SECTION: CAPITOL HILL HEARING TESTIMONY

LENGTH: 4180 words

COMMITTEE: HOUSE BUDGET

HEADLINE: CHALLENGES FACING WORKING FAMILIES

TESTIMONY-BY: RON HASKINS, SENIOR FELLOW,

AFFILIATION: BROOKINGS INSTITUTION, WASHINGTON, DC

BODY:
August 1, 2001

Testimony of

Ron Haskins Senior Fellow, Brookings Institution, Washington, DC Senior Consultant, Annie E. Casey Foundation, Baltimore

Before the Committee on the Budget U.S. House of Representatives

Chairman Nussle, Ranking Member Spratt, and Members of the Committee on the Budget:

My name is Ron Haskins. I am a Senior Fellow at the Brookings Institution in Washington, DC and a Senior Consultant at the Annie E. Casey Foundation in Baltimore. Until January of this year, I was a staffer with the Committee on Ways and Means where I had the great privilege of working on the seminal welfare reform law of 1996. Thus, it is a great privilege to appear before you today to discuss the effects of this mighty piece of legislation on working families.

Figure 1 provides extensive information about the effects of the 1996 reforms. The graphs in Figure 1 show, respectively, that the rolls of the Temporary Assistance for Needy Families (TANF) block grant and its predecessor program (Aid to Families with Dependent Children; AFDC) have declined by nearly 60 percent and that there has been a huge increase in the number of single mothers who work. The increase in working mothers has been especially impressive among never-married mothers; these mothers exhibited a 40 percent jump in employment in just the three years leading up to 1999. This spectacular rise in employment is especially important because in the past never-married mothers were the least likely to complete high school or to have job experience and the most likely both to go on welfare and to stay on welfare for long spells. The figure also shows that child poverty has declined greatly both as measured by the official Census Bureau measure of poverty and by a broader poverty measure that takes into account more public benefits including the Earned Income Tax Credit and food stamps. Finally, the figure shows that by 1999 poverty among all female- headed families and among black female-headed families was the lowest it has ever been.

Figure 2 provides additional information on poverty. The first panel shows that as the welfare rolls declined so dramatically every year between 1995 and 1999, both overall child poverty and poverty among black children also declined every year. In fact, the declines in black child poverty in both 1997 and 1999 are the biggest single year declines in history and by the end of 1999 black child poverty was the lowest ever. It is useful to reflect on the patterns depicted in this panel. At the same time that the nation is experiencing the greatest declines in the welfare rolls in history, child poverty is also declining more than at any time since the 1960s. In fact, as the second panel in Figure 2 shows, under a broad Census Bureau measure, child poverty declined more than twice as much during the economic expansion of the 1990s than it declined during the expansion of the 1980s. Both the 1980s and the 1990s expansions generated a net increase of about 19 million jobs, but the expansion of the 1990s was accompanied by a much greater decline in child poverty. The reason is simple. In the 1980s, single mothers stayed on welfare and did not take advantage of the growing economy. In fact, during the expansion of the 1980s the welfare rolls actually grew by over 10 percent. By contrast, as we have seen, in the 1990s about two million mothers left welfare or avoided welfare altogether and most of them found jobs. Obviously, only when mothers join the economy can they take advantage of economic opportunity. The third panel in Figure 2 shows that even deep poverty (poverty at half the official poverty level or about $7,300 for a family of three in 2001) has fallen dramatically.

Changes in welfare receipt, work, income, and poverty are not the only possible effects of the 1996 reforms. Unlike previous welfare reform legislation, the 1996 reforms aimed specifically at decreasing the number of children born outside marriage and increasing the formation of two-parent families. In creating the TANF block grant, Congress formulated four goals that states should use the block grant funds to achieve. One of these, of course, was to provide cash assistance to destitute families. But an explicit goal of the legislation was to reduce welfare dependency both by increasing work and by increasing marriage. In fact, three of the four TANF goals addressed issues of family formation.

The data shown in Figure 3 provide some evidence that an exceptionally important shift in nonmarital births and family living arrangements is now beginning to occur. The top panel shows that after more than a half century of increases, the number of births outside marriage, the nonmarital birth rate per 1,000 unmarried women, and the percentage of all children born outside marriage leveled off in about 1995. Part of the explanation for this felicitous result is found in the second panel of Figure 3. Due in part to widespread recognition that teen pregnancy is harmful to both the adolescent mother and her child, as well as to federal, state, and local policies aimed at reversing the tide of teen births, the nation is now in the midst of a decade of progress in reducing teen births. Equally impressive, the third panel of Figure 3 shows the most recent and perhaps the most hopeful of the increasingly positive trends in nonmarital births and family formation. This trend, based on national data from the Census Bureau, shows that the percentage of children under age 6 living with their married mother halted its 30-year decline in 1995 and has increased every year but one since then (Bavier, 2001). Similar results have been reported recently from another national survey being conducted by the Urban Institute in Washington, DC (Acs & Nelson, 2001).

It is necessary to avoid overclaiming for the effect of the 1996 reforms on these very positive developments in family formation. However, there are several reasons to believe that welfare reform may be playing at least some role in these historic shifts. First, even though many of these developments in family formation began before the federal reforms were enacted in 1996, by that time a welfare reform movement in the states was well underway. By 1994, half the states had mounted welfare reform demonstration programs under waivers from federal law, an approach to reform that was initiated by the Reagan Administration and had been picking up speed during the Bush and Clinton Administrations. Many of the state demonstration programs involved mandatory work reforms that were similar to those later required by the 1996 federal reforms.

The mandatory work requirements of state and federal legislation are designed in part to discourage nonmarital births and to encourage marriage. By requiring most mothers to work shortly after joining welfare, and by limiting cash benefits for most recipients to five years, mothers are forced to realize that they cannot depend on permanent welfare benefits like they could under the old AFDC program when millions of mothers stayed on welfare for more than a decade. Once mothers understand that they cannot permanently depend on welfare, they begin to realize that they must have other sources of income. The major means of achieving income for most of these mothers is work. However, marriage can also increase the mother's income if she marries a man who is employed. The data in the last panel of Figure 3 suggest that a growing number of poor and low-income mothers are taking the course of marriage.

Again, these trends must be watched closely and Congress would be well advised to carefully review these and other data on family composition as part of the welfare reform reauthorization debate that will take place next year. Given the encouraging data on family formation, it seems wise for Congress to consider additional actions that might be taken to provide even more movement away from births outside marriage and toward marriage and parenting by both biological mothers and fathers.

Regardless of the interpretation one gives to the data on nonmarital births and family formation, virtually no one doubts that welfare reform is playing a major role in the substantial increases in employment and declines in child poverty shown in Figures 1 and 2. Why have the welfare rolls and poverty declined while work increased so much in recent years? Nearly every analyst agrees that the answer involves three factors that have combined to produce these exceptionally favorable changes. The three factors are the welfare reform movement that swept the states and the federal government in the early- and mid-1990s, the growth of a federal system of benefits for working families, and a strong economy. Scholars do not agree about the relative importance of these factors, nor will they ever (Blank and Schmidt, 2001). But no one disputes the claim that two of the factors were produced by deliberate decisions of policymakers at the federal and state level. I hardly need to tell members of this Committee about the historic federal welfare reform legislation of 1996. After a partisan and even somewhat rancorous debate, Congress approved the 1996 welfare reform legislation with a bigger bipartisan majority than the bipartisan majority that enacted Medicaid in 1965. Let's be clear, the 1996 reforms were highly bipartisan and were signed by a Democratic President who had campaigned on the promise to "end welfare as we know it."

But less well known than the famous 1996 reform legislation is the work support system that Congress has created with dozens of legislative initiatives spanning more than a quarter century. The work support system consists of a variety of programs that provide benefits to poor and low-income working families. By definition, any program that provides benefits to working families is part of the work support system, but I would emphasize seven programs (see Figure 4) that now provide extensive help to working families. These include child care, Medicaid and the relatively new State Child Health Insurance Program (SCHIP), food stamps and child nutrition, the earned income tax credit (EITC), the child tax credit, and housing programs. As a recent study from the Congressional Budget Office (CBO, 1998) shows, owing mainly to legislative expansions of these programs since roughly the mid-1980s federal spending on working poor and low-income families has increased greatly in recent years. The CBO analysis reveals that if Congress had not made statutory changes to expand the work support programs since 1985, in 1999 the federal government would have spent a mere $6 billion to support poor and low-income working families. However, taking into account all the expansions (plus the child tax credit and the SCHIP program which are new), CBO estimated that the federal government would spend nearly $52 billion on working families in 1999. This projection of $52 billion was almost surely an underestimate of the amount actually spent to help working families in 1999 because so many more mothers left welfare than was predicted, thereby greatly increasing EITC spending. In addition, the CBO study examined only entitlement spending which does not include some child care programs, housing, and several other benefits.

To gain a more complete understanding of the work support system and how it has grown, consider child care, Medicaid, and the EITC, three of its major components. In 1993, across all child care programs including Head Start, the federal government spent about $9.5 billion. By 2000, the total federal dollars available for child care had exploded to around $18 billion, nearly twice as much as just seven years earlier. The two major reasons for this increase were that in the 1996 welfare reform law, Congress included an additional $4.5 billion over six years in money for the child care block grant. Further, Congress allowed states to spent TANF dollars on child care either directly out of the TANF block grant or by transferring up to 30 percent of their TANF funds into the child care block grant. Thus, Congress gave states both increased funding of programs designed specifically for child care as well as the flexibility to use TANF funds for child care. As a result, there is general agreement that states have been able to meet all the demand for child care of families still on TANF as well as provide TANF-supported child care for a period of time, usually at least one year, after families leave TANF (Besharov & Samari, 2001).

Changes in Medicaid are equally impressive. It might be argued that one of the original sins of American health care policy for the poor was that the only way to gain entrance to Medicaid was through either the AFDC or the Supplemental Security Income (SSI) programs. In other words, to obtain federal support for health care a family had to go on welfare. This approach to rationing federal support for health care was never a good idea. But as Congress and the American people became more and more concerned about welfare dependency, maintaining a Medicaid system that literally required a family to go on welfare before its members could receive help became unsustainable. Beginning in the mid- 1980s, Congress enacted a series of reforms, designed primarily by Henry Waxman, that expanded eligibility for Medicaid to people, especially children, outside the AFDC and SSI programs. These coverages were gradually expanded until the 1996 welfare reform law completely separated eligibility for the TANF program from Medicaid eligibility. Now nearly all children who live in families with incomes under the federal poverty level ($14,630 for a family of three in 2001), and many children who live in families with much greater income, are eligible for Medicaid health insurance. Moreover, in 1997 Congress and President Clinton created the SCHIP program that offers coverage to children above the Medicaid eligibility level. In fact, because of the SCHIP rules for handling work disregards, some states are providing coverage to children in families with incomes in excess of $35,000.

The EITC shows the same pattern as child care and Medicaid; namely, large increases in both total spending and in the number of families receiving benefits. In 1984, about 7 million families received an average credit of about $260; total federal spending was $1.6 billion. By 1999, after Congress expanded the credit in 1986, 1990, and 1993, usually on a bipartisan basis, about 18 million families received an average credit of $1,700 at a total federal cost of $31 billion.

It would be difficult to exaggerate the importance of the combined effect of the 1996 welfare reform legislation and the expansion of the work support system by Congress and several presidents. Through repeated legislative action involving a broad array of programs and though its approval of substantial increases in spending, Congress has shown a deep commitment to changing the rules of the game for low-income families. In the old days, welfare often provided a better deal than low-wage work. Now Congress has substantially restricted welfare use and greatly increased the returns to low-wage work. As a recent study by the Urban Institute in Washington, DC shows, in most states mothers who take a minimum-wage job and work half-time are better off financially than they would be on welfare (Coe and others, 2001).

This system of limited welfare and expansive public support for low-wage work is the best approach the federal government has yet found for both reducing welfare dependency and fighting poverty. Taken together, these two sets of changes constitute a revolution in federal policy toward poor and low-income families.

There is, of course, no public policy that works flawlessly. Thus, it is not surprising that this new approach to fighting dependency and poverty has been shown to have some problems (Haskins and Blank, 2001). Four of the problems are especially notable and should be a major focus of discussion during the welfare reform reauthorization debate in Congress next year. All four could raise issues for the Budget Committee and for next year's budget resolution.

The first problem is one that seems solvable. Because Congress has so carefully constructed the work support system described above, it is disappointing to find that two vital parts of the system are not working well. Families leaving welfare or low- income families that avoid welfare are not receiving the Medicaid and food stamp benefits for which they qualify. After several years of increases, administrative data for both programs show declines in enrollment of either families with children or children. In Medicaid, for example, national data show that the number of children enrolled declined between 1995 and 1998 (Herz & Baumrucker, 2001). A recent national study of families leaving welfare found that only about 40 percent of families eligible for food stamps actually received the benefit (Zedlewski, 2001). It hardly seems necessary to emphasize how important these two benefits are to poor and low-income families and how essential they are to creating an effective national work support system. Thus, Congress and the states should carefully review the caseload declines and determine whether changes are called for in statutes, regulations, or administrative procedures. It would appear that part of the food stamp problem could be solved by changes in the federal quality control system which provides an incentive for states to be less than aggressive in ensuring that working families receive food stamps (Haskins, 2001). With regard to Medicaid, as testimony before the Committee on Ways and Means last year demonstrated, states such as Florida, Oklahoma, and Ohio that took aggressive administrative action had substantial increases in the number of children and families receiving coverage (Committee, 2000). The major thrust of administrative action in these states seems to have been measures designed to ensure that families knew they were eligible and that made it as easy as possible for working families to apply for and maintain their eligibility. Florida, for example, changed its procedures so that families could apply for and maintain their eligibility entirely over the phone.

A second important problem is that some families are worse off as a result of welfare reform. Although the children's poverty rate has declined substantially, some families appear nonetheless to be falling through the cracks. Wendell Primus (2001) of the Center on Budget and Policy Priorities in Washington, DC, based on national data from the Current Population Survey, estimates that as many as 700,000 families with children have less income in 1999 than in 1995. Not all of this increase should be attributed to welfare reform. Even so, it seems clear that the 1996 reforms have led to declines in income for some families. Under the AFDC program, mothers could and often did stay on welfare for long spells. Indeed, Bane and Ellwood (1983) showed that up to 65 percent of the families on the rolls at a given time would eventually be on the rolls for 8 years or more. But the TANF program that replaced AFDC has serious restrictions on length of spells. The federal 5-year time limit provides incentive for families to get off welfare as soon as possible. Many states have adopted time limits that are even shorter than the federal 5-year limit; some families in these states have already hit the state time limit and have left welfare. State policy on sanctions is probably even more important that time limits in causing families to leave welfare (Pavetti and Bloom, 2001). More than 35 states have sanction policies that result in complete termination of cash welfare benefits. The point is that whereas families could stay indefinitely on AFDC, they must meet TANF requirements or have their benefits reduced or terminated. Congress may want to use the reauthorization debate to review what actions states are taking to help these families and to determine whether they can return to welfare once they fulfill whatever requirement resulted in their sanction. Similarly, Congress may want to determine whether states are prevented from helping families that reach the 5- year time limit because federal policy allows a maximum of 20 percent of the state caseload to be exempt from the time limit.

The next two issues are not so much problems as they are opportunities that need to be further exploited. The first of these is the family formation goal of the 1996 reforms. Although the number and percentage of children born outside wedlock has leveled off and the number of children living with two biological parents has increased slightly, research shows that states have not been very aggressive in designing and mounting programs to reduce nonmarital births or increase marriage (Gais and others, 2001). There is every reason to believe that state and local programs designed to increase marriage and reduce nomarital births would produce even greater success. Especially important would be programs that offer services to young couples at the time of a nonmarital birth. Recent research shows that about half of these couples live together and a total of more than 80 percent describe themselves as being in a mutually monogamous relationship (McLanahan, 1999). However, research also shows that the majority of these relationships fall apart within a year or two (Rangarajan and Gleason, 1998). Thus, job training and employment assistance, counseling, and other services may prove beneficial at the time of the birth when the parents are committed to each other and their baby. A major goal of these programs should be helping fathers find jobs or acquire skills to qualify for better jobs.

The second issue is that many, perhaps most, of the mothers leaving welfare take low-wage jobs. Both experience and research show that mothers now working in low-wage jobs can, with proper assistance, move up the job ladder. However, as Gary Burtless (1995) of the Brookings Institution has shown, most mothers will not improve their income without external assistance. If Congress wants to help these mothers improve their income, there is probably no choice except to find ways to help them acquire more training and education. The Congressional debate on this issue may be somewhat misleading because members may simply assert that if education and training are made available, all will be well. But this claim is false.

Two education and training strategies make sense at this point. First, states and localities should mount programs in which they cooperate with low-wage workers and educational institutions to arrange courses that lead to specialized short-term training or even two-year degrees. It is essential that the training be relevant to employment available in the local economy. Educational institutions, especially junior and community colleges, must be flexible in the length of courses and the times at which they are offered in order to meet the needs of single working parents.

The second strategy is to work with employers to identify the types of training that are required for low-wage workers to advance in their organization. Some programs of this type already exist, but many more are needed. Employers in need of skilled workers are often agreeable to arrangements of this sort and may even pay at least part of the training costs and grant their workers released time for training. In some cases, it may prove possible to work with both employers, who can identify the types of training and the skills required for advancement, and with local post-secondary institutions that can provide the training.

The 1996 welfare reforms, in combination with expansions of the work support system over the past quarter century, are now providing a solid basis for using public policy to simultaneously minimize welfare dependency, increase employment, and reduce poverty. Although it is too early to make any definitive claims, it appears that these policies are also associated with stable or slightly declining rates of nonmarital births and are having a positive impact on family formation. Given these acheivements, it would be a serious mistake for Congress to make big changes in either welfare reform or the work support system at this point. Especially if Congress wants states to tackle the remaining problems and thereby produce even more progress against welfare dependency, poverty, and family dissolution, it would be a grave mistake to reduce funding for the TANF block grant. Federal welfare and work support policies are producing great success. There is every reason to believe that if allowed to continue with only minor adjustments and modest additional investments, they will produce even more growth in employment and income and greater declines in welfare dependency and poverty in the years ahead.



LOAD-DATE: August 3, 2001