1
Excerpts of TAL Agreement,
Cancun
• Subject to unspecified amendments, TAL
interconnect and access cost models were stable
enough to software application phase.
• Revised methodologies will constitute Appendices
3& 4 of Recommendation D 400 R.
• Deadline date for submission of contributions to
revision of models prior to progress to software
phase was August 31, 2003.
• February 2004 initial date for completion of
software.
2
Revision, TAL Interconnect
Model
Objectives
To simplify basic interconnect services for
software application in terms of:
a) Unit costs of single transit interconnect
traffic.
b) Unit costs of Double transit interconnect
traffic.
3
General Objective
Regulator
Service Provider
Service Provider
Investment
Opportunity
Fair Proxy Market price
reflective of reasonably
efficient cost
Incumbent
Service Provider
Consumer
4
Traffic Definition
According to Recommendation D 600 R :
• 1.4.1. Incoming national, single transit:
Traffic terminated on the network of an
Operator in a country which originated on
the network of another Operator in the same
country and where the interconnection point
between the two networks is located within
the local tariff zone of the Operator which is
terminating the traffic. Example: traffic
between telephone G and A or telephone K
and B at Diagram 1.
•
5
Zone 1
Telephone "B"
Telephone "D"
Telephone "A"
Zone 2
Telephone "C"
6
Distant Intertional
IGW
Distant International
Satellite
Telephone "E"
IGW
Telephone "F"
Neighbouring country
Telephone "C"
h
IGW
Point of Interconnection
Telephone "B"
Neighbouring country
Point of Interconnection
Telephone "A"
Telephone "D"
National network operator XCloud
Telephone "G"
National Network Operator Y
Telephone "H"
Telephone "K"
7
Double Transit Traffic
• Incoming national, double transit: Traffic
terminated on the network of an Operator in a
country which originated on the network of
another Operator in the same country and where
the interconnection point between the two
networks is located outside the local tariff zone of
the Operator which is terminating the traffic.
Example: traffic between telephone G and B or
telephone K and A at Diagram 1.
8
Network & Non-network
Resources Used
• Switching Components (including signaling
network).
• Transmission Components.
• Other resources used to support network
and general delivery of the services.
• Unit Cost = ƒ(cost caused due to usage of
network and other resources to deliver
traffic in period n/Volume of traffic
9
delivered) in period n.
Stages of cost-orientated Unit Costs
•Cost of
of
network
components
components
•Operation
and
maintenance
costs
costs
•Service
traffic
•Amortization
rules
•Equipment
price trends
•Cost of
capital
•Cost of
functional
support
•Identifiable
direct and
indirect costs
costs
•Other
common costs
•Routing table
•Cost
distribution
distribution
COSITU - The ITU tariff model
•Unit
endogenous
cost of
services
23
10
Two Basic Costing Principles
• Causality:
– The demonstration of a clear cause-and-effect
relationship between service delivery, on the one
hand, and the network elements and other
resources used to provide it, on the other hand,
taking account of relevant cost determinants
(cost inducers/drivers).
• Efficiency:
– The provision of a forecast of cost reductions
that result from a more efficient combination of
resources.
11
Diagram 2
LRAEC
11/11/2003
Marginal Cost
Cost
Subtitle
11/11/2003
Efficient Cost
C1
Quantity
Q1
12
Parent Formula
•
•
•
•
•
•
•
•
•
General LREC formula:
C = A(ß)/Mo
Where:
C= per minute cost of traffic
A= total cost (direct, indirect and common)
ß = Efficiency factor = x/y
x = required capacity
y = installed capacity
13
Mo = traffic throughput.
throughput
Cost Adjustment?
• Historical costs: Based on the cost price of
equipment and services,
• Current costs: Take account of the changing
environment: falling prices of
telecommunication equipment, currency
depreciation. The TAL Model utilizes
current costs as confirmed on the relevant
market.
14
Fully Distributed or Incremental
Costing
Accounting Methods
Cost Methods
Historical Costing
Actual Cost
incurred
(includes
inefficiencies)
Current Costing
Hybrid Costing
(Adjustment toward
Current costs)
Current Costs of
Providing Service
(Reduction of over
pricing)
Actual Costs
adjusted for market
changes and
efficiency
Fully Distributed Costing
Incremental Costing
Total Cost is allocated to
Services
Long Run Incremental
Costs are allocated to
Incremental services
15
Clarity re consistency with
Fundamental Costing Principles
•
•
•
•
•
•
Parent TAL formula:
C = A(ß)/Mo
Interconnect Model Formulae:
LEC = (A/Mo ) *(∑∝)
IEC = (? A/? Mo ) *(∑∝)
The efficiency factor ∝, is estimated in terms
of adjustment in OPEX and or adjustment for
avoidable excess network capacity.
16
LEC & IEC
ü EEC = Aå/M1
ü Aå = total cost
adjusted for
unavoidable diseconomies of scale
(inefficiency).
ü M1= total traffic
(current +
expected)
ü IEC= A@/M-1
ü A@ = cost of
additional
components to
deliver interconnect
traffic, subject to
adjustment for
inefficiency
ü M-1= interconnect
17
traffic
Current Cost Adjustment
Effect of Exchange rate adjustment on cost may be
estimated by:
CAD = DA[(1+ τ)DP/2 /(1-ε)DP/2 –1]
Where:
CAD = Adjustment to current cost;
DA = Depreciation allowance;
τ
= Average annual variation in equipment
price;
•
18
DP =
Depreciation period;
ε = Average per annum rate in currency
movement.
19
Cost of Capital
Combined effect of debt and equity
(i) Creditors demand interest
(ii) Owners demand dividends
Firms normally raise capital through:
q Long term & short term debt
q Sale of Preferred stock
q Sale of common stock
q Retained earnings
These cost are basic determinants of the cost of
capital.
20
Cost of Debt
Cost of debt is determined by:
• Interest rate on debt; subject to:
• Adjustment for taxes where interest payment
on debt are deductible for income tax
purposes.
• After tax cost of debt can be expressed as:
(interest rate) * (1- tax rate)
• E.g. If interest rate is 9% and tax rare of 25%
• The after tax cost of debt = (0.09)*(1- 0.25)
= 0.0675
21
Marginal Cost of Debt
Cost of debt should be calculated at marginal
cost:
• Current interest rates on the market
• Taking into account interest rates on
similar enterprise on the international
market
22
Cost of Equity
• Required rate of return (s ) on the common stock
of the firm
• Firms use Expected rate of return.
• The Expected Rate of return on a stock is a risky
variable.
• s = risk-free rate of return σ(RF) + Risk Premium
(RP).
• RF is normally reflective of interest on
government security.
• RP is the difference between the required rate &
risk free rate of return
23
•
•
•
RF = ƒ(rate on government securities )
RP = ƒ(different methodology of estimation)
Interest on firms long term bonds (iC ) + a
percentage-points addition based on historical
returns per year on the financial market (iA). Let
RF,iC & iA be 6%, 9% and 5% respectively:
• iC + iA = (0.09 + 0.05) = 0.14
• RP = 0.14 – 0.06 = 0.08
• s = 0.14
24
Jeorgenson Method
qThe cost of capital d is calculated using the
formula:
formula
• d = D i* (i-t) + E s
D+E
D+E
Where:
D = amount of medium and long term debt;
σ
= before tax return on capital;
E = equity; and
i = average interest rate.
t = rate of corporation tax.
25
Operating and maintenance costs
q Cost of inputs
ü Purchases and variations in stock ;
ü Transport;
ü Outside services
q Personnel costs
q Taxes and levies (corporation Tax not
included
q Other charges
q Other Financial and similar charges
q Operating provisions
26
Total Network
Costs
Cost of Network
Elements
Operating Costs
Depreciation
Expenses
Capital Cost
Operating &
Maintenance costs
(OPEX)
Text
Network
TextAsset Valuation
Technology Text
(Adjusted)
Cost
relationship
Text
Text
Cost
Drivers
Depreciation
Profile
(time, method etc)
Text
Text
Other Operating
Costs
Contribution to
Interest rate:
Text Text Service
TextText
equity & debt
prooduction
Routing:
:Component
Usage
Production
Efficiency
Text
Disaggregated
Component
Costs
Interconnect
Service Costs
27
Total Product/Service Efficiency
i, Mi, Ler, Ai, Ws,Ti, µi)
Oz
ƒ
(
S
e
• Total Service
Efficiency (Tse) could Where:
Si = Relative Salaries and wage
be expressed as:
levels;
• Tse = OZe + Ne
Mi = Managerial competences;
Where:
Ler = Line/employee ratio;
OZe = Organization and Ai = Employees work ethic;
Operational
Ws = Skill of Average Worker;
Ti = Availability state-of-theefficiency.
art
technology at work place;
Ne = Network efficiency.
µi = Other factors.
28
Network Efficiency (Ne)
• Ne = tc + tc+1 -? = 0
Where:
Ø tc is Current traffic;
Ø ti is Expected traffic where i = 1 to n;
Ø ? is Network capacity.
capacity
•
29
Evaluating Total Efficiency
Factor
• Let the network efficiency be expressed as
Q1(used capacity +backup capacity ) /Q2(installed capacity)
=ß
Total Service Efficiency, (T
( se) is:
Tse = (ß + Si+ Mi + Ler+ Ai+ Ws + Ti+ µi)
= (ß + OZe)
30
Issue /Decision
a) Except for the element: employee line
ratio, the other elements that comprise OZe
are extremely difficult to model.
b) The fundamental question: do the elements
that comprise OZe satisfy the SG3
practicability criterion for inclusion in any
cost model??????.
c) ß is measurable, with some degree of
difficulty.
31
Practicability
1. Sg3 documents the need for a model to
be practicable.
2. Two fundamentals of Practicability
are
a) Availability of data; and
b) Simplicity of methodology for
quantitative estimation.
estimation
32
Evaluating Network Efficiency
Assumptions:
qEfficiency requires network dimensioning
to accommodate full-capacity busy hour
traffic (including factor for traffic growth
and breakdown).
qIf not, the network is inefficient either in
terms of excess or under capacity.
qAn important concern, in terms of cost, of
the regulator is excess capacity.
33
Estimating Network factor, ϖ
– TBHEj = ß j (1+ g + ? ) = Required capacity
•
•
•
•
•
Where:
ß j is current busy hour traffic in erlangs.
g is the growth factor for busy hour traffic;
? = network backup factor
j = 1 to n. So:
ϖ = ß j (1+ g + ? )
•
____________
•
IC (installed capacity)
34
•
Likely Outcome
ϖ
ϖ
ϖ
=1
< 1
>1
OPTIMA
% Excess capacity
% Under capacity
35
Avoidable/Unavoidable Excess Capacity
Key efficiency questions:
1. Is excess network capacity avoidable?
2. Example when not and when:
i. If computed required capacity for
consistent busy hour traffic demand at
a particular switch is Q, but closest
manufacturing specification for that
switch is of capacity Q1where Q1 > Q,
the excess capacity Q1- Q is
unavoidable.
36
ii. However, if the operator elects to
install a switch with capacity Q2,
where Q2 > Q1, there is a gap Q2 - Q1
iii. If pQ2 > pQ1, the efficiency gap is
negative and avoidable and requires
an efficiency adjustment factor:
ϖu = (Q1/ Q2)
So the LEC unit cost of interconnect
traffic is:
(Total induced cost of Traffic)*(ϖu)
• Volume of traffic
37
Possible Cost Model Architecture for (PSTN) Fixed Network
Sources of Data
Extract following
data from the
Plant Accounts
Determine
network cost of
services
(4)
(5)
(6)
Technology
Cost Pool
Cost Driver
Services
Element / Functionality
Local Loop
Underground Conduit
Cable
Poles
etc.
Determine
weighted
average cost of
capital
Apply cost drivers and
routing factors
(3)
Annualized
Capital
Cost
Switch
Group Elements into
Cost Pools
Loop
distance
# lines in service
Line miles in service
sensitive
costs
Line termination
Signalling
Signalling and Setup
& Setup/
Minutes of Use
No. of call attempts
Routing Factors
Identify network
support costs
from general
ledger
Unbundle Technology
Groups to Technology
Elements
(2)
Engineering Studies
• (1)
• Opening &
Closing Net
Book Values
• Depreciation
Charge for the
period
• Date of
Commissioning
of Assets
Determine capital
works in progress
by asset class
Reclassify to
Technology Groups
Classify plant account
records into following
technology groups and
determine annualized
capital cost
Services
Trunk Termination
Minutes of Use
Conveyance
Transmission
Electronics
Satellite Equipment
Direct
Microwave Equipment
Fibre Optic Equipment
38
National Network
39
Unit Service Cost
Interconnect formula per individual service
becomes:
(Total induced cost of Traffic)*(ϖ u) Rij
• Volume of traffic of Service, j
• Rij is the factor which represents the
intensity and frequency of use of the
network in the delivery of service, j where
j = 1 to n
40
COSITU REVISION
• Old Efficiency factor COSITU =
Used capacity/Installed capacity.
• Efficiency Factor TAL =
Used capacity + backup capacity /
unavoidable installed capacity.
This adjustment was effected in COSITU.
41
42