Reinventing Welfare… Again: The latest version of reform needs a tune-up

Gary Burtless and R. Kent Weaver

Welfare reform was the focus of fierce partisan debate in the last Congress. President Clinton promised to “end welfare as we know it” in his 1992 campaign, but failed to submit reform legislation in time for congressional Democrats to act on it before they lost their majority in the 1994 election. Last winter, Clinton vetoed the Republican Congress’s budget reconciliation bill and a stand-alone welfare reform bill. Both would have fundamentally restructured the safety net on GOP terms. Last August Congress passed a modified version of the Republican plan—the Personal Responsibility and Work Opportunity Act—and the president signed it over the objections of many Democrats in Congress and senior officials in his own administration.

While the new law may appear to settle the issue of welfare reform, at least temporarily, political pressures and implementation problems could soon put it back on the congressional table. Even as he signed the welfare bill Clinton promised to try to soften some of its harshest provisions. Social conservatives, disappointed that most of their preferred remedies for illegitimacy were left out of the new law, may also press for change.

Welfare has been on the nation’s agenda for more than two decades. It is deeply unpopular. Most voters believe that the old Aid to Families with Dependent Children program discouraged work and encouraged illegitimacy and family breakup. It provided too little help to keep families from falling into poverty, but too few incentives to push able-bodied adults into self-sufficiency. The 1996 reform addresses some of AFDC’s worst problems, but it creates some big new problems for state and federal policymakers. And it imposes serious and unnecessary risks on the nation’s poorest children. To improve their prospects, welfare should be fixed—again.

The New Welfare Law

The new welfare law changes the nature, organization, and financing of a vital part of the U.S. safety net. Under AFDC, Washington offered states open-ended grants for cash welfare benefits for needy children and their adult caretakers. States had to match the federal dollars to get the grants, but federal spending had no fixed limit. States were free to define need, establish benefits, and determine eligibility, leading to a great deal of interstate variation.

The new law replaces AFDC with a federal block grant called Temporary Assistance for Needy Families. Though small exceptions will be made for low-income states with fast-growing populations and states in recession, most states’ TANF grants will be determined by their federal AFDC grants during the past few years. The new law ends the individual entitlement to benefits. Under new state programs, poor children may no longer be automatically entitled to cash benefits. The new law gives states more program flexibility in many areas, but it also imposes new federal requirements. For example, each state must ensure that a rising percentage of its adult aid recipients engages in approved work. The head of each family on welfare will now be required to work within two years after cash payments begin. Work hours requirements are stringent, and states will face increasingly harsh penalties for failing to meet them. The Congressional Budget Office estimates that states would have to invest about $14 billion of their own money over 1997-2002 to meet the new work requirements.

States cannot use the TANF grant to pay benefits to a family for more than 60 months; they can limit benefits to fewer than 60 months if they choose. They can use existing federal social service funds to provide noncash benefits to families after 60 months. Up to 20 percent of a state’s caseload can be exempted from the 60-month limit for hardship reasons. Unlike earlier versions of the GOP welfare bill, the new law does not require (but does permit) states to deny benefits to unmarried teen mothers and to impose family caps. (A cap denies higher monthly payments to families into which children are conceived or born while the mother is receiving welfare.)

Predicting the impact of the new law is not easy. It is not clear, for example, how quickly or well states will implement the new work programs. No one knows how states or aid recipients will react to incentives in the new law. How many recipients, facing the five-year time limit, will find work in the private sector? How many states will use their own resources to pay for aid to recipients who exhaust their eligibility for federal benefits? Will competition among states lead to lower monthly payments or tighter time limits? How many states will impose a family cap or deny benefits to unwed teen mothers? Will such measures reduce illegitimacy and caseloads? How will recessions affect states’ willingness to pay for family assistance?

Many specialists believe the new law will, in time, increase poverty and deepen distress among people who are already poor. Some of the hardship will result from cuts in food stamps, Medicaid, and Supplemental Security Income. The new law abolishes or sharply curtails most forms of public aid to noncitizen immigrants, trims food stamps, and restricts SSI and food stamp eligibility for some classes of recipients. To reduce the risks the new law poses to poor children, several features should be revised.

Revisiting the Issues: Entitlement

Converting AFDC into a block grant probably settled the entitlement issue with respect to family support payments for the foreseeable future. But ending the entitlement will cause problems. State disparities in eligibility for cash benefits will almost surely increase. Some states may even decide to vary eligibility within their borders, as they do in state-financed general assistance programs. In states that do not make aid a legal entitlement, low-income children may have no assurance of receiving benefits. Even in states that do, monthly benefits may fluctuate if the state decides to operate within a fixed welfare budget. The distribution of cash assistance within and across states could become quite capricious.

The first GOP reform package allowed states to convert federal food stamp funds into a federal nutrition block grantþeffectively ending the federal entitlement to food stamp benefits in states taking up this option. The new law dropped this provision. Future efforts to wring savings out of entitlement programs may revive the effort to allow states to turn food stamp funds into a block grant. That would be a serious mistake.

Food stamps are the only program guaranteeing a uniform level of aid throughout the nation to indigent families in all kinds of circumstances. Food stamp spending is also highly responsive to unpredictable needs arising from recessions, natural disasters, and demographic change.

Time Limits

The new law contains two kinds of time limits. States must require adults to work not later than 24 months after cash benefits begin. Federal funds cannot be used for cash aid to families whose adult head has received aid for more than 60 months. The limits send a clear signal to beneficiaries that they must become self-sufficient in the labor market and set an unambiguous timetable for achieving that goal. But the harsh reality is that many adult recipients will not become self-sufficient in the foreseeable future. Some lack skills, others have no work experience, still others face family circumstances or health problems that make steady work impossible. A number live in areas—rural counties and Indian reservations, for example—where jobs are few and far between. Given the inherent instability of many jobs in the low-wage labor market, many current and future aid recipients will suffer repeated spells of unemployment, some of them long. If no government-financed cash assistance is available, strict time limits could cause highly visible hardship and even homelessness among unskilled parents and their children.

To mitigate hardship under time limits, the law should be changed in several ways. States could provide cash assistance after five years to those willing to engage in workfare or subsidized jobs—a move consistent with the Clinton administration’s original welfare proposal. If Congress is unwilling to offer this flexibility, it could authorize federal waivers permitting states to extend the five-year time limit under certain conditions.

Allowing states to set time limits shorter than five years, as the new law permits, is particularly risky. Some states may be tempted to export their welfare problem by adopting tighter time limits than their neighbors. To avoid a “race to the bottom” in time limits, states should be required to get federal approval to impose time limits of less than five years.

Work Requirements

Political competition between President Clinton and congressional Republicans ensured that the new welfare law would feature stiff work requirements. Work requirements enjoy overwhelming support among the public and surprisingly strong support among assistance recipients themselves. They are essential to maintaining public support for cash and nutrition assistance programs over the long run. But we must be realistic about what such requirements can accomplish and about the capacity of the low-skill labor market to absorb new workers. Research has shown that welfare-to-work programs can increase both earnings and work among welfare recipients, but not enough to allow most to become self-sufficient. Further, many recipients will endure long spells of joblessness after completing the programs. Without access to publicly provided jobs and supplemental aid, such as child care, some single mothers will find it impossible to support their children.

On the details of work requirementsþ how many hours of work, how much of the caseload to cover—it is tempting for federal policymakers to be overly prescriptive. They can earn easy political points that way, but they may also strip states of the flexibility needed to implement effective programs. Welfare-to-work experiments find that pushing recipients into jobs instead of training achieves the highest short-term earnings gains. But these findings may not be valid for mothers with very young children, who will increasingly become subject to stiff work requirements. States need greater flexibility than the new law provides to set the number of hours that welfare recipients must work and the mix of work and training that they will be offered.

Reducing Illegitimacy

The new law requires that teenage mothers live with a parent or responsible adult to receive cash benefits. As long as exceptions can be made in case of parental abuse or other hardship, the requirement is appropriate. The law also offers small incentives for states to reduce out-of-wedlock birth rates and, as noted, allows states to impose family caps. Flexibility on family caps is sensible because evidence on their impact, drawn mainly from recent experience in New Jersey, is slim. Reductions in births to welfare recipients appear modest, but the evidence is incomplete.

Many conservatives argue that states should deny cash assistance to teenage mothers who bear children outside of marriage. The new law does not require states to take this step, but it allows them to do so. It would be more prudent to permit strictly limited and carefully evaluated trials of the policy. While there is little doubt that out-of-wedlock teen pregnancies are costly, not only for society but for the children they produce, we have no evidence that denying benefits to unwed teen mothers will cause illegitimate births to fall. Such a ban has never been tested, and the potential harm to affected children could be large.

State Fiscal Effort

Under the new block grant formula, states can reduce their welfare spending significantly, thereby risking a race to the bottom in eligibility and benefits. To qualify for full federal block grant funds under the new law, states that meet the work participation targets need spend only 75 percent of what they spent in the past. The “maintenance-of-effort” requirement is even less stringent than it appears because states can spend part of the TANF block grants on services that may not go to low-income dependent children or their parents. In addition, inflation will erode the value of previous state efforts, making the requirement less onerous over time.

The shift to block grant funding for family assistance, combined with continued availability of federal food stamps, will present states with a powerful inducement to reduce cash benefits. By cutting cash benefits and spending some of its TANF grant on other services once funded by state tax dollars, a state can free up money for tax cuts or other nonwelfare spending. The amount of its block grant would remain unchanged, and the cut in assistance payments would be partially offset because food stamp payments, based on families’ cash income, will grow. The temptation for states to reduce spending on cash assistance and shift the funds to other purposes or tax relief will be particularly strong in the next couple of years. Although AFDC caseloads in many states have dropped noticeably from their peak in the early 1990s, future state block grants are set equal to the federal AFDC payments states got when caseloads were at their peak. With more federal money than they need to maintain benefits for current caseloads, many states will be able to shift some welfare spending to other needs. But when welfare again becomes a pressing fiscal problem, they will find it politically difficult to recapture these monies.

A good case can be made for strengthening state maintenance-of-effort requirements. Effective programs that move recipients toward self-sufficiency are costly. So, too, are the new administrative arrangements needed to implement time limits and other features of the new system.

State Funding Allocations

The TANF block grant has three big shortcomings. It provides uneven levels of federal support to children in rich and poor states. It is not responsive to states’ changing demographic needs. And it may not offer much help in serious recessions. The first two flaws grow out of the decision to allow states to base their grant allocations on the highest of their 1994, 1995, or average 1992-94 spending levels. The decision, based not on compelling policy arguments, but on the political need to avoid visible winners and losers, will lock into place a distribution of federal funds that favors wealthy states. Recently, for example, the federal government spent $1,800 per poor child on AFDC in Connecticut but only $300 in Mississippi. In time, the new allocation formula will help states that are losing low-income residents and hurt those with growing numbers of poor children. If the block grant is to be retained, Congress should make it more equitable. Over a 10-year period the allocation of federal funds could be revised so that spending is based on the number of poor children in a state, not on the amount of federal spending in the first half of the 1990s.

To address the problem of recessions, Congress should increase the contingency grant fund. Budgeted at $2 billion for fiscal 1997-2001, the fund will be available to states with high and increasing unemployment and states whose food stamp caseloads jump at least 10 percent. To compensate states for recession-related burdens, the fund should probably be doubled or tripled. States using the fund should be required to maintain pre-reform funding levels and to match federal dollars they draw from the fund.

Efforts to increase the contingency fund or the TANF block grants will face the same obstacle that blocked the Clinton administration’s welfare reform plan in 1994. New appropriations for the fund must meet deficit neutrality requirements under the Budget Enforcement Act of 1990. Increasing taxes or cutting other programs to increase spending for family assistance is likely to be tough politically—just as it was when policymakers tried to reform AFDC.

Reinventing Reform

Few public programs arouse as much passionþor antipathy—as welfare. Broad dissatisfaction with welfare has led to many calls for reform over the past quarter-century. Some reform efforts have yielded modest adjustments in the structure of welfare, but until 1996 none produced fundamental change.

The Personal Responsibility and Work Opportunity Act is a decisive break with the past. It removes the guarantee of federal support for cash aid to indigent children and their parents, toppling a pillar of social protection that stood for more than six decades. While it is not certain how states will respond to the incentives in the new law, it seems safe to predict that most will offer aid with more strings attached. Cash assistance will be linked to recipients’ efforts to find and keep a job. The emphasis on work is a significant and welcome change in the orientation of welfare. It conforms with Americans’ deeply held belief that able-bodied adults should toil for their daily bread, at least eventually, if they are to earn the right to public support.

But the new law has serious shortcomings. By abolishing the entitlement of needy children to cash assistance, it places a large and extremely vulnerable population at risk. By allowing states to cut spending on welfare, it tempts them to divert resources to other uses, including tax reductions or benefits for more affluent citizens. And it offers scant fiscal relief to states falling into recession.

Sensible reform can correct these defects. If states fail conspicuously to protect the interests of poor children, the impetus for reform will be strong. Because states are most likely to fail when facing economic hard times, welfare should be reinvented one more time—before the next recession.