The Consequences of Teenage Childbearing: Consistent Estimates When Abortion Makes Miscarriage Non-random
We are grateful to Carlos Sepulveda-Rico and Greg Hutko for excellent research assistance and to Joshua Angrist, Mingli Chen, Donna Ginther, Jonathan Guryan, John Johnson IV, Karen Norberg, Robert Pollak, the editor, Steve Pischke, and two anonymous referees and participants in informal labour lunches at Boston University and MIT and workshops and seminars at George Washington University, Harvard, the NBER and Washington University for helpful comments and suggestions. This research was supported in part by NSF grant SEC-0339149 and NIH grant R03 HD056056-1. The opinions expressed are those of the authors and not necessarily those of the Federal Reserve Bank of New York or of the funding agencies.
Miscarriage, even if biologically random, is not socially random. Willingness to abort reduces miscarriage risk. Because abortions are favourably selected among pregnant teens, those miscarrying are less favourably selected than those giving birth or aborting but more favourably selected than those giving birth. Therefore, using miscarriage as an instrument is biased towards a benign view of teen motherhood, whereas ordinary least squares (OLS) on just those giving birth or miscarrying has the opposite bias. We derive a consistent estimator that reduces to a weighted average of OLS and IV when outcomes are independent of abortion timing. Estimated effects are generally adverse but modest.