Abstract

This thesis comprises four essays on important public health issues. The first essay studies how social interactions can spread petty corruption in the health sector. Using a Vietnam dataset, I find that social interactions measured by advice on hospital choice increase the propensity of patients to give bribes to hospital staff as well as raises the bribe amount. There is also evidence on the information-transmitting role of social networks. The second essay evaluates long term health impacts of Agent Orange exposure in the VietnamWar on the Vietnamese population. I use a unique dataset that includes both self reported hypertension and objectively measured blood pressure. The results indicate that exposure to Agent Orange significantly increases the risk of having hypertension and reduce height, with the largest burden falling on the cohort born during the spraying period and on the most heavily sprayed areas. I also show that using self-reported hypertension data may lead to upward bias in the estimate of the effects of Agent Orange on hypertension. There is also evidence that exposure to Agent Orange and herbicides during the Vietnam War increases risk of cancer and mental illness. The third essay investigates a possible link between hypertension and happiness by examining possible impacts of neighborhood wealth on individuals' hypertension. Using both self-reported and objective hypertension data to proxy for happiness, I find that self-reported hypertension rate is much lower than objectively measured hypertension rate which lead to a large discrepancy between results obtained from self-reported and objective hypertension data. Moreover, I find that high neighborhood wealth raises hypertension risks for people aged 55-65 and not for younger or older age groups. The fourth essay provides a theoretical rationale for smoking bans by proposing a theoretical model of maximizing behaviour on the part of smokers. It also empirically evaluates effects of smoking bans imposed at home and in workplace. Both calibrated model simulations and empirical results suggest that, with the exception of heavy smokers, workplace bans have relatively minor impacts on smokers while restrictions on smoking in the home are found to be of an order of importance greater.