Why Clinton's Workfare Won't Work

President Clinton's plan to "end welfare as we know it" goes over big with middle-class voters caught between rising taxes and falling real incomes. His most controversial idea, a two-year limit on welfare benefits, is a particularly effective way for a Democrat to score points with conservatives.

Unfortunately, however, the concept is unrealistic and unworkable. The plan would rely on job training to get nearly a third of welfare mothers off the dole. Studies show that such training does help some recipients get work but not enough to achieve the President's goal. And forcing more than a million low-skilled women to seek jobs, as the plan implies, would wreak havoc at the bottom of the labor market. A more practical approach would be to whittle down welfare rolls by investing as much as possible in training--while encouraging more recipients to work by raising the roughly $700 a month they can earn before they lose benefits in most states.

MORE SPENDING? This isn't very splashy, of course--and it would mean ditching the two-year cutoff. But politics aside, there is little reason for welfare to provoke so much worry in the first place. Although Washington spends about $14 billion a year on welfare, that's only 1% of the federal budget. And that's down from 1.4% in 1970, largely because benefits have lagged behind inflation (charts). "The Administration's plan wouldn't be a bad idea in the best of all possible worlds, where jobs and money [for training] were plentiful," says Heidi I. Hartmann, director of the Institute for Women's Policy Research, a Washington-based group that has studied welfare. "But we don't have that."

The time-limit idea was proposed in 1988 by prominent Harvard University public-policy professor David T. Ellwood. He's now an official at the Health & Human Services Dept., where he co-chairs the Administration's welfare planning group. Ellwood suggested combining time limits with extensive training for welfare mothers, plus child-care and health-care assistance once they landed a job. He also wanted public jobs for those who couldn't find jobs through these so-called workfare programs. Either way, a recipient's benefits would have ended after two years.

The problem is that training can do only so much. For instance, since 1985, California has run one of the country's first large-scale workfare programs. It slashes welfare grants for recipients who don't participate in state-provided education, training, and job searches. These efforts helped participating welfare mothers earn 24% more over two years than mothers who remained on regular welfare, according to a 1993 study by the Manpower Demonstration Research Corp. (MDRC), a nonprofit New York-based research group. However, that amounted to a mere $519 a year--hardly enough to launch anyone out of welfare. As a result, the program cut welfare payments by only 7% a year. "Although the results are encouraging, you won't eliminate welfare this way," says MDRC President Judith M. Gueron.

Of course, the Administration could apply the two-year cutoff anyway. But government spending might actually increase. About 30% of the 5 million adults on welfare leave the rolls before two years, studies show. Two million more would be excluded from workfare because their children are too young. So imposing a cutoff would throw 1.5 million people into the labor market.

The result: The unemployment rate for black and Hispanic women, who account for 55% of all welfare mothers, might double, to 20%. Wages for both groups would plummet, knocking more families into poverty and into other programs such as food stamps, Medicaid, and unemployment insurance. Indeed, all 20 million workers earning $6.50 an hour or less would be undercut. "If you put that many people into the market, the equilibrium wage for low-wage workers would fall," comments David Neumark, a University of Pennsylvania economist who studies the minimum wage.

TRUE REASON. Creating make-work government jobs would ameliorate some of these problems. But 1.5 million minimum-wage jobs would cost at least $15 billion a year. That isn't politically feasible. Even workfare for that many people would cost at least $5 billion more a year. The Administration recently suggested a tax on welfare grants and food stamps to raise funds. But a meaningful tax on families getting $375 a month, the average welfare grant, could cause many recipients to lose their housing. "The real reason we have welfare is that it's cheaper to have poor mothers take care of their own children than to get them job-ready, create jobs for them, and pay someone else to look after their kids," says Hartmann.

Clinton should pursue the core of his plan, which is workfare. It makes sense to spend $5 billion or so a year on training and child-care assistance, backed up by threats to cut benefits of nonparticipants. But unless he can find an extra $15 billion, his pledge to end welfare does little more than stir up ire against society's poorest members.

Before it's here, it's on the Bloomberg Terminal.