Transaction vs. switching costs: Comparison of three core mechanisms for mobile markets

The fast growth in demand of mobile Internet urges mobile network operators (MNOs) to rapidly increase the wireless network capacity. For this purpose, governments are allocating large parts of the valuable low frequency spectrum to MNOs. This expansion also adds pressure to better optimize the intra-MNO and inter-MNO spectrum usage. Regulators are concerned about problems such as network blackouts, coverage disparities and congestion. Latest technology developments provide new mechanisms to address these problems through two alternative evolution paths, operator-driven and user-driven. The operator-driven path permits operators to trade network capacity and spectrum through, for instance, national roaming and dynamic spectrum access mechanisms, respectively. On the other hand, the user-driven path enables users (and traffic) to rapidly switch between networks through an end-user multihoming mechanism which intensifies retail competition. MNOs are reluctant to adopt these mechanisms if they involve risks. However, regulators can facilitate the deployment of these mechanisms by guiding the level of inter-MNO transaction costs and end-user switching costs. This paper analyzes the market dynamics of these three core mechanisms by employing agent-based modeling. Initial results indicate that each mechanism improves allocative efficiency on a dynamic basis and that such mechanisms become necessary if the current static market model based on vertically integrated MNOs is not able to meet the requirements of service quality, capacity and coverage. One promising use case of the proposed mechanisms is the indoor femto-cellular deployment which suffers from coverage disparity due to static single-MNO base stations. Moreover, either end-user multihoming or national roaming may provide MNOs a feasible business case for building indoor infrastructure by solving coverage disparity problems, by means of competition or cooperation, respectively. Dynamic spectrum access may work as an extension of the previous mechanisms for solving congestion; however, it requires higher technical and business coordination.