The effect of overspending on tariff choices and customer churn: Evidence from mobile plan choices

Understanding the relation between consumer overspending, tariff choices and customer churn is extremely important in many industries. If consumers are insensitive to their overspending and keep making mistakes in tariff choices, firms may profit from overspending and do not have to worry about customer churn. In this paper, using a rich panel data from a major mobile network operator in China, we show evidence that consumers respond to financial incentives and try to reduce overspending by switching to another plan or terminating the relationships with the operator. Given the trade-off between profiting from overspending and reducing customer churn faced by the operator, we conduct a simple policy simulation and show that the operator should encourage consumers to reduce their overspending. We also show that the effects of switching decisions on future overspending and the hazard of churn differ among upward switchers (i.e. consumers who switch to a larger plan) and downward switchers (i.e., consumers who switch to a smaller plan). Specifically, we find that consumers who switch down are more likely to reduce their future overspending, while consumers who switch up are less likely to churn.

Automated summary

Understanding the relationship between consumer overspending, tariff choices, and customer churn is crucial in various industries. This study, based on panel data from a major mobile network operator in China, demonstrates that consumers respond to financial incentives by attempting to reduce overspending through switching tariffs or terminating their relationships with the operator. The findings suggest that operators should encourage consumers to reduce overspending in order to balance profit and customer retention. Moreover, the effects of switching decisions differ for upward and downward switchers, with downward switchers more likely to reduce future overspending, and upward switchers less likely to churn.