Panel Paper: Does Access to Emergency Contraception Affect Unintended Pregnancy Rates? New Evidence from STD Data

Saturday, November 8, 2014 : 3:50 PM
Enchantment Ballroom E (Hyatt)

*Names in bold indicate Presenter

Inna Cintina, University of Hawaii and Petru S. Stoianovici, Charles River Associates
For the past several decades, the highest unintended pregnancy rates in the United States were among teenage and young women, particularly among poor and minority groups.  Access to birth control and relatively easier access to emergency contraception (EC), like Plan B, could not only reduce the number of unintended pregnancies, but might also mitigate some of the social costs associated with teen childbearing.  However, the identification of the potential effects of EC is a challenging task, since a reduction in the overall cost of EC also reduces the overall cost of unprotected sex, which in turn could change the sexual behavior.  Depending on which changes in behavior dominate, the reduction in the cost of EC can decrease, increase or have no effect on the incidence of unintended pregnancies.  Economists agree that a reduction in the EC costs is associated with an increase in the incidence of sexually transmitted diseases (STDs), indicating a relative shift toward riskier sexual behavior.  Yet, there is no consensus in the economic literature about the effect on the incidence of unintended pregnancies, typically measured by the incidence of abortions.  These mixed EC results could be driven by differences in the data used, but also by methodological differences.  We contribute to the unintended pregnancies and EC literature by evaluating the consistency between the results of the unintended pregnancy models used in the literature, and the results of the effects on the STDs models.  Given the consensus of the effect for the STDs outcome, by using the same data for the unintended pregnancies and STD models, this paper offers a simple test of the validity of results reported in the EC studies, and can shed light on the sources of differences in the EC results in the literature. Provisional and partial results suggest that the unintended pregnancies models that are consistent with the STD models show a statistically significant effect of EC on pregnancies.