About the Author

Avatar photo

Professor Lawrence Mead

Why Welfare Reform Succeeded

Print Friendly and PDF
Posted on

Welfare reform was a great but incomplete triumph. It moved the welfare poor toward work. It was a major achievement for government. And its political effects could well make good its shortcomings, provided the poor mobilize politically.


“Welfare reform” connotes all the changes government made in family welfare in the 1980s and 1990s, chiefly to move adult recipients into jobs. This includes the radical Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) of 1996, which recast Aid to Families with Dependent Children (AFDC) as Temporary Assistance for Needy Families (TANF), changed the program from an entitlement to a block grant, stiffened work and child support requirements, limited families to five years of aid, denied coverage to many noncitizens, and gave states more control over welfare.

But reform also includes earlier efforts to enforce work, going back to the Family Support Act (FSA) of 1988 or before. Where FSA put most welfare mothers in school or training, PRWORA demanded that they “work first”—take available jobs.

Reform also included efforts to “make work pay” through expansions of the Earned Income Tax Credit (EITC) and greater health and child care funding. States also reformed welfare, again with the chief goal of work. Most toughened work tests and sanctions, but many also improved work incentives and child care. Reform was a national movement, including PRWORA but going much beyond it.


Whether reform succeeded should be judged first by its own goals. We should infer these from what government actually did, not its rhetoric. The most important goal by far was to enforce work. The strongest public grievance against welfare always was, not its cost, but the fact that few adult recipients worked in return for aid (Gilens, 1999). PRWORA’s leading mandate was that states move at least half of their cases into “work activities” by 2002, on pain of cuts in their federal grants. The new child care, health, and EITC benefits were also aimed chiefly at employment.

Second to work in importance was to reduce dependency. This goal was not mandated like employment, but states were allowed to satisfy their work targets through caseload fall. Thus, reduced dependency was valued chiefly as a means of enforcing work, not as an end in itself.

A third priority was to promote marriage. PRWORA’s preamble stresses the importance of marriage, but the law did less to promote it than work. To deter unwed pregnancy, PRWORA required teen-aged mothers to live at home, among other steps. It promised states a financial bonus if they reduced nonmarital births, but it did not penalize them for not doing so. In practice, most states ignored marriage (Goggin Orth, 2002). PRWORA also contains many provisions to improve child support enforcement. Finally, reformers hoped to reduce poverty, but this was to follow from the other provisions and was not an explicit goal of PRWORA.


Reform achieved its two leading goals. Work levels among poor mothers soared, both on and off the rolls. The share of welfare cases meeting the work participation standards under FSA and PRWORA rose from 15 percent to about 33 percent.1

The share of recipients working in unsubsidized jobs while on welfare rose from 8 percent in 1994 to 26 percent in 2001 (U.S. Congress, 2004, p. 7.81). In the population, the share of all poor mothers with children who worked jumped from 44 percent in 1993 to 64 percent in 1999, while the share working full-time year-round rose from 9 to 17 percent. Those figures fell to 54 and 16 percent by 2005, partly due to the 2001 recession.2 Around 71 percent of families leaving aid worked at some time afterward, although fewer worked continuously (Acs, Loprest, Roberts, 2001). Still more dramatically, between 1994 and 2001, the AFDC/TANF caseload plummeted by around 60 percent, from over 5 million cases to 2 million, vastly its largest fall ever.

Welfare reform met its other goals less fully. Unwed pregnancy rates and child support payment did improve, driven in part by reform, but these were slow developments that predated PRWORA. Overall poverty rates fell from 15 percent in 1994, when the caseload started to fall, to 11 percent in 2000, before rising back to 13 percent in 2005. For child poverty, the drop was greater—equivalent figures of 22, 16, and 18 percent. For black children—those most exposed to reform—the figures were 44, 31, and 34 percent, the lowest levels ever recorded (U.S. Department of Commerce, 2006, tables B1, B2). But caseloads fell much more than poverty.

Welfare reform did not harm children’s development, as many had feared. In evaluations of welfare work programs in the 1990s, there were some suggestions that young children do better in school, but teenagers worse, if welfare mothers work. But these effects are small (Morris, Huston, Duncan, Crosby, Bos, 2001).


By liberals’ standards, reform appears less successful. It did not force states to raise cash benefits, which remained low in the South, although there was no “race to the bottom” where states competed to cut benefits, as some had feared.

Nor did reform assure that families leaving welfare had enough to live on. While most leavers’ incomes rose after welfare, most remained poor and some lost income, at least initially (Acs, Loprest, Roberts, 2001; Primus, Rawlings, Larin, Porter, 1999). Reform made work pay more than welfare, and leavers who worked consistently gained income (Danziger, Heflin, Corcoran, Oltmans, Wang, 2002; Loeb Corcoran, 2001). But typically, leavers avoided poverty only if they worked steadily and claimed EITC and Food Stamps. Failure to do those things might well leave them worse off. Conservatives typically thought it fair to make those demands, while many liberals did not.

It did not address the rising inequality that has hit earnings and incomes in recent decades, when most real wage gains have gone to the better-off. But as we note below, some solutions to these problems may emerge from the political effects of reform.


Some think the real credit for reform’s achievements should go to the superb economy the nation enjoyed in the late 1990s—not only low unemployment, but real wage growth, even for the low-skilled. Among reform measures, some think the EITC and child care subsidies achieved more than work requirements.

Studies weight these factors differently, but it is reasonably clear that welfare reform outweighed the economy in driving rolls down, and that work enforcement caused more change than work incentives (Grogger Karoly, 2005). Thus, even in a worse economy than the 1990s, reform would have made progress toward its major goals.


Reform was a triumph, not just for social policy, but for American government. Seldom has a major social reform been instituted so successfully. This was a credit to elected leaders and administrators at all levels.

In the 1960s and 1970s, when welfare first became a national issue, how to reform it was deeply divisive. Liberals and Democrats wanted to liberalize benefits and coverage, while Republicans resisted. In turn, conservatives demanded work requirements, and liberals resisted. The impasse prevented all fundamental change. But over time, these passions cooled. To judge from congressional hearings, welfare reform came to be treated more practically, as a problem to be solved. This more mature style dominated deliberations on FSA, and it remained important even during PRWORA, when partisan division rebounded due to the sharp changes Republicans imposed on the federal role (Mead, 2006).

A concordat emerged where liberals abandoned entitlement, the idea that aid should be given based on need without demanding work, while conservatives abandoned downsizing government. Rather, both sides focused on promoting work.

That concurrence, though clearest at the state level, is visible also in Washington, where PRWORA was attacked by liberals but accompanied by improved support benefits (Mead, 2004a; Winston, 2002). Aside from the cuts in aliens’ benefits, spending on welfare rose, instead of falling. The new welfare system was more demanding than the old, but also more generous. The recent reauthorization of TANF continued this approach.

Administrators implemented the new system unevenly across the states, and not without mishap. Yet their progress was far quicker than past research would have anticipated (Gais, Nathan, Lurie, Kaplan, 2001). In most places, a work test became real “on the ground.” Reform at its most ambitious—in Wisconsin—rebuilt aid entirely around work, and also reinvented the bureaucracy (Mead, 2004a).

Large diversion effects resulted. Poor mothers got a message that they were now expected to work. Millions of them left welfare for jobs, often before they were told to, while millions more went to work without ever seeking aid. Change was much greater than evaluations of past work programs anticipated. It was this political dynamic—not economic incentives—that finally transformed welfare (Mead, 2004a, chap. 9). That American government had this capacity was profoundly encouraging. A similar combination of benefits and demands might help solve other social problems, notably the work problems of low-skilled men.


Politics may also hold the answer to the shortcomings of reform. The income and opportunity problems of mothers who have left welfare are no longer distinct from those of low-wage workers in general. To address them requires no longer welfare reform, but social reform of an older kind. Just as the movement of welfare politics has been away from partisanship, so the movement of a broader social politics should be toward it. Before the 1960s, Democrats and Republicans disputed how much government intervention in the economy was good for workers. The issue was not so much macroeconomic policy in the current sense—tax cuts and budget deficits—as whether a regulated or unfettered marketplace best helped workers get ahead. Democrats wanted more government (and trade union) intervention, Republicans less. That sort of debate was muted after poverty became a major issue in the 1960s.

Leaders focused instead on how to assuage the inner city. Few poor adults worked, weakening their claims on government. More of them had to become employed before the question of how to help them as workers could get back on the agenda. That is just what welfare reform has begun to achieve. By thrusting welfare mothers into jobs, it forced new attention to how to help ordinary working Americans.

On that issue, both parties have a case, but the debate has recently tended left despite a conservative administration. Welfare reform has created more “working poor” and “working families,” and helping them through government is a lot more popular than the old welfare. Aside from spending more on Medicare and schools, the Bush administration accepted higher child care funding in TANF reauthorization, and it is spending more on fathers. A “living wage movement” is pressing local government to force employers to improve pay and benefits for low-skilled workers. Congress faces pressure to raise the minimum wage. Those steps would be unimaginable if 5 million families still lived on welfare without work expectations. So welfare reform, although initiated by conservatives, may finally benefit the left (Mead, 1992).

This new economic politics may be welfare

reform’s greatest achievement. The change is still nascent, however, because today’s poor no longer vote and organize as they did in the 1960s or before. Rather, advocates and the better-off usually speak for them. That has nothing like the same impact as the poor themselves marching on Washington. An elitist politics cannot undo inequality (Mead, 2004b). The political message of reform to the poor was to give up claims on government based on weakness. Rather, make claims based on contribution, above all, by working. Government will do more for you off welfare than it ever did on. That might include a larger welfare state or a smaller one, but you must demand it. The needy must assert themselves, both at work and in politics. Finally, what reform enforced was not work, but citizenship.

This article first appeared in the Journal of Policy Analysis and Management, 26, Spring 2007.


Acs, G., Loprest, P., Roberts, T. (2001). Final synthesis report of findings from ASPOE ‘Leavers’ grants. Washington, DC: Urban Institute.

Danziger, D., Heflin, C. M., Corcoran, M. E., Oltmans, E., Wang, H. (2002). Does it pay to move from welfare to work? Journal of Policy Analysis and Management, 21, 671–692.

Gais, T. L., Nathan, R. P., Lurie, I., Kaplan, T. (2001). Implementation of the Personal Responsibility Act of 1996. In R. Blank and R. Haskins (Eds.), The new world of welfare: An agenda for reauthorization and beyond. Washington, DC: Brookings.

Gilens, M. (1999). Why Americans hate welfare: Race, media, and the politics of antipoverty policy. Chicago: University of Chicago Press.

Goggin. M. L., Orth, D. A. (2002), State and county implementation of the family formation and pregnancy prevention goals of PRWORA. East Lansing, MI: Michigan State University, Department of Political Science.

Grogger, J., Karoly, L. A. (2005). Welfare reform: Effects of a decade of change. Cambridge, MA: Harvard University Press.

Loeb, S., Corcoran, M. (2001). Welfare, work experience, and economic self-sufficiency. Journal of Policy Analysis and Management, 20, 1–20.

Mead, L. M. (1992). The new politics of poverty: The nonworking poor in America. New York: Basic Books.

Mead, L. M. (2004a). Government matters: Welfare reform in Wisconsin. Princeton: Princeton University Press.

Mead, L. M. (2004b). The great passivity. Perspectives on Politics, 2, 6715.

Mead, L. M. (2006). Welfare politics in Congress: Hearings, 4th draft. New York: New York University, Department of Politics.

Morris, P. A., Huston, A. C., Duncan, G. J., Crosby, D. A., Bos, J. (2001). How welfare and work policies affect children: A synthesis of research. New York: Manpower Demonstration Research Corporation.

Primus, W., Rawlings, L., Larin, K., Porter , K. (1999). The initial impacts of welfare reform on the incomes of single-mother families. Washington, DC: Center on Budget and Policy Priorities.

U.S. Congress, House, Committee on Ways and Means (2004). 2004 Green Book: Background material, and data on the programs within the jurisdiction of the Committee on Ways and

Means. Washington, DC: U.S. Government Printing Office.

U.S. Department of Commerce, Bureau of the Census (2006). Income, poverty, and health insurance coverage in the United States: 2005. Washington, DC: U.S. Government Printing Office.

Winston, P. (2002). Welfare policymaking in the States: The devil in devolution. Washington, DC: Georgetown University Press.

  1. Journal of Policy Analysis and Management DOI: 10.1002/pam Published on behalf of the Association for Public Policy Analysis and Management 1 Data from the U.S. Administration for Children and Families. Roughly, FSA required welfare mothers with children aged 3 or more to participate for 20 hours a week. TANF required all welfare mothers to participate at rates that rose from 20 to 30 hours a week by 2000, although states could exempt mothers with children under age 1. 
  2. Data from the U.S. Bureau of the Census, March Current Population Survey for the years after the indicated years. Nor did reform assure that leavers would move up to better jobs over time. It ignored the employment problems of poor men, many of them the fathers of welfare families.