Population Services International

T. Paul Schultz, Ph.D.

MANDY MCANALLY: In your view, have development agencies and nongovernmental organizations (NGOs) become more sophisticated in their use of research and data in the last 20 years?

T. PAUL SCHULTZ: I think research focused on health and education and social support policies, such as the conditional cash transfer programs, have clearly advanced in the last decade since the Mexican Progresa – Oportunidades program was initiated and scientifically evaluated. Some NGOs have recognized the value of evaluating their programs, while others have assumed they have the best solution to their problem and followed their intuition without evaluation. Many international and national bilateral programs have strategically set their policies in accord with political pressures from their constituencies, but some have begun to incorporate plans to evaluate their initiatives, at least in terms of short-run consequences.

It may take the perspective of an NIH [National Institutes of Health] to undertake what would be valuable in the form of long-run panel surveys to evaluate the consequences of randomized social experiments in family planning, health and female education.

MM: What is your view on market-driven solutions to family planning riddles, such as social marketing?

PS: My expectation is that much of the effect of family planning on fertility and family welfare occurs because it improves knowledge about this evolving technology and thereby informs women (and men) of these options and reduces their fears of side effects or stigma associated with use. The actual monetary cost and local access to birth control supplies still emerge from several studies as factors affecting demand for birth control and perhaps fertility. But once a population learns to use modern birth control methods as in Indonesia in 1996, the subsequent financial crisis, which increased the public prices for birth control substantially, did shift users to the private sector, but had modest if any effects on birth rates. And I would expect even smaller effects on completed family size.

The challenge with social marketing, which I interpret you to mean public subsidies for private distribution of birth control, is that the effects may today be modest in many low-income countries. Demand for birth control may not be very responsive to monetary price in a poor middle class. Certainly, there must be exceptions where demand for birth control and hence fertility is responsive to price (elastic), and those are settings where public subsidies and social marketing could be justified and should be concentrated.

MM: About 10 years ago, you wrote eloquently about how the paucity of data made setting priorities in health and education in Africa daunting. Are we now better-equipped to make policy and evaluate it, or less?

PS: I think the data for Africa in many respects has improved enormously in the last decade or two. But evaluating the important consequences of health and family planning programs is difficult and makes the debates about returns to schooling seem less challenging. Nonetheless, returns to schooling are still debated vigorously because we do not have the time series changes in educational opportunities occurring in an independent way that facilitates our evaluation of their consequences on the population that must then have completed their schooling and established their record of working for wages.

You can see the challenge with evaluating health is even more of a barrier, where the inputs to health are self-selected by parents and allocated to children of different health endowments (e.g. the sick, bright or more promising gender). And we are ultimately interested from an economic perspective in the long-run consequences of the intention of the programs to treat the health of children as it affects the children's productivity as adults.

MM: What does social science tell us about the efficacy of family planning interventions on standards of living?

PS: There have been very few evaluations of family planning that estimate the effect of intention to treat on lifetime or completed fertility, because the implementation of family planning in a rigorous experimental design has been rare. The lack of scientific evaluation studies of family planning does not suggest that gains in "standard of living" did not occur, but that there is no convincing evidence to show where they occurred and why. Cross-country regressions or even panel studies of countries or regions are generally not capable of confirming causal effects to answer your question, because the spread of family planning has not unfolded in a way that is independent of other development processes. It is likely that institutions of family planning are also supplying their services in response to the private demand for birth control in the population.

© 2014 PSI
1120 19th Street, NW, Suite 600
Washington, DC 20036
p (202) 785-0072 | f (202) 785-0120