|Title||Essays on Optimal Portfolio Choice and Unemployment Insurance|
|Year of Publication||2007|
|University||University of California, Los Angeles|
|Keywords||Employment and Labor Force, Health Conditions and Status, Net Worth and Assets|
This dissertation consists of two essays exploring optimal portfolio choice over the life cycle and optimal unemployment insurance program. The first essay explores the effects of uninsurable risk of health expenditures as well as labor income risk on portfolio choice in a realistically calibrated life-cycle model. Most of the existing literature that examines labor income risk and its effect on portfolio composition cannot explain continued declines in risk-taking with age after retirement. This paper uses MEPS (Medical Expenditure Panel Survey) and HRS (Health and Retirement Study) data to calibrate uncertain medical expenses for the retired. With the consideration of idiosyncratic health expense risk in addition to labor income risk, the model can generate declining financial risk-taking with age after retirement, and therefore fits the data much better than those studies which consider labor income alone. Additionally, regressions on simulated data also show that investors with poorer health tend to hold a smaller share of stocks in their portfolios, which is consistent with the empirical pattern of portfolio choice. Finally, using the model I have developed, I predict the impact of changes in government health insurance programs, such as the expansion of the Medicare program, on individuals' portfolio choices. Simulated results show that a more generous Medicare policy significantly increases the proportion of financial wealth held in equities for the retired. The second essay proposes the optimal design of the unemployment insurance contract in an environment with consumption commitments in which people cannot substitute freely among different goods within a single period. The optimal plan I obtain involves a relatively flat decreasing sequence of insurance payments over some duration, which is then followed by a large drop to a very low level of transfer. The results fit current policy well, and therefore give an explanation to justify the current policy. Additionally, the model predicts that if we change from the current unemployment program to the optimal contract, the government will only save 1.7% in unemployment payments, which shows that current policy is not as flawed as researchers have traditionally believed. In fact, to achieve efficiency, the efficient unemployment transfers should include a jump, similar to what we observe in practice.
|Endnote Keywords|| |
|Endnote ID|| |
|Short Title||Essays on Optimal Portfolio Choice and Unemployment Insurance|