Bundle Pricing of Congested Services

Abstract

Bundle pricing is commonly adopted by service firms managing multiple congestion-prone service facilities. Under bundle pricing, the firm charges a single price for all the services as a whole, and customers either purchase the bundle or do not purchase at all. This scheme is in contrast to a la carte pricing, under which the firm sets a separate price for each service, and customers can choose which service to purchase, if any. The existing theory generally sees bundle pricing as more lucrative, especially when customers' valuations for different services are negatively dependent, pointing to bundling's ability to reduce customer valuation dispersion and thus enable surplus extraction. However, despite its prevalence in practice, little is known about bundle pricing in service systems that involve congestion-driven delay. We find that the prescriptive guidelines from the existing theory can be reversed in such congested service settings. Specifically, bundling fails to make more revenue than a la carte pricing when the market size is large or customers are highly delay-sensitive, regardless of whether customer valuations are independent or negatively dependent across services. Moreover, negative dependence of customer valuations may even further decrease bundling revenue.