Academic journal article Journal of Risk and Insurance

Claiming Behavior in Workers' Compensation

Academic journal article Journal of Risk and Insurance

Claiming Behavior in Workers' Compensation

Article excerpt

ABSTRACT

Using administrative data on workers' compensation claims in Michigan combined with data collected from a sample of workers identified by physicians as having work-related pain in their backs, wrists, hands, or shoulders, this article provides evidence that a substantial number of potentially eligible workers do not file workers' compensation claims. Multivariate analysis identifies the effects of various factors on the probability of filing a workers' compensation claim, conditional on having a work-related health problem. We find that the severity of the worker's condition and the worker's general health are the most important determinants of the decision to file, and that the generosity of wage loss benefits also affects the decision of an eligible worker to file. Finally, claims propensities vary considerably across workplaces, holding all other measured factors constant.

INTRODUCTION

The broad objectives of workers' compensation are straightforward: to provide medical care and wage-replacement benefits to employees injured or made ill in the course of work. However, the availability of benefits provides a less straightforward set of behavioral incentives to employees. Because wage-replacement benefits essentially compensate individuals for not working, employees have an incentive to exaggerate the severity of existing injuries, miss more work than necessary, and/or to inaccurately attribute an injury to work. (1) This risk is termed "moral hazard" and arises out of information asymmetries, where the worker has full knowledge of the cause and severity of his/her injury but the employer does not. (2)

Typically, the presence of moral hazard has been tested by examining the change in claims rates in response to increases in benefits. In general, the research has supported that an increase in benefits leads to an increase in claims (Meyer, Viscusi, and Durbin, 1995; Kreuger, 1990). (3) One drawback to the moral hazard research, however, has been an inability to distinguish between injury and claims rates, so that one cannot tell whether claims increased because of some underlying shift in work hazardousness or because employees were taking advantage of more generous benefits. The purpose of this article is to examine claiming behavior independently of injury effects.

There is a substantial empirical literature that explores the implications of moral hazard, in particular the question of whether changes in the generosity of benefits affect various measures of workplace safety or program usage. Much of the research has been based on data on injury rates or workers' compensation claims rates at the state, industry, or occupational level, although more recent studies have made use of individual-level data. (4) This literature has identified a number of factors that influence injury rates and the number of claims filed, and offers fairly compelling evidence of the existence of a relationship between the generosity of benefits and reported injury or claims rates. However, the existing data sources do not allow comparisons between the number of workers actually injured and the number of workers' compensation claims filed.

In this article we combine administrative data on workers' compensation claims in Michigan with data collected from a sample of workers who were identified by physicians as having work-related pain in their backs, wrists, hands, or shoulders. (5) We present evidence that a substantial number of workers who have been independently identified as having these work-related injuries do not file workers' compensation claims. We use a multivariate analysis to estimate the effects of various factors on the probability that workers with such work-related ailments file for wage-loss benefits. Specifically, we examine four possible effects. Because workers' compensation is designed to offer benefits during periods of ill health, we examine the effect of health and injury severity on claim probability. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.