Werner Neu

1
Cost Modelling for Mobile Termination Rates

• Werner Neu
• Presentation at CBKE - CIL - WIK
  ConferenceWroclaw, 18-20 October 2006

2
Themes

• Regulatory motivation for modelling of cost of
  mobile termination
• Bottom-up cost modelling
• Alternative Top-down cost determination
• Concluding remarks

3
Regulatory Motivation for cost modelling

• Markets for termination are monopoly markets in
  EU framework
• Thus, prices for termination are not determined
  according to forces of competition
• Regulators are called upon to set prices on basis
  of cost
• Cost determination using cost accounting records
  is beset with many difficulties
• There is thus a need for a cost determination
  that is as much as possible based on objective
  information and data
• Bottom-up cost modelling is the tool best suited
  for this purpose

4
Bottom-up cost modelling (1)

• Started in late 1980s and early 1990s in the US
• At the beginning was mostly an academic exercise
• Regulators got interested when they found that
  cost data submitted by regulated firms were not
  transparent
• Is implemented to reflect Long-Run Incremental
  Costs (LRIC)
• LRIC the cost standard that a competitive
  multi-product firm needs to adhere to
• Is based on the premise that the investments in
  network elements are the principal cost drivers
• Network and its elements to be derived on the
  basis of objective measures of demand
• Originally used for fixed, now also increasingly
  for mobile networks
• Following slide shows schematic view of modelling
  process

5
Bottom-up cost modelling (2)
Investmentprices
Annuityfactors
Structuralparameters
Costs of capital
Dimensioningof networkelements
Investmentcalculation fornetwork elements
Network structure hierarchy
Operating costs
Coverage Traffic
Operating costsfactors
Network Design Module
Cost Module
Cost pernetwork element
Intensity of use factors
Mark-up overhead common costs
LRICof service
Model output
Model input
6
Bottom-up cost modelling (3)

• Comments
• Network for all services need to be modelled in
  order to capture the effect of economies of scale
  due to total volume of services
• The design and modelling of the network is the
  intellectually most demanding part
• Once the quantities of network elements, and the
  investments in them are known, it is
  conceptually easy to determine
• CAPEX interest on and depreciation of the
  investment
• OPEX the current expenses of maintaining the
  netw
• Both CAPEX and OPEX are in this approach a
  function of the current value of the investment
• Getting hold of relevant data, in particular
  prices of equipment and facilities, poses
  greatest challenge for implementation

7
Bottom-up cost modelling (4)

• Typical services usually included in a model of a
  2G network
• Voice
• SMS and MMS
• Basic data service with 9.6 kbit/s
• GPRS
• High speed data services with n9.6 kbit/s
• For all of above
• on-net
• originating for off-net
• terminating from off-net lt Service if
  interest here

8
Bottom-up cost modelling (5)

• Network structure and network elements that need
  to be accounted for in the model
• Cell deployment
• Base station subsystem (BSS)
• Base stations
• Base station controllers
• Network switching subsystem (NSS)
• Mobile switching centres
• Registers (HLRs, VLRs)
• Transmission links between various nodes

9
Bottom-up cost modelling (6)

• Structure for a GSM mobile network

10
Bottom-up cost modelling (7)

• Essentially cost in a bottom-up approach is
  determined according to the following equation
• C I r d f
• where C cost of relevant network
  element I value of investment in relevant
  network element r required return on
  investment the WACC d rate of
  depreciation of investment f relation of
  OPEX to value of investment
• Cost per unit of services is then
• c C / q
• where q volume of relevant
  services for a particular network
  element, all measured in the same
  dimension (usually minutes of service)

11
Bottom-up cost modelling (8)

• Often r d is combined when annuity
  approach is used
• One coefficient for both CK a I
• where a is derived below
• Annuity approach to determine a C 1/(?
  ?2 ?n ) I
• where 1/(? ?2 ?n ) a
• n length of economic life of network element
• ? 1 / (1 r ) in simple case
• (1 ?p/p) (1 g) / (1 r ) in tilted
  annuity approach
• where ?p/p expected average change in price
  of equipment items
• g expected growth of service
  delivered by the network element

12
Bottom-up cost modelling (9)

• Cost of termination
• ct S wi.t ci
• where ct per-minute cost of termination
• wi.t weight of per-minute cost of network
  element i in termination
• ci per-minute cost of network element i

13
Bottom-up cost modelling (10)

• Schematic view of mobile network by network
  elements

Core-facingport
IC-facingport
Backbonelink
BSC-facing port
BSC-MSCbackhaul link
MSC-facing port
BTS-facing port
BTS-BSCaggregationlink
BSC-facing port
14
Bottom-up cost modelling (11)

• List of required data
• Demand both in terms of geographical extent and
  distribution
• Network parameters
• Prices for equipment to be installed in the
  network
• Requirements in terms of reserves, redundancy for
  security reasons, etc.
• Empirical relationships between OPEX and
  investment value

15
Bottom-up cost modelling (12)

• Makes regulator to a large degree independent of
  the information from the cost accounting system
  of the regulated firm
• Provides a real knowledge of the production
  process of regulated services
• Allows to reflect all complex linkages in the
  production process
• Offers a rather detailed view of cost structures
• If right expertise is used, knowledge advantage
  of regulated firm is greatly reduced

16
Bottom-up cost modelling (13)

• Bottom-up cost modelling is
• (1) costly
• (2) time-intensive
• (3) knowledge-intensive
• In discussions with regulated firm, lack of
  relevant knowledge is often greatest drawback
• Different views between two sides about
• efficient network
• values of essential input parameters
• To the non-experts in the process, lack of
  transparency due to complexity

17
Alternative Top-down cost determination (1)

• Is the traditional tool when cost information is
  to be obtained from a company
• For the lay person comes naturally to mind and is
  therefore relatively easy to justify in any
  public debate
• Reflects actual transactions and can claim
  completeness with respect to all categories of
  relevant costs
• If done competently and transparently, can
  almost provide the same type of information as
  bottom-up cost modelling
• Courts have been known to be partial to cost
  determinations based on actual cost figures
• May be quicker - provided regulated firm has
  functioning cost accounting system in place

18
Top-down cost determination (2)
Data Capture
Balance Sheet
Profit Loss
Cost Accounting Records
AllocationStep 1
Services
Related Functions
Other Functions
Network Components
Step 2
Services
Network Components
Related Functions
Step 3
Services
Network Components
Step 4
Services
19
Top-down cost determination (3)

• Reconciliation with bottom-up approach
• In both approaches, the focus is on the network
  and the investment in its various elements
• Once their numbers are known, the various network
  elements are the cost drivers
• Order of objectives in a reconciliation
• Agreement regarding quantities of network
  elements
• Determination of cost of operations and
  maintenance on the basis of records used for
  top-down approach
• Here lies the strength of top-down approach
• Allows to trace costs in terms of concrete
  activities involved in operation and maintenance
• Makes possible to verify benchmarks used in
  bottom-up models

20
Top-down cost determination (4)
Data Capture
Balance Sheet
Profit Loss
Cost Accounting
AllocationStep 1
Services
Related Functions
Other Functions
Network Components
Step 2
Services
Capex
Opex
Step 3
Services
Capex
Step 4
Services
21
Concluding remarks

• Cost modelling has established itself as a
  powerful tool of the regulator
• Bottom-up modelling is the most effective in
  terms of providing relevant information
• Top-down modelling has its greatest value in
  providing information on the costs of operation
  and maintenance

22
wik-Consult GmbHPostfach 200053588 Bad
HonnefTel 02224-9225-0Fax [phone]eMail
info_at_wik-consult.comwww. wik-consult.com
23
Bottom-up cost modelling (2)

• Concrete steps of determination of cost of
  termination
• (1) Demand for all services
• (2) Corresponding network structure with all
  quantities of network elements
• (3) Required capacity of network elements for all
  services
• (4) Value of investments in the network
• (5) Capital and operating costs of network
  elements
• (6) Unit (usually per minute) costs of every
  network element
• (7) Determination of the cost of termination
  according to its use of various network elements