What makes the workplace safe

Less than a month ago, Bloomberg BNA published a man named David Michaels as saying that “employers recognize that managing safety is useful not only to prevent injuries and fatalities, but in fact leads to a more profitable company.” Although all of this is true, the implication behind the statement of Mr. Michaels, the head of the Labor Department’s Occupational Health and Safety Administration (OSHA)—that such safety is best brought about by state action—is a bit more problematic.

If it’s true that state intervention is a necessary precondition for workplace safety, then there are a couple of things we would expect to see. In fact, the first thing we would probably want to do would be to look at the number of workplace injuries before and after organizations such as the Occupational Health and Safety Administration (OSHA) came into existence. In the pre-OSHA period for which National Safety Council data is available (1928-1970), workplace fatalities fell from 15 for every 100,000 workers to 6.8 per 100,000, a drop of 55 percent. Since OSHA came into existence, workplace fatalities have continued to drop at almost the exact same rate, down as low as 3.5 per 100,000 in 1993 (a drop of 49 percent from the 1970 benchmark). These data suggest OSHA has had little impact on safety in the workplace, and that it cannot reasonably be given credit for the improvements that have been made during its tenure.

Even if OSHA is not responsible for the dramatic decline in workplace fatalities in America over the course of the 20th century, it may have been superior to voluntary market arrangements in a number of other ways. It’s possible, for instance, that OSHA has helped to advocate for workers laboring under dangerous conditions in a way that is not reflected by fatality, injury and illness statistics (the latter two of which have stagnated for the duration of OSHA’s existence). Perhaps OSHA has provided an otherwise non-existent check on dangerous employer practices?

Again, say the data, not likely. Although OSHA did manage to exact $149 million in penalties for safety violations in 2002, workers’ compensation premiums from just a year earlier absolutely shatter that figure by more than two orders of magnitude, at $26 billion. In fact, 2004 estimates of risk-based wage premiums—the extra bump in wages that employers have to provide in order to get laborers to take on riskier jobs, and the “purest” market solution to the problem of danger in the workplace—obliterate even that figure by an order of magnitude, at an astonishing $245 billion. It seems that, at least in fiscal terms, OSHA lands firmly and by a large margin in second place again.

Okay, but what about awareness? A lot of market solutions work in theory, but they assume that workers (or women, or consumers, and on and on) are fully rational and have full information, don’t they? And, since that’s not the case, they must fall apart when applied to the real world.

Aside from the sheer and utter absurdity of the claim that markets require such unrealistic and meaningless conditions in order to fully function, it also turns out that laborers are smarter than supporters of organizations such as OSHA tend to give them credit for. According to one study of 496 workers, the proportion of workers in any given industry who identified their environment as dangerous was correlated with risk of injury in said industry. In high-risk industries like meat production and logging, in which 40 or more disabling injuries occur for every million hours, a total of 100 percent of employees identified their environment as dangerous. And, “of all workers who quit manufacturing jobs, more than one-third do so when they discover that the hazards are greater than they initially believed.”

If all of this is true—and it is—then why is it that the voluntary employment behaviors of both employers and employees alike get such a bad rep? Why is it that employers—experts, after all, on employing people—can’t be trusted to employ people, and employees—also experts, on their own employment—can’t be trusted to be employed? Why is it that the relationship between two people can only be allowed to exist insofar as it is approved by a third party?

A small part of the reason is that market detractors are, well, correct: Markets are sub-optimal, employees aren’t always totally informed and workplace injuries do occur. A larger part of the reason, however, is that detractors often wind up comparing these truths to an ideal situation, in which all needs are always met, all employees are totally taken care of, and no one is ever hurt on the job, rather than the reality of middling non-factors such as OSHA. This is known as the “Nirvana fallacy,” and it tends to come up a lot—admittedly, in both directions—in discussions of markets and state intervention.