Consider a market with a countable number of MNPs i, i = 1, ..., N, i 6= -i. Customers
are of mass 1 and have randomly chosen one MNP. As in Dewenter and Haucap (2005)
customers receive the same gross-utility a from calling but no utility from being called.3
Calls are assumed to be balanced across customers.
Each customer in network i demands a-b(si)pi,j = a-bipi,j calling minutes to customers
in network j, j = 1, ..., N, i 6= j, where pi,j is the per-minute price for outgoing calls to
network j , bi is a scale parameter for price-elasticity increasing in the investor’s market
share si, b0(si) > 0. The larger bi the lower is the willingness to pay for one unit of off-net
calling. As I will only consider off-net traffic the on-net/off-net pricing strategies and
the utility of being with a particular MNP need not be further specified. Customers only
decide on the call length dependent on prices. The short-run demand function deviates
from the model in Dewenter and Haucap (2005) where the representative off-net demand
depends on the average off-net prices weighted by competitors’ market shares.4
MNPs are active in a calling-party-network-pays regime (cpp regime). The per-minute
termination rate ti is the wholesale price which one MNP asks another for terminating a
call. The per-unit costs ci are identical for call origination and termination (as assumed
in A-LRT). Finally, I assume the long-run market to be sufficiently less concentrated.
For this setting this means si < 2 P-i s-ib-i.
3.1 Short-run price choice
With a linear pricing scheme provider i’s short-run off-net profits from call origination
and termination are given by:
πi(pi,j,ti) = πii,j + πji,i = ((pi,j -tj - ci)sisj(a- bipi,j))
j j j (1)
+ ((ti - ci)sisj(a - bjpj,i))
j
3 For the analysis of quality investments one could think of identical gross-utilities for customers per
provider ai .
4 A more detailed discussion on customers demand for off-net calls can be found in Dewenter and
Haucap (2005) and Hoernig (2007). Nevertheless, it is obvious that by adopting Dewenter and
Haucap’s assumptions, bi = bj = b and pi = ci, one receives similar results with the model of this
paper.
More intriguing information
1. Knowledge, Innovation and Agglomeration - regionalized multiple indicators and evidence from Brazil2. A Multimodal Framework for Computer Mediated Learning: The Reshaping of Curriculum Knowledge and Learning
3. A Regional Core, Adjacent, Periphery Model for National Economic Geography Analysis
4. Tourism in Rural Areas and Regional Development Planning
5. The name is absent
6. Correlates of Alcoholic Blackout Experience
7. The name is absent
8. Urban Green Space Policies: Performance and Success Conditions in European Cities
9. Tariff Escalation and Invasive Species Risk
10. The name is absent