What Works: Gender Equality by Design

Iris Bohnet

385 pages, Harvard University Press, 2016

Buy the book »

Yes we can. That is the most significant takeaway from What Works: Gender Equality by Design, Iris Bohnet’s eminently readable book on design strategies to reduce gender inequality. Bohnet, a behavioral economist who directs the Harvard Kennedy School’s Women and Public Policy Program, encourages us to resist the temptation to say, “Oh, but that couldn’t possibly apply to me,” and consider the possibility that we can all contribute to creating a more inclusive and equal society. She argues that it’s easier than we might think both to acknowledge our inherent biases and to act to diminish their effect using evidence-based approaches.

The book’s pithy, practical chapters are firmly grounded in data. First, Bohnet walks us through the problem of unconscious bias that affects every one of us. Yes, that includes well-meaning, self-styled progressives like me who have led feminist organizations! Early on, she recommends that we assess our assumptions using tools like implicit association tests, short quizzes that show we all make unconscious, nearly instantaneous judgments. Such biases help us understand why, as Bohnet puts it, “When performance is observable, successful women are rated as less likable than men,” but, “When performance is ambiguous, successful women are rated as less competent than men.” A sobering finding, particularly in a year that may offer the US electorate, for the first time, a choice in a general presidential election between a male candidate from one major party and a female candidate from the other.

Although the book focuses on gender-based structural bias, and even more specifically on distinctions between those who self-identify as men and women, it draws on examples of structural discrimination against many other marginalized groups, from African Americans in the United States to so-called untouchables in India. Bohnet explains, for instance, why traditional diversity training programs often backfire owing to the “halo” effect, a form of moral licensing. In essence, people who believe they already have made significant efforts to be tolerant, by attending a diversity workshop or rooming with a person of another race, often then feel entitled to continue being just as biased against those they consider “other,” or even more so. One dismaying study Bohnet cites found that in 2008 people who said they would vote for Obama were then more likely to discriminate against African Americans.

Colleagues of mine belonging to “minority” groups have raised related concerns in conversations about how to advance race and gender equity. They worry that diversity work could lead to a backlash of sorts in a field where they are just beginning to gain a foothold— fears that are clearly not unfounded. Indeed, as a philanthropic leader currently helping the Ford Foundation take its work on reducing inequality to the next level, I found myself wishing that Bohnet more explicitly addressed issues of intersectionality. This term, coined by the global feminist movement, reminds us that different kinds of identity-based structural bias, such as racism and sexism, do not function in a vacuum but rather interact and affect one another. Checking for and assessing implicit bias along just one dimension is usually inadequate.

Regardless of this missed opportunity, Bohnet offers useful suggestions for walking the talk on equality. One chapter focuses on the need to “collect, track, and analyze data to understand patterns and trends and make forecasts” regarding personnel. It turns out that although workplaces are good at sorting production and consumption data, few are rigorous about analyzing data on their employees. You can’t make changes if you don’t really know how many women are in senior-level positions, or how many members of minority groups you hired last year. A superb section on changing interview procedures includes a simple model inspired by Atul Gawande’s now-famous checklist to improve health outcomes in hospitals. All questions prepared beforehand? Check. Interviewers organized to interview one-on-one rather than as group? Check. Rating matrix ready for completion post interview? Check. Having hired many people over the years, I was chastened to learn how easily many interview models favor groupthink and implicit bias.

Chapters on improving assessment tools, increasing transparency, and physically changing the workplace to ensure a more level playing field are equally eye-opening. Bohnet describes how, in the last ten years, Harvard University has changed its interior décor, having learned that buildings full of pictures of former leaders, all white men, diminish women’s ability to see themselves as successful leaders. As a woman who grew up in India under Indira Gandhi as prime minister, I can attest to the effect of visible cues in schools and workplaces. My “normal” was that a diminutive woman in a sari led the nation and was my naval officer father’s boss. And that simple reality—her portrait hanging in offices—helped me counter patriarchal prejudice that came my way.

Bohnet doesn’t waste time trying to impress us with academic jargon or pages of charts, although her book is meticulously footnoted and documented. Her straightforward tools for designing inclusive and diverse workplaces and institutions are a boon to anyone who hopes to study, work, and live in an atmosphere of openness and civility.