Research Description
Job market paper:
"Unemployment duration and
part-time versus
full-time reemloyment and wages"
Adviser: Dr. Guilkey
Abstract: The main goals of
my research are to estimate the effects of unemployment
duration on formerly unemployed workers' wages, to analyze the
determinants
of part-time versus full-time reemployment, and to measure the impact
of
part-time versus full-time work on wages. Using the 1996 and 2001
panels of
the Survey of Income and Program Participation, I find that for the
typical man
a one month increase in the duration of unemployment leads to a 0.4%
decrease
in reemployment wages. In contrast, for the typical woman, I fail to
find any
effect of unemployment duration on wages. Furthermore, as unemployment
progresses, holding everything else equal, part-time reemployment is
more likely
than full-time reemployment for both genders. In addition, for men, my
results show
that there is no part-time and full-time wage differential for workers
with less than
a high school education, and for the same group of women, there exists
a positive
part-time wage premium. With the exception of women with an advanced
degree,
for both women and men, I find evidence of the existence of a full-time
wage
premium for higher levels of education, and the premium increases with
the level
of education. In this paper, I also simulate a short-term macroeconomic
policy for
a reduction of the state unemployment rate by 2 percentage points,
which would
alleviate the negative effects of prolonged unemployment on men's
reemployment
and wages. The simulation results show that a reduction in the state
unemployment
rate decreases the average duration of unemployment, and slightly
increases the log
of hourly wage rate for men.
Working paper:
"Testing
the validity of self-reported disability as a measure of true disability"
Abstract:
The main question of this research is whether self-reported disability
is systematically
biased relative to the SSA disability measures. In other words, are the
self-reported
disability measures in surveys conducted by independent organizations
reliable? There
is no doubt that many individuals may overestimate their disability
conditions for the
SSA due to rationalization factors. However, Benitez-Silva et al., by
introducing the RUR
hypothesis, believe that the applicants of the disability programs are
fully aware of the
SSA's disability criteria and their self-reported disabilities in a
non-governmental survey
are adjusted by the SSA's disability norm. In this paper, I have tested
the RUR hypothesis
by revisiting Benitez-Silva et al. (2004). First, I replicated
Benitez-Silva et al.'s analysis
as accurately as possible and then, using information from the
additional three waves of
the Health Retirement Study, I make updates to the analysis. The main
implications from
the non-parametric and parametric tests are that for the majority of
individuals' observed
characteristics, assuming that the SSA and applicants use the same
characteristics in
disability determination, the SSA and individuals probably use
different weights, which
is a direct contradiction to the RUR hypothesis.
"The causal relationship
between unemployment duration and unemployed worker's health"
Abstract: The main goal of this paper is to
explore the impact of impaired health on the duration of
unemployment. It is a common belief that self-reported health
correlates with unobserved
factors in the duration of unemployment equation. To solve this
endogeneity problem,
I propose the empirical model where I jointly estimate the health and
duration equations
controlling for unobserved worker heterogeneity using a discrete factor
method proposed by
Mroz and Guilkey(1995) and Mroz (1999). Using the 1996 and 2001 panels
of the Survey of
Income and Program Participation, I find that for men and women, a
worker with impaired
health has a significantly longer unemployment spell than a healthy
counterpart. Furthermore,
the results show that a worker with impaired health more likely ends up
with a part-time job.
