Association for Public Policy Analysis & Management

The theory of representative bureaucracy posits that when bureaucrats share demographic or social identity with citizens, they are more likely to act in ways that benefit minority groups (Meier and Nigro 1976; Riccucci and Meyers, 2004).Empirical research looking at the transformation of passive representation into active representation is mostly applied within the United States context and few policy areas (e.g. education, law enforcement and welfare policy areas). Beyond limited application of the theory in policy areas relevant to governments and citizens, the existing literature lacks clarity on, if representation has differential effect on bureaucrats and citizens and on different citizen groups.

To explore the relationship between representative bureaucracy and organizational outcomes in a socio-cultural context that is different than the United States, this study applies it in a country that is a part of the Sub-Sharan Africa region. In addition, the study applies the theory in a much less explored policy area – public health policy. The study applies the theory in Tanzania and within the context of family planning service delivery.

The objective of this paper is three-fold. First, by applying the theory in the Sub-Sharan Africa region, it aims to uncover effects of representation (gender matching between clients and providers) on organizational outcomes (family planning service utilization). Here, the study intends to learn if outcomes of representation will be different when applied within this specific policy and socio-cultural contexts. Second, the study aims to uncover if representation (gender matching) has similar or different effects on bureaucrats’ and citizens’ behavior. Last, it aims to explore if the effects of representation are similar for different groups (first-time service users and non-first-time service users). The aforementioned objectives are tested using a nationally representative Demographic and Health Survey (DHS) data with specific information on Service Provision Assessment (SPA). A series of Logistic regression analysis indicate that representation (gender matching) is negatively associated with first-time service users’ (clients’) decision to adopt a family planning method. For the same group, the study also reports that bureaucratic behavior is positively associated with clients’ (citizens’) decision to adopt family planning method. In the contrary to the effect of representation (gender matching) on clients’ outcome, the analysis reports that representation is positively associated with bureaucrats’ behavior. The findings suggest that for this particular population, representation has negative effect on citizens’ outcome but a positive effect on bureaucratic behavior. It also suggests that representation has a positive effect on one group (first-time service users) but shows no statistically significant effect on the other (non-first-time service users).