Well-Being of Children in Working Poor and Other Families: 1997 and 2004: Research Brief

By:
Richard Wertheimer, Kristin Anderson Moore, and Mary Burkhauser
Child Trends

Prepared for:
U.S. Department of Health and Human Services (HHS)
Office of the Assistant Secretary for Planning and Evaluation (ASPE)

Overview

When Congress reformed the welfare system in 1996, major goals of the legislation were to increase employment and income of needy families and to decrease child poverty.  Another major goal was to improve child outcomes through increased parental employment and earnings along with other provisions of welfare reform.⁽¹⁾  However, there was also concern that increased work effort by single mothers would lead to less time spent with their children and that some child outcomes might deteriorate.⁽²⁾

Drawing on work requirements in the welfare reform legislation, we have defined substantial work effort as 1,820 hours per year (based on the requirement of 35 hours of work per week) for two-parent families and 1,040 hours per year (based on the requirement of 20 hours of work per week) for single-parent families. Poor families making a substantial work effort are designated working poor; those not making a substantial work effort are designated non-working poor.  Our analysis finds that, between 1997 and 2004, the well-being of children in working poor families improved significantly for 10 of the 15 measures available in both years and remained stable for the remaining measures.  In contrast, the well-being of children in non-working poor families improved significantly for only five measures and deteriorated significantly for four measures.  Moreover, whereas the well-being of children in working poor families was not consistently better than the well-being of children in non-working poor families in 1997, by 2004, the well-being of children in working poor families was better than for children in non-working poor families for 12 of the 17 measures available for that year. These patterns held when social and economic factors are accounted for statistically.

These findings suggest that the increase in working poor families share of all poor families⁽³⁾ has not led to deteriorating child outcomes and indeed is more consistent with the reverse  that increased work effort among low-income families  is associated with better child outcomes.

Introduction

When the U.S. welfare system was reformed in 1996, two explicit objectives were increasing employment and earnings of needy families and decreasing child poverty.  It was hoped that achieving these objectives would lead to improved outcomes for children.  Outcomes measured in this brief fall into three categories: (1) measures of how well a child is developing;  (2) measures associated with how well a child will develop; and (3) measures of interaction with the community. Measures of well-being used in this study are from the Survey of Income and Program Participation and are shown in the box on the next page.

The twin purposes of this brief are to:

1. Explore the changes in the well-being of children in working poor families and other families (of all income levels) between 1997 and 2004; and
2. Compare the 2004 well-being of children in working poor families with non-working poor families.

Changes in child well-being between 1997 and 2004

The well-being of children in working poor families improved significantly for 10 of 15 measures that were available in both years.  For the remaining five measures, there was no statistically significant change.

In contrast, for the same 15 measures, the well-being of children in non-working poor families improved significantly for only five of 15 measures and deteriorated for four measures.

For example, as shown in Figure 1, among children in working poor families, the percentage in special classes for gifted students increased from 9 percent in 1997 to 14 percent in 2004.  Measures which improved for children in working poor families (see Tables 1-3) include:

How well a child is developing

• Special classes for gifted students

How well a child will develop

• Meals with mother
• Meals with father
• Parental aggravation
• Mother involvement
• Father involvement

Interaction with community

• Participation in extracurricular activities
• School engagement
• Negative parental attitude toward community
• Ever attended kindergarten

In contrast, among children in non-working poor families, the percentage of children in special classes for gifted students decreased from 12 percent to 9 percent.  Detailed 1997-2004 comparisons for all measures are presented in Tables 1-3.  These tables also present results for children in working families with incomes between 100 percent and 200 percent of the official poverty threshold and for children in working families with incomes greater than 200 percent of the poverty threshold.⁽⁴⁾

Figure 1.
Percentage of children in special classes for gifted students by family work and poverty status, 1997 and 2004

Differences in well-being between children in working poor families and children in non-working poor families in 2004

The well-being of children in working poor families was better than for children in non-working poor families in 12 of the 17 measures available in 2004 that reflect child well-being or have been shown by research to be associated with child well-being.  Children in working poor families have an advantage over children in non-working poor families in the following areas:

How well a child is developing

• Special classes for gifted students
• Ever repeated a grade
• Ever suspended or expelled from school

How well a child will develop

• Parental aggravation in parenting
• Mother involvement with child
• Mothers educational aspirations for child
• Fathers educational aspirations for child
• Ever lived apart from parents

Interaction with community

• Childs participation in extracurricular activities
• Childs school engagement
• Positive parental attitude toward the community
• Negative parental attitude toward community

See Figure 2 and Tables 1-3.

Figure 2.
Selected well-being indexes by family work and poverty status, 2004

In contrast, in 1997, children in working poor families had an advantage over children in non-working poor families only in the two variables measuring the parents attitude toward the community and were at a disadvantage in four measures.⁽⁵⁾

Discussion

The well-being of children in working poor families improved between 1997 and 2004 for 10 of the 15 measures included in the Survey of Income and Program Participation. Moreover, the well-being of children in working poor families was better than the well-being of children in non-working poor families for 12 of 17 measures  in sharp contrast with findings for 1997 which indicated that children in working poor families had an advantage in only two measures and a disadvantage in four.⁽⁵⁾

It is also well established that the employment rate of single mothers substantially increased around the time of and immediately after welfare reform legislation was passed in 1996.

At the time welfare reform was implemented in 1997, there was concern that increased work by mothers, in response to the work requirements imposed by the new Temporary Assistance to Needy Families (TANF) program, might lead to deterioration for measures of child well-being, since mothers would have less time available to spend with their children.  While work effort of single mothers did indeed increase, the findings reported here suggest instead that many child well-being measures have improved for working poor families.  Deterioration of child well-being measures was confined to the children in non-working poor families.

Because we were concerned that some of the differences in well-being between children in working poor families and children in non-working poor families might be due to differences between these two groups in their composition by parental education, race/ethnicity, family structure, and parental age, we performed analyses for selected measures⁽⁶⁾ of well-being in which these variables were controlled.  For 10 of the 13 measures subjected to these analyses, differences that were statistically significant in Tables 1-3 were also significant in the analyses after these variables were controlled.⁽⁷⁾  Our analyses indicate that most of the differences in child well-being between children in working poor families and children in non-working poor families are not due to differences in the composition of these groups by parental education, race/ethnicity, family structure, and parental age.  These findings are consistent with considerable research indicating that, after infancy, maternal employment is not related to poorer development for children and, indeed, is often related to better child outcomes for lower income women.⁽⁸⁾ Nevertheless, we note that causal conclusions based on these data are not possible because the data are not experimental, and numerous other social and economic changes were occurring over these same years.

In conclusion, our findings suggest that increased work effort by poor families is not associated with deteriorating child outcomes and indeed is more consistent with the reverse  that increased work effort is associated with improved child outcomes.

Endnotes

¹ The Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996 (H.R. 3734-9), Sec. 401, states that one of the purposes of the Temporary Assistance for Needy Families (TANF) block grants is to promote job preparation, work, and marriage. It also provides that recipients participate in work activities while receiving TANF benefits and requires that a beneficiary work after receiving benefits for 24 months.  It also provides (Section 411) that the Secretary of the Department of Health and Human Services transmit to Congress an annual report describing whether the States are meeting . . .  the objectives of increasing employment and earnings of needy families . . . .  Finally, Section 413 provides that the Secretary  may assist States in developing . . . innovative approaches for reducing welfare dependency and increasing the well-being of minor children living at home with respect to recipients of assistance under programs funded under this part.

² For example, a 1998 review of research studies that predate welfare reform raised concerns about findings of negative outcomes for children in low-income families when employment is initiated during the first year of a childs life; and that child outcomes among employed mothers vary according to maternal wage level and that the quality of the home environment provided to young children can decline when mothers begin jobs that are low-wage and involve repetitive, unstimulating tasks.  See Zaslow, M., Tout, K., Zaslow, M. & Moore, K.A. (1998), Welfare Reform and Children: Potential Implications.  Washington, DC: The Urban Institute.

³ The percentage of children in working poor families as a percentage of children in all poor families increased from 37 percent in 1997 to 57 percent in 2004.  Thus, the working poor in 2004 probably include a sizable number of families who, in the absence of welfare reform, would have not made a substantial work effort.

⁴ For children in working families with incomes between 100 percent and 200 percent of the poverty threshold, eight of 15 measures increased, while the corresponding ratio for children in working families with incomes over 200 percent of the poverty threshold was nine of 15 measures.

⁵ Wertheimer, R., Long, M., & Jager, J.  2002.  Children in working poor families: update and extensions. Child Trends report to the Foundation for Child Development.

⁶ Multivariate analyses were restricted to well-being measures that couldnt have been observed prior to the year (2004) in which work and income were observed.  For example, a child could have been suspended or expelled from school in 2003 or even several years before 2004.  Thus, for a variable measured this way, it was not logically possible that the work and income of the family in 2004 affected the likelihood of a child being suspended or expelled unless the child was suspended or expelled in 2004.  In contrast, the survey asks about television rules in 2004  the same year that it asks about family work and income. Thus, this variable could be included in the multivariate analyses.

⁷ In nine of the 10 measures where the difference remained significant, the differences decreased somewhat; in the other case it increased.

⁸ Wertheimer, R., Jekielek, S., Moore, K.A., Redd, Z, & Wertheimer, R.F. (2005). Tradeoffs among work, family, health and well-being: a social-demographic perspective, in Raley (ed.), Work, Family, Health and Well-Being.  Mahway, N.J.: Lawrence Erlbaum Associates.