Quality Competition with Stochastic Demand and Costly Search: Theory and Evidence from the Video Rental Industry

<p> A theoretical model is developed to predict optimal service rates in markets where firms compete in availability. We show that firms are more likely to stock&hyphen;out of popular products as the cost of consumer search increases. Carlton (1978) showed that, in a zero&hyphen;profit competitive environment, firms balance the risk of not being able to serve a particular customer against the cost of holding excess capacity and that this balancing act will result in an equilibrium in which not all customers are served. The model was later adapted to oligopolistic competition by Peters (1984) and Deneckere and Peck (1995). This paper extends this literature on competition under stochastic demand by developing a model that incorporates 1) the possibility that customers may be able to purchase from another firm in the case of a stock&hyphen;out and 2) the option for firms to offer an imperfect substitute in order to persuade some customers to make a purchase when the first choice product is out of stock. Empirical evidence is presented in support of the theoretical model using data collected from video rental outlets in a midsize southeastern US city.</p>