Study: “The Effects of Health and Demographic Change on Economic Growth: Integrating Micro and Macro Perspectives, III. Composition Bias in Estimating the Effect of Fertility Reduction on Economic Growth”
PI(s): Rubinstein, Yona
Co-PI(s): Weil, David; Wilde, Joshua
Affiliation(s): Brown University
Institutional Partner(s): PRB Center
Project Dates:
Start: 2007
End: 2010
Data Source(s): Multiple Surveys & Sources
Methods: Cross-Country Regression
Geographic Location(s): Cross-Country Analysis (with a focus on India)

Description:
This project is the third subproject of “The Effects of Health and Demographic Change on Economic Growth: Integrating Micro and Macro Perspectives.ʺ This project assesses how systematic bias may cause traditional cross-country regression analyses to understate the economic benefits of fertility reduction. The bias results from the common observation that reductions in fertility do not affect all parts of the income distribution equally. Specifically, poorer parts of a society usually lag in reducing their fertility. The goal of this study is to assess the magnitude of these compositional biases by adjusting growth rates of income per capita in a country to reflect composition changes from that country and by adjusting the coefficients derived from cross-country estimates of the effect of fertility reduction on income.

At first one might think that this composition bias would lead to the conclusion that lower fertility is not good for economic outcomes, since it leads to more poor people relative to rich. The researchers hypothesize, however, that the bias might lead an observer to think that fertility reduction was not as economically beneficial as it actually was. To see this perspective, consider an example in which there are two countries, one of which has a reduction in fertility (biased toward high-income families) and one of which does not. Suppose that. in fact. fertility has no effect on family economic outcomes so that all families in both countries experience the same rate of income growth. Because of the composition bias, the average level of income will grow more slowly in the country with rapid fertility reduction than in the country where fertility was constant. This program and project will produce results that contribute to a more nuanced understanding of the mechanisms that underlie the relationship between health, fertility, and economic growth.