Diversity management has become an increasingly important element of modern organizational life: approximately 90 percent of U.S. federal government agencies and subagencies have diversity management programs and each year organizations spend some $8 billion on diversity programming. As it has grown in prominence, public administration scholars have sought to understand what effect, if any, diversity management has on government functioning. In essence, these efforts put aside the social equity case for diversity management, and outcomes like equal opportunity and advancement, and test the business case: is diversity management associated with government performance? In fact, there is evidence that it is: agencies that manage diversity well are likely to also perform well. These studies, to some extent, legitimate the cost and attention devoted to managing diversity.

But they also raise an important question: what explains the success of diversity management efforts? Why do some employees see management as handling diversity better than others? Although nearly all federal agencies have diversity programs, they have varied success. For example, in a 2011 survey, the average employee at the Department of Education’s Office of Elementary and Secondary Education (OESE) was much less likely to agree that supervisors did a good job of managing diversity than the average employee at the Department of Treasury’s Office of the Inspector General (OIG). As the performance studies just mentioned would predict, the OIG performed considerably better than the OESE. Results like this suggest the importance of explaining variation in employees’ perceptions of diversity management. However, few studies have done so.

This paper contributes to the diversity management literature in two ways. First, using an open system model, it theorizes about what it means to successfully manage diversity and how organizations might do so. This model generates a variety of personnel- and management-based hypotheses. Second, using two levels of analysis, and cross-sectional and longitudinal data, it empirically investigates these hypotheses using U.S. federal employee survey and personnel data collected by the U.S. Office of Personnel Management.

The paper’s findings show that across the federal government there are considerable differences in how employees perceive diversity management: generally speaking, minorities and women are less content than whites and men with managers’ diversity efforts. However, subagencies appear to be able to counteract this trend: women and minorities in subagencies with robust diversity programming and just procedures perceive that diversity is managed well. The paper’s longitudinal analysis shows that traditions of diversity management within organizations are strongly inertial over time: an organization that succeeds at managing diversity at time one is very likely to succeed at time two. As such, it may be difficult for managers to alter their employees’ perceptions of diversity management. However, despite this inertia, the analysis reveals that subagencies can improve how diversity is managed over time by creating more robust diversity programming and managing more fairly. The paper concludes by considering the implications of these findings.