Over the past 30 years, welfare and other public policies for families living in poverty have developed a primary objective of increasing parents' self-sufficiency by requiring and supporting employment, culminating in the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA). This legislation gave states considerable latitude in designing their welfare policies. At the same time, however, there has been very little research on the effects of welfare policies on children to inform decisions policymakers are making. Fortunately, there now is consistent evidence from well-designed studies about whether promoting work among low-income single parents helps or hurts children, and under what conditions it does so. This policy report summarizes the results on children from a synthesis of nearly a dozen welfare experiments aimed at increasing the self-sufficiency of low-income parents. The study finds that: (1) Welfare policies that increase employment, but do not affect income, have few effects on children. These findings are consistent with the generally neutral effects of maternal employment for low income children found in the developmental literature, and should be reassuring to those concerned about the negative effects to children of requiring mothers to go to work; and (2) Welfare policies, when designed in ways that increase both parents' employment and income, appear to benefit elementary school children, particularly children's school achievement. The effects are small, but notable, analogous to increasing children's test scores from the 25th to the 30th percentile. These findings present policymakers with a choice: either to implement policies that reduce welfare caseloads, increase employment and have limited effects on young children, with limited government costs or to implement policies that increase employment and income, and, in turn, benefit young children, with greater cost to the government. One caution emerges, however: Emerging findings on adolescent children suggest that programs may be less beneficial for adolescents than for children in middle childhood, suggesting that the way in which programs affect children may be shaped by children's developmental stage. Adolescents were only examined systematically in two studies, but those studies suggested increased adolescent problem behavior (drinking, smoking, minor delinquency) and increased problems in school when parents move from welfare into employment. These findings point to an effective role for the Temporary Assistance for Needy Families (TANF) surpluses, should they be maintained. For states that are interested in using welfare policy to improve children's school achievement, policies that supplement the earnings of low-income workers might be an important complement to programs aimed directly at improving the school outcomes of children. [This issue of "Social Policy Report" also contains "Special Commentary: 'U.S. Attacked.' Headline of 'New York Times,' September 12, 2001. Children's Responses" (Lonnie Sherrod and Jeanne Brooks-Gunn).]