The paper is organized as follows: The next section provides an overview over the
existing literature on off-net mobile competition. Afterwards I introduce the theoretical
model. Hypotheses are derived which will be tested for the EU 15 countries as well
as for Norway and Switzerland. In section 4 the alternative estimation approaches are
introduced and compared with regard to alternative estimation outcomes. Section 5
provides the estimation results and discusses them in more detail. Section 6 concludes
and provides ideas for further extensions.
2 Literature Review
Alternative aspects of mobile interconnection have been analyzed in an extensive range
of literature which is mainly based on the framework of three seminal papers, Armstrong
(1998), Laffont, Rey and Tirole (1998a) and Laffont, Rey and Tirole (1998b), hereafter
A-LRT. Assuming a symmetric two-provider model, the papers provide insight into mul-
tiple fundamental outcomes concerning competition in network-based markets and open
a wide range for extending research. Concerning off-net traffic Laffont et al. (1998a)
show that in a common per-unit pricing system the increase in total outgoing traffic
corresponds to a reduction in incoming traffic. Thus, one provider’s decision to reduce
off-net prices is a decision at the margin. Alternatively, allowing for on-net/off-net price
discrimination network providers choose higher off-net prices, thus, affecting customers’
network choice as shown in Laffont et al. (1998b). In consequence, a raising-rivals-costs
strategy in the sense of increasing termination rates need not result in a change of retail
prices but in a promotion of competition for market shares.
The issue of infrastructure investments in network-based markets has been analyzed in
a comparably low number of papers mostly assuming a vertically integrated upstream
monopolist competing with one or more downstream entrants. Central results of these
papers have been proved to exist also with network competition. Foros (2004) shows
for a vertically integrated upstream monopolist and a downstream retailer that the level
of quality investments depends on the substitutability of downstream services. The
higher the degree of substitutability the lower is the investment incentive for the net-
work provider. Foros assumes investments to take place before the regulation stage. As
the investor does not know the implemented regulation in advance the under-investment
challenge becomes even stronger with regulation than in the situation with no regulation
which negatively affects total welfare. On the other hand, if downstream substitutabil-