Abstract
The paper examines the effect of the Performance-Based Funding (PBF) on the performance of health facilities in North Benin, within the context of lack of control group. To account for the lack of control group, we base on the well accepted theory arguing that the PBF affects performance through the process and the structure settled in the health facility to develop a strategy to access the effect of PBF: mediation variables approach. We consider three performance variables (less than 5 years old children completely vaccinated, family planning and delivery) measured in four different ways. Using data from a survey on 55 health facilities observed over 14 trimesters, we find that the PBF has no effect on performance variables regardless of how they are measured. However, the program has some partial mediation effects that sometimes cancel each other out. We also find that a process that is designed for a given performance variable rather affects another performance variable.