Abstract
We model competition between two firms selling identical goods to customers who arrive in the market stochastically. Shoppers choose where to purchase based upon both price and the time cost associated with waiting for service. One seller provides two separate queues, each with its own server, while the other seller has a single queue and server. We explore the market impact of the multi-server seller engaging in waiting cost-based-price discrimination by charging a premium for express checkout. Specifically, we analyze this situation computationally and through the use of controlled laboratory experiments. We find that this form of price discrimination is harmful to sellers and beneficial to consumers. When the two-queue seller offers express checkout for impatient customers, the single queue seller focuses on the patient shoppers thereby driving down prices and profits while increasing consumer surplus.
Highlights
In many markets there are more customers desiring attention than can be accommodated at one time, which means that some customers have to wait, be it in line to checkout at the grocery, on hold with technical support, or for a table at a restaurant
We find that when multi-line sellers are prohibited from engaging in price based express checkout, prices and profits are higher than when sellers can engage in this form of price discrimination
The observed behavior is in line with previous Bertrand competition experiments, which have found that sellers are typically competitive, but rarely so competitive that they push prices all the way down to the predicted level. When we began this project, we were struck by the puzzling rarity of time-cost based price discrimination in situations where customers have to physically queue for service in competitive markets
Summary
In many markets there are more customers desiring attention than can be accommodated at one time, which means that some customers have to wait, be it in line to checkout at the grocery, on hold with technical support, or for a table at a restaurant. If individuals value their time differently, there is a socially optimal allocation of scarce queue slots, and traditional first come first serve queuing will be inefficient. These arguments are not unique to service industries; users of communication and transportation networks have heterogeneous waiting costs as certain packets of information and cargo are more time-sensitive than others, and all queuing systems that ignore preference heterogeneity are inefficient
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
