My research aims to better understand the relationship between health insurance costs and labor market trends in the U.S. I will be testing the extent to which employers respond to rising health insurance costs by lowering wages, changing the composition of workers, and adjusting the generosity of the health insurance packages they offer. My hypothesis is that employers simultaneously utilize all of these mechanisms to compensate for higher health insurance costs. My analysis will primarily rely on estimating a firm level fixed effects model with the National Compensation Survey employer level panel data set. Using these estimates, I will look at the relationship between health insurance costs and average wages, the number of employees hired, the generosity of the insurance plan offered, and the decision to offer health insurance to its employees. This analysis will be accompanied by an individual fixed effects model where I use individual panel data from the Medical Expenditure Panel Survey to look at how workers of different skill levels and demographic groups see their wages and hours worked change in response to rising health insurance costs. The results of these two estimations will be presented in multivariate regression tables that provide insight into the labor market effect of rising health insurance costs.
The proposed project will examine the relationship between automatic enrollment and employee compensation. Autoenrollment has been shown to be highly effective at increasing employee participation in 401(k) s. However, the increase in pension participation generated by autoenrollment will increase employers cost of offering a matchbecause they now have to match the contributions of previously unenrolled workers. Our hypothesis is that firms who adopt autoenrollment will lower their match rates or other compensation to offset their higher costs. We propose to test this hypothesis using the National Compensation Survey (NCS) micro data pooled for 2005 and 2009. The analysis will require: employer level data on establishment size, region, and industry; pension plan-level data on plan type, match structure, match rates, contributions, participation rates, and automatic enrollment; and job-level data on unionization, full-time/part-time status, occupation, and compensation (including wages, defined benefit pensions, defined contribution plans, health insurance, life insurance, short-term and long-term disability insurance, paid holidays, paid sick leave, paid vacations, paid jury duty service, quality of life benefits, and nonproduction bonuses).
Over the past two decades, there has been a dramatic rise in employers accommodating workers who have suffered on-the-job injuries. Workplace accommodation is expected to reduce the time injured workers spend away from work, thereby reducing Workers Compensation (WC) costs for experience rated employers. In this paper, we examine whether employers are more likely to accommodate injured workers when employer costs for WC rise. We will conduct our analysis both at the aggregate level and the individual level. At the aggregate level, we identify the rate of accommodation at the state-year level using the publicly available series, Incidence Rates of Nonfatal Occupational Injuries from the Survey of Occupational Injuries and Illnesses. At the individual level, we use the restricted use geocode NLSY79 and identify those individuals injured at work and which of those individuals were accommodated by their employer. The main covariate of interest will be a state-year measure of employer costs for WC, drawn from the micro data of the National Compensation Survey.
This research project investigates two important questions related to total compensation for employees with disabilities: 1) When total compensation is defined more completely to include compensation beyond wages and salaries (e.g., health insurance, pensions), how large is the total compensation gap between employees with disabilities and employees without disabilities?, and 2) How is the mix of total compensation structured for employees with disabilities relative to those without disabilities? To answer these questions we will link data collected in the Employer Costs for Employee Compensation by occupation to other demographic survey data
The principal aim of this study is to investigate the extent to which occupational fatalities among youth under age 18 involve activities prohibited by the US child labor laws and identify the case characteristics most commonly associated with deaths due to prohibited activities. Investigators will utilize a case series approach to identify and describe all occupational fatalities among youth involving prohibited activities between 1992 and 2008 recorded in the Census of Fatal Occupational Injuries (CFOI) data. Prohibited activities will be determined by examining decedent age, relevant employment variables (i.e., family business, industry), time of injury, and decedent activity at the time of injury (from narratives) and cross-referencing this information against the applicable child labor laws.
The value of statistical life (VSL) is a measure of peoples willingness to tradeoff wealth for risk. With few exceptions, analysts apply uniform VSL estimates to policy evaluations, regardless of the characteristics of the targeted population. Using the hedonic wage approach, this study examines regional differences in VSL in the United States, reflecting variation in wealth, demographics, and preferences across the states. Specifically, we calculate VSL based on the tradeoff between labor market wage and job-related fatality risk. Previous studies have revealed differences in VSL between developed and developing countries, with lower VSL values for the latter. These studies suggest that health is a normal good that is positively related to wealth. We hypothesize higher VSLs in wealthier regions and lower VSLs in poorer regions. Access to confidential microdata will allow construction of more complete and accurate measures of fatality risk; complete measures of fatality counts by state and industry are not currently released due to small cell suppressions.
Climate change has emerged as the biggest global health threat of the 21st century and increases in average ambient temperatures are expected globally. Heat illness in the general population has received research attention due to recent deadly heat waves. However, impacts of climate change on worker health have not been well-characterized. This research will study occupational fatalities due to exposure to environmental heat and overexertion. The following hypotheses will be tested:
Three analytical approaches will be employed to test these hypotheses. A descriptive study examining six years (2005-2010) of Census of Fatal Occupational Injuries (CFOI) data will quantify and describe heat-related occupational fatalities. CFOI data will be queried for relevant event codes. Data from the National Climatic Data Center (NCDC) will be used to assign the maximum daily temperature for each fatality based on temporal and spatial boundaries. A similar descriptive analysis will examine overexertion-related fatalities identified through CFOI. The third analysis will examine excess morbidity and mortality to quantify and describe overall occupational illnesses, injuries, and fatalities occurring during heat waves. In addition to a morbidity analysis of three state emergency department databases, CFOI data will be queried to calculate excess mortality. NCDC data will identify three heat wave events during 2005-2010. The results will present excess mortality during the heat wave period compared to a referent time period. Subgroup analysis of the data will examine distributions of excess mortality by cause/event, age, occupation, and gender. In addition to determining areas for further research, results could inform policy and standard updates as well as adaptation measures to protect vulnerable workers from excess ambient heat.
A firm's financial policy and condition can have a significant effect on overall firm value through the indirect costs that they impose on firm stakeholders. While there is a large body of research in financial economics examining the how capital structure affects the conflicts between equity holders and debt holders, very little research has been done examining the impact on other stakeholders in the firm. This paper proposes to examine how firm employees, as stakeholders, are affected by the financial policy and condition of the firm. Using the micro level panel data from the Survey of Occupational Injuries and Illnesses, this paper will examine the statistical relationship between financial policy and changes in workplace injuries. First, we propose to examine whether firms that become financially distressed, as a result of high debt levels, see an increase in workplace accidents. Second, we propose to examine whether manufacturing plants taken private in leveraged buyouts experience improvements or declines in workplace safety. Lastly, we propose to examine whether stronger banking relationships and scrutiny by financial intermediaries has a measurable effect on workplace safety. The answers to these questions should prove very revealing to financial economists interested in the effects of stakeholder conflicts within the firm, as well as workplace safety organizations wishing to model the determinants of workplace injuries.
A large and influential literature exists that uses wage-risk tradeoffs in the labor market to estimate the willingness to pay for small reductions in the risk of a fatality. This research proposal addresses two major shortcomings in the existing empirical literature and seeks to extend our knowledge on decision making under uncertainty. The first shortcoming of past empirical studies is that it, without exception, fatal and non-fatal injuries are treated as separate processes. We challenge this approach and develop an empirical model based on structural actuarial loss models of the injury and death process in the labor market. Secondly, past labor market wage studies uniformly ignore the heterogeneity in injury severities. By treating injuries as homogeneous, the past literature has (i) provided no information on the value of reducing the severity of an injury and (ii) likely resulted in biased estimates of the value of reducing fatality risks since they are likely correlated with the injury severity. We propose a different approach that explicitly incorporates both the probability that an accident occurs and the expected severity of an accident at the sub-fatal level. By explicitly recognizing that both the severity and probability of an accident matter in determining compensating wage differentials, we hope to reduce bias in estimates of both the value of reducing fatal and non-fatal risks. A second focus of this proposal is to use our empirical model to extend our knowledge on decision making under uncertainty by testing competing theories on this subject. Prior research relies on observational data from financial markets, gambling behavior, and laboratory experiments over monetary gambles. Our proposed research is the first to test competing models of decision-making under uncertainty using data on tradeoffs between money (wages) and physical risks.
Worker safety is an ongoing concern in construction. While there are studies on hand injuries in the general working population, few have focused specifically on the construction workforce. A review of the literature thus far shows that hand injuries occur at high frequencies and have higher incidence rates than expected. Further research is warranted to determine the rates of hand injuries in construction overall and within its subcategories (residential, commercial, highway), by age, by occupation, by length of employment, and by ethnicity. This proposal will identify areas where target research is needed by describing the rates of hand injuries in construction and by looking at medical costs and lost time from work. It will also target trade-specific (carpenters) rates of hand injuries to determine rates in high risk populations working in the industry where prevention and intervention efforts might be most effective. The Annual Survey of Occupational Injuries and Illnesses (SOII) and Current Population Survey (CPS) datasets will be used for the analyses of construction related hand injuries for the years 2003-2010. The SOII and CPS datasets will be used to determine rates of hand injuries in construction as a whole and within sub-categories of construction such as type of construction, age, ethnicity, job tenure, and type of occupation. A large dataset of 30,000 union carpenters from Washington State, from 1989 to 2008, will be examined which includes hours worked and information on workers compensation claims. This dataset will provide rates of hand injuries within this cohort, associated medical costs, and lost time from work. The National Electronic Injury Surveillance System (NEISS) dataset will be used to determine hand injuries within construction and within the subcategories of construction.
Last Modified Date: January 9, 2013