Understanding the Fertility-Economy Link for Teenagers

Rand Corporation, Santa Monica CA

Investigators

Abstract

DESCRIPTION (provided by applicant): During the 1990s, the economy strengthened and teenage fertility and the rate of sexual activity and the lack of contraception decreased sharply. Black female teenagers, who generally experienced greater relative economic gains than white female teenagers, had even greater declines in fertility, the rate of sexual activity, and the lack of contraception. These patterns suggest that there may be a link between the economy and fertility-related outcomes. In this project, we will combine state aggregate economic data with individual behavior data from the National Longitudinal Survey of Youth from 1997 to estimate how changes in the economy affect fertility and its proximate determinants--the rate of sexual activity, contraception use, pregnancies, and abortions-- for all teenagers and across race/ethnicity. We will use a difference-indifference method to control for unobserved heterogeneity across states and generic factors that would influence teenage behavior across the country (e.g., Magic Johnson's announcement that he has AIDS). Previous researchers have examined the economy-fertility-outcomes link. However, besides using more current data that spans the build up to the peak of the business cycle along with the following recession, we make several contributions. Previous studies used narrowly defined economic measures (e.g., state teenage employment-population ratio) that are subject to sampling error from being based on small samples and subject to endogeneity in that exogenous fertility shocks could affect the economic measure. One of our contributions is to use (1) the aggregate state unemployment rate, which should hardly be affected by teenage fertility shocks and is less subject to sampling error than the measures used in previous studies and (2) a state employment growth rate based on the industrial composition of the state and national output growth rates across states. Another contribution is that we will perform simulations to determine how much the rates of sexual activity, lack of contraception, pregnancies, and abortions changed in the 1990s as a result of the economic changes, and we will calculate how much of the patterns for all teenagers and across race/ethnicity can be explained by the economic changes.