Consumers' Dynamic Switching Decisions in the Cellular Service Industry

This paper develops an empirical framework to analyze consumer's dynamicswitching decision in the cellular service industry. It firstincorporates the sequential problem of quantity, plan and firmsubscription choice in the presence of switching costs into a dynamicstructural model, which allows for fully heterogeneous consumers andmultiple switching possibilities across networks. The model is estimatedusing the data set on the number of switching consumers and theevolution of observed plan/firm characteristics over time. Based on theBLP-style estimation methods, we combine a nested technique that usesparametric assumptions with the structural estimation algorithm. Themagnitude of switching costs is estimated and it turns out thatswitching costs vary across networks. A dynamic model with restrictednumber of switching is likely to underestimate the switching costs.Lower switching costs encourage consumers to switch relatively early.Change in the variety of optional plans and plan characteristics alsoplay a great role in the consumers' switching decision.