| dc.description.abstract |
<p>In this dissertation, I discuss two important factors in individuals' decision-making
processes: subjective expectation bias and time-inconsistent preferences. In Chapter
I, I look at how individuals' own subjective expectations about certain future events
are different from what actually happens in the future, even after controlling for
individuals' private information. This difference, which is defined as the expectation
bias in this paper, is found to have important influence on individuals' choices.
Specifically, I look into the relationship between US elderly's subjective longevity
expectation biases and their smoking choices. I find that US elderly tend to over-emphasize
the importance of their genetic makeup but underestimate the influence of their health-related
choices, such as smoking, on their longevity. This finding can partially explain why
even though US elderly are found to be more concerned with their health and more forward-looking
than we would have concluded using a model which does not allow for subjective expectation
bias, we still observe many smokers. The policy simulation further confirms that if
certain public policies can be designed to correct individuals' expectation biases
about the effects of their genes and health-related choices on their longevity, then
the average smoking rate for the age group analyzed in this paper will go down by
about 4%.</p><p>In Chapter II, my co-author, Hanming Fang, and I look at one possible
explanation to the under-utilization of preventive health care in the United States:
procrastination. Procrastination, the phenomenon that individuals postpone certain
decisions which incur instantaneous costs but bring long-term benefits, is captured
in economics by hyperbolic discount factors and the corresponding time-inconsistent
preferences. This chapter extends the semi-parametric identification and estimation
method for dynamic discrete choice models using Hotz and Miller's (1993) conditional
choice probability approach to the setting where individuals may have hyperbolic discounting
time preferences and may be naive about their time inconsistency. We implement the
proposed estimation method to US adult women's decisions of undertaking mammography
tests to evaluate the importance of present bias and naivety in the under-utilization
of mammography, controlling for other potentially important explanatory factors such
as age, race, household income, and marital status. Preliminary results show evidence
for both present bias and naivety in adult women's decisions of undertaking mammography
tests. Using the parameters estimated, we further conduct some policy simulations
to quantify the effects of the present bias and naivety on the utilization of preventive
health care in the US.</p>
|
|
