I f there's one thing most Americans agree on, it is the ideal of giving all children a fair opportunity to succeed in life. Government programs such as Head Start and election-year slogans such as "Leave no child behind" invoke the time-honored metaphor of a contest that every child has a chance to win. The very idea of free public schooling is based on this goal--and to the extent that schools don't succeed in providing equal opportunity, they are vulnerable to criticism from liberals and conservatives alike.
But we're a long way from the "opportunity society" that our Horatio Alger-style political rhetoric extols. What we have is an "inequality society." The average income of the richest 20 percent of American families, for example, amounts to about 12 times the average of those in the bottom 20 percent. Families at the bottom spend more than they bring in, accumulating debt. Families at the top spend a lower percentage of their income, saving money and accumulating wealth. Economists estimate that most families spend about the same percentage of their total expenditures on their children. The end result is that at least five times as much private capital gets invested in kids at the top of the income spectrum as in those at the bottom.
Government is supposed to help level the playing field; as a practical matter, this requires providing more assistance to poor children than to rich ones. But public spending does little to counterbalance inequality. Overall, children in poor families receive only slightly more public support than children in rich ones, with families in the middle receiving significantly less than those at either end. But if one looks just at education, government support tends to benefit the affluent more than the poor. And our social safety nets for children provide more generous benefits to those who lose a parent through death than to those who lose one through desertion. Our system of public support for children is inconsistent, inefficient, and unfair.
Benefits for Rich and Poor
Most European countries provide support for children through a family allowance system in which parents receive a regular check in the mail. The United States tries to accomplish something similar through tax deductions and credits. The complexity of our tax system, however, means that different families get different amounts per child. And few citizens--indeed, few policy makers--really understand who gets what.
Let's take a look at how families with children are treated at different income levels. Those with income below the poverty level don't owe income taxes. Instead, low-income workers can take advantage of the Earned Income Tax Credit (EITC), which gives them a "refund" from the IRS greater than the taxes that were withheld from their pay. This benefit applies to families in which one or both parents are earning income; the levels increase sharply for families with one child, and again (less sharply) for families with two. No additional benefits are provided for families with three or more children. In 1999 a family with two children eligible for the maximum EITC would have received $3,816 (see companion article on page 23).
Families who earn enough income to owe federal taxes can subtract from their taxable income an exemption for every dependent. In 1999 this exemption was set at $2,750. By subtracting that amount, families in the 15 percent tax bracket would save about $412 a year. The same exemption would save families in the 31 percent tax bracket (with adjusted gross income between $104,050 and $158,550 for a married couple filing jointly) more than twice as much--$852. In addition to the exemption, families can deduct a $500 credit per child from their taxes. Thus, the overall benefit for a family in the 31 percent tax bracket with two kids is $2,704. That's less than the maximum EITC. But a high-income family with three kids would save $4,056, a more generous benefit than a family with three kids would receive from the EITC.
Here's the kicker: Tax benefits linked to child rearing follow a U-shaped pattern; they are highest at both ends of income distribution and lowest in the middle. Thus, a family in the 15 percent tax bracket with two children that is not eligible for the EITC enjoys a tax benefit of only $1,824.
Furthermore, relatively few families receive the maximum EITC because it phases out quickly as a family earns more than about $13,000 a year. This phaseout has the same effect as a higher tax rate for working families in the lower-middle part of the income distribution. Such inequity has prompted some policy makers to argue that eligibility for the Earned Income Tax Credit should be expanded in a way that increases public support for child rearing among near-poor and middle-income families. David Ellwood and Jeffrey Liebman of the Kennedy School of Government at Harvard University advocate this approach. Robert Cherry of Brooklyn College at the CUNY Graduate Center and Max Sawicky of the Economic Policy Institute have developed a detailed proposal for a Universal Unified Child Credit: They suggest combining the dependent exemption, child credit, and EITC into a single credit that would initially rise along with earnings, and then phase down to a minimum benefit of $1,270 per child for all families.
Schools, Better and Worse
The most important thing government does for children is fund schools. Families who make use of public elementary and secondary schools receive benefits unavailable to those who choose private schools (except in some states where voucher programs are under way). The average level of government spending per child is currently about $6,000 per year. But levels of spending vary considerably, both across and within states, from more than $8,000 per student to less than $4,000. Because most schools are primarily financed by property taxes, rich communities have rich schools. In comfy suburbs with rising real estate values, even relatively low tax rates can generate high revenues.
Under pressure from grass-roots groups, more than a dozen states have moved toward financial equalization in recent years, redistributing funds from rich districts to poor ones. As well, some federal funds (under the Title I program) are targeted specifically to children from low-income families for tutoring and similar programs to improve educational performance. But these redistributional efforts are modest, and it is unlikely that they fully counteract the inequalities built into the larger school finance system. The extent to which such inequalities affect educational outcomes is hotly debated, but many progressive policy makers argue that "leveling up" should be a high priority.
Early-education programs offer especially important benefits to children from poor families by improving their reading readiness and better preparing them for the first grade. The Rand Corporation estimates that such programs, targeted to low-income states, would more than pay for their costs. Even the conservative economist James Heckman (recent winner of the Nobel Prize) argues that we should spend more in this area. Yet public investments remain low. Al Gore campaigned on a promise of $50 billion over 10 years in matching grants to states, or less than $5 billion a year. George W. Bush mumbled about the possibility of expanding Head Start.
About $4 billion was spent on Head Start in 1998. Despite its political popularity, the program has never enrolled more than about 40 percent of eligible children; and because it is typically organized on a half-day basis, it does not meet the needs of many mothers who work for pay. Furthermore, as families earn more income, they lose their eligibility for Head Start in a phaseout similar to that of the EITC.
Federal block grants have enabled states to increase spending on child care, but slots remain scarce: Less than 15 percent of families eligible for child care assistance actually receive it. Efforts have been targeted to mothers leaving the welfare rolls, so almost-but-not-quite-poor families are least likely to receive assistance with child care costs. Many states have opted for quantity over quality--providing vouchers to large numbers of families but setting voucher levels so low that they cannot pay for high-quality center-based care.
The higher its income, the more likely a family can afford child care with an explicitly educational component. Only about half of four-year-olds from households with incomes of $10,000 or less attend center-based programs. Attendance is even lower among working-poor families who do not qualify for Head Start or state assistance. By contrast, more than 75 percent of four-year-olds from households with incomes of $50,000 or more are learning their ABCs in center-based programs. And affluent parents are more likely to be able to make use of the Child and Dependent Care Credit (which is worth as much as $720 for families with incomes of $75,000 or more with two children). Upscale professional and managerial parents also are well positioned to ask their employers for pretax accounts that allow them to set aside up to $5,000 per year for dependent care; this money is exempt from income and payroll taxes (a subsidy worth about $2,000 a year per child to families in the highest income bracket).
The low level of public support for child care means that in many cities it actually costs more to send a four-year-old to an early-education program than to send an 18-year-old to college. The subsidies built into community college and state university tuition rates, combined with the extensive financial aid available for enrollment at private colleges, cover about 60 percent of total expenditures. Students at public universities enjoy a subsidy of about 80 percent. In contrast, only about 30 percent of the total costs of child care are subsidized by the government.
This huge difference in public support has momentous distributional implications: Children from low-income families, less likely to graduate from high school and go on to college, are less able to take advantage of higher-education subsidies. In 1992 only about 28 percent of high school graduates from the poorest quartile of families enrolled in a four-year college within 20 months of graduation, compared with 66 percent from the richest quartile. Family income had a significant effect even among graduates with identical high school records and test scores.
Through the Safety Nets
Federal policy provides two different safety nets designed to protect children against economic misfortune. One is survivors' insurance, a feature of Social Security that provides benefits to covered families who suffer the death of a wage earner, with virtually no restrictions. The other is Temporary Assistance for Needy Families (TANF), a welfare program that imposes strict eligibility requirements, time limits, and work requirements. However different these programs may appear, both are funded by tax revenues and both represent a form of social insurance.
The amounts of survivor assistance are generous. In 1997 the average annual benefit per child was $6,012. The average annual benefit per widowed father or mother was $6,264. For a widowed father or mother with two children, the total came to $18,288. Compare this number with the maximum annual TANF benefit in the typical state for a family of three at that time: $4,548. That's less than a third as much. The family receiving survivors' insurance does not have to qualify as poor to receive this help. There is no requirement that the surviving spouse find a job.
Economist Gary Becker, who writes a regular column for BusinessWeek, denounces the effect of welfare as follows: "It's bad for children to grow up in a family where all they know is that a check comes from the government every month. That's destructive of the child's self-respect and self-esteem." But is it bad for children or their families to receive survivors' insurance or life insurance money? Or, for that matter, a trust fund check? Do popular magazines or TV shows ever interview young middle-class widows or rich coupon clippers who are ashamed of receiving a monthly check? The amount of federal money docketed for TANF in 1997 was $16.5 billion; state expenditures were about the same, for a total of $33 billion. The amount spent on survivors' insurance that year was about $55 billion.
The bottom line is that we provide more adequate--and more expensive--insurance against the death of a parent than against the desertion or economic inadequacy of a parent, which are the primary reasons single mothers and their children require public assistance. Child support enforcement efforts have improved but vary enormously from state to state. Private detectives are willing to track down a delinquent parent in return for a percentage of the child support owed--if it's a large amount. Such bounty hunters seldom see low-income parents as attractive clients.
Consider, as well, the way housing policies help different families. A variety of government programs subsidize housing for poor families, and the children in these families benefit. At the same time, families who own their homes can deduct the interest they pay from their taxable income, a benefit that costs the federal government more than twice as much as is spent on low-income housing assistance and low-rent public housing.
Similarly, employer-provided health insurance is tax-exempt--an important benefit for the families of those workers who enjoy it, few of whom are at the bottom of the income distribution. Medicaid has traditionally financed care for poor children. Since 1996 it has been disconnected from welfare, and its coverage has been reduced. An Urban Institute study shows that many eligible children are not participating, perhaps because their families are unaware of their eligibility. Similarly, the Children's Health Insurance Program (CHIP), enacted in 1997 to help kids whose parents did not qualify for Medicaid and were too poor to afford commercial plans, failed to deliver the goods. A spate of publicity earlier this fall revealed that many states had not fully used the federal funds they were offered. Need was not lacking. Political commitment was.
Poverty and Progressivity
Poverty rates in the United States are higher than in any other industrialized country. The reason is simple: We don't provide the level of public assistance that countries such as France and Sweden do. Child poverty rates vary widely within the United States. In the 1990s, they ranged from a low of 6.9 percent in Maryland to a high of 45 percent in Washington, D.C. States with high levels of income inequality among children also have high levels of poverty among children, according to the Urban Institute. Much of this variation is attributable to differences in state "policy effort." In other words, those states that have made concerted efforts to reduce poverty and inequality for kids have succeeded.
But the complexity of our public policies, which include tax benefits and public assistance as well as programs that are actually labeled "insurance," makes it difficult for ordinary citizens to understand who is getting what. Furthermore, the U-shaped pattern of the most visible tax benefits--higher at both ends than in the middle--may undermine support for more generous profamily policies. One way to address this political problem is to fill in the middle--for example, by supporting the proposal for a Universal Unified Child Credit. But it is important to remember that this proposal will do little to directly help children in poor families.
The only true way to build an "opportunity society" is to thoroughly reform our current social-insurance and education policies. If we want to link the profamily agenda to the ideal of equal opportunity for kids, we need to explain why some Horatio Algers won't have a chance without some early help from Robin Hood. ¤
