Employer responses to compensation pressures

The economic rationalist approach assumes that compensation pressures provide incentives for employers to improve their occupational health and safety performance and that employers respond accordingly. But this, as we have seen, is a problematic assumption. This chapter looks at how employers in fact respond to these pressures. In Chapter 4 we examine the limited reach of the compensation system and the way in which many types of employment and many health and safety risks are in principle beyond its influence. Chapter 5 looks at the broader safety pays approach and explores the argument that other economic considerations, for example productivity, can serve to focus employer attention on health and safety. 

In asking whether employers in fact respond to premium incentives by improving their OHS performance, our first resort is to the research literature. Experience-based premiums—that is, premiums which reflect the individual employer s compensation costs—are common in the United States for employers above a certain size. Establishments below this threshold undergo what is called manual rating in which premiums are simply set in accordance with the experience of the industry as a whole. There have been a number of studies which attempt to measure the impact of this system. 

The methodological basis of many of these is as follows. Benefit 

One important study in this tradition avoids this problem. Moore and Viscusi (1989) relate benefit levels not to lost-time injury claims but to fatality rates. The reporting of fatalities is presumably not affected by the level of benefits and consequently, they argue, any relationship can be assumed to reflect safety incentive effects rather than reporting effects. The authors in fact find a strong negative relationship between benefit (and hence premium) levels on the one hand and fatality rates on the other, thus lending support to the hypothesis that experience rating works to make employers more safety conscious. 

Two studies using a slightly different methodology have since suggested that both the hypothesised effects—increased claims reporting and decreased injury rates—are present in the lost-time injury claims data (Kniesner and Leeth 1989 Butler and WorraU 1991). However, in one other study which also avoids the interpretation problem discussed above, Chelius and Smith found that experience rating had no observable effect on employer behaviour (1983, p. 136). 

---