Emerging trends in telecom Rohan Samarajiva, with contributions from team
24 August 2006, CFA/Amba Research
Agenda Introduction Next market Fixed-mobile convergence Spectrum management Local loop unbundling
If I know the future . . . Why would I tell you? Why wouldn’t I run to my broker and convert that knowledge into value?
I do not know the future.
But I do think about trends in a broad sense . . .
Incomplete knowledge, scenarios . . . Based on ongoing research
LIRNEasia
and sister networks RIA! (Africa)and DIRSI (Latin America & Caribbean)
Other
Observation of fact patterns Not very different from what CFAs do?
But
less emphasis on financials
So, what is the plan for today? Scenarios, based on incomplete evidence Discussion from multiple perspectives Hopefully leading to useful knowledge; new questions, if not new answers . . .
Not very different from what we do at LIRNEasia everyday . . .
Agenda Introduction Next market Fixed-mobile convergence Spectrum management Local loop unbundling
One starting point: 3G Dialog’s announcement First in the region, in 2006 Big fuss re 3G was around 2000 Problem?
3G?
Region?
Products (services) succeed because they
Fill a gap in the product space that needs to be filled
Gap
may have arisen “naturally” or been created
Gaps located in time and space
A
“window” that is open for a period of time and then closes
Importance of time
The gap that existed in 2000 may have been filled/redefined by 2006
GMPCS
(Iridium, Globalstar, ICO) aimed to fill a gap that was being filled fast by GSM Too slow to market GMPCS has become an appendage of GSM What is Globalstar? ICO?
No universal answer for 3G
Even if the window is closing in developed markets, not necessarily the case in emerging markets
Mature
markets have plenty of broadband connectivity at low prices
Emerging markets don’t have capacity even for a price
What does 3G offer consumers?
Mobile voice and data?
What does WiFi offer consumers?
Voice and data anywhere?
Who will win?
In the hands of
Terminal
manufacturers
Regulators Freeing up frequencies Pricing
Innovators
in terms of price and availability
Watch Taipei WiFi story Major WiFi push in country with 2nd highest mobile penetration in Asia
Telekom Malaysia affiliates: Growth in pre- and post-paid customers Sri Lanka Indonesia
Bangladesh
TM affiliates EBITDA & EBITDA margins Sri Lanka Indonesia
Bangladesh
TM affiliates’ monthly Average Revenue Per Users (ARPUs) Sri Lanka, Rupees Indonesia, Rupiah (‘000)
Bangladesh, Taka
Reality: Growth in prepaid almost swamping postpaid; ARPUs Î USD 5; EBITDA margins holding at around 5060% ARPU per month: Prepaid vs. postpaid Sri Lanka Prepaid
Sri Lanka Postpaid
Indonesia Prepaid
Indonesia Postpaid
B'desh Prepaid
B'desh Postpaid
0.00
5.00
10.00
15.00
20.00
25.00
30.00
USD
All TM data: http://www.telekom.com.my/about_tm/investor_relation/
What’s the next market? 3G for the rich? or SEC groups C, D and E?
The next billion . . . .
Their mode of access will be wireless They will for the most part be from the Asia Pacific
From
the emerging economies in the Asia Pacific: high growth in low-teledensity countries
Pakistan and Bangladesh growing at above 100% a year now India has enormous potential; even if China’s rapid growth moderates, multiple millions . . . .
From among the poor in the emerging economies
Teleuse on a shoestring 1
LIRNEasia’s 2005 survey research project looking at persons with income < USD 100/month
In
2006-07, study will be conducted in 5 countries (India, Pakistan, Philippines, Sri Lanka & Thailand)
Random samples drawn to represent financially constrained population in country Supplemented by diaries and focus groups
Study
completed in 2005 in 11 locations in India (over 2000 respondents) and Sri Lanka (over 1000); six languages
Exploratory; not representative of country
Screening questions
Did you use a telephone in past three months?
Less
than 1% screened out in LK by this question
Less than 15% in India
Monthly income less than USD 100
Many people do not own the phone that they use 100
Own
Don't Own
77%
76%
80
52% 49%
60
51% 49%
40
24%
23%
20 0 < USD50
USD50-100
< USD50
Sri Lanka
India
Sri Lanka What people own Landline Only
USD50-100
India
< USD 50 2
USD50100 20
< USD 50 19
USD50100 37
Mobile Only
21
28
3
8
Landline + Mobile
1
4
1
6
Phone ownership doubles in higher income groups
How do the financially constrained communicate? Fixed (49%)
21%
Mobile
2%
11%
(19%)
3% 3%
23% 37%
‘Public’ access
Base: 3199
(66%)
Getting connected: Financing fixed line connections Obtaining a fixed line: paying in slightly installments How respondent paid for fixed line harder in Sri Lanka connection Use of installment plans
100
100
80
80
60
% of owners
% of fixed line owners
Where owners got the money from
40
60
40
20 20
0 Sri Lanka
India
Had to get finance from somewhere else I had the money (or someone in my household did
Base: full sample: 847 SL: 212 India: 647
0 Sri Lanka paid full amount at once
India Paid in installments
Base: full sample: 847 SL: 212 India: 647
Reflects connection charges for fixed phones
Sri Lanka incumbent (min.) Sri Lanka entrants
India incumbent
=
~USD 200
=
~USD 100
=
USD 16
Opposite case in mobiles: less people in India had money available for connection Where owners got money for the connection from 100
% of mobile owners
80
60
40
20 Base: full: 509 SL: 322 India: 187
0 Sri Lanka
India
Had to get finance from somewhere else I had the money (or someone in my household did
% of mobile owners
Getting mobile handsets: 10% got it free; 33% of Indians bought 2nd hand 100
80
60 Sri Lanka total India total 40
Bases: full:509 SL: 322 India: 187
20
0 bought it brand new bought it second hand received it as a gift / got it free
% of fixed + public access users who spend...
71% fixed + public access users spend USD12 Base Mobile users (Jaffna): 125
‘On average, around the world, people spend about 2-3% of their income on telecom’
Telecommunications Regulation Handbook, infoDev; module 6
Rural respondents: fixed phones more affordable; less likely to change use if price changed Perception of current costs
Base: fixed (only) users
High
80
Affordable
60 40 20 0 Sri Lanka Urban
Sri Lanka Rural
N India Urban
N India Rural
S India Urban
S India Rural
S India Urban
S India Rural
Change in usage if costs were halved W ill increase usage W on't change
80
60
40
20
0 Sri Lanka Urban
Sri Lanka Rural
N India Urban
N India Rural
•Reflects telecom/transportation trade-off •Access deficit charge in India ensures lower cost of rural fixed phones
Mobiles more costly; use would increase if costs halved; rural users less unhappy re cost Base: mobile (only) users
Perception of current costs 100
High
80
Affordable
60 40 20 0
Sri Lanka Urban
Sri Lanka Rural
N India Urban
N India Rural
S India Urban
S India Rural
Change in usage if costs were halved 100
Will increase usage
Won't change
80 60 40 20 0
Sri Lanka Urban Sri Lanka Rural
N India Urban
N India Rural
S India Urban
S India Rural
Reflects reasons for choice of mobile : convenience
In sum . . .
Great majority approached were telecom users, though 58% used other peoples’ phones
Next
market; primed and ready!
Still mostly through public phones
Could
be Indian idiosyncrasy; new research may yield different results
Barriers to participation in market exist
Biggest
barrier is availability of service
Lower handset prices, better 2nd hand markets likely to accelerate ownership
In sum . . .
Poor seem willing to spend more than the norm on telecom high in some cases (1/3rd of Jaffna respondents spend more than 12%) that questions being redesigned (income Î expenditure)
Rural perceptions of greater affordability
So
Telecom-transport tradeoff?
Demand aplenty; will policy/regulation & suppliers (service & equipment) respond?
Agenda Introduction Next market Fixed-mobile convergence Spectrum management Local loop unbundling
Fixed-mobile convergence Are two products seen as substitutes by consumers? This is the key question, not what policy says or regulators do
Though
policy/regulation can affect the way people see products
India versus Pakistan & Sri Lanka
India has announced license unification as policy
But
not all the pieces are in place yet, three years later
Yet major progress on substitution side
Pakistan & Sri Lanka seeking to maintain differences through licensing and related actions
Prices
Perception is that fixed and mobile have converged in India; not yet in Sri Lanka Most objective way of comparing is by use of baskets (OECD methodology)
Problem: South Asian mobile use is much higher than in OECD
Therefore, we compare only the OECD high-user mobile basket (1800 mts/yr) v. fixed (1200 calls/yr or 4400 mts/yr)
But, cost per call/minute may be more important psychologically than basket
Will know when research is complete
Only preliminary results given here (suggestions welcome)
India Fixed - Residential Basket vs Mobile - Prepaid and Postpaid High User Baskets (Annual Totals) 1,200.00
1,000.00 16% cheaper than fixed
USD PPP
800.00
600.00
32% cheaper than fixed
400.00
200.00
-
Fixed
Mobile - Postpaid
Mobile - Prepaid
-
79.63
99.99
Usage
776.30
627.02
638.75
Rental
326.00
213.84
-
4.31
8.98
3.56
SMS
Connection
Connection
Rental
Usage
SMS
Comparison of Sri Lankan Mobile Prepaid and Postpaid Tariffs (Annual Totals) 1,000.00 900.00 800.00
USD PPP
700.00 600.00 36% cheaper
500.00 400.00 300.00 200.00 100.00 -
Mobile - Postpaid
Mobile - Prepaid
SMS
43.92
43.92
Usage
579.51
501.79
Rental
250.53
-
Connection
12.10
21.78 Connection
Rental
Usage
SMS
Counter-intuitive results According to OECD high-user comparison in USD PPP, IN fixed > IN postpaid > LK postpaid > IN prepaid > LK prepaid Unable to easily compare with LK fixed because of minutes bundled with rental
IN & LK pre- & post-paid mobile: OECD high-user annual baskets compared 1000
900
800
700
USD PPP
600
500
400
300
200
100
0
IN Mobile - Postpaid
IN Mobile - Prepaid
LK Mobile - Postpaid
LK Mobile - Prepaid
Proof of the pudding . . .
Nearly 90,000 BSNL subscribers have surrendered their connections in Calcutta between April and November. “Over three years, about 100,000 subscribers have surrendered their phones every year. This time, the number is even bigger,” a senior CalTel official said. “This is a worldwide phenomenon, the growth is in the mobile phone sector,” BSNL chairman and managing director A.K. Sinha said. He was in the city today to inaugurate an information and media centre at Telephone Bhavan and a customer service centre at Alipore. Sinha said across India, 25 lakh BSNL subscribers have surrendered their fixed phones till November. From Indian Telegraph http://www.lirneasia.net/2005/08/fixed-line-substitution/
No convergence in Sri Lanka?
Sri Lanka is RPP, though a majority of incoming calls may not be charged
Fundamentally
different
Also, people believe mobile >fixed
Perception matters . . .
Otherwise,
why would anyone buy a CDMA phone for LKR 7700 x 3 installation charge from SLTL?
But on the other hand, mobiles have fashion and status connotations (at least among the Indian poor?). ..
Reasons for choosing to use a mobile It improves my social status 80
Most economical way to receive calls
It is fashionable to use one
70
72
60
Most economical way to make calls
57 60
50
I can use value added services like SMS etc.
40 35
49
To control my communication expenditures
30
34
20
30
7
7
10 40
11 32
To save travel time and cost
It allows privacy
26
72 25
0
13
64
42
53
The connection is clear
41 57
48 59
It is the only available service/ No other choice
71
It is easier to use 57
67 78
I can use it while on the move
It is easy for me to access 79
I can use it at any time
Sri Lanka
India
www.lirneasia.net
Agenda Introduction Next market Fixed-mobile convergence Spectrum management Local loop unbundling
Spectrum management Critical to everything in telecom Philosophical/technological issues Situation in Sri Lanka
Big picture
Old centrally planned and managed framework under threat from two directions
Property
rights and decentralized decision
making
Commons and unlicensing
Current regime is good for those who have frequencies
Quasi property rights regime
Cannot
sell frequencies, but can sell the company with the frequencies to capture value
(Lanka Bell with CDMA frequencies) – (LB without) = Value of 2.5 MHz of CDMA 800 But 200,000 new users
Creates
solid barriers to entry
No incentives to increase efficiency of use
Downside is potential rent seeking from Minister down
Well designed auctions can alleviate But not being adopted (why is 5th license USD 4m?)
President has ordered refarming, but not a change in paradigm
Can be done
As
was done with CDMA 800 and GSM 1800 in 2002-05 Consultation papers Informed input Government officials educated on replacement costs vs depreciated costs
Hypothetical refarming process Step
Main policy actions
Parallel policy actions
1
Government sets overall policy and authorizes negotiations with seven operators (O1 – O7)
2
System and frequency license modifications negotiated (Modifications include removal of technology restrictions from O1, O2, etc.; and may include extending license term of O4 (which will gain no benefits but has to yield frequencies)
3 & 3A
O4 and O5 release GSM 900 frequencies; O1, O2, and O3 will also be requested to agree to phased release of frequencies to enable overall ordering of the bands
1800 MHz Tender Board releases funds for band clearing (some 1800 MHz frequencies have been auctioned to GSM operators)
4 & 4A
O6 assigned GSM 900 frequencies & releases CDMA 800 frequencies
1800 GSM and 1900 CDMA bands fully cleared
5 & 5A
O1, O2, and O3 assigned CDMA 800 frequencies
Auction frequency slots that may be used for CDMA 1900 or GSM 1800 to current operators but possibly also to newcomers
Agenda Introduction Next market Fixed-mobile convergence Spectrum management Local loop unbundling
Local loop unbundling
Major regulatory issue
Go
too slow; new entrant voice and data providers will
die
Go too fast; network buildout will stall
But how relevant in LK and similar economies?
Talk
to me after backbone access assured to all
Interesting twist in LK: who owns the SLTL local loop?
Customer
who paid distance-based installation
charge?
If yes, what is the problem?
More info: Rohan Samarajiva www.lirneasia.net (e.g., search “shoestring”)
[email protected] (re shoestring1 or 2)
[email protected] (basket methodology)
[email protected]