Date of Completion


Embargo Period



Labor, Health Limits, Divorce, Earnings, Employment

Major Advisor

Kenneth Couch

Associate Advisor

Dennis Heffley

Associate Advisor

David Simon

Field of Study



Doctor of Philosophy

Open Access

Open Access


The dissertation considers the influence that the arrival of work limiting health conditions has on labor market activity but also on broader social behavior of those that experience them. The primary data source used in the analysis is the 2004 Survey of Income and Program Participation (SIPP).

The first chapter examines demographic correlates of the evolution of work limiting health conditions in the United States. Beyond a basic description of the onset and incidence of specific health conditions and their association with common demographic covariates, a set of specific health conditions that arrive largely unexpectedly are identified. The primary method used is logistic regression.

The second chapter then divides the reported health conditions in the SIPP into those that are less predictable (exogenous) versus those that are more predictable (endogenous). How the onset of work limiting health conditions affects the subsequent divorce behavior is studied across those groups. Using retrospective histories contained in the topical module on “work disability history” of SIPP, I find that for men and women divorce behavior is not explained by the onset or evolution of exogenous health conditions while it is closely related to the onset or evolution of a broader, arguably, endogenous set of health conditions. The patterns of response are shown to vary by race and origin. The primary estimation method is a panel linear probability model with fixed effects.

In the third chapter, the effects of exogenous health conditions and more predictable health conditions on employed people’s earnings and employment are examined. Using information contained in all waves of the core data and the topical module on “work disability history”, I find that people who are observed employed and later experience the onset of any work related health conditions tend to have lower subsequent earnings and a lower probability of being employed compared to the people who stay healthy. The adverse impact is even greater for people with exogenous health conditions. The impact of any work limiting health condition exists among different demographic groups to varying degrees. The primary estimation method is a difference-in-differences regression model with person and year fixed-effects.