Here is my CV (pdf) if you are interested
Abstract: Certificate of Need (CON) regulations were designed to limit "wasteful" hospital competition. These regulations affect the number of hospitals in a community as well as the number of services that hospitals can provide. The intention of these regulations is to reduce medical costs and improve quality. I use data from the National Center for Health Statistics Linked Birth and Infant Death file (which includes almost all of the births and neonatal deaths in the U.S.) to show that the regulation of obstetrics or neonatal intensive care units by CON does not improve neonatal health. In contrast, I find that mothers in states which repealed CON regulations in entirety are slightly more likely to have a healthy baby as compared to mothers in states with CON regulations. Analysis of newborn mortality suggests that extremely premature babies may be slightly worse off in the years immediately following a state's CON repeal. Taken as a whole, the empirical evidence indicates that Certificate of Need regulations have very little impact on health outcomes. This undermines the claims of CON proponents. CON regulations do not ensure health care quality.
pdf
Abstract: Policy makers and economists have worried about the problem of defensive medicine for a number of years now without being certain of its magnitude or of its existence. Previous research has used tort reforms and malpractice insurance rates to proxy for the potential pressure of litigation on doctor's behaviors. This paper takes a different approach and directly examines how doctors respond to being sued for malpractice. Using hospital discharge data and court records, the effects of a malpractice lawsuit on physician behavior is estimated. Because it is possible that malpractice lawsuits are correlated with unobserved physician characteristics, a difference in difference approach is used to circumvent endogeneity problems. This paper uses a dataset of labor and delivery doctors and the empirical results indicate that the first time a doctor is sued for malpractice increases certain defensive medicine practices. The percent of births for which a cesarean section is performed increases significantly, while estimates for other obstetrical procedures lack statistical significance.
Abstract Economists have long been interested in the extent to which economic resources affect decisions to marry and divorce. For married couples, an increase in resources can either provide a stabilization effect or, alternatively, can enable divorce by allowing the couple to overcome costs associated with divorce. Similarly, while economic theory predicts that an increase in income makes an unmarried person more attractive to potential marriage partners, it may also make single life more attractive. However, answering these questions empirically has been difficult due to a lack of exogenous income shocks. We overcome this problem by exploiting the randomness of the Florida Lottery and comparing recipients of large prizes to those of small prizes. Results indicate that while positive income shocks of $25,000 to $50,000 do not cause statistically significant or economically meaningful changes in divorce rates, single women are less likely to marry as a result of the additional income.
pdf
Abstract: A fundamental question faced by policymakers is how best to help individuals who are in financial trouble. This paper examines the consequences of the most basic approach: giving people large cash transfers. To determine whether this prevents or merely postpones bankruptcy, we exploit a unique dataset of Florida Lottery winners linked to bankruptcy records. Results show that although recipients of $50,000 to $150,000 are 50 percent less likely to file for bankruptcy in the two years after winning relative to small winners, they are equally more likely to file three to five years afterward. Furthermore, bankruptcy filings indicate that even though the median winner of a large cash prize could have paid off all of his unsecured debt or increased equity in new or existing assets, she chose not to do either. Consequently, although we cannot be sure other recipients of financial assistance would react in the same way lottery players did, our results do suggest that some skepticism regarding the long-term effect of cash transfers may be warranted.
pdf
Link to NY Times blog