Year of Publication

2015

Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation

College

Business and Economics

Department

Economics

First Advisor

Dr. Christopher Bollinger

Abstract

This dissertation examines heterogeneity in the value of a statistical life (henceforth VSL) stemming from employer-provided health insurance (henceforth EHI) and worker sorting. The dissertation consists of three essays.

In the first essay (Chapter 2), I investigate the effect of health-driven productivity on the wage compensation for mortality risk, and how EHI influences VSL using the US labor market data. In this chapter I build a framework showing that the level of job risks influences the incentive of employers to provide EHI. The basic notion of the framework is that health insurance is an investment in health and health is a form of general human capital. Employers are willing to invest in employees' health and pay the associated costs as long as they can recoup the costs of health investment. Occupational hazards, however, are harmful to health; productivity gains from health tend to decline as risk increases, resulting in lower health investment made by employers. As a result, the workers in risky jobs have to contribute more to their health investment in the form of lower wages than do workers in safe jobs. This behavioral response pushes down the wage offer curve of the insured in high risk occupations. Consequently, workers with health insurance, on average, accept a lower risk premium, leading to a lower VSL. Empirical findings from this dissertation suggest evidence of heterogeneity in VSL due to health insurance status: the estimated VSL for workers with health insurance is lower than those without one.

In the second essay (Chapter 3), I extend the framework of the second chapter into the United Kingdom (the UK) labor market. Different from the US, the UK has universal health care system in which all eligible individuals (almost all the UK citizens) are covered by publicly-provided health care. This chapter also provides evidence that private medical insurance in the universal health care system affects the risk premium. Despite the fact that the UK and the US have different institutional settings in health coverage, findings from the UK are, to some extent, qualitatively similar to the US.

A major issue in estimates of VSL is that people are not randomly assigned to jobs. That is, heterogeneous people would sort into jobs based on their preferences on risk and safety-related skills. Thus, failure to account for heterogeneity in both risk preferences and safety-related skills will bias the estimated VSL. In the third essay (Chapter 4), I discuss worker sorting and how it may affect the mortality risk premium. In this chapter, I focus on the role of personality traits in safety-related skill and their influence on worker sorting based on job risk. I use Five-Factor Model of personality or also known as the ‘’Big Five” personality traits. The big 5 personality traits are extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience. In my framework, these personality traits are inputs and the technology of skill formation transforms the traits into safety-related skill.

Share

COinS