How common is sexual harassment?

FOR social scientists the apparent epidemic of propositioning, pinching, groping and flashing that is gripping America brings a rare opportunity to observe a new norm, around how men behave towards women, being created in real time. First though, they must figure out the extent of the problem. The most extreme example—rape—is hard enough to count. Government statistics produced by the Centres for Disease Control and Protection suggest that one in five women and one in 60 men have been a victim of rape or an attempted rape in their lifetime. On the other end of the scale, sexual harassment—a charge that rarely carries criminal punishment—is far more common, and harder still to count.

It is also a fairly new phenomenon, in the sense that there was no phrase that described it until the late-1970s. That coincided with women’s growing importance in the labour force as their share of jobs rose from 33% to 42% over the preceding two decades. A formal legal definition arrived in 1980 when the Equal Employment Opportunity Commission (EEOC), a federal agency, stated that unwelcome sexual advances which affected an individual’s work were grounds for a complaint.

The law was slow to take hold. Just 16 cases of sexual harassment were received between 1980 and 1985. But after a Supreme Court ruling in 1986 decided that sexual harassment was covered by the 1964 Civil Rights Act, and held companies liable for the harassment of employees even if they were not themselves aware of it, the number of EEOC harassment cases shot up. In the intervening 30 years there have been 400,000 complaints against companies. Even before the Harvey Weinstein story broke the dam, the number of cases had risen by 5% since 2014.

Yet since the majority of sexual harassment goes unreported, its true prevalence can only be guessed. A 2016 report by the EEOC provided a range of 25%-85% for the share of women who have faced sexual harassment at some point in their working lives. Victims tend to be young, junior and working in male-dominated industries. They are overwhelmingly women: just 17% of EEOC sexual harassment cases are brought by men. Once their relative workplace status and occupations are accounted for, non-white women are no more at risk of harassment than whites.

Training staff to prevent sexual harassment, which companies often do to lower their risk of being sued, has been ineffective. The most important single thing is for organisations to have a credible threat of sanction for the perpetrators. That also means protecting staff who speak out. Lilia Cortina, a professor of women’s studies at the University of Michigan, finds that this is rarely done well. To that end the EEOC has begun emphasising the role of what it calls “bystander intervention”—a technique shown to be effective in sexual assaults on university campuses.

The creation of new norms around acceptable behaviour is likely to prove more effective. And there is already evidence that attitudes to harassment are changing. YouGov, a pollster, asked Americans on behalf of The Economist about their views on acceptable physical contact between the sexes (see chart). Younger men are likely to be more respectful than older men in their behaviour towards women, and they are also more likely to agree with women about what is acceptable. On this subject at least, younger Americans are more conservative than their elders.