*This blog was co-authored by Barbara Seligman, consultant to CGD
In his recently released book, Fatal Misconceptions: The Struggle to Control World Population, Matthew Connelly, an associate professor of history at Columbia, tells a cautionary tale about the arrogance that marked the origins and early history of international 'population control' programs. Of the many topics on which we could post, we focus here on his conclusions about the role of organized family planning programs in helping to reduce fertility rates. Professor Connelly repeats the assertion that family planning efforts explain "less than five percent of fertility levels in developing countries" (p.338). The reader might unwittingly conclude that there is widespread agreement regarding the "five percent" attribution, which is certainly not the case. Here we take a closer look at the analysis where this attribution first appeared, and at a more recent study using "gold standard" data from a controlled experiment that suggests the benefits of organized family planning efforts are significant and go well beyond 'births averted.'
1994, the year of the landmark 4th International Conference on Population and Development held in Cairo, marked a shift away from a narrow view of curbing rapid population growth through organized family planning efforts to a more holistic reproductive health approach. That year economist Lant Pritchett, then with the World Bank and now at Harvard and a Senior non-resident Fellow at the Center, published results of a cross-national study in which he concluded that only five to 10 percent of differences in fertility levels across countries could be attributed to family planning programs or lack of access to contraceptives. Pritchett attributed the vast majority of the difference in fertility levels to demand factors, which he maintained were not affected by contraceptive supply. In repeating this assertion without further discussion Connelly may lead some readers to conclude that in spite of billions of dollars of cumulative investment (and the indignities and physical harm it sometimes caused), international family planning assistance didn't make a difference in slowing fertility and curbing population growth rates.
In his 1994 study Pritchett uses cross-national regression analyses to test the hypothesis that demand for smaller family size - not supply of contraceptives - explains variations in fertility levels across countries. He also reviews evidence from the most important family planning controlled experiment available, the Family Planning and Health Services Project (FPHSP) in Matlab, Bangladesh which was designed to effectively eliminate the costs of using contraception. These costs included the time and money spent traveling to a clinic or other source for contraceptive supplies or advice as well as the costs of social opprobrium of using contraception, which were significant in this conservative district where many women practiced purdah and family planning workers were sometimes stoned. FPHSPâ€™s "doorstep delivery" program ensured the women of Matlab uninterrupted and reliable access to their contraceptive method of choice even during the worst weeks of the rainy season. Pritchett acknowledges that Matlab proves that family planning can have a role in determining fertility. But, he argues, 'fertility is not invariant with respect to the cost of contraception, just that it is sufficiently inelastic to make cost variations an unlikely source for explaining or causing major demographic changes.' Generally speaking, Pritchett contends contraceptive cost should be a minor consideration affecting a person's decision to have another child. To underscore the relative insignificance of contraceptive cost relative to the cost of raising a child, he asks his readers, presumed to be mostly from industrialized countries, 'How many additional automobiles would people buy if motor oil were free?' (p.25)
Since Pritchett published his article, fellow economists Shareen Joshi (University of Chicago) and T. Paul Schultz (Yale University) have completed an exhaustive evaluation of fertility and other health and social consequences associated with the Matlab program. The authors found fertility levels in the program area were 15 percent lower and that fertility was at least one child lower for women between the ages of 30 and 55 (in 1996) compared to the comparison area. Contraceptive use increased and fertility declined more in the program area than in the comparison area where government family planning efforts were underway and fertility declined from more than six to less than four children per woman. Joshi and Schultz' results indicate that reducing the cost of contraceptives, particularly for poorer women who, for a variety of reasons, may not be able to buy contraceptives or for whom the costs of contracepting are especially high, appears to make a significant difference for voluntary reduction of fertility. While Dr. Pritchett acknowledged that reducing the cost of contraception and contracepting in the situation of Matlab, Bangladesh did increase contraceptive use and reduce fertility, he contends that the very high costs of making contracepting 'cheaper than free' as was done under the program mean that it cannot be replicated at the national level. Since 1994 when Pritchett published his article, the FPHPS effort has been extended to the national level. The costs of the 'door-step' model did not prevent this kind of scaling up of the program as Pritchett predicted would be the case.
Joshi and Schultz also look at benefits from the FPHSP that go beyond 'births averted.' They find the program is associated with important non-fertility benefits - not considered by Dr. Pritchett - including improvements to women's health, their economic productivity outside their household and their household assets not present in the comparison area. Professor Connolly's book might have also considered the value of some of these non-fertility benefits associated with organized family planning efforts, especially in settings like Bangladesh where they served as catalysts for important social changes. The realization that fertility was something over which a woman might reliably exercise some control would seem to be a vital first step in encouraging investments in daughters for whom life might hold the possibility of non-reproductive roles. Other side benefits in Bangladesh of the organized family planning programs included that women counselors be hired (eventually more than 25,000 women would serve as Family Welfare Advisers) with civil service benefits otherwise enjoyed mostly by men, and that unrelated men and women worked together in a professional environment.
So, did organized family planning programs do more harm than good? Professor Connelly implies that they did, and advises the public health crusaders of today not lose sight of the important lessons from the family planning past. His wise words and lessons from the past have special significance we believe for the public health 'battlefields' of the 21st century, especially HIV/AIDS. However, we urge Professor Connolly to pay more attention to the evidence regarding some of the good that family planning programs brought to the lives of poor women. In using history to avoid the mistakes of the past in the future, one must recognize the good as well as the bad.
Professor Connelly will appear at the Woodrow Wilson International Center on Tuesday, April 22 from 3:00-5:00, along with Mr. Bob Engleman, Vice President for Programs at the Worldwatch Institute.
Mary Arends-Kuenning, 2002, "Reconsidering the Doorstep-delivery System in the Bangladesh Family Planning Program," Studies in Family Planning, vol. 33, no.1. March.
Joshi, Shareen and T.Paul Schultz, 2007. Family Planning as an Investment in Development: Evaluation of a Programâ€™s Consequences in Matlab, Bangladesh. Yale University Economic Growth Center, Discussion Paper No.951. February.