The effect of consumer switching costs on market power of cable television providers

This paper empirically evaluates the effect of consumer switching costs on market power of cable television providers. To do so I propose an algorithm to estimate supply side parameters when the demand side is represented by consumers with persistent heterogeneity in tastes and state-dependent utility. Under such conditions, regardless of whether consumers are forward-looking or behave in a myopic way, the demand schedule introduces dynamics into the producer problem. Presence of multiple state variables makes full solution to the producer dynamic programming problem difficult or computationally infeasible. Therefore, I estimate supply side parameters from the optimality conditions for the dynamic controls. Using data from the paid television markets in US, I estimate cost functions of the cable television providers. In these markets, persistent consumer heterogeneity and state dependence due to switching costs requires producers to keep track of the entire distribution of consumer types which enters states space of a dynamic programming problem. Preliminary estimates of the consumer switching costs are $149 and $238 for cable and satellite providers respectively. The estimates of the supply side parameters imply average cost of providing service per subscriber of $2.19 and average per subscriber price-cost margin of $16.52. Counterfactual simulations suggest that without switching costs cable prices would be 28 percent higher with satellite competitors and 51 percent higher in case of static monopoly scenario.