The Personal Responsibility and Work Opportunity Act of 1996 (“welfare reform”) converted the traditional welfare program known as Aid to Families with Dependent Children (AFDC) into a new one called Temporary Aid to Needy Families (TANF).
The reform capped federal spending on TANF at its 1996 level and required states to impose work requirements on recipients. Twenty years later, complaints of a safety net under attack — in retreat, thinning, being shredded — are reaching a shrill crescendo. But they badly misrepresent the broad expansion in the safety net that has actually occurred.
According to the Congressional Budget Office, federal spending on AFDC totaled $21.4 billion in 1996. By 2013, spending on TANF had declined to $17.1 billion. The core of the case for a shrinking safety net turns on a reduction of less than $5 billion over nearly 20 years.
States also spend on welfare but, unlike the federal government, their spending was not capped by reform and in fact rose in nominal terms. The decline in state spending, in real terms, was only from $18.6 billion to $15.4 billion. Taken together, total federal and state spending declined from $40 billion to $32.6 billion, or less than 20 percent.
Because the immediate objective of welfare reform was to move families off welfare and into the workforce, the number of program recipients dropped more precipitously — from 12.6 million recipients to 4 million, or 68 percent. The share of TANF spending dedicated directly to cash payouts therefore fell as well, from $33.8 billion to $9.7 billion, or 71 percent.
But spending on other social services for many of the same families therefore increased substantially. Work supports like child care, transportation, work preparation and subsidies in the form of tax credits increased fivefold — from less than $2 billion to more than $10 billion per year. Spending on other social services, including the promotion of two-parent families, increased as well.
Critics of the reform lament both the absolute reduction in resources and the substitution of other services for cash. They also highlight situations in which states have used federal TANF funds for their own pre-existing anti-poverty programs to free up their own state funds, representing in effect a further reduction in total anti-poverty spending. Yet even taken together, these changes are barely perceptible against the backdrop of rapid growth in other programs — often targeted toward the same population.
In 1995, the year prior to welfare reform, federal and state spending on anti-poverty programs totaled $540 billion. Of that total, Medicaid accounted for $219 billion, SNAP (food stamps) and other nutrition programs for $51 billion, housing for $42 billion, and SSI (disability) for $36 billion. The $47 billion spent on family services programs (primarily AFDC) accounted for less than 10 percent of the total. Even had all other program spending been held constant for the subsequent 20 years, the decline in family services spending to $37 billion would have represented a total safety net cut of 2 percent — less than 0.1 percent per year.
But other safety net spending did not stop growing. Medicaid alone added $349 billion to its budget, while the creation of CHIP (Children’s Health Insurance Program) added an additional $13 billion — CHIP alone more than offset the total TANF cut. Total enrollment across Medicaid and CHIP increased from 33 million in 1995 to 72 million at the end of 2015. Spending on SNAP climbed by $37 billion as its enrollment increased from 27 million to 46 million.
Even cash benefits increased substantially. Total SSI spending (primarily a cash benefit program) increased by $17 billion, almost fully offsetting the TANF decline. Rapid growth in other forms of cash support — the Earned Income and Child Tax Credits — increased by $58 billion. Together, these increases in cash support to low-income households exceeded the decrease in TANF cash support by a factor of three.
All of this also ignores the dramatic increases in earned income now flowing toward households that might formerly have received AFDC benefits, thanks to the strong incentives TANF created to enter the workforce. As Scott Winship noted recently in a related Manhattan Institute Reality Check, the employment rate for never-married mothers has risen dramatically since welfare reform, while the share living below the poverty line has declined.
In short, the $1 trillion safety net of 2015 was 102 percent larger than in 1995, but only 1 percent smaller than had TANF spending not declined. Claims that this evolution represents a thinning, shredding or retreat of the safety net, imply the safety net of 1995 was better than that of 2015 and the 1 percent reduction has hurt more than the 102 percent increase has helped. That seems unlikely.
A more reasonable concern would be that AFDC provided a particular benefit to a particular subset of welfare recipients who may be less well served by the safety net’s reformed configuration. That assertion, unlike those accusing Americans in general or those on the Right in particular of abandoning the poor, would invite a reasonable discussion about reforms to further strengthen the safety net.
Decrying any reduction in any anti-poverty program as disastrous for the poor only frustrates the prospects for effective reform. It creates a one-way ratchet where resources once dedicated to a particular purpose can never be moved, locking inefficient allocations into place and discouraging budget-conscious policymakers from adding spending on things that work.
The recent drama surrounding the expiration of SNAP eligibility for as many as 1 million childless adults in April 2016 illustrates the challenge well. “This thinning (of the safety net) goes beyond welfare,” wrote Alana Samuels at The Atlantic. “The numbers of people receiving food stamps will drop, and thousands more people won’t be able to eat, or survive. But to the policymakers who look for a shrinking welfare program, the changes will be considered a success.”
At The Week, Emily Hauser described the “abrupt increase in the misery of hundreds of thousands of Americans” in a column titled “The grotesque injustice of starving 1 million unemployed Americans.”
Yet those losing food stamps had only been eligible at all because of waivers triggered by the high unemployment rates of recent years. Further, the removal of 1 million recipients compares to an increase of 20 million recipients since 2005. The result of the “cut,” or rather the return to the pre-waiver standard for normal economic times, left 19 million more Americans receiving food stamps than had received them 10 years earlier.
That additional support is ignored by critics who complain the safety net was weakening over the period. But as soon as that same support is even marginally contracted, the same critics declare it to be of the utmost importance. New benefits count for nothing on the way up, but everything on the way down.
The American safety net of the early 21st century is the most generous it has ever been, providing more benefits to more people than ever before. It can surely do better. But the lesson of recent history, implicit in the strongest criticisms, is that how money is spent matters more than how much. That lesson, not demands to write ever-larger checks, should guide future reforms.