U.S. Abortion Policy and Fertility
Citation: Klerman, J. A. U.S. Abortion Policy and Fertility.
This paper explores the potential effect on fertility of differing abortion policies: legalization and Medicaid funding. Using individual-level birth-certificate data compiled by the National Center for Health Statistics (NCHS) combined with Census estimates of state populations for the period where the funding of abortions changed in 1977 and the end of the data, 1992, Klerman produces estimates of the birth rate for a year of conception disaggregated by race (black or nonblack), age (grouped into five-year bands), state, and sometimes parity (first birth or higher order birth). The population estimates are not disaggregated by parity. The dependent variable is the log of the birth rate in a state in the calendar year of conception. The regressions are stratified by race, age, and sometimes parity and include variables for the policy of interest (legalization of abortion and Medicaid funding of abortions), the AFDC payment standard, economic conditions (the employment rate and, for the later period, the unemployment rate), dummy variables for state and year, and state-specific linear time trends. In the period between 1977 and 1981, Medicaid funding of abortions had no effect on the birth rate. However, in the period following 1982, Medicaid funding has a large negative effect on the fertility of blacks. The effect is largest for women in their twenties and for higher-order births. The results for whites follow a similar pattern, but are much smaller. This pattern of results is consistent with the eligibility requirements of Medicaid and the incentives that they provide. Klerman hypothesizes that the reason that no effect appears in the earlier period before court cases established that states were not required to fund abortions is because the states that changed their policies generally had funded relatively fewer abortions per woman. In contrast, the states that were most sympathetic to funding abortions funded abortions continuously throughout this period and funded relatively more abortions per woman. Klerman computed the TFR for each state that was implied by the results of the regressions and compares these computed values with the actual values. For whites, all of the effects are small. Starting at legalization and full funding, ending funding would increase the TFR by 2 percent. Making abortion illegal would increase the TFR by an additional 3 percent. For blacks the effects are larger; ending Medicaid funding would increase fertility by 10%, while making abortion illegal would increase fertility by an additional 5%. Parity-specific estimates are even larger. Klerman concludes that though the stated goals of the Personal Responsibility and Work Opportunities Act of 1996 (PRWORA) are to lower nonmarital fertility without increasing abortions, these two goals are in conflict. Medicaid funding of abortions to lower the birth rate would raise the number of abortions. Banning abortions would lower the abortion rate but increase the fertility rate.