In brief

In 1996, the Clinton Administration reformed the US welfare system with the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), which created the Temporary Assistance for Needy Families (TANF) programme. It was supported by a large majority of politicians and the public for its cost-cutting efforts and changed the US benefits system for families by reducing the number of welfare cases by over 60 percent.

The challenge

Discussions about reforming the US’s 1935 social welfare policy on financial assistance to children in low-income families began in the 1950s. The Aid to Families with Dependent Children (AFDC) programme was expensive for the government and was believed to be ineffective in breaking the poverty cycle. The perception was that low-income households had become dependent on benefits payments and were unwilling to seek employment. The reform became a major goal of Bill Clinton during his 1992 presidential campaign, as he aimed “to end welfare as we know it”.[1]

The initiative

In 1996, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) was passed into law. It instituted the Temporary Assistance for Needy Families (TANF) programme, which was intended to replace the AFDC as the main welfare provider for low-income families in the US by the following year.[2]

When PRWORA was passed, it delegated greater control of welfare policy to the states via a federal block grant. This meant that additional federal funds should no longer be used to pay for TANF benefits “to any adult for more than a total of sixty months during that person’s lifetime”.[3]

Legislators backed the reform because it was meant to achieve several objectives that would in their view improve family welfare (see also Clarity of Objectives): "Legislators articulated the following goals for reform: (1) provide assistance to needy families so that children may be cared for in their own homes or in the homes of relatives; (2) end the dependence of needy parents on government benefits...; (3) prevent and reduce the incidence of out-of-wedlock pregnancies; and (4) encourage the formation and maintenance of two-parent families".[4]

The public impact

As PRWORA was meant to reform the US social care system for children, the TANF initiative had a significant impact on society. The major change attributed to TANF was the dramatic decline in the number of families on social welfare. "The total caseload fell more than 60 percent from the peak of 1994 to 2002, a period roughly contemporaneous with the sustained economic expansion of 1992-2000. [Additionally,] participation among single mothers has dropped dramatically, from 25 percent in 1996 to 9 percent [in 2004]."[5]

The decline of the caseload indicates that American families were made to rely less on government welfare services overall. However, there were additional impacts that affected the livelihood of low-income families as a whole.

Firstly, while welfare payments decreased, it made recipients actively seek employment: “among welfare recipients, the percentage that reported earnings from employment increased from 6.7 percent in 1990 to 28.1 percent by 1999”.[6]

Secondly, TANF had an impact on related government programmes. Foreign nationals were excluded; it required states to provide Medicaid to families receiving support under TANF; and the distribution of food stamps was changed. “It gave states more control over food stamp operations and coordination with family cash aid, added a work rule for able-bodied adults without dependents and expanded existing work rules, cut future benefits and tightened financial and non-financial eligibility tests, expanded penalties for violating rules and controls over trafficking, and encouraged electronic delivery of benefits.”[7]

Thirdly, it was a major challenge for the legislature to transition from a federal to a state-led approach to funding. The federal block grant that was introduced meant that individual states had to consider their welfare spending habits carefully, as every additional dollar had to be paid for out of their own pockets.[8]

Have an idea for a case study? Print

What did and didn't work

All cases in our Public Impact Observatory have been evaluated for performance against the elements of our Public Impact Fundamentals.


Stakeholder Engagement Good

While political and public discourse about welfare and its reform was changing constantly over several decades, a number of stakeholders remained involved throughout, including state governments, the federal government, local welfare agencies, social workers, low-income families, and NGOs.

Generally, all parts of government through “federal laws, state laws, and state and local policies, have sought to change the purposes and operations of welfare programmes practically since the original federal law was passed over 60 years ago in 1935".[9]

Specifically, states legislatures began to initiate the first welfare-to-work programmes as early as 1981. These programmes granted state agencies the freedom to experiment with job search, public service and employment training programmes in exchange for benefits. The Ford Foundation became involved to offer a large-scale evaluation of the welfare-to-work schemes in all states, resulting in the conclusion that welfare-to-work programmes had “modest effects on… recipients finding jobs”.[10] The results of these evaluations and policy experimentation gave state governors, such as Bill Clinton, the idea of promoting welfare reform during his presidential campaign.

Political Commitment Good

Bill Clinton advocated for welfare reform during his 1992 campaign, but it was the Republicans who actively pushed the reform forwards after they won a majority in Congress in the 1994 congressional elections.

On the Republican side, the 1994 "Contract with America" provided the groundwork to pursue welfare reform. It was a policy proposal collection intended to convince voters during the congressional elections and proposed a stricter method of cutting social welfare. Its proposals found favour with many of their fellow members of congress, who drafted their own legislation without the involvement of the White House. The legislation pushed for a strong connection between benefits payments and employment. It was "based on a simple compact designed to reinforce and reward work. Each recipient will be required to develop a personal employability plan designed to move that individual into the workforce as quickly as possible.”[11]

President Clinton initially submitted draft legislation to Congress in early 1994 in the form of the Work and Responsibilities Act (WRA). However, he was soon busy amending US healthcare legislation, and the Republicans accused him of stalling on welfare reform. Officials secretly admitted that Bill Clinton aimed “to delay welfare legislation so it does not interfere with the Administration's primary goal of healthcare”.[12] The team he put in place in the White House was sceptical of the Republican proposal, as they estimated it would “push more than a million children into poverty".[13]

Under increased pressure from Congress, President Clinton had to make a commitment to find a deal after twice vetoing Congress’s proposal. When the welfare bill was signed, three influential assistant secretaries, Mary Jo Bane, Peter B. Edelman, and Wendell E. Primus, resigned in protest and expressed their "deep concerns" about it.[14]

Public Confidence Good

Public opinion prior to the reform was positive, because the American public thought that the existing welfare system was too expensive. It was a commonly held belief that the poor should proactively seek jobs to break their dependency on social benefits. According to a national public opinion poll on welfare in 1995, a majority of the American public “[thought] the government is spending too much on welfare."[15] This view was reflected in the state legislatures, where “the broadest support was for imposing strict and extensive employment obligations on adult recipients of welfare benefits".[16]

Welfare workers themselves were not initially opposed to the proposed changes under PRWORA. They believed that the Act would give them the authority to encourage jobseekers to apply for employment. In New York, a welfare administrator remarked: “We now have permission to be ‘be real’ with clients, to make them understand they have an obligation to work, to help themselves".[17]


Clear Objectives Fair

PRWORA did not formulate the objectives of the reform clearly nor did it set out a specific schedule. As Bill Clinton had to govern with a Republican-led Congress, vetoes meant that the original bill had to be amended to obtain bipartisan agreement.[18]

The final policy document stated the following objectives:

  1. "Provide assistance to needy families...
  2. “End the dependence of needy parents on government benefits by promoting job preparation, work, and marriage.
  3. “Prevent and reduce the incidence of out-of-wedlock pregnancies.”
  4. “Encourage the formation and maintenance of two-parent families."[19]

Evidence Strong

A lot of research had been published prior to the welfare reform. The determining factors identified were caseloads and costs: between 1989 and 1993, AFDC caseloads grew by 33 percent, while welfare costs in general had grown by 116 percent since 1980.[20]

Since the 1980s, America had witnessed a broader discourse on social benefits that had served to convince Congress that reform was necessary. The Manpower Demonstration Research Corporation (MDRC) argued in their study of the topic that “requiring welfare mothers to work could have positive effects on their employment and earnings, while reducing their dependency... [Charles] Murray contended in Losing Ground that the disincentives against good behaviour created by welfare and other social programmes were the principal reason for the demoralisation of the poor; the only solution was to abolish most aid for the working-aged.”[21]

As a result of these debates, projects in the 1980s and early 1990s experimented with setting demands for its participants. The MDRC asserted that “putting clients to work in available jobs, even if low-paid, outperformed those that stressed training or education for better-paid positions”.[22] Bill Clinton adopted this argument when advocating PRWORA, emphasising that better childcare, healthcare and child support services were needed to compensate for the lack of benefits.

Strong criticism of the welfare reform bill came from a number of Democrats, who objected that it did not serve its purpose. Peter Edelman, former assistant secretary for planning and evaluation at the Department of Health and Human Services, who resigned in protest of the welfare reform, remarked: “It does not promote work effectively, and it will hurt millions of poor children by the time it is fully implemented. What’s more, it bars hundreds of thousands of legal immigrants – including many who have worked in the United States for decades and paid a considerable amount in Social Security and income taxes – from receiving disability and old-age assistance and food stamps, and reduces food-stamp assistance for millions of children in working families.”[23]

Further projections estimated that 12.8 million people on welfare at the time were at risk of "sinking further into poverty and homelessness".[24]

Feasibility Fair

There was broad bipartisan agreement on reforming the system, based on its increasing costs, and from that perspective PRWORA and the TANF programme were very effective. Block grants ensured that states no longer bore their share of the unlimited federal administrative costs. This meant that that any additional benefit had to be paid by state funds and would not increase the federal government’s burden.

PRWORA also required additional management resources in work and employment related programmes, which were initially unavailable, and increased states' welfare payments. Hence, the law permitted individual states “to set aside any amount of their TANF grants they like”[25] for contingency planning. This increased the risk of their being even fewer benefits available.[26]

However, PRWORA and the TANF programme's goals conflicted. Michael Wiseman from the University of Wisconsin questioned whether a dependency on welfare benefits, which are intended to help citizens avoid poverty, could be mitigated by employment. Accordingly, many of the causes of poverty remained unchanged because they resulted from the unavailability of suitable jobs rather than welfare recipients' unwillingness to work. "No state had yet been successful in generating the quantity of public employment that the WRA would likely require."[27]


Management Good

The implementation of PRWORA was the responsibility of individual states, which now had to set their own objectives and timelines. This was the deliberate intention of the reform, which shifted power from the federal to the local level.

Existing social services agencies, which administered the old AFDC programme, were targeted in a majority of states. In Arizona, the Senate planned “to shrink the existing bureaucracy involved in the delivery of welfare service (most prominently the Department of Economic Security) and almost entirely privatise the new temporary assistance and job training programmes".[28] In New York, the entire Department of Social Services was dissolved and its functions integrated into the Departments of Health, Labour, and Taxation and Finance. Responsibilities were devolved as far as county level in order to have a welfare service structure that was as close to local needs as possible.[29]

To compensate for the leaner social services bureaucracy, employment agencies were granted greater control over the TANF programme's operations in a majority of states. This changed the institutional character of welfare, as employment agencies' primary function was to "move people into the workforce". “By locating welfare offices in job centres, states can signal that people can and should get a job and that welfare is part of a service system for jobseekers, rather than a single programme unto itself.”[30] This led to a situation where employment, labour or workforce development agencies were deciding which families were eligible for TANF, and were also administering financial assistance.

Measurement Fair

At the federal level, PRWORA’s success was measured mainly by the social welfare caseloads, which were recorded by the Department of Health and Human Services. By June 2000, the number of cases had fallen by 53 percent to 6.5 million. Given that different states had completely different systems in place, it remains hard to measure the overall success. While the law did provide for data collection through household surveys by the states, “[they] simply lack the capacity to collect it".[31]

Independent studies often focused on the broader economic indicators, such as the unemployment rate, the poverty rate, the number of poor female-headed families, and the number of individuals being paid a poverty wage. It was hard to attribute the effects on these indicators to PRWORA and TANF alone. For that reason, independent ethnographic research was also used to evaluate impact. On the one hand, the number of poor female-headed families with children dropped by 22 percent; some states reported a drop in the poverty rate by almost 9 percent; and unmarried women with children became more readily employed. On the other hand, it is argued that there was no rise in the living wage, so that the poor increasingly converged with "the working poor". Additionally, “because of the nature of jobs available… and their poor preparation for the labour market, many low-skilled former TANF recipients find jobs that pay poorly and do not last".[32]

Alignment Weak

Alignment at state level was weak because the discretion of state legislatures led to a variety of different eligibility and restriction criteria being applied in different US states.

Many state agencies were eager to enact the new reforms because “political and economic signals demonstrated to state and local bureaucracies that the emphasis on work was not going away easily or soon”.[33] States appeared to have enough leeway to adapt the reforms to their own circumstances. They were able to alter some of the eligibility criteria for cash assistance and have the discretion to select which providers could carry out their programmes.

Taken together, all the states generally accepted TANF provisions on work requirements in exchange for benefits. It was reported, that a “majority of states adopted stiffer initial work requirements, and a large minority of states strengthened the federal sanction policies and cut the time limits”.[34]

While states initially experimented with the possibility of  redesigning the law to some extent, either through access-enhancing or access-restricting amendments, most reportedly settled their legislation around 1998. This trial period however, did not lead to a common set of shared practices but to dramatic differences in their overall policy.[35]


America's Changing Attitudes Toward Welfare and Welfare Recipients 1938-1995, Laurie MacLeod, Darrel Montero, Alan Speer, June 1999, Volume 26, Issue 2, The Journal of Sociology and Welfare

Assessing the Impact of Welfare Reform on Single Mothers, Hanming Fan and Michael P. Keane, January 2004, Brookings Papers on Economic Activit, Brookings Institution

Clinton Puzzle: How to Delay Welfare Reform Yet Seem to Pursue It, Jason De Parle, 5 January 1994, The New York Times

Implementing the Personal Responsibility Act of 1996: A First Look, Richard P. Nathan and Thomas L. Gais, Federalism Research Group, The Nelson A. Rockefeller Institute of Government, State University of New York

Interview: Welfare Reform, 10 years later, Ron Haskins, 24 August 2006, Brookings Institution

Is Devolution Working? Federal and State Roles in Welfare, Richard P. Nathan and Thomas L. Gais, 1 June 2001, Brookings Institution

Research and Welfare Reform, Lawrence M. Mead, 8 July 2004, New York University

Short History of the 1996 Welfare Reform Law, 7 February 2001, CRS Report for Congress

State Policy Choices Under Welfare Reform, R. Kent Weaver and Thomas Gais, 2 April 2002, Brookings Institution

The Worst Thing Bill Clinton Has Done, Peter Edelman, March 1997, The Atlantic

United States National Educational and Social Development Policy Handbook, Volume 2 – Social Policy: Important Programs and Regulations, 2015, International Business Publications

Welfare-Reform Critics Were Wrong, Robert Rector, 3 March 2003, The Heritage Foundation

Welfare Reform: How Do We Measure Success?, Daniel T. Lichter and Rukamalie Jayakody, 2002, Annual Review of Sociology

Welfare Reform in the United States: A Background Paper, Michael Wiseman, 1996, Volume 7, Issue 4, Housing Policy Debate

Welfare Reform under PRWORA: Aid to Children with Working Families?, Pamela Loprest, Stefanie Schmidt and Ann Dryden Witte, January 2000, Tax Policy and the Economy: Volume 14, National Bureau of Economic Research

Sign up to stay updated on news about our meetings, our insights and our other activities.
Back to top