Twenty years ago this week, President Bill Clinton signed into law a historic welfare reform bill that transformed anti-poverty policy. Rather than being entitled to receive cash benefits from the federal government every month, beneficiaries would be expected to eventually work, and time limits were imposed for receipt of benefits. States were given fixed pots of money to serve poor, mostly female-headed, families with children, along with broad discretion to run their programs as they wished.
It has become an article of faith among liberals that the legislation failed and has left the children of single parents worse off.
“It turns out,” Sen. Bernie Sanders tweeted in April, “that the welfare reform bill has been an absolute disaster.” This belief is so widespread on the left that Hillary Clinton felt it necessary to distance herself from her past support for the legislation. An influential book last year by sociologists Kathryn Edin and Luke Shaefer claiming to show that welfare reform dramatically increased the number of families getting by on just $2 a day per person solidified liberal consensus.
But as I show in a new study, child poverty today is at an all-time low, and practically no households in the United States are managing on $2 a day per person. How is it that these facts have eluded welfare reform’s critics? One problem is that our official poverty statistics ignore the parts of the safety net that have expanded since 1996. While federal benefits taking the form of cash — like welfare and unemployment insurance — are counted as income in assessing whether a family is above or below the poverty line, noncash benefits, like food stamps, housing subsidies and Medicaid, are not.
A second problem: Official estimates rely on an annual adjustment in the poverty line that raises it faster than the cost-of-living increases. Nor does the official measure account for the savings cohabiting couples — no less than married couples — realize by living together and pooling expenses. Finally, because income is grossly underestimated among poor households participating in the survey from which official poverty figures are drawn, the deeper the level of poverty considered, the less reliable are the results.
The official figures indicate that child poverty was higher in 2012 than in 1996. After improving the poverty data to account for their shortcomings, rather than rising from 20.5% of children to 21.8%, the child poverty rate falls from 9.5% to 5.4%. Meanwhile, the share of children living under half the poverty line is essentially the same in both years — about 1 in 100 children. The share living under $2 a day per person? One in one thousand in both 1996 and 2012.
None of this means that welfare reform cannot be improved upon. Some states have made it overly difficult for eligible families to get welfare benefits. Requiring states to provide cash benefits to some fraction of children in deep poverty would be one way to remedy this issue. But the idea that the facts indicate that we need to roll back reform or institute a massive public jobs program is wholly unwarranted.
While welfare reform succeeded at increasing work and reducing poverty, it was less successful reversing the tide of out-of-wedlock childbearing. One lesson from reform is that big, clear financial incentives may be more effective than government programs that try to change people’s behavior. A way to apply that lesson to nonmarital childbearing would be to expand the Child Tax Credit — but only for married couples. That would leave single parents no worse off, but encourage delayed, planned and marital childbearing.
Another solution would be to end the marriage penalty within the Earned Income Tax Credit. Today, one or both members of an unmarried couple may qualify for the credit individually but see it reduced or eliminated if the couple marries, combining their incomes. Basing the EITC on individual earnings would eliminate this penalty, and, by expanding its generosity to childless men, make them more attractive marriage partners.
Far from rolling back reforms, other lessons of the welfare overhaul should be extended to parts of the safety net that continue to embody perverse incentives, discouraging work and independence. Federal disability programs help millions of adults with serious work limitations and impaired children, but they also support a growing population of able-bodied adults and children with questionable diagnoses. Food stamps and housing programs have minimal to no work requirements or time limits.
We ought to build into these programs provisions to discourage long-term dependence, generous exemptions from work requirements and time limits to protect families with the greatest challenges — and financial support for the working poor to ensure that work leaves people better off.
Welfare reform was the most successful anti-poverty legislation since the national expansion of food stamps in 1974. History will regard it as a model, not a mistake.
Winship is the Walter B. Wriston Fellow at the Manhattan Institute and author of the report “Poverty After Welfare Reform.”
