.2	24.6	25.9	27.1
Education
Less than high school	45.4	41.4	43.5	49.7
High school or GED	36.7	41.2	37.9	33.5
Some college	15.7	12.2	17.1	15.7
College or more	2.2	5.2	1.5	1.1
Child is male	53.8	56.7	55.5	51.6
Wave 4 characteristics
Earned less than $15,000	53.9	34.9	61.1	52.1
Employed	65.5	79.5	61.2	69.4
Survey waves since stopped coresiding	3.2	2.9	3.3	2.9
Survey waves since relationship ended	2.6	2.6	2.6	2.5
Mom has new partner	51.3	52.3	51.4	49.4
Dad has new partner	50.4	51.6	49.6	47.5
Mom has new children by different partner	24.8	17.9	26.1	24.6
Dad has new children by different partner	37.6	44.9	36.4	34.5
N	2,019	183	1,301	455

NOTE: All figures weighted by national sampling weights. The overall descriptive statistics include the three major racial and ethnic groups and a residual “other race” group. All values are percentages, unless otherwise noted.

Qualitative Data

Between 1995 and 2001, Timothy Nelson, Kathryn Edin, and a team of graduate students conducted in-depth repeated intensive interviews with 165 low-income fathers who had at least one nonmarital child younger than eighteen. These fathers were drawn from three high-poverty communities within the Philadelphia metropolitan area: East Camden, Kensington, and North Philadelphia. Roughly equal numbers of African American, white, and Puerto Rican fathers (Philadelphia’s dominant Hispanic group) were interviewed. The sample was also stratified by age: roughly equal portions within each racial and ethnic group were thirty or younger, while the remainder were older than thirty. The inclusion of fathers of multiple racial and ethnic groups and both younger and older fathers offers a considerable advantage over most prior qualitative work, which has typically focused only on black fathers and/or on very young fathers.

Unmarried fathers in these neighborhoods were often not stably attached to households. Rather than employ a random sample in each of the three neighborhoods, researchers contacted a wide range of third parties who could act as trusted intermediaries. These included leaders of a variety of local nonprofit and government sector organizations, including churches, settlement houses, and social service agencies. The intermediaries reported that most men were not attached to their organizations, however, so researchers also recruited subjects through street contacts and via referrals from fathers who had already participated in the study. No more than five referrals came from any given source.

Most fathers were interviewed at least twice at a place and a time of their choosing. Interviews were semistructured to ensure consistency across cases, but interviewers were encouraged to change the order and wording of questions to fit with the flow of the conversation. The first portion of the interview was focused on the father’s life history, focusing specifically on transitions into and out of schooling, employment, romantic relationships, and fatherhood roles. In the second part of the interview, fathering views and behaviors as well as father’s financial situations were captured. Interviews ranged from one and a half to seven hours and were transcribed verbatim and electronically coded by topic. The results presented here rely on analyses of full transcripts.

Survey Findings on Father Involvement

Table 2 details the proportions of nonmarital children who have contact with their biological father at one, three, and five years. Overall, involvement rates among unmarried fathers begin high but decline for all racial groups throughout the first five years of a child’s life. The high initial rates of involvement reflect relatively high rates of coresidence among parents of nonmarital children at first, mostly in the context of cohabiting unions, although some eventually marry. But the table also reveals large declines in coresidence over time. More than half of nonmarital children reside with their father around the time of their first birthday, but this figure declines to only 35 percent by the time of their fifth birthday.

TABLE 2.

FATHER INVOLVEMENT AFTER A NONMARITAL BIRTH, BY RACE AND ETHNICITY

	Year 1				Year 3				Year 5
	Overall	White	Black	Hispanic	Overall	White	Black	Hispanic	Overall	White	Black	Hispanic
Resident fathers	52.1	58.8	46.9	57.9	43.9	48.1	39.4	49.2	35.5	39.7	29.3	41.8
Nonresident fathers	47.9	41.2	53.1	42.1	56.1	51.9	60.6	50.8	64.5	60.3	70.7	58.2
All nonresident fathers
Saw child since previous survey	87.2	89.3	92.8	79.9	73.3	72.7	78.3	67.2	66	69.6	73.1	56.8
Saw child in past month	63.2	58.5	67.4	59.7	49.0	53.1	51.6	42.9	45.9	56.3	52.4	36.6
Saw child at least eight days in past month	36.4	28.7	35.6	39.4	29.8	32.2	30.2	27.7	25.8	35.1	28.1	20.6
Mean number of days father saw child in past month	8.9	6.8	9.3	8.5	6.6	5.9	7.2	6.1	5.8	7.1	6.7	4.3
Fathers who saw child in past month
Saw child at least eight days in past month	57.6	49.1	52.8	65.8	60.7	65.9	59	64.6	56.3	62.4	53.6	56.3
Mean number of days father saw child	13.7	11.8	13.8	14.2	13.4	11.3	13.9	14.1	12.6	12.5	12.8	11.7

NOTE: Weighted by national sampling weights for each survey year. All values are percentages unless otherwise indicated.

The second panel of the table considers only nonresident fathers and shows that by the child’s first birthday, 36 percent see their fathers several times each week, and nearly six in ten saw their father in the past month. By the child’s third birthday, these figures have fallen to 30 percent with several visits each week, with rates of monthly involvement at about 50 percent. By the time children reach age five, only about a quarter (26 percent) still see their father several times a week, though 45 percent still have regular contact. At each survey wave, African American children are less likely to live with their fathers than are Hispanics and whites. Thus, fewer African American children experience “automatic” fatherhood—a situation where there are few, if any, impediments to frequent father-child contact.

Although coresident and romantically involved fathers’ involvement is “automatic”—an almost inevitable result of maintaining a coresidential romantic relationship with the mother—men outside of a coresidential or romantic bond with the mother must negotiate a fathering relationship, making each hour of father–child contact more “expensive” in terms of transaction costs (see McLanahan 2008 [this volume]). This is particularly true if the father’s children live in multiple households. Because nonmarital African American children are significantly less likely to live in a coresidential union with both of their biological parents than their Hispanic or white counterparts, black fathers must work harder to maintain similar levels of intensive and regular contact, which they do. In fact, among nonresidential fathers, African Americans have higher rates of at least weekly, at least monthly, and at least yearly contact with their child until age three, after which the racial differences diminish. In sum, African American fathers accomplish their high overall rates of involvement mainly through negotiated visitation, while white and Hispanic fathers are more likely to achieve high rates of involvement through coresidence.

Thus far, we have not taken into consideration the amount of elapsed time since the father lived with the mother or was romantically involved with her. whether the “breakup” is defined as the point at which the father stopped core-siding with the mother or when romantic involvement ended, Table 3 shows that father involvement drops very dramatically in the aftermath of breakup, as the package deal hypothesis would predict. For fathers of all racial and ethnic groups, involvement in the first year after breakup is relatively high, but these rates decline rapidly as more time passes.

TABLE 3.

INTENSIVE FATHER INVOLVEMENT FOR NONMARITAL CHILDREN BY TIME SINCE RELATIONSHIP ENDED

	Overall	White	Black	Hispanic
Time since parents stopped coresiding
1 wave	49.1	47.1	52.1	44.7
2 waves	34.2	26.9	37.8	29.5
3 waves	24.4	20.6	26.8	19.7
4 waves	17.9	17.2	19.2	14.9
Time since parents were romantically involved
1 wave	42.1	43.2	42.4	40.9
2 waves	25.6	22.7	27.8	23.4
3 waves	19.5	16.8	21.1	16.7
4 waves	8.1	13.8	9.5	3.6

NOTE: Percentages are unweighted and pooled across survey waves in person–period format. Intensive father involvement is defined as seeing child at least eight days in past month.

Interestingly, the decline in father involvement after coresidence ends is steeper for white and Hispanic fathers than for African American fathers. This difference is in part an artifact of the fact that black fathers are more likely to maintain romantic involvement outside of a coresidential union. The decline in involvement by the number of waves since the parents were romantically involved is even larger, and this large decline is of about the same magnitude for all racial groups.

Next, we consider what accounts for the decline in involvement over time once the relationship ends. If traditional notions of fatherhood as a package deal hold sway, we would expect that much of the decline in involvement could be traced to entry of mothers and fathers into subsequent romantic partnerships, especially those containing children, for fathers in such arrangements could once again father the children in these households—either social or biological children—“automatically.”

Table 4 shows the predicted probability of father involvement by subsequent relationship status for white, black, and Hispanic fathers. These predicted probabilities are calculated from logistic regressions that control for exogenous characteristics including father’s age, education, employment, earnings, and time since parents stopped coresiding. Coefficients for the full logistic regression models are included in the appendix. These calculations show the steep drop-off in the probability of intensive father involvement for all groups as parents become involved with new partners and have new children with these partners. The declines in involvement are especially drastic for whites and Hispanics. It is important to note, however, that in this study and others like it, the direction of causality is unclear.

TABLE 4.

PREDICTED PROBABILITY OF INTENSIVE FATHER INVOLVEMENT BY PARENTS’ SUBSEQUENT RELATIONSHIP STATUS AT FIVE-YEAR FOLLOW-UP

	Overall	White	Black	Hispanic
Parents are single	.29	.24	.30	.27
Mom repartnered	.20	.18	.21	.18
Dad repartnered	.22	.19	.23	.20
Mom had subsequent children by someone else	.14	.15	.17	.07
Dad had subsequent children by someone else	.20	.08	.24	.15

NOTE: All parents were unmarried at focal child’s birth and father is presently nonresidential. Predicted probabilities are based on logistic regressions that control for father’s earnings, age, education, current employment status, child gender, and time since parents stopped living together, and are evaluated at the sample means for these variables. Full model parameters are listed in the appendix.

APPENDIX.

COEFFICIENTS FROM LOGIT REGRESSION MODELS PREDICTING AT LEAST WEEKLY FATHER INVOLVEMENT AT FIVE-YEAR FOLLOW-UP

	Parents Single	Mom Repartnered	Dad Repartnered	Mom Subsequent Children by New Partner	Dad Subsequent Children by New Partner
Race
Non-Hispanic black	0.59** (0.26)	0.62** (0.29)	0.66** (0.29)	0.69* (0.36)	1.34** (0.63)
Hispanic	0.11 (0.29)	0.10 (0.33)	0.09 (0.33)	−0.66 (0.59)	0.49 (0.68)
Non-Hispanic other race	0.53 (0.61)	0.26 (0.72)	1.19** (0.56)	0.75 (0.93)	1.05 (1.03)
Education
High school or GED	0.17 (0.18)	0.12 (0.21)	0.08 (0.21)	0.37 (0.31)	−0.05 (0.29)
Some college	0.33 (0.22)	0.40 (0.27)	0.13 (0.29)	0.69 (0.43)	−0.27 (0.39)
College or more	0.29 (0.51)	0.43 (0.65)	1.09 (0.72)	0.43 (1.17)	—a
Earned less than $15,000	−0.31 (0.24)	−0.79*** (0.28)	−0.82*** (0.26)	−0.25 (0.41)	−0.22 (0.36)
Employed	0.99**** (0.19)	0.78*** (0.25)	0.37 (0.25)	1.32*** (0.38)	0.84 (0.33)
Age (in years)	−0.004 (0.01)	−0.03 (0.02)	−0.01 (0.02)	−0.01 (0.03)	0.02 (0.02)
Child is male	−0.02 (0.15)	0.07 (0.19)	0.04 (0.19)	0.10 (0.29)	−0.06 (0.27)
Survey waves since parents stopped coresiding	−0.43**** (0.07)	−0.45**** (0.09)	−0.48**** (0.09)	−0.29* (0.17)	−0.37*** (0.14)
Constant	−0.64 (0.49)	0.13 (0.63)	0.10 (0.63)	−2.05* (1.08)	−2.08** (1.04)
N	944	810	684	439	389
Pseudo R2	.10	.09	.08	.10	.07

NOTE: Omitted reference categories are Non-Hispanic white and less than high school. All parents were unmarried at focal child’s birth, and dad is presently nonresidential.

^aThere are no college-educated fathers in this cell.

^*p < .10.

^**p < .05.

^***p < .01.

^****p < .001.

Table 4 reveals other interesting differences by race. Among African Americans, for example, a father’s subsequent partnerships and parental roles are less strongly associated with declining involvement than the mother’s subsequent relationships are, particularly if the mother goes on to have an additional child. This gendered pattern suggests a willingness on the part of the father to remain involved regardless of his other familial commitments, but less willingness on the part of the mother to facilitate that involvement once she establishes a new family. For Hispanics, the story is much the same, but the declines are more dramatic. For whites, it is the fathers’ own subsequent partnerships and especially their new parental roles that are associated with the strongest declines.

In sum, we have shown that for urban children born outside of marriage, ongoing rates of involvement are relatively high among all racial and ethnic groups. Black nonmarital children are far less likely to have parents living in coresidential unions, but are somewhat more likely to have parents in an ongoing, romantic partnership that is not coresidential. Once we limit our focus to only children whose parents are no longer living together or romantically involved, we see that the longer the father and the mother have been apart, the more father involvement wanes, though all else equal, this tendency is stronger for white and Hispanic fathers than for their African American counterparts. Finally, there is strong evidence that subsequent partners and parental relationships can account for a large portion of the decline, but again the impact is larger for whites and Hispanics than for African Americans.

Qualitative Findings on Father Involvement

We now turn to the qualitative data drawn from 165 low-income noncustodial fathers in the Philadelphia metropolitan area. Using these life history narratives, we explore the mechanisms that underlie the strong statistical associations between both men’s behaviors and men’s and women’s transitions into subsequent relationships. We begin by presenting 3 of the 165 cases in some detail, and then discuss broader themes derived from the qualitative analysis, drawing on these cases as examples.

William’s story

“We were friends for all our life,” said William, a white twenty-seven-year-old father, describing his relationship with Tiffany, the mother of his eleven-year-old daughter Brittney. “Then when I was about twelve or thirteen years old we started seeing each other. We were seeing each other for a while; then we broke up. Six months later,” William recounted, “she comes around saying ‘I’m pregnant, it’s yours.’ Of course I denied it. I was young. I said, ‘It’s not my kid. …’ I wasn’t there for the pregnancy. I wasn’t there for the childbirth, nothing. I guess after a while, [I admitted to myself that she was my child].”

“For seven years I didn’t have the right to see my child,” William told us, as he described how Tiffany retaliated in the face of his initial denials by refusing to allow William to have any contact with his child. William admitted that given his ongoing drug habit and the repeat incarcerations for petty drug offenses and parole violations during these years, her reluctance was perhaps justified. As Brittney was approaching her seventh birthday, however, William was remanded to a residential drug rehabilitation program and, newly clean and sober, began writing and calling his daughter, apologizing for the years he was absent and asking for another chance. “I started writing letters and calling my daughter all the time when I was at the rehab. [Then Tiffany] let my mom bring my daughter up once a week [to visit], so we started building a relationship. … We had a relationship when I left … that was really tight.”

Just before William’s release, Tiffany broke up with the father of her five-month-old infant son, BJ. William and Tiffany decided to try again, and after one month in a halfway house, he moved in with her and the two children. They spent nearly two years together parenting Brittney and BJ while he worked as a packer in a tropical fish warehouse. He quickly took on the father role with BJ, which was intensely meaningful to William since he was able to participate more fully in fatherhood as the baby grew. “Like BJ, when he was growing up that was great, because I wasn’t there for Brittney. [I loved] being able to see him do things and stuff like that, seeing him grow up.”

Meanwhile, BJ’s biological father was proving to be a problem. William boasts about how he quickly intervened when the man tried to work his way back into his son’s life, an attempt that William saw as a flimsy excuse for trying to rekindle his relationship with Tiffany. “It got to the point where he was harassing her all the time, calling on the phone calling her all kinds of names and stuff like that. So I had to get a restraining order against him.” Once, when William found Tiffany and BJ’s father together in his car, talking amorously in his view, “I got into a big fight with him … and tried to kill him.” Not surprisingly, the visits between BJ and his father ended.⁵ William, meanwhile, became closer than ever to BJ. William explains, “I have a son, but he’s not mine. She had a baby by somebody else and he was only like six months old when I came home and I practically raised him. … So he knows me as his father. He knows me as his daddy.”

Just after BJ’s second birthday, William was again locked up on a drug charge and served nearly two years. Tiffany told him she was through with him and did not want the kids around him either. Nonetheless, when William was released, Tiffany allowed him to spend two to three hours each day with the children after his work shift ended at 3:30, helping them with homework and taking them to the park. Eager to have his children on overnight and weekend visits, he used his first month’s wages from a warehouse job to rent a two-bedroom second-floor apartment. For several months, he saw his children daily and had them with him most weekends.

Then William became romantically involved with the first-floor tenant, a woman with daughters aged two and five. This relationship had its ups and downs as the new woman, Jessica, was not sure she should trust a repeat drug offender. But eventually Jessica asked William to move in with her. Six months later, when Jessica announced that she was pregnant, William planned to propose marriage, seeing his chance to finally do fatherhood right. “I want to be there for the pregnancy. I want to be there through everything. When she goes to the doctor, when she has the baby, to wake up with the baby in the middle of the night. It’s just something that I’ve been wanting for a long time.”

However, William’s happiness was marred by how his hopeful new life had come into conflict with his old. Tiffany, outraged that William’s attempts to rekindle their romance stopped abruptly when Jessica invited him to move in, began to deny him visitation, threatening to get a restraining order if he approached her apartment building. Tiffany and Jessica did not get along and had a few violent fights. Tiffany did not approve of Jessica, and said she was unwilling to let her children come under Jessica’s influence. William considered going to court to establish paternity so he could demand visiting rights, yet he hesitated—this would only allow him access to Brittney, his biological child, and not to BJ, and he did not want to gain legal access to Brittney only to lose out on contact with the boy he insisted on calling his son.

Brittney, age eleven, also played a role. “We were sitting down talking and she started running her mouth about me stepping in and out of her life. ‘You want to be a father to the little girl downstairs!’ [she said]. I said, ‘I’m not being a daddy, I’m just there to support her because her father’s not there.’” Thus, within six months of his relationship with Jessica becoming serious, his visits with Brittney and with BJ, as well as his financial contributions, fell to zero.

Like the vast majority of fathers in his situation, William was heartbroken by this state of affairs. “I can’t see my kids and I really hate it because I want a relationship with them. … My kids are the world. … I love doing their homework with them. I love being there for them. … Being a father’s great. They have [changed my life], they really have. They gave me a sense of responsibility. [Without them] I probably would be goofing off, hanging around on the corner or something like that. Now that I got responsibilities and things like that, it kind of keeps me in line, you know what I’m saying … ? My dad wasn’t there for me. My dad’s dad wasn’t there for him. … I don’t want to repeat the cycle again.”

Apple’s story

Apple, a twenty-six-year-old African American father, was proud that he was “in love and everything” with Gloria, the mother of his three children (ages eleven, nine, and five), during the eight and a half years the two were together. At first, they saw each other only casually, but within eight months she was pregnant with his daughter Vanessa. Apple, who had to repeat both seventh and eighth grade, had dropped out of school by this point and worked full-time as a drug dealer, but stopped two months shy of Vanessa’s birth. His determination to “go straight” was solidified when the baby was born, and as there was an outstanding warrant for his arrest, he decided that the right thing to do was to turn himself in. He and Gloria fought violently over this decision, which she saw as a desertion, and the altercation landed him in the emergency room from a knife wound in the cheek.

When Apple returned home after serving his sentence in a juvenile facility, he moved with Gloria and Vanessa, now nine months old, into a North Philadelphia row house that Gloria inherited from her grandmother. Everything was “lovey dovey” for a brief period of time—long enough for the conception and birth of a second child. During this time Apple worked twelve-hour days as a sandwich maker at a convenience store. During a store robbery he was injured with a gunshot wound and, because he had no insurance, was left with a large debt to the hospital. There was also some trouble in the relationship—Gloria admitted that she had been seeing another man and was pregnant by him, though she terminated the pregnancy—but she also soon conceived a third child by Apple.

Around the time this third child was born, Gloria became a Muslim and prohibited any drinking in the couple’s home. Things went well for a while, but a fourth child was then born that looked nothing like Apple. For a while, Apple convinced himself that he was the child’s father, but then Apple was caught failing to comply with the drinking prohibition. Another violent fight ensued and Gloria revealed the truth: Apple was not the fourth baby’s father. During this fight, a broken bottle used as a weapon caused serious wounds to his hands and arms that landed him in the emergency room again. Several weeks later, the two had yet another altercation on a trip to the Jersey shore with the kids in Gloria’s car. This time, Gloria called the police and accused Apple of carjacking. Apple’s bail was set at $35,000, and since he did not know anyone with enough money to pay a bail bondsman, he spent two weeks in jail before the charges were dropped.

Because of these two weeks in jail, Apple lost both jobs. Desperate for money, he decided to sell marijuana and was caught and incarcerated briefly, as this was his first adult conviction. Meanwhile, Gloria abruptly married a fellow Muslim, which devastated Apple, who still insists that Gloria was his “first love.” Upon his release, Apple moved in with his mother and began searching for work, finally securing a full-time job making sandwiches at a hoagie shop. He also found a new girlfriend, Jennifer, who had a job and her own apartment nearby. Apple moved in with Jennifer, and fourteen months later they conceived a child, who was born with a heart condition that qualified her for a disability payment of just over $1,000 a month. Jennifer quit her job to take care of the child full-time. With the $200 or more Apple cleared each week from the job plus the disability benefits, the two could cover their living expenses.

Meanwhile, Gloria, who left her husband and began to collect welfare, named Apple as the father of the oldest three children. Given Gloria’s history, his family suggested that he demand a blood test, but Apple decided against it. “I just never wanted to get the blood work just in case one of the [children wasn’t mine]. I would not have felt good about that. Then depression would have set in. So I guess I waived my rights.” Meanwhile, once Apple became involved with Jennifer, any direct contact between Gloria and Apple seemed to result in violent fighting. “I wish I could see all four, you know. I pray … we can work it out. But [Gloria], she just talk vicious to me like, threatens me.” Thus, he visits his children only rarely, though his daughter, the oldest, calls him daily. In fact, the last time he saw them was at a Father’s Day barbecue Gloria threw three months prior, a party to which Jennifer and the baby, Jade, were not invited.

Apple could barely contain his joy over life with his baby daughter. He felt his relationship with Jennifer, who was staying home full time with the baby, was “airtight,” and he gloried in his relationship with Jade, the eight-month-old. Despite his troubles with Gloria, Apple said, “I am glad I had four children, regardless [of whether] I’m with their mother or whatever. I’m not a rich daddy or the best daddy, but I’m still entitled, still have four children.”

Holloway’s story

Holloway was a thirty-nine-year-old African American father of a six-year-old. He showed up six days a week at 6:00 a.m. at a day labor agency in North Philadelphia. About four times a week, he succeeded in getting work and was transported to a work site where he did manual labor for $5.15 per hour. The son of a stably employed brick mason whose older brother—a welder—had a union job, this laborer still viewed his situation as better than average, as his two younger brothers were both serving substantial prison sentences. “They were out hustling man. They wanted me to hustle with them [but] I never really did. I’m the only one [in my family] that never really did. I’ve never been to jail. Not even locked up. I never been arrested at all.”

After dropping out of high school because he could not keep up with the work, Holloway began to train for a trade through Job Corps. “They sent me to Gary, Indiana and I stayed out there for about five months. I [got] in hot water, ’cause by me being from Philly, I had to hang with the guys from Philly ’cause if I didn’t, there was no protection for me. People would be like, not killed but stabbed, beat up real bad. They end up kicking [the Philly guys] out.” At nineteen, Holloway returned to Philadelphia and moved back in with his mother, picking up day labor work. At twenty-two, he joined the National Guard, but quit due to “problems with finishing stuff. I’m messed up like that.”

At twenty-three, Holloway had a stroke of luck and landed a full-time job with a building management and janitorial firm cleaning downtown office buildings for $6.75 an hour, plus double pay for plentiful overtime hours. A year later, he met Linda at a downtown club, a woman with two sons aged eight and ten. They moved into a row home that Linda’s aunt owned, and the three years that followed were Holloway’s best. “It was like I was the man of the house. Bringing in my little pay and stuff like that, it was a family. She made me feel like I was the boss. [I was] giving her half of my check when I had it.”

Though Holloway and Linda stopped using birth control early on, it took two years for them to “accidentally” conceive a child. Eight months into the pregnancy, two crises ensued that drove this relatively stable couple apart. First, a Thanksgiving weekend turned tragic when faulty wiring sparked a house fire. Holloway saved himself, jumping out the second-story window onto the fire escape, while the rest of the occupants fled onto the roof, where they were later rescued by firefighters. Linda screamed to Holloway that another child—the son of a cousin who was staying with them overnight—was still in the house. Holloway attempted to go back into the burning building, but the smoke and the heat were so intense that he quickly abandoned his search for the child, who eventually suffocated while hiding in a closet. Holloway spent the next three days in the hospital.

A week later Holloway received the news that his company had decided to downsize, and Holloway, along with many of the other more experienced workers, received a pink slip. This job loss, and the failure to quickly find a substitute, had devastating consequences. By the time the baby was born, Linda and her family, who already blamed Holloway for the death of the four-year-old child in the house fire, had firmly decided that he was “a loser,” and Linda gave Holloway “the boot.”

Over the next six years, Holloway made valiant attempts to stay connected with his child, and came by whenever he had enough in his paycheck to buy her a treat or go shopping for something she needed. As he lived with his brother and had modest living expenses, he expended the majority of his resources on these outings. “I like seeing her grow up. Smile. I guess it makes me feel … it’s like I’ve achieved something. ’Cause like for me, I didn’t think I could make something as pretty as my daughter. That’s what I think, it’s great to see something that you made to grow up.”

But these visits stopped abruptly once Linda acquired a new “friend.” Linda stopped letting Holloway come around the house to visit anymore, afraid her new man would think that she and Holloway “are up to something.” And because he lived with his brother—his mother moved to New York—she would not let her visit there either because his brother was once charged with the rape of a former girlfriends’ daughter—charges that were later dropped. Now the only time Holloway sees his daughter is through a chance neighborhood encounter.

As Linda’s new romance progressed, it became clear to Holloway that these were not the only threats to his parenthood: Linda’s new boyfriend “happens to have taken a liking to my daughter,” he confided. “He seems like he’s trying to take my daughter from me.” Just before we met him, Holloway walked around the corner in hopes of encountering the child and spied Linda’s boyfriend buying ice cream for his daughter. According to Holloway, “[I felt like saying,] ‘I have a couple of dollars, I might wanna buy ice cream. …’ I wanted to talk to him several times, and pull him to the side and say, ‘Look man, she’s my daughter, you don’t really gotta buy her ice cream. You know, I do work sometimes.’” Holloway continued, “And this guy she’s with, he got kids somewhere else. … He lost his family, so he gotta take mine. [And] he has the power [to] because he has a good job. He’s like a big shot.”

“I suppose every father that’s not with his kids’ mother or whatever, probably go through the same thing I go through. … Being like broke up from the family, I feel abandoned. That’s maybe because I had a baby by the wrong person. … My brother, he talking about don’t go see my daughter or nothing, leave them both alone. But see I got a problem with that. I don’t want to abandon them. I ain’t much, but at least she knows that she has a father.”

Themes from the qualitative analyses

The narratives offered by William, Apple, and Holloway illustrate several themes that are common in our analysis in the 165 in-depth interviews with unmarried fathers in Philadelphia.

Salience of the father role

For all three men, the importance of their role as a father was high. All but William embraced fatherhood from the beginning. William, who denied paternity initially, is more the exception than the rule in our data. If anything, like Apple, the men in our study seemed eager to claim as many children as they could, even when they had some reason to believe that a child might not be theirs biologically. This is perhaps another indication of the salience of the father role.

Rejection of the package deal

Most fathers, but especially African American fathers, firmly rejected the “package deal” noting that a father’s parental relationship is contingent upon his relationship with the mother, although many end up living by it nonetheless. This is reflected in the high rates of contact after breakup. Holloway’s disdain of Linda’s new partner, who had abandoned responsibility for his own biological children, was typical. As Sullivan (1993) also found, men who “step off” from their responsibility as fathers are nearly uniformly condemned, while those who continue to care for their children are generally valued by their peers.

Declines in involvement as unintended consequence of new relationships

No father in our study entered into a new relationship with the intention of shedding commitments to children from past relationships. Instead, the decline in involvement was an unintended consequence of the transition to the new partnership. Faced with real limitations of resources and time, fathers often found that if they wished to make a go of the new relationship, and particularly if they wanted to play the father role to their full satisfaction, the new relationship required most of the resources they had to invest.

Increased transaction costs

Subsequent partnerships can dramatically increase the transaction costs of visiting children from past relationships. As Apple’s story richly illustrates, romantic relationships among unmarried parents with children are often quite volatile even from the beginning (Reed 2006; Hill 2007). Sexual jealousy between new and old partners is often a theme, as is evident in both William’s and Holloway’s cases, but these stories also richly illustrate how many different people have to cooperate for visitation to occur—not only the mother and father but the new partner and the children themselves.

Contradictory definitions of fatherhood

On one hand, when claiming father status with nonbiological children they live with and help raise, many of these fathers firmly rejected the notion that fatherhood is primarily biological. This is most evident among our African American fathers. As we listened to these men talk about their roles as social fathers, and sometimes had an extended opportunity to watch them enact these roles, we became convinced that in some cases, men’s roles as social fathers may be as meaningful to them as those with their biological children, as was the case for Apple. However, when it came to their own children by blood, these same men claimed the primacy of the biological tie, and were outraged when their ex-partner and the new man in her life failed to respect him as such.

Motivations to claim social children

As both William’s and Holloway’s stories show, albeit from different angles, men who enter into new relationships with women who have children by prior partners have strong motivation to claim fatherhood for the children in that household, or have another child by the new partner, especially if they can “do fatherhood right” by participating in their upbringing from an early age. As we see in Linda’s relationship with her new “friend,” in the typical courtship scenario men tend to woo women by wooing their children as well, in part because women often use the way a man treats “another man’s child” as a test of his worthiness as a partner (Edin and Kefalas 2005).

Although many men who play this role insist that their intent was not to push out the biological father, this is sometimes a consequence. A parallel study of low-income single mothers in Philadelphia (Edin and Kefalas 2005) shows a similar pattern for women—though they rarely take on new partners with the intent of displacing the father, the time and attention bestowed on the children by a new romantic partner often creates an unfavorable comparison.

As Elliot Liebow (1967) argued four decades ago, our research shows that even if both men are making equal contributions, the offerings of the man without the biological child are valued more because they are not measured against his obligation to the child, a mental accounting that almost always places the biological father’s performance in the red and the nonbiological father’s performance in the black. When the new partner is a “big shot” like Linda’s new partner and biological fathers like Holloway are only minimally employed, the contest can seem impossible to win.

A portrait of continuous fathering

Except for periods in men’s life course when they are struggling with addiction or are incarcerated, an examination of the 165 cases reveals a portrait of almost continuous intensive fatherhood. This portrait is consistent with Lerman and Sorensen’s (2000) analysis of data from six waves of the NLSY79, collected between 1984 and 1992, which assesses how many fathers of nonmarital children are intensively involved with any of their biological children at any point in time. They conclude that “the striking reality is that about two of three fathers (under 35 years old) who have fathered a child out of wedlock have a close involvement with at least one of their nonmarital children. Many of those who do not … have married someone else and are living with a marital child” (p. 145).⁶ Similarly, most fathers in our study felt they were living out the high value they placed on fatherhood by intensively involving themselves in fathering activities for at least one child—whether biological or social—at any given time. Ironically, these narratives often show that part of the motivation to enter subsequent relationships is to enact the father role in a more complete and satisfying way.

Conclusion and Discussion

International comparisons show that among U.S. couples, cohabiting unions among parents with children are extraordinarily fragile—far more fragile than marital unions and far more fragile than unmarried parental unions in other countries in the industrialized world (Andersson 2001). The findings in this paper support the theory that this may, in part, be due to strong norms that support the traditional notion of fatherhood as a “package deal,” especially for Hispanics and whites. This support is not only evident in the strong falloff in father involvement after breakup but in the large effect of mothers’ and fathers’ subsequent partner and parenthood roles in declining father involvement.

As both the mother and the father of a child born outside of marriage move further away from their failed partnership and enter new partnerships and new parental roles, the qualitative data show that new normative expectations are often set into motion that are in sharp competition with the old. Especially for mothers, new partnerships seem to provide a strong motivation to give the new partner the role of father, particularly once the mother has a child with that partner. For his part, the father may be under considerable pressure to use his scarce emotional and financial resources to fulfill the demands of his new partner and parenting roles, which he can enact within the context of a conjugal relationship.

Thus, while the conventional wisdom might assume that unmarried fathers are uninvolved because they are eager to evade responsibility for their progeny, our results suggest a very different story. This analysis suggests that declining rates of fathers’ involvement are primarily due to unmarried women’s and men’s eagerness to enact a cultural ideal of parenting that views it as part of a package deal. Indeed, it may, in part, be women’s and men’s desire to demonstrate competence in subsequent partner relationships and parenting roles that leads to diminished involvement with children from other relationships.

The findings also support the Mincy and Pouncy (2007) hypothesis of greater institutionalization of the “baby father” role among African Americans than among other U.S. racial and ethnic groups. It may be that stronger norms guide unmarried African American fathers as they enact the father role, which sustain their involvement with the child even after the relationship with the child’s mother ends. African American fathers are therefore more likely to remain in regular contact with their children even after entering into relationships with new partners and having children with them.

Past analyses have considered a wide array of factors correlated with the large decline in father involvement among the fathers of nonmarital children over time, but less attention has been paid to the mechanisms involved. No other analysis of unwed parents we know of has focused on the role of men’s and women’s subsequent romantic relationship transitions, whether marital or not, and their subsequent transitions to new parenting roles. Nor has any study we are aware of looked at mothers’ and fathers’ relationship transitions simultaneously, so that the relative importance of each can be ascertained. Our analysis suggests that the ways in which fatherhood is defined, both in the culture at large and within racial and ethnic subcultures, hold significant sway.

So was Moynihan right to worry about the impact of rising rates of nonmarital childbearing on future generations? On one hand, we document relatively high rates of father involvement with nonmarital children, which contradicts common conceptions of the “hit-and-run” father. But answering this question requires us to go beyond the father’s view and to look at the situation from the point of view of the children. From this vantage point, we concur with Moynihan, though clearly the impact now reaches beyond African American children to encompass a significant minority of all American children. Because fatherhood is generally enacted in a meaningful way within the context of a conjugal union, because the fragility of these unions is high, and because repartnering and subsequent childbearing is common, children are likely to experience fatherhood as a game of musical chairs, a series of temporary commitments rather than a lifelong obligation.

As stability is critical for child well-being, the shifting cast of fathers and father figures in children’s lives is likely to detract from, not add to, their well-being (Fomby and Cherlin 2007). Although social fathers might sometimes add to the well-being of their nonbiological children, comparisons of children in single and married stepfather homes suggest that this is not typical (McLanahan and Sandefur 1994). It is unlikely that many children in this situation will receive the same level of emotional or financial investment enjoyed by those who live stably with both their biological mothers and fathers.

Biographies

Kathryn Edin is a professor of social policy at the John F. Kennedy School of Government at Harvard University. Her research focuses on urban poverty and family life, social welfare, public housing, child support, and nonmarital childbearing and the economic lives of the poor. Her most recent publication is Promises I Can Keep: Why Poor Women Put Motherhood before Marriage (with Maria J. Kefalas; University of California Press 2005). She is also coauthor of Making Ends Meet: How Low Income Single Mothers Survive Welfare and Low Wage Work (Russell Sage Foundation 1997). Current projects include an in-depth study nested within the interim evaluation of the Moving to Opportunity Experiment and an in-depth longitudinal study of the Gautreaux Two program.

Laura Tach is a doctoral candidate in sociology and social policy at Harvard University. Her research examines how various social contexts, particularly family structures and neighborhoods, influence the well-being of individuals. She is presently involved in projects studying mixed-income neighborhoods, multiple-partner fertility and father involvement, and trajectories of marital quality.

Ronald Mincy is the Maurice V. Russell Professor of Social Policy and Social Work Practice at the Columbia University School of Social Work and director of the Center for Research on Fathers, Children, and Family Well-Being. He has published widely on the effects of income security policy on child and family poverty, family formation, child well-being, responsible fatherhood, the urban underclass, and urban poverty. His most recent book is Black Males Left Behind (Urban Institute 2006). He is currently working on a multiyear study of the New York State Non-custodial Parent Income Tax Credit and a random clinical trial of an intervention that uses nonresident fathers to delay sexual debut among African American males.