Three essays on public policy simulations

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




Richard V. Burkhauser


microsimulation, minimum wage, labor

Subject Categories

Economic Theory | Labor Economics | Public Administration


This dissertation is composed of three separate essays that use estimations and simulations to evaluate the impact of various public policies. My first essay focuses on the population with disabilities. Using data from the Survey of Income and Program Participation (SIPP), I compare the labor force activity and economic status of people with disabilities to those without disabilities. I find that while people with disabilities are less likely to work and more likely to receive government transfers, a majority of men with disabilities and a significant share of women with disabilities work and do not receive any form of government transfers. Despite the importance of work for people with disabilities, a large percentage are enrollees in multiple government transfer programs. I find that the high effective tax rates of these programs make it unlikely that enrollees will rejoin the labor market. In the second half of this essay I develop a work-based policy for people with disabilities called the Disabled Worker Tax Credit that would serve as an alternative to some of these long term transfer programs.

In my second essay I analyze the employment and distribution impacts of minimum wage increases. I use an estimating equation from the research of leading proponents of the view that minimum wage increases do not cause employment losses. However, rather than using annual data from one month, I test the employment impacts of minimum wage increases using monthly data from both the SIPP and the Current Population Survey. I find the traditional result that neoclassical theory would predict: minimum wage increases create employment losses concentrated among less valued workers. In the second part of this essay, I shift my analysis to the distributional impacts of minimum wage increases. I find that the majority of benefits from minimum wage increases are not distributed to the working poor when an alternative measure of economic status is used. The overall benefits of these increases are even smaller when the estimated elasticities from the SIPP and CPS are added to control for employment losses.

In the final essay of my dissertation, I develop a microsimulation model to simulate a future population of retirees in the year 2010. The basis of my microsimulation model is a two step panel fixed effects and fixed factor econometric model that controls for individual level heterogeneity and allows for correlations across a series of outcomes. Unfortunately, the microsimulation model based upon this estimation technique produced forecasts that varied widely from actual data. At least part of this difference can be explained from two flaws in my econometric approach. First, the use of fixed effects estimators in unbalanced panels for simulation can distort the expected values of variables in the overall population. Second, the fixed factor equations that simulate the estimated fixed effect in the microsimulation model are not estimated with a great deal of precision. Future work with this microsimulation model calls for a new econometric approach that addresses these issues.


Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.