Markkula Center of Applied Ethics

Family Change and Welfare Reform

Jane Mauldon1

I. Introduction

The accelerating growth of single-parent families poses special challenges for policy makers in the United States. Since the Colonial Period, the support and rearing of children has been defined as the almost exclusive prerogative and obligation of parents. The state has played a very circumscribed role vis-`a-vis families, chiefly as a monitor of children's well-being. The flip side of parental autonomy is an expectation of self-sufficiency. Public education constitutes the only major exception to the overall dearth of public resources available to most children. Otherwise, the state provides very limited material and other help to most families with children. There are, for example, no child allowances in the United States, no mandatory paid maternity leave, no universal health programs for children, little publicly provided child care, and so forth.

Consistent with this country's reluctance to assist families directly, most public discussion about "the family" has focused on policies to influence adult behavior towards children rather than on strategies to provide material support for children. In the current Congress, efforts to tie assistance to "appropriate" or "moral" parental behavior are reaching levels unprecedented in national policy. Politicians are proposing to deny assistance to children conceived while the mother is on Aid to Families with Dependent Children ("AFDC") and to children conceived by unwed mothers who are under the age of eighteen. Other proposals seek to cut or deny aid for families if the mother does not comply with program work requirements, if she provides any false information on her application for aid, or if she does not cooperate with official efforts to collect child support from the child's father. Some states are reducing a family's public subsidy grant if the children do not attend school since this is behavior that is expected to be under the parents' control. Other states cut a teen parent's AFDC grant if she does not attend school. Under all of these initiatives, public assistance to a family is tied to parental behavior and not solely to children's need.

Two demographic trends have fueled these policy proposals: (1) increases in the number of unmarried women giving birth; and (2) the accompanying rise in the number of fathers who appear to have abandoned their children. In this paper, we describe these demographic changes in some detail and place them in context. We review some of the reasons for the trends and assess how likely they are to be slowed or reversed. We outline some likely consequences of proposals to "reform" AFDC, and conclude with a review of other policy strategies to assist families with children. Because adult behavior towards children, rather than the lives of children themselves, is the center of policy interest and the object of policy concern, the presented data focuses on adult living arrangements and behavior.

II. The Demographic Picture

In the United States in 1993, there were 9.3 million female-headed families with minor children, of which nearly four in every ten (3.4 million families) were headed by a woman who had never been married.2 The median income of a female-headed family was $13,445, far below that of married couples. About half of these families were below the official poverty threshold, causing many to depend on public assistance.3 This data shows the hardships experienced by many children in single-parent families, and provides a clear rationale for the current levels of policy concern.

A. The Rise in Unwed Childbearing

The proportion of births out-of-wedlock has risen quite sharply in the past two decades, from 11% in 1970 to nearly 30% in 1992.4 The large majority (69%) of births to teenagers are outside marriage, while four out of ten births to women aged twenty to twenty-four and 17% of births to older women are also out-of-wedlock.5

The increase in nonmarital births can be linked directly to the growing number of young women who are unmarried. In 1990, nearly two-thirds (63%) of women aged twenty to twenty-four had never married, whereas in 1970, only 28% had never married. Currently, both men and women have the highest median ages at first marriage ever recorded, 23.9 and 26.9, respectively (the data goes back to 1890).6 Since 1970, the median ages at first marriage have increased a full three years for both men and women.7

These changes have lengthened the years of singlehood for sexually mature young women and men by nearly fifty percent. Young women now have more years in which to get pregnant while still single. The early twenties have generally been a time of life characterized by high rates of childbearing. Since the period of unmarried sexual activity has been lengthened, there has been a modest decline in the ages at which young people become sexually active. While sexually active teenagers today are more likely to use birth control,8 the increases in contraceptive use have not kept pace with the increased numbers of unmarried people who are having sex. Thus, we find a sharp rise in nonmarital pregnancies.

If women do become pregnant, their options are broader than in the past. About half of nonmarital conceptions are now aborted, an option not readily available to most women before 1973. Unmarried motherhood is also more socially and economically feasible than in the past. Marriages to legitimate a conception, "shotgun marriages," are becoming a thing of the past. As recently as the 1960's, more than half (fifty-three percent) of women who conceived a child premaritally, a rarer event than now, were married before the birth of the child. Even sex between unmarried people was described as though the couple would later marry since it was known as "premarital sex." By the 1985-1989 period, only one-fourth (twenty-seven percent) of premarital conceptions were legitimated through marriage.9

Table 1 summarizes birth and marriage data since 1970 for women aged fifteen to thirty-five, accounting for 90% of all births. The table shows the number of births attributed to unmarried women for the age-group twenty to twenty-four increasing from 10% of births in 1970 to 41% in 1992. The table also shows two trends that have led to a higher percentage of births occurring outside marriage. The trends are, first, the rising number of women who are unmarried, now standing at 68% of women aged twenty to twenty-four, and, second, their rising propensity to give birth when unmarried. For comparison, the table also shows the fertility rate in each age-group for all women, married and unmarried.

Table 1:
Births to Unmarried and Married Women,
By Age Group: 1970-1990

[Space]

[Space]

Percent of Births that were to Married Women,
by Age of Mother:


Percent of Women who were Unmarried,
by Age:


Unmarried Birth Rates (Unmarried births
per 1,000 Unmarried Women):


Overall Birth Rates (All Births per 1,000
Women in Age-Group):


Percent of Nonmarital Births that were
in each Age-Group:

Age of Woman
1970
1980
1992
[Space]
15-19

29%
48%
70%

20-24

10%
20%
41%

25-29

6%
10%
20%

30-34

7%
8%
14%

1970
1980
1992
[Space]
88%
93%
95%
45%
58%
68%
24%
33%
41%
20%
23%
30%
1970
1980
1992
[Space]
22.4
27.6
44.6
38.4
40.9
68.5
37.0
34.0
56.5
27.1
21.1
37.9
1970
1980
1992
[Space]
68.3
53.0
60.7
167.8
115.1
114.6
145.1
111.4
117.4
73.3
61.9
80.2
1970
[Space]
29%
36%
19%
10%

While the fertility rates of unmarried women have been rising steadily, the overall fertility rates, independent of marital status, fell in most age-groups between 1970 and 1980. Since 1980, they have held steady for women in their twenties while rising for teens and women in their thirties. The increase in births among teens is a result of the increasing number of teens who are sexually active. The increase for older women largely represents delayed childbearing rather than a return to larger families. The final row of Table 1 shows the fraction of nonmarital births to women in the different age groups.10 Contrary to the stereotype, teenage births are a minority, consisting of twenty-nine percent of all nonmarital births. Teenage births are important for several reasons. First, they represent the onset of childbearing. Second, they may sometimes disrupt other important transitions into adulthood such as schooling or employment. Third, women later regret having had their first child young, and some teens would become better equipped to be parents if they delayed childbearing. However, numerically, women aged twenty to twenty-four have the largest share of nonmarital births, accounting for some forty percent of the total. Overall, about two-thirds of all nonmarital births are to women under twenty-five.

These tables present data on two aspects of the current transformation of American families: marriage and childbearing. A third important aspect is the acceleration of the divorce rate during the 1960's and 1970's, to a point where about half of all marriages now end in divorce. The consequence of these changes is that large numbers of women are now raising children alone, while large numbers of men are neither living with nor supporting their offspring.

B. Fewer Men Living With Children

Table 2 represents these trends since 1973 in terms of the living arrangements of men and women in the prime child-rearing years, between eighteen and forty-five.11

Table 2:
Living Arrangements of Adults of Childbearing
Age, 1973-1993

[Space] [Space]
1973
1983
1993
Women, Age 18-45 [Space]
Married with chiildren
Single parent
[Space]
55%
9%
44%
13%
41%
15%
Married, no chiildren
Unmarried and childless (family-free)
[Space]
13%
23%
12%
32%
13%
30%
Total [Space]
100%
100%
100%
Percent of women living with children [Space]
64%
57%
57%
Men, Age 18-45 [Space]
Married with chiildren
Single parent
[Space]
52%
1%
40%
1%
37%
2%
Married, no chiildren
Unmarried and childless (family-free)
[Space]
12%
35%
12%
46%
13%
47%
Total [Space]
100%
100%
100%
Percent of men living with children [Space]
52%
42%
40%
Median age [Space]
28
29
32

In 1973, a majority of both men and women in this age-group were married with children. That proportion has since fallen fourteen percentage points, or by about one-fourth. Now, more young and middle-aged adults are unmarried and childless than are married with children. The proportion of adults age eighteen to forty-five who are living apart from parents, children, and spouses, and generally apart from all family members, has doubled in the past twenty years.12

This data clearly shows men's "flight from the family." Not only are six of every ten young and middle-aged men not routinely involved in child-rearing, but most of these men are not married. When the 2% of men living on military bases and the 1.5% of men confined to prison are included in the analysis, it is clear that a majority of men aged eighteen to forty-five are not living in families of their own. This means that they do not live with a spouse or child of their own, a status that we will term hereinafter "family-free." Nearly half (45%) of these family-free young adults, both men and women, are living in their parents' homes, while the rest appear to be living entirely outside families.

Although the overall trends are similar in all groups, the changes in the family have occurred at different rates among different racial and ethnic groups.13 A majority of men, regardless of ethnicity, are not living with children. This proportion ranges from 58% of Whites to 72% of African-Americans. The proportion of men who are family-free ranges from 46% of whites to 65% of African-Americans. While female headship and unmarried childbearing are most likely to occur among black women, the majority of nonmarital births and female-headed households are white.14

Increasingly, men and women are going through dramatically different experiences as they mature into young adults. Among the youngest adults, aged eighteen to twenty-four, more than a quarter (26%) of the women are living with and raising their own children, while only 8% of men are. Men never seem to catch up to women. In the entire eighteen to forty-five age-group, more than half (57%) of women are raising their own children, and more than one-fourth of them (25%) are doing so alone.15 In contrast, only 40% of men in this age-group are raising their own children, and only 2% are doing so alone. Men are now 40% more likely than women to not have children or to be living apart from them. They are 70% more likely to be entirely family free.

The gender gap in child-rearing is most marked among African-Americans. Only 28% of black men are living with their children on a full-time basis, while nearly two-thirds (62%) of black women are. Again, the experiences of the youngest adults are the most divergent. Thirty-seven percent of black women aged eighteen to twenty-four are raising children, whereas only 5% of black men are living with their own children.

This large gender gap is directly related to the lower marriage rates among African-Americans. The majority of black women with children are not currently married. To the extent that African-American men cohabit with their children's mother in her home instead of marry, these tables will exaggerate black men's absence from their children's lives. Intermittent cohabitation, as well as other types of involvement in children's lives, such as frequent visiting, cannot show up in our data either. Some ethnographic data16 suggests that these behaviors are quite common in low-income African-American families, in which case these tables overstate the gender gap.

A sharp disjuncture now exists between the child-rearing experiences of men and women, and it is greatest among the youngest adults, those under age twenty-five. This gendered disjuncture has always existed because family responsibilities have been split along gender lines, and child-rearing responsibilities have fallen-at least since the Colonial Period- almost entirely to women in both single and two-parent families. However, men who live apart from their children are, for the most part, absent from child-rearing to a much greater extent than the prototypical "fifties father" who brought home the family paycheck but was emotionally distant from his children.

C. Why the Decline in Marriage?

These rapid demographic changes pose complex and difficult challenges for policy makers. One response is to cast about for ways to slow or reverse the trends towards single-parent families. A favorite scapegoat is the availability of AFDC. However, as we will see, the evidence supporting this theory is weak at best. The changes in family structure, specifically the decline in marriage among young adults and the growth of nonmarital childbearing, result from a number of intersecting forces in which AFDC is only a bit player.

One set of forces is the opportunities facing young adults in terms of education, employment, and wages. The more advantaged and talented young adults are delaying marriage in favor of education, as they attend college or postgraduate training in increasing numbers.17 Marriage has traditionally followed, rather than preceded or accompanied, schooling.

An even greater barrier to marriage may be unemployment, especially male unemployment. Among young people who do not go to college, unemployment is chronic and severe. In 1992, 20% of the young people not enrolled in school were not in the labor force either, and 13% of those who were in the labor force were unemployed. In sum, 30% of young adults not in school were also not working. And what fraction of those aged sixteen to twenty-four fit the classic profile of a marriage prospect? The fraction in 1980 was 40%. By 1992, this fraction had dropped to 34%. For whites, the decrease was from 42% to 36%, and for African-Americans, from 30% to 26%.18

These low employment rates and accompanying low wages translate into incomes that are completely inadequate to support a family. By the late 1980's, one in two white men (52%), three in four African-American men (76%), and two in three Latino men (60%) aged twenty to twenty-four had earnings below the poverty level for a family of three.19 Even for men in their late twenties and early thirties, the proportions were nearly one-quarter of white men (24%), more than one-third of black men (37%), and just under one-half of Hispanic men (40%) aged twenty-five to thirty-four with earnings below poverty.20 These proportions were all much higher than a decade earlier, by some ten to thirty percentage points.

The problem of intolerably low earnings is the most severe for young men whose education stops at high school graduation or before. The average earnings of a male high school dropout fell by more than forty percent between 1973 and 1986, and the earnings of a male high-school graduate fell by nearly one-third.21 Not surprisingly, the least educated young men and women have the lowest marriage rates, the young women have the fewest reasons to delay childbearing, and consequently these parents have the highest rates of non-marital childbearing.

A second set of forces discouraging marriage has to do with changing attitudes. Today's high labor force participation rates for married women have dramatically changed family life and the very character of marriage in the United States.22 Although young men, especially those who are less-educated, now have less to offer economically than in the past, one might argue that young women can contribute more financially to the marriage today. Thus, the relationship should be able to go forward anyway. Single and married mothers now have very similar probabilities of working. In 1990, 72% of married mothers living with children under eighteen were employed, compared to 70% of single mothers.23

An egalitarian type of marriage seems more available to young people than in the past, even if the traditional "man as breadwinner" model is precluded by the high unemployment and low wages facing young men. However, an egalitarian marriage is probably easiest to sustain if no children are present at the outset. Moreover, a large body of evidence indicates that in most marriages, "egalitarian" more often characterizes the income flow into the home than the spouses' contributions to the work within the home. Even men whose wives are employed full-time rarely participate equally with their wives in the work of the home.

The data suggests that young women are assessing their potential marriage partners, and are finding themselves disappointed. They may not want an egalitarian marriage, it may not be offered to them, and the men around them may not have much to offer materially. Young women are also, however, finding themselves not ready for marriage and finding the institution of marriage itself a risky proposition.24 They know that a majority of young marriages end in divorce. They have often had personal experience as children of their parents' divorce or separation, and they may wish to protect themselves and their child from the trauma of what they see as an inevitable separation.

The polarized child-rearing experiences of young adults, especially of young African-Americans, can only contribute to a lack of understanding and trust between men and women, and between prospective spouses. Recall that 37% percent of black women aged eighteen to twenty-four are raising children, while only 5% of black men are living with their own children.

Meanwhile, as marriage seems less and less permanent and appears to offer fewer and fewer advantages to young people, unmarried childbearing has lost much of its stigma.25 The increase in unmarried childbearing has occurred among all racial and ethnic groups, and at all ages. Moreover, it is common to all industrialized and industrializing nations. Indeed, unmarried childbearing is even more common in several European nations than in the United States, and is associated with the growth in women's economic and social independence as well as the concomitant decline in marriage that are widely shared international trends.

III. Public Policy Responses

The substantial gender gap in the child-rearing behavior of young Americans, along with the high poverty rates among families with children, has drawn attention to policies governing AFDC. On the one hand, men are seen as getting away without supporting the children they helped conceive. On the other hand, women are seen as able to live off the state instead of getting married. Although recent and proposed changes in the laws governing these two policy areas are often advocated as ways to stem the increase in mother-child families, these claims may be nothing more than vain hopes. In the following section, we analyze some current policy initiatives with an eye to their likely consequences for children, and their possible impacts on adult behavior, particularly as they pertain to the household structure.

2A. The AFDC Program

AFDC is often cited as the most important, or even the sole, cause of the increase in female headship. The program provides cash benefits to low-income, primarily single parents with children. Because very little public aid exists for adults without children, and because AFDC is much less readily available to two-parent than single-parent families, the program is widely believed to encourage divorce and nonmarital childbearing.26 Consequently, reforming welfare has become a central policy issue for both Democrats and Republicans.

The evidence supporting the claim that welfare caused the increase in mother-child families is not strong. Although some recent studies have found modest but statistically significant increases in unwed childbearing or divorce among certain groups of women associated with higher levels of AFDC payments,27 the majority of studies have found no effects at all of AFDC on family formation behavior.28 The correlational evidence tends to run in the wrong direction. The growth in single-parent families accompanied a decline, not growth, in average AFDC payments per family for the last two decades. 29 Unmarried birth rates tend to be highest in the states that have the lowest AFDC grants. Moreover, the fraction of single-parent families utilizing AFDC has not been rising.

The conclusion reached by Hilary Williamson Hoynes in 1994 from a study of twenty years of longitudinal data is typical of many studies.30 "This study clearly shows that there is no evidence [in the data considered] that AFDC benefits play any role in female headship decisions."31 Among scholars who have found AFDC to play a role, none have suggested that those effects could account for more than a small fraction of the overall growth in divorces and out-of-wedlock births.

Nevertheless, both Democratic and Republican politicians now seem to believe that only a complete overhaul of the program will stem the growth of out-of-wedlock childbearing. Republicans are proposing many changes to AFDC, of which I will focus on only two: (1) providing federal financial support to the states in the form of block grants, rather than as a subsidy based on the number of AFDC cases served by the state; and, (2) making children born to unmarried women under eighteen permanently ineligible for AFDC.

The proposal to shift the federal component of AFDC financing from an open-ended entitlement basis to a set of capped block grants to individual states should be viewed against a demographic backdrop. A state's AFDC caseload is a function of the number of children living in the state, especially the number of young children since families tend to use AFDC when their children are very young. The number of young children varies with the number of women in their prime childbearing years, usually their twenties. Other important factors are the proportion of families that are single-parent families, which is itself a function of the age and ethnic composition of the state's population, and the strength of the state's economy.

These demographic and economic factors influencing the size of the AFDC caseload can shift quite sharply over time and differentially among states. States in which the population of young adults is growing will have more and more new families and additional children every year, thus enabling more people to be eligible for AFDC. A funding formula that is tied to the state's past AFDC caseload will not be responsive to the growth in the number of children.

In a sharp break with the past, block grants will not provide a state's AFDC program with the necessary federal funds to expand with demographic trends, or with local unemployment rates. California's recent history offers a fine example of both problems. California saw its AFDC rolls grow sharply between 1990 and 1992. This was a consequence of the large number of women in their late teens and twenties living in the state, the resulting large number of births, and the statewide recession that pushed many parents out of the work force and into temporary dependency on AFDC.32

A block grant system would not have provided the kind of flexibility California needed to respond to those demographic and economic circumstances. The demographic and economic trends that propel families onto AFDC are not under the direct or even indirect control of individual state governments. Yet these are the proximate factors driving the AFDC caseload. If these factors change so as to generate higher AFDC caseloads, states will have to manipulate other policy tools, such as the level of the grant or the eligibility criteria for the program, in order to bring expenditures back into line.

The proposal to deny aid to babies born to unmarried minors is intended to reduce the AFDC caseload by making some children ineligible for the program. It is also intended to reduce poverty in the long run by discouraging those women who are clearly unable to support a family from having children. It is part of a broader agenda to discourage unmarried births by cutting off financial aid to these babies.

Examining the last point first, this proposal will have only a small impact on the overall unmarried birth rate, since nearly ninety percent of nonmarital births are to women aged eighteen and over. Even though the number of children officially supported by AFDC grants may fall to some extent, the number of AFDC cases and expenditures are not likely to shrink appreciably for the following reasons.

In 1992, there were 160,000 births to unmarried girls under eighteen. We assume that between 60% and 80% of these mothers are on AFDC during the baby's first year, so these infants account for between one-sixth and one-quarter of all infants on AFDC. If these children make up a roughly constant fraction of AFDC recipients as they age, we may assume that eventually one-fifth of all children who otherwise would receive AFDC will be ineligible under this "exclusion of minors" rule.

However, the number of families and the volume of expenditures on the program will not shrink by as much as one-fifth. Data from the Current Population Survey indicates that only about one-fifth of minor parents on AFDC live on their own as heads of their own AFDC cases.33 The remaining four-fifths live with their parents, who are almost always themselves recipients of AFDC grants. In the past, a young mother's new baby has been included in the grandparent's AFDC grant. However, under the new policy, this will not be the case. The new policy will prevent only a few new AFDC cases from being opened, perhaps only four percent fewer than in the past. The main effect of the proposal will be to compel older women already on AFDC to care for a new grandchild without an increase in their grants.

Because the fiscal impact of the minor-exclusion rule will typically be felt by the young mother's own mother and by the family as a whole, rather than by the young woman alone, its deterrent effect on childbearing is likely to be weak. As we have already seen, the steady drop in the value of the AFDC grant has not discouraged unmarried childbearing among teens or older women in the past. In effect, this new policy simply ratchets the grant down another notch, forcing families to cope as they have done in the past.

Once the young mother turns eighteen her situation changes in two ways. First, she is no longer eligible for AFDC on her mother's grant or will no longer be once she turns nineteen, depending on her state of residence and her educational status. More importantly, only if she has another baby will she become eligible to head her own AFDC case. Thus, one result of this policy may be to encourage a young unwed mother to have a second child as quickly as possible after her eighteenth birthday. A birth at that point entitles her to a welfare grant which is intended for a family with one child rather than two.

We have considered only two elements in the Republicans' proposals to revamp welfare. In essence, these proposals would manipulate the program to lower the grants across the board or for certain types of families. Other proposals, such as the two year limit, would involve removing large numbers of people from the program. These policies are unlikely to have more than a marginal impact on the rate of unmarried childbearing. It is hard to see why they would encourage marriage, given that past cuts have been associated with fewer marriages.

Indisputably, these proposals would inflict harm on children, with the extent of the harm commensurate with the magnitude of the cuts. There is abundant evidence that poverty is bad for children and severe or long-term poverty is even worse for them. Poor children are much more likely than affluent children to grow up malnourished, developmentally impaired, unhappy, in poor physical and mental health, under-educated, and ultimately under-employed or unemployable. AFDC is a safety net that does not lift recipients out of poverty but keeps them from complete destitution. Every AFDC cut causes recipients to fall deeper into poverty and pushes some families into homelessness or other kinds of severe hardship.

If we, as a society, want children to grow up with sufficient human and material resources, policies must be adopted that make marriage economically and socially viable for disadvantaged young parents. Many of the strategies that would be good for children could also encourage marriage or discourage nonmarital childbearing. It is to these policies that we now turn.

2B. Policies to Support Young Parents

A focused strategy is needed to improve the lives of the fifty percent of young people who do not attend college. College attendance offers young people many advantages apart from the education that subsequently translates into increased earning potential. As long as they are still students, young people receive shelter and an extension of the protection and counsel provided by parents. Most college teachers are very conscious of their dual roles as scholars and as mentors. Students who flounder and fail typically get second chances, and the institution offers various sorts of assistance to help them get back on track and successfully complete their education.

Young people who do not attend college have no such institutionalized safe havens in which to mature. Rather, they face an exceptionally brutal job market, which is far less open and forgiving than the job market facing better-educated people. Within this harsh environment, any physical, emotional, or mental setbacks, such as the depression, uncertainty, irresponsibility, and impulsiveness often displayed by adolescents, can have serious consequences. Thus, while better employment, education, and training programs are obviously needed for non-college-bound youth, so too are a range of other buffers and assistance. In particular, we should invest in drug and alcohol treatment programs which have been found to pay off fiscally. Additionally, recreational and mentoring programs for people in their late teen and early twenties should be instituted.

Drug and alcohol problems can seriously interfere with employment and family life among young adults. Colleges and universities try to deal with these problems among their students. Middle-class parents also have access to resources in order to address such problems. Similar resources should be directed to less advantaged youth.

Finally, reproductive health services explicitly designed for young men and women are essential. They are necessary not only to prevent unwanted pregnancies which are at epidemic proportions, but also to stem the equally troubling epidemic of sexually transmitted diseases, particularly AIDS, among young adults. These investments in the skills, education, and health of young adults will be costly, but so is higher education. Billions of public dollars flow into our universities and colleges every year.

We also need to expand systems of income support that can support two-parent as well as single-parent families in poverty. Chief among these is the Earned Income Tax Credit, a subsidy to working parents. A refundable tax credit, perhaps one scaled to family income so that lower-income families received a larger credit, would also direct resources toward two-parent as well as single-parent families with children. For two-parent families with low or no earnings, however, AFDC should be as readily available as it is to single-parent families. Under current law, a couple with low earnings will not receive much assistance from the Earned Income Tax Credit and will be ineligible for AFDC if the primary worker works more than 100 hours in a month, no matter how low the parents' earnings. These rules should be changed.

If low-income families are to escape poverty through employment, subsidized and high quality child care must become widely available. At present, the people at the top of the priority list for this highly coveted public benefit are women who have recently left AFDC for a job or are combining AFDC and employment after having been on welfare for quite a long time.

The supply of public dollars for this service falls far short of the demand, and consequently, many women who would want to work cannot do so. In particular, two-parent families find it much harder to qualify for subsidized services than do single-parent families.

Child support policies also need reform. If young men became convinced that they would have to pay child support for every child they conceived, some might become more interested in preventing unplanned pregnancies. The reforms instituted under the 1988 Family Support Act were intended to increase the number of child support awards, particularly for unmarried mothers, to reduce judicial discretion in setting the levels of awards, and to increase collections. The next set of changes should focus on reducing the considerable variation among states in award levels, particularly the payments expected from low-income fathers. These policies should permit mothers on AFDC to receive a larger fraction of the child support payment. They should also develop better strategies to deal with the difficulties facing low-earning men in arrears for large amounts of child support.

In sum, what is needed is a comprehensive strategy to improve the lives of low-income young adults with a special focus on reproductive health, mental health, and substance abuse services. We need to implement changes in our means-tested programs for employment assistance, income supplementation, and child care that would make marriage look like a viable economic solution for these young adults. Finally, child support laws must be changed to impose fair, manageable, and unavoidable payments on noncustodial fathers.


Footnotes

1Graduate School of Public Policy, University of California Berkeley. The author would like to thank Kristin Luker, Joan Hollinger, Rondi Gilbert, and Patty Dacey for helpful comments on earlier drafts.
Back

2Steve W. Rawlings, U.S. Bureau of the Census, Current Population Report P20-477, Household and Family Characteristics: March 1993, at 92 tbl. 10 (1994) [hereinafter Rawlings 1993].
Back

3U.S. Bureau of the Census, Statistical Abstract of the United States: 1994, at 471 tbl. 718, 479 tbl. 736 (1994) [hereinafter Statistical Abstract].
Back

4Id. at 80 tbl. 100.
Back

5See generally id. Among women older than 25 the fraction of births that are non-marital hovers around 17% for all age groups. Id.
Back

6U.S. Bureau of the Census, Current Population Reports P23-181, Households, Families and Children: A 30-Year Perspective 6 fig. 2 (1992) [hereinafter A 30-year Perspective].
Back

7Id.
Back

8See generally Alan Guttmacher Institute, Sex and America's Teenagers (1994); Jane Mauldon & Kristin Luker, Contraception among America's Teens: The News Is Better than You Think (1995) (on file with the Berkeley Institute for Research on Policy Solutions, Working Paper PS10, Graduate School of Public Policy, University of California Berkeley).
Back

9A 30-Year Perspective, supra note 5, at 10 fig. 6.
Back

10Women between 15 and 34 represent 94% of nonmarital births and 90% of all births, so if we had included the full age-range 15-45 the results would not differ much from those shown. See id. at 11 fig. 7.
Back

11See generally Rawlings 1993, supra note 1; Steve W. Rawlings, U.S. Bureau of the Census, Current Population Reports P20-388, Household and Family Characteristics: March 1983 (1984) [hereinafter Rawlings 1983]; Arlene F. Saluter, U.S. Bureau of the Census, Current Population Reports P20-478, Marital Status and Living Arrangements: March 1993 (1994); Arlene F. Saluter, U.S. Bureau of the Census, Current Population Reports P20-389, Marital Status and Living Arrangements: March 1983 (1984); U.S. Bureau of the Census, Current Population Reports P20-258, Household and Family Characteristics: March 1973 (1974) [hereinafter Household and Family characteristics 1973]; U.S. Bureau of the Census, Current Population Reports P20-255, Marital Status and Living Arrangements: March 1973 (1974). The data describe only the civilian non-institutional population, and thus exclude people in prison or in the military. The data slightly understate the prevalence of unmarried parents living with their children, in that unmarried couples had to select one person to be identified as the household head and parent of the child(ren), and the other parent's relationship to the children went unrecorded. These two-parent families show up in the data as one-parent families and the second parents show up as living neither with spouse nor with children. Fortunately, from other studies we know the number of these families is small. See Larry L. Bumpass & R. Kelly Raley, Redefining Single-Parent Families: Cohabitation and Changing Family Reality, 32 Demography 97, 97-109 (1995).
Back

12These changes would be even greater except that the median age of this population has risen as the baby boom cohort has aged: the median age of 18-45 year-olds was 28 in 1973, 29 in 1983 and 32 in 1993. See Rawlings 1993, supra note 5; Rawlings 1983, supra note 10; Household and Family Characteristics 1973, supra note 10.
Back

13

Table 3:
Living Arrangements of Adults of Childbearing Age, 1993:
By Race/Ethnicity and by Age
[Space] [Space]
Age 18-45,
White Non-
Hispanic
Age 18-45,
African-
American
Age 18-45,
Hispanic
Age 18-45,
All Races
Age 18-24,
All Races
Women [Space]
Married with chiildren
Single parent
[Space]
44%
12%
23%
39%
44%
20%
41%
15%
13%
13%
Married, no chiildren
Unmarried and childless
(family-free)
[Space]
15%

29%

6%

32%

10%

26%

13%

30%

10%

64%

Total [Space]
100%
100%
100%
100%
100%
Percent of women living
unmarried, childless and
apart from parents
[Space]
16%
14%
13%
15%
21%
Percent of women living
with children
[Space]
56%
62%
64%
57%
26%
Men [Space]
Married with chiildren
Single parent
[Space]
39%
2%
25%
3%
38%
3%
37%
2%
7%
1%
Married, no chiildren
Unmarried and childless
(family-free)
[Space]
13%

46%

7%

65%

9%

51%

13%

47%

6%

86%

Total [Space]
100%
100%
100%
100%
100%
Percent of men living
unmarried, childless and
apart from parents
[Space]
24%
32%
29%
25%
26%
Percent of men living
with children
[Space]
42%
28%
40%
40%
8%

Back

14In 1993, there were 5.9 million one-parent families with children maintained by white women, 3.1 million maintained by African American women and 1.3 million headed by Latinas. Statistical Abstract, supra note 2, at 62 tbl.72. In 1992, there were 467,500 births to unmarried white women, 449,000 to unmarried black women, and 251,500 to unmarried Latina women. These births constituted, respectively, 19% of births to white women, 39% of births to Latinas, and 68% of births to African American women.
Back

15It is not possible to establish from these data what fraction of the men who are not with children of their own actually have biological offspring, but certainly a substantial fraction of them do; the children of single mothers obviously have fathers somewhere.
Back

16See Carol B. Stack, All Our Kin: Strategies for Survival in a Black Community (1974).
Back

17In 1980, among young adults aged 16-24, 21% were in college, 22% were in some other schooling (including high school), and 57% were not enrolled in school. In 1992, 28% of the age-group were in college, 24% in other schooling and 49% were not enrolled in school. See generally Statistical Abstract, supra note 2, at 400 tbl. 623.
Back

18See generally Statistical Abstract, supra note 2 at 400 tbl. 623.
Back

19See William T. Grant Foundation, The Forgotten Half: Pathways to Success for America's Youth and Young Families (1988).
Back

20Suzanne M. Bianchi, Children of Poverty: Why Are They Poor?, in Child Poverty and Public Policy 91, 91-125 (Judith A. Chafel ed., 1993).
Back

21See William T. Grant Foundation, supra note 18.
Back

22See James A. Sweet & Larry L. Bumpass, American Families and Households (1987).
Back

23See A 30-Year Perspective, supra note 5, at 40 fig. 33.
Back

24Frank F. Furstenberg, Jr., Teenage Childbearing Reconsidered, in Kids Having Kids: The Consequences and Costs of Teenage Childbearing in the United States (The Robin Hood Foundation, forthcoming 1996).
Back

25Deanna L. Pagnini & Ronald R. Rinfuss, The Divorce of Marriage and Childbearing: Changing Attitudes and Behavior in the United States, 19 Population & Dev. Rev. 331, 331-47 (1993).
Back

26See Charles Murray, Losing Ground: American Social Policy, 1950-1980 (1984).
Back

27See David Ellwood & Mary Jo Bane, The Impact of AFDC on Family Structure and Living Arrangements, 7 Research in Labor Economics 137, 137-208 (Richard Ehrenberg ed., 1985); Robert Moffitt, The Effect of the U.S. Welfare System on Marital Status, 41 J. Pub. Econ. 101, 101-24 (1990); Robert Plotnick, Welfare and Out-of-Wedlock Childbearing: Evidence from the 1980s, 52 J. Marriage & Fam. 735, 735-46 (1990).
Back

28Lyle Groeneveld, et al., Final Report of the Seattle Denver Income Maintenance Experiment, in 1 Design & Results (1983).
Back

29

Table 4
Year
[Space]
Average Monthly AFDC
Benefit per Family (In
1993 Dollars)
[Space]
Births per 1,000
Unmarried Women
1970
1975
1980
1985
1990
1992

Ratio: 1992: 1970

[Space]
$663
$564
$480
$455
$428
$394

0.59

[Space]
25.4
24.5
28.4
32.8
43.8
45.2

1.78

Back

30Hilary Williamson Hoynes, Does Welfare Play Any Role in Female Headship Decisions? (1995) (unpublished paper, on file with Department of Economics, University of California Berkeley).
Back

31Id.
Back

32California had the third highest rate of births per 1,000 population in the nation in 1990-1992, exceeded only by Alaska and Utah. It also had the third highest unemployment rate.
Back

33Cara Lesser, Can They Go Home Again? Requiring Minor Parents to Live at Home Is Unlikely to Reduce Welfare Dependency (1994) (unpublished paper prepared for U.S. General Accounting Office, on file with the Graduate School of Public Policy, University of California Berkeley).
Back

34Currently the welfare program claims the entire child support payment from the father except for $50 a month, an amount that has not changed in 12 years.
Back