Reevaluating the costs of teenage childbearing: Response to Geronimus and Korenman

Commentary is provided on some of the differences and reasons for the differences between the Geronimus and Korenman findings on the economic effects of teenage childbearing as well as on the authors own interpretation of existing data. The disagreements pertain to technical details involving sampling and data analysis. A review of the history and background of research by both research groups is presented. Geronimus and Korenman presented their first results in 1989 and 1991 based on data from the National Longitudinal Survey of Young Women (NLSY). The findings were that the effects of early childbearing were smaller than previously estimated. In addition the effects were statistically insignificant for economic status probability of high school graduation and other factors. These controversial findings were replicated by Geronimus and Korenman for a sample of NLSY data and data front the Panel Study of Income Dynamics (PSID). Hoffman et al. also analyzed PSID data. The findings from the three replications are very consistent; i.e. fixed-effect estimates are smaller than previously estimated. There are statistically significant and quantitatively important effects on economic status the probabilities of being officially poor and of graduating from high school. The probability of acquiring postsecondary education is also negative. The sister-comparison data....... support the contention of Geronimus and Korenman that the effects of teen childbearing are negligible. A table of the Hoffman et al. results shows that the NLSY results were outliers. Geronimus and Korenman argue that the NLSY data are correct. Concern is raised that the NLSY sisters-comparison data are unrepresentative because of selective inclusion of older women the smaller sample size in the PSID the underrepresentation of women aged 20-24 years the underrepresentation of married women and the misidentification of women without sisters. The conclusion is that it is too early to say that the true effects of teen childbearing are small.