Academic

Scholars, particularly economists, are of two minds when it comes to employment regulations aimed at protecting vulnerable groups. The goal behind these laws, of course, is to equalize the status of groups seen as vulnerable to economic discrimination, such as women or the disabled. One prominent position amongst economists, however, is that these policies often are counterproductive, raising the cost of employing the relevant groups and thus imposing unneeded costs upon them. In a new paper with Jasmin Sethi, Kauffman Fellow Daniel Chen finds evidence that sexual harassment law does help reduce gender inequality in the workforce.

It's useful first to explore the mechanics of why theorists posit such laws might or might not operate to the benefit of the groups they're attempting to aid. One position holds that the laws, by imposing increased enforcement costs on firms, disincentivize them from hiring the group in question. For persons already hired, these laws act as forced benefits -- if a firm willing to spend X amount of resources on a given employee, the amount they are forced to expend to stay in compliance with specific laws or regulation will simply be taken away in other areas (salary, benefits, hours, etc.). This model also posits that the market should resolve invidious discrimination, because firms which don't discriminate should be able to seize a competitive advantage by exploiting the labor of otherwise superior workers irrationally excluded by their competitors.

With regards to sexual harassment, however, an alternative view emerges. Harassment can be seen as a form of "economic warfare" waged by insiders (i.e., men who already have jobs) against outsiders who might underbid them and thus deprive them of established benefits. By withdrawing their cooperation (i.e., harassing), the men can lower the productivity of the competing group, removing their competitive advantage and prompting firms to either refrain from hiring or dismissing women. The dissenting position explains why firms don't quash discrimination themselves by noting that the insiders may be risk-averse to any changes in the employment structure. It is also possible that the managers in a position to enforce an anti-harassment norm may also be the very "insiders" benefiting from it (indeed, there is research indicating that women in supervisory positions are more likely to face harassment than non-manager women).

To test these models, Chen and Sethi examined how labor markets shifted in response to sexual harassment decisions by federal courts, specifically, how firms responded to plaintiff "wins" at the appellate level (the authors use the idiosyncratic variation in pro plaintiff decisions provided by the random assignment of judges (of particular gender and party affiliations) to panels). Disaggregating the results amongst the federal circuits helped provide variation in results for comparison purposes. They found that sexual harassment law reduced gender inequality by 1.3-15% in the areas of employment status, hours worked, wages, and management status -- seemingly buttressing the dissident view and differentiating harassment law from other areas, such as disability and maternity benefits, where the dominant view had been borne out via empirical research.

Digging deeper into the results, at least two further findings stand out. First, the reduction in gender inequality was enjoyed predominantly by women who were not already in the workforce. This is consistent with the insider/outsider model forwarded above: that harassment law prevents entrenched actors in sex-segregated occupations from locking out newcomers. Reducing harassment prompts women who were previously outside the workforce to enter, but women who already have jobs see little change. Second, the results demonstrated that firms respond to victories by sexual harassment plaintiffs regardless of how large or small the damage award is. This indicates that the response by the firms was motivated less by a desire to avoid the monetary sanctions of a damage award, and more by reputation or other less tangible costs associated with the risk of being seen as a harassment-friendly workplace.