Next Steps in Welfare Reform

Testimony Welfare

Next Steps in Welfare Reform

April 11, 2002 13 min read
Robert Rector
Senior Research Fellow
Robert is a leading authority on poverty, welfare programs and immigration in America.

Before I begin, let me first thank the committee for the opportunity to speak before you today. While I serve as Senior Research Fellow on Welfare and Family Issues at The Heritage Foundation, I must stress that the views I express are entirely my own, and should not be construed as representing the position of The Heritage Foundation.


Six years ago this month, President Clinton signed legislation overhauling part of the nation's welfare system. This legislation, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), replaced the failed Aid to Families with Dependent Children (AFDC) program with a new program called Temporary Assistance to Needy Families (TANF). The reform legislation had three goals: 1) to reduce dependence and increase employment; 2) to reduce child poverty; and 3) to reduce illegitimacy and strengthen marriage.

At the time of its enactment liberal groups passionately denounced the new welfare reform legislation, predicting it would result in substantial increases in poverty, hunger and other social ills. Contrary to these predictions, welfare reform has been effective in meeting each of the three goals listed above.

  • Overall poverty, child poverty and black child poverty have all dropped substantially. While liberals predicted that welfare reform would push an additional 2.6 million persons into poverty, in fact there are 4.2 million fewer people living in poverty today than when welfare reform was enacted.
  • Child poverty has declined substantially as well; some 2.3 million fewer children in poverty today than in 1996.
  • The decreases in poverty have been the greatest among black children. The poverty rate for black children has fallen by a third and is now at the lowest point in U.S. history. There are 1.1 million fewer black children in poverty today than in the mid-1990's.
  • The poverty rate for single mothers has been cut by one third and is now at the lowest point in U.S. history.
  • Hunger among children has almost been cut in half. According to the U.S. Department of Agriculture there are nearly 2 million fewer hungry children today than at the time welfare reform was enacted.
  • Welfare caseloads have been cut nearly in half.
  • Employment of the most disadvantaged single mothers has increased from 50 to 100 percent.
  • The explosive growth of out-of-wedlock childbearing has come to a virtual halt.
  • The share of children living in single mother families has fallen and the share living in married couple families has increased, especially among black families.

    Some attribute these positive trends to the strong economy in the late 1990's. While a strong economy was a contributing factor to some trends, most of the positive changes greatly exceed those in prior economic expansions. The difference is welfare reform. The challenge before Congress is to build on and strengthen the 1996 reform.

Work Requirements:

The key to the success of welfare reform so far has been work requirements. Under the reform act, a certain portion of TANF recipients were required to undertake constructive activities aimed at self-sufficiency: supervised job search, training, and community service work. When TANF recipients were required to undertake constructive activities as a condition of getting aid, the number of new applications to welfare fell precipitately and the length of stay on welfare dropped dramatically. Caseloads plummeted. As caseloads fell, the employment level of single mothers soared. Child poverty in single mother families dropped by a third to the lowest level in U.S. history.

To continue this success, we need to improve upon the work requirements in the current TANF law. To do this, five points are important.

Retain and update the primary goal of caseload reduction:

The current TANF law required states to reduce their TANF caseloads by roughly 50 percent over five years. States which failed to meet this goal were required to have the "residual caseload" engaged in work activities. (The "residual" caseload is the number of cases that exceed the caseload reduction targets.) Thus if a state cut its caseload by 30 percent, the residual caseload to be engaged in work activities would be 20 percent (i.e., the difference between the reduction goal of 50 percent and the actual reduction of 30 percent).

The caseload reduction standards have worked well but they are now obsolete. These standards should be updated. Specifically, states should be required to reduce their 2001 caseload levels by 80 percent over the next five years. (Cases without parents should be excluded from this standard.)

If all states met this goal, there would be only 300,000 TANF cases with parents left nationwide by 2007. As with current law, if any state fails to meet its caseload reduction goal, it would be required to have the extra or residual caseload engaged in work activities. States that fail to meet their caseload reduction goal and work-activity requirements should be subject to the same financial penalties as under current law.

Do not count case "exits" as a performance measure:

Caseload "exits" (the number of persons leaving the caseload) are a misleading performance measure. Typically, both AFDC and TANF have had around 2 million case closures or exits each year. There have been millions of exits even in years where the caseload is rising rapidly. Thus the count of exits creates an illusion of success even when dependence is increasing. Dependence is reduced only when the number of exits is greater than the number of entries or case openings; dependence is reduced only when the size of the overall caseload is falling. To evaluate the success of welfare by the volume of exits can create the misperception that the nation is reducing dependence even when caseloads are constant or rising.

Performance measures based on the "quality" or "success" of exits are also misleading. These measures include items such as: the average wage of persons leaving TANF or the average length of employment of TANF leavers. While these measures may appear reasonable, in reality, they are very misleading as measures of performance. This is because the "success" of persons leaving TANF is largely determined by the characteristics of persons entering TANF. For example, the average wage of persons leaving TANF is largely determined by the education levels of those entering TANF. States with liberal entrance rules and higher benefit levels will draw better educated and more employable person into the front door of TANF, and this will obviously result in higher average wages among those exiting out the rear door. By contrast, conservative states with low benefits and strict entrance policies will tend to have only those who truly need aid entering the program. These entrants will have lower average education levels and will tend to have lower average wages when they leave TANF. "Exit" performance measures that reward states for having higher welfare benefits and more liberal entrance rules should be avoided.

Require full check sanctions for parents who refuse to perform required activities:

In most states, individuals who refuse to undertake required activities (such as job search, community service or training) have their entire TANF check cut off; this is called "full check sanction". Unfortunately, sixteen states sanction only the "adult portion" of the TANF check, which is generally worth about $50 per month. In these states non-compliant parents continue to receive the bulk of their welfare check even after refusing to engage in activities required by the state. Truthfully, these states have "work suggestions" rather than "work requirements" since parents receive little more than a slap on the wrist if they refuse to comply with the state's "requirements". Regrettably, states that have only a weak or partial sanction for non-cooperating parents include large states such as: New York, California, Pennsylvania, and North Carolina. Overall, the states with weak sanction policies now have 53 percent of the national TANF caseload. Since welfare caseloads are falling less rapidly in these states with "work suggestions" rather than work requirements, within a few years states with weak sanction policies will have the vast majority of adult TANF cases. The federal law should insist on work requirements not "work suggestions"; all states should be required to establish full check sanction policies.

Require continuous constructive activity by TANF recipients:

Another important element of a successful work-based welfare system is to require continuous supervised activity by recipients. States that require TANF recipients to continuously engage in activities directed toward self-sufficiency are far more successful in reducing welfare dependence than those who do not. Allowing a recipient to sit idly on the welfare rolls is good for neither the recipient nor the taxpayer. This principle should be a key element in the next stage of welfare reform. The new law should state that once a TANF recipient has signed a self-sufficiency plan, the recipient should not be permitted to remain idle on the rolls, but should be required to be continuously engaged in constructive activities aimed at self-sufficiency.

Recognize that work requirements do not require added government funds:

Ever since President Bush announced his welfare reform plan, critics have charged that implementation of the work requirements in the plan will require increased government outlays. This is not a new charge. The same charge was used to block work requirements for welfare recipients during most of the 1980-1995 period.

Identical arguments were raised against the 1996 welfare reform law. Those arguments were false then, and they are false now. The idea that work requirements (requiring welfare recipients to engage in community service work or other constructive activities) cost more than simply mailing benefit checks to recipients has been decisively refuted by the record since 1996.

The central fallacy in this argument is the assumption of static caseloads. If caseloads remain static, then requiring job search, training or community service work does cost more than the traditional welfare, because the government must pay the welfare benefit check as well as the costs of daycare while the recipient is performing the required activities.

But the central lesson of the 1996 PRWORA reform is that caseloads are not static: work requirements dramatically reduce caseloads. Declining caseloads lead to reduced cash benefits. Reduced cash benefits generate savings that can be redirected to pay for daycare and other ancillary social services. By this principle welfare reform in the 1996 to 2002 period has been largely self-financing. The same principle will apply to the next stage of reform.

In addition, it is important to note that, with respect to work requirements, the next stage of TANF reform will be on a much smaller scale than the PRWORA reform. Between 1995 and 2001, the AFDC/TANF caseload fell by three million (from 5 million cases to 2 million). During the next stage of reform, the TANF caseload is likely to drop only by one million cases from the current level of 2 million cases. This would leave around one million cases on TANF in 2007: around 300,000 adult cases and 700,000 child only cases (which are exempt from work requirements.)

Thus, even if one accepts the argument that work requirements demand greater government spending, such spending would be far more modest over the next five years than the last, because the size of the caseload affected by the requirements is so much smaller.

Promoting Healthy Marriages:

Today nearly one third of all American children are born outside marriage. That's one out-of-wedlock birth every 35 seconds. This collapse of marriage is the principal cause of child poverty and a host of other social ills. A child raised by a never-married mother is seven times more likely to live in poverty than a child raised by his biological parents in an intact marriage. Overall, some 80 percent of the long-term child poverty in the U.S. occurs in broken or never-formed families. In addition, children in these families are more likely to become involved in crime, to have emotional and behavioral problems, to be physically abused, to fail in school, to abuse drugs, and to end up on welfare as adults.

The growth of single parent families has an enormous impact on government. Indeed, the modern welfare state, as it relates to children, has grown up largely as a support system for single parenthood. For example, in programs such as Food Stamps and public housing, some 80 to 90 percent of aid to children flows to single parent families.
Overall, some 75 percent of government means-tested aid to families with children goes to single parent families. At present, federal and state governments spend some $150 billion per year in means-tested aid for single parent families. Yet, despite this considerable expenditure, little or nothing is devoted to reducing the underlying force which propels this expenditure: the decline of marriage itself. For each thousand dollars which government spends subsidizing single parenthood only one dollar is spent trying to limit its debilitating spread.

With the enactment of welfare reform in 1996, Congress did take the first tentative steps to redress this historic neglect. The reform law established, for the first time, formal national goals of reducing out-of-wedlock childbearing and strengthening marriage. State governments were charged with the task of using federal Temporary Assistance to Needy Families (TANF) welfare funds to carry out this new mission. Unfortunately, states have ignored the purposes of the reform law; only a handful have developed even meager pro-marriage programs.

The next stage of reform, therefore, must be to actually carry out the original goals of the 1996 legislation. To revitalize marriage we must reduce illegitimacy and divorce and improve the quality of existing marriages. Pro-marriage reforms must endeavor to restore the culture of marriage in at risk communities. They must seek: to instill an understanding of the value of marriage; to provide skills needed to sustain successful long-term marital relations; and to reduce the disincentives to marriage implicit in most government welfare programs.

Understanding How the Existing Welfare System Penalizes Marriage:

The U.S. welfare system is currently comprised of over 70 means-tested aid programs providing cash, food, housing, medical care, and social services to low income persons. While it is widely accepted that welfare is biased against marriage, relatively few understand how this bias operates. Many erroneously believe that welfare programs have eligibility criteria that directly exclude married couples. This is not true.

Nevertheless, welfare programs do penalize marriage and reward single parenthood because of the inherent design of all means-tested programs. In a means-tested program, the benefits are reduced as non-welfare income rises. Thus, under any means-tested system, a mother will receive greater benefits if she remains single than if she is married to a working husband. Welfare not only serves as a substitute for a husband, it actually penalizes marriage because a low-income couple will experience a significant drop in combined income if they marry.

For example: the typical single mother on Temporary Assistance to Needy Families receives a combined welfare package of various means-tested aid benefits worth about $14,000 per year. Suppose this typical single mother receives welfare benefits worth $14,000 per year while the father of her children has a low wage job paying $15,000 per year. If the mother and father remain unmarried, they will have a combined income of $29,000 ($14,000 from welfare and $15,000 from earnings.) However, if the couple marries, the father's earnings will be counted against the mother's welfare eligibility. Welfare benefits will be eliminated or cut dramatically and the couple's combined income will fall substantially. Thus means-tested welfare programs do not penalize marriage per se, but instead implicitly penalize marriage to an employed man with earnings. Nonetheless, the practical effect is to significantly discourage marriage among low-income couples.

This anti-marriage discrimination is inherent in all means-tested aid programs including: TANF, Food Stamps, Public Housing, Medicaid, and the Women Infants and Children (WIC) food program. The only way to eliminate the anti-marriage bias from welfare entirely would be to remove means-testing from all welfare programs, making all mothers eligible irrespective of their husbands' earnings. This, however, would cost tens of billions of dollars.

While we cannot completely eliminate the anti-marriage bias in welfare, we can take constructive steps to reduce the bias and to increase the number of healthy marriages. The welfare reform proposal designed by President Bush and introduced by Congressman Wally Herger would provide $300 million a year for pilot programs to strengthen marriage. These programs would provide marriage skills training to couples interested in marriage. The proposed policy would also permit experiments in reducing the penalties against marriage in welfare programs. This proposal is a critical first step in efforts to improve child well-being in our nation.

Will Increasing Healthy Marriages Reduce Child Poverty?:

It is indisputable that increasing healthy marriages will dramatically improve child well-being. Healthy marriages help children by reducing: child abuse, juvenile delinquence, drug abuse, depression, behavior problems, school failure, and future crime and welfare dependence.

Recently, however, certain vocal groups have challenged the idea that increasing stable healthy marriages will substantially reduce child poverty over the long term. This challenge seems perplexing since child poverty is four to seven times higher in single parent than in married couple families.

My colleagues and I at The Heritage Foundation have recently completed some research on the effects of marriage on poverty using the Current Population Survey of the Census Bureau. We find that the growth of single parenthood since the 1960's is responsible for around a third of current child poverty. In addition, we find that if existing single mothers were married to men, who were identical with the mothers in race, age and level of education, the marriages would be effective in eliminating child poverty in about 80 percent of cases. Similar research by Dr. Sara McLanahan, using data from the Fragile Families survey of non-married women at the time of birth, shows that marriage to the child's father could be effective in reducing, by up to 90 percent, the level of child poverty. These data indicate that restoration of marriage over the long term would have a strong positive effect in reducing child poverty.

Similarly, some assert that increasing the education levels of single mothers would be a far more effective way of reducing child poverty than would be increasing healthy marriages. This is untrue. For example, a child, raised in an intact married family with a mother who is a high school graduate, is two thirds less likely to be poor than is a child raised by a never-married mother who is a college graduate.


The key elements of future welfare reform must be: requiring work, promoting healthy marriages, and improving child well-being. The welfare reform plan developed by President Bush and Congressman Herger is a strong positive step forward in this regard.

The Heritage Foundation is a public policy, research, and educational organization recognized as exempt under section 501(c)(3) of the Internal Revenue Code. It is privately supported and receives no funds from any government at any level, nor does it perform any government or other contract work.

The Heritage Foundation is the most broadly supported think tank in the United States. During 2013, it had nearly 600,000 individual, foundation, and corporate supporters representing every state in the U.S. Its 2013 income came from the following sources:

Individuals 80%

Foundations 17%

Corporations 3%

The top five corporate givers provided The Heritage Foundation with 2% of its 2013 income. The Heritage Foundation’s books are audited annually by the national accounting firm of McGladrey, LLP.

Members of The Heritage Foundation staff testify as individuals discussing their own independent research. The views expressed are their own and do not reflect an institutional position for The Heritage Foundation or its board of trustees.


Robert Rector
Robert Rector

Senior Research Fellow