Evaluating Permanent Disability Ratings Using Empirical Data on Earnings Losses

Abstract

ABSTRACT Workers' compensation systems are typically designed to assign higher permanent disability benefits to workers with more severe disabilities. However, little or no scientific work exists to guide the design of ratings systems to properly account for the amount of earnings power lost due to disability. In this article, we examine the effectiveness of disability ratings using matched administrative data on ratings and earnings for a large, representative sample of permanent disability claimants in California. We find that while workers with higher ratings do experience larger earnings losses on average, there are large and persistent differences in average earnings losses for similarly rated impairments in different parts of the body. We then explore how adjusting permanent disability ratings to reflect cross-impairment differences in earnings losses can affect the equity of permanent disability benefits. Adjusting disability ratings to account for typical earnings losses reduces cross-impairment differences substantially. The adjusted ratings result in a more equitable distribution of disability benefits across workers with different impairments. INTRODUCTION Individuals disabled by workplace injuries or illnesses are usually eligible for permanent disability (PD) benefits from workers' compensation. Estimates suggest that there are over 500,000 PD claims nationwide per year (Blum and Burton, 2003). These workers suffer from a wide range of injuries of varying severity, and determining an equitable scheme to compensate these workers is one of the most daunting policy challenges facing any workers' compensation system. Ideally, an equitable PD ratings system should be structured so that, when it is used to determine the level of PD benefits, more compensation is provided to workers with impairments that cause greater earnings losses (Berkowitz and Burton, 1987). However, basing PD benefit payments directly upon the actual earnings losses suffered by disabled workers creates work disincentives (cf. Krueger, 1990; Butler, 1994), so most states base benefits at least partially upon physician evaluations of the severity of impairment. (1) The main purpose of PD ratings scales is to convert medical information into a form that permits the necessary comparisons of impairment severity across and within disabilities occurring in different parts of the body. However, the extent to which these scales actually produce equitable results in practice is an open question. In this article, we use matched administrative data on the disability ratings and earnings of a large sample of PD claimants in California to assess the effectiveness of PD ratings. We study the extent to which workers with similarly rated injuries suffer similar postinjury outcomes. Furthermore, we examine the extent to which incorporating information about predicted earnings losses into the disability rating system can improve the equity of compensation for PD. An important limitation that nearly all current disability ratings systems face is a lack of scientifically validated data measuring the relationship between impairment, disability, and earnings loss (cf. Spieler et al., 2000; Cocchiarella, Turk, and Andersson, 2000). In the absence of such data, the current assignment of disability benefits is based on an implicit, unproven assumption that workers with greater medically detectible signs of impairment will also suffer more severe earnings losses. In fact, past work has cast some doubt on the ability of disability ratings systems to accurately assess the impact of different injuries on earnings capacity. Three 1968 studies that focused on Wisconsin, Florida, and California exhibited mixed results about the relationship between earnings losses and disability ratings (Berkowitz and Burton, 1987). Park and Butler (2000) use data from Minnesota to show that impairment ratings are able to explain only a small fraction of earnings losses subsequent to a disability. …