Introduction

On August 22, 1996, President Clinton signed the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, which replaced the old welfare system with a new program, Temporary Assistance to Needy Families (TANF), designed to focus on work and responsibility and to provide states with flexibility to create the best approaches for their individual circumstances. Overall, between August 1996 and March 1998 there has been a 27 percent decrease in the number of families and recipients on the rolls. The percent of the US population receiving assistance in March 1998 is the lowest since 1969. This report compiles early data about welfare caseloads, family employment and earnings, and state policy choices, to give a preliminary picture of these first two years of welfare reform.

While we will know much more over the coming months and years, the early information presented in this report suggests that:

We have made dramatic progress as a nation on the critical goal of moving families from welfare to work.

Whether it's the employers who have joined the President's Welfare to Work Partnership, the state and local workers who help families find jobs and training, the bus drivers, child care providers, and volunteer mentors who provide day-to-day support to families, or families on welfare themselves, individuals have come together to achieve the goal of moving families from welfare to work.

There is evidence that the Federal government is also doing its share. Under the Clinton Administration, the Federal workforce is the smallest it has been in thirty years. Yet, this Administration believes that the Federal government, as the nation's largest employer, must lead by example. The President asked the Vice President to oversee the Federal government's hiring initiative in which Federal agencies have committed to directly hire at least 10,000 welfare recipients between 1997 and 2000. Already, Federal agencies have hired over 5,700 welfare recipients, or 54 percent of this goal. As a part of this effort, HHS has already hired 250 people, or 83 percent of its goal.

This report brings together a substantial body of national and state evidence suggesting that there has been early progress on that critical goal and in particular that employment among parents who have received welfare and who are at risk of receiving welfare has increased dramatically. For example:

Next year, there will be important new information on employment, as states submit reports on the number of welfare recipients in each state who got jobs this year, how long they held those jobs, and how much their earnings changed over time. States will be submitting these reports in order to apply for the High Performance Bonus, a provision of the statute which allocates $1 billion dollars over five years in prize money to the best-performing states.

We have dramatically increased child support collections and made major progress towards our goal that every child will have the financial and emotional support of both parents.

In most states, state policy and spending choices have reflected a focus on work rather than a race to the bottom.

The next steps on welfare reform are to invest in all families, including those who have the hardest time finding employment, and to ensure that families have the supports they need -- such as affordable, quality child care -- to hold onto a job, reach self-sufficiency, and avoid the need to return to welfare.

The report includes early information that suggests that the job of welfare reform is not yet done and that everyone -- states, employers, local communities, families, the faith and nonprofit communities -- must continue their commitment. For example, early information suggests that as caseloads drop, the proportion of long-stay families on state welfare caseloads is increasing. State welfare evaluations show that while some state welfare policies have strong effects on the employment of families with more barriers to employment, others mostly have effects on those who are the easiest to employ. We must ensure investments that will enable all families to move to work and self-sufficiency.

In 1997, the President issued a challenge to the Governors, saying, "We have continuing responsibility because it's still a national priority..... We ought to take a look at how we're doing -- our successes, our shortcomings and our continuing challenges -- in four areas: jobs, child care, transportation and child support." This remains critical today: we must continue to invest in families to build on our progress to date and ensure welfare reform that works over the long haul. We must invest in all families, not just those who have the easiest time moving into the work force. And, we must invest in the critical supports that families need to hold onto that first job and succeed at work -- supports such as child care, transportation, child support and ongoing training and mentoring.

Below are more extensive highlights describing the information available to date.

EMPLOYMENT OF NEEDY FAMILIES

A substantial body of evidence suggests employment among the parents at risk of welfare recipiency is increasing dramatically. Between 1992 and 1996, the proportion of previous year AFDC recipients employed the following March increased from 19 percent to 25 percent, and jumped dramatically to almost 32 percent in 1997. This percentage increase means that 1.7 million people were working in 1997 who had received AFDC in 1996. Although the strong economy is undoubtedly a major factor in this increase, there is also evidence that welfare reform efforts - - starting with the waivers the Administration granted to 43 states even before the enactment of the 1996 law - - are playing an important role.

The proportion of low income (under 200% of poverty) employed single mothers caring for children under 18 has increased significantly. Employment among this population has increased from 44 percent in 1992 to 54 percent in 1997. The average annual increases in 1996 and 1997 were over twice as large as in the previous three years.

Low income single mothers, the population of parents most likely to be affected by welfare policy, increased their employment substantially compared to comparable married mothers. In 1992 the proportion of all low income mothers with children under 6 that were employed was 35 percent for both single and married mothers. By 1997, the proportion for single mothers had risen to 50 percent. While employment for all women with children has increased, this trend has been sharper for single women. By 1997, the proportion of employment of low income single mothers with young children increased at a rate 3 times faster than for low income married mothers--15 percent versus 5 percent.

To date, evaluations of specific state programs suggest that increased employment of welfare recipients is a result of implementation of welfare policy change. The increases in employment are frequently in the range of 8 to 15 percentage points. These state specific studies are important because they isolate the effects of state policies from external factors such as the economy. Although they are not nationally representative, the approaches being evaluated are quite typical of state TANF policies. Recently, several promising HHS studies have been released, including a Manpower Demonstration Research Corporation study of the Portland, Oregon JOBS program and Urban Institute study of five states' implementation of work first programs. These studies confirm the view that states are moving individuals into better than minimum wage jobs, however, they were released too recently to include in the report. They, and other studies, will be included in subsequent reports.

The proportion of people working in the period following welfare receipt higher than in the past. Approximately 50 to 60 percent of individuals leaving the welfare rolls are working in the period following welfare receipt. This is comparable to or slightly higher than the 45 to 50 percent of those who left welfare who were working after leaving AFDC based on point in time studies. These studies are suggestive but do not rigorously isolate the extent to which this increase in work results from the strong economy in contrast to policy changes.

TRENDS IN CASELOADS AND EXPENDITURES

There have been dramatic declines in welfare caseloads. Overall, between August 1996 and March 1998 there has been a 27 percent decrease in the number of families and recipients on the rolls. The percent of the U.S. population receiving assistance in March 1998 is the lowest since 1969. As Tables 1:1 and 1:2 show, these declines are spread across almost all of the states.

Early data tells us that although states are reducing spending on welfare programs in the aggregate, some states are actually spending more per family given the reduction in caseloads. Thirteen states spent more per family in 1997 than in 1994, recognizing that a work-based system can require up-front investments. States are using these extra dollars in a variety of ways including investing in child care, up-front diversion, rainy day funds, cash and work-based assistance and on state earned income tax credits. In FY 1997, almost half of the states reported spending more than they are required to spend of their own funds. Specifically, 22 states, or 43%, reported spending above the required maintenance of effort level. Five states reported expenditures which were more than 125% of the maximum required: Alaska, Arkansas, Delaware, Missouri, and South Dakota.

CHILD SUPPORT COLLECTIONS

The nation's child support program has made dramatic improvements in critical areas. In 1997, the state and federal child support enforcement program collected a record $13.4 billion for children, an increase of 68% from 1992, when $8 billion was collected. Not only are collections up, but the number of families that are actually receiving child support has also increased. In 1997, the number of child support cases with collections rose to 4.2 million, an increase of 48% from 2.8 million in 1992.

There has been a substantial increase in the number of paternities established. The state and federal child support enforcement programs established a record 1.3 million paternities in 1997, two and a half times the 1992 figure of 510,000. Much of this success is due to the in-hospital voluntary paternity establishment program begun in 1994 which encourages fathers to acknowledge paternity at the time of the child's birth. For the first time, we have established paternities equal to the number of out-of-wedlock births.

OUT-OF-WEDLOCK BIRTHS

The birth rate for unmarried women aged 15-44 years decreased slightly between 1995 and 1996. This rate decreased from 45.1 births per 1,000 women in 1995 to 44.8 in 1996. One of the goals of the PRWORA legislation is to prevent and reduce the incidence of out-of-wedlock pregnancies. The Administration for Children and Families issued proposed regulations implementing section 403(a)(2) of the Social Security Act which establishes a bonus to reward decreases in out-of-wedlock births on March 2, 1998, and expects to release final regulations this fall.

Data show that teenage birth rates (under 20 years) have declined. Nationally, the birth rate for teenagers continued to decline in 1996, and has now fallen by 12 percent to 54.4 births per 1,000 women aged 15-19 years, compared with 62.1 in 1991. Teenage birth rates by state vary substantially, from 28.6 (New Hampshire) to 102.1 (District of Columbia).

CHILD POVERTY

According to the official poverty measure, child poverty has declined since 1993 but remained constant between 1995 and 1996. Approximately 20.5 percent of all children were poor in 1996.

The EITC (Earned Income Tax Credit) lifts one out of seven children out of poverty. The decline in child poverty since 1993 is much sharper using a more inclusive measure of income which includes the EITC rather than the official measure. Since 1993, the maximum size of the EITC has increased by 140 percent for a family with two children. Children moved out of poverty by the EITC constitute 14.5 percent of the children who would have been poor in the absence of government programs. Using a more inclusive measure of income that includes FICA employee payroll and income taxes, the EITC and in-kind benefits, including Food Stamps and Housing Assistance, the child poverty rate for 1996 was 16.1 percent.

DEMOGRAPHIC AND FINANCIAL CHARACTERISTICS OF FAMILIES RECEIVING ASSISTANCE

Long term recipients are an increasing percentage of state caseloads. Since FY 1994, there has been a small but steady decline in the percentage of the caseload who have been on assistance for one year or less (36% to 33%) and a corresponding increase in the percentage of the caseload on assistance five years or more (19% to 24% ).

Earned income has increased among welfare recipients. The average earnings per family on welfare has increased by 7 percent, from $466 in FY1996 to $500 in FY1997. The percentage of families with earned income has increased steadily from 9 percent in FY 1994 to 13 percent in FY1997.

The average monthly TANF payment has decreased slightly. Monthly TANF payments averaged $362 per family in FY 1997, compared to $370 in FY 1996.

Child only cases accounted for 23 percent of the TANF caseload. While this is a dramatic increase from 10 percent in 1988, the trend seems to be slowing. The 1997 figure is only one percentage point above the 1996 figure of 22 percent and the absolute number of child only cases actually declined from 1996 to 1997.

The average size of TANF families has remained constant. The TANF families averaged 2 recipient children, which is consistent with the data from recent years. Three in every four families had only one or two children. Seventy percent of families had only one adult recipient, and 7 percent included two or more adult recipients.

STATE POLICY CHOICES

Most states have changed the way they count income under TANF to enable working recipients to keep more of their check. Forty-two states have enacted policies to change the way income is counted in determining eligibility and benefits. Most of these have increased their earnings disregards.

Most states have maintained their benefit levels. According to state TANF plans, nine states have increased while eight have decreased their benefit levels.

States policies increasingly support working families. Thirty-eight states raised their general resource limits in order to promote accumulation of assets to achieve self-sufficiency, and 47 states have raised their automobile resource limits. To help families transition off assistance, 29 states indicate they are extending child care benefits for more than 12 months, and 12 states provide transitional medical assistance for more than 12 months.

States are beginning to focus more attention on the hard-to-serve and fragile families. For example, 26 jurisdictions have elected the Family Violence Option to ensure that victims of domestic violence receive appropriate protections and services, and most states exempt parents of infants from work requirements.

States are beginning to turn their welfare offices into employment offices and are taking a variety of steps to reinforce the work message. Almost all states have adopted a "Work First" model for setting individual expectations and responsibility and for structuring employment and training services. This approach emphasizes early entry into the job market. Thirty-two states expect parents to participate in work within six months (compared to the statutory standard of 24 months).

DATA AVAILABILITY

The first annual report to Congress on the TANF program provides the most current data available and attempts to set the groundwork for subsequent reports which will contain a greater amount of information. Not all states were required by the statute to report data under new TANF reporting requirements for any part of FY1997, and among those that were, no state had to report for more than the last quarter of the fiscal year. Furthermore, states are in the process of fully implementing their new data reporting systems.

As might be expected, converting from the old AFDC system to the new TANF system has proven to be very challenging for states and HHS, and all parties have been working diligently to update the data reporting system. However, some data, including information on participation rates in the last quarter of FY 1997, that will be included in future years, are not yet available.In many areas where HHS is dependent on non-TANF reporting systems to provide data, e.g., measurement of child poverty and out-of wedlock births, data also are not yet available for periods after the implementation of TANF.

Where there is no post-TANF implementation data to report to Congress, earlier data from the AFDC Program has been provided. This will serve as a valuable baseline for subsequent reports. In addition, other important information regarding TANF has been included although it is not required to be in the report by the statute. As data become available throughout the year it will be made public.

OTHER REPORTS

The Bureau of the Census will continue to collect data on the Survey of Income and Program Participation. This data will enable interested persons to continue to evaluate the effects of the TANF program on recipients of assistance and other low income families. The Bureau of the Census will pay particular attention to the issues of out-of-wedlock birth, welfare dependency, the beginning and end of welfare spells, and the causes of repeat welfare spells.

This is one of several reports which have been mandated in PRWORA in order to monitor the well being of children and families on public assistance, and to gauge the success in moving families to self-sufficiency. HHS will also be issuing other reports in the future including an annual ranking of states and a review of the most and least successful work programs, and an annual ranking of states and review of issues relating to out-of-wedlock births. Furthermore, HHS will report beginning 3 years after the date of enactment of PRWORA on the circumstances of certain children and families. This report will provide information on individuals who were children in families that have become ineligible for assistance by reason of having reached a time limit, children born after such date of enactment who have not attained 20 years of age, and individuals who became parents before attaining 20 years of age.