Welfare reform replaced Aid to Families with Dependent Children (AFDC) with Temporary Assistance for Needy Families (TANF) 30 years ago this month. At the time, opponents warned that child poverty would surge as time limits and work requirements curbed reliance on welfare benefits. In reality, as noted by our colleague, Matt Weidinger, falling dependency was accompanied by surging employment among single mothers and declining child poverty.
However, protecting these gains proved challenging in the face of progressives’ denial of welfare reform’s success. They created a revisionist history of the post-AFDC era and promoted the view that providing benefits with no strings attached, as AFDC largely did, is what’s best for low-income families. While conservatives have largely fended off these efforts, the events of the past 10 years show that preserving the gains of welfare reform requires vigilance and responsiveness.
As Weidinger describes, the early evidence of welfare reform’s success was so strong a 2002 New York Times editorial called it “obvious.” The welfare reform debate quieted down for much of the next decade-and-a-half, with Congress never making more than modest changes to TANF.
However, in the wake of the Great Recession, voices on the left increasingly claimed that welfare reform had been harmful. While some families had been able to adjust to the new work-oriented safety net, a sizable minority fell into “extreme poverty.” The solution to this problem was for the United States to enact a child allowance — universal per-child benefits (perhaps becoming less generous for upper-income families) that didn’t depend on whether parents worked or not. In other words, policymakers should provide unconditional cash aid to low-income families without a work requirement attached, just as AFDC had done.
The foundational text for these advocates, 2015’s $2 a Day: Living on Almost Nothing in America, by Kathryn Edin and Luke Shaefer, made three bold claims. First, 1.5 million American households were living on $2 a day per person. Second, this figure had risen steadily since the mid-1990s. And third, that increase was caused by welfare reform.
Late the same year, progressive Representatives Barbara Lee and Lucille Roybal-Allard successfully inserted language in an appropriations bill directing President Barack Obama’s Department of Health and Human Services to request a National Academy of Sciences (NAS) report on reducing child poverty. The NAS panel began its work on the report in 2017, a year that also saw Democrats in the House and Senate sponsor child allowance bills.
When the NAS report was issued in 2019, it found that converting the existing child tax credit (CTC) to a child allowance would do more to reduce child poverty than any of the other options the panel considered. (Three of the panel members had joined Edin and Shaefer as authors of a 2018 paper advocating a child allowance.) The NAS report estimated that the reform would raise up to four million children out of poverty.
Key to that figure, however, was a little-noticed assumption in the statistical modeling — that few families would respond to no-strings-attached benefits by working less. Proponents heralded the report’s findings. The problem with AFDC, they argued, was that it had phased out benefits as beneficiaries increased their earnings, creating a work disincentive that trapped them on AFDC. A child allowance, being universal for all but the richest families, featured no such disincentive, since higher earnings did not reduce benefits received.
The culmination of this multipronged campaign arrived in early 2021, when President Joe Biden signed into law a temporary conversion of the CTC to a child allowance. Democrats’ attempts to make the policy permanent later that year failed, but just barely. As the fight over the policy neared its climax, more than 400 economists signed a letter advocating the child allowance be made permanent. That letter, co-organized by one of the NAS panel members, cited the panel’s conclusion that a child allowance would not reduce work by much.
However, the campaign to reverse the harms supposedly done by welfare reform rested on flawed evidence.
First, the Edin and Shaefer extreme poverty research was quickly debunked, first by one of us (Winship) and then by Bruce Meyer and his colleagues. Winship found that essentially no children resided in families that lived on $2 a day per person in the U.S. after accounting for various measurement problems. If one wanted to naïvely believe the estimates that produced an increase in extreme child poverty after the mid-1990s, one had to believe that extreme poverty had risen not just among single-mother families, but among groups unaffected by welfare reform, such as the elderly and college-educated married couples. The fact that the measured “rise” began in the 1970s also exonerated welfare reform.
Meyer found evidence of extreme poverty only among a very small number of single childless adults, a group unaffected by welfare reform. Those families naively deemed extremely poor by error-prone survey data indicated they experienced no more hardship than other poor people did, reinforcing the conclusion that they only looked extremely poor in the data.
For its part, the NAS panel’s assumption that moving to a child allowance would not substantially disincentivize work was based on a basic misunderstanding of how the policy change would alter incentives. The panel reasoned that since there was no gradual phase-in of a child allowance as people went from not working to having higher earnings, there was little payoff to choosing one over the other. However, what matters for evaluating the impact of a policy change is how the work incentive changes. The CTC includes a phase-in that incentivizes moving from nonwork to work. Taking away that phase-in would be expected to reduce work for the same reason that AFDC’s phase-out discouraged it.
This issue was raised by one of us (Corinth) and Meyer in a working paper released a month after the economist letter. Corinth, Meyer, and their colleagues drew on estimates from the academic literature of how sensitive employment decisions are to changes in the payoff to working. They found that nearly 1.5 million parents would stop working due to the replacement of the CTC with a child allowance, reducing the policy’s impact on child poverty by a third. They and others also pointed out that the effects on marriage, fertility, and living arrangements could also work against poverty reduction.
Corinth and Meyer were subjected to a wave of criticism as policymakers considered the permanent child allowance. However, their contention that the NAS panel had made an error and that moving to a child allowance would significantly reduce work was vindicated by a follow-up report this year from another NAS panel. Unlike the previous one, the new NAS report concluded that parents would strongly respond to the change in work incentives, affirming the Corinth–Meyer argument. The earlier 2019 report that was so influential had erred in its modeling.
Unfortunately, these weaknesses in the evidence have done little to diminish the campaign’s influence on policymakers, many of whom remain interested in a child allowance. Cities across the country have experimented with temporarily providing a guaranteed income to low-income families, generating interest in and momentum toward a universal basic income. If enacted in lieu of a work-oriented safety net, that would undo many of the positive effects welfare reform had on poor families.
The lesson of welfare reform was that a work-promoting safety net is the most effective tool we have to reduce both dependency and poverty. It would be unfortunate if the only way policymakers relearn this lesson is by embracing policies that undo welfare reform and witnessing the consequences.
Editor’s note: This is the second installment in a three-part series on welfare reform. Read the first here.
Scott Winship is a senior fellow and director of the Center on Opportunity and Social Mobility at the American Enterprise Institute. Kevin Corinth is the Daniel C. Searle Chair and a senior fellow at the American Enterprise Institute.
