Service Operations in the Presence of Strategic Consumer Behavior
- Author(s): Ashley, Katherine
- Advisor(s): Feldman, Pnina
- et al.
This dissertation analyzes pricing and information provision in the presence of strategic consumer behavior, with a focus on service industries. Using both analytical and empirical methods, I analyze two tools that firms may use to match supply with demand when customers are strategic: reservation no-show fees and inventory announcements.
First, I study a common problem in the service sector: reservation no-shows. Many firms allow customers to make reservations for service at a specified time in the future. Reservations are valuable to consumers because they insure against the service not being available, but are costly for the firm because valuable capacity may be held for customers who fail to show up. No-show fees, payable only if a reservation-holder fails to show up in the service period, are one tool that firms can use to mitigate the loss from wasted capacity. I analyze the performance of a no-show fee pricing scheme for a capacity-constrained service firm in a market with both forward-looking customers and a walk-in demand segment. I characterize three types of optimal policies that emerge in equilibrium and examine the social welfare implications of charging an optimal fee. In equilibrium, social welfare may strictly increase under a no-show fee policy; that is, the utility gain from using a no-show fee to allocate capacity more efficiently may be greater than the cost it imposes on consumers.
In the second half of the dissertation, I analyze the information content and sales impact of inventory announcements, i.e. messages such as ``3 seats left at this price.'' I use an expansive original dataset to study the impact of these announcements in the market for airline tickets. My data include itinerary listing and seat map information
for thousands of flights on three major U.S. carriers, with travel dates in the winter and spring of 2015. I measure the ``informativeness'' of announcements, which I define as their ability to improve customer predictions of future price changes, and estimate the impact that these messages have on ticket sales. I find that announcements contain significant information about future price changes (both direction and magnitude), and are most informative early in the booking horizon. To measure the customer response, I use an instrumental variables model to address potential endogeneity between sales and announcements. My findings provide evidence that for at least some airlines, consumer ticket purchases increase immediately after new inventory messages, a result that is consistent with rational consumer behavior.