Please use this identifier to cite or link to this item:
Czajkowski, Mikołaj
Sobolewski, Maciej
Year of Publication: 
Series/Report no.: 
27th European Regional Conference of the International Telecommunications Society (ITS), Cambridge, United Kingdom, 7th - 9th September 2016
Recent theoretical models of network competition with call externalities demonstrate strategic incentives of incumbent providers to reduce receiver benefits in rival network by excessive off-net pricing. Such anti-competitive pricing practices have a potentially damaging impact on financial standing of a late entrant, leading to non-convergence of long-run market shares – an outcome that has been observed in many European mobile markets. The theoretical reasoning behind call externalities assumes that receiving calls contribute to consumer utility hence, receiver benefits drive subscription choices. So far no attempts have been made to test this critical assumption in a rigorous manner. We use data elicited from prepaid and postpaid users of mobile telephony in Poland in a discrete choice experiment designed specifically to model subscription choices when operators set termination-based discriminatory tariffs under calling party pays regime. Receiver benefits are controlled with an incoming price – a variable informing about the cost of off-net calls paid by subscribers originating a call from other networks. The model also accounts for switching costs and network effects. We find that call externalities are significant driver of subscription choices, albeit their influence has smaller magnitude than direct price effects. Next, we assess the impact of excessive off-net pricing on the structure of market shares of mobile operators in Poland and estimate customer base stealing effect encountered by the late entrant. Our empirical findings support a widespread view that call externalities might have indeed limited market competition and late entrants' growth in many European countries.
Call externalities
personal network effects
switching costs
mobile telephony
stated preference
discrete choice experiment
random parameters logit model
Document Type: 
Conference Paper

Files in This Item:

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.