This post was written by Member Scholar Thomas O. McGarity and Senior Policy Analyst Matt Shudtz.
The Mercatus Center has recently published a report on OSHA that simply rehashes the same old discredited arguments that industry apologists in academia and think tanks have been making for thirty years. Not surprisingly, they reach the conclusion that voluntary compliance programs and worker education efforts are better uses of OSHA’s limited resources than rulemaking and enforcement.
The report contains no original research, and (with one exception) it relies exclusively on studies finding little or no correlation between OSHA activity and reductions in worker injures. At the same time, the report ignores much of the evidence tending to show OSHA regulations and enforcement are effective. The simple (and frustrating) fact of the matter is that it is almost impossible to design a study using available occupational injury statistics to measure with much confidence the extent to which enforcement of OSHA standards is or is not associated with a reduction in workplace injuries or deaths. It is therefore not surprising that the studies reach mixed results. The Mercatus report ignored some reports showing a positive correlation and belittled a recent study showing a highly positive correlation.
By law, the agency has reviewed a number of standards issued over the last forty years. The cotton dust standard virtually eliminated byssinosis, at a cost to industry far less than expected. The standards controlling exposure to ethylene oxide resulted in reduced risk to employees and lower-cost sterilizers available to employers, even as industrial production of the chemical increased. OSHA’s inspections have also been proven effective, with studies (among others, here, here, and here) indicating that injuries and standards violations decrease following the inspections – by as much as 50 percent.
When effectively employed by a deteremined agency, deterrence-based enforcement works. Even though OSHA inspects fewer than one percent of U.S. workplaces annually and penalties available under the law are paltry, the threat of OSHA enforcement still has a deterrent effect on employers. With more resources, higher potential penalties, and fewer loopholes for small farms and other businesses, OSHA could have a much larger impact on worker health and safety. Those changes will require an act of Congress, which the companies that support the Mercatus Center will vigorously resist.
At the heart of the Mercatus report is a simplistic case for focusing governmental occupational safety and health resources exclusively on improving workers’ knowledge of hazards. Certainly, education and training are good things and all employers should be responsible for providing them to workers. But it is impossible to give workers enough knowledge to assure complete protection. Think about the workers who died on the Deepwater Horizon or at the Texas City Refinery. In both cases, bad process safety management decisions led to massive explosions that killed workers who probably did not know of the risky decision making that led to their deaths. But would it be cost-effective or efficient to educate every worker about every aspect of the processes at those facilities that could have catastrophic impacts? Of course not. So education and training only go so far. Better adherence to process safety management rules, based on an expectation that those rules will be enforced, will do more to improve worker safety than Mercatus’s proposed education and outreach.
Fortunately, catastrophic incidents are rare. As Mercatus rightly notes, many more workers suffer illnesses caused by health hazards like toxic air contaminants. But again, while Mercatus proposes government-sponsored education and outreach, those are necessary but not sufficient elements of a good occupational health and safety strategy. For example, after dozens of workers were sickened by certain chemicals used in microwave popcorn manufacturing, employers eventually replaced those chemicals. But the best strategy for protecting workers requires employers to design processes, choose chemicals, and operate machinery with a goal of reducing risks – not just swapping out a known risk for a different risk.
Instead of providing OSHA with adequate resources to do its job and authorizing it to assess meaningful penalties, the Mercatus report advocates greater reliance on the incentives provided by workers compensation insurance, the civil justice system, and the free market. This is nothing new. Industry apologists in academia and conservative think tanks have been trotting out the arguments that fill the pages of the Mercatus report for thirty-five years. Two CPR scholars (McGarity and Shapiro) refuted them in an article published fifteen years ago, when the Contract with America Congress was considering massive changes to the Occupational Safety and Health Act that would have effectively turned it into the information and education institution that the Mercatus report advocates. Congress had the good sense to reject those arguments then, and it should reject them now.
As the Mercatus study recognizes, workers compensation insurance premiums are not performance-based for the small firms that make up some of the most hazardous workplaces in the country, and workers compensation premiums are only modestly based on performance even in larger firms. As premiums did begin to rise for larger firms, instead of making workplaces safer, they lobbied state legislatures to exclude workplace illnesses and injuries that were resulting in large awards and to reduce the amount allowed for any given award.
The tort system was tried in the last quarter of the nineteenth century, and by itself it failed miserably to protect workers. As the Mercatus report notes, the courts recognized three defenses (contributory negligence, assumption of the risk and the fellow servant rule) that made recovery from negligent employers virtually impossible. The Mercatus report advocates a return to those days with one significant modification – and in the wrong direction. Instead of placing the burden on the injured worker of proving that the employer was negligent, it would require the employee to prove that the employer was “grossly negligent,” a higher state of culpability that would otherwise justify an award of punitive damages in most jurisdictions. It is hard to imagine that such an expensive and debilitated compensation system would send a message to employers to invest in safety.
The free market has also been tried and found wanting. Unskilled workers are in no position to bargain up wages for risky jobs. Many of the riskiest jobs in America (roofers, fishermen, agricultural workers) pay the poorest wages. The Mercatus report’s analysis is in fact reminiscent of the astonishing observation of Frank Knight (a founder of the Chicago School of economics) that the difference between a laborer and his employer is that the laborer prefers to risk his safety while employer prefers to risk his capital. In any event, the recent horrific fire at a textile facility in Bangladesh (just over a century after a very similar tragedy at the Triangle Shirtwaist facility in New York inspired occupational safety legislation in that state) should make it clear beyond cavil that the free market does not provide adequate safety incentives.
What Mercatus has done, in essence, is to reverse the hierarchy of generally accepted approaches to ensuring workplace safety. Standard occupational health and safety practice involves an established order of preference for selecting the means to reduce occupational hazards. Employers focus first on the elimination of hazardous conditions and substitution of hazardous agents. When elimination or substitution is not feasible, employers are to develop engineering and administrative controls that reduce workers’ exposure to hazards enough to keep them safe and healthy. Only when elimination, substitution, engineering and administrative controls are insufficient should employers rely on personal protective equipment to protect workers. The key to all of this is that it is the employer’s responsibility to find and fix the hazards. Mercatus is suggesting that employers should only be responsible for describing hazards, then leaving it to workers to demand changes. As evidenced by the thousands of worker deaths and millions of injuries and illnesses suffered each year, the U.S. workforce simply does not have that kind of bargaining power.