Please Share:: 
India Equity Research Reports, IPO and Stock News
Visit http://indiaer.blogspot.com/ for complete details �� ��
It’s time to focus on Bharti’s revenue
growth, not negative sector newsflow
It offers the best potential return of all
India large caps; competition, regulatory
risk-reward profile also improving
Upgrade Bharti to OW and raise TP to
INR425 from INR370
Look beyond the headlines. The backdrop to this
overpopulated, troubled industry is messy, even if, unlike
consensus, we think the worst is over for revenue per minute
after a decline of 30% in the last two years. So, perhaps not
the best time to buy the biggest stock in the sector? We
disagree. Market leader Bharti offers the best potential return
among Indian large caps at a time when competitive and
regulatory risks for the market leader are also diminishing.
Focus on earnings growth at Bharti. We estimate FY12e
earnings growth at c27% and FY13e at c37% (versus c18%
for Sensex over the next two fiscal years and c20% earnings
growth at other large caps). A large part of the growth comes
from improvements in voice pricing strategy, a pick-up in
2G value-added services and the company’s tower business.
The rate of tariff decline has been falling and we expect this
to improve further and revenue per minute to also rise.
Regulatory and competitive risks diminishing. The
government needs a healthy telecom sector to generate
revenue from taxes and spectrum auctions. Bharti stands to
benefit most from reforms that we think will help weed out
weaker rivals, many of which already face huge additional
fines and extra costs for contract infringements related to the
2G licence scandal. For example, it is now much tougher for
the new entrants to raise financing. Bharti is well positioned
to benefit from the inevitable consolidation that must follow,
even if the process takes a while.
Top pick: Bharti. We upgrade Bharti to OW and raise our
TP to INR425 (from INR370). We remain N(V) on Idea and
RCOM. Our TPs for all three stocks factor in the possible
negative impact of recommendations by the regulator, TRAI.
These include charging incumbents for excess spectrum as well
as a one-time payment when they renew licences and the refarming
of 900 MHz spectrum.
Summary
Focus on Bharti’s revenue growth, not negative sector newsflow
Unlike consensus, we think the worst is over for revenue per
minute and expect improvements over time
Upgrade Bharti to OW and raise our TP to INR425 (INR370);
reiterate N(V) on Idea Cellular and RCOM
A brighter outlook
The fragmented industry structure and the issue of
new licences in 2007 have raised serious questions
about sector returns and the ability of operators to
monetise mobile connectivity. However, things are
looking up. Positive signals include the fact that
there have been no major cuts in tariffs since 2010
and revenue per minute has stabilised over the past
two quarters. Moreover, sector revenues improved
by 5% in 3Q FY11, driven by robust increases in
minutes of use (MOU) – the highest growth since
December 2008.
We believe the competitive landscape is changing.
All eight new players which entered the sector in
2007 are facing significant financial and
regulatory challenges. Since the probe into the 2G
spectrum allocation scandal was announced,
several have been fined for failing to meet rollout
obligations and some may even face the
possibility of having their licences cancelled. The
mere presence of the eight newcomers has kept
tariffs in check but we expect this trend to change
– but it will take time.
A range of positive factors
Over and above the challenges faced by new
entrants, there are several other factors driving our
positive view. These include:
MNP is now behind us without causing any
meaningful damage.
The winners of 3G licences will need to invest
in their networks to improve voice revenue.
Limited upside to MOU per subscriber and
the fact that simply adding subscribers is no
longer the solution. Since June 2009, the
sector has added 348m subscribers but
incremental revenue has been only USD840m
– revenue growth of 13% compared to
subscriber growth of 46%. Operators need to
extract more revenue from voice subscribers.
Expect policy reforms
While regulatory risks remain (both from the 2G
corruption probe and TRAI spectrum pricing
recommendations), the telecom sector is
significant for the government coffers. Tax
receipts are estimated at USD5bn and the
government also raised cUSD24bn via the
spectrum auctions in the last fiscal year. Add to
this the fact that the government is targeting a
fiscal deficit of c4.6% in FY12 and there will be
constant need to bring the deficit down going
forward. Against this backdrop, we expect the
government to introduce policy reforms to create
sustainable revenue streams from the sector.
We estimate the telecom sector ROE is 5%,
significantly below average returns from other
sectors (see table). We believe it is important for
the regulator and the government to pay attention
to the financial health of the sector. Given this, we
see little chance of TRAI’s recommendations
being part of the new telecom policy the
government plans to deliver this year.
Moreover, we believe TRAI’s recommendations
are a major shift from current policies and are
unlikely to be accepted by the Department of
Telecoms (DoT) in their present form. That said,
our target prices for all three stocks have been
adjusted to take into account possible damage
from TRAI’s recommendations.
Tariffs
We believe the market has a very pessimistic view
on tariffs. The bulls are betting on 3G, which we
don’t expect to be earnings accretive in the near
term. The stability in revenue per minute is being
viewed as a good downside support but the
possibility of tariffs moving up is being
completely ruled out. We believe that pricing
power can return to the sector and revenue per
minute should improve over the next 3-4 quarters.
Now this need not take the route of big
movements in the headline rates. Given the mood
of the Indian regulator, we don’t believe that
operators will do anything drastic in the near term,
preferring tweaks to current subscriber plans.
Valuation
BHARTI, OW, TP INR425
We are turning positive on Bharti, upgrading to
OW and raising our target price to INR425 from
INR370. We roll our multiples forward to FY13e,
in line with our methodology on other Indian
stocks under coverage. Our target price is adjusted
by an additional INR18 to factor in TRAI
recommendations, taking the total negative impact
to an estimated INR46.
We believe Bharti is the best way to play the
brighter outlook given its quality subscribers,
generation of c2bn daily minutes on its wireless
platform, deep coverage and network advantage.
Our new target price implies a PE of c16x on
FY13e EPS. The stock is currently trading at 13x
on our FY13e estimates. Our numbers for FY13e
are c2% above consensus. Bharti’s 12-month
forward average trading multiple has been 14x.
As we factor in higher depreciation charges on
account of 3G we have cut our FY13e earnings by
c4.5%. We estimate FY12e earnings growth at
c25% and FY13e at c35%. A large part of the
growth comes from improvements in voice rates,
a pick up in 2G VAS and scale benefits from the
tower business. We remain cautious on the Bharti
Africa business and value it at -INR33 per share.
Upsides risks to our call include the possibility of
improvement in headline tariffs, visibility on
regulations pertaining to sector consolidation and
faster than estimated pick up in 3G. Further ability
to monetize tower assets also remains a key
upside risk to our valuations. News reports
suggest that both Vodafone and Idea may consider
merging their residual towers in the Indus Tower
Company; this should be positive for valuations.
Downside risks include the use of the 3G platform
by challengers for voice, a slower than estimated
pick up in 3G services and acceptance of TRAI
recommendations in the current format.
Idea Cellular, N(V), TP INR72
We retain our Neutral (V) rating on Idea Cellular
but cut the target price from INR75 to INR72 as
we make adjustments for the possible acceptance
of TRAI recommendations. Idea is most
vulnerable to TRAI recommendations and as such
we see limited upside. Our new target price
implies PE of 29x on FY12e EPS and 20x on
FY13e EPS. The stock trades at c19x on our
FY13e estimates. Our numbers for FY12e are 3%
above consensus and 8% below for FY13e.
Upsides risks to our call include the possibility of
improvement in headline tariffs, visibility on
regulations pertaining to sector consolidation and
faster than estimated pick up in 3G. Downside
risks include the use of the 3G platform by
challengers for voice, slower than estimated pick
up in 3G services and acceptance of TRAI
recommendations in the current format.
RCOM, N(V), TP INR122
We retain our target price of INR122 and Neutral
(V) rating. Our target price implies PE of 16x on
FY12e EPS and 12x FY13e EPS. The stock is
currently trading at c11x on our FY13e estimates.
Our numbers for FY12e are in line with consensus
but FY13e numbers are c9% below. While tariff
improvements should be positive for the company,
the upside is capped given poor subscriber quality
and weakening of the CDMA business with MNP.
Separately, the possible sale of tower assets remains
the only catalyst in the near term; we believe
unlocking tower assets will be difficult for the
company, given low network utilization on both
GSM and CDMA.
The key upside risks are the ability to monetize
tower assets and sell a 26% stake to a strategic
partner over the next 6-9 months. The key downside
risks are slower-than-estimated ramp-up in 3G and
rapid subscriber churn in CDMA.
Risks
More spectrum to operators like RCOM and
Tata. Given the corruption probe, we think this
is unlikely to happen without an auction.
Regulator approves MVNOs. Under the
current tariffs, spreads are not wide enough
for MVNOs (mobile virtual network
operators that provide services but do not
have spectrum) to be viable. The failure of
Virgin Mobile supports our argument.
3G used for discounting voice. Telcos have
spent a lot of money to obtain 3G spectrum.
Using 3G for voice will extend the payback
period by several years, more so if discounts
are offered. The pricing on data so far
suggests that telcos have no intention of
playing a value destructive game in 3G.
It is getting tough for the
competition
We believe the competitive landscape is changing given the various
challenges faced by new entrants
MNP has highlighted the weakness for CDMA; improvement in
tariffs positive for RCOM and Tata given challenges on CDMA
Stretched balance sheets, declining investments from new entrants
and structural issues with RCOM and Tata suggest possibilities of
improvement in revenue per minute
Competitive landscape is
changing
Concerns about the Indian telecom sector have been
driven by fragmentation and risks from overcapacity.
Immediately after the issue of licences in 2007, there
was a phase of serious tariff cuts. This was initiated
by GSM incumbents in an attempt to capture
growth, a sort of land grab strategy aimed at leaving
little room for new entrants.
Bharti cut local call rates in early 2008 from INR1.9
to INR1 and later the STD rates by c40%. The rate
cuts were replicated across the sector. Then Tata-
DoCoMo announced it was moving to per second
billing, which cut per minute voice realisations 15-
20%. Something which was introduced as a 90-day
offer soon became a permanent tariff.
The fragmented industry structure and tariff cuts
raised serious questions about sector returns and the
ability of operators to monetise mobile connectivity.
While the stability in revenue per minute over the
past three quarters has raised hopes, with 14 players
still in the market the key question remains the
trajectory of voice pricing.
The outcome of the 3G auctions has changed the
structure of the industry. All seven 3G players are
under equal pressure to invest and take a hit in terms
of recurring 3G amortization charges. At the same
time, recent regulatory developments raise doubts
about the very existence of the eight new players.
Not only have their business models never been
viable in our view, but they now face regulatory and
financial challenges they may not be able to
overcome. We view weakening of competition as a
positive for the Indian wireless space.
What is making us positive?
No headline rate cuts
There have been no meaningful headline rate cuts
since the launch of per second billing by Tata
DoCoMo in September 2009. This is evident from
the stability witnessed in the revenue per minute
over the last three quarters (Figure 10).
Funding becoming tough
Funding for new entrants is becoming tight and this
is positive. According to the Economic Times, S
Tel’s plan to raise cUSD200m seems to be in trouble
as banks appear unwilling to provide finance.
The Central Bureau of Investigation (CBI) inquiry
has revealed instances where banks lent money to
new operators to pay for spectrum in 2008. This
raises the risk of banks being drawn into the 2G
scam enquiry. Public sector banks provided over
INR115,000m to Unitech and STel based on the
licences issued by the Department of Telecom
(DoT). The two licences are among the 85 that
have been declared illegal by the Comptroller and
Auditor General (CAG). Of the INR100,000m
loan to Unitech, the major portion was disbursed
by the State Bank of India (INR80,500m).
Other banks that have lent to Unitech are:
Corporation Bank (INR5000m), Allahabad Bank
(INR5000m), South Indian Bank (INR4,000m),
Canara Bank (INR1,200m), Oriental Bank
(INR700m), Central Bank of India (INR700m),
Punjab National Bank (INR1200m), Standard
Chartered Bank (INR1000m) and Yes Bank
(INR700m). STel got the INR15,380m loans from
IDBI and IDBI Trusteeship Services Ltd, during
between July and November 2009.
In early January 2011, the finance ministry
cleared all banks that lent money to telcos for 2G
licences in 2008. While the Supreme Court has
asked the CBI to broaden the scope of its
investigation into the 2G scam by looking into
bank lending to the sector, the finance ministry
and the central bank have both said there were no
systemic risk issues related to telco lending.
Separately the SBI has tightened rules for loans to
telecom firms by making it mandatory for telecom
companies to submit a utilization plan, a
repayment schedule and details of their service
rollout. The SBI suggested that it will ensure that
the money is used only for the stated purpose.
We note that project-based funding could mean a
rise in the cost of bank loans for telcos and make
it very difficult for the new entrants to raise
money. The situation is such that the telecom
minister has pleaded with the finance minister and
the prime minister to ask banks to lend to the
telecom sector. With the launch of 3G services,
this issue will become even more critical.
In a separate development, new entrant Uninor, a
subsidiary of Norwegian telecom Telenor Group
and Unitech Group, is considering raising
additional capital. There is disagreement between
Telenor and Unitech over the nature of the
financing. Tata-DoCoMo is also considering
raising USD700m and it may try and tap the
equity markets. In our view, the fact that both
Uninor and Tata Tele are looking to raise capital
through a rights issue suggests that external
financing may not be available.
That said, there have been some exceptions. For
example, RCOM recently borrowed USD1.9bn
from China Development Bank (the loan comes
with conditions to buy Chinese equipment). The
funding doesn’t resolve all of RCOM’s problems
and its stretched balance sheet remains a concern
for the company.
Cancellation of licences and payment
for spectrum
Under the terms of the contract, the licensee is
required to cover at least 90% of the area in the
metro service areas of Delhi, Mumbai, Kolkata and
Chennai within a year of the allocation of spectrum.
For non metro areas, it is required to cover 10% in
the first year and 50% within three years.
TRAI has suggested to the DoT that 69 licences
be cancelled in cases where rollout obligations
have not been met. The DoT said it will take a call
shortly on these recommendations. Separately, the
apex court will also decide on the same issue.
In a recent development, the Enforcement
Directorate has filed cases against five companies,
including Swan Telecom, Loop Telecom and
Loop Mobile and STel for alleged foreign
exchange violations. The companies face fines for
allegedly routing cUSD960m into the country
from overseas in contravention of regulations.
We believe these uncertainties will further delay
any expansion plans of new entrants. Moreover, if
the licences are not cancelled, additional fines
cannot be ruled out, further damaging the
prospects of new entrants. The DOT has already
collected cUSD70m in fines from new entrants.
The report from the Comptroller Audit General
(CAG) estimates that the scandal cost USD39bn
in terms of lost government revenue (the CBI
recently put the figure at a more modest
USD6.5bn). The CAG also suggested that many
of the new entrants obtained licences despite
improper documentation.
While the Indian regulator’s decision may not be
final or binding, it raises the possibility of new
GSM entrants having to pay one-time charges for
start-up spectrum and also losing their licences.
We calculate that new operators may be asked to
collectively pay USD6bn for 2G spectrum and
USD9bn for 3G
MNP threat has subsided
Concerns that mobile number portability (MNP)
would disrupt pricing have proved to be
unfounded. Aimed at high end subscribers of
GSM incumbents, MNP has actually turned out to
be positive for three operators.
At the same time MNP has exposed the
vulnerability of CDMA operators. While there is
still a possibility that MNP will still cause minor
disruption, we think the overall impact will be
very limited.
Balance sheets are tight
Post the 3G auctions, the top six players are
operating with tight balance sheets (Figure 15),
limiting price flexibility. This is reflected in stable
revenue per minute and rational 3G prices. Our
analysis suggests that present 3G prices are in line
with global pricing norms (Figure 16).
Murky outlook limits any further
overcapacity
We believe the new entrants are now not only
operating with unviable business models but are
also facing questions about their ability to survive.
This will prevent any serious investments by the
bottom eight players, reducing the risk of further
overcapacity. The industry fragmentation and
overcapacity has not gone down well with
regional and global investors. The possibility of
improvement in revenue per minute will be
viewed as positive for the sector, increasing the
chances of a possible re-rating, in our view.
Pricing theory
An operator making cuts to tariffs needs to be
confident about making a decent profit from
selling cheaper minutes. We do not believe that
pricing is set on the basis of cost plus, but rather
by what might be considered “game theory”.
Challengers will be reluctant to cut tariffs unless it
allows them to steal a march on the established
operators. Otherwise they work on the assumption
that any cuts they make will be matched by the
larger competitors, making it a zero sum game.
If we see things in today’s Indian context, there is
nothing the challengers can do that incumbents
would find it difficult to match. Our analysis
suggests that even the likes of Tata-DoCoMo need
to price outgoing calls at least c10% higher to
achieve EBITDA-breakeven on GSM.
By further cutting tariffs challengers will be only
hurting themselves. For the likes of RCOM and
Tata, taking market share is no longer the main
priority as it has made little impact on their
revenues as tariff cuts were matched by
incumbents. This is supported by the fact that
50% of Tata subscribers and c34% of RCOM subs
are not active (Figure 20).
In our view both RCOM and Tata have a big task on
their hands in terms of achieving EBITDA
breakeven on GSM, investments in 3G and rescuing
CDMA. Our view is supported by the fact that since
per second billing was introduced none of the
challengers have resorted to price cuts.
RCOM and Tata face
structural issues
The biggest challenge for Tata and Reliance is
their low quality subscriber base, which is more of
a structural issue than a question of tactics. The
poor subscriber quality is driven by their late
entry into GSM spectrum in the 1800 MHz and
poor investment in the network.
In our view, price cuts attract teenagers rather
than quality subscribers – hence the high
percentage of inactive subscribers. Both Tata and
Reliance have also missed out on MNP and
instead face continued churn on the CDMA
network which has quality issues.
Idea Cellular did the right thing
The most interesting case study in the Indian
GSM space is Idea Cellular. It started late but,
with the right kind of focus and execution, it has
emerged as a strong pan-India GSM player.
Back in Q3 2008, Idea Cellular was a regional
operator with a presence in 11 areas with revenues
way below that of RCOM, Bharti, BSNL and
Tata. However, Idea’s revenue market share has
jumped by 360bps since then thanks to its
investments in its network (Figure 18).
Figure 20: Active subscribers
Operator Active subs %
Bharti 93%
Idea 90%
Vodafone 78%
RCOM 66%
Aircel 61%
BSNL 54%
MTS 51%
HFCL 50%
Videocon 50%
Tata Tele 50%
Uninor 47%
Loop 44%
STel 41%
MTNL 37%
Etisalat 34%
Source: TRAI
CDMA getting weaker
MNP exposed the weakness of CDMA networks.
This particularly applies to Tata and RCOM, two
operators that still have little traction in GSM. Given
the various challenges they face, we believe Tata and
RCOM have little choice other than to try to increase
the rate per minute.
We are not worried about Aircel as they have not
used tariffs to gain market share. They are the only
new entrant to use a VAS strategy to which the
incumbents could not respond in the 2G
environment, given the spectrum crunch. However,
the launch of 3G services will allow the incumbents
to respond to the VAS offered by Aircel.
Dual SIM phenomenon dying down
The basis for the dual SIM phenomenon was
arbitrage between the tariffs provided by the new
entrants and incumbents. As incumbents have
responded to the tariffs of new players, the arbitrage
seems to have narrowed. That said, the likes of
Uninor are still focused on the dual SIM strategy
through their location based discount schemes.
However, the fact that it has gained little traction
suggests that dual SIM is no longer going to be a
viable strategy.
Voice revenue to improve
We expect voice revenue to improve in the medium term; we
expect the first signs of this to be visible from 3Q FY12
We believe this trend will be driven by investments in 3G, the low
quality of new subscribers and limited upside to MOUs
Revenues can also be improved by reducing freebies such as
special vouchers and free airtime
Tariffs should improve – but
how and when?
We believe the market has a very pessimistic view of
the current tariffs. Those who are bullish are betting
on 3G, which we don’t expect to be earnings
accretive any time in the near term. The stability in
revenue per minute is being viewed as a good
downside support but the possibility that there are
chances of tariffs moving up is being completely
ruled out. We believe that pricing power can return
to the sector and revenue per minute should improve
over the next 3-4 quarters.
Now this need not take the route of big
movements in the headline rates. Given the mood
of the Indian regulator, we don’t believe that
operators will do anything drastic. In our view
there are other ways to boost revenue per minute
and tariff upside will be driven by:
Investments in 3G.
The fact that subscriber additions are not very
value accretive.
3G in its early stages needs to be supported
by growth in voice business.
MOU growth is on a declining trend.
While the sector has benefited from stellar
subscriber growth, the impact on revenue from
new subscribers has been limited. New
subscribers are not only poor in quality, but also
relatively expensive to acquire. In fact it is the
retailers that have benefited from the competition
for subscribers. The Indian telecom space has
added about 362m subscribers since June 2009
(when Tata Docomo launched the per second
billing plan) but the corresponding revenue has
been only cUSD890m, an increase of 13% versus
subscriber growth of c46% (annualised revenues
for the sector are estimated at USD27bn).
Note that a large part of this 13% revenue growth
came from higher usage by older subscribers. The
strategy will now shift from subscribers to
extracting more revenue from the existing set of
subscribers via voice and usage. We are of the
view that a large part of the usage or minutes in
the system today are inelastic and this allows
telcos to take away freebies and tweak various
offers to improve the revenue per minute.
3G will require steady investment
We expect most incremental network capex to focus
on adding to operators’ existing 3G systems. 3G will
require significant investment going forward, both in
backhaul and the core network. The focus of capital
investment, however, remains a contentious point.
With 3G, we believe all aspects of the network
should see higher spending.
While the Indian voice network coverage is stable
(at least with the incumbents, leaving aside the
spectrum crunch issues), subscribers will have
similar expectations about the 3G network. This
will put pressure on the 3G winners to invest in
networks or they may fail to capture the data
growth. We believe the need to invest in 3G
networks will not only drive rationality but also
raise the need to work on improving realised rates.
Voice revenues have to support 3G
We believe 3G business is EPS dilutive for at least
the next 3-4 years for the overall sector and for every
player. Our analysis suggests the industry requires
c90m subscribers to achieve EBITDA-breakeven on
3G, with monthly ARPU assumption of cINR175-
200. Given this, there are very limited options other
than to improve voice revenues.
MOU growth on a declining trend...
becoming inelastic in nature
Our analysis suggests that the MOU per
subscriber are on a declining trend (Figure 22).
We are modelling for an annual c2% decline in
sector MOUs from FY11 to FY15e.
If we were to treat the subscribers with Bharti at
end of March 2009 as old subscribers and all
incremental subscribers as marginal, our study
suggests that MOUs of the existing subscribers
have grown by 46% in the past seven quarters and
grown at a quarterly CAGR of 6%. This suggests
that the growth in minutes of usage is being driven
by the old subscribers, and new subscribers are
MOU dilutive. For the purpose of our analysis we
are assuming that MOU of marginal subscribers
have declined from 260 to 190 per sub.
Analysis with similar assumptions for Idea Cellular
suggests that MOU per sub of old subscribers have
increased 80% over the last seven quarters and
grown at quarterly CAGR of 9%.
Analysis for both telcos suggests that MOUs from
old subscribers have been phenomenal in the past
seven quarters, but we believe this trend cannot
continue for much longer. While the usage growth
will continue, the rate of growth should decline in
our view. As such we believe that we have
reached a scenario where the bulk of usage is
inelastic in nature and there is not much left for
the telecom operators to achieve by cutting rates.
Instead, some tweaking of rates will help the
telcos to benefit at the net level. Bears may argue
that telcos may opt to continue with present tariffs
or provide discounts in an attempt to boost usage
from the marginal subscriber segment. We believe
that limited purchasing power is the biggest
challenge in this regard and telcos will get little
benefits from such a strategy.
How will rpm improve?
We don’t expect the improvements in revenue per
minute to be immediate. The MNP overhang may
linger for a couple of quarters but in our view
some investors may miss out on the first stages of
improving rpm, which we believe could happen in
the following possible ways:
Withdrawal of free airtime
Free talk time pulls down the realized rates as
paid minutes are replaced by non-paying minutes.
Withdrawal of free airtime will result in an
improvement in effective realized rates and
should be positive.
Withdrawal of special vouchers
Withdrawal of discounts (Figure 23) could result
in improvements in revenue per minute. We
believe night calling plans or other offers which
lead to network utilization in non-peak hours will
continue. But there is huge amount of discounts
on the on net calls and SMS tariffs and we believe
some of these plans can be tweaked.
Improvement in headline rates…this will be the
last to come
Last, not but least, we expect an announcement on
headline rates at some point. This is unlikely to
happen in the near term as it will be difficult to
get past the regulator in the current climate. We
don’t expect any announcement until the
operators have complete clarity on the policy
framework, particularly on M&A policy.
Telcos are already charging for call
centre calls
There are already a few promising signs. Telcos
have been charging subscribers for calls made to
call centres for 9-10 months. Separately, the likes
of Bharti have been charging SMS interconnect
fees. Both these examples highlight the various
possibilities which exist for operators to improve
revenue per minute.
Potential surprises
A complete withdrawal or tweaking of per second
billing plans. For instance, operators could bill for
every two seconds. For this to happen, some kind
of industry consensus will be required. A
withdrawal of per second billing could improve
effective rates by 10-15% and it would be easier
to get this past the regulator than a rise in headline
rates. We note India has a policy of forbearance
and tariffs are driven by free market mechanisms
and the regulator at present has no say on tariffs.
VAS will grow
Another aspect where we are looking for growth
is VAS. While 3G services will cater to high end,
the 3G ecosystem is likely to throw up various
niche products which were completely absent in
the 2G regime. The 3G network comes with two
advantages – the ability to do video calling and
faster download speeds. Apart from these two
services, there are not many services/products
which cannot be provided on a 2G platform.
The present VAS contribution to overall revenues
is c14%; 80% of the present revenue comes from
caller tunes and 20% from data services. With the
3G product portfolio aligned on the data side,
there is a solid case for 2G VAS revenues to
benefit from the development of the 3G
ecosystem. We believe that going forward the
contribution from data-centric products should
drive growth in the 2G VAS space and we see
contribution from data-specific products moving
from the present c20% to 50% by FY15e.
The regulatory climate
Regulations cannot work in isolation; the policy framework needs to
create sustainable revenue streams aligned with the competitive
landscape
We don’t expect the Indian telecom regulator to control tariffs and
believe it will continue with a policy of forbearance
There is a need for reforms to sustain tax receipts and generate
benefits from steady spectrum auctions
The sector’s financial health
There is a huge amount of scepticism about various
regulatory issues facing the Indian telecom sector.
But before we get into those discussions let’s
understand the relative importance of the telecom
sector for the Indian government. Our analysis
suggests that the government – excluding spectrum
auctions – collected a total USD5bn in taxes from
the sector and cUSD24bn from spectrum fees last
year. While it is very apparent that the Indian
government is very much focused on raising taxes
from the sector, we believe government may lose out
in the longer term by not paying attention to the
sector’s health. The government is in the process of
drafting the National Telecom Policy 2011, and it is
important this issue is taken into consideration.
The sector is in a poor state overall and, except for
Bharti, all the telecom operators are working below
their cost of capital, in our view. ROEs have come
down for Bharti as well, suggesting that incremental
prepaid business is challenging and diluting the
overall ROE.
While the raising of USD15bn from 3G spectrum
auction was good news for the Indian government,
we believe the ability of government to benefit from
such high premium on 3G spectrum was driven by
presence of 14 players. However, this cannot be
sustainable and in our recent regulatory
developments will reduce the number of players to
6-7 in the near to medium term.
The 3G auctions have raised two questions:
Why was no single player bidding for 3G
spectrum pan India, despite a pan India
2G presence?
Why was no new foreign player investing
in 3G?
Add to this USD10bn investments made by Bharti
in Africa, there seems to be some suggestion that
Indian telcos are a bit hesitant to place bets in the
Indian telecom space given the present regulatory
environment. We believe a more conducive
environment and a policy framework which
considers the financial health of the sector will be
positive and also maximise the tax collection for
the government
Time to create sustainable revenue
stream
The best piece of policy so far by the Indian
government has been the consideration of a revenue
sharing regime for the telecom sector. This
encouraged participants to invest in telecom
infrastructure. However, the industry fragmentation
in 2007 caused by the issue of eight new licences
resulted in a loss of USD998mn to the government
as the revenue per minute declined by c40%. This is
over and above the loss from the fact that 2G
spectrum was not put out to auction (CAG estimates
the loss at USD39bn).
The Indian government has set some tough fiscal
deficit targets and it goes without saying that it needs
to continuously invest in the economy to drive
growth in excess of 7-8%. This creates the need for
the government to create sustainable revenue
streams from an emerging sector like telecoms.
Globally, continuous auction of spectrum has been
the source of incremental revenues from the telecom
sector. Post realizing USD23bn from the 3G and
BWA spectrum auctions, government has budgeted
cUSD4bn as non-tax revenues from the telecom
sector this fiscal year. This is good news and we
believe government plans to raise this money by way
of spectrum sales, possibly auctions of 2G spectrum.
We believe that in the Indian context spectrum
auctions can prove useful to the government. Taking
the US as an example, the regulator, the FCC, has a
very clear roadmap in terms of what it needs to refarm
and how. For instance it plans to re-farm about
300MHz of spectrum in the next five years and the
plan also lays out the various bands which will meet
the requirement.
The fact that in India, 40% of the total available 2G
spectrum and 60% of the total 3G spectrum is not
with the government but with the Indian defence
forces or other government sectors, highlights that
Indian telecom regulator needs to have a longer term
perspective and a robust re-farming policy in place.
The process of spectrum re-farming is very complex,
long and expensive.
Data will drive demand for spectrum
Incremental demand for spectrum will be driven
by both voice and data. But what will create more
demand is data and not voice from a medium term
perspective. The situation in India could be even
worse than global conditions as the first release of
3G spectrum was limited and none of the operators
have pan India 3G spectrum. Most of them will be
keen to ramp up with a pan India presence, in our
view. The spectrum is 5MHz for all operators
However, it will take time before the telcos make
money from 3G and their ability to participate in
auctions for 3G spectrum will have to be supported
by the cash flows from the voice business. This
again raises the significance of improvement in
revenue per minute and how indirectly this will aid
the government as well. Any policy change by the
government with respect to tariffs may prove to be
counterproductive and as such we believe the Indian
telecom regulator/government will be better placed
by sticking to its current approach of forbearance.
Regulator has to depend on
incumbents, why?
Bears will argue that the government may decide not
to depend on incumbent operators and may
encourage fresh investments in the market. We don’t
think so; there was hardly any participation in 3G by
any additional foreign player. And the failure of
Hutch3’s standalone 3G business is a great exampleTime to create sustainable revenue
stream
The best piece of policy so far by the Indian
government has been the consideration of a revenue
sharing regime for the telecom sector. This
encouraged participants to invest in telecom
infrastructure. However, the industry fragmentation
in 2007 caused by the issue of eight new licences
resulted in a loss of USD998mn to the government
as the revenue per minute declined by c40%. This is
over and above the loss from the fact that 2G
spectrum was not put out to auction (CAG estimates
the loss at USD39bn).
The Indian government has set some tough fiscal
deficit targets and it goes without saying that it needs
to continuously invest in the economy to drive
growth in excess of 7-8%. This creates the need for
the government to create sustainable revenue
streams from an emerging sector like telecoms.
Globally, continuous auction of spectrum has been
the source of incremental revenues from the telecom
sector. Post realizing USD23bn from the 3G and
BWA spectrum auctions, government has budgeted
cUSD4bn as non-tax revenues from the telecom
sector this fiscal year. This is good news and we
believe government plans to raise this money by way
of spectrum sales, possibly auctions of 2G spectrum.
We believe that in the Indian context spectrum
auctions can prove useful to the government. Taking
the US as an example, the regulator, the FCC, has a
very clear roadmap in terms of what it needs to refarm
and how. For instance it plans to re-farm about
300MHz of spectrum in the next five years and the
plan also lays out the various bands which will meet
the requirement.
The fact that in India, 40% of the total available 2G
spectrum and 60% of the total 3G spectrum is not
with the government but with the Indian defence
forces or other government sectors, highlights that
Indian telecom regulator needs to have a longer term
perspective and a robust re-farming policy in place.
The process of spectrum re-farming is very complex,
long and expensive.
Data will drive demand for spectrum
Incremental demand for spectrum will be driven
by both voice and data. But what will create more
demand is data and not voice from a medium term
perspective. The situation in India could be even
worse than global conditions as the first release of
3G spectrum was limited and none of the operators
have pan India 3G spectrum. Most of them will be
keen to ramp up with a pan India presence, in our
view. The spectrum is 5MHz for all operators
However, it will take time before the telcos make
money from 3G and their ability to participate in
auctions for 3G spectrum will have to be supported
by the cash flows from the voice business. This
again raises the significance of improvement in
revenue per minute and how indirectly this will aid
the government as well. Any policy change by the
government with respect to tariffs may prove to be
counterproductive and as such we believe the Indian
telecom regulator/government will be better placed
by sticking to its current approach of forbearance.
Regulator has to depend on
incumbents, why?
Bears will argue that the government may decide not
to depend on incumbent operators and may
encourage fresh investments in the market. We don’t
think so; there was hardly any participation in 3G by
any additional foreign player. And the failure of
Hutch3’s standalone 3G business is a great example
of how telcos have realized that there is no
standalone 3G business case. We believe that to
create sustainable revenue streams from spectrum
auctions, the regulator will need to rely on the top
five/six operators.
Post 3G, all the top five players have taken on a lot
of debt. So the government’s ability to monetize
spectrum auctions may also be limited.
Risks to our calls / what we
may be missing
More spectrum issued to new
entrants Tata and RCOM
We believe it will be difficult for the government to
issue spectrum to any player without an auction. But
the new entrants argue that licence conditions allow
the minimum spectrum up to 6.2 MHz and, as such,
there are possibilities of more spectrum going to new
entrants, or at least RCOM and Tata.
Bears may argue that additional capacity creates the
risk of another tariff war. We don’t think there is risk
to tariffs if more spectrum is allocated to the likes of
RCOM or Tata because their GSM networks are not
suffering from undercapacity, so they are unlikely to
cut tariffs. Separately, we believe more spectrum
will be the last chance for both RCOM and Tata to
position for high-end subscribers, rather than playing
another tariff-disruptive move and giving away
capex savings. We believe they are likely to be
rational and focus on structural issues.
MVNO approach
There have been some suggestions that the
government may decide to boost MVNOs to save
the business models of new entrants. We don’t think
that the MVNO approach works in low cost, low
tariff markets like India. With call rates at INR0.5,
termination at INR0.2, the spread is very thin for
operators with low scale to operate. Virgin mobile is
a clear example, having failed to create any impact
so far in any format. As we suggested, for new
entrants outgoing tariffs should be at least c15%
premium to the present tariff levels to be EBITDA
breakeven, it highlights that if new GSM entrants
decide to share a percentage from the tariffs earned
they will only delay their profitability.
Bears may further argue that new GSM entrants may
get into contracts with MVNOs to earn one time
fees, as their network costs are fixed and this allows
them some breathing space.
In this context we would like to highlight quality
issues. The biggest problem for RCOM and Tata in
GSM has been late entry and their GSM spectrum in
1800 MHz. It will be either the new entrants (which
have hardly any network to offer) or the likes of
RCOM and Tata which may adopt the MVNO route.
With spectrum in GSM at a mere 4.4MHz, it is not
enough to share/lease capacity. At the end of the day
MVNOs will be dependent on the 1800MHz of the
new entrants which is poor quality.
VOIP
VoIP has been a concern for investors in mobile for
several years. However, the inroads that it has made
into voice revenues have been relatively slow,
largely thanks to technical issues. The highest profile
VoIP service remains Skype, which has an
enormous following on desktop and laptop
platforms. Operators with very little market share
like 3UK have been pushing Skype. But note that
calls made via Skype over 3UK’s network are –
confusingly– not actually VoIP calls. The call is
circuit switched over 3UK’s radio network, and only
once it reaches the core does it become IP. This
should serve to underline the difficulties inherent in
getting VoIP to work adequately in a mobile
environment – certainly to the degree of quality of
service to which users will be accustomed.
However, VOIP is coming into its own with LTE.
Because this platform is all-IP, it has no capability to
carry switched voice calls; hence all voice
communication must be carried by VoIP. Therefore
the timing of the mass market LTE is potentially
most significant for VOIP to make an impact on the
voice tariffs.
Conversely, the longer it takes for LTE-powered
VoIP to arrive, the better the industry’s prospects for
stable pricing. However, the threat is smaller in
emerging markets like India, where the tariffs are
already the lowest and the GSM ecosystem is strong
with robust handset choices.
Further it will take a lot of work to get a VoIP LTE
solution as reliable as 2G and 3G platforms. LTE is
an all-IP platform, which does not have channels
dedicated to carrying voice traffic. Instead of the
circuit switched approach, it uses an end-to-end IP
connection from the user’s device back into the core
network. In the early LTE standards releases, it is
clearly data rather than voice that is the priority.
VoIP did not even make it into LTE as standardised
in 3GPP UMTS Release 8, albeit in part because the
relevant technology is simply not ready. There are
various alternative solutions for LTE VoIP, though
only two are currently approved within the 3GPP
standard. These are voice over IMS (VoIMS) and
circuit switched fall-back (CSFB).
The former was the preferred choice. However, it
relies on IMS-based core networks, and has run into
real difficulties given the failure of IMS to gain
momentum. The 3GPP designated alternative for
voice over LTE is CSFB, which is – quite simply, in
our view – a compromise, routing the call instead
over the switched network via 2G or 3G.
Furthermore, far from being a pragmatic short-term
alternative, it would appear to have some grave
practical defects. For instance, it is likely to involve a
very significant increase in call set-up times – as the
handset must start up its 2G or 3G radio and connect
to the appropriate cell site at the appropriate
frequency. Plus there is a good chance that any LTE
data connection running in the meantime will be
dropped (just as 2G to 3G transitions – or vice versa
– often result in problems). There could also be
some very deleterious implications in terms of
battery life. Given this, we believe it will still be
some time before VOIP becomes a meaningful
threat to voice tariffs.
Our global team argued in Frequonomics that the
rise of VoIP over LTE would likely take years, and
we have heard little in the interim to change this
view. The challenge here is for the handset industry
to get multi-mode devices capable of supporting 2G,
3G and 4G technologies, with adequate battery life
and with the necessary on-board multi-tasking
processing capability to handle VoIP as one of a
number of concurrent user applications.
3G used for discounting voice.
3G spectrum is six times more efficient than 2G and
as such per unit of capacity created with 3G
spectrum is much cheaper. This may be attractive
enough for the likes of RCOM and Tata to use 3G
for voice and go ahead with cuts in tariffs, given that
the cost of producing minutes in 3G is very low.
While the proposition may seem logical, one needs
to factor in that telcos need to recover the price paid
for 3G spectrum as well. If 3G is used for voice, the
payback period for the operators may take decades,
more so if that too is discounted from the present
levels. Moreover, the subscriber volume on 3G
platform will be restricted to those with 3G enabled
handsets. And these will be the high end subscribers,
less sensitive too lower tariffs on voice, suggesting
that this strategy could fail.
TRAI recommendations
We don’t expect the TRAI recommendations to be accepted in the
current format
We believe they are a major shift from the present policy framework
Re-farming of spectrum in 900 MHz is the most significant
recommendation; we believe it will be not easy for the regulator to
implement this and that it faces several challenges
Unlikely to be accepted in the
current format
While the probe into the 2G scam has been negative
for new entrants, the TRAI recommendations have
created problems for the top GSM players, namely
BSNL, Bharti, Idea and Vodafone.
We believe they represent a major shift in thinking.
To us, it appears that TRAI wants to change the rules
midway through the process, after billions of rupees
have already been spent. The recommendations
could lead to another round of legal battles which
would bring uncertainty and adversely impact future
investment in the sector. We don’t believe the Indian
government can afford such an outcome.
One-time payment for excess
spectrum beyond 6.2 MHz
We believe it will not be easy for TRAI to get a
consensus on this issue. The spectrum in excess of
6.2 MHz has been obtained by the incumbents by
meeting subscriber-linked criteria. Given this, we
believe this recommendation could be challenged
by GSM incumbents. Despite this, if the
recommendations are implemented, GSM
incumbents may decide to give up the excess
spectrum in select markets to avoid
financial damage.
Separately, GSM operators don’t believe in the
concept that there should be a maximum amount
of spectrum a single operator can hold. We are not
aware of any other market which sets an upper
limit to the spectrum an operator can hold.
Spectrum pricing
Another related concern is spectrum pricing,
which has implications for both the licence
renewal issue and the payout for spectrum in
excess of 6.2MHz. TRAI recommended 3G
spectrum prices as the benchmark for arriving at
the payout for the operators in both instances.
In a recent development, TRAI revisited its
estimates on spectrum pricing and the latest
recommendations have lowered the payout for
metros and increased the payout for operators in
non-metros. In some of these non-metros the
spectrum payout exceeds the payment for 3G
spectrum. The underlying rationale is that the
non-metros have less of a data business case as
their markets are predominantly about voice;
hence, 2G spectrum generates higher payouts in
such markets.
We believe pure voice markets have upside to
margin, whereas data in the medium term can
allow operators to improve margins. Moreover,
the spectrum pricing recommendations have been
strongly criticised by the GSM operators and we
think it is unlikely that they will be accepted in
the current format.
Licence renewals
The TRAI has suggested very stringent conditions
for the renewal of licences, including:
At the time of renewal, spectrum held by the
operators in the 900MHz band to be replaced
by an equal amount of spectrum in the 1800
MHz band.
The renewal of existing licences for another
10 years instead of 20 years and one-time
payment for the entire spectrum which is
renewed along with the license
We believe spectrum re-farming should be
given a longer time frame; it should be based
on price incentives. If someone is ready to get
spectrum re-farmed then that operator should
be given certain price incentives. But
reframing should be done with a certain
degree of facilitation caveats rather than
penalties, which is the case with the TRAI
recommendations in our view
Furthermore, TRAI recommendations in the context
of spectrum pricing suggest that 900 MHz spectrum
is more efficient than 1800 MHz and as such the
spectrum in the 900 MHz band should be charged at
a premium to the spectrum in the 1800 MHz band .
Given the same rationale, the licence renewal
conditions should be factoring in the same approach
and as such those operators who give up spectrum in
the 900 MHz band must be compensated with 50%
more spectrum in the 1800 MHz band. However,
the TRAI recommendations fail to address
such inconsistencies.
We believe the limitations of the dynamics of timing
cannot be an excuse; operators which came in first
got spectrum in the 900 MHz band, and those which
came in later got spectrum in the 1800 MHz band.
When operators were allocated spectrum they knew
what they were getting and taking away the
spectrum now from early movers into the sector may
not send positive signals to other industries and may
impact foreign investment as well.
Overall, we believe that refarming spectrum will
be complex, time-consuming and expensive. It
could also have a significant adverse impact on
the sector as it would lead to additional capex.
Our view is that the complexities involved in the
process of re-farming will make it a next to
impossible process for the government. There will
be a serious amount of subscriber discomfort and
possibilities of serious disruption to essential
telecom services cannot be ruled out.
We believe the Indian telecom regulator needs to
take these factors into consideration before finalising
any policy changes.
The biggest losers of the re-farming process will be
BSNL and MTNL as these operators have 900 MHz
spectrum across India. They would face higher
capex and their business viability may suffer
significantly. With the financial performance of the
PSUs already a worry, we believe the government
will find it difficult to introduce the refarming policy
suggested by TRAI.
BSNL is having a very difficult time. In FY10, the
company posted its first ever loss of INR182bn. In
an effort to balance its social commitments of
providing services in rural areas and competing
with private players in urban areas, its
profitability has taken a severe hit (Figure 26).
Also, the company had to pay INR1,850bn for 3G
and broadband wireless spectrum in FY2011,
reducing its cash balance to cINR1,150bn
excluding losses in FY11. If BSNL applies for
900 MHz licence in all its 23 circles then there is
a very high probability that licence fees coupled
with ongoing operational losses will result in a
cash deficit. Hence, we believe that the
government will structure 900 MHz licence fees
to take into consideration BSNL’s cash position.
Leaving out PSU from re-farming and only
making it valid for private players will not be easy
and will only lead to litigation, in our view.
Last, but not least, to implement re-farming the
Indian government would require c434Mhz of 2G
spectrum in the 1800 MHz, or about 20MHz across
each circle. If new players prevail and there are no
licence cancellations, total spectrum required would
be 736MHz, or c33MHz across each circle.
Assuming nothing comes from defence, the shortfall
will be c522 MHz, and if c20 MHz comes from
defence in each circle, the shortfall would still be
c90MHz. It is well known how cumbersome it has
been for the telecom ministry to obtain a mere 15
MHz in 3G from the defence department.
To sum up, regulators across other markets have
simplified rules and if the TRAI recommendations
are accepted than it would be one of the few nations
with such stringent telecom regulations. Examples
that support our view are mentioned below:
In the United States, licences are issued for a
10-year period but are renewed routinely at a
nominal cost
In the UK and the Nordic region, annual
licence fees are not applicable for the licences
awarded by auction
India Equity Research Reports, IPO and Stock News
Visit http://indiaer.blogspot.com/ for complete details �� ��
It’s time to focus on Bharti’s revenue
growth, not negative sector newsflow
It offers the best potential return of all
India large caps; competition, regulatory
risk-reward profile also improving
Upgrade Bharti to OW and raise TP to
INR425 from INR370
Look beyond the headlines. The backdrop to this
overpopulated, troubled industry is messy, even if, unlike
consensus, we think the worst is over for revenue per minute
after a decline of 30% in the last two years. So, perhaps not
the best time to buy the biggest stock in the sector? We
disagree. Market leader Bharti offers the best potential return
among Indian large caps at a time when competitive and
regulatory risks for the market leader are also diminishing.
Focus on earnings growth at Bharti. We estimate FY12e
earnings growth at c27% and FY13e at c37% (versus c18%
for Sensex over the next two fiscal years and c20% earnings
growth at other large caps). A large part of the growth comes
from improvements in voice pricing strategy, a pick-up in
2G value-added services and the company’s tower business.
The rate of tariff decline has been falling and we expect this
to improve further and revenue per minute to also rise.
Regulatory and competitive risks diminishing. The
government needs a healthy telecom sector to generate
revenue from taxes and spectrum auctions. Bharti stands to
benefit most from reforms that we think will help weed out
weaker rivals, many of which already face huge additional
fines and extra costs for contract infringements related to the
2G licence scandal. For example, it is now much tougher for
the new entrants to raise financing. Bharti is well positioned
to benefit from the inevitable consolidation that must follow,
even if the process takes a while.
Top pick: Bharti. We upgrade Bharti to OW and raise our
TP to INR425 (from INR370). We remain N(V) on Idea and
RCOM. Our TPs for all three stocks factor in the possible
negative impact of recommendations by the regulator, TRAI.
These include charging incumbents for excess spectrum as well
as a one-time payment when they renew licences and the refarming
of 900 MHz spectrum.
Summary
Focus on Bharti’s revenue growth, not negative sector newsflow
Unlike consensus, we think the worst is over for revenue per
minute and expect improvements over time
Upgrade Bharti to OW and raise our TP to INR425 (INR370);
reiterate N(V) on Idea Cellular and RCOM
A brighter outlook
The fragmented industry structure and the issue of
new licences in 2007 have raised serious questions
about sector returns and the ability of operators to
monetise mobile connectivity. However, things are
looking up. Positive signals include the fact that
there have been no major cuts in tariffs since 2010
and revenue per minute has stabilised over the past
two quarters. Moreover, sector revenues improved
by 5% in 3Q FY11, driven by robust increases in
minutes of use (MOU) – the highest growth since
December 2008.
We believe the competitive landscape is changing.
All eight new players which entered the sector in
2007 are facing significant financial and
regulatory challenges. Since the probe into the 2G
spectrum allocation scandal was announced,
several have been fined for failing to meet rollout
obligations and some may even face the
possibility of having their licences cancelled. The
mere presence of the eight newcomers has kept
tariffs in check but we expect this trend to change
– but it will take time.
A range of positive factors
Over and above the challenges faced by new
entrants, there are several other factors driving our
positive view. These include:
MNP is now behind us without causing any
meaningful damage.
The winners of 3G licences will need to invest
in their networks to improve voice revenue.
Limited upside to MOU per subscriber and
the fact that simply adding subscribers is no
longer the solution. Since June 2009, the
sector has added 348m subscribers but
incremental revenue has been only USD840m
– revenue growth of 13% compared to
subscriber growth of 46%. Operators need to
extract more revenue from voice subscribers.
Expect policy reforms
While regulatory risks remain (both from the 2G
corruption probe and TRAI spectrum pricing
recommendations), the telecom sector is
significant for the government coffers. Tax
receipts are estimated at USD5bn and the
government also raised cUSD24bn via the
spectrum auctions in the last fiscal year. Add to
this the fact that the government is targeting a
fiscal deficit of c4.6% in FY12 and there will be
constant need to bring the deficit down going
forward. Against this backdrop, we expect the
government to introduce policy reforms to create
sustainable revenue streams from the sector.
We estimate the telecom sector ROE is 5%,
significantly below average returns from other
sectors (see table). We believe it is important for
the regulator and the government to pay attention
to the financial health of the sector. Given this, we
see little chance of TRAI’s recommendations
being part of the new telecom policy the
government plans to deliver this year.
Moreover, we believe TRAI’s recommendations
are a major shift from current policies and are
unlikely to be accepted by the Department of
Telecoms (DoT) in their present form. That said,
our target prices for all three stocks have been
adjusted to take into account possible damage
from TRAI’s recommendations.
Tariffs
We believe the market has a very pessimistic view
on tariffs. The bulls are betting on 3G, which we
don’t expect to be earnings accretive in the near
term. The stability in revenue per minute is being
viewed as a good downside support but the
possibility of tariffs moving up is being
completely ruled out. We believe that pricing
power can return to the sector and revenue per
minute should improve over the next 3-4 quarters.
Now this need not take the route of big
movements in the headline rates. Given the mood
of the Indian regulator, we don’t believe that
operators will do anything drastic in the near term,
preferring tweaks to current subscriber plans.
Valuation
BHARTI, OW, TP INR425
We are turning positive on Bharti, upgrading to
OW and raising our target price to INR425 from
INR370. We roll our multiples forward to FY13e,
in line with our methodology on other Indian
stocks under coverage. Our target price is adjusted
by an additional INR18 to factor in TRAI
recommendations, taking the total negative impact
to an estimated INR46.
We believe Bharti is the best way to play the
brighter outlook given its quality subscribers,
generation of c2bn daily minutes on its wireless
platform, deep coverage and network advantage.
Our new target price implies a PE of c16x on
FY13e EPS. The stock is currently trading at 13x
on our FY13e estimates. Our numbers for FY13e
are c2% above consensus. Bharti’s 12-month
forward average trading multiple has been 14x.
As we factor in higher depreciation charges on
account of 3G we have cut our FY13e earnings by
c4.5%. We estimate FY12e earnings growth at
c25% and FY13e at c35%. A large part of the
growth comes from improvements in voice rates,
a pick up in 2G VAS and scale benefits from the
tower business. We remain cautious on the Bharti
Africa business and value it at -INR33 per share.
Upsides risks to our call include the possibility of
improvement in headline tariffs, visibility on
regulations pertaining to sector consolidation and
faster than estimated pick up in 3G. Further ability
to monetize tower assets also remains a key
upside risk to our valuations. News reports
suggest that both Vodafone and Idea may consider
merging their residual towers in the Indus Tower
Company; this should be positive for valuations.
Downside risks include the use of the 3G platform
by challengers for voice, a slower than estimated
pick up in 3G services and acceptance of TRAI
recommendations in the current format.
Idea Cellular, N(V), TP INR72
We retain our Neutral (V) rating on Idea Cellular
but cut the target price from INR75 to INR72 as
we make adjustments for the possible acceptance
of TRAI recommendations. Idea is most
vulnerable to TRAI recommendations and as such
we see limited upside. Our new target price
implies PE of 29x on FY12e EPS and 20x on
FY13e EPS. The stock trades at c19x on our
FY13e estimates. Our numbers for FY12e are 3%
above consensus and 8% below for FY13e.
Upsides risks to our call include the possibility of
improvement in headline tariffs, visibility on
regulations pertaining to sector consolidation and
faster than estimated pick up in 3G. Downside
risks include the use of the 3G platform by
challengers for voice, slower than estimated pick
up in 3G services and acceptance of TRAI
recommendations in the current format.
RCOM, N(V), TP INR122
We retain our target price of INR122 and Neutral
(V) rating. Our target price implies PE of 16x on
FY12e EPS and 12x FY13e EPS. The stock is
currently trading at c11x on our FY13e estimates.
Our numbers for FY12e are in line with consensus
but FY13e numbers are c9% below. While tariff
improvements should be positive for the company,
the upside is capped given poor subscriber quality
and weakening of the CDMA business with MNP.
Separately, the possible sale of tower assets remains
the only catalyst in the near term; we believe
unlocking tower assets will be difficult for the
company, given low network utilization on both
GSM and CDMA.
The key upside risks are the ability to monetize
tower assets and sell a 26% stake to a strategic
partner over the next 6-9 months. The key downside
risks are slower-than-estimated ramp-up in 3G and
rapid subscriber churn in CDMA.
Risks
More spectrum to operators like RCOM and
Tata. Given the corruption probe, we think this
is unlikely to happen without an auction.
Regulator approves MVNOs. Under the
current tariffs, spreads are not wide enough
for MVNOs (mobile virtual network
operators that provide services but do not
have spectrum) to be viable. The failure of
Virgin Mobile supports our argument.
3G used for discounting voice. Telcos have
spent a lot of money to obtain 3G spectrum.
Using 3G for voice will extend the payback
period by several years, more so if discounts
are offered. The pricing on data so far
suggests that telcos have no intention of
playing a value destructive game in 3G.
It is getting tough for the
competition
We believe the competitive landscape is changing given the various
challenges faced by new entrants
MNP has highlighted the weakness for CDMA; improvement in
tariffs positive for RCOM and Tata given challenges on CDMA
Stretched balance sheets, declining investments from new entrants
and structural issues with RCOM and Tata suggest possibilities of
improvement in revenue per minute
Competitive landscape is
changing
Concerns about the Indian telecom sector have been
driven by fragmentation and risks from overcapacity.
Immediately after the issue of licences in 2007, there
was a phase of serious tariff cuts. This was initiated
by GSM incumbents in an attempt to capture
growth, a sort of land grab strategy aimed at leaving
little room for new entrants.
Bharti cut local call rates in early 2008 from INR1.9
to INR1 and later the STD rates by c40%. The rate
cuts were replicated across the sector. Then Tata-
DoCoMo announced it was moving to per second
billing, which cut per minute voice realisations 15-
20%. Something which was introduced as a 90-day
offer soon became a permanent tariff.
The fragmented industry structure and tariff cuts
raised serious questions about sector returns and the
ability of operators to monetise mobile connectivity.
While the stability in revenue per minute over the
past three quarters has raised hopes, with 14 players
still in the market the key question remains the
trajectory of voice pricing.
The outcome of the 3G auctions has changed the
structure of the industry. All seven 3G players are
under equal pressure to invest and take a hit in terms
of recurring 3G amortization charges. At the same
time, recent regulatory developments raise doubts
about the very existence of the eight new players.
Not only have their business models never been
viable in our view, but they now face regulatory and
financial challenges they may not be able to
overcome. We view weakening of competition as a
positive for the Indian wireless space.
What is making us positive?
No headline rate cuts
There have been no meaningful headline rate cuts
since the launch of per second billing by Tata
DoCoMo in September 2009. This is evident from
the stability witnessed in the revenue per minute
over the last three quarters (Figure 10).
Funding becoming tough
Funding for new entrants is becoming tight and this
is positive. According to the Economic Times, S
Tel’s plan to raise cUSD200m seems to be in trouble
as banks appear unwilling to provide finance.
The Central Bureau of Investigation (CBI) inquiry
has revealed instances where banks lent money to
new operators to pay for spectrum in 2008. This
raises the risk of banks being drawn into the 2G
scam enquiry. Public sector banks provided over
INR115,000m to Unitech and STel based on the
licences issued by the Department of Telecom
(DoT). The two licences are among the 85 that
have been declared illegal by the Comptroller and
Auditor General (CAG). Of the INR100,000m
loan to Unitech, the major portion was disbursed
by the State Bank of India (INR80,500m).
Other banks that have lent to Unitech are:
Corporation Bank (INR5000m), Allahabad Bank
(INR5000m), South Indian Bank (INR4,000m),
Canara Bank (INR1,200m), Oriental Bank
(INR700m), Central Bank of India (INR700m),
Punjab National Bank (INR1200m), Standard
Chartered Bank (INR1000m) and Yes Bank
(INR700m). STel got the INR15,380m loans from
IDBI and IDBI Trusteeship Services Ltd, during
between July and November 2009.
In early January 2011, the finance ministry
cleared all banks that lent money to telcos for 2G
licences in 2008. While the Supreme Court has
asked the CBI to broaden the scope of its
investigation into the 2G scam by looking into
bank lending to the sector, the finance ministry
and the central bank have both said there were no
systemic risk issues related to telco lending.
Separately the SBI has tightened rules for loans to
telecom firms by making it mandatory for telecom
companies to submit a utilization plan, a
repayment schedule and details of their service
rollout. The SBI suggested that it will ensure that
the money is used only for the stated purpose.
We note that project-based funding could mean a
rise in the cost of bank loans for telcos and make
it very difficult for the new entrants to raise
money. The situation is such that the telecom
minister has pleaded with the finance minister and
the prime minister to ask banks to lend to the
telecom sector. With the launch of 3G services,
this issue will become even more critical.
In a separate development, new entrant Uninor, a
subsidiary of Norwegian telecom Telenor Group
and Unitech Group, is considering raising
additional capital. There is disagreement between
Telenor and Unitech over the nature of the
financing. Tata-DoCoMo is also considering
raising USD700m and it may try and tap the
equity markets. In our view, the fact that both
Uninor and Tata Tele are looking to raise capital
through a rights issue suggests that external
financing may not be available.
That said, there have been some exceptions. For
example, RCOM recently borrowed USD1.9bn
from China Development Bank (the loan comes
with conditions to buy Chinese equipment). The
funding doesn’t resolve all of RCOM’s problems
and its stretched balance sheet remains a concern
for the company.
Cancellation of licences and payment
for spectrum
Under the terms of the contract, the licensee is
required to cover at least 90% of the area in the
metro service areas of Delhi, Mumbai, Kolkata and
Chennai within a year of the allocation of spectrum.
For non metro areas, it is required to cover 10% in
the first year and 50% within three years.
TRAI has suggested to the DoT that 69 licences
be cancelled in cases where rollout obligations
have not been met. The DoT said it will take a call
shortly on these recommendations. Separately, the
apex court will also decide on the same issue.
In a recent development, the Enforcement
Directorate has filed cases against five companies,
including Swan Telecom, Loop Telecom and
Loop Mobile and STel for alleged foreign
exchange violations. The companies face fines for
allegedly routing cUSD960m into the country
from overseas in contravention of regulations.
We believe these uncertainties will further delay
any expansion plans of new entrants. Moreover, if
the licences are not cancelled, additional fines
cannot be ruled out, further damaging the
prospects of new entrants. The DOT has already
collected cUSD70m in fines from new entrants.
The report from the Comptroller Audit General
(CAG) estimates that the scandal cost USD39bn
in terms of lost government revenue (the CBI
recently put the figure at a more modest
USD6.5bn). The CAG also suggested that many
of the new entrants obtained licences despite
improper documentation.
While the Indian regulator’s decision may not be
final or binding, it raises the possibility of new
GSM entrants having to pay one-time charges for
start-up spectrum and also losing their licences.
We calculate that new operators may be asked to
collectively pay USD6bn for 2G spectrum and
USD9bn for 3G
MNP threat has subsided
Concerns that mobile number portability (MNP)
would disrupt pricing have proved to be
unfounded. Aimed at high end subscribers of
GSM incumbents, MNP has actually turned out to
be positive for three operators.
At the same time MNP has exposed the
vulnerability of CDMA operators. While there is
still a possibility that MNP will still cause minor
disruption, we think the overall impact will be
very limited.
Balance sheets are tight
Post the 3G auctions, the top six players are
operating with tight balance sheets (Figure 15),
limiting price flexibility. This is reflected in stable
revenue per minute and rational 3G prices. Our
analysis suggests that present 3G prices are in line
with global pricing norms (Figure 16).
Murky outlook limits any further
overcapacity
We believe the new entrants are now not only
operating with unviable business models but are
also facing questions about their ability to survive.
This will prevent any serious investments by the
bottom eight players, reducing the risk of further
overcapacity. The industry fragmentation and
overcapacity has not gone down well with
regional and global investors. The possibility of
improvement in revenue per minute will be
viewed as positive for the sector, increasing the
chances of a possible re-rating, in our view.
Pricing theory
An operator making cuts to tariffs needs to be
confident about making a decent profit from
selling cheaper minutes. We do not believe that
pricing is set on the basis of cost plus, but rather
by what might be considered “game theory”.
Challengers will be reluctant to cut tariffs unless it
allows them to steal a march on the established
operators. Otherwise they work on the assumption
that any cuts they make will be matched by the
larger competitors, making it a zero sum game.
If we see things in today’s Indian context, there is
nothing the challengers can do that incumbents
would find it difficult to match. Our analysis
suggests that even the likes of Tata-DoCoMo need
to price outgoing calls at least c10% higher to
achieve EBITDA-breakeven on GSM.
By further cutting tariffs challengers will be only
hurting themselves. For the likes of RCOM and
Tata, taking market share is no longer the main
priority as it has made little impact on their
revenues as tariff cuts were matched by
incumbents. This is supported by the fact that
50% of Tata subscribers and c34% of RCOM subs
are not active (Figure 20).
In our view both RCOM and Tata have a big task on
their hands in terms of achieving EBITDA
breakeven on GSM, investments in 3G and rescuing
CDMA. Our view is supported by the fact that since
per second billing was introduced none of the
challengers have resorted to price cuts.
RCOM and Tata face
structural issues
The biggest challenge for Tata and Reliance is
their low quality subscriber base, which is more of
a structural issue than a question of tactics. The
poor subscriber quality is driven by their late
entry into GSM spectrum in the 1800 MHz and
poor investment in the network.
In our view, price cuts attract teenagers rather
than quality subscribers – hence the high
percentage of inactive subscribers. Both Tata and
Reliance have also missed out on MNP and
instead face continued churn on the CDMA
network which has quality issues.
Idea Cellular did the right thing
The most interesting case study in the Indian
GSM space is Idea Cellular. It started late but,
with the right kind of focus and execution, it has
emerged as a strong pan-India GSM player.
Back in Q3 2008, Idea Cellular was a regional
operator with a presence in 11 areas with revenues
way below that of RCOM, Bharti, BSNL and
Tata. However, Idea’s revenue market share has
jumped by 360bps since then thanks to its
investments in its network (Figure 18).
Figure 20: Active subscribers
Operator Active subs %
Bharti 93%
Idea 90%
Vodafone 78%
RCOM 66%
Aircel 61%
BSNL 54%
MTS 51%
HFCL 50%
Videocon 50%
Tata Tele 50%
Uninor 47%
Loop 44%
STel 41%
MTNL 37%
Etisalat 34%
Source: TRAI
CDMA getting weaker
MNP exposed the weakness of CDMA networks.
This particularly applies to Tata and RCOM, two
operators that still have little traction in GSM. Given
the various challenges they face, we believe Tata and
RCOM have little choice other than to try to increase
the rate per minute.
We are not worried about Aircel as they have not
used tariffs to gain market share. They are the only
new entrant to use a VAS strategy to which the
incumbents could not respond in the 2G
environment, given the spectrum crunch. However,
the launch of 3G services will allow the incumbents
to respond to the VAS offered by Aircel.
Dual SIM phenomenon dying down
The basis for the dual SIM phenomenon was
arbitrage between the tariffs provided by the new
entrants and incumbents. As incumbents have
responded to the tariffs of new players, the arbitrage
seems to have narrowed. That said, the likes of
Uninor are still focused on the dual SIM strategy
through their location based discount schemes.
However, the fact that it has gained little traction
suggests that dual SIM is no longer going to be a
viable strategy.
Voice revenue to improve
We expect voice revenue to improve in the medium term; we
expect the first signs of this to be visible from 3Q FY12
We believe this trend will be driven by investments in 3G, the low
quality of new subscribers and limited upside to MOUs
Revenues can also be improved by reducing freebies such as
special vouchers and free airtime
Tariffs should improve – but
how and when?
We believe the market has a very pessimistic view of
the current tariffs. Those who are bullish are betting
on 3G, which we don’t expect to be earnings
accretive any time in the near term. The stability in
revenue per minute is being viewed as a good
downside support but the possibility that there are
chances of tariffs moving up is being completely
ruled out. We believe that pricing power can return
to the sector and revenue per minute should improve
over the next 3-4 quarters.
Now this need not take the route of big
movements in the headline rates. Given the mood
of the Indian regulator, we don’t believe that
operators will do anything drastic. In our view
there are other ways to boost revenue per minute
and tariff upside will be driven by:
Investments in 3G.
The fact that subscriber additions are not very
value accretive.
3G in its early stages needs to be supported
by growth in voice business.
MOU growth is on a declining trend.
While the sector has benefited from stellar
subscriber growth, the impact on revenue from
new subscribers has been limited. New
subscribers are not only poor in quality, but also
relatively expensive to acquire. In fact it is the
retailers that have benefited from the competition
for subscribers. The Indian telecom space has
added about 362m subscribers since June 2009
(when Tata Docomo launched the per second
billing plan) but the corresponding revenue has
been only cUSD890m, an increase of 13% versus
subscriber growth of c46% (annualised revenues
for the sector are estimated at USD27bn).
Note that a large part of this 13% revenue growth
came from higher usage by older subscribers. The
strategy will now shift from subscribers to
extracting more revenue from the existing set of
subscribers via voice and usage. We are of the
view that a large part of the usage or minutes in
the system today are inelastic and this allows
telcos to take away freebies and tweak various
offers to improve the revenue per minute.
3G will require steady investment
We expect most incremental network capex to focus
on adding to operators’ existing 3G systems. 3G will
require significant investment going forward, both in
backhaul and the core network. The focus of capital
investment, however, remains a contentious point.
With 3G, we believe all aspects of the network
should see higher spending.
While the Indian voice network coverage is stable
(at least with the incumbents, leaving aside the
spectrum crunch issues), subscribers will have
similar expectations about the 3G network. This
will put pressure on the 3G winners to invest in
networks or they may fail to capture the data
growth. We believe the need to invest in 3G
networks will not only drive rationality but also
raise the need to work on improving realised rates.
Voice revenues have to support 3G
We believe 3G business is EPS dilutive for at least
the next 3-4 years for the overall sector and for every
player. Our analysis suggests the industry requires
c90m subscribers to achieve EBITDA-breakeven on
3G, with monthly ARPU assumption of cINR175-
200. Given this, there are very limited options other
than to improve voice revenues.
MOU growth on a declining trend...
becoming inelastic in nature
Our analysis suggests that the MOU per
subscriber are on a declining trend (Figure 22).
We are modelling for an annual c2% decline in
sector MOUs from FY11 to FY15e.
If we were to treat the subscribers with Bharti at
end of March 2009 as old subscribers and all
incremental subscribers as marginal, our study
suggests that MOUs of the existing subscribers
have grown by 46% in the past seven quarters and
grown at a quarterly CAGR of 6%. This suggests
that the growth in minutes of usage is being driven
by the old subscribers, and new subscribers are
MOU dilutive. For the purpose of our analysis we
are assuming that MOU of marginal subscribers
have declined from 260 to 190 per sub.
Analysis with similar assumptions for Idea Cellular
suggests that MOU per sub of old subscribers have
increased 80% over the last seven quarters and
grown at quarterly CAGR of 9%.
Analysis for both telcos suggests that MOUs from
old subscribers have been phenomenal in the past
seven quarters, but we believe this trend cannot
continue for much longer. While the usage growth
will continue, the rate of growth should decline in
our view. As such we believe that we have
reached a scenario where the bulk of usage is
inelastic in nature and there is not much left for
the telecom operators to achieve by cutting rates.
Instead, some tweaking of rates will help the
telcos to benefit at the net level. Bears may argue
that telcos may opt to continue with present tariffs
or provide discounts in an attempt to boost usage
from the marginal subscriber segment. We believe
that limited purchasing power is the biggest
challenge in this regard and telcos will get little
benefits from such a strategy.
How will rpm improve?
We don’t expect the improvements in revenue per
minute to be immediate. The MNP overhang may
linger for a couple of quarters but in our view
some investors may miss out on the first stages of
improving rpm, which we believe could happen in
the following possible ways:
Withdrawal of free airtime
Free talk time pulls down the realized rates as
paid minutes are replaced by non-paying minutes.
Withdrawal of free airtime will result in an
improvement in effective realized rates and
should be positive.
Withdrawal of special vouchers
Withdrawal of discounts (Figure 23) could result
in improvements in revenue per minute. We
believe night calling plans or other offers which
lead to network utilization in non-peak hours will
continue. But there is huge amount of discounts
on the on net calls and SMS tariffs and we believe
some of these plans can be tweaked.
Improvement in headline rates…this will be the
last to come
Last, not but least, we expect an announcement on
headline rates at some point. This is unlikely to
happen in the near term as it will be difficult to
get past the regulator in the current climate. We
don’t expect any announcement until the
operators have complete clarity on the policy
framework, particularly on M&A policy.
Telcos are already charging for call
centre calls
There are already a few promising signs. Telcos
have been charging subscribers for calls made to
call centres for 9-10 months. Separately, the likes
of Bharti have been charging SMS interconnect
fees. Both these examples highlight the various
possibilities which exist for operators to improve
revenue per minute.
Potential surprises
A complete withdrawal or tweaking of per second
billing plans. For instance, operators could bill for
every two seconds. For this to happen, some kind
of industry consensus will be required. A
withdrawal of per second billing could improve
effective rates by 10-15% and it would be easier
to get this past the regulator than a rise in headline
rates. We note India has a policy of forbearance
and tariffs are driven by free market mechanisms
and the regulator at present has no say on tariffs.
VAS will grow
Another aspect where we are looking for growth
is VAS. While 3G services will cater to high end,
the 3G ecosystem is likely to throw up various
niche products which were completely absent in
the 2G regime. The 3G network comes with two
advantages – the ability to do video calling and
faster download speeds. Apart from these two
services, there are not many services/products
which cannot be provided on a 2G platform.
The present VAS contribution to overall revenues
is c14%; 80% of the present revenue comes from
caller tunes and 20% from data services. With the
3G product portfolio aligned on the data side,
there is a solid case for 2G VAS revenues to
benefit from the development of the 3G
ecosystem. We believe that going forward the
contribution from data-centric products should
drive growth in the 2G VAS space and we see
contribution from data-specific products moving
from the present c20% to 50% by FY15e.
The regulatory climate
Regulations cannot work in isolation; the policy framework needs to
create sustainable revenue streams aligned with the competitive
landscape
We don’t expect the Indian telecom regulator to control tariffs and
believe it will continue with a policy of forbearance
There is a need for reforms to sustain tax receipts and generate
benefits from steady spectrum auctions
The sector’s financial health
There is a huge amount of scepticism about various
regulatory issues facing the Indian telecom sector.
But before we get into those discussions let’s
understand the relative importance of the telecom
sector for the Indian government. Our analysis
suggests that the government – excluding spectrum
auctions – collected a total USD5bn in taxes from
the sector and cUSD24bn from spectrum fees last
year. While it is very apparent that the Indian
government is very much focused on raising taxes
from the sector, we believe government may lose out
in the longer term by not paying attention to the
sector’s health. The government is in the process of
drafting the National Telecom Policy 2011, and it is
important this issue is taken into consideration.
The sector is in a poor state overall and, except for
Bharti, all the telecom operators are working below
their cost of capital, in our view. ROEs have come
down for Bharti as well, suggesting that incremental
prepaid business is challenging and diluting the
overall ROE.
While the raising of USD15bn from 3G spectrum
auction was good news for the Indian government,
we believe the ability of government to benefit from
such high premium on 3G spectrum was driven by
presence of 14 players. However, this cannot be
sustainable and in our recent regulatory
developments will reduce the number of players to
6-7 in the near to medium term.
The 3G auctions have raised two questions:
Why was no single player bidding for 3G
spectrum pan India, despite a pan India
2G presence?
Why was no new foreign player investing
in 3G?
Add to this USD10bn investments made by Bharti
in Africa, there seems to be some suggestion that
Indian telcos are a bit hesitant to place bets in the
Indian telecom space given the present regulatory
environment. We believe a more conducive
environment and a policy framework which
considers the financial health of the sector will be
positive and also maximise the tax collection for
the government
Time to create sustainable revenue
stream
The best piece of policy so far by the Indian
government has been the consideration of a revenue
sharing regime for the telecom sector. This
encouraged participants to invest in telecom
infrastructure. However, the industry fragmentation
in 2007 caused by the issue of eight new licences
resulted in a loss of USD998mn to the government
as the revenue per minute declined by c40%. This is
over and above the loss from the fact that 2G
spectrum was not put out to auction (CAG estimates
the loss at USD39bn).
The Indian government has set some tough fiscal
deficit targets and it goes without saying that it needs
to continuously invest in the economy to drive
growth in excess of 7-8%. This creates the need for
the government to create sustainable revenue
streams from an emerging sector like telecoms.
Globally, continuous auction of spectrum has been
the source of incremental revenues from the telecom
sector. Post realizing USD23bn from the 3G and
BWA spectrum auctions, government has budgeted
cUSD4bn as non-tax revenues from the telecom
sector this fiscal year. This is good news and we
believe government plans to raise this money by way
of spectrum sales, possibly auctions of 2G spectrum.
We believe that in the Indian context spectrum
auctions can prove useful to the government. Taking
the US as an example, the regulator, the FCC, has a
very clear roadmap in terms of what it needs to refarm
and how. For instance it plans to re-farm about
300MHz of spectrum in the next five years and the
plan also lays out the various bands which will meet
the requirement.
The fact that in India, 40% of the total available 2G
spectrum and 60% of the total 3G spectrum is not
with the government but with the Indian defence
forces or other government sectors, highlights that
Indian telecom regulator needs to have a longer term
perspective and a robust re-farming policy in place.
The process of spectrum re-farming is very complex,
long and expensive.
Data will drive demand for spectrum
Incremental demand for spectrum will be driven
by both voice and data. But what will create more
demand is data and not voice from a medium term
perspective. The situation in India could be even
worse than global conditions as the first release of
3G spectrum was limited and none of the operators
have pan India 3G spectrum. Most of them will be
keen to ramp up with a pan India presence, in our
view. The spectrum is 5MHz for all operators
However, it will take time before the telcos make
money from 3G and their ability to participate in
auctions for 3G spectrum will have to be supported
by the cash flows from the voice business. This
again raises the significance of improvement in
revenue per minute and how indirectly this will aid
the government as well. Any policy change by the
government with respect to tariffs may prove to be
counterproductive and as such we believe the Indian
telecom regulator/government will be better placed
by sticking to its current approach of forbearance.
Regulator has to depend on
incumbents, why?
Bears will argue that the government may decide not
to depend on incumbent operators and may
encourage fresh investments in the market. We don’t
think so; there was hardly any participation in 3G by
any additional foreign player. And the failure of
Hutch3’s standalone 3G business is a great exampleTime to create sustainable revenue
stream
The best piece of policy so far by the Indian
government has been the consideration of a revenue
sharing regime for the telecom sector. This
encouraged participants to invest in telecom
infrastructure. However, the industry fragmentation
in 2007 caused by the issue of eight new licences
resulted in a loss of USD998mn to the government
as the revenue per minute declined by c40%. This is
over and above the loss from the fact that 2G
spectrum was not put out to auction (CAG estimates
the loss at USD39bn).
The Indian government has set some tough fiscal
deficit targets and it goes without saying that it needs
to continuously invest in the economy to drive
growth in excess of 7-8%. This creates the need for
the government to create sustainable revenue
streams from an emerging sector like telecoms.
Globally, continuous auction of spectrum has been
the source of incremental revenues from the telecom
sector. Post realizing USD23bn from the 3G and
BWA spectrum auctions, government has budgeted
cUSD4bn as non-tax revenues from the telecom
sector this fiscal year. This is good news and we
believe government plans to raise this money by way
of spectrum sales, possibly auctions of 2G spectrum.
We believe that in the Indian context spectrum
auctions can prove useful to the government. Taking
the US as an example, the regulator, the FCC, has a
very clear roadmap in terms of what it needs to refarm
and how. For instance it plans to re-farm about
300MHz of spectrum in the next five years and the
plan also lays out the various bands which will meet
the requirement.
The fact that in India, 40% of the total available 2G
spectrum and 60% of the total 3G spectrum is not
with the government but with the Indian defence
forces or other government sectors, highlights that
Indian telecom regulator needs to have a longer term
perspective and a robust re-farming policy in place.
The process of spectrum re-farming is very complex,
long and expensive.
Data will drive demand for spectrum
Incremental demand for spectrum will be driven
by both voice and data. But what will create more
demand is data and not voice from a medium term
perspective. The situation in India could be even
worse than global conditions as the first release of
3G spectrum was limited and none of the operators
have pan India 3G spectrum. Most of them will be
keen to ramp up with a pan India presence, in our
view. The spectrum is 5MHz for all operators
However, it will take time before the telcos make
money from 3G and their ability to participate in
auctions for 3G spectrum will have to be supported
by the cash flows from the voice business. This
again raises the significance of improvement in
revenue per minute and how indirectly this will aid
the government as well. Any policy change by the
government with respect to tariffs may prove to be
counterproductive and as such we believe the Indian
telecom regulator/government will be better placed
by sticking to its current approach of forbearance.
Regulator has to depend on
incumbents, why?
Bears will argue that the government may decide not
to depend on incumbent operators and may
encourage fresh investments in the market. We don’t
think so; there was hardly any participation in 3G by
any additional foreign player. And the failure of
Hutch3’s standalone 3G business is a great example
of how telcos have realized that there is no
standalone 3G business case. We believe that to
create sustainable revenue streams from spectrum
auctions, the regulator will need to rely on the top
five/six operators.
Post 3G, all the top five players have taken on a lot
of debt. So the government’s ability to monetize
spectrum auctions may also be limited.
Risks to our calls / what we
may be missing
More spectrum issued to new
entrants Tata and RCOM
We believe it will be difficult for the government to
issue spectrum to any player without an auction. But
the new entrants argue that licence conditions allow
the minimum spectrum up to 6.2 MHz and, as such,
there are possibilities of more spectrum going to new
entrants, or at least RCOM and Tata.
Bears may argue that additional capacity creates the
risk of another tariff war. We don’t think there is risk
to tariffs if more spectrum is allocated to the likes of
RCOM or Tata because their GSM networks are not
suffering from undercapacity, so they are unlikely to
cut tariffs. Separately, we believe more spectrum
will be the last chance for both RCOM and Tata to
position for high-end subscribers, rather than playing
another tariff-disruptive move and giving away
capex savings. We believe they are likely to be
rational and focus on structural issues.
MVNO approach
There have been some suggestions that the
government may decide to boost MVNOs to save
the business models of new entrants. We don’t think
that the MVNO approach works in low cost, low
tariff markets like India. With call rates at INR0.5,
termination at INR0.2, the spread is very thin for
operators with low scale to operate. Virgin mobile is
a clear example, having failed to create any impact
so far in any format. As we suggested, for new
entrants outgoing tariffs should be at least c15%
premium to the present tariff levels to be EBITDA
breakeven, it highlights that if new GSM entrants
decide to share a percentage from the tariffs earned
they will only delay their profitability.
Bears may further argue that new GSM entrants may
get into contracts with MVNOs to earn one time
fees, as their network costs are fixed and this allows
them some breathing space.
In this context we would like to highlight quality
issues. The biggest problem for RCOM and Tata in
GSM has been late entry and their GSM spectrum in
1800 MHz. It will be either the new entrants (which
have hardly any network to offer) or the likes of
RCOM and Tata which may adopt the MVNO route.
With spectrum in GSM at a mere 4.4MHz, it is not
enough to share/lease capacity. At the end of the day
MVNOs will be dependent on the 1800MHz of the
new entrants which is poor quality.
VOIP
VoIP has been a concern for investors in mobile for
several years. However, the inroads that it has made
into voice revenues have been relatively slow,
largely thanks to technical issues. The highest profile
VoIP service remains Skype, which has an
enormous following on desktop and laptop
platforms. Operators with very little market share
like 3UK have been pushing Skype. But note that
calls made via Skype over 3UK’s network are –
confusingly– not actually VoIP calls. The call is
circuit switched over 3UK’s radio network, and only
once it reaches the core does it become IP. This
should serve to underline the difficulties inherent in
getting VoIP to work adequately in a mobile
environment – certainly to the degree of quality of
service to which users will be accustomed.
However, VOIP is coming into its own with LTE.
Because this platform is all-IP, it has no capability to
carry switched voice calls; hence all voice
communication must be carried by VoIP. Therefore
the timing of the mass market LTE is potentially
most significant for VOIP to make an impact on the
voice tariffs.
Conversely, the longer it takes for LTE-powered
VoIP to arrive, the better the industry’s prospects for
stable pricing. However, the threat is smaller in
emerging markets like India, where the tariffs are
already the lowest and the GSM ecosystem is strong
with robust handset choices.
Further it will take a lot of work to get a VoIP LTE
solution as reliable as 2G and 3G platforms. LTE is
an all-IP platform, which does not have channels
dedicated to carrying voice traffic. Instead of the
circuit switched approach, it uses an end-to-end IP
connection from the user’s device back into the core
network. In the early LTE standards releases, it is
clearly data rather than voice that is the priority.
VoIP did not even make it into LTE as standardised
in 3GPP UMTS Release 8, albeit in part because the
relevant technology is simply not ready. There are
various alternative solutions for LTE VoIP, though
only two are currently approved within the 3GPP
standard. These are voice over IMS (VoIMS) and
circuit switched fall-back (CSFB).
The former was the preferred choice. However, it
relies on IMS-based core networks, and has run into
real difficulties given the failure of IMS to gain
momentum. The 3GPP designated alternative for
voice over LTE is CSFB, which is – quite simply, in
our view – a compromise, routing the call instead
over the switched network via 2G or 3G.
Furthermore, far from being a pragmatic short-term
alternative, it would appear to have some grave
practical defects. For instance, it is likely to involve a
very significant increase in call set-up times – as the
handset must start up its 2G or 3G radio and connect
to the appropriate cell site at the appropriate
frequency. Plus there is a good chance that any LTE
data connection running in the meantime will be
dropped (just as 2G to 3G transitions – or vice versa
– often result in problems). There could also be
some very deleterious implications in terms of
battery life. Given this, we believe it will still be
some time before VOIP becomes a meaningful
threat to voice tariffs.
Our global team argued in Frequonomics that the
rise of VoIP over LTE would likely take years, and
we have heard little in the interim to change this
view. The challenge here is for the handset industry
to get multi-mode devices capable of supporting 2G,
3G and 4G technologies, with adequate battery life
and with the necessary on-board multi-tasking
processing capability to handle VoIP as one of a
number of concurrent user applications.
3G used for discounting voice.
3G spectrum is six times more efficient than 2G and
as such per unit of capacity created with 3G
spectrum is much cheaper. This may be attractive
enough for the likes of RCOM and Tata to use 3G
for voice and go ahead with cuts in tariffs, given that
the cost of producing minutes in 3G is very low.
While the proposition may seem logical, one needs
to factor in that telcos need to recover the price paid
for 3G spectrum as well. If 3G is used for voice, the
payback period for the operators may take decades,
more so if that too is discounted from the present
levels. Moreover, the subscriber volume on 3G
platform will be restricted to those with 3G enabled
handsets. And these will be the high end subscribers,
less sensitive too lower tariffs on voice, suggesting
that this strategy could fail.
TRAI recommendations
We don’t expect the TRAI recommendations to be accepted in the
current format
We believe they are a major shift from the present policy framework
Re-farming of spectrum in 900 MHz is the most significant
recommendation; we believe it will be not easy for the regulator to
implement this and that it faces several challenges
Unlikely to be accepted in the
current format
While the probe into the 2G scam has been negative
for new entrants, the TRAI recommendations have
created problems for the top GSM players, namely
BSNL, Bharti, Idea and Vodafone.
We believe they represent a major shift in thinking.
To us, it appears that TRAI wants to change the rules
midway through the process, after billions of rupees
have already been spent. The recommendations
could lead to another round of legal battles which
would bring uncertainty and adversely impact future
investment in the sector. We don’t believe the Indian
government can afford such an outcome.
One-time payment for excess
spectrum beyond 6.2 MHz
We believe it will not be easy for TRAI to get a
consensus on this issue. The spectrum in excess of
6.2 MHz has been obtained by the incumbents by
meeting subscriber-linked criteria. Given this, we
believe this recommendation could be challenged
by GSM incumbents. Despite this, if the
recommendations are implemented, GSM
incumbents may decide to give up the excess
spectrum in select markets to avoid
financial damage.
Separately, GSM operators don’t believe in the
concept that there should be a maximum amount
of spectrum a single operator can hold. We are not
aware of any other market which sets an upper
limit to the spectrum an operator can hold.
Spectrum pricing
Another related concern is spectrum pricing,
which has implications for both the licence
renewal issue and the payout for spectrum in
excess of 6.2MHz. TRAI recommended 3G
spectrum prices as the benchmark for arriving at
the payout for the operators in both instances.
In a recent development, TRAI revisited its
estimates on spectrum pricing and the latest
recommendations have lowered the payout for
metros and increased the payout for operators in
non-metros. In some of these non-metros the
spectrum payout exceeds the payment for 3G
spectrum. The underlying rationale is that the
non-metros have less of a data business case as
their markets are predominantly about voice;
hence, 2G spectrum generates higher payouts in
such markets.
We believe pure voice markets have upside to
margin, whereas data in the medium term can
allow operators to improve margins. Moreover,
the spectrum pricing recommendations have been
strongly criticised by the GSM operators and we
think it is unlikely that they will be accepted in
the current format.
Licence renewals
The TRAI has suggested very stringent conditions
for the renewal of licences, including:
At the time of renewal, spectrum held by the
operators in the 900MHz band to be replaced
by an equal amount of spectrum in the 1800
MHz band.
The renewal of existing licences for another
10 years instead of 20 years and one-time
payment for the entire spectrum which is
renewed along with the license
We believe spectrum re-farming should be
given a longer time frame; it should be based
on price incentives. If someone is ready to get
spectrum re-farmed then that operator should
be given certain price incentives. But
reframing should be done with a certain
degree of facilitation caveats rather than
penalties, which is the case with the TRAI
recommendations in our view
Furthermore, TRAI recommendations in the context
of spectrum pricing suggest that 900 MHz spectrum
is more efficient than 1800 MHz and as such the
spectrum in the 900 MHz band should be charged at
a premium to the spectrum in the 1800 MHz band .
Given the same rationale, the licence renewal
conditions should be factoring in the same approach
and as such those operators who give up spectrum in
the 900 MHz band must be compensated with 50%
more spectrum in the 1800 MHz band. However,
the TRAI recommendations fail to address
such inconsistencies.
We believe the limitations of the dynamics of timing
cannot be an excuse; operators which came in first
got spectrum in the 900 MHz band, and those which
came in later got spectrum in the 1800 MHz band.
When operators were allocated spectrum they knew
what they were getting and taking away the
spectrum now from early movers into the sector may
not send positive signals to other industries and may
impact foreign investment as well.
Overall, we believe that refarming spectrum will
be complex, time-consuming and expensive. It
could also have a significant adverse impact on
the sector as it would lead to additional capex.
Our view is that the complexities involved in the
process of re-farming will make it a next to
impossible process for the government. There will
be a serious amount of subscriber discomfort and
possibilities of serious disruption to essential
telecom services cannot be ruled out.
We believe the Indian telecom regulator needs to
take these factors into consideration before finalising
any policy changes.
The biggest losers of the re-farming process will be
BSNL and MTNL as these operators have 900 MHz
spectrum across India. They would face higher
capex and their business viability may suffer
significantly. With the financial performance of the
PSUs already a worry, we believe the government
will find it difficult to introduce the refarming policy
suggested by TRAI.
BSNL is having a very difficult time. In FY10, the
company posted its first ever loss of INR182bn. In
an effort to balance its social commitments of
providing services in rural areas and competing
with private players in urban areas, its
profitability has taken a severe hit (Figure 26).
Also, the company had to pay INR1,850bn for 3G
and broadband wireless spectrum in FY2011,
reducing its cash balance to cINR1,150bn
excluding losses in FY11. If BSNL applies for
900 MHz licence in all its 23 circles then there is
a very high probability that licence fees coupled
with ongoing operational losses will result in a
cash deficit. Hence, we believe that the
government will structure 900 MHz licence fees
to take into consideration BSNL’s cash position.
Leaving out PSU from re-farming and only
making it valid for private players will not be easy
and will only lead to litigation, in our view.
Last, but not least, to implement re-farming the
Indian government would require c434Mhz of 2G
spectrum in the 1800 MHz, or about 20MHz across
each circle. If new players prevail and there are no
licence cancellations, total spectrum required would
be 736MHz, or c33MHz across each circle.
Assuming nothing comes from defence, the shortfall
will be c522 MHz, and if c20 MHz comes from
defence in each circle, the shortfall would still be
c90MHz. It is well known how cumbersome it has
been for the telecom ministry to obtain a mere 15
MHz in 3G from the defence department.
To sum up, regulators across other markets have
simplified rules and if the TRAI recommendations
are accepted than it would be one of the few nations
with such stringent telecom regulations. Examples
that support our view are mentioned below:
In the United States, licences are issued for a
10-year period but are renewed routinely at a
nominal cost
In the UK and the Nordic region, annual
licence fees are not applicable for the licences
awarded by auction
No comments:
Post a Comment