[Economic Report of the President (1998)] [Administration of William J. Clinton] [Online through the Government Printing Office, www.gpo.gov] [DOCID: f:erp_c3._] Economic Report of the President - - - - - - - - - - - - H. Doc. 105-176 [From the online service of the the U.S. Government Printing Office] [wais.access.gpo.gov] CHAPTER 3 -- The Economic Well-Being of Children AFTER DROPPING SHARPLY IN THE 1960s, the official poverty rate among children trended upward from 1969 to 1993, reflecting increases in the share of children in single-parent families and declines in real wages at the bottom of the income distribution. Recently, however, the picture has improved, as the child poverty rate declined by 2.2 percentage points from 1993 to 1996. Other measures of children's material well-being have improved recently as well. Since the early 1990s the share of children in households reporting that they did not have enough food to eat has decreased, the share of households with children living in inadequate housing has fallen, and children's use of basic health services has increased. Continued declines have also been recorded in infant and child mortality rates, and some measures of children's educational achievement have improved. Nevertheless, many children remain economically vulnerable. In 1996 one in five children was officially poor; in 1995 one in nine lived in households paying more than half of their income for housing, and one in seven did not have health insurance; and in 1994 one in 56 lived in households where children experienced hunger due to inadequate resources for food. These factors place children at risk of both current and future hardship, as many of these factors may be critical to children's long-term development. This Administration has adopted a range of strategies to improve the economic and social well-being of children. The first of these is to put in place a system that guarantees that children's basic needs are met. For this reason the Administration has supported major initiatives to expand health insurance, child care, and subsidized housing and has substantially strengthened the child support enforcement system. The second strategy is to provide financial support to needy families in a way that promotes work and personal responsibility. To achieve these goals the Administration proposed and the Congress enacted increases in the minimum wage, an expansion in the earned income tax credit (EITC), and a major restructuring of the welfare system. The third strategy is to invest in programs that enrich children's educational opportunities, by expanding the Head Start program and encouraging higher standards for elementary and secondary schools. This chapter begins by describing recent economic and demographic trends that have influenced the economic well-being of children, and recent programs designed to improve the economic status of families with children. The second section reports on changes in two key components of children's material well-being: food sufficiency and housing conditions. The third section describes recent changes in children's health outcomes, access to health care, and health insurance coverage rates. The concluding section discusses child care and educational programs for children. TRENDS IN THE ECONOMIC WELL-BEING OF CHILDREN THE LINK BETWEEN INCOME AND CHILDREN'S WELL-BEING The adequacy of family income is a critical predictor of both the present and the future well-being of children. Children who grow up in low-income families score lower on standardized academic achievement tests, are less likely to complete high school, complete fewer years of school, and are likely to have lower earnings when they enter the labor market than children who grow up in higher income families. Most studies find that family income is more strongly correlated with children's achievement than parental schooling or family structure. Low family income might affect children's well-being for any of several reasons. One is that low-income parents may not be able to afford to invest as much in the things that improve their children's well-being, such as food, shelter, medical care, and education. A second is that the poor are more likely to live in neighborhoods with a high concentration of poverty and thus may face higher crime rates, poorer quality schools, and more limited connections to mainstream economic activity. A third is that low-income families may experience higher levels of emotional distress, due to the economic pressures of living on a limited income. Finally, income may be highly correlated with children's achievement not because income directly affects child well-being, but because income is associated with some other, hard-to-measure factor that does affect child well-being, such as the value that parents place on education. For example, children from high-income families may perform better on standardized academic tests than children from low-income families not because they have higher incomes but because their parents encouraged them to work harder in school. If this is so, increasing family income will not necessarily increase children's academic achievement unless it also increases parents' commitment to their children's education. A recent study found that more sophisticated techniques that control for potential differences across families produced lower estimated impacts of family income on child well-being than suggested by simple correlations of family income and child well-being. MEASURING TRENDS IN CHILD POVERTY RATES One of the most commonly used measures of the adequacy of family income is the poverty rate. The poverty rate is the percentage of the population who live in families with before-tax cash incomes below a defined level of need, called the poverty line. The official poverty line in use today was devised in the early 1960s and based on the minimum cost of a nutritionally adequate diet. This amount was multiplied by three, because data from the late 1950s suggested that the typical family spent one-third of its income on food. Since then the poverty line has been updated annually for inflation using the consumer price index for all urban consumers. In 1996 the poverty line for a family of two adults and two children was $15,911. Many people have argued that the official poverty measure should be modified to account for family income and family income requirements differently. Box 3-1 describes a recent proposal by the National Research Council to change the way we measure poverty. ------------------------------------------------------------------------------ Box 3-1.--How Does Our Poverty Measure Affect Our Conception of Poverty? A 1995 report by the National Research Council (NRC) recommended a number of changes in the way we measure poverty. Its recommendations include the following: Defining income. On the one hand, the definition of family income should be expanded to include other important sources of purchasing power, such as the EITC, food stamps, and housing subsidies. On the other hand, some necessary expenditures that reduce a family's resources available for basic consumption needs should be subtracted from income, such as taxes, necessary child care and other work-related expenditures, child support payments, and out-of-pocket medical expenditures. Setting a threshold. Poverty thresholds should be adjusted to provide a more accurate measure of family income requirements. First, the consumption bundle used to derive thresholds should be based on food, clothing, and shelter, not food consumption alone. Second, thresholds should reflect regional variations in housing costs. Third, thresholds should be adjusted for family size in a more consistent way than currently. Finally, thresholds should be updated to reflect changes in expenditure patterns over time. Box 3-1.--continued A recent study used key elements of the NRC proposal to estimate alternative poverty rates from 1991 to 1996. These estimates produced increases in poverty from 1991 to 1993 similar to, and decreases in poverty from 1993 to 1996 somewhat larger than, those under the official measure. In addition, under the alternative measure a smaller proportion of the poor are children, black, or members of female-headed households. These changes reflect the fact that the new measure more completely accounts for in-kind transfers, such as food stamps and housing benefits, and for work-related expenditures. As a result, the new measure tends to decrease the relative poverty rate of persons who are more likely to receive in-kind transfers, and to increase the relative poverty rate of employed low-income persons with higher work-related expenses. ------------------------------------------------------------------------------ Chart 3-1 shows that the poverty rate for children declined sharply from 1960 to 1969 and has since trended upward, with peaks in 1983 and 1993. (Throughout this chapter ``children'' refers to persons under age 18 except where stated to the contrary.) Chart 3-1 also illustrates an experimental poverty measure developed by the Bureau of the Census, which includes taxes and in-kind transfers in family income. Including taxes and transfers reduces the estimated poverty rate in each year but does not substantially affect the trend in poverty from 1979 to 1993. Since 1993, poverty rates have declined more rapidly under the experimental than under the official measure. This is largely attributable to recent expansions in the EITC, which affect the experimental but not the official measure of poverty. Estimates of changes in the poverty rate are sensitive to the method used to adjust the poverty line for inflation. Some have argued that the price index used in the official poverty measure has overestimated the actual level of inflation. (See the 1997 Economic Report of the President for a discussion of this issue.) Under some alternative estimates of poverty that incorporate different inflation estimates, poverty trends for children are flat or down over the last several decades. Table 3-1 shows poverty rates for children of different ages, races, family status, and incomes in 1996. In that year 20.5 percent of all children were poor, 9.0 percent were in extreme poverty (defined here ~ TABLE 3.1-C~hildren with Family Incomes Below Different Income Cutoffs, 1996 [Percent of children in each category] ------------------------------------------------------------------------------ Family income Demographic ~ category Less than half Less than Less than twice of poverty line poverty line poverty line ~------------------------------------------------------------------------------ Age of child:�1A1 Under 6 . . . . . . . . . . . 10.5 22.7 46.2 6-17. . . . . . . . . . . . . 7.4 18.3 40 7 Race/ethnicity of child: White . . . . . . . . . . . . 6.6 16.3 38.2 Black . . . . . . . . . . . . 20.6 39.9 68.0 Hispanic. . . . . . . . . . . ~14.7 40.3 72.0 Family status: Female~-headed . . . . . . . . 25.8 49.3 76.3 Married couple. . . . . . . . 2.8 10.1 31.0 All children. . . . . . . . . 9.0 20.5 43.2 ------------------------------------------------------------------------------ 1�1AChildren in families. Note.~--Income is before-tax cash income. Source: Department of Commerce (Bureau of the Census). as family income less than half the poverty line), and 43.2 percent were poor or near poor (defined here as family income less than twice the poverty line). The poverty rate for children under age 6 was 1.25 times that for children aged 6-17, the rate for black and Hispanic children was 2.5 times that for white children, and the rate for children in female-headed households with no husband present was nearly five times that for children in married-couple families. Over one-quarter of children in female-headed families lived in families with incomes below half the poverty line, and over three-quarters of this group lived in families with incomes below twice the poverty line. EXPLAINING RECENT CHANGES IN CHILD POVERTY Changes in family structure, labor market opportunities, and transfers are all likely to have an important influence on child poverty. This section briefly describes each of these three factors. It then estimates the relative influence of these factors on changes in child poverty since 1979. Family Structure Changes in family structure are likely to have had a substantial influence on child poverty rates over the past few decades. The percentage of children living with their mother only has nearly tripled since 1960. This reflects increases both in the percentage of children living with a mother who is divorced or who is married but not currently living with her husband and in the percentage living with a never-married mother (Chart 3-2). Because children in female-headed households have poverty rates nearly five times those of children in married-couple households, an increase in the percentage of children in female-headed households is likely to increase child poverty. Children in female-headed households have a higher risk of living in poverty for two reasons. First, most single mothers do not receive child support and must therefore rely on a single income. Nearly two-thirds of all divorced, separated, or never-married women with children did not receive child support payments in 1991. Second, that single income is likely to be lower in a female-headed than in a male-headed household, because women typically earn less in the labor market than men. In 1996 the median earnings of full-time, year-round workers were about 25 percent lower for women than for men. Macroeconomic and Labor Market Conditions Both macroeconomic and labor market conditions can affect child poverty because they influence the quality and quantity of jobs available to parents. Table 3-2 shows how changes in child poverty rates are related to two key indicators of macroeconomic performance: the unemployment rate and the economic growth rate, as measured by annual growth in real gross domestic product per capita. The marked decline in child poverty from 1959 to 1969 coincided with both high economic growth rates and decreases in unemployment. Some studies have attributed the decline in poverty over this period to these strong macroeconomic conditions. TABLE 3-2.--Changes in Child Poverty Rate and Selected Macroeconomic Indicators ____________________________________________________________________________ Change in official Average annual Change in poverty rate of growth rate in unemployment Period children real GDP rate~ (percentage per capita (percenta~ge points) (percent) points} ------------------------------------------------------------------------------ 1959-69 . . . . . . . . . -13.3 3.0 -2.0 1969-79 . . . . . . . . . 2.4 2.1 2.3 1979-89 . . . . . . . . . 3.2 1.8 -.5 1989-93 . . . . . . . . . 3.1 .3 1.6 1993-96 . . . . . . . . . -2.2 1.8 -1.5 ----------------------------------------------------------------------------- Sources: Department of Commerce (Bureau of the Census and Bureau of Economic Analysis Department of Labor (Bureau of Labor Statistics). Despite continued strong rates of economic growth, the child poverty rate increased during the 1970s and 1980s. This was partly a result of the increase in the unemployment rate from 1969 to 1979, and partly attributable to an increase in wage inequality since the 1970s. (See Chapter 4 of the 1997 Economic Report of the President for a discussion of these trends.) Two recent studies have concluded that the increases in overall poverty in the 1980s were largely attributable to increases in wage inequality and to decreases in the real wages of low-wage workers. During the 1990s changes in child poverty have been more closely aligned with changes in unemployment and economic growth rates. The increase in child poverty from 1989 to 1993 coincided with increases in unemployment and low economic growth rates, whereas the decrease from 1993 to 1996 coincided with lower unemployment and higher rates of economic growth. Changes in Transfer Policy One other factor that has had an important influence on child poverty is changes in the generosity of the tax-and-transfer system. The real value of cash transfers available to low-income families with children has deteriorated significantly since the 1970s. This reflects a decline in the real value of benefits under the Aid to Families with Dependent Children (AFDC) program, the main cash assistance program for low-income families with children until 1996. Although this decline was offset somewhat by expansions in food stamps, combined benefits from both AFDC and food stamps have also decreased (Chart 3-3). Since 1993, expansions of the EITC have increased the transfers available to low-income working families. In addition, the welfare system has been restructured to promote work and family responsibility. These changes are described further below. Assessing Relative Magnitudes--A Decomposition Table 3-3 presents estimates of the impact of changes in family composition, earnings and other before-tax-and-transfer income, and taxes and transfers on the change in child poverty since 1979. It presents two measures of poverty: the official measure, which is based on before-tax cash income, and an alternative measure, which includes both taxes and means-tested food and housing transfers. These estimates may not accurately reflect the full impact of each of these factors on poverty over this period, because they assume that the observed changes in family composition have not been influenced by changes in underlying poverty rates for married couples and single-parent families. These estimates may also be sensitive to the order in which each income source is accounted for in the analysis. The first line of the table shows the impact of changes in family structure on changes in the child poverty rate for various periods since 1979. It shows how changes in the percentage of children living in each of three family types--married couples, female householders with no husband present, and male householders with no wife present--would have affected child poverty in each period, if the poverty rates of each group had not changed. The second line shows the impact of market earnings. It shows the effect of changes in before-tax-and-transfer poverty rates of children in each family category. The third line shows the impact of cash transfers (social insurance and welfare payments). It shows the effect of changes in the percentage of children within each family category whose incomes are brought above the poverty line when cash transfers are included in income. The fifth and sixth lines show similar calculations for means-tested food and housing transfers and for taxes. TABLE 3-3.--Accounting for Changes in Child Poverty~~ [Percentage points] ------------------------------------------------------------------------------ Factor 1979-89 1989-93 1993-96 1979-96 ------------------------------------------------------------------------------ Change in ~official poverty measure attributable to changes in: Family structure . . . . . . . . . 1.2 0.8 0.3 2.3 Earnings and other before-tax-and transfer income . . . . . . . . . 1.1 3.5 -3.0 1.6 Social insurance and welfare payments. . . . . . . . . . . . . 1.0 -1.1 .5 .4 Total change in official poverty measure . . . . . . . . 3.2 3.1 -2.2 4.1 Change in extended poverty measure attributable to changes in: Means-tested food and housing transfers . . . . . . . . . . . . .4 -.3 .0 .1 Taxes~. . . . . . . . . . . . . . . .3 .0 -2.5 -2.2 ~ Total change in extended poverty measure . . . . . . . . . . . . 4.0 2.9 -4.7 2.2 ------------------------------------------------------------------------------ Note.--A positive number indicates an increase, and a negative number a decrease, in the child poverty rate resulting from that factor. Detail may not add to totals because of rounding. Sources: Department of Commerce (Bureau of the Census) and Office of Management and Budget.~ ~~~ These calculations imply that the increase of 3.2 percentage points in the official poverty rate from 1979 to 1989 is attributable to changes in family structure (1.2 percentage points), increases in before-tax-and-transfer poverty (1.1 percentage points), and decreases in social insurance and welfare payments (1.0 percentage point). Changes in food and housing transfer payments further increased the extended poverty measure by 0.7 percentage point. By contrast, the 2.9-percentage-point increase in the extended child poverty measure from 1989 to 1993 is mainly attributable to an increase in before-tax-and-transfer poverty (3.5 percentage points) and to changes in family composition (0.8 percentage point). These factors were offset by transfers, which tended to decrease poverty over this period. Finally, the table suggests that the 4.7-percentage-point decline in the after-tax-and-transfer extended child poverty measure from 1993 to 1996 is attributable to both a 3.0-percentage-point decrease in before-tax-and-transfer poverty and a 2.0-percentage-point increase in the proportion of children moved out of poverty by taxes and transfers (primarily the EITC). Changes in family structure had a small impact on child poverty during this period, increasing the poverty rate by 0.3 percentage point from 1993 to 1996. Overall, this table suggests that changes in family structure have put upward pressure on child poverty rates since 1979. The long-term increase in wage inequality has been reflected in an increase in before-tax-and-transfer poverty since 1979, although there have also been cyclical fluctuations in before-tax-and-transfer poverty in the 1980s and 1990s. Finally, the tax-and-transfer system did more to reduce child poverty in 1996 than in 1979. This largely reflects the recent expansion of the EITC. POLICY INITIATIVES TO SUPPORT FAMILY INCOMES This Administration has established a number of initiatives to expand the resources available to families with children in a way that creates positive incentives for work and personal responsibility. Recent policy initiatives include: A higher minimum wage. The Congress and the Administration increased the minimum wage from $4.25 to $4.75 per hour in October 1996, and to $5.15 per hour in September 1997, bringing the minimum wage to its highest level in real terms since 1984. An expanded EITC. In 1993 the Congress approved the Administration's proposal to expand the EITC. The EITC is a refundable tax credit designed primarily for low-income working families with children. In 1997 the maximum credit for a family with one child was $2,210 (a 54-percent increase since 1993), and that for a family with two or more children was $3,656 (a 140-percent increase since 1993). Welfare reform. The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 dramatically restructured the welfare system to promote work and personal responsibility, and to allow States greater flexibility in designing welfare assistance programs. The law converted the Aid to Families with Dependent Children program into a new block grant, called Temporary Assistance for Needy Families. To promote work and personal responsibility, the law includes new work requirements, a 5-year limit on the length of time that families can receive assistance, and new measures to strengthen child support enforcement. The law also expands funding for child care to make it easier for families to move from welfare to work. Even before the new welfare law was enacted, the Administration had granted waivers to 43 States to allow them to reform their welfare systems to require work, make work pay, and encourage parental responsibility. Increased child support enforcement. Federal legislation in 1993, 1994, and 1996 included measures designed to make the child support enforcement system more effective. Key reforms include streamlined procedures to establish paternity, a new-hire reporting system to track delinquent parents across State lines, uniform interstate child support laws, computerized registries of statewide child support collections, and tough new penalties, such as revocation of the driver's licenses of delinquents. These initiatives have already had noticeable effects. The combination of a higher minimum wage with an expanded EITC guarantees a more adequate income to working families with children. A single mother with two children can now earn enough, if she works full-time, to bring her income with the EITC above the poverty line. In addition, because the EITC benefit increases the financial reward to working, it encourages parents to enter the labor market. Research has shown that expansions in the EITC have been associated with increases in the employment rate of single mothers with children. The changes to the welfare system have already been associated with a large decline in welfare caseloads. The number of children on welfare declined by 23 percent from January 1993 to June 1997. Research suggests that the decline in welfare caseloads from January 1993 to January 1997 was due to a strong economy and to recent State welfare reforms that the Federal waiver process facilitated. Finally, the child support enforcement system has become more effective in guaranteeing that absent parents fulfill their obligations to pay child support. From 1992 to 1996, child support collections increased by 50 percent, to a record $12 billion. In addition, the number of paternities established for children born to unmarried women rose to 1 million in 1996, almost double the number in 1992. These changes are reflected in a substantial increase from 1992 to 1996 in the percentage of never-married mothers who received child support payments. OTHER MEASURES OF CHILDREN'S MATERIAL WELL-BEING Two alternative measures of the adequacy of housing conditions and the sufficiency of household food supply suggest that the material well-being of children has improved since 1989. These trends are consistent with the reduction in child poverty over this period, under the extended measure of child poverty that includes taxes and means-tested food and housing benefits. AVAILABILITY OF FOOD One alternative measure of children's material well-being is the adequacy of household food supply. It has been shown that households that report having an insufficient food supply consume smaller quantities of essential nutrients than other households. Estimates from the Department of Agriculture's Continuing Survey of Food Intakes by Individuals (CSFII) provide information on the percentage of children living in households where there sometimes or often was not enough to eat in the last 3 months. These estimates may overstate the prevalence of hunger among children, if many adults typically go without food before they let their children go hungry. Estimates from the CSFII suggest that the percentage of children living in households without enough to eat has fallen from 4.1 percent in 1989-91 to 3.0 percent in 1994-96 for all children, and from 13.5 percent to 9.4 percent for children in households with incomes below 130 percent of the poverty line (Table 3-4). ~ TABLE 3-4.~--Children in Households Reporting That There Was Sometimes or Often Not Enough to Eat During the Last 3 Months [Percent] ~~--------------------------------------------------------------------------- Category 1989-91 1994-96 --------------------------------------------------------------------------- Children in households at or below 130 percent of poverty line . . . . . . . . . . . . . . . . 13.5 9.4 Children in households above 130 percent of poverty line . . . . . . . . . . . . . ~. . . .8 .6 All children�1A . . . . . . . . . . . . . . . . . . 4.1 3.0 --------------------------------------------------------------------------- Source: Department of Agriculture. The Department of Agriculture's Food and Nutrition Service has recently developed a detailed questionnaire that makes it possible to separately identify whether children or adults in the household experienced hunger. This questionnaire was incorporated into the April 1995 Current Population Survey. The survey found evidence that hunger is more likely among adults than among children: among households with children, 4.3 percent had adult members who had been hungry in the last year due to insufficient resources, but only 1.8 percent had children who had been hungry in the last 12 months because the parents could not afford to buy more food. This survey also suggests that it is unusual for children to go without food for an extended period: 0.9 percent of households with children reported that at least one of their children had skipped a meal in the past year, and 0.2 percent reported that their children had not eaten for a whole day at least once in the past year because of insufficient resources. ADEQUACY OF HOUSING Estimates from the Annual Housing Survey and the American Housing Survey present a mixed picture of the changes in housing conditions of households with children. On the one hand, the percentage of these households that have high rent burdens has increased substantially. The percentage of households with children that spend more than half their income on housing has increased from 6.5 percent in 1978 to 11.5 percent in 1995 (Table 3-5). For renters with very low incomes (defined as households with incomes below 50 percent of the median income for their area), this figure has increased from 31.0 percent to 37.6 percent. TABLE 3-5.--Housing Problems Among Households with Children [Percent of all households with chil~~dren] --------------------------------------------------------------------------- Housing cost or condition 1978 1989 1993 1995 ----------~~~----------------------------------------------------------------- All households with children: Housing costs more than 50 percent of income . . . . . . . . . . . . 6.5 8.7 10.9 11.5 Housing costs 31-50 percent of income . . . . . . . . . . . . 8.3 15.1 15.6 16.7 Severe physical problems with housing . . . . . . . . . . . . . 3.1 3.2 2.0 2.2 Moderate physical problems with housing . . . . . . . . . . . . . 5.6 5.5 5.1 5.0 Crowding (more than 1 person per room) . . . 9.4 7.0 6.3 6.6 Very low income households with children: Housing costs more than 50 percent of income . . . . . . . . . . . . 31.0 36.1 38.2 37.6 Housing costs 31-50 percent of income . . . . . . . . . . . . 28.0 30.6 29.1 30.6 Severe physical problems with housing . . . . . . . . . . . . . 7.5 5.8 3.6 3.4 , Moderate physical problems with housing . . . . . . . . . . . . . 10.5 12.0 10.1 9.5 Crowding (more than 1 person per room) . . 21.9 16.7 14.2 17.0 --------------------------------------------------------------------------- ~ Note.--Income is before-tax cash income. Very low income is defined as family income below half the median for the area. Source: Department of Housing and Urban Development (Office of Policy Development and Research). ~ On the other hand, measures of housing quality for households with children have shown improvement. The percentage of all households with children living in housing with either moderate or severe physical problems fell from 8.7 percent in 1978 to 7.2 percent in 1995, and from 18.0 percent to 12.9 percent for very low income renters. In addition, households with children were less likely to live in crowded living situations in 1995 than in 1978: the percentage living in housing with more than one person per room dropped from 9.4 percent to 6.6 percent for all households and from 21.9 percent to 17.0 percent for very low income renters. These findings are consistent with the results of a recent study which found that many measures of the housing quality of children in households in the bottom income quintile had improved since the early 1970s. These children were more likely to live in modern (that is, post-1940 vintage) housing and more likely to have air conditioning in the 1990s than in the early 1970s. On the other hand, this study also found that children in the bottom household income quintile were more likely to live in neighborhoods where their parents reported that crime was a serious problem in 1991-93 than in 1973-75. NEW HOUSING POLICY INITIATIVES This Administration will continue to make housing quality and affordability high priorities in fiscal 1999. Key initiatives include: Expanding the low-income housing tax credit. The low-income housing tax credit gives States the authority to allocate a fixed pool of tax credits to developers of affordable housing. For fiscal 1999 the President has proposed increasing the total amount of tax credits available to each State from $1.25 to $1.75 per State resident. The HOME Investments Partnerships Program. The Administration has expanded funding for the HOME program by 50 percent since 1993, to $1.5 billion in fiscal 1999. This program offers funding to States, cities, and counties to develop affordable housing options for low-income families. These funds can be used for rehabilitation of existing housing, new housing development, and tenant-based rental subsidies. To date, almost 310,000 families have been awarded assistance through this program. HEALTH STATUS AND HEALTH INSURANCE HEALTH OUTCOMES Over the past 10 years infant and child mortality rates have continued to decline. Other measures, however, such as the incidence of chronic health conditions and of low birthweight, are stable or increasing. One of the most frequently cited measures of children's health status is the infant mortality rate. Infant mortality has continued its decades-long decline in the 1990s, falling from 9.8 deaths per thousand live births in 1989 to 7.2 per thousand in 1996 (Chart 3-4). During the same period the incidence of low birthweight (weight at birth below 2,500 grams, or about 5.5 pounds) has increased slightly, from 7.0 percent of live births in 1989 to 7.3 percent in 1995. This shows that the improvement in infant mortality in the 1990s has been entirely due to factors other than reductions in low-birthweight births. This improvement has been attributed in part to two key interventions. First, new medical treatments have been developed for infant respiratory disorders, an important cause of mortality among preterm infants. Second, there has been a marked decline in sudden infant death syndrome since researchers discovered that these deaths could be prevented by placing babies on their backs to sleep. The slightly increased incidence of low birthweight since the mid-1980s is due in part to an increase in multiple births and births to older women. Failure to prevent low-birthweight births is costly to society because many of these infants require more-expensive medical interventions than do infants born at normal weight. In addition, low-birthweight infants have a much higher risk of infant mortality. In 1995 nearly two-thirds of all infant deaths occurred among the 7.3 percent of all infants born at low birthweight. Mortality rates of older children have also fallen throughout the 1990s, continuing a steady decline since 1960 (Chart 3-5). Much of the decline in the 1960s has been attributed to medical interventions, which have minimized the risk of death from such conditions as congenital anomalies of the heart, infectious diseases, and certain childhood cancers. By contrast, the reduction in mortality during the 1980s has been primarily attributed to a decline in injury-related mortality. The one exception to this pattern is among those aged 15-19, whose mortality rates have increased since 1985. This largely reflects the more than doubling in homicides among adolescents since 1985. The homicide rate for black male adolescents has nearly tripled since 1985, and in 1994 it was nearly seven times higher than the homicide rate for all adolescents. Other indicators of child health have shown less progress in the 1980s and 1990s. The prevalence of asthma among children has increased substantially since the mid-1980s. The prevalence of most other chronic health conditions has fluctuated, without a consistent upward or downward trend. HEALTH INSURANCE One of the critical determinants of children's access to health care is whether or not they have health insurance. Research has shown that children with health insurance coverage are more likely to receive preventive and primary care, and more likely to have a regular relationship with a primary care provider, than uninsured children. Insured children are also more likely to receive treatment for such conditions (when they are present) as injury, asthma, and acute earache. They are less likely to be hospitalized for conditions that appropriate outpatient care could have prevented, and they receive less intensive hospital services when admitted to the hospital. Since 1987 the Congress and the Administration have substantially expanded children's access to health insurance through Medicaid, the primary government program offering health insurance to low-income children. Before 1987 Medicaid was mainly restricted to children in very low income, single-parent families. Since then a series of legislative initiatives have extended Medicaid to much broader groups of children. By 1996 all pregnant women and all children under age 6 who had family incomes below 133 percent of the poverty line, and all children age 13 and younger with family incomes below 100 percent of the poverty line, were eligible for Medicaid. Coverage will continue to be phased in for all children born after September 1983 until poor children of all ages are covered. In addition, many States have expanded children's eligibility for Medicaid beyond these federally required levels. The net result of these expansions is that a much larger share of the child population are now enrolled in Medicaid. Estimates from the Current Population Survey suggest that the proportion of children who are enrolled in Medicaid increased from 16 percent in 1989 to 23 percent in 1995 (Table 3-6). For children under age 6 the increase in Medicaid coverage was even larger, and by 1995, 30 percent of all children under age 6 were covered by Medicaid. ~~ TABLE 3-6.-Children with Health Insurance, by Age of Child and Type of Coverage [Percent] ------------------------------------------------------------------ Age of child Any insurance Medicaid Private insurance 1989 1995 1989 1995 1989 1995 ------------------------------------------------------------------ Current Population Survey: Under 6 . . . . .~ 87.2 86.7 20.3 29.6 70.6 60.4 6-17. . . . . . ~ 86.4 86.0 13.2 19.9 75.3 69.1 All children. . ~ 86.7 86.2 15.7 23.2 73.6 66.1 Health Interview Survey: Under 6 . . . . .~ 84.9 88.4 18.7 28.2 67.9 60.1 6-17~. . . . . . . 85.5 85.7 10.9 16.6 74.0 68.7 All children. . 85.3 86.7 12.8 20.6 71.8 65.7 ----------------------------------------------------------------- Sources: Department of Commerce (Bureau of the Census) and Department of Health and Human Services (National Center for Health Statistics). ~~~~~~~~~~~~ At the same time that Medicaid eligibility has increased, there has been a reduction in children's private insurance coverage. This decline is likely to have been driven in part by a general reduction in employer-provided health insurance, which has affected both low- and high-income families. It may also have resulted in part from newly eligible families dropping their private insurance coverage in order to enroll in Medicaid. Studies have tried to estimate the extent to which the expansion in Medicaid eligibility has crowded out private insurance coverage. These studies have produced a wide range of estimates. One study found that at least one person dropped private insurance coverage for every two persons made eligible for Medicaid. Two other studies estimated that anywhere between 0 and 25 percent of new Medicaid coverage displaced existing private coverage. Table 3-6 presents estimates of the change in insurance coverage for children from 1989 to 1995 from two large household-based surveys, the March Current Population Survey and the Health Interview Survey. Both surveys suggest that the percentage of children under age 18 with health insurance from any source remained constant from 1989 to 1995. These surveys provide a somewhat different picture, however, of the change in total insurance coverage for young children. Whereas the Health Interview Survey finds a moderate increase in insurance coverage for children under age 6, the Current Population Survey finds no increase for this group. This may result from changes in question content in the Health Interview Survey in 1990. It may also reflect differences in the way the two surveys ask about health insurance coverage. Even though there have not been substantial increases in the percentage of children with health insurance, the recent increase in the percentage of children covered by Medicaid may have important implications for the quality of children's medical care. On the one hand, since Medicaid requires no copayments from the insured, covers prescription drugs, and in many States covers services such as dental care, it may promote greater utilization of medical care than private insurance. On the other hand, since provider reimbursement levels tend to be much lower under Medicaid than under other insurance plans, fewer providers may be willing to provide care to children with Medicaid coverage than to children with private insurance. Although research to date on the impact of the recent Medicaid expansion on children's utilization of medical care is inconclusive, Table 3-7 suggests that basic medical care services received by low-income children have increased. The average number of physician visits per year rose by more than 30 percent for poor children in fair or poor health, and by more than 10 percent for poor children in excellent or good health, from 1987-89 to 1993-95. By contrast, during the same period the average annual number of physician visits decreased for children with family incomes above twice the poverty level. ~ TABLE 3-7.--Average Number of Physician Contacts in Last Year for Children Under 15, by Family Income [Number of contacts] ------------------------------------------------------------------------------ Good or excellent health Fair or poor health Family income 1987-89 1993-95 1987-89 1993-95 ------------------------------------------------------------------------------ Below poverty line . . . . . . 3.6 4.0 10.8 14.2 Poverty line to twice poverty line. . . . . . . . . . . . . 3.8 3.9 15.2 16.2 Above twice poverty line . . . 5.0 4.9 22.6 20.7 ------------------------------------------------------------------------------ Note.--Income is before-tax cash income.~ Source: Department of Health and Human Services (National Center for Health Statistics).~ This evidence is also consistent with data pointing to an increase in the share of pregnant women and children receiving at least a minimal level of primary care. The percentage of children under age 5 who did not see a doctor during the previous year fell from 8 percent in 1983 to 5 percent in 1994, while the comparable percentage of children aged 5-17 decreased from 27 percent to 21 percent (Chart 3-6). In addition, the percentage of pregnant women who initiated care in the first trimester of pregnancy increased to a record high of 82 percent in 1996, from 76 percent in 1989. RECENT INITIATIVES TO EXPAND CHILDREN'S ACCESS TO HEALTH INSURANCE Despite recent efforts to expand children's access to health insurance, a large share of children remain uninsured. Estimates from the Current Population Survey suggest that 15 percent of all children did not have health insurance in 1996. Table 3-8 presents estimates from the Current Population Survey on the characteristics of these children. Roughly 21 percent are potentially eligible for Medicaid, because they TABLE 3-8.-Uninsured Children by Family Income, 1996 ~~ ---------------------------------------------------------------------------- Family income and age of child Percent of all ~ uninsured children ----------------------------------------------------------------------------- Below poverty line . . . . . . . . . . . . . . . . . . . . 32.6 Under 6 . . . . . . . . . . . . . . . . . . . . . . . . . 9.8 6-11 . . . . . . . . . . . . . . . . . . . . . . . . . . 10.9 12-17 . . . . . . . . . . . . . . . . . . . . . . . . . 12.0 Between 100 and 150 percent of poverty line . . . . . . . 20.0 Between 150 and 200 percent of poverty line . . . . . . . 16.4 ~ Above twice poverty line . . . . . . . . . . . . . . . . . 30.9 ---------------------------------------------------------------------------- Note.-Income is before-tax cash income. ~ Source: Department of Commerce (Bureau of the Census) ~ are under age 12 and have family incomes below the poverty level. An additional 48 percent either are poor children aged 12 and over or have family incomes between 100 percent and 200 percent of the poverty level. Their family incomes are low enough so that the cost of health insurance may pose a significant barrier, but probably not low enough to guarantee them eligibility for Medicaid. This Administration has developed a two-pronged effort to continue progress in increasing insurance coverage for low-income children. The first component is to extend insurance coverage to more low-income children through the new Children's Health Insurance Program (CHIP). This program offers $24 billion in new Federal funding to States over the next 5 years to expand health insurance programs for uninsured low-income children. States have the option of using the funding to expand children's health insurance coverage through Medicaid, separate State programs, or a combination of the two. In general, States must use their allocation to cover children in families below twice the poverty level, or within 50 percent of the State's current Medicaid income limits for children if the State already covers children at or near twice the poverty level. States must contribute some of their own funds to the program in order to receive these Federal funds, and they must maintain Medicaid eligibility standards at least equal to those in effect in June 1997. In addition, the President will make enrolling eligible uninsured children in both Medicaid and CHIP a priority. In partnership with the States, health care providers, and business and community groups, the Administration will identify and encourage successful outreach campaigns to enroll up to 5 million uninsured children. To facilitate this effort, the proposed budget for fiscal 1999 includes an option for States to determine presumptive eligibility for Medicaid among children at sites such as schools and day care centers. It will also allow States to receive Federal funding for outreach activities at a 90-percent matching rate from a fixed pool of funds. These legislative proposals will be complemented by administrative actions to simplify enrollment of uninsured children. CHILD CARE AND EDUCATION CHILD CARE Over the past two decades the adequacy and affordability of day care have become increasingly salient issues, as a growing proportion of mothers with young children have entered the work force. Whereas 30 percent of married mothers with one or more children under age 6 were working in 1977, by 1997 that figure had risen to 61 percent. This increase in employment of women with young children has translated into a substantial increase in nonparental child care. Between 1977 and 1993 the number of children under 5 in nonparental child care whose mother was working more than doubled. By 1993, 47 percent of young children with employed mothers had their primary day care arrangement in a day care center or a family day care home, and only 5 percent were cared for by a nonrelative in the child's home. The affordability of child care is an especially critical issue for low-income working families. Although Federal programs subsidize day care costs for low-income families, a large share of these families do not have access to subsidized care. Approximately 1 million low-income children under 13 received federally subsidized care in fiscal 1995. This compares with the approximately 10 million children under 13 with employed mothers and family incomes below 200 percent of the poverty level. For families without access to subsidies, the cost of child care can represent a substantial financial burden. In 1993 child care expenditures represented 25 percent of annual income for those families with annual incomes below $14,400 with employed mothers and preschool children in paid child care. Comparable families with annual incomes above $54,000 spent only 6 percent of their income on child care. The quality of child care is also a critical issue. Two recent studies of regulated child care providers offer reason for concern. One study found that 86 percent of child care centers surveyed provided mediocre or poor care when judged from the perspective of child development, and 12 percent were of such poor quality that the children's health and safety needs were only partly met. The second study, of family day care homes, found that 91 percent were of only adequate quality or less. The President's fiscal 1999 budget includes a dramatic increase in Federal investments in child care to increase its affordability and quality. Key initiatives would: Expand child care subsidies. The proposed budget builds on the increases in child care subsidies legislated in 1996, by expanding funding for the Child Care and Development Block Grant Program by $7.5 billion over 5 years. These new funds, combined with funds provided in welfare reform, would allow States to provide child care subsidies to more than 2 million low-income children by 2003--more than double the number of children served in fiscal 1995. Increase tax credits for child care expenses. The proposed budget would increase tax subsidies for working parents who pay for child care expenditures, by expanding the child and dependent care tax credit. The President's proposal would offer more help to 3 million families with annual incomes below $59,000, providing nearly $5 billion in aid over the next 5 years. The President's proposal also includes a new tax credit for private employers that offer child care services for their employees. Expand after-school care for school-age children. The President's proposed budget includes $800 million in new funding over 5 years to dramatically expand the 21st Century Community Learning Program. This program provides funding to school-community partnerships to establish or expand before- and after-school programs for school-age children. The program will serve up to half a million children each year. Improve early learning and child care quality. The President's proposed Early Learning Fund would provide $3 billion over 5 years in challenge grants to communities for programs that improve early learning and the quality and safety of child care for young children. The President's proposed budget also includes funding for scholarships for up to 50,000 child care providers per year, and for improved enforcement of State health and safety standards. EARLY CHILDHOOD EDUCATION Early childhood education programs can play a critical role in preparing children aged 3-5 for entry into school and can have an important effect on children's short-run and long-run development. Research has found that children who participate in early childhood education programs show large short-run gains in IQ, which persist until entry into kindergarten. Research has also linked early childhood education to a number of longer term outcomes, such as grade retention and placement in special education programs. Studies of a smaller number of programs have also shown early childhood education to be associated with increases in high school graduation rates and post-high school monthly earnings, and with a lower probability of teen pregnancy. One of the principal Federal programs supporting early childhood education for disadvantaged children is the Head Start program. The major focus of Head Start is support for enriched preschool programs and development services for children aged 3-5. Federal guidelines require that 90 percent of all children served be from families with incomes below the poverty line. Head Start offers disadvantaged children and their parents a range of services that focus on education, social and emotional development, and health and nutrition. With the establishment of the Early Head Start program in 1994, for which disadvantaged children under age 3 are eligible, the range of Head Start services was extended to younger children as well. A recent nationwide study found that participation in Head Start was associated with increased performance on the Peabody Picture Vocabulary Test and a reduction in grade repetition for white and Hispanic children, although it did not find similar gains for black children. The study also found that both white and black children who participated in Head Start were more likely to be immunized against measles than nonparticipating children from the same family. The President's proposed budget includes $3.8 billion in additional funding over 5 years to help reach the goal of expanding participation in Head Start to 1 million children in 2002, from 714,000 in fiscal 1993. This funding would also allow a doubling of participation in Early Head Start, to 80,000 children by 2002. ELEMENTARY AND SECONDARY EDUCATION One of our society's most important investments in children is elementary and secondary education. Elementary and secondary schools play a critical role in preparing children for college and for entry into the labor market. Research has shown that increases in educational attainment are associated with increases in labor market earnings: each year invested in elementary or secondary education is estimated to increase annual earnings by 5 to 12 percent. Investments in elementary and secondary education can also achieve other important social goals, such as the development of an informed electorate. Measures of Student Performance One of the key tools used in assessing the performance of our elementary and secondary school students is the National Assessment of Educational Progress (NAEP). The NAEP has two parts: the long-term trends assessment, which has repeated the same set of questions since the early 1970s to provide a consistent record of progress over time, and the main assessment, a more recent set of tests designed to reflect current testing methodology and educational content. The main assessment also groups students into three levels of achievement based upon collective judgments about what students should know and be able to do in each subject area. Evidence from the NAEP long-term assessment suggests that achievement in science, mathematics, and reading has improved since the late 1970s. Charts 3-7 and 3-8 show average NAEP long-term assessment scores for 13-year-old white, black, and Hispanic students in these subjects. Chart 3-7 reveals increases in mathematics and science scores since the late 1970s, which have been larger for black and Hispanic than for white children. Improvement in reading scores has been somewhat less dramatic overall (Chart 3-8), but both white and black children have recorded measurable improvement since 1971. Despite these recent gains, significant challenges remain for the Nation's educational system. Evidence from the main NAEP assessments suggests that many students do not achieve basic competency in mathematics, science, or reading. The most recent main assessments found that 38 percent of eighth-grade children performed below the basic level in mathematics, as did 40 percent in science and 30 percent in reading. In addition, U.S. students do not perform well in comparison with students in other countries. According to the Third International Mathematics and Science Study, a study of half a million children in 41 countries, U.S. eighth-graders had average mathematics scores that were below those of 20 other countries. Although U.S. eighth-graders performed better in science, they were still outperformed by students in nine other countries. A second challenge facing the Nation's educational system is the substantial variation in the performance of schools across the country. A recent study of first-grade students found that those attending the top quarter of schools with respect to student performance had average scores in both reading and mathematics nearly 75 points higher than those of students in the bottom quarter. This difference is approximately equal to the average achievement gain of students from the spring of first grade to the spring of second grade. In other words, by the end of second grade the average student's achievement in the bottom-ranked schools will just about equal that of students finishing first grade in the top schools. Differences in student performance are evident across States as well. For example, in 1994 the share of fourth-graders in public schools who scored at or above the basic level in reading ranged across 39 States from a low of 40 percent to a high of 75 percent, and in 1996 the share of eighth-graders in public schools who attained at least the basic level of proficiency in mathematics ranged across 40 States from a low of 36 percent to a high of 77 percent. Impact of School Inputs on School Performance A substantial body of research has investigated the extent to which school quality is related to measurable inputs, such as expenditures per pupil, pupil-teacher ratios, or the level of teacher training. This research has had mixed results. On the one hand, most studies that attempt to relate school resources to students' achievement on standardized tests tend to find only weak evidence that these resources do influence school quality. This may be because most research in this area is based on samples that are not large enough to find a statistically measurable effect. A recent study which combined the results from a large number of other studies found stronger evidence that school expenditures per pupil are positively associated with student achievement. On the other hand, studies that estimate the relationship between school resources and students' earnings later in life tend to find much larger effects. A recent study used data from the 1980 Census to estimate the relationship between the average level of school resources in the State in which workers were born and their subsequent earnings. This study found that workers who had been born (and probably attended school) in States with more abundant school resources earned higher rates of return to each additional year of schooling than other workers. A decrease in the pupil-teacher ratio by five students was associated with an increase in the rate of return to each additional year of school of 0.4 percentage point, and a 10-percent increase in teachers' pay was associated with a 0.1-percentage-point increase. (The average rate of return for all workers in the sample was 5 to 7 percent.) This literature also found that there are important dimensions of school and teacher quality that are unrelated to school expenditure patterns. After controlling for student and parent characteristics likely to affect student performance, and even after controlling for measurable characteristics of schools and classrooms, it is clear that students in particular schools, or enrolled in particular teachers' classes, consistently perform better than average. This suggests that certain aspects of teacher or school quality that are not easily measured, such as the teacher's level of enthusiasm or the school's management style, may be critically related to student performance. Recent Federal Initiatives in Primary and Secondary Education Recognizing that the quality of primary and secondary education can have an important influence on children's later economic opportunities, the Administration has developed and supported a number of initiatives to improve the quality of America's schools. In his 1998 State of the Union address the President proposed two major new initiatives that would increase the financial resources available to public schools: Smaller classes with qualified teachers in grades 1-3. The President is proposing that $12.4 billion be devoted over 7 years ($7.3 billion over 5 years) to reducing class sizes in public schools in grades 1 to 3 from a nationwide average of 22 pupils to an average of 18, and to helping local school districts hire an additional 100,000 well-prepared teachers. The initiative will also provide funds to States and local school districts to test new teachers, develop more rigorous teacher testing and certification requirements, and train teachers in effective reading instruction. This initiative will help ensure that every child receives personal attention, learns to read independently, and gets a solid foundation for further learning. New construction and renovation of school buildings. The President is proposing Federal tax credits to pay interest on nearly $22 billion in bonds to build and renovate public schools. This initiative provides more than double the assistance of the Administration's earlier school construction proposal, which covered half the interest on an estimated $20 billion in bonds. Half of this new bond authority would be allocated to the 100 school districts with the largest number of low-income children, and the other half would be allocated to the States. These proposals build on a number of ongoing Administration efforts to improve the quality of primary and secondary education. To increase the educational opportunities of disadvantaged children, the Administration has expanded the Title I program, which targets resources to children in high-poverty schools. In addition, since it is clear that some important differences in the quality of individual teachers and schools are not directly related to the level of a school's financial resources, the Administration has supported initiatives to change the way schools operate, to better reward performance, and to grant schools more flexibility in meeting measurable performance standards. These are the key emphases of the Goals 2000 and the Charter Schools programs. Finally, the Administration has developed two new initiatives to improve literacy and to increase students' access to the Internet: Title I--Education for the Disadvantaged. Title I provides funds to raise the achievement of disadvantaged children. In 1994 the President proposed, and the Congress adopted, changes to Title I to focus resources on schools with a high percentage of children from poor families, to raise standards of achievement for disadvantaged students, and to give schools greater flexibility in helping students meet these standards. The appropriation for Title I grants to local education agencies was increased by about 20 percent from fiscal 1993 to fiscal 1998. Goals 2000. Enacted in 1994, the Goals 2000 program encourages States to set rigorous academic standards for student performance and to determine whether students are making progress in meeting these goals. It also provides funding to support reform of individual schools and for parental information and resource centers in each State, to help parents become more involved in their children's education. Charter Schools. The Federal Charter Schools program supports the efforts of parents, teachers, and communities to develop innovative public schools that are free from most of the rules and regulations that apply to most public schools and are held accountable for raising student achievement. Since the program's inception in 1995 over 700 charter schools have been established, and Federal funding has increased from $6 million in fiscal 1995 to $80 million in fiscal 1998. The America Reads Challenge. The proposed America Reads program is a multipronged effort to help States and communities ensure that all children are reading well and independently by the end of the third grade. Key initiatives include recruiting and training volunteer reading tutors and helping families help their children build literacy skills. In addition, the Administration has recruited work-study students in 800 universities to help with tutoring initiatives. Technology. The Technology Literacy Challenge Fund and the Technology Innovation Grants program are aimed at meeting four goals: to connect all schools to the Internet, to provide teachers with professional development in the use of technology, to put modern computers in all schools, and to provide challenging software that encourages children to learn more. These initiatives should help prepare our children for the 21st century and keep the Nation competitive in a global economy. CONCLUSION It is clear that children have shared in the benefits of the economic recovery of the past 3 years. The child poverty rate fell from 1993 to 1996, and under an extended poverty measure that includes taxes and means-tested food and housing benefits, the rate was lower in 1996 than in 1989. Other measures of well-being, such as health status, educational achievement, food sufficiency, and housing quality have also shown improvements during the 1990s. Yet many children remain vulnerable, either because they have low family incomes, or because they lack access to health insurance, or because they are not learning basic mathematics, science, and reading skills in school. For this reason the Administration will continue to invest in initiatives to improve the well-being of children. Key initiatives for fiscal 1999 will focus on increasing access to child care and early childhood education, improving the quality of primary and secondary education, and increasing access to affordable housing for families with children.