The return of newly manufactured items/materials has been estimated to cost over $35 billion in the US and $137 billion worldwide. The flow of returns is immediate from consumer side to the manufacturer side, a type of reverse logistics. However, our focus is not on end-of-life products. We present a strategic mixed integer programming model to find the most efficient network morphology. The model also determines the most profitable assignment of reverse logistics tasks to the supply chain players who may include original equipment manufacturers, contract manufacturers, and third party logistics service providers. We study the sensitivity of these decisions to changes in parameter values using problems faced by our industrial collaborators.