The Business Ethics Blog

Diversity and equality of opportunity are good things. And discrimination, on the other hand, is both morally repugnant and economically foolhardy. And yet it persists.

So how on earth could programs designed to encourage diversity and opportunity and to discourage discrimination be a bad thing? That’s exactly the question asked and answered by Harvard prof Frank Dobbin and Alexandra Kalev of Tel Aviv University, in research summarized in their Harvard Business Review piece, Why Diversity Programs Fail.

The goal of diversity programs is a laudable one, namely to increase diversity as a way of fighting back against systemic discrimination. The corporate world is in many ways still a male world, and a white male world at that. In spite of advances, women and minority groups still make up disproportionately small proportion of managers at big companies. If change is coming, it is coming painfully slowly. As Dobbin and Kalev point out, “Black men have barely gained ground in corporate management since 1985. White women haven’t progressed since 2000.” And it’s not for lack of qualified management candidates. “[B]oth groups,” the authors point out, “have made huge educational gains over the past two generations.”

So it’s easy to see why some might think that good intentions aren’t enough, and that proactive diversity programs would be a useful thing. Except they aren’t. For evidence, Dobbin and Kalev looked at a range of programs designed to encourage diversity—including diversity training, formal grievance procedures, as well as hiring tests and performance rating systems—and their conclusion about them is resoundingly negative. The authors reach this conclusion based on literally thousands of academic studies that have found, time after time, that diversity programs not only don’t work, they tend to be counterproductive. At companies that have instituted them, diversity has typically actually been reduced.

Why don’t such programs work? Dobbin and Kalev suggest three problems. One is that such programs tend to be negative, focusing for example on legal implications. If we’re caught discriminating, we could be sued! People tend to react badly such reasoning. Second, some companies make diversity training courses compulsory, and employees tend to result compulsory training, and then (so the hypothesis goes) blame the very disadvantaged groups the programs were aiming to help. Finally, Dobbin and Kalev hypothesize that when managers see such programs instituted, they feel blamed, and react badly to that. The result in all three cases is the potential for backlash and for managers to find end-runs around programs they don’t like.

So why do big companies persist in using such programs? Dobbin and Kalev point to fear of legal liability. That is, managers need to look like they’re doing something, even if there’s no evidence that that “something” really works. It’s very much like a physician ordering extra, unnecessary diagnostic tests. The only thing that seems worse than doing it would be not doing it and then having trouble surface later.

The fact that such programs don’t work is further evidence for the truism that management, pretty generally, is a difficult task. Coordinating and motivating people to work together to achieve a goal—whether the goal is increased sales or increased diversity—is not easy. More specifically, it’s an example of the principle that the best way to institute change isn’t always the most straightforward-seeming way, which is to exert direct control by telling people what to do. As lawyer and legal scholar Scott Killingsworth argues, “command-and-control” approaches to compliance come with a number of inherent limitations and adverse side effects. When command-and-control doesn’t work, the better route is through the long, slow road of cultural change.

With regard to diversity, what does work? Dobbin and Kalev recommend three broad strategies, none of which focuses on control. First, they suggest that companies “engage managers in solving the problem.” For example, get them to act as mentors to people in disadvantaged groups, and get them personally involved in for example recruiting a wider range of diverse job candidates. The second strategy is to make use of what psychologists call the “mere exposure effect,” a psychological mechanism according to which merely being exposed to a person, idea, or group tends to result in positive feelings about them. So, expose employees to people from different groups (for example by having them work together on diverse, self-managing teams). Finally, Dobbin and Kalev suggest making managers feel personally accountable for change. Not accountable in a legalistic way; accountable in a social way that comes with the feeling that people around you area aware of your behaviour. To this end, the authors recommend department-level transparency about stats regarding who gets hired and who gets promoted, and the institution of diversity task-forces with members drawn from various departments.

Perhaps most fundamentally, Dobbin and Kalev recommend that it be made clear, within the organizations, that top managers are paying attention to the issue of diversity. That is, what matters is not that the boss is telling you what to do; what matters is that the boss cares, and cares enough to pay personal attention.