A Structural Estimation Approach to Study Agent Attrition

Abstract

Worker attrition is a costly and operationally disruptive challenge throughout the world. Although large bodies of research have documented drivers of attrition and the operational consequences of attrition, managers still lack an integrated approach to understanding attrition and making decisions to address it on a forward-going basis. To fill this need, we build a structural model that both captures the firm’s decision to terminate a worker’s employment (involuntary attrition) and uses an optimal stopping problem process to model a worker’s decision to leave the firm (voluntary attrition). We then estimate the parameters of the model and conduct counterfactual analyses on the population of 1,118 agents serving one client over 3 years for an Indian business process management company. Our model reveals a number of interesting findings. We find that supervisors have a strong impact on whether employees stay because they reshape the way that agents make their decisions. We also find that the impact of supervisors on agent attrition is more significant than the impact of salary. For example, increasing salary by 20% decreases the total attrition level by 5%. However, if agents were managed by the best supervisors, among those that manage similar agents, the attrition rate decreases by 10%. Altogether, our paper contributes to the burgeoning literature on people operations and managerial practice. This paper was accepted by Serguei Netessine, operations management.