Children in Poverty: Profile, Trends, and Issues

November 25, 2008 (RL32682)

Contents

Figures

Tables

Appendixes

Summary

Child poverty persists as a social and economic concern in the United States. In 2007 12.8 million children (17.6% all children) were considered poor under the official U.S. definition. In records dating back to 1959, the incidence of poverty among related children in families has ranged from a peak of 26.9% in 1959 to a low of 13.8% in 1969. Poverty affects a child's life chances; by almost any indicator, poor children fare worse than their nonpoor counterparts.

Family living arrangements, indicated by the presence of just one or both parents, greatly affect the chances that a child is poor. In 2007, 43.0% of children in female-headed families were poor, compared to 8.5% of children in married-couple families. In that year, 24% of children were living in female-headed families, more than double the share who lived in such families when the overall child poverty rate was at its historical low (1969). Children who are racial or ethnic minorities are at particular risk of being poor. In 2007, a little more than one-third of black children (34.2%) and almost three out of ten Hispanic children (28.3%) were poor, compared to about one in ten white non-Hispanic children (9.7%).

Work is the principal means by which families with children support themselves. Without family earnings, a child is almost certain to be poor. However, earnings often fail to overcome poverty. In 2007, about one-third (32.4%) of all poor children lived with at least one adult who was a full-time, full-year worker; another one-third were in families with a worker who either worked part-year or (less likely) part-time; another third lived in families without an adult who worked during the year. Higher child poverty rates were observed for those whose parents had less, rather than more, education. Children of younger parents, with less potential time and experience in the workforce, were more likely to be poor than children of older parents. Additionally, dramatic gains have occurred in recent years in work by lone mothers—especially among those with preschool children. Employment rates of single mothers with children under age 3 rose from 35.1% in March 1993 to 59.1% in March 2000, but have since remained below their 2000 level, standing at 54.5% in March 2008. Nonetheless, many of these working single mothers (and their children) remained poor.

The social safety net for children consists of (1) earnings-based social insurance programs and (2) need-based transfers of cash and noncash benefits. Need-tested benefits have undergone a radical transformation during the past 20 years, capped by the 1996 welfare reform law. Cash welfare caseloads have plummeted since the reforms of the mid-1990s, so that many families receiving need-tested aid only receive noncash benefits (e.g., Medicaid and food stamps) whose value is not reflected in official poverty statistics. Further, the welfare reforms of the mid-1990s were accompanied by expansions of the Earned Income Tax Credit (EITC), which supplements the earnings of low-income families with children. (The value of the EITC is also not considered in official poverty statistics.) The result has been to curtail benefit availability for nonworking families while raising the returns to work. This report will be updated annually, when new Census Bureau data are released.


Children in Poverty: Profile, Trends, and Issues

Introduction

Child poverty persists as a social and economic concern in the United States. In 2007,12.8 million children were considered poor under the official U.S. Census Bureau definition. The child poverty rate (the percent of all children considered poor) stood at 17.6%—well below the most recent high of 22% in 1993, but still well above its historic low of 13.8% in 1969. In 2007, 1.8 million more children were counted as poor than in 2000, when 15.6% of children were poor.

Child poverty reflects both family circumstances in which children reside and economic conditions and opportunities in communities where children live. Family living arrangements, indicated by the presence of just one or both parents, greatly affect the chances that a child is poor. Children who are racial or ethnic minorities are at particular risk of being poor.

To avoid poverty, most children need an adult breadwinner. The economic well-being of children usually depends on how well their parent(s) fares in the labor market. Children most at risk of poverty are in families without an earner. However, some are poor despite the work—even the full-time work—of a parent. They are among the working poor.

Poverty affects children's life chances, their prospects of realizing their full potential, and their ability to successfully transition into adulthood.1 By almost any indicator, poor children fare worse than their nonpoor counterparts. Poor children are at greater risk of poor physical health, delayed cognitive development, poor academic achievement, and risky behavior (particularly among children growing up in poor neighborhoods). Poor adolescent girls are more likely to become teenage mothers than their nonpoor counterparts, contributing to a cycle of poverty from one generation to the next.

While income poverty is associated with poor child outcomes, lack of income may account for only part of the reason why poor children face poor future prospects. Other factors are arguably as important, if not more so, than income, per se, in affecting children's life chances. Prolonged or deep income poverty among families with children may signal more chronic problems than merely a lack of income. Income support policies may help alleviate economic distress among such families, helping to provide families' basic needs; however, without other social supports or other, more fundamental changes in children's family circumstances, children's prospects may remain limited.

Numerous programs operate in the U.S. to aid children without a breadwinner, or with one whose earnings are low. These programs, together, constitute the main threads of this nation's income/social safety net:

Government is challenged to maintain family self-support through work-based policies to promote parents' work and family economic self-sufficiency, while at the same time maintaining a safety net that prevents children from falling into abject poverty when parents are unable to work or their efforts are insufficient.

Child Poverty in 2007

In 2007, 12.8 million children2 out of a total of 72.8 million lived in families whose pre-tax money income that year fell short of the poverty threshold. This translates into a child poverty rate of 17.6%.

How is Child Poverty Measured?

The Census Bureau measures poverty by comparing money income of the child's family with income thresholds that vary by family size and composition. For this purpose, the Bureau counts only money income before taxes (and any capital gains). For 2007, the poverty threshold for a single parent with two children was $16,705; for two parents with two children, it was $21,027.

The official count of the poverty population takes into account all cash income (other than any capital gains) before taxes. Thus, it includes social insurance cash payments and cash welfare. It excludes noncash aid and tax credits. The Census Bureau also issues alternate poverty counts that assign value to major noncash benefits and tax credits, but they are not included in this CRS report.

As a group, children are more likely to be poor than are either the aged (persons aged 65 or older) or nonaged adults (persons 18 to 64 years old). Figure 1 compares 2007 poverty rates of children, the aged, and nonaged adults. It shows that children were 80% more likely to be poor than the aged and that the incidence of poverty among nonaged adults was only slightly higher than that of the aged.

There are some well-known correlates to child poverty. A child's risk of being poor varies by family structure and size and by race/ethnicity. Further, since children primarily rely on adult workers for income, child poverty also varies by the educational attainment and age (which is related to work experience) of the family's head, and (if present) spouse.

This section provides a profile of child poverty in 2007. Shown are child poverty rates by characteristics of the family, child, or family head and spouse. This perspective answers the question, among children, who are most at risk for being in poor families? This section also provides the composition of child poverty by those characteristics, answering the question who are the poor children?

Figure 1. Poverty Rates of Children, the Aged, and Nonaged Adults, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Child Poverty by Family Type

Living arrangements of U.S. children have undergone a dramatic shift since the 1960 census. In 1960, 88 out of 100 children lived with two parents. The proportion living with two parents averaged 81% during the 1970s, 75% in the 1980s, and 70% in the 1990s. Corresponding increases were registered by children living with mother only (the family type most afflicted with money income poverty): 8% in the 1960 census; 15% during the 1970s; 20% in the 1980s, and 23% in the 1990s. The data for children living with father alone: 1% in the 1960 census, 3% on average during the 1970s and the 1980s, and 4% during the 1990s. Finally, the share of children living with neither parent averaged 3% during the 1970s and 1980s, but rose to 4% in the 1990s.3

Figure 2 presents data about poverty rates of children who lived in families in 2007. The bar chart presents 2007 child poverty rates by type of family. As it shows, children in married-couple families had a much lower poverty rate (8.5%) than those in single parent families headed by a woman (43.0%) or a man (21.3%). The pie chart shows the composition of the 12.8 million related poor children. Female-headed families held the majority of poor children—58.9%, or 7.5 million. Children in husband-wife families, despite their relatively low risk of poverty, comprised 34.0% of all poor children, or 4.3 million. The remaining 0.9 million poor children (7.1%) were in male-headed families (no wife present).

Figure 2. Poverty Among Related Children, By Family Type, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Because of the relatively high poverty rate among children in female-headed families, much of the policy discussion about child poverty centers on this group. Therefore, much of this remaining profile provides information on the poverty rate and composition of poverty for children in single parent families as well as for all children.

Among children in female-headed families, poverty rates vary by whether the single mother had ever been married (see Figure 3). Children with mothers who never married are about twice as likely to be poor (a 54.8% poverty rate) as children whose mothers divorced (poverty rate of 27.6%). Moreover, the number of all children in families where the single mother never married exceeds the number of children in families where the mother divorced. Thus, children in families with never-married mothers account for a large share of poor children, about half (51.3%) of all poor children in female-headed families.

Figure 3. Poverty Among Related Children in Female-Headed
Families, by Marital Status of the Female Head, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Child Poverty by Race and Ethnicity

Children in racial and ethnic minorities tend to have higher poverty rates than white children. Figure 4 shows that in 2007, the poverty rate among African-American (non-Hispanic) children was 34.2%—3.5 times the poverty rate for white (non-Hispanic) children of 9.7%. Hispanic children had lower poverty rates than African-American children, but higher poverty rates than white children.

Despite the variance in poverty rates by race and ethnicity, the populations of poor white, African-American, and Hispanic children are of similar size. The larger population of white children, when multiplied by the lower poverty rates of this group, yields a number of poor white children not too different from the number of poor children in each of the two other racial/ethnic groups.

Figure 4. Poverty Rates Among Related Children, By Race and Ethnicity, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Figure 5 shows poverty rates and the composition of poor children by race and ethnicity for children in female-headed families. As in the overall child population, minority children have higher rates of poverty than do white children—the poverty rates for African-American (49.9%) and Hispanic (51.6%) children are both about 1.5 times that of white children in female-headed families (32.4%). However, African-American children comprise the largest group of poor children in female-headed families—accounting for about four out of ten poor children in female-headed families.

Figure 5. Poverty Among Related Children in Female-Headed
Families, By Race and Ethnicity, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Additionally, Hispanic children are highly likely to be in married couple families. (Not shown on the figure.) Among children in married-couple families, Hispanic children have a poverty rate (19.3%) higher than that of whites (4.7%) or African-American children (11.0%). Therefore, Hispanics account for the largest share of poor children in married-couple families (45.9%).

Thus, Hispanic children in female-headed families have a slightly higher poverty rate than African-American children in female-headed families, and also have a substantially higher poverty rate than African-American children in married-couple families. However, their poverty rates among all children are lower than those for African-Americans. This is because Hispanic children are more likely than African-American children to be in married-couple families, and children in married-couple families have lower poverty rates than children in female-headed families.

Figure 6. Poverty Rates Among Related Children in Married Couple
Families, By Race and Ethnicity, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Child Poverty by Annual Work Experience of the Family Head
or Spouse

If a family has no earnings, a child is almost certain to be poor. The poverty rate in 2007 for children without a working parent (or spouse of the family head) was 70.0% (bar chart in Figure 7). However, millions of children are poor even though the family head (or spouse) works full time, year round. In 2007, about seven out of 100 children with such a worker were poor. The number of these children totaled 4.2 million, and they accounted for 32.4% of all poor children (pie chart in Figure 7). Most children (77.4%, not shown on the chart) live in families where either the head or, if present, the spouse is a full-time, full-year worker. Hence, these children account for a relatively large share of all poor children even though their poverty rate is low. The bar chart shows that the incidence of poverty is 35.1% among children in families with a parent who works full-time part-year and 40.0% among those in which a parent works part-time, full-year.

Figure 7. Poverty Among Related Children, By the Work Experience
of the Family Head or Spouse, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

The relationships between work experience for family adults over the year and child poverty also hold for children in female-headed families (though these families are without a potential spouse to supplement the work of the family head). Among children in female-headed families where the head works full-time, year-round, 18.7% were poor (bar chart in Figure 8). Of children in single, female-headed families without an earner, 76.2% were poor.

The pie chart in Figure 8 depicts the composition of all poor children in female-headed families. Of these children, 20.5%—totaling 1.5 million children—lived in families where the mother was a full-time worker, year round. However, the pie chart also shows that 42.9% of all poor children in these families had a mother who did not work during the year. Though work among lone mothers has increased dramatically in recent years (discussed later in this report), 24.2% of all children in female-headed families had a mother who did not work. In comparison, only about 2.6% of all children in married-couple families had neither adult heading the family in the workforce.

Figure 8. Poverty Among Related Children in Female-Headed Families, by Work Experience of the Female Head, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Child Poverty by Educational Credential of the Family Head or Spouse

From 1960 to 2007, the share of the population (at least 25 years old) with a high school diploma (or more) has more than doubled, from 41.1% to 85.7%. In the same period the share with 4 or more years of a college education more than tripled, from 7.7% to 28.7%.4 The returns to education have also increased over time, as the average wages of those with a college degree have increased relative to the average wages of those without such a degree. Increasingly, job applicants must have postsecondary credentials. By some measures, high school graduates and those who failed to complete high school have seen declining real wages.

Thus, child poverty rates depend in part on the educational level of the family head (or spouse). In 2007, almost half (49.6%) of children whose family head had not completed high school were poor. This group made up one-third of all poor children (see Figure 9). Children whose family head (or the family head's spouse) had completed high school, but not gone beyond it, had a poverty rate of 27.3% and represented 37.2% of all poor children. If the family head or family head's spouse achieved an associates degree, the child poverty rate was sharply lower (9.7%) than that of one with some postsecondary education, but no degree (16.0%).

Figure 9. Poverty Among Related Children, by the Educational Attainment of the Family Head or Spouse, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

In 2007, more than two out of three children (68.3%) in families headed by a mother who failed to complete high school were poor (see Figure 10). This group represented 33.6% of all poor children in female-headed families. Attainment of a high school diploma reduced the child poverty rate, but it still was almost one-half (49.2%). As with the overall child population, poverty rates were much lower when the female head earned a college degree. Children in families where the female head received an associates degree had a poverty rate of 24.6%, compared with 35.5% for those who had some post-secondary education but no degree. Poverty rates were relatively low (13.9%) for children with female heads who had a bachelors degree or an advanced degree.

Figure 10. Poverty Among Related Children in Female-Headed Families, by Educational Attainment of the Female Head, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Child Poverty by the Age of the Head or Spouse

Child poverty rates are highest in young families with children. The poverty rate for preschool children (children under the age of six) was 20.8% in 2007, compared with a 16.0% poverty rate for older children. This is because, in part, these children have on average younger parents. Younger parents, who have less job experience in the workforce, tend to earn less than older adults with more experience.

Figure 11 shows the child poverty rate by the age of the household head or spouse if present. For married-couple families, the age of the older adult was used to determine the age of the head or spouse. It shows that child poverty rates tend to mirror the "life-cycle" pattern of earnings of adults. That is, earnings tend to be low in the early years, peak in middle age, and decline as adults approach and reach retirement age. This explains some of the pattern shown on the figure. However, children in families with a never-married female head (the group with the highest poverty rate) also tend to be in families where the head was young.

Children in families with the head or spouse under age 25 have the highest poverty rates, almost 50%. For children in families with a head or spouse aged 30 to 34, the poverty rate drops to 22.4%. Among poor children, 45% are in families with the head and spouse younger than age 35. Child poverty rates drop below the average rate for all children (17.6%) for age groupings with the head or spouse over 35 and younger than age 64, but are about the same as the overall rate for children in families with a head or spouse age 65 and older.

Figure 11. Poverty Among Related Children, by Age of the Family Head or Spouse, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

The basic relationships between child poverty and age of the family's adults shown in Figure 11 also tend to hold for children in female-headed families: the highest poverty rates are for children with younger parents. (No figure is shown.) However, poverty rates for children in female-headed families are higher than for all related children for all age categories of the family head.

Child Poverty by Immigrant Status of the Family Head or Spouse

Figure 12 shows that slightly more than one out of every four poor children is the child of a parent born outside the United States. The poverty rate for these children in 2007 was 26.8%, compared with a rate of 15.7% for children whose family head or spouse was native born. Any children born in the United States are citizens, regardless of their parents' citizenship status.

Figure 12. Poverty Rates Among Related Children, By Immigrant Status of the Family Head and Spouse, 2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

Trends in Child Poverty

Children have been more likely than any other age group in the U.S. to be poor since 1974, when their poverty rate first topped that of the aged. In records dating back to 1959, the incidence of poverty among related children in families has ranged from a peak of 26.9% (1959) to a low of 13.8% (1969). (See Figure 13). In 2007, the rate was 17.6%. Child poverty rates display both cyclical and longer term trends. Except for the 1961-1962 recession, child poverty rates rose during economic slumps, peaking in the year or two after the end of the recession.

Figure 13. Poverty Rates for Related Children, 1959-2007

Source: Congressional Research Service (CRS) with data from the U.S. Bureau of the Census and the National Bureau of Economic Research.

Note: Shaded areas indicate periods of economic recession.

During the years covered by Figure 13, the poverty rate for the aged (not shown) fell from 35.2% in 1959 to 9.7% in 2007.

Trends in Child Poverty Rates, by Type of Family

Figure 14 shows poverty rates of related children, by family type, from 1959-2007. Poverty rates for children in female-headed families have been higher than those for children in male-present (married couple or families with a male-head but no spouse) since poverty data have been recorded. The figure shows that poverty rates for both female-headed families and male-present families fell in the early period (1960s). Since then, much of the variation in poverty rates among children in male-present families has been cyclical.

Poverty rates for children in female-headed families show little cyclical variation in the first three recessions shown (1961, 1970-1971, 1974-1975). However, by the 1982-1983 recession, poverty rates for children in female-headed families do begin to exhibit cyclical increases and decreases, likely attributable to increased labor force participation of women. Poverty rates for children in female-headed families also show pronounced secular (noncyclical) patterns. Rates during the entire economic expansion of the 1980s were higher than in the mid- and late-1970s, coincident with the increase in the number of children of never-married mothers, who have high poverty rates compared to children in other types of female-headed families.

From the mid-1990s to 2000, in the wake of the 1996 welfare reform law, the drop in the poverty rate was more pronounced for children in families headed by a lone mother than for children in families with a male present. However, the increase in the poverty rate for children in single parent families from 2000 to 2007 was also more pronounced than the increase in the rate shown for children in families with a male present.

Figure 14. Child Poverty Rates by Family Type, 1959-2007

Source: Congressional Research Service (CRS) with data from the U.S. Bureau of the Census and the National Bureau of Economic Research.

Note: Shaded areas indicate periods of economic recession.

Trends in the Number and Composition of Poor Children

In 1959, the first year of official poverty data, 17.2 million children were counted as poor (see Figure 15). The poor child population declined through most of the 1960s and hovered around 10 million during the 1970s (with an all-time low of 9.5 million in 1973). During the 1980s the peak number was 13.4 million (1983); and during the 1990s, 15 million (1993). It rose in 2004 to 12.5 million. Since 1970, the number of poor children has fluctuated because of both the economy and demographic trends. The number of related children under 18 fell from 70 million in 1968 to 62 million in 1978. The number of related children began to rise again in 1988, reaching 72.8 million in 2007.

Figure 15 shows the number of poor children from 1959 to 2007 by family type. It shows both the variation in the number of poor children, and its changing composition. In 1959, most poor children and most children lived in married-couple families. Since 1972, the majority of poor children have been in female-headed families.

Figure 15. Number of Poor Related Children, By Family Type, 1959-2007

Source: Congressional Research Service (CRS) with data from the U.S. Census Bureau.

The historical trend toward an increased prevalence of children living in female-headed families has resulted in higher overall child poverty rates than would have otherwise been the case had children's living arrangements been unchanged over the past several decades. In order to approximate the effects of historical changes in living arrangements on the overall child poverty rate, we estimate overall child poverty rates based on the relative composition of children by family type that existed in 1960, while maintaining historically observed child poverty rates by family type.5 Effectively, the adjusted poverty rates present a crude approximation of what the overall child poverty rate might have been had family composition remained unchanged from its 1960 level. Figure 16 shows the effects of these adjustments. The top line of the figure shows historical child poverty rates, whereas the bottom line shows the overall adjusted child poverty rate had child family living arrangements been the same as those observed in 1960. The figure shows, for example, that in 2007 the child poverty rate was 17.6%, but had family composition in 2007 been the same as in 1960, the overall adjusted child poverty rate would have been 12.6%; instead of the observed 12.8 million children being counted as poor in 2007 had family composition remained unchanged from 1960, the number of poor children estimated by this method would have been 9.2 million, or 3.6 million fewer than the number observed.

Figure 16. Child Poverty Rates: Historic Rates and Adjusted Rates, Holding Family Composition at 1960 Levels

Source: Prepared by the Congressional Research Service (CRS). Based on analysis of U.S. Census Bureau Historical Poverty Tables from the Current Population Survey (CPS), Table 10. - Related Children in Female Householder Families by Poverty Status; available at http://www.census.gov/hhes/www/poverty/histpov/hstpov10.html.

Rise in Work by Lone Mothers

Dramatic gains have occurred in recent years in work by lone mothers—especially among those with preschool age children. Employment rates of single mothers with infants or toddlers (under age 3) increased markedly from 1993 through 2000, rising from 35.1% to 59.1% over the period (see Figure 17).6 In 2000 their rate of employment overtook that of their married counterparts—in earlier years these single mothers' rate of employment had lagged behind their married counterparts by as much as 18 percentage points. Similarly, single mothers with somewhat older preschool age children (age 3 to 5) also experienced significant employment gains over most of the 1990s. Among these women, their employment rate rose from 54.1% in 1992, to 72.7% in 2000, overtaking their married counterparts in 1999. Employment rates among single mothers have yet to rebound to the peak levels attained prior to the last economic (2001) recession, marked as beginning in March 2001 and ending in November of that year.

Figure 17. Employment Rates of Married and Single Mothers, by Age of Youngest Child, March 1988 to March 2007

Source: Prepared by the Congressional Research Service (CRS). Based on analysis of U.S. Census Bureau Mar. 1988 through 2007 Current Population Survey (CPS) data.

Note: r = Revised estimates based on expanded CPS sample and 2000 decennial census-derived.

Factors encouraging work by single mothers include a healthy economy during much of the 1990s, the transformation of the family cash welfare program into a work-conditioned and time-limited operation, increases in the EITC and increases in the federal minimum wage. TANF—and preceding state-waivered programs of Aid to Families with Dependent Children (AFDC)—converted cash assistance from a needs-based entitlement to a program of temporary help aimed at promoting work and personal responsibility. Under the TANF block grant, most states reward work by permitting recipients to add to their benefits some (or all) of their earnings, at least for a time. And most states have increased sanctions for failure to perform required work. Increases in the EITC, passed by Congress in 1993 and phased in between 1994 and 1996, have increased the financial incentive for single mothers to work.7 Other factors, such as increased funding for child care subsidies, may also have contributed to making work possible for more single mothers. For more details about trends in welfare, work, and the economic well-being of lone-mother families with children, see CRS Report RL30797, Trends in Welfare, Work, and the Economic Well-Being of Female-Headed Families with Children: 1987-2006, by [author name scrubbed] (pdf).

How Many Single Mothers Are Poor Despite Working?

Figure 18 shows the trend in the distribution of annual hours worked by poor lone-mothers from 1987 to 2007. The figure shows, for example, that poor lone-mothers increased their job attachment during these years, based on annual hours of work. The picture of hours worked mirrors the employment rate numbers depicted earlier, in Figure 17. Hours worked among poor lone-mothers increased over the same period in which their employment rates were increasing. Lone-mothers are not only more likely to be working in the later years than in earlier years, but are likely to be working more hours. For example, in 1995, half of all poor lone mothers did not work, as indicated by the median number of hours worked (i.e., estimated hours worked was zero at the 50th percentile). By 1999 and 2000, half of all poor single mothers (the 50th percentile) were working nearly 480 hours per year or more, and in 2000 20% (80th percentile) were working 1,760 or more hours. The figure shows the decline in hours worked from 2000 to 2005, for all but the top 10% of poor working mothers (90th percentile), probably reflecting the effects of the past (2001) recession on poor mothers' work attachment. Annual hours worked have increased somewhat since 2005 but are still below their 2000 peak.

Figure 18. Annual Hours Worked by Poor Mothers in Mother-Only Families at Various Percentile Rankings of Hours Worked, 1987 to 2007

Source: Estimates prepared by the Congressional Research Service based on U.S. Census Bureau Mar. 1988 through 2008 Current Population Survey (CPS) data.

Overall, among all types of families with related children, Census Bureau data show that the incidence of "full-time work poverty" increased somewhat from 1987 to 2007. Full-time full-year workers are considered to have worked 35 or more hours per week for 50 or more weeks during the year. Among all children, the rate at which at least one of their parents were full-time, full-year workers increased from 71.8% in 1987 to 79.8% in 2000, and stood at 77.4% in 2007 (see Table 1). The decline since 2000 was due, at least in part, to the 2001 recession and unemployment in its aftermath. Tracking this trend, the percent of poor children in families where one parent was a full-time, year-round worker rose from 19.6% of poor children in 1987 to 35.4% of poor children in 2000, before falling to 30% in 2003. It stood at 32.4% in 2007. The share of children in families with lone mothers who worked full-time year-round also increased fairly sharply, from 8.3% in 1987 to 21.6% in 2000, falling to 16.6% in 2003, before rising again to 20.4% in 2007.

Table 1. Percent of Related Children in Families Below the Poverty Threshold with at Least One Full-Time, Year-Round Worker, 1987-2007

Year

Percent of Children in Families with a Full-Time, Year-Round
Working Parent

Percent of Poor Children with a Full-Time,
Year Round Working Parent

All Families

Married Couple Families

Mother-Only Families

1987

71.8

19.6

36.5

8.3

1988

72.7

22.6

40.8

10.3

1989

72.3

22.4

41.2

9.4

1990

72.5

22.6

39.2

9.7

1991

71.3

20.8

37.3

10.4

1992

71.7

19.9

36.0

9.9

1993

71.4

21.5

38.8

9.6

1994

72.7

24.3

45.5

10.9

1995

73.9

25.4

44.6

14.1

1996

74.9

24.8

47.3

10.2

1997

75.5

25.4

45.7

13.2

1998

76.6

31.5

55.0

17.6

1999

78.4

31.4

51.4

18.6

2000

79.8

35.4

57.8

21.6

2001

78.4

31.9

53.3

18.7

2002

77.6

33.4

49.1

19.0

2003

76.6

30.0

52.6

16.6

2004

77.4

32.7

55.7

18.6

2005

77.8

32.1

56.3

17.9

2006

78.1

32.8

57.9

19.2

2007

77.4

32.4

54.4

20.4

Source: Table prepared by the Congressional Research Service, based on data from the U.S. Census Bureau.

Government "Safety Net" Policy

The framework for federal cash income support policy can be found in the Social Security Act and the Internal Revenue Code. A two-tier safety net was put into place in the 1935 Act. The first tier, consisting of Old-Age Insurance (usually thought of as social security) (Title II) and unemployment insurance (Title III) aimed to protect families from the economic risks associated with the retirement or unemployment of their workers. Workers earned rights to these social insurance benefits by paying payroll taxes in a covered job. The second tier made grants to states to help make means-tested payments to specified categories of needy persons not expected to work, namely, the aged (Title I),8 children (with only one able-bodied parent in the home) (Title IV), and the blind (Title X).

Added to the social insurance tier over time were Survivors' Insurance (1939 Social Security Act Amendments), Disability Insurance (1956 Amendments), and Medicare (1965 Amendments) health insurance for the elderly. Added to the welfare tier over time was aid for needy persons who were permanently and totally disabled (1952), replaced in 1972 by a 100% federal cash program for the aged, blind, and disabled called SSI. Health insurance for low-income aged, blind, and disabled persons and for needy families with children was added in the form of the Medicaid program (1965 Amendments). Health insurance for low-income children ineligible for Medicaid was added by enactment of the State Children's Health Insurance Program (SCHIP) (Balanced Budget Act of 1997). Also, the cash welfare program for needy children was opened up to some unemployed two-parent families (1961) and, finally (1996) was replaced by a block grant for temporary assistance.

The social insurance programs are designed to provide benefits when a family loses earnings because its breadwinner is permanently or temporarily out of the labor market (through death, disability, unemployment). At the outset the welfare programs were restricted to persons not expected to work, but over time, work requirements have been added to the cash welfare program for families with children, and it now emphasizes moving families from welfare to work.

In 1975, Congress enacted a program to explicitly support and supplement the income of working poor parents—the EITC. This provision of the tax code makes payments from the Treasury to parents whose credit exceeds any income tax liability. In tax year 2005, the EITC was claimed by 22.8 million tax filers, with credits totaling $42.4 billion.9

In addition, child care subsidies have been expanded for families receiving cash welfare by the Family Support Act of 1988, the Omnibus Budget Reconciliation Act of 1990, which created child care programs for the working poor, and the 1996 welfare reform law, which consolidated child care funding. FY2008 appropriations for child care totaled $5 billion ($2.9 billion for the mandatory child care block grant and $2.1 billion for the discretionary Child Care and Development Block Grant).10

Finally, the federal child support enforcement program (begun in 1975) has shifted its role from reimbursing federal and state governments for welfare costs to facilitating income transfers to families with children where one parent (usually the father) is noncustodial. The increased role of child support is particularly important in light of the growth of female-headed families. In FY2006, child support enforcement offices collected $24 billion: $0.9 billion for TANF cash welfare families, $9.5 billion for former TANF cash welfare families, $3.1 billion for families receiving Medicaid (and no TANF), and $11.3 billion for families that never received TANF cash welfare.11

Social Security

As noted above, the Social Security Act established an old-age insurance system for workers; and later, social security was enlarged to cover dependents and survivors of retired or disabled workers.12 With these additions, social security became a system of comprehensive insurance for all families with a worker who paid social security payroll taxes.13 If the family breadwinner died or became disabled, social security would provide cash for his dependents and survivors. It was widely hoped that coverage by work-related social insurance eventually would eliminate most need for cash relief to families who lost their breadwinner.

In September 2008, there were 50.6 million social security recipients, of which 3.1 million were children.14 This means the federal government makes social security payments to about 4% of U.S. children. Total benefits to children totaled nearly $1.5 billion in that month, a rate of $17.5 billion per year. In addition, some adult social security beneficiaries are in families with children, so that this program reaches an even greater proportion of the child population.15

Unemployment Insurance

The Social Security Act also established the federal-state unemployment insurance (UI) program. This program provides temporary unemployment benefits to workers who are unemployed through no fault of their own, provided they have earned a state-determined sum of wages during an established base period, usually the first quarter of the last five completed calendar quarters, and are available for work.

Historically there has been concern that UI eligibility rules present barriers to low-wage workers with unstable work histories. To what extent might unemployment benefits be available for former welfare recipients who lose a job?16 Studies of former cash recipients in a number of states conclude that most recipients who leave welfare for work have sufficient earnings to qualify for UI at some point after leaving welfare.17 However, having sufficient earnings is only one of the qualifying conditions for receiving UI upon losing a job. The job loss must be considered as occurring through no fault of the potential recipient, a definition that varies among the states. A study of welfare leavers in New Jersey, considered a fairly liberal state with respect to UI eligibility rules, found that as many as 60% of those with monetary eligibility might have been disqualified for other reasons—especially the high rates of voluntarily quitting a job. It found that about half of the job quitters did so for a personal reason, such as a health problem, having to care for a child at home, or a transportation issue. Advocates of broadening UI rules to qualify more low-wage workers have recommended that states permit job quits for "good reason."

Cash Public Assistance for Children

Before passage of the Social Security Act in 1935, some states offered "mothers' pensions" so that needy mothers could stay home to raise their children. This aid was largely restricted to "paternal orphans," children whose father had died.18 In response to the Great Depression, the Social Security Act provided federal funds to enable states to help certain needy groups, including children whose second parent was dead, incapacitated, or continually absent from home—Aid to Dependent Children (ADC).19 The original purpose of AFDC was to help states enable needy mothers to be full-time caregivers at home. Thus, it had no expectation of mothers' work and no requirement for it. In fact, the law penalized work by requiring states to reduce benefits by the full amount of any earnings (including those spent on work expenses).

However, this changed over the years. Gradually the assumption that AFDC mothers belonged at home faded as more and more nonwelfare mothers went to work and as rising numbers of unwed mothers joined AFDC, altering the character of the caseload.

Since the early 1960s, Congress has tried to promote work and self-support of welfare families by means of work requirements, financial incentives,20 and various services. The first work rule (1961) applied to unemployed fathers (a group that Congress admitted to AFDC—at state option—that year). The rule required states to condition AFDC for Unemployed Fathers (AFDC-UF) on acceptance of work. In 1967, Congress established a Work Incentive Program (WIN) for AFDC families and required states to assign "appropriate" persons to this education and training program. The 1967 law required states to give AFDC parents who went to work a financial reward: disregard of some earnings when calculating benefits.21 In 1971, Congress removed from states the discretion to decide who must work or train. Instead, it specified that states must assign to WIN all able-bodied custodial parents except those with a preschool child, under age six. In the Family Support Act of 1988, Congress lowered the young child age threshold for work exemption. It required states to assign to a new Jobs and Basic Skills (JOBS) Training program for AFDC families all able-bodied custodial parents except those with a child under age three (under age one, at state option). However, the law required states to "guarantee" child care for children below age six.

Finally, in 1996, Congress replaced AFDC with the TANF block grant. TANF law requires states to engage certain percentages of adult recipients in specified "work activities." The law exempts no one from work participation, but it permits states to exempt a single parent caring for a child under age one (and to exclude that parent in calculating the state's work participation rate). In addition, the law bars a state from penalizing a single parent with a child under age six for failure to engage in required work if the person cannot obtain needed child care because appropriate care is unavailable (or available care is unsuitable). States decide individual participation rules.

In June 2008, 3.0 million children were in families that received cash benefits from TANF programs or separate state-funded (TANF) programs. This is down from a historic peak of 9.6 million children in families receiving cash benefits from AFDC in 1994. In FY2006, federal and state spending on cash welfare benefits totaled $9.9 billion—a little less than half the amount spent on cash welfare by the federal government and states back in the mid-1990s.22

Money Income Poverty Rates of Children, by Income Source

Although there are numerous government "safety net" programs that aid families with children, the chief component of money income for most families with children is earnings. The first bars in Figures 19 and 20 show what child poverty rates in 2007 would have been if families had no cash income other than earnings. The rates would have been 21.8% for all related children and 52.4% for related children in female-headed families. The last bars show the official money income poverty rates, 17.6% for all related children and 43.0% for children being raised by the mother alone. In succession, the intervening bars show the poverty-reducing contributions of (1) cash from other family effort (property income, private pensions, child support and alimony), and (2) government cash transfers (social insurance, other cash benefits, and cash welfare). In general, the official poverty rates are about 20% lower than market income (earnings only) poverty rates. Figures 19 and 20 show only sources of pre-tax money income (income counted in determining official poverty rates). They exclude noncash aid and tax benefits.

Figure 19. Poverty Rates for All Related Children, 2007, by Source of Income

Conclusion

This report examined both the role of work and the role of government income supports as factors affecting the official child poverty rate. Work is the principal means by which families with children support themselves. Child poverty rates are correlated with factors associated with wage rates, such as parent(s)' educational attainment and work experience; the amount of work done during the year; and family type, which affects the likelihood that a family will have a second earner. Without family earnings, a child is almost certain to be poor.

However, earnings alone often fail to overcome poverty. In 2006, one-third of all poor children lived with at least one adult who was a full-time, full-year worker. The economy has many jobs that pay low wages. Parents in such jobs may escape poverty only through job advancement that results in higher wages, possibly only by upgrading skills and education while working. Often, it takes work of both parents to move their children and families out of poverty. This often comes at the cost of arranging child care to permit both parents to work outside of the home. Additionally, work is not always steady, with some parents experiencing spells of joblessness for part of the year.

Access to the first tier of the "safety net," social insurance, is restricted to families with a current or previous wage earner. Social insurance benefits are established through work, and are characterized as earned rights. Although these programs are not targeted to the poor per se, these programs—which partially replace earnings because of the death, old age, disability or involuntary unemployment of a worker—have a significant impact on reducing poverty among families with children. Over the past 20 years, the social insurance tier of the safety net has remained relatively static. The protections of its cash benefit programs have generally been maintained. However, the social insurance tier has not been revised to take into account changes in the economy, family structure, and work patterns of parents.

In contrast, the second tier of the "safety net," programs targeted to low-income families and persons, has undergone a radical transformation over the past 20 years. Cash welfare for needy families with children has increasingly been tied to a philosophy of mutual obligations—adult recipients have been expected to engage in either work or activities to move them into work. In the wake of welfare reforms made at the federal and state levels in the mid-1990s, the cash welfare rolls plummeted, so that children in families receiving cash assistance represent a small and decreasing share of both the overall and poverty child populations. Need-tested assistance increasingly is paid to families only in the form of noncash benefits (Medicaid and food stamps) whose value is not reflected in the official child poverty statistics. These programs cast an uneven net of basic support to families. Moreover, when parents receiving need-tested benefits go to work, benefits are often significantly reduced as earnings increase, at times creating a perverse disincentive to work. For many families with limited earnings capacity, the route toward self-sufficiency can be a steep climb, yielding only small economic gain for additional work effort.

Along with the transformation of need-tested benefits have come expansions of refundable tax credits for families with children, particularly the Earned Income Tax Credit (EITC). As a work support, the EITC helps offset the financial disincentives relating to work that poor families have traditionally faced under welfare programs. For some families with very low earnings, the EITC may supplement up to 40 cents on every dollar a parent earns. EITC payments (either advance payments, tax refunds, or reductions in tax liabilities that would otherwise be owed) are not included in the official poverty statistics. However, for a single parent with two children, who works full-time, all year at the 2006 federal minimum wage, those earnings combined with food stamps and the EITC are sufficient to lift that family to just above the poverty threshold.

Thus, recent changes to taxes and benefits targeted to low-income families have sought to increase the rewards to work. However, the world of work does not guarantee steady income. Working parents face risks to financially supporting their children, stemming from job loss associated with either cyclical or structural changes in the economy or interrupted periods of work because of their own or a family members' illness. Moreover, there remain a small but not insignificant group of families with children where no parent has an attachment to the labor force. This creates a series of policy dilemmas which include how to balance protections against economic risk while maintaining a policy that rewards work, and helping children in those families without a worker without undermining the efforts of working parents.

Appendix. Support Tables to Selected Figures in This Report

Table A-1. Poverty Rates and Number of Poor Children, by Family Type, 1959-2007

(Support Table for Figures 13-15)

Year

Poverty Rates for Related
Children Under Age 18

Number of Poor Related
Children under Age 18
(in thousands)

All Families

Female-Headed Families

Male-Present Families

All Families

Female-Headed Families

Male-Present Families

1959

26.9

72.2

22.4

17,208

4,145

13,063

1960

26.5

68.4

22.3

17,288

4,095

13,193

1961

25.2

65.1

21.0

16,577

4,044

12,533

1962

24.7

70.2

19.9

16,630

4,506

12,124

1963

22.8

66.6

18.0

15,691

4,554

11,137

1964

22.7

62.3

18.2

15,736

4,422

11,314

1965

20.7

64.2

15.7

14,388

4,562

9,826

1966

17.4

58.2

12.6

12,146

4,262

7,884

1967

16.3

54.3

11.5

11,427

4,246

7,181

1968

15.3

55.2

10.2

10,739

4,409

6,330

1969

13.8

54.4

8.6

9,501

4,247

5,254

1970

14.9

53.0

9.2

10,235

4,689

5,546

1971

15.1

53.1

9.3

10,344

4,850

5,494

1972

14.9

53.1

8.6

10,082

5,094

4,988

1973

14.2

52.1

7.6

9,453

5,171

4,282

1974

15.1

51.5

8.3

9,967

5,387

4,580

1975

16.8

52.7

9.8

10,882

5,597

5,285

1976

15.8

52.0

8.5

10,081

5,583

4,498

1977

16.0

50.3

8.5

10,028

5,658

4,370

1978

15.7

50.6

7.9

9,722

5,687

4,035

1979

16.0

48.6

8.5

9,993

5,635

4,358

1980

17.9

50.8

10.4

11,114

5,866

5,248

1981

19.5

52.3

11.6

12,068

6,305

5,763

1982

21.3

56.1

13.0

13,139

6,696

6,443

1983

21.8

55.4

13.5

13,427

6,747

6,680

1984

21.0

54.0

12.5

12,929

6,772

6,157

1985

20.1

53.6

11.7

12,483

6,716

5,767

1986

19.8

54.4

10.8

12,257

6,943

5,314

1987

19.7

53.7

10.6

12,275

7,019

5,256

1988

19.0

52.9

10.0

11,935

6,955

4,980

1989

19.0

51.1

10.4

12,001

6,808

5,193

1990

19.9

53.4

10.7

12,715

7,363

5,352

1991

21.1

55.4

11.1

13,658

8,065

5,593

1992

21.6

54.6

11.8

14,521

8,368

6,153

1993

22.0

53.7

12.4

14,961

8,503

6,458

1994

21.2

52.9

11.7

14,610

8,427

6,183

1995

20.2

50.3

10.7

13,999

8,364

5,635

1996

19.8

49.3

10.9

13,764

7,990

5,774

1997

19.2

49.0

10.2

13,422

7,928

5,494

1998

18.3

46.1

9.7

12,845

7,627

5,218

1999

16.6

41.5

9.1

11,510

6,602

4,984

2000

15.6

40.1

8.6

11,005

6,300

4,705

2001

15.8

39.3

8.8

11,175

6,341

4,834

2002

16.3

39.6

9.2

11,646

6,564

5,082

2003

17.2

41.8

9.6

12,340

7,085

5,255

2004

17.3

41.9

9.7

12,473

7,152

5,321

2005

17.1

42.8

9.3

12,335

7,210

5,125

2006

16.9

42.2

9.0

12,299

7,341

5,958

2007

17.6

43.0

9.5

12,802

7,546

5,256

Source: Table prepared by the Congressional Research Service, based on data from the U.S. Census Bureau.

Table A-2. Employment Rates of Single Mothers and Married Mothers, by Age of Youngest Child, March 1988 to March 2008

(percent of single mothers employed in March, Support Table for Figure 17)

Year

Single Mothers

Married Mothers

With a Child under Age 18

Youngest Child under
Age 3

Youngest Child Age 3 to 5

Youngest Child Age 6 to 17

With a Child under Age 18

Youngest Child under
Age 3

Youngest Child Age 3 to 5

Youngest Child Age 6 to 17

1988

57.4

35.1

52.9

69.1

61.8

50.7

58.1

69.6

1989

58.2

37.9

53.1

70.0

63.0

51.4

60.8

70.6

1990

60.3

38.0

61.0

70.9

63.4

52.7

60.9

70.8

1991

58.1

36.6

55.7

70.2

63.1

52.7

60.5

70.5

1992

57.3

35.2

54.1

69.8

63.9

53.1

59.4

71.9

1993

57.3

35.1

54.8

70.1

63.9

53.2

59.4

71.9

1994

58.0

37.7

55.2

69.3

65.5

56.0

61.2

72.6

1995

61.1

43.1

58.5

70.5

67.1

57.4

63.9

73.4

1996

63.5

44.7

60.4

72.9

67.6

58.2

63.3

74.2

1997

65.6

51.5

64.3

72.0

68.5

58.3

64.4

75.2

1998

68.8

54.8

63.7

76.4

67.9

58.3

64.1

74.2

1999

70.7

55.8

69.8

77.1

67.9

57.0

63.1

75.1

2000

72.8

59.1

72.7

78.5

68.4

56.8

66.0

75.0

2001

73.0

56.1

74.4

79.8

68.5

57.1

64.7

75.4

2001r

72.5

57.6

71.3

79.1

68.0

56.0

64.2

75.1

2002

71.2

57.9

71.0

76.3

66.7

54.9

61.7

74.1

2003

69.6

54.8

69.3

75.4

66.3

53.5

61.7

74.2

2004

69.7

54.1

69.5

75.7

65.3

52.4

62.3

72.7

2005

68.9

53.7

66.7

75.7

65.9

54.9

61.9

72.8

2006

69.6

57.0

68.0

75.1

66.1

55.4

61.8

73.1

2007

70.0

56.5

68.6

76.0

67.3

56.9

63.0

74.2

2008

69.1

54/5

68.5

75.5

67.0

56/1

63.6

73.7

Source: Congressional Research Service (CRS) estimates based on analysis of U.S. Census Bureau Mar. 1988 to 2008 Current Population Survey (CPS) data.

Note: r = Revised estimates based on expanded CPS sample and 2000 decennial census-derived weights.

Table A-3. Annual Hours Worked by Poor Mothers in Mother-Only Families at Various Percentile Rankings, 1987 to 2007

(Support Table for Figure 18)

Year

10th Percentile

20th Percentile

30th Percentile

40th Percentile

50th Percentile (Median)

60th Percentile

70th Percentile

80th Percentile

90th Percentile

1987

0

0

0

0

0

80

450

1,040

1,680

1988

0

0

0

0

0

120

480

936

1,716

1989

0

0

0

0

0

128

480

1,040

1,664

1990

0

0

0

0

0

208

600

1,080

1,680

1991

0

0

0

0

0

192

546

1,040

1,800

1992

0

0

0

0

0

120

550

1,040

1,664

1993

0

0

0

0

0

120

480

1,040

1,760

1994

0

0

0

0

0

252

640

1,150

1,820

1995

0

0

0

0

0

340

784

1,320

2,080

1996

0

0

0

0

80

432

800

1,300

1,840

1997

0

0

0

0

255

600

1,040

1,456

2,080

1998

0

0

0

21

400

700

1,040

1,584

2,080

1999

0

0

0

132

480

840

1,200

1,600

2,080

2000

0

0

0

60

480

875

1,280

1,760

2,080

2001

0

0

0

0

320

780

1,144

1,664

2,080

2002

0

0

0

0

320

780

1,170

1,664

2,080

2003

0

0

0

0

260

660

1,040

1,560

2,080

2004

0

0

0

0

225

640

1,040

1,600

2,080

2005

0

0

0

0

180

600

1,040

1,560

2,080

2006

0

0

0

0

209

680

1,152

1,640

2,080

2007

0

0

0

0

320

784

1,200

1,610

2,080

Source: Estimates prepared by the Congressional Research Service based on U.S. Census Bureau, Mar. 1988 through 2008 Current Population Survey (CPS) data.

Footnotes

1.

For an overview of research on the consequences of poverty for children, see Duncan, Greg J. and Jeanne Brooks-Gunn ed. Consequences of Growing Up Poor. Russell Sage Foundation, 1997. Also, for an overview of the effects of poverty on health and social status and economic growth, see: U.S. Government Accountability Office. Poverty In America: Economic Research Shows Adverse Impacts on Health Status and Other Social Conditions as well as the Economic Growth Rate, GAO-07-344, January 2007.

2.

The Census Bureau defines related children under age 18 as children who are related to the household head. In March of 2007, there were 74.0 million persons under the age of 18, with 72.8 million meeting the definition of a related child. The remaining 1.2 million persons under the age of 18 who are not considered related children include (1) those who live in families all of whose members are not related to the head of the household; (2) children who are not living with relatives (e.g., foster children); and (3) persons under the age of 18 who are themselves household heads or are spouses of household heads.

3.

U.S. Census Bureau, Living Arrangements of Children Under 18 Years Old, 1960 to Present, Table CH-1, Internet release date September 15, 2004. See http://www.census.gov/population/socdemo/hh-fam/tabCH-1.pdf.

4.

U.S. Census Bureau, data at http://www.census.gov/population/www/socdemo/educ-attn.html.

5.

For example, in 1960, 9.2% of children lived in female-headed families, but in 2007, 24.1% lived in such families. In 2007, the poverty rate for children in female-headed families was 43.0%. If, in 2007, 9.2% of children had lived in female-headed families (i.e., the same share as in 1960), rather than the 24.1% observed, they would have numbered 6.7 million, rather than the 17.6 million observed. Assuming that the child poverty rate remained at its 2007 level (i.e., 43.0% of children in such families being considered poor), then an estimated 2.9 million children in female-headed families would have been poor (i.e., .430 x 6.7 million) rather than the 7.5 million observed (i.e., .430 x 17.6 million). Considering then children living in male-present families, their estimated number in 2007, assuming 1960 incidence rates, would have been 66.1 million, rather than the 55.2 million observed. Applying the child poverty rate of 9.5%, in 2007, for children living in male-present families, to the number of children who would have been estimated to be living in such families when adjusted to 1960 incidence levels (66.1 million), the number of poor children in such families would be 6.3 million, rather than the observed 5.3 million. Although the estimated number of poor children in male-present families would be about 1 million higher, if family composition in 2007 was the same as in 1960, the number in female-headed families would be 4.7 million lower, for a net effect of 3.6 million fewer poor children in 2007, than what was actually observed. Based on these adjustments, the total number of poor children in 2007 would have been 9.2 million, rather than the observed 12.8 million, resulting in a child poverty rate of 12.6%, rather than the observed 17.6%.

6.

Employment rates are for March of the indicated year.

7.

Bruce D. Meyer and Dan T. Rosenbaum attribute 60% of the increase in single mothers' weekly and annual employment between 1984 and 1996 to the EITC. See their paper, "Welfare, the Earned Income Tax Credit, and the Labor Supply of Single Mothers," NBER Working Paper No. 7363, September 1999.

8.

Aid to the needy aged was placed in Title I as the most popular provision of the Social Security Act.

9.

See CRS Report RL31768, The Earned Income Tax Credit (EITC): An Overview, by [author name scrubbed].

10.

CRS Report RL30785, The Child Care and Development Block Grant: Background and Funding, by [author name scrubbed].

11.

See CRS Report RS22380, Child Support Enforcement: Program Basics, by [author name scrubbed].

12.

Child survivor benefits were first paid in January 1940.

13.

Until 1950, social security omitted farm and domestic workers.

14.

Source of this data is the Social Security Administration's web site, http://www.ssa.gov/OACT/ProgData/icp.html

15.

See CRS Report RL33289, Social Security's Effect on Child Poverty, by [author name scrubbed].

16.

For a discussion of unemployment insurance issues raised by the welfare-to-work movement, see Rethinking Income Support for the Working Poor, Perspectives on Unemployment Insurance, Welfare, and Work, edited by Evelyn Ganzglass and Karen Glass (Washington, DC, National Governors Association, 1999).

17.

See Anu Rangarajan, Carol Razafindrakoto, and Walter Corson, Study to Examine UI Eligibility Among Former TANF Recipients: Evidence from New Jersey, Final Report (Princeton, N.J., November 2002, Mathematica Policy Research, Inc.). This study was prepared for the U.S. Department of Health and Human Services and the Department of Human Services of New Jersey. See also Anu Rangarajan and Carol Razafindrakoto, Unemployment Insurance as a Potential Safety Net for TANF Leavers: Evidence from Five States (Princeton, N.J., September 2004, Mathematica Policy Research, Inc.). Prepared for the Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation.

18.

In 1931, widows made up 82% of the needy mothers who received state aid (mothers' pensions) and whose marital status was known. Winifred Bell, Aid to Dependent Children (New York and London: Columbia University Press, 1965), p. 9.

19.

In 1962, the program was renamed Aid to Families with Dependent Children (AFDC).

20.

In 1962, Congress removed the financial work penalty by requiring states to permit working parents to add to their AFDC grant "any expenses reasonably attributed to the earning of income."

21.

States were to disregard the first $30 monthly in earnings plus one-third of the remainder. By regulation, the "reasonable" work expense deduction, including child care cost (already in law) was added to the new disregard and taken last. In 1981, Congress capped the general work expense disregard, added a capped child care disregard, and specified that these disregards must be taken first; it also put a time limit on the one-third disregard. These changes reduced the earnings level at which a person lost AFDC eligibility.

22.

For updated statistics on cash welfare, see CRS Report RL32760, The Temporary Assistance for Needy Families (TANF) Block Grant: Responses to Frequently Asked Questions, by [author name scrubbed] (pdf).