Health behaviors engender delayed health consequences. Time preferences (commonly measured in terms of time discounting) influence how an individual views immediate short term costs over long term benefits. Individual differences in time preferences are shown to be associated with addictive health behaviors such as smoking or substance abuse. However, evidence on how such preferences influence beneficial health behaviors, is mixed. This proposed dissertation research proposes two specific aims - i) to understand the factors that influence time preferences and how time preferences influence healthy behaviors;and ii) to understand whether time preferences show a 'present bias'which may be the reason for observed procrastination tendencies using behavioral economics paradigm. The data used are from Mexican Family Life Survey, a longitudinal nationally representative survey of Mexicans and Mexican Migrants to the United States. The first research question utilizes an instrumental variable approach to address the simultaneous determination of time preference and health behaviors including diet, physical activity, and smoking. The main explanatory variable - discount rate - will be calculated using answers to the hypothetical gambles assuming traditional exponential discounting. The second research question will explore whether present bias (as evidenced by quasi-hyperbolic discounting as opposed to exponential discounting) explains individual differences in health behaviors using non-linear regression models. Additionally, a successful social program (Conditional Cash Transfer) from Mexico will be examined to understand whether its success in promoting healthy behaviors might be attributed to alleviating present bias. Using the proposed framework taking into account the simultaneous determination of health behaviors and discounting would further our knowledge of how individual discounting behaviors, which is an essential element of cost effectiveness and other economic analyses. Utilizing behavioral economics paradigm addresses some of the observed anomalies that are not explained by traditional models. Using multiple measures of time discounting (one based on hypothetical gambles and another traditional proxy measure) facilitates our understanding of issues related to measurement methods of time discounting. Additionally, this study will be one of the original studies addressing discounting specifically in a minority population where problems related to health behaviors such as obesity are increasing at alarming rates.
The proposed study furthers our understanding of discounting of health by individuals, and specifically by minority populations by utilizing conceptual models from standard economics as well as from behavioral economics. This study further investigates how measurement methods impact discounting of health, which is crucial to comparative effectiveness analyses and other economic analyses. Understanding mechanisms involved in how inconsistent and myopic preferences adversely influence health behaviors is essential to devising health care policies geared towards helping individuals overcome such suboptimal behaviors.
