Getting to Know the Future Customers of the Office of Child Support
Projections Report for 2004 and 2009
Table of Contents
- Outline of the Report
- Primary Findings and Results
- Demographic Characteristics of Custodial Parents
- Economic Characteristics of Custodial Parents
- Demographic Characteristics of Nonparent Custodians
- Economic Characteristics of Nonparent Custodians
- Demographic Characteristics of Noncustodial Parents
- Demographic Characteristics of Children Eligible for Support
- Hispanic and Asian Child Support Populations
- Hispanic Child Support Populations
- Hispanic Custodial Parents
- Hispanic Nonparent Custodians
- Asian and Pacific Islander Child Support Populations
- Asian and Pacific Islander Custodial Parents
- Underlying Population Trends
- Appendix A. Detailed Tables
- Appendix B. Models
- Appendix C. Demographic Differences Among Custodial Parents
Hans P. Johnson
The rapid growth of the child support population of the United States over the past decade is projected to slow between 1998 and 2009. Our projections suggest that the growth of the child support population over the next ten years will continue to outpace the total population growth in the United States, but will not be as remarkable as the growth of the previous ten years. In 1998, the child support population of the United States reached an estimated 62 million people, a 32 percent increase from 1988. By 2009, we project that the child support population of the United States will reach 72 million people, a 15 percent increase from 1998. More than 19 million custodians will be the primary caregivers for 30 million children, with a projected 22 million noncustodial parents.
The rate of growth of the child support population is projected to slow for three reasons. First, the population of the United States is aging out of the prime parenthood years. Over the next ten years, the large cohorts of baby boomers will begin to be replaced by the much smaller cohorts of the baby bust. Second, nonmarital fertility has been fairly constant over the past five years and is projected to remain so (albeit at a very high level by historic standards). And third, divorce rates in the United States have declined from their peak in the early 1980s.
Despite the slowing in the overall rate of growth of the child support population, certain segments of the population will continue to experience rapid growth rates. For example, the number of custodial fathers will increase dramatically, as will the number of nonparent custodians (including grandparent custodians), and never-married custodians. Each of these subgroups is projected to grow substantially faster than the overall child support population. Likewise, Hispanics and Asians will experience faster growth rates than any other race or ethnic group. Finally, growth will be stronger in the early part of the projection period, from 1998 to 2004, than in the latter part of the projection period from 2004 to 2009.
In this report we present the numbers of custodial parents, nonparent custodians, and noncustodial parents projected for the years 2004 and 2009, examining the characteristics of each group as well as the underlying demographic and economic trends driving these projections. As the future of the child support population becomes clearer, so will the ability to anticipate the needs and to plan for the changes. Anticipating the future can mean changing the future for thousands of children.
Children need to be financially supported by their parents. Enforcing child support is a crucial component in the effort to maintain and raise the economic and social welfare of children in the United States. The number of children residing with only one biological parent has increased significantly in the past ten years. Persistently high divorce rates and high levels of nonmarital births suggest that the proportion of children living in a household without both parents will only increase. Of course, meeting a child’s financial needs can be a daunting task for a single parent or a combined family. More families will need help and, as welfare reform has made public assistance less available, more families will come to rely on child support payments. Making sure that noncustodial parents contribute financially to the upbringing of their children will not only further the maintenance of parental bonds but will help ensure that children in poor families are kept out of poverty.
In order to anticipate the needs of families, it is necessary to calculate the future growth of the child support population—the number of children, the number of custodians, and the number of noncustodial parents. In this report, we develop projections of the child support population in 2004 and in 2009. These projections build on the estimates developed for 1988 and 1998 in the baseline report, “Getting to Know Child Support’s Future Customers: Baseline Estimates for 1988 and 1998.” Thus, altogether, we have estimates and projections that span more than twenty years, a period of remarkable change in the child support population of the United States.
Of course, projecting future populations entails some uncertainty. Human behavior is not always easy to understand in the present, let alone in the future. In general, demographers rely on historical trends to forecast future populations. For example, the United States Census Bureau uses patterns and trends in fertility, mortality, and migration to forecast future populations. The past is a better predictor of the short-range future than it is of the long-range future, and thus projections for shorter time horizons (five or ten years) are likely to be more accurate than projections for longer time horizons. Similarly, projections for large areas with large populations tend to be more robust than projections for small areas with small populations. For example, because national population projections do not require projections of internal migration, projections for the entire United States are more robust than projections for individual states or regions.
The projections we have developed for the child support population are short-term projections relying on large data sets and established trends covering broad regions. Unless there are dramatic and unexpected changes in the social and economic climate, we expect our projections to be robust.
Outline of the Report
In the second section of this report we present our primary findings, namely that while the overall number of adults and children in the child support population will continue to increase, the overall rate of increase will slow. Certain segments of the child support population will continue to experience strong growth, and we identify those segments. In section three we discuss the Hispanic and Asian child support populations. In section four, we discuss the underlying trends that drive the population projections. Section five provides details of our methodological approach and includes sensitivity tests of the assumptions employed in the model. We conclude the report in section six.
Primary Findings and Results
- Demographic Characteristics of Custodial Parents
- Economic Characteristics of Custodial Parents
- Demographic Characteristics of Nonparent Custodians
- Economic Characteristics of Nonparent Custodians
- Demographic Characteristics of Noncustodial Parents
- Demographic Characteristics of Children Eligible for Support
The child support population in the United States is projected to reach 72 million by 2009, a 15 percent increase over the 1998 estimate of 62 million. By 2009, the child support population will include 17 million custodial parents, almost 3 million nonparent custodians, 22 million noncustodial parents, and over 30 million children eligible for support. Although the overall number of custodians and children in the child support population remains high, the projections for the year 2009 indicate a significant slowing of the growth of the child support population. Still, certain segments of the child support population will continue to experience strong rates of increase. For example, the number of nonparent custodians will increase dramatically, as will the number of custodial fathers and never-married custodians. Other groups, such as divorced custodial parents, will experience little growth.
We will first discuss the results of our projections for each of the components of the child support population: custodial parents, nonparent custodians, noncustodial parents, and children eligible for child support. Throughout the report we will indicate the underlying historical trends and economic conditions driving these projections. 
The child support population of the United States encounters singular economic challenges. Characterized by higher rates of poverty and lower levels of education than the rest of the population, the child support community faces hardships that do not necessarily derive from the burdens of supporting children. The more information we have about the demographic trends and economic needs of the families requiring child support the better informed our programs and policy debates will be.
Demographic Characteristics of Custodial Parents
A custodial parent is a parent who lives with and cares for at least one child in a household separate from that of the other biological or legal parent. We project that the number of custodial parents will reach 16.1 million by 2004 and 16.8 million by 2009. Thus, in absolute terms, the number of custodial parents will increase moderately. As shown in Figure 2.1, however, average annual changes in the number of custodial parents will be substantially lower in both projection periods (1998-2004 and 2004-2009) than during the previous ten years. The rate of growth of the child support population has slowed for three reasons. First, the population of the United States is aging out of the prime parenthood years. Over the next ten years, the large cohorts of baby boomers will begin to be replaced by the much smaller cohorts of the baby bust. Second, nonmarital fertility has been fairly constant over the past five years and is projected to remain so (albeit at a very high level by historic standards). And third, divorce rates in the United States have declined from their peak in the early 1980s. Even though the rate of growth has slowed, the number of custodial parents in the United States will continue to rise at a faster rate than the nation’s overall population growth rate between 1998 and 2009. The number of custodial parents will increase by 14 percent, while the nation’s population will grow by only 10 percent.
Figure 2.1. Average Annual Change in the Number of Custodial Parents
The projections of the numbers of custodial parents and the analysis of their characteristics, combined with the estimates developed in the baseline report, portray a population experiencing remarkable growth and change. Between 1988 and 2009, the number of custodial parents in the United States will have increased by almost 50 percent. Not only is the number of custodial parents changing rapidly, so too are their characteristics. For example, in 1988 the typical custodial parent was a 32-year-old divorced mother living in a central city of a metropolitan area; by 2009, the typical custodial parent will be a 35-year-old, never-married mother living in a suburb. Likewise, there have been dramatic increases in the number of custodial fathers and in the number of Hispanic custodial parents. Below, we discuss the increases in the number of custodial fathers and the geographical distribution of custodial parents.
One of the most noteworthy changes in the child support population of the United States has been and will continue to be the large increases in the number of custodial fathers. From 1988 to 2009, the number of custodial fathers is projected to increase by 33 percent. In contrast, the number of custodial mothers will increase by only 11 percent. These large increases in the number of custodial fathers are a reflection of the increasing probabilities of being a custodial parent among men. Figure 2.2 shows trends in the probability of being a custodial parent by gender, with the 1988 probabilities by gender indexed to 100. Among adult men, the probability of being a custodial parent was 1.4times higher in 1998 than in 1988.
Figure 2.2. Indexed Trends in the Probability of Being a Custodial Parent
During the late 1990s, as the probabilities of being a custodial parent were increasing for men, the probabilities were decreasing for women. In particular, white men, either divorced or separated, were much more likely to be custodial fathers in 1998 than in 1988. Despite these increases, it is important to recall that women remain much more likely to be custodial parents. In the late 1990s, almost 12 percent of adult women in the United States were custodial parents, compared to less than 3 percent of adult men. We project that in 2009 four out of every five custodial parents will be a custodial mother.
Figure 2.3. Percent Change in Custodial Parents by Race/Ethnicity
The race and ethnic composition of custodial parents will change in the coming decade. Hispanics and Asians have been the fastest growing groups and are projected to remain so through 2009 (see Figure 2.3). Not surprisingly, these increases in the number of Hispanic and Asian custodial parents are driven primarily by the overall population growth of these groups within the United States. In contrast, whites and African Americans will experience little growth; for some key age groups the number of white and African American custodial parents actually declines. Furthermore, among African American adults, the probability of being a custodial parent declined from the mid 1990s to the late 1990s (from almost 18 percent to less than 16 percent) and is projected to continue to decline slightly. Still, African Americans are much more likely to be custodial parents than other groups, and, despite increasing probabilities of being a custodial parent, Asians and Pacific Islanders will remain the group least likely to be a custodial parent. By 2009, whites will still comprise a majority of custodial parents (56 percent), African American custodial parents are projected to comprise 22 percent, Hispanics 18 percent, Asians and Pacific Islanders 3 percent, and all others 4 percent.
The percentage of custodial parents living in specific regions of the United States will have changed little between 1988 and 2009. What little change does occur reflects overall trends in regional population growth. While the number of custodial parents is projected to increase in each region, the rate of increase in the slowest growing region, the Northeast (7 percent increase between 1988 and 2009), will be much lower than in the fastest growing region, the West (23 percent increase). By 2009, the West will replace the Midwest as the second most populated region for custodial parents. The South will remain home to almost 40 percent of custodial parents; it has a larger population than any other region in the United States and the likelihood of being a custodial parent is greater in the South than in other regions. Although regional percentages do not change substantially, this does not mean that there has been no geographical change in the child support population, only that the change cannot be detected at the regional level. If we were to look at each individual state we would no doubt find greater differences in the growth and decline of the child support populations of specific states.
The regional stability in the distribution of custodial parents further belies a dramatic geographical shift from urban to suburban areas. In 1988, central cities of metropolitan areas were the leading places of residence of custodial parents. Thirty-eight percent of all custodial parents lived in these highly urban environments. Thirty-seven percent lived in other parts of the city. By 2009 the suburbs will have replaced the central city as the leading location of custodial parents. Forty-nine percent of custodial parents will live in the suburbs and only 29 percent will live in the central city. The child support population is thus no longer primarily a phenomenon of the central city but rather a suburban phenomenon. This twenty-year trend occurs for at least two reasons. First, the population of the United States has gravitated toward the suburbs; more people live in the suburbs than in urban city centers. Second, the incidence of nonmarital childbearing has increased throughout the population, becoming more widespread among all races and in all regions, including the suburbs.
Although the location of the custodial households has changed, there is virtually no change in the number of children per custodial household. In 1998 each household including a custodial parent had an average of 1.82 children. In 2009 the figure is projected to be 1.81. Thus the number of children per household has neither increased nor decreased dramatically in years. This constancy in the number of children per custodial household reflects the stable levels of fertility in the United States over the last twenty years, even though substantial changes have occurred in terms of the marital status of parents and the presence of both parents in the same household.
Clearly the most dramatic demographic change projected for custodial parents in 2009 is the increase in the number of never-married custodial parents. The number of never-married custodial parents increased from 3.1 million in 1988 to 4.9 million in 1998, and is projected to rise to 6.6 million in 2009. While the percentage of divorced custodial parents in the total custodial parent population is projected to decline from 49 percent in 1988 to 37 percent in 2009, the number of custodial parents that have never been married is projected to increase from 27 percent in 1988 to 39 percent in 2009. These dramatic changes reflect increases in the number of nonmarital births in the United States over the past fifteen to twenty years and declines in the divorce rate. Though nonmarital fertility leveled off a bit between 1993 and 1998, it did so at a high level. That high level is expected to continue over the next ten years. (See section four for a more complete discussion of underlying trends). In 1975, 14 out of every 100 babies were born to unmarried women. In 1995, 32 out of every 100 babies were born to unmarried women. Thus, new cohorts born today are more likely than those born twenty years ago (and as likely as those born five years ago) to be born to unmarried women. As older cohorts of children age out of the child support system, the younger cohorts replacing them are more likely to be born to unmarried women. Hence, high nonmarital fertility rates will continue to lead to substantial increases in the child support population.
Economic Characteristics of Custodial Parents
Forecasting the economy is more difficult than forecasting the demographic characteristics of a population. The United States has sustained record economic growth during the mid to late 1990s. Economists believe that a downturn will eventually occur, but it is hard to predict with any accuracy when a recession will hit. In order to keep our economic projections reliable, we have compared similar points in the business cycle, such as the strong economies in the late 1980s and the late 1990s. If current trends continue—if we do not experience a recession in 2004 or 2009—the socioeconomic outlook for custodial parents is mostly favorable.
Educational attainment levels for custodial parents jumped considerably between 1988 and 1998 and are expected to continue to improve at a more moderate rate as we approach 2009. In 1988, fewer than one out of three custodial parents had attended or graduated from college. By 1998, 43 percent had been to college, and, by 2009, nearly one out of every two custodial parents will have graduated from or attended college. In contrast, 16 percent will have never completed high school. These higher levels of educational attainment reflect increases that are occurring throughout the United States, with each successive generation completing more schooling than preceding generations.
From 1988 to 1998, labor force participation rates for custodial parents increased substantially. By 1998, almost 80 percent of custodial parents were in the labor force (either employed or looking for work). The decline in the share of custodial parents not in the labor force was especially dramatic in the last part of the 1990s, undoubtedly due to a continuing strong economy and also, at least in part, due to welfare reform. We project that this trend toward increasing labor force participation will continue at a slower pace, so that by 2009 about 84 percent of custodial parents will be in the labor force. These projections assume that 2009 will not be a recession year. Custodial parents are not only of prime childbearing and childrearing years, they are also of prime working age.
Figure 2.4. Poverty Rates for Custodial Parents
We also project that poverty rates for custodial parents will continue their long- term decline, though remain at high levels. One out of every three custodial parents fell below the poverty level in 1988. With the recession of the early 1990s, those rates reached 36 percent. The strong economy throughout the rest of the 1990s led to tremendous declines in poverty rates among custodial parents, so that by 1999 the poverty rate had fallen to 28 percent (see Figure 2.4). Still, this is more than double the nation’s overall poverty rate of 12 percent for the same year. By 2009, we project that one out of every four custodial parents will live in poverty, a projection that assumes a slowing of the rate of decline in poverty. As in the past, certain groups will experience much higher levels of poverty than others. In particular, never-married custodial parents will continue to experience poverty rates about seven times higher than married custodial parents.
Forecasting the receipt of public assistance is even more difficult than developing projections for other economic indicators. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) and the strong economy have led to dramatic declines in public assistance utilization among all segments of the population, including custodial parents (see Figure 2.5).
Figure 2.5. Public Assistance Utilization Rates Among Custodial Parents
Whether these declines continue into the future depend on the economy and future welfare policy. Indeed, the relative importance of welfare reform policies, on the one hand, and the economy, on the other hand, in reducing welfare caseloads is the subject of much debate. In a review of the studies for congressional testimony, Moffit (1999) concluded that research suggests just over half of the 1993-1996 decrease in caseloads could be explained by the declining unemployment rate. However, up through 1996, it appeared that the welfare caseload was becoming more sensitive to the business cycle over time; Moffit believes that welfare reform may in turn reduce the sensitivity of the caseload to the economy. In their research, Ziliak and Figlio find that a recession similar to the last one could increase welfare rolls by up to 8 percent (a small increase relative to the current downturns). Mayer (2000) also surveyed the literature on welfare caseloads and reached five main conclusions: 1) the best single predictor of caseloads in any given year is their level in the prior year; 2) a one point increase in the unemployment rate would increase caseloads by 4 to 6 percent; 3) higher benefits increase caseloads (though there is little consensus on the role of other parameters such as work requirements); 4) the number of single mothers is an important determinant of caseloads; and 5) models explain only a small fraction of the variance in caseload changes and do a worse job for recent years. Given these difficulties in understanding recent declines in public assistance use, let alone projecting such use for the future, we have simply held public assistance utilization rates at 13 percent for custodial parents.
Demographic Characteristics of Nonparent Custodians
A nonparent custodian is the primary caregiver of a child who lives without a biological or legal parent in the household. Foster parents are not included in this category. Between 1998 and 2009, we project that the number of nonparent custodians will increase from 2.4 million to 2.9 million. Although the number of nonparent custodians is increasing at a less rapid rate than in the past, the growth rate for nonparent custodians is faster than the growth rate of custodial parents (23 percent versus 14 percent), and more than double the growth rate of the entire United States population. The determinants of this slowdown are the same determinants that drive the slowing of the rate of increase in the child support population as a whole: declining rates of divorce, the aging of the baby boom, and the relative stability in nonmarital fertility rates. The pool of children eligible for child support will not increase as quickly in the next ten years as in the past ten years.
Just as the percentage of men who are custodial parents has increased in the past ten years, so has the number of men who are nonparent custodians. In 1988, 48 percent of nonparent custodians were men. By 1998, the number had risen to 53 percent. We project that the percentage of male nonparent custodians will change little to 2009, increasing to 55 percent.
Both male and female nonparent custodians tend to be older than their custodial parent counterparts. In 1998 four out of ten nonparent custodians were aged 50 or older. Almost half were grandparents. Many nonparent custodians are baby boomers, and as baby boomers age, so will nonparent custodians. From 1988 to 2009, the proportion of nonparent custodians less than 30 years of age will have decreased from 28 percent to 21 percent. From 1998 to 2009, the proportion of nonparent custodians aged 60 and over will have increased from 20 percent to 24 percent. To the extent that many nonparent custodians are grandparents, the increase in age of nonparent custodians also reflects increasing generational lengths as the average age at birth for parents has increased over the past two decades.
The race and ethnic composition of these older nonparent custodians will continue to change at a moderate pace in the next ten years. Most dramatic will be increases in Hispanic nonparent custodians, increasing from 18 percent of all nonparent custodians in 1998 to a projected 23 percent in 2009. The large growth in the number of nonparent custodians who are Hispanic and Asian reflects the rapid population growth for these groups in the United States. In fact, the number of Asian and Hispanic children will grow even faster than the number of Asian and Hispanic adults. (See section three for an extended discussion of the Asian and Hispanic child support population.) As shown in Figure 2.6, Asian and Hispanic nonparent custodians experienced the fastest growth between 1988 and 1998, and, though their growth rates will slow, they will continue to be the fastest growing groups to 2009.
Figure 2.6. Percent Change in Nonparent Custodians by Race and Ethnicity
In the late 1980s and early 1990s, the nonparent custodial population of the United States was concentrated in the South, with over half of all nonparent custodians living in the South in 1988. By 2009, the regional distribution of the nonparent custodial population of the United States is projected to be much closer to that of the custodial parent population, with 43 percent of nonparent custodians residing in the South, 20 percent in the Midwest, 15 percent in the Northeast, and 22 percent in the West. These trends partly reflect regional population growth, but also the increased probability of being a nonparent custodian in all regions of the country except the South.
Like custodial parents, nonparent custodians are increasingly found in the suburbs. By 2009, almost four in every ten nonparent custodians will reside in the suburban portion of a metropolitan area. In contrast, in 1988 the leading location of residence for nonparent custodians was the central city. The shift toward the suburbs is not quite as dramatic for nonparent custodians as it is for custodial parents. In part, this is because nonparent custodians are already more likely than custodial parents to live completely outside of urban and suburban areas. Twenty-seven percent of nonparent custodians will live in a non-metropolitan area in 2009; twenty-two percent of custodial parents will make this same choice. Again, the child support population will continue to increase in suburban portions of metropolitan areas simply because more people are moving to and living in the suburban United States.
The demographic trend away from the central cities has not affected the average number of children in the nonparent custodial households. From 1998 to 2009 the number of children per custodial household is projected to drop negligibly, from 1.82 to 1.81. Thus, even though there are more people deciding to have children without getting married, the number of children per custodial household remains remarkably consistent.
Nonparent custodians, as with custodial parents, are increasingly less likely to have ever married in the twenty-year period we are examining. While only 27 percent of custodial parents were never married in 1988, the proportion is projected to increase to 39 percent in 2009. For nonparent custodians, the percentage who have never married is expected to increase as well, from 22 percent to 29 percent over the same time period. Thus, taken together, the rise in the number of never-married adults raising children is quite striking. In one respect, the change in marital status of nonparent custodians is more dramatic than the change in marital status of custodial parents. One of every four custodial parents are married, both in the estimates for 1988 and in the projections for 2009. In contrast, 50 percent of nonparent custodians were married in 1988, but, by 2009, only 39 percent will be married and 61 percent will be either divorced, separated, never-married, or widowed. Thus the percentage of nonparent custodians that are married will decrease significantly while the percentage of custodial parents that are married is projected to remain relatively constant. Of course, the increase in the proportion of never-married custodial parents remains overall more striking than the decrease in the number of married nonparent custodians because there are far more custodial parents in the child support system than there are nonparent custodians.
Economic Characteristics of Nonparent Custodians
Despite the rapid increase in divorced and separated nonparent custodians, the socioeconomic outlook for nonparent custodians is one of general improvement. If current trends continue, nonparent custodians will be less likely to live in poverty and will be better educated than in the past.
Educational attainment levels will increase for nonparent custodians. By 2009, more than one in every three will have either attended or graduated from college. This represents a doubling in the proportion of better educated nonparent custodians from 1988. The improvements in educational attainment levels of nonparent custodians is at least in part a cohort effect. Younger adults in the United States tend to be better educated than older adults. Over time, these younger, better- educated cohorts will replace the older, less-educated nonparent custodians of today. Still, the distribution of educational attainment levels of nonparent custodians is projected to be bipolar, with large numbers of very poorly educated nonparent custodians projected to be of similar size to those at the other end of the education spectrum. We project that by 2009, one of every three nonparent custodians will have failed to complete high school, a rate much higher than that of the general adult population in the United States.
Figure 2.7. Poverty Rate for Non-Parent Custodians
For nonparent custodians, we also project that poverty rates will decline, though still remain at high levels. From the late 1980s to the late 1990s, poverty rates for nonparent custodians declined from 35 percent to 28 percent (see Figure 2.7). Our projections for 2009 anticipate that about one in every four nonparent custodians will live in poverty, a substantial improvement over the 1988 level of 35 percent. These projections assume that long-term trends will continue. We are comparing similar periods with strong economies (the late 1980s and late 1990s). However, if 2004 or 2009 proves to be a recession year, then the projected poverty rates will be too low.
Demographic Characteristics of Noncustodial Parents
Noncustodial parents are those parents who do not live with and are not the primary caregivers of their biological or legal children. Noncustodial parents could be either the former spouse of a custodial parent, or the parent of a child whose primary caregiver is neither the mother nor the father. As shown in Table 2.1, the number of noncustodial parents is projected to surpass 22 million by 2009. The rate of growth in the noncustodial parent population of the United States is expected to slow to 16 percent between 1998 and 2009, half the rate of increase in the previous ten years. This slower growth is the direct result of the slowing in the rate of increase in custodial parents and nonparent custodians discussed earlier. As in the past, the number of noncustodial parents that are the parents of children for whom neither parent is the primary caregiver is projected to grow at a faster rate than the number of noncustodial parents who are the spouses of custodial parents (23 percent versus 14 percent between 1998 and 2009).
|Spouses of custodial parents||11,574,000||14,667,000||16,142,000||16,786,000|
|Parents of children not living with either parent||2,991,000||4,536,000||5,065,000||5,575,000|
Figure 2.8. Percent of Noncustodial Parents that are Mothers
The vast majority of noncustodial parents are fathers, though the proportion of noncustodial parents that are mothers is increasing (see Figure 2.8). Between 1998 and 2009, the number of noncustodial mothers is projected to increase 28 percent while the number of noncustodial fathers is projected to increase only 13 percent. The more rapid growth for noncustodial mothers is the direct consequence of increasing probabilities of fathers serving as primary caregivers. Still, it is important to note that the vast majority of custodial parents are mothers, and will be so in the future. Indeed, our estimates and projections suggest that noncustodial fathers can expect their children to be cared for by the mother, while only about half of noncustodial mothers can expect their children to be cared for by their fathers. Our projection for 2009, for example, indicates that for 83 percent of noncustodial fathers, the mothers of their children are the primary caregivers; on the other hand, only about half (53 percent) of noncustodial mother’s children live with the children’s father.
Demographic Characteristics of Children Eligible for Support
The children eligible for child support are those children who have at least one parent not living with them. Again, this category excludes children living in group quarters and foster children. The number of children in the child support system grew dramatically from 1988 to 1998, increasing by 17 percent. The number of children in the United States as a whole increased by only 10 percent. Thus the rate of growth for children in the child support system was almost twice the rate of growth of the number of children in the general population. In the next ten years, the rate of increase will slow, but the overall number of children in the child support system will continue to rise. The U.S. Census Bureau projects that there will be 72 million children in the United States in 2009, and we project that 30 million children will live in households eligible for child support (see Figure 2.9). The overall number of children in the United States increases in large part due to cohort effects—a larger number of adults reaching childbearing age—not because the average adult is having more children than in the past. The United States population as a whole continues to increase, despite a fertility rate that hovers just below replacement rate, because of these cohort effects and also because of immigration.
Figure 2.9. Number of Children in the United States and Children in Child Support Households (in thousands)
Still, the number of children in the United States is projected to increase only 3 percent between 1998 and 2009, with most of that growth occurring between 1998 and 2004. These small increases in the number of children in the United States are accompanied by, and indeed largely determine, the slowing of the rate of increase in the child support population. Between 1998 and 2004, we project that the number of children eligible for child support will increase 9 percent, and between 2004 and 2009 we project that the increase will be less than 4 percent. The faster growth of the child support population than the number of children in the United States can be understood as a consequence of historic and ongoing changes in nonmarital fertility and divorce. In 2009, a 17-year-old child will be almost three times as likely to have been born to an unmarried mother as a 17-year-old child was in 1988. Similarly, a 17-year-old in 2009 would have been raised in an era of substantially higher divorce rates than a 17-year-old in 1988. It is these older children, those between the ages of 14 and 17, who will experience the largest increases in population between 1998 and 2009. These older children are the last of the children of the baby boomers.
Hispanic and Asian Child Support Populations
- Hispanic Child Support Populations
- Hispanic Custodial Parents
- Hispanic Nonparent Custodians
- Asian and Pacific Islander Child Support Populations
- Asian and Pacific Islander Custodial Parents
Despite a general slowing of the increase in child support populations in the United States, we project that Hispanic and Asian child support populations will continue to experience rapid increases. Because the child support population mirrors the ethnic composition of the nation, the large increases in the Asian and Hispanic child support populations reflect the growth of these minorities in the United States.
As noted earlier, the number of children in the United States is expected to increase little between 1988 and 2009, increasing only about 3 percent. However, as shown in Figure 3.1, growth rates will be quite strong for Hispanics and Asians and Pacific Islanders. Indeed, were it not for large increases in the number of Hispanic children (up 4.1 million between 1998 and 2009) and Asian and Pacific Islander children (up 1.0 million over the same time span), the number of children in the United States would actually decline by several million. For Hispanics, there are two reasons for these increases: immigration, and relatively high rates of childbearing. Hispanics account for the single largest group of immigrants to the United States, with especially large flows in the 1980s and 1990s. The vast majority of these immigrants are young adults of childbearing age. Most come from countries with relatively high rates of childbearing, and fertility rates of Hispanics in the United States are substantially higher than for other race and ethnic groups. For Asians and Pacific Islanders, the primary reason for increasing numbers of children is an increase in the number of women of childbearing age. This increase, in turn, is attributable to large flows of Asian immigrants over the past two decades.
Figure 3.1. Percent Change in Child Population of the United States by Race/Ethnicity, 1998-2009
Hispanic Child Support Populations
The broad ethnic category “Hispanic” comprises several distinct subgroups. We report findings for Mexicans, Cubans, Puerto Ricans, other Central and South Americans, and other Hispanics. These groups vary substantially not only in demographic and socioeconomic characteristics but also in the likelihood of being a part of the child support population. Within these subgroups, there are important differences by nativity as well. Thus, to understand more fully the Hispanic child support population, it is necessary to develop a portrait of each of these distinct subgroups. In this section, we develop these portraits.
Because sample sizes in the Current Population Surveys (CPS) are not particularly large for some subgroups, we have combined CPS samples from 1994 through 1999 (hereafter referred to as the late 1990s) to develop estimates of the characteristics of Hispanic custodial parents and nonparent custodians. By combining surveys across the late 1990s, our final sample consists of almost 8,000 Hispanic custodial parents and 1,700 Hispanic nonparent custodians (see Table 3.1). Sample sizes are sufficient to consider characteristics of all custodial parent subgroups, though the Cuban sample will not support detailed cross tabulations. For nonparent custodians, sample sizes for Cubans and other Hispanics are not large enough to support developing profiles. Thus, our profile for nonparent custodians combines all Hispanic subgroups together.
First we consider the characteristics of custodial parents. Second, we consider the characteristics of nonparent custodians.
|Nonparent Custodians||Custodial Parents|
Hispanic Custodial Parents
By 1998, the number of Hispanic custodial parents had reached 2.1 million. Almost one in every ten Hispanic adults is a custodial parent, a proportion substantially higher than among whites and substantially lower than among African Americans. The majority of Hispanic custodial parents are of Mexican descent, with Puerto Ricans and Central/South Americans together comprising about a quarter of the total (see Figure 3.2). The numerical dominance of Mexicans among custodial parents simply reflects their numerical dominance among the population of Hispanics in the United States. The majority of Hispanics in the United States are of Mexican descent, comprising six of every ten Hispanics over the age of 15. The likelihood of being a custodial parent varies substantially by Hispanic subgroup: less than 5 percent of Cubans over the age of 15 were custodial parents in the late 1990s, compared to almost 15 percent among Puerto Ricans (see Table 3.2). These differences reflect differences in divorce and nonmarital childbearing between the subgroups.
Figure 3.2. Hispanic Custodial Parents by Subgroup
|Note: Restricted to those aged 15 and over in the late 1990s|
The majority of Hispanic custodial parents are immigrants, with almost 60 percent foreign born. About four of every five Central/South American custodial parents are immigrants, whereas almost all Puerto Rican custodial parents were born in the United States (see Table 3.3). For each subgroup, immigrants were less likely to be custodial parents than U.S. born residents. Thus, the dominance of immigrants among Hispanic custodial parents is solely a function of their greater numbers in the adult Hispanic population of the United States, and not due to greater probabilities of being a custodial parent among immigrants as compared to Hispanics born in the United States.
|U.S. born||Foreign born|
|All Hispanic custodian parents||58%||42%|
Hispanics in the United States are a young population. Most Hispanic immigrants arrived in the United States in the last two decades, and came to the United States at young ages. Relatively high rates of fertility among Hispanics also leads to a young population, with many children and young adults. In the late 1990s, about two of every five Hispanic custodial parents was in his/her teens or twenties. Mexican and Puerto Rican custodial parents have the youngest age structure, while Cubans have the oldest (see Table 3.4).
|Mexican||Central/South American||Puerto Rican||Cuban||Other Spanish||Total Hispanic|
|Note: a = less than one percent|
Like other custodial parents, the vast majority of Hispanic custodial parents are custodial mothers. In the late 1990s, 85 percent of Hispanic custodial parents were custodial mothers. Custodial fathers comprised only 10 percent of Cuban and Puerto Rican custodial parents, 15 percent of Central/South American custodial parents, and 17 percent of Mexican custodial parents.
We can examine the geography of Hispanic custodial parents in two ways: first by considering where certain subgroups are most likely to reside (e.g., where do most Mexicans live), and second by considering the ethnic distribution within a region (e.g., which subgroup is largest in the Midwest). As shown in the top half of Table 3.5, Mexican custodial parents are most likely to live in the West, Puerto Rican custodial parents are most likely to live in the Northeast, and Cuban custodial parents are most likely to live in the South. For every Hispanic subgroup except Mexicans, the Midwest is the least favored region. The bottom half of Table 3.5 shows the subgroup distribution of Hispanic custodial parents within each region. In the Midwest, South, and particularly in the West, Mexican custodial parents represent the largest single group. In the Northeast, Puerto Ricans are the single largest subgroup among Hispanic custodial parents.
|Subgroup Distribution of Hispanic Custodial Parents by Region|
Hispanic custodial parents are concentrated in metropolitan areas. In the late 1990s, 90 percent of Hispanic custodial parents lived in a metropolitan area, compared to 76 percent for all custodial parents. Not only are Hispanic custodial parents concentrated in metropolitan areas, within metropolitan areas Hispanic custodial parents are concentrated in the central cities. Indeed, while the rest of the custodial parent population in the United States has largely become a suburban population, Hispanic custodial parents are still more likely to live in central cities than in any other type of location. As shown in Table 3.6, the type of location varies tremendously for Hispanic subgroups.
Puerto Ricans are particularly concentrated in central cities, while Cubans are most likely to live in the suburbs. Mexicans and other Spanish custodial parents are more likely to live in rural areas than other Hispanic custodial parents, though the percentages are small.
Compared to other custodial parents, Hispanic custodial parents are more likely to have never married and less likely to be currently married. Puerto Rican custodial parents have especially low levels of current marriage, with almost half never married. Cuban custodial parents are more likely to be divorced than are other Hispanic custodial parents (see Table 3.7).
|Puerto Rican||Cuban||Other Spanish||
|Note: a = less than 1 percent|
The number of children in households with a custodial parent is slightly higher for Hispanics than for other ethnic groups. As shown in Table 3.8, Puerto Rican custodial parents have the greatest number of children living in their households, while Central/South Americans have the fewest.
|Puerto Rican||Cuban||Other Spanish||All Hispanics|
Public assistance utilization among Hispanic custodial parents has declined substantially over the past five years. In 1994, 35 percent of Hispanic custodial parents reported receiving public assistance, compared to only 17 percent in 1999 (see Figure 3.3). While all the subgroups experienced dramatic declines in public assistance utilization, the rates of use remained much higher for Puerto Ricans than for the other subgroups. For Mexicans and Central/South Americans, public assistance utilization rates were about the same for immigrants as for U.S. born custodial parents, whereas rates were much higher for Cuban and other Spanish immigrant custodial parents than U.S. born custodial parents of those groups. For all groups, custodial fathers had very low rates of public assistance utilization, about one-fifth those of custodial mothers. Separated and never-married custodial parents use public assistance at a rate four times higher than currently married custodial parents.
Labor force participation rates have been increasing dramatically for Hispanic custodial parents, but still remain lower than for non-Hispanic custodial parents. By 1998, 72 percent of Hispanic custodial parents were in the labor force, compared to only 61 percent in 1995. Labor force participation rates for custodial parents of any race or ethnic group in the United States had reached 79 percent by 1998. Labor force participation rates are substantially higher for Cubans than for other Hispanic custodial parents, and are substantially lower for Puerto Ricans. Even with their relatively low labor force participation rates, Puerto Rican custodial parents experienced the most dramatic increase in labor force participation in the late 1990s; in 1995, less than half of Puerto Rican custodial parents were in the labor force, by 1999 about two of every three Puerto Rican custodial parents were in the labor force. Since Puerto Rican custodial parents are more likely to have received public assistance than other Hispanic subgroups, the substantial increases in labor force participation among Puerto Rican custodial parents is at least partly a result of welfare reform and its work requirements.
Poverty rates are notably high for Hispanic custodial parents. In 1998, 43 percent of Hispanic custodial parents lived in poverty, compared to 30 percent for all custodial parents. Poverty rates for Hispanic custodial parents are more than three times higher than for the overall population of the United States (see Figure 3.4). These very high poverty levels, combined with the fact that over 70 percent of Hispanic custodial parents are in the labor force, suggest that a large share of Hispanic custodial parents are among the working poor. Among Hispanic subgroups, poverty rates are especially high for Puerto Ricans (more than half were in poverty in 1998) and somewhat lower for Cubans (though still above the average for all custodial parents in the United States). Mexicans and Central/South Americans have poverty rates similar to those for all Hispanic custodial parents.
Low levels of educational attainment are one reason Hispanic custodial parents have such high poverty rates. As shown in Table 3.9, in the late 1990s almost half of all Hispanic custodial parents had not graduated from high school. Only Cubans had high school completion rates that were close to those of all custodial parents.
|Some college||College graduate|
Figure 3.5. Percentage of Custodial Parents not Completing High School
These low levels of educational attainment are remarkable, and only partly attributable to immigrant populations. As shown in Figure 3.5, immigrants are much less likely to have graduated from high school than U.S. born Hispanic custodial parents. This is true for each of the subgroups, with seven of every ten foreign-born Mexican custodial parents not having completed high school. But even among U.S. born Hispanic custodial parents, over one-third have not graduated from high school, and only U.S. born Cubans have very high proportions of high school graduates.
Hispanic Nonparent Custodians
In 1998, Hispanics comprised 18 percent of all nonparent custodians. While our projections indicate that the growth rate of Hispanic nonparent custodians will be lower from 1998 to 2009 than in the preceding ten years, Hispanic nonparent custodians will still be one of the fastest growing segments of the child support population in the United States. Between 1998 and 2009, we project an increase of over 50 percent for Hispanic nonparent custodians, reaching a population of over 600,000 by 2009.
Table 3.10 provides a brief profile of this population as of the late 1990s. In general, nonparent Hispanic custodians have worse economic outcomes than other nonparent custodians. For example, poverty rates for nonparent Hispanic custodians are substantially higher than for other nonparent custodians. Educational attainment levels are especially low, even lower than those of Hispanic custodial parents; two of every three nonparent Hispanic custodians had not completed high school.
|Less than high school||68%|
|High School grad||20%|
|In the labor force||3%|
|Receive public assistance||3%|
Asian and Pacific Islander Child Support Populations
Asians and Pacific Islanders are less likely than other race and ethnic groups to be a part of the child support population. For example, less than 5 percent of adult Asians and Pacific Islanders are custodial parents, slightly lower than the percentage for whites, half the percentage of Hispanics, and one-third the percentage of African Americans. Still, Asians and Pacific Islanders are one of the fastest growing populations in the United States. We project that the number of Asian and Pacific Islander custodial parents will increase by over 80 percent from 1998 to 2009. This rapid increase is primarily a reflection of large increases in the number of Asians and Pacific Islanders in the United States, and not due to increases in the likelihood of being a custodial parent among Asians and Pacific Islanders.
Asians and Pacific Islanders are a diverse population. Ideally, we would be able to develop child support population estimates and projections for subgroups such as Chinese, Japanese, Filipino, Asian Indian, and Southeast Asians. Unfortunately, sample size issues are problematic when considering Asian and Pacific Islander child support populations. In the late 1990s, about 3,300 Asians and Pacific Islanders aged 15 and over were included in the March Current Population Surveys each year. Of those, less than 4 percent were custodial parents, and just about 1 percent were nonparent custodians. To present characteristics of custodial parents and nonparent custodians, we therefore combine six years of Current Population Survey data (1994 through 1999), giving us a total sample of 716 Asian and Pacific Islander custodial parents, and 191 nonparent custodians. With such small samples, breakdowns for subgroups are not possible. In some instances, we are able to contrast immigrant Asians and Pacific Islanders with U.S. born Asians and Pacific Islanders. Estimates for nonparent custodians are especially uncertain, and are not reported here.
Asian and Pacific Islander Custodial Parents
Among the race and ethnic groups we consider in this report, Asians and Pacific Islanders have the lowest probabilities of being a custodial parent. Overall, only 3.6 percent of Asians and Pacific Islanders aged 15 and over in the United States were custodial parents in the mid to late 1990s. These low rates are a reflection of low divorce rates and low rates of nonmarital childbearing.
Most Asian and Pacific Islander custodial parents are immigrants simply because the vast majority of adult Asians and Pacific Islanders in the United States are immigrants. In the United States, 75 percent of all Asians and Pacific Islanders aged 15 and over are foreign born, and 65 percent of all Asian and Pacific Islander custodial parents are foreign born. Thus, while first-generation immigrants are less likely to be custodial parents than second- and third-generation Asians and Pacific Islanders, they still comprise the majority of Asian and Pacific Islander custodial parents.
Asian and Pacific Islander custodial parents tend to be slightly older than other custodial parents. In the late 1990s, only 24 percent of Asian and Pacific Islander custodial parents were less than 30 years old, compared to 31 percent for all custodial parents. Foreign-born Asian and Pacific Islander custodial parents tend to be older than those born in the United States (see Table 3.11).
All Asian and
|U.S. Born||Foreign born|
The vast majority of Asian and Pacific Islander custodial parents live in the West (see Figure 3.6). This reflects the geographic distribution of Asians and Pacific Islanders in the United States. The West, and California in particular, has long been the primary destination of immigrants from Asia to the United States. Subsequent generations tend to settle near their parents, where social networks and contacts are strong.
Figure 3.6. Regional Distribution of Asian and Pacific Islander Custodial Parents
Asian and Pacific Islander custodial parents are concentrated in the suburbs. Over half (54 percent) of Asian and Pacific Islander custodial parents lived in the suburbs in the late 1990s. Asian and Pacific Islander custodial parents are especially unlikely to live outside a metropolitan area, with only 8 percent doing so in the late 1990s.
Asian and Pacific Islander custodial parents are slightly more likely to be married and slightly less likely to have never married than other custodial parents (see Table 3.12). Because these statuses are associated with economic outcomes (married custodial parents fare much better than other custodial parents, and never-married custodial parents fare the worst), some of the indicators that suggest Asian and Pacific Islander custodial parents are relatively well off stem from their favorable mix of marital statuses.
Public assistance utilization rates for Asian and Pacific Islander custodial parents are slightly lower than for other groups, and appear to exhibit the same strong downward trend during the mid-to-late 1990s. The decline in public assistance use seems to be greater among immigrant Asian and Pacific Islander custodial parents, though small sample sizes do not allow us to attach statistical significance to the differences.
Poverty rates are high for Asian and Pacific Islander custodial parents, though not quite as high as for other custodial parents. In the late 1990s, 26 percent of Asian and Pacific Islander custodial parents lived in poverty, compared to about 30 percent for custodial parents of any race or ethnic group. Foreign-born Asians and Pacific Islanders had lower poverty rates than U.S. born Asians and Pacific Islanders (24 percent and 29 percent, respectively).
Asian and Pacific Islander custodial parents tend to be better educated than other custodial parents. One of every four Asian and Pacific Islander custodial parents has graduated from college, for example, compared to only one of every eight custodial parents of any other race or ethnic group. Immigrant Asians and Pacific Islanders exhibit a bipolar educational distribution, with a large proportion with little education (24 percent had not graduated from high school) and a large proportion with college degrees (30 percent in the late 1990s). Asians and Pacific Islanders born in the United States had very high proportions graduating from high school (92 percent), but lower percentages of college graduates (15 percent) than immigrant Asians and Pacific Islanders. Overall, these high levels of education help explain why Asian and Pacific Islander custodial parents have low poverty rates compared to other custodial parents.
Underlying Population Trends
The relatively slow growth of child support populations in the United States over the next ten years can be attributed to several underlying trends in the population of the United States. Those trends include the aging of the nation’s population, a slight decline in the high rates of divorce, and the leveling off of nonmarital childbearing at a historically high level.
One of the central features of the population of the United States over the past fifty years has been the creation and aging of the baby boom generation. Defined as the cohort of people born between 1946 and 1964, baby boomers comprised more than one-fourth of the U.S. population in 1998. The implications of the aging of the baby boom are well known for the social security system, but their aging also has implications for the child support populations of the country. Baby boomers are increasingly becoming “empty nesters,” older adults whose children have moved out of the home. In 1988, baby boomers were between the ages of 24 and 42, prime ages for parenthood. By 2009, baby boomers will be between the ages of 45 and 63, and many will no longer have children living at home. Figure 4.1 and Figure 4.2 illustrate the aging of the baby boom. The number of people between the ages of 30 and 44 will decline by almost 10 percent between 1998 and 2009.
Figure 4.1. U.S. Population by Age
Figure 4.2. Percent Change in U.S. Population by Age, 1998-2009
The aging of the baby boom matters for child support populations because of the age pattern of parenthood. In particular, adults between the ages of 30 and 39 are those most likely to be custodial parents (see Figure 4.3). As baby boomers age out of theseprime custodial (and noncustodial) parent age groups, all else equal the number of custodial parents would decline. The aging of the baby boom is the single most important factor in the slowing in the rate of increase of child support populations. Indeed, precisely because of the aging of the baby boom, the Census Bureau projects that the number of children in the United States will change very little over the next ten years. As noted earlier in this report, the number of children in the United States is projected to increase only 3 percent between 1998 and 2009.
Figure 4.3. Percentage of Custodial Parents Among Adults in 1998, by Age Group
Divorce rates are another important determinant of child support populations. As shown in Figure 4.4, divorce rates, after almost tripling from 1960 to 1980, have actually declined. From its peak in 1979 of 22.8 divorces per 1,000 married women, the divorce rate dropped 17 percent to 19.5 divorces per 1,000 married women in 1996. Still, the annual number of divorces remains at or near record levels, and divorce rates are still very high by long-term standards. The high divorce rate leads to increases in the child support population, though the increases are not as great as they would have been had divorce rates continued to increase into the twenty-first century. About 1.2 million divorces occur each year, involving approximately 1 million children. In 1996, we estimate that 15 out of every 1,000 children will be involved in a divorce, compared to 19 per 1,000 in 1981 and only 6 per 1000 in 1951. By the age of 18, over 20 percent of American children will experience the divorce of their parents.
Figure 4.4. Divorces and Divorce Rates 1940-1997
Nonmarital childbearing also plays a role in determining child support populations. As shown in Figure 4.5, after dramatic increases from the 1970s to the early 1990s, nonmarital childbearing has leveled off in recent years. Still, the levels of nonmarital births, nonmarital birth rates, and the proportion of all births that are to unmarried mothers are at or near record levels.
Figure 4.5. Nonmarital Births, 1970-1998
By the late 1990s, about one in every three births in the United States was to an unmarried mother (compared to only one in every ten births in 1970). During the early part of the 1990s, the rate of nonmarital childbearing was fairly stable, but the proportion of births to unmarried women continued to increase (because marital fertility declined during this period). A continuation of these high levels of nonmarital childbearing will lead to increases in the child support population of the United States, although the increases are not as large as they would have been had the levels continued to rise throughout the 1990s and into the first decade of the twenty-first century. In addition, past increases in nonmarital fertility have not yet worked themselves through the entire child support population. A simple example illustrates this point: in 1998, less than 20 percent of 17-year-old children were born to unmarried mothers, by 2009 almost one-third of 17-year-olds will have been born to unmarried mothers. Thus, the proportion of children born to unmarried mothers will continue to rise, even as nonmarital fertility rates remain relatively constant.
Trends in living arrangements have obvious and direct implications for child support. Increases in single-parent families lead to increases in support eligibility. The number of two-parent families with children less than 18 years of age has changed very little over the past ten years in the United States, increasing only 3 percent. One-parent families grew by 28 percent during this same time period, and now comprise one-third of all families with children. Much of the growth in one-parent families has been fueled by increases in families headed by single fathers. The number of single-parent father families almost doubled in size from 1988 to 1998, increasing from 1.2 million to 2.1 million. Thus, the trends discussed earlier for child support populations reflect broader societal changes in the make-up and constitution of families in the United States.
Economically, families tended to be slightly better off in 1998 than in 1988. In 1988, 11.6 percent of all Americans living in families lived in poverty, compared to 11.2 percent in 1998. Both 1988 and 1998 were years with strong national economies. The intervening recession years saw poverty rates for people in families peak at 13.6 percent in 1993. Poverty rates for people living in married couple families are lower than for people who live in non-family households. People in non-family households, a growing share of the child support population, have poverty rates almost twice as high as those in family households (20.8 percent versus 11.2 percent in 1998).
While poverty rates have declined only slightly, the number of families receiving public assistance has fallen dramatically. The number of families in the United States receiving welfare (Aid to Families with Dependent Children in 1988, and Temporary Assistance to Needy Families in 1998) declined from 3.8 million to 3.2 million between 1988 and 1998. This 17 percent decline is almost identical to the 18 percent decline in the number of custodial parents receiving public assistance. Estimates from the Department of Health and Human Services indicate an incredible 20 percent decline in the number of families receiving welfare in 1999, as compared to just one year earlier. The strong economy and especially welfare reform have led to these remarkable declines. These declines in welfare caseloads suggest that child support is becoming an increasingly important potential source of income for poor families.
We know that some groups of people are much more likely to be in the child support population than other groups. For example, women are much more likely to be custodial parents than men. These population subgroups can be defined by gender, race or ethnicity, age, and marital status. Large and dramatic differences in the likelihood of being in the child support population exist within each of these groups (see Appendix C). For each group, there are also temporal changes in the likelihood of being a custodial parent. By identifying characteristics associated with being a custodian and by examining temporal trends, we can develop projections of child support populations and understand the forces that are driving changes in the child support population.
Figure 5.1. Percentage of Custodial Parents Among Adults, by Year
In developing projections, we look for stability in historical probabilities of being in the child support population. As shown in Figure 5.1, the percent of adults that were custodial parents rose from around 6 percent in the late 1980s to 7.5 percent by 1995, and, though declining, has remained above 7 percent since 1993. This temporal variation in the probability of being a custodial parent is dwarfed by the variation associated with other characteristics, such as gender, race, age, and marital status. For the purposes of projecting the number of custodial parents in the future, this relatively small temporal variation is good. The much larger variation associated with other factors means that if we can project the population by age, race, gender, and marital status, we can produce reasonable projections of the number of custodial parents.
Because no single data set provides all the information necessary, we used various data sets and methods to develop projections of child support populations. In brief, our method relies on Census Bureau projections of the population by marital status, age, and gender for future populations of the United States. Then, using historic trends in the probability of being in the child support population generated primarily from Current Population Surveys, we forecast and apply those probabilities to the Census Bureau population projections. We disaggregate the trends and hence the projections by race, ethnicity, age, gender, and marital status. We consider the sensitivity of the projections by applying various assumptions and alternative specifications of the projection model. Below, we discuss our approach in more detail.
Our first step is to develop population projections for our various population groups. While the United States Census Bureau provides estimates and projections of the nation’s population by marital status, it does not include race and ethnicity detail. Because the likelihood of being in the child support population varies considerably by race and ethnicity, it is necessary to disaggregate the Census Bureau’s marital status projections into race and ethnic categories. To do so, we use a three-step process. First, we use historical trends in race and ethnic proportions by marital status to project those proportions to 2004 and 2009. Specifically, within distinct marital statuses, we use logistic regression models to identify time trends in the likelihood of belonging to certain race and ethnic groups by age. We extend these trends to 2004 and 2009 by using linear extrapolations. Second, we apply the projected race and ethnic proportions to the Census Bureau’s projections by marital status and age to obtain population projections. Finally, we adjust those projections so that they are consistent with the latest Census Bureau national population projections by age and race/ethnicity. The Census Bureau projections by marital status are relatively old. We evaluated the projections in light of recent estimates, and adjusted our race and ethnic totals to be consistent with the Bureau’s more recent population projections by race and ethnicity (those recent projections do not include marital status). We also discussed with Census Bureau demographers some early results from a new and as of yet unreleased set of population projections by marital status, and were assured that the changes are minimal for our projection horizon (2009).
Our second step involves projecting future likelihoods of being in the child support population. To do so, we analyze past trends in the probability of being a custodial parent and in the probability of being a nonparent custodian. Projections of noncustodial parents and children eligible for child support are derived from the projections of custodial parents and nonparent custodians. We use regression models to identify trends in custodial status among people aged 15 and over. Specifically, using the March Current Population Surveys from 1988 through 1999 and the April Child Support supplements of the Current Population Survey, we develop logistic regression models of the probability of being a custodian. Separate models are developed for custodial parents and nonparent custodians, each of four race and ethnic groups, three marital statuses, and for men and women. This procedure takes advantage of the large variations in the likelihood of being a custodian according to race and ethnicity, gender, marital status, and age. The regression results provide us with twelve-year trends (1988 through 1999) in the probability of being a custodian. These trends are then used to project future probabilities of being a custodian. Two sets of projections are developed. One set holds the probabilities constant throughout the projection horizon, and the other continues the trends observed during the past five years into the next ten years. These two sets of probability projections for each race and ethnic group, both genders, and each marital status, are then applied to the projected base populations to derive two sets of projections of custodians. The final series of projections then averages these two sets of projections.
To project characteristics not associated with our population groups, we simply use historic trends. In most cases, we use ten-year average annual changes to project changes to 2004 and 2009. One important exception is in the case of public assistance utilization rates among child support populations. As discussed in more detail earlier in the report, we hold those rates constant at 1999 levels.
Finally, we conduct sensitivity tests to evaluate how changes in assumptions and the model would alter the final projections. In general, we find that the projections of custodial parents are less sensitive to changes in assumptions and modeling approach than projections of nonparent custodians. In most cases, the model allowing probabilities to vary over time results in a higher child support population than the model which holds probabilities constant. For custodial parents, the difference in the two sets of projections is less than 5 percent in 2009. For nonparent custodians, the difference is less than 10 percent in 2009. Still, these differences are not insignificant, especially considering the relatively slow growth of the child support population in the United States. They suggest that changes in nonmarital childbearing and divorce could have substantial impacts on future child support populations.
The Office of Child Support Enforcement will be better equipped to meet the needs of the next generation of children if it has specific knowledge about the demographic characteristics of the population it serves. On the one hand, the fact that children are born every day seems immutable What could be more constant than the succession of generations? Even the dreary fact that not all children will be adequately cared for seems an old story. On the other hand, the households in which children are raised can and have changed considerably over time. Who has children, whether they marry or divorce, and the age of their respective cohorts can fluctuate over time, changes which cumulatively contribute to sizable differences in the child support population.
In the twenty-year period we examine, the changes in the child support population have been and will be dramatic. The rapid growth of the child support population in the 1980s and 1990s will slow in the first decade of the twenty-first century, but the increase in the absolute number of children, custodians, and noncustodial parents remains momentous, indicating a serious need for a strong program. The growth rate of the child support population has slowed because of the aging of the baby boomers, a decline in divorce rates, and a leveling of the rate of nonmarital fertility.
That the rates of increase have slowed from the steep rates of the 1980s and 1990s does not mean that the future population of the child support community will not have striking differences from the population in the past. Certain segments of the child support population will experience consequential growth. There will be a large increase in the number of custodial fathers, and an even greater increase in the number of children living with never-married parents. The number of Hispanic and Asian custodial parents will also jump considerably as the ethnic composition of the United States changes. Finally, the population of nonparent custodians, including grandparents, will continue to grow at a faster rate than that of custodial parents.
While the economic characteristics of custodians will improve slightly over the coming decade (younger adults coming into the system tend to be better educated than their older counterparts) the population remains, in general, a poor population and hence more vulnerable to economic downturns and to cuts in the funding of public assistance. In the coming decade there will be more children in the child support system, fewer children born to married parents, and perhaps fewer families receiving public assistance. Under these circumstances, child support enforcement will be more important than ever. Children need financially stable environments. Making sure that parents, when able, accept financial responsibility for their children will foster social bonds, assist single parents, and help to alleviate child poverty in the United States.
Appendix A. Detailed Tables
|American Indian and other||2%||3%||4%||4%|
|Asian and Pacific Islander||1%||2%||3%||3%|
|Type of area|
|Other part of metro||37%||42%||46%||49%|
|Outside a metro||25%||24%||23%||22%|
|Number of children age 17 or less|
|Four or more||7%||6%||6%||6%|
|Current Marital Status|
a = less than 0.5%
The race/ethnic group "Other" consists primarily of American Indians, Eskimos, and Aleuts.
Married includes married living separately.
|1.00-1.24 times pov.||6%||7%||8%||8%|
|1.25-1.50 times pov.||5%||6%||6%||6%|
|1.50+ times pov.||54%||57%||60%||62%|
|Labor Force Status|
|Not in labor force||29%||21%||18%||16%|
|Less than h.s.||26%||19%||17%||16%|
a = less than 0.5%.
Labor force status is among those eligible (persons aged 16 and over).
Public assistance includes AFDC and "other assistance" and is determined at the family level for receipt in the year prior to the survey.
|American Indian and other||4%||4%||5%||6%|
|Asian and Pacific Islander||2%||3%||4%||5%|
|Current Marital Status|
|Less than high school||54%||43%||38%||33%|
|High school graduate or GED||30%||33%||31%||31%|
|Labor Force Status|
|Not in the labor force||40%||39%||38%||38%|
|Below poverty level||35%||30%||27%||26%|
|1.00-1.24 times poverty||6%||7%||8%||8%|
|1.25-1.50 times poverty||7%||6%||7%||7%|
|Above 1.50 times poverty||52%||56%||58%||59%|
|Type of area|
|Central city of a metropolitan area||39%||35%||34%||33%|
|Other part of a metropolitan area||34%||36%||38%||39%|
|Outside a metropolitan area||27%||29%||28%||27%|
|Number of children age 17 or less|
|Four or more||10%||7%||6%||5%|
a = less than 0.5%.
Some custodians are both custodial parents and other custodians. They are included in this table and the custodial parent table. Married includes married living separately.Labor force status is among those eligible (persons aged 16 and over). Public assistance includes AFDC and "other assistance" and is determined at the family level for receipt in the year prior to the survey.
Appendix B. Models
To estimate trends in the probability of being a custodial parent, we use logit models. The model takes the form:
log(p/(1-p)) = a + b1x1 + b2x2
where p is the probability of being a custodial parent, x1 is a vector of dummy variables for seven age groups and x2 is a vector of dummy variables for years 1988 through 1999. Because the CPS sample sizes are large, we are able to develop separate models by gender, race and ethnicity, and marital status. The coefficients for the time dummies are then used to develop projections for each gender, race and ethnic group, and marital status.
The model for nonparent custodians is similar, except fewer age groups and fewer race and ethnic groups are included in the vectors of the dummy variables.
Appendix C. Demographic Differences Among Custodial Parents
In this appendix, we present some figures that illustrate the dramatic differences in the probability of being a custodial parent according to certain demographic characteristics. These differences are shown for key demographic characteristics and temporal differences are shown for some groups.
Figure C.1. Percentage of Custodial Parents Among Adults in 1998, by Gender
Figure C.2. Percentage of Custodial Parents Among Adults in 1998, by Race/Ethnicity
Figure C.3. Percentage of Custodial Parents Among Adults, by Marital Status
Figure C.4. Percentage of Custodial Parents among Adults, by Gender
Figure C.5. Percentage of Custodial Parents among Adults, by Age Group
Figure C.6. Percentage of Custodial Parents among Adults, by Race/Ethnicity
Jennifer Cheeseman Day. Copyright © 1996. Projections of the Number of Households and Families in the United States: 1995 to 2010, U.S. Bureau of the Census, Current Population Reports. 25-1129. Government Printing Office, Washington D.C.
Irwin Garfinkel, Sara S. McLanahan, and Thomas L. Hanson. Copyright © 1998. “A Patchwork Portrait of Nonresident Fathers,” in Fathers Under Fire: The Revolution in Child Support Enforcement. Russel Sage Foundation, New York.
Susan E. Mayer. Why Welfare Caseloads Fluctuate: A Review of Researach on AFDC, SSI, and the Food Stamps Program. JCPR Working Paper 166 04-01-2000.
Robert Moffit. Copyright © 1999. Is Welfare Reform Working?: The Impacts of Economic Growth and Policy Changes.
Copyright © 2000 National Center for Health Statistics. Health, United States, 2000; With Adolescent Health Chartbook. Hyattsville, Maryland.
Copyright © 1998 National Center for Health Statistics. Health, United States, 1998; Hyattsville, Maryland.
Copyright © 1998 National Center for Health Statistics. National Vital Statistics Report. Vol. 47, No. 4.
Copyright © 1998 National Center for Health Statistics. Monthly Vital Statistics Report. Vol. 43, No. 9, March 1995.
Gregory J. Robinson, Ahmed Bashir, Prithwis Das Gupta, and Karen A. Woodrow. Copyright © 1993. Estimation of Population Coverage of the 1990 United States Census Based on Demographic Analysis., Journal of the American Statistical Association. . 88:423:1061-1071.
Gregory J. Robinson and Bashir Ahmed. Copyright © 1992. Utility of Synthetic Estimates of Census Coverage for States Based on National Demographic Analysis Estimates.
Gregory J. Robinson, Bashir Ahmed, Prithwis Das Gupta, and Karen A. Woodrow. Copyright © 1991. Estimating Coverage of the 1990 United States Census: Demographic Analysis.. Proceedings of the Social Statistics Section. American Statistical Association.
Elaine Sorensen. Copyright © 1999. Obligating Dads: Helping Low-Income Noncustodial Fathers do More for their Children. Washington, D.C.. Urban Institute.
Elaine Sorensen. Copyright © 1997. A National Profile of Nonresident Fathers and Their Ability to Pay Child Support. Journal of Marriage and the Family. vol. 59 (November). 785-97.
U.S. National Center for Health Statistics. Monthly Vital Statistics Report. Vol. 46, No. 12. July 1998.
U.S. Bureau of the Census. Copyright © 2000. Population Projections of the United States by Age, Sex, Race, Hispanic Origin, and Nativity: 1999 to 2100. Population Projections Program, Population Division, U.S. Census Bureau, Washington, D.C.
U.S. Bureau of the Census. Copyright © 1999. Statistical Abstract of the United States: 1998. Washington, D.C.
U.S. Bureau of the Census. Copyright © 1999.National Estimates: Quarterly Population Estimates, 1980 to 1990; Quarterly Intercensal Civilian Population Estimates 4/1/88.
 Appendix A contains tables with detailed results for custodial parents and nonparent custodians.
 See section four for a thorough treatment of these underlying determinants.
 We discuss this notable trend in greater detail in section three.
 In nonparent custodial households, we designate only one adult as the nonparent custodian. In married couple households, the reference person for the survey (often the husband) is designated as the nonparent custodian. This may account for the relatively high proportion of males among nonparent custodians.
 We have excluded foster children from our estimates and projections. Thus, biological parents of children in foster care are not included in our estimates and projections of noncustodial parents.
 The estimates and projections in this report are for the 50 states and the District of Columbia. The Commonwealth of Puerto Rico is not included in the estimates and projections, though Puerto Ricans living in one of the states or in the District of Columbia are included.
 We use the term “Mexican” to include any person who identified as Mexican American, Mexican, Mexicano, or Chicano.
 Puerto Ricans born in Puerto Rico are born in the United States.
 Because public assistance utilization rates have changed so much during the late 1990s, estimates pooled across the sample years would be misleading. For the largest subgroups, we report annual estimates. Relative levels of utilization between subgroups are consistent for each year of the data, despite small sample sizes. Some of our discussion is based on these general relationships, which hold across years even as the levels decline.
 Non-family households are households in which no members are related. In the child support population, a child living with an unrelated adult is considered a non-family household.
 See Appendix B for the form of the regression models.