Ideas with Impact

A Recipe for Analyzing Fertility

I began my career with the Population Council in 1973, and in 2015 I retired as its president. One of the joys of working at the Council was being surrounded by public health and social science all-stars. Council researchers have changed the way the world thinks about important health and development issues and influenced policies and practice on family planning, the rights and health of girls, and population change. A highlight of my career has been seeing Population Council vice president and distinguished scholar John Bongaarts speak before standing-room-only-audiences in New York, New Delhi, and Addis Ababa. Like a small number of superstars in other professions, John can fill a hall in any of the world’s capital cities.

John’s books and articles, which lead people around the world to crowd into rooms to hear him speak, analyze complex problems, discern their key components, and then elucidate simple but comprehensive models for understanding them. The best-known and most influential of these models is the framework he developed to explain how marriage patterns, contraception use, breastfeeding, and abortion impact fertility.

Proximate Determinants of Fertility

These proximate determinants of fertility—factors that have a direct effect on fertility, as opposed to indirect determinants of fertility like socioeconomic, cultural, and environmental variables—were first described in a 1956 article by University of California sociologists Kingsley Davis and Judith Blake. Davis and Blake asked how differences in fertility between developed and less-developed countries could be explained. They proposed a list of factors—intercourse, conception, and gestation variables—that determine fertility, but their relative importance and their relation to one another were vague. In effect, early attempts to describe the determinants of fertility were like saying “Take some flour, butter, milk, eggs, and sugar, mix them together, bake, and you’ll have a cake.” John’s achievement was to detail precisely how varying these ingredients affected the cake’s outcome. Using data from censuses and surveys, he created an elegant but straightforward framework that researchers could apply to analyze the determinants of fertility.

This framework, which was published in 1978 in the Council journal Population and Development Review, highlighted four factors—the proportion married, the prevalence of contraception, the extent of induced abortion, and the length of postpartum amenorrhea—that largely determine fertility rates. Since its publication, John’s framework has become a mainstay of demographic investigations. It has been incorporated into hundreds of studies that collect and report data used by policymakers and program managers and is taught to students of demography around the world.

John’s framework allows researchers to answer important questions about fertility patterns. For example, I was with John in Addis Ababa when an audience member asked how fertility can be low when contraceptive use is also low. The answer lies in the other proximate determinants. If fertility and contraceptive prevalence are both low, there must be an older age at marriage, higher abortion rates, and/or more breastfeeding and therefore longer intervals of postpartum amenorrhea.

John’s algorithm also helps clarify policy and programmatic options. If lower fertility is a goal, interventions must work through the proximate determinants, by increasing contraceptive use, promoting older age at marriage, encouraging breastfeeding, or allowing more abortions.

At the same time that John was improving the analytical techniques used to study reproductive behavior and showing how data could be powerfully interpreted and used, researchers began paying more attention to data-collection efforts. Sample surveys were being improved with better sampling designs, better questionnaires, and more robust measures of the variables that John identified.

These improvements—the specification of more precise theoretical models (the more exact recipes) and the improvements in sample surveys (the raw ingredients for analysis, if you will) have made a huge difference in social science and public health research.

Trajectory of Population Growth

In another path breaking paper, this one published in Science in 1994, John developed a widely cited model to explain the trajectory of population growth. John presented data on three factors that were decisive for future growth: unwanted fertility, high desired family size, and population momentum—the tendency for population growth to continue after replacement-level fertility is reached because of the concentration of young people in the childbearing years. As with the proximate determinants framework, this new framework organized complex facts into an easy-to-understand paradigm that explained the effects of varying the components and their policy implications. And John’s remarkable productivity has continued. His most recent article describes the shocking and tragic case of missing women due to sex-selective abortion, neglect, and excess female mortality.

John’s success is based on mastery, imagination, and, as befits his position at the Population Council, relevance to population, health, and development policies and programs. He has a thorough command of the scientific literature, the available data, and a wide range of analytical techniques. And while many other researchers do, too, what sets John apart is what he does with the armamentarium of modern demography. He asks important questions, uses the best data, and develops innovative analyses to produce papers and reports that are always sound and thoughtful and rich with important and sometimes unexpected implications for science and public policy.

John is just one of the reasons that for more than 40 years I have been dazzled by the Council’s work to change the world for the better. I’m proud to have been a part of the Population Council family and look forward to its continued success in improving health, saving lives, and advancing development.

Peter J. Donaldson served as president of the Population Council from 2005 to 2015.