Emily Oster - Career

Career

Oster is perhaps most well known for her PhD dissertation, "Hepatitis B and the Case of the Missing Women," in which she suggests that biology can be used to reveal the truth about the missing-women puzzle. Oster points to findings that areas with high Hep B rates tend to have higher male-to female birth ratios. The fact that Hep B can cause a woman to conceive male children more often than female, she says, accounts for a bulk of the "missing women" in Amartya Sen's famous 1990 essay, "More Than 100 Million Women Are Missing." Sen, on the other hand, attributed the "missing women" to societal discrimination against girls and women in the form of the allocation of health, educational, and even food resources. The use of Hep B vaccine in 1982 led to a sharp decline in the male-to-female birth ratio, she notes in her dissertation.

These conclusions have been disputed by Avraham Ebenstein, whose work (focused on the more recent period, after Oster's analysis) finds that unnaturally high male-to-female ratios occur only for children born after the first-born. This suggests that sex-selective abortion is mostly responsible for the "missing women".

In April 2008, Oster released a working paper "Hepatitis B Does Not Explain Male-Biased Sex Ratios in China" in which she evaluates new data and admits that her old research was incorrect. " This has been seen as a sign of integrity by Freakonomics author Steven Levitt.

Oster's current work focuses primarily on HIV in Africa. In a 2009 Ted Talk, she discusses how an incentive-based analysis can illuminate public policy in trying to reduce the spread of HIV. For example, life expectancy in Africa is relatively low; therefore, more individuals may decide that taking precautionary measures to reduce the chances of death from AIDS aren't worth the effort. The implication is that AIDS education and preventive programs are more likely to succeed in areas with longer life expectancies, e.g. in areas with less malaria. Oster also notes that it is possible to compare AIDS prevalence in the absence of random HIV testing by looking at mortality data (since much of the data on people with AIDS in Africa focuses on pregnant women and intravenous drug users). By comparing mortality rates in the 20 to 40 year old age groups, where AIDS kills more than other diseases (of the very young or old), it is possible to estimate the extent of the AIDS epidemic even without direct HIV testing.