UW News

June 26, 2003

Social program benefits ‘spill over,’ study shows

Social-welfare programs may help many more people than previously thought, UW research indicates.

One maternal and child health program in Bangladesh not only improved the health of the mothers and infants it targeted, but a UW economist found that it also made the grandmothers and great-grandmothers living with them healthier.

Yet this kind of “spillover benefit” is rarely taken into account when judging programs’ success, said Anoshua Chaudhuri, who will present the findings June 27 to the Econometric Society meeting at Northwestern University in Evanston, Ill.

“I really do believe we’ve been missing something in standard cost-benefit analyses,” said Chaudhuri, who performed the research for her UW doctoral dissertation in economics. “You might even call it a ‘trickle down effect.’ ”

Chaudhuri plans look for similar side benefits from U.S. social programs such as the Women, Infants and Children (WIC) program and Head Start. Though Americans are less likely than rural Bangladeshis to have multiple generations living under the same roof, American children may gain spillover benefits from programs targeted to their siblings.

But how does this spillover effect work?

The goods or services provided by a social program can free up family resources to spend on needs — such as food or education — of other household members, Chaudhuri said. Plus, a so-called positive contagion effect can benefit everyone in the house when a program prompts changes that make for a healthier and more hygienic environment.

Such was the theory. Chaudhuri found a place to test it in a cholera-plagued region of southern Bangladesh where the International Center for Diarrhoeal Disease Research launched an intensive maternal-health program in 1982 (with an equally impoverished area nearby serving as a control population). The program provides immunizations, vitamins, prenatal care and health and hygiene instruction to all treatment-area women of childbearing age and children below the age of 5.

A survey of 4,364 households — roughly half in the target area and half in the control area — was available for Chaudhuri to analyze.

Not only did children in the area served by the program grow a half-centimeter taller on average than their non-served peers, but compared to the control area, 20 percent more women aged 60 and above in the targeted households were in the healthy range of the Body Mass Index, a common indicator of adult health in the developing world. The maternal-health program did not exist when those older women were in their childbearing years.

Chaudhuri said her origins in a region of India bordering Bangladesh helped her to understand and interpret the findings. But the theoretical framework and microeconomic evaluation methods set out in her study, she said, could be readily used to evaluate the spillover effects of programs in a variety of other settings.

“This is a novel contribution,” said Elaina Rose, a UW associate professor of economics and one of Chaudhuri’s dissertation advisers. “This shows that cost-benefit analyses of programs which don’t account for spillover benefits may understate the overall benefits considerably.”

The conference paper is available online at http://students.washington.edu/anoshuac  (click on “research.”)