EVIDENCE FOR EFFECTIVENESS OF WELFARE REFORM STRONG

Date: Saturday, December 18, 2004

Child poverty dropped sharply after welfare reform went into effect, but that's a long way from proof that welfare reform deserves the credit.

Lots of other things were going on in the late 1990s that might have contributed to the same welcome result, including rising wages, a booming stock market and plentiful jobs.

In a study for the Manhattan Institute, economists June O'Neill and Sanders Korenman, both of Baruch College at the City University of New York, try to tease out the various factors that are likely to have influenced child poverty rates. They estimate that reform may be responsible for as much as half of the decline among black and Hispanic households headed by single mothers, two groups that had high rates both of welfare participation and child poverty (since being on welfare pretty much guaranteed being poor).

You might wonder whether the abrupt end of the '90s boom, and the slow recovery in jobs, undid the gains of the good years. The answer is mostly no. The study is based on Census Bureau figures through 2002, the most recent available. In a last-minute footnote about preliminary 2003 data, the authors say that child poverty increased from 16.7 percent to 17.6 percent from 2002 to 2003, but that's still three percentage points lower than in 1995, a pre-reform year with lower unemployment than 2003. Since job numbers improved during 2004, the authors think poverty is unlikely to have increased in 2004 and may have started back down.

Factors the authors identify as influential in the decline of child poverty include "increased work participation among single mothers; a rise in the level of wages; increases in parental education; declining family size; and, for some groups, a rise in the proportion of children living with two parents."

Clearly these factors are related in many ways, not all of them direct or obvious. Many people meet their spouses at work, so a single mother who stays home with her children, especially if most of her neighbors are like her, may have fewer opportunities to meet a future husband than a single woman with children who takes a job. And in the absence of long-term welfare, she might also have more incentive to accept a marriage proposal if she gets one.

People interested in the mathematical analysis can go look at the paper

(at www.manhattan-institute.org/html/cr_ 44.htm online). I have room only to suggest a couple of the reasons it is intuitively plausible to attribute declining child poverty at least in part to the effects of welfare reform.

One is that the decline was sharpest precisely in the groups where welfare was most prevalent, that is, black and Hispanic children of single mothers. From 1993 to 2002, the poverty rate, calculated from household income, for white, non-Hispanic children declined from 12 percent to 8 percent. For Hispanic children, the drop was from 39 percent to 26 percent, and for black children from 44 percent to 30 percent. (Higher rates for Hispanics are largely associated with less education, and for blacks with a smaller fraction of two-parent families, relative to whites.) Also, the number of children living with two parents, who are much less likely to be poor, increased among black and Hispanic children but not among whites.

Another factor suggesting that welfare reform is partly responsible for the decline, and not merely associated with it by chance, is that it reverses a trend of climbing child poverty rates that prevailed for a quarter of a century.

From 1959 to 1969, the authors note, the child poverty rate was cut by nearly half, from 27 percent to 14 percent. But so was the poverty rate for people 18 to 64, which is lower. Starting in 1969, though, the gap between children and adults began to widen. The poverty rate went up during recessions, and down in recoveries, but the rate for children went up more and down less, getting as high as 23 percent in 1993. Since then, the gap has narrowed significantly.

Mothers of black and Hispanic children were more likely to be working in 2002 than in 1995, by about 10 percentage points. As a result, the proportion of children living with a working parent rose to 77 percent for black children and 85 percent for Hispanic children. For whites, it was 92 percent.

Why is this timely right now? Because Congress will be debating the reauthorization of the 1996 welfare reform act. The evidence indicates, O'Neill and Korenman conclude, that reform was a significant improvement over its predecessor, with no evidence of harm to children's well-being and considerable evidence of good effects.

The current system, they say, "has struck an appropriate balance between providing a helping hand to the needy and improving incentives for self-sufficiency" and the act "appears to be one of the rare policy prescriptions that actually exceeded expectations."

They recommend staying on course. "Our (bipartisan) advice to policymakers, then, is this: If it's working, don't fix it." Good advice.