Commentary | Inequality and Poverty

Holes in welfare reform

Opinion pieces and speeches by EPI staff and associates.


Holes in welfare reform

By  Jared Bernstein and Mark Greenberg

In a presidential election dominated by a few big issues, two that got very little attention were poverty and welfare reform. With a new term, though, comes the opportunity to  take a fresh approach  to issues. To start, the administration should consider what’s happened to welfare reform since the economy entered a slowdown in 2001.

Routinely, the federal government announces new declines in the nation’s welfare caseload, treating each drop as further evidence of success. But for the past three years the caseload and the nation’s poverty rate have been moving in opposite directions: Caseloads have fallen as  poverty increased. At the same time that the welfare caseload  declined, the nation’s poverty rate grew from 11.3 percent in 2000 to 12.5 percent in 2003, with over 4 million more people in poverty, including 1.3 million more children. The continuing drop in welfare rolls wasn’t a result of more families achieving self-sufficiency.

Why have welfare caseloads fallen while poverty grows? Our analysis of poverty and employment data suggests that we’re now seeing the other side of welfare reform, a system that generally performed well during the tightest labor market in 30 years but has been far less effective amid the slack labor market that has prevailed since the 2001 recession.

In the late 1990s employment among single-parent families surged for a number of reasons: the strong economy, with high demand for low-wage workers; welfare reform; expanded child-care subsidies for low earners; an increased minimum wage; and an expanded earned-income tax credit (EITC), which increased income for low-wage workers. Between 1994 and 2000 the share of low-income single moms at work grew 20 percentage points, from about 40 percent to 60 percent (the average for all  people went up by 1.9 points). Single mothers increased their annual hours at work by 460 hours, almost three full-time months. Their annual earnings more than doubled, as did their EITC,  pushing their poverty rates down to the lowest levels on record.

Since 2000 these trends have all been reversed. Low-income single mothers are still working at much higher levels than they were before welfare reform, but their job and hours losses have led to lower earnings and EITC receipts, driving their drops in income and increases in poverty. Unemployment compensation has helped a little but not nearly enough to offset the losses in earnings.

Even though the economic recovery is officially a  3-year-old, this toddler is still crawling. True, we’ve been adding jobs lately, but not enough to tighten up the job market. When the recovery began in November 2001, the unemployment rate was 5.6 percent. As of this October, the rate was essentially unchanged at 5.5 percent. That may not sound too high, but remember that it wasn ‘t until the jobless rate headed down toward 4 percent in the late 1990s that the employment and earnings gains enumerated above began to materialize. By the second half of last year, the persistently slack job market led to slower wage growth, and the inflation-adjusted wages of many workers began to fall.

Why have welfare caseloads kept falling despite the weak labor market? Welfare reform involved a two-pronged approach: expanding supports for low-earning families while making it harder for families that aren’t working to get help. Even when the economy was creating large numbers of new jobs, some families were hurt by this strategy, but the overall approach proved a highly potent anti-poverty combination. Recent experience suggests that our current policies may be too pro-cyclical, boosting the economic fortunes of the vulnerable in boom times, but failing to help when jobs are less plentiful.

What should the administration do? First, when employment is falling and poverty is rising, it should stop treating a drop in welfare caseloads as cause for congratulations and start treating it as a sign that the system is failing to respond to increased needs. Second, when the welfare law comes up for reauthorization  again next year, the administration should promote changes that reward states for success in promoting employment, not just cutting caseloads. Third, the administration should ensure that Congress maintains funding for the program, because states under fiscal stress will be least likely to extend help to families losing jobs.

Changes outside of welfare could also make a big difference. If the unemployment compensation system reached more people who had lost their jobs, fewer of their families would need welfare. Moving back to full employment and truly tight labor markets would lead to sharp declines in poverty like those in the 1990s. In the meantime, we should recognize that falling welfare caseloads in tandem with rising poverty are a signal that the safety net is failing to do its job.

Jared Bernstein is a senior economist at the Economic Policy Institute. Mark Greenberg is director of policy for the Center for Law and Social Policy.



See related work on Wages, Incomes, and Wealth | Inequality and Poverty

See more work by Jared Bernstein