The Paradoxical Effects of Diversity

Is diversity good for business performance? Well, it depends.

That's the conclusion reached by the Diversity Research Network, a group of scholars drawn from six universities who recently completed the largest field-based study of this topic to date. “The Effects of Diversity on Business Performance: Report of the Diversity Research Network,” appears in the Spring 2003 issue of Human Resource Management. After examining four Fortune 500 companies in depth, the researchers found that a variety of contextual variables, including an organization's culture, strategy and human resource practices, helps to determine whether diversity boosts performance or drags it down.

In the first case study, which involved an information-processing company with more than 26,000 employees, professor Karen Jehn and lecturer Katerina Bezru-kova of the University of Pennsylvania's Wharton School found that racial diversity was associated with poor performance in business units that shared three traits: a competitive organizational culture, a growth-oriented business strategy and human resource practices that focused on honing job-related skills. However, this relationship disappeared when groups promoted collective achievement, emphasized stability or customer relationships rather than growth, and used training opportunities to impart diversity-related values to employees.

Similarly, an analysis of a large financial services firm, conducted by associate professor Robin Ely and professor David Thomas, both of the Harvard Business School, suggested that how groups approach diversity is at least as important as the degree of diversity they display. On the basis of employee surveys and sales and customer satisfaction data collected from 480 of the company's retail branches, Ely and Thomas concluded that racial diversity enhanced performance in units that treated diversity as a resource for innovation and learning. But diversity typically had a negative impact in groups that showed no evidence of what Ely and Thomas term the “integration-and-learning perspective.”

These effects were substantial. When Ely and Thomas focused on branch offices with equal numbers of white and nonwhite employees, they found that on average, groups that appeared to endorse the integration-and-learning approach outperformed units lacking this perspective by 37%. What's more, the same high scorers outperformed racially homogeneous branches by a margin of nearly 30%. (In all cases, performance was measured against branch-specific goals, which allowed the researchers to control for local factors that might affect sales and customer satisfaction, such as the wealth of the surrounding community.)

The third study, by Susan Jackson, professor of human resource management at Rutgers University, and Aparna Joshi, assistant professor at the Institute of Labor and Industrial Relations, University of Illinois at Urbana-Champaign, identified team leaders as a further influence on the relationship between diversity and performance. After analyzing sales teams at another large information-processing firm, they determined that ethnic diversity was associated with superior performance when teams were led by minority-group members, but not when managers were white. However, the opposite pattern appeared when diversity was defined in terms of gender, rather than ethnicity or race. Greater diversity, which in this case meant a higher proportion of female employees, was associated with poor performance for teams led by women, but not for teams led by men.

The broader finding — that gender and racial diversity can have different effects —received support in the first two studies as well. For example, Jehn and Bezrukova reported that gender diversity had a positive impact on group processes and consequently on performance, while racial diversity had negative effects — unless accompanied by intensive training in career development and diversity management.

Taken together, says Thomas Kochan of MIT's Sloan School of Management, coordinator of the multiyear project, these results underline the importance of developing a more nuanced understanding of how different forms of diversity affect how people work. In addition, they suggest the need to move beyond the simple version of the “business case” for diversity. Proponents of this view, which achieved prominence in the 1990s, contend that diversity benefits businesses by enhancing team performance and increasing their ability to serve a diverse customer base. But as these analyses show, when it comes to team output, diversity can cut both ways. Moreover, according to the last of the project's four studies, simply matching a company's workforce to its market is unlikely to increase the odds of success. After analyzing data from a national retail chain, David Levine and Jonathan Leonard, both professors at the Haas School of Business, University of California, Berkeley, and Aparna Joshi found no evidence that most customers cared whether or not they were served by people of the same gender or race.

Consequently, the Diversity Research Network members conclude, it is time to recast the terms of the diversity debate. There is little point in continuing to ask whether diversity's impact is naturally good or bad. Instead, managers and researchers alike should recognize that diversity has become an inescapable social fact and figure out how to maximize its benefits while minimizing its negative effects. Companies can work toward this goal by supporting internal experimentation and evaluation, while human resource managers should focus on developing analytical tools to pin down the links between specific programs and business performance.

Few organizations are currently up to this challenge, according to MIT's Kochan, although a small group of companies with a deep commitment to diversity (including the four case study subjects) may be ahead of the curve. “Managers generally don't think about diversity in the same analytic way that they think about investments in finance, new products, technologies, or even increasingly, human capital,” he explains. As a result, companies often waste money on misguided initiatives that are useless or even worse. “Their intentions are in the right direction, but they haven't taken the next step of allocating their resources in the way that's going to get the maximum payoff,” Kochan observes. “They don't even know if they're allocating their resources effectively, so there's a very, very long way to go.”