Syndicate content

Add new comment

A BAD Conference

Varun Gauri's picture

Last week, I attended a conference at the Rotman School of Management at the University of Toronto. It was BAD, and it was primarily about gender. (By BAD, I of course mean it was about “Behavioral Approaches to Diversity”.) The topic is obviously relevant to World Bank goals, both internally and for our clients, and to the work of the Mind, Behavior, and Development Unit (eMBeD). Here are some selected highlights.  

Alice Eagly of Northwestern University argued, using a series of US opinion surveys dating back several decades, that American stereotypes of men and women have changed in some respects. Stereotypes of men and women are now about equal when it comes to competence and intelligence. But it is still the case that men are seen as more “agentic” (assertive, aggressive, take charge), and women are seen as communal. The reason for this, Eagly argued, is that although women now have higher levels of education than men, occupational segregation still shunts women into the caring professions, which prolongs the stereotype of women as communal and men as agentic. The stereotype, in turn, plays an important role, she argued, in the preponderance of men in leadership positions because people want leaders who are not only competent but assertive and aggressive.  

How can behavioral science help gender diversity in organizations? Iris Bohnet presented a number of evidence-based approaches drawn from her book, What Works: Gender Equality by Design. For instance, the book argues that explicitly comparing real candidates forces evaluators to focus on individual performance instead of stereotypes, and that blinding the identities of applicants can level the playing field. What was most striking to me, however, was some new research about what not to do. Employers often include equal opportunity statements in job ads, but these do not work, and sometimes backfire, reducing job applications from minorities!  Still, Iris (who is now academic dean) could not bring herself to remove Harvard Kennedy School’s diversity and equal opportunity statements. Instead, she has opted for making them more personalized, explaining why Harvard cares about diversity. Let’s see if that works. Relatedly, Katy Milkman described a paper that finds that 42% of top public companies have exactly two women on their boards of directors, which is 45% more than you’d expect if the number of female directors were distributed by chance. It’s “twokenism,” an example of the dangers of checkboxes, and the power of descriptive social norms (since having two women on boards is the modal number among big US firms). 

Dolly Chugh addressed the problem of trying to be good, which she assesses in her new book The Person You Mean to Be: How Good People Fight Bias. She notes that when we aim to be good and fail (e.g., when someone tells us we have implicit biases, no matter how hard we’ve tried not to), we are justifiably disappointed, and sometimes angry. Instead of trying to be good, we should ask ourselves to be “goodish” - to recognize our bounded ethicality and reflect on it.  

Kathleen McLaughlin, Senior VP and Chief Sustainability at Walmart, described how Walmart uses its $300b in outsourcing as development funds. It has spent $25b of that money to promote women-owned businesses. It has also spent $1b on the training of women entrepreneurs. The latter has been subject to an impact evaluation led by the Tufts Labor Lab 

I presented eMBeD’s work, together with MENA, on measuring social norms related to women’s labor force participation in Jordan. We find that 60% of non-working women want to work, which means that social norms against women working are not deeply internalized, and the right interventions and leadership could make a difference. Jordanian men and women both overestimate the extent of opposition to women working, which suggests that media and information campaigns, perhaps soap operas, could correct that misimpression and weaken the social norms. We’ll report more on that and related studies soon. In the meantime, just remember eMBeD is looking to do more work like this because, of course, we also want to be BAD.