Essays on Disability, Public Assistance Programs and Employment

This dissertation explores the relationship between disability, public assistance programs
and employment. The first chapter investigates the impact of chronic pain on the labor market
participation decisions of people nearing retirement age. The results from estimating a random
effect discrete-time hazard model suggest that chronic pain greatly enlarges the hazard of exiting
labor market even after controlling for many diseases that contribute to pain, and this effect
increases with the severity of pain level. I further estimate a discrete-time competing risks model
to explore the association between chronic pain and labor market exits due to different reasons.
The finding indicates that chronic pain plays a more important role in the labor market exit
decisions of those who are disabled, rather than those who retire normally, implying that chronic
pain is a good measure of health-related work capacity.
The second chapter examines a potentially important spillover effect of raising the
minimum wages. I estimate the impact of minimum wage increases on SSI program participation
and employment outcomes among non-elderly adults with disability. Working age adults with
disability are at the margin of being affected by the wage floor. On the one hand, a large portion
of this population may not have the capacity to work. On the other hand, this is the group that may
be paid exactly at the minimum wage rate, on the premise that their health conditions do not
completely stop them from working. Using the variation in minimum wages across states during
the period 2004-2013, the findings suggest that higher minimum wages may cutback SSI program
participation through the channel of employment effect for younger adults with disability. No
discernible effect is found on the program participation or the employment outcome of adults aged
50 and older.
The last two chapters examine the effects of a policy change in the Medicaid eligibility rule
on various individual outcomes, including Medicaid coverage, employment, earnings and
household savings. The third chapter estimates the effects of eliminating Medicaid asset tests on
Medicaid enrollment and labor market outcomes of non-elderly, low-income parents using a
Difference-in-Difference method. The identification strategy exploits the exogenous variation in
the timing of asset tests removal among states. The finding indicates that removing Medicaid asset
tests only play an important role in mothers’ outcomes. Asset test removal is associated with a 4.4
percentage point increase in the probability of Medicaid enrollment for mothers, but the expanded
access to Medicaid slightly lowers their earned income, and decreases the probability of full-time
employment and any employment by 2.3 and 2.8 percentage points respectively.
The last chapter assesses the effect of eliminating Medicaid asset tests for low-income
families on household saving behavior. I use Difference-in-Difference method and data on loweducated household heads with children from the Survey of Income and Program Participation
(SIPP) to estimate the elimination effects. The findings indicate a strong positive impact of
eliminating asset tests on household holdings of liquid assets, and the effects are greater for states
with stricter lower asset limit levels before the elimination. However, the effect is not robust to the
inclusion of state specific time trends in the model, which may imply that the elimination of
Medicaid asset tests correlates with other trends in state level output, and it may not be possible to
disentangle the causal effect of the asset test removal from these underlying trends. This study
finds no evidence of the elimination effect on non-liquid assets or household net worth (excluding
the value of primary residence and first car).