Dissertation Title: "Essays in Health Economics: Understanding Risky Health Behaviors"
This dissertation presents three papers applying health economics to the study of risky behaviors. The first uses data from the 1979 National Longitudinal Survey of Youth to examine the relationship between adverse events and risky behaviors among adolescents. Substance use responses to experiencing either of two adverse events--violent crime victimization or death of a non-family member one felt close to--explain 6.7 percent of first cigarette use, and 14.3 percent of first use of illegal drugs other than marijuana. Analyses of exercise, a positive coping mechanism, find shock-responses consistent with a coping-response, but not with rational, time-inconsistent, or non-rational drivers considered here. I conclude that distressing events lead to risky behaviors, with a coping response contributing to this effect.
Using National Health Interview Survey data, the second paper considers the mechanism behind growth in smoking's education gradient between 1950 and 1980. Regressions test for education differentials in initiation and cessation responses to cigarette advertising, prices, brand-specific risk information, and public health information on smoking. Differential advertising-responses explain 39 and 27 percent of growth in smoking's education gap among males and females, respectively, while a differential response to brand-specific tar and nicotine information explains a further 13 and 8 percent. Brand-choice analyses find an education gradient among smokers: more educated smokers favor lower risk cigarettes, prefer the more modern high filtration brand-image, and smoke fewer cigarettes per day. These analyses suggest education differentials in demand for risk-reduction and brand-image responses.
My third paper considers the extent to which gateway effects, dual use, and harm reduction shape the relationship between youth smoking and electronic cigarette use. Using National Youth Tobacco Survey data on high school students ages 14 to 18, the analysis estimates propensities to be a current smoker absent access to electronic cigarettes, and considers the impact of changes in electronic cigarette availability on smoking rates in different propensity groups. Harm reduction is evident in the high propensity group, wherein a one percentage point increase in electronic cigarette use is associated with a 0.5 percentage point drop in the current smoking rate. There is no evidence of a gateway effect.