Sector note
Egyptian Telecom Sector Telecoms | Egypt
MENA research
A yield game
Telecom Egypt
Target price (EGP) 21.1 Current price (EGP) 17.6 Potential return 20.0% Bloomberg ETEL EY Reuters ETEL.CA MCap (USDbn) 5.2 Daily volume (USDm) 3.9
Real penetration of 95% implies limited subscriber growth going forward, mainly driven by an increase in dual/triple SIMs and a young population
A relatively high telecom spending to GDP implies low revenue growth for the sector (2011–14e CAGR of 2.8%), making it no longer a growth story We remain Overweight on TE (TP EGP21.1/share, 2010e FCF and dividend yield of 16.6% and 8.1%) and Neutral on Mobinil (TP EGP179.6/share)
The penetration rate in Egypt is higher than it seems as real penetration is expected to reach 95% by the end of 2010e. We forecast an addressable population of 55.7m (adjusted for age and income distribution) versus a total population of 81.3m. Based on our calculations, we estimate dual SIMs stand at 39% of the total subscriber base, which implies 53.1m human subscribers versus reported subscribers of 74.9m. Accordingly, we expect limited subscriber growth, with a 2011–14e CAGR of 5.9% mainly driven by dual/triple SIMs (reaching 53% by 2012e) and new additions (1.4m per year) from the youth segment (21% of the population). We expect ARPU dilution to continue on the back of price cuts into 2011e as operators fight for subscribers. Blended ARPU is expected to dilute at a 4.5% 2011–14e CAGR, offsetting market subscriber growth. Accordingly, we expect Telecom Egypt (TE), Mobinil, and Vodafone Egypt’s (VFE) revenue to grow at a 2011–14e CAGR below 2.7%. Our revenue growth estimates are supported by Egypt’s relatively high telecom spending to GDP (3.3%) compared to other emerging and developed markets. We believe the Egyptian telecom story suits value investors, as such TE remains our top pick. We reiterate our Overweight rating on TE (TP EGP 21.1/share) a defensive stock with a high yield. We expect dividend yields of 8.1% for 2010e and 8.7% for 2011e, supported by strong cash flow generation (FCF yield of 16.6% and 11.2%) and a solid balance sheet (TE’s cash balance is expected to reach 17% of total assets in 2013e). Our cash flow assumptions take into account an MVNO license fee (EGP550m) in 2011e, related annual CAPEX of EGP150m onwards and further broadband expansion. TE is cheap on multiples, trading at a 10% discount to peers on 2011e PE and 36% on adjusted EV/EBITDA. We maintain our Neutral rating on Mobinil (TP EGP179.6/share) as we expect FCF yield and net profit growth to come under pressure given expected high CAPEX and margin compression due to competition. We expect a dividend yield of 4.3% for 2010e and 5.8% for 2011e (payout ratio of 50% and 70%). Upside risks to our valuation are lower ARPU dilution and Orascom Telecom exercising its put option, thereby forcing France Telecom to launch a mandatory tender offer to minorities at EGP221.7/share.
Overweight
Mobinil
Neutral
Target price (EGP) 179.6 Current price (EGP) 166.7 Potential return 7.7% Bloomberg EMOB EY Reuters EMOB.CA MCap (USDbn) 2.9 Daily volume (USDm) 2.6 All prices as of 12 December 2010
14 December 2010
Karim Khadr Analyst +971 4 293 5381 karim.khadr@hc‐si.com
Sarah Shabayek Analyst +971 4 293 5389 sarah.shabayek@hc‐si.com Disclaimer: See page 40
Sector note Telecoms | Egypt 14 December 2010
Muted growth outlook
Limited subscriber growth as real penetration is expected to reach 95% in 2010e, dual/triple SIMs to drive 2011–12e growth, the young to remain long‐term drivers Continuing ARPU dilution (2011–14e CAGR of ‐5%) as operators are subscriber acquisition oriented
Relatively high telecom spending to GDP indicates growth at par with real GDP, albeit at a lower rate in the short term due to competition, 2011–14e CAGR of 2.8%
Real penetration of 95% in 2010e Egypt’s penetration rate is higher than it seems since the addressable market is considerably smaller than Egypt’s population (55.7m versus 81.3m). We calculated the penetration rate by taking into account age and income distribution and subtracting those below 9 years of age, those above 75, and half of those under the poverty line (to avoid double counting). Dual SIM users are expected to represent 39% of total subscribers (supported by Mobinil and TE’s management guidance of c35%) leading to human subscribers of 53.1m versus reported market subscribers of 74.9m. Dual/triple SIMs to drive growth in 2011e and 2012e We believe price competition in the mobile sector ensures some short‐term growth as it stimulates dual and even triple SIMs. Mobinil is back in the game fighting for its market share, and VFE changed its strategy from focusing on value‐added propositions to subscriber acquisitions. The fight on market leadership and Etisalat’s efforts to further increase its market share ensure that competition remains in the market and that the percentage of dual SIMs could increase to 50% in 2011e and 53% in 2012e from 39% in 2010e. We believe the pricing environment will stabilize by 2012e and therefore going forward headline prices should remain unchanged. The on‐net/off‐net price differential should be maintained and therefore subscribers would have no incentive to purchase another SIM or drop one they own. New additions of 1.4m/annum from the youth segment to continue to drive growth in the long term Some growth should still take place driven by those below 9 years of age. The Egyptian population’s growth rate is c1.9% (most CEMEA countries are experiencing negative growth), and a significant part of the population is under the age of 9 (21% of the total population). New subscribers are expected to be the main source of market growth (c1.4m new subscribers per year). We increased our estimates for market subscribers in 2010e and 2011e to reflect continuing intense competition but lowered the rate by which subscribers grow in the long run to reflect market maturity. 2
Sector note Telecoms | Egypt 14 December 2010
Adjustments made to market subscribers and population to arrive at real human penetration (‘000) Population (a) Market subscribers (b) Market subscriber growth Market net adds (c) SIM penetration Population 75 (d) 50% of population below poverty line (e) Addressable population = (a‐d‐e) = (f) Age and poverty adj. penetration % of dual SIM Dual SIMs (g)(1) Market sub adj. for dual SIM (h) Real penetration = (h/f)
2008a 78,066 42,554 37% 11,503 54.5% 17,947 7,338 52,781 81% 0% ‐ 42,554 81%
2009a 79,667 57,020 34% 14,466 71.6% 18,064 7,329 54,273 105% 14% 7,860 49,160 91%
2010e 81,298 74,861 31% 17,842 92.1% 18,207 7,398 55,692 134% 39% 21,726 53,135 95%
2011e 82,929 84,242 13% 9,381 101.6% 18,351 7,464 57,115 147% 50% 28,405 55,837 98%
2012e 84,489 89,206 6% 4,964 105.6% 18,424 7,520 58,546 152% 53% 31,299 57,907 99%
2013e 86,071 91,659 3% 2,453 106.5% 18,524 7,574 59,973 153% 53% 32,006 59,653 99%
2014e 87,607 93,999 3% 2,340 107.3% 18,590 7,622 61,394 153% 53% 32,764 61,235 100%
Source: IMF, AlembicHC Note (1): Derived from the formula : c = {f – (b – g)} + Δf + Δg
Revised mobile market subscriber estimates (m)(1) 90
120%
85 106%
106%
80
107%
108%
102%
75
100%
92%
89
70
110%
92
94
96 90%
84
65
80%
75 72%
60
70%
57
60%
55 2009a Market subs (old est.)
2010e
2011e Market subs (new est.)
2012e 2013e Penetration rate (old est.)
2014e Penetration rate (new est.)
2015e Real penetration
Source: AlembicHC Note: (1) Real penetration deducts those below 9 years and below the poverty line from the total population and adjusts market subscribers to dual SIM usage
Operators’ subscribers and net adds (‘000) Subscribers Mobinil VFE Etisalat Net adds Mobinil VFE Etisalat Market share Mobinil VFE Etisalat
2008a
2009a
2010e
2011e
2012e
2013e
2014e
2015e
20,115 17,600 4,839
25,354 23,325 8,341
30,339 30,734 13,789
33,298 33,742 17,201
34,591 35,062 19,553
35,313 35,797 20,549
36,027 36,522 21,449
36,689 37,196 22,343
4,997 4,267 2,239
5,239 5,725 3,502
4,985 7,409 5,448
2,959 3,009 3,413
1,293 1,319 2,352
722 735 996
713 726 901
662 674 894
47.3% 41.4% 11.4%
44.5% 40.9% 14.6%
40.5% 41.1% 18.4%
39.5% 40.1% 20.4%
38.8% 39.3% 21.9%
38.5% 39.1% 22.4%
38.3% 38.9% 22.8%
38.1% 38.7% 23.2%
Source: AlembicHC
3
Sector note Telecoms | Egypt 14 December 2010
Demographics indicate limited growth going forward compared to previous years
SIM penetration vs subscriber growth
110% 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%
Penetration rate adjusted to demographics 110% 100% 90% 80% 70%
60% 50% 40% 2007a
2008a
2009a
2010e
2011e
SIM penetration rate
2012e
2013e
2007a
2014e
Market subs growth
Source: US Consensus Bureau, AlembicHC
2008a
2009a
2010e
2011e
SIM penetration
2012e
2013e
2014e
Real penetration
Source: US Consensus Bureau, AlembicHC
Compared to other CEMEA countries, Egypt still has some market growth
Population age groups in absolute numbers (m)
Age groups as a percentage of total population
80
100%
3%
80%
2%
60%
1%
40%
0%
20%
‐1%
0%
‐2%
70 60 50 40 30
20 10 ‐
60+
50‐59
30‐49
20‐29
10‐19
0‐9
0‐9
Source: US Consensus Bureau, AlembicHC
10‐19
20‐29
30‐49
50‐59
60+
pop. growth Pop. growth
Source: US Consensus Bureau, AlembicHC
Breakdown of forecasted market net adds (‘000) 20,000 18,000 16,000 14,000 12,000 10,000
1,479
7,860
13,866
8,000 6,000 4,000
10,227
5,114
2,000
6,679
1,492
2,557
‐ 2008a
2009a
Reamining addressable population
2,894
1,419
2010e
1,423 1,278
1,431 639
2011e
2012e
Youth (10 years old)
706 1,427
759 1,421
2013e
2014e
Increase in dual SIM users
Source: AlembicHC
4
Sector note Telecoms | Egypt 14 December 2010
ARPU to dilute by a 2011–14e CAGR of 5% New subscribers are younger than 9 years or hold another SIM We believe ARPU dilution will continue as long as market volatility persists, which is expected to end in 2012e. The fact that real penetration is expected to reach 95% by the end of 2010e means that every additional subscriber should be harder to add since he/she would either be less than 9 years old, already subscribed to another operator, or one of the few remaining from the addressable population. We believe the existing fierce competition (based on our calculation for incremental ARPU) is eroding value, suggesting that existing subscribers for Mobinil and VFE are spending less.
Incremental ARPU Calculation (EGP)(1)
ARPU forecast (EGP)
50
75
46
70
40 30
‐12%
65 33
37
60
‐22%
55 28
20 10
50
14
‐24%
45
7
35
2008a Mobinil
2009a
‐10% 2007a
2010e VFE
Source: Mobinil, VG, TE, AlembicHC Note: (1) Negative incremental ARPU means that existing subscribers are spending less
‐1%
1%
‐21%
20
‐5 2007a
‐5%
‐15%
25 (10)
‐14%
30
1
0
‐14%
40
2008a
2009a
2010e
2011e
Mobinil
‐5%
‐1%
2012e
2013e
0% 2014e
VFE
Source: Mobinil, VG, TE, AlembicHC
Tariff plans mirror operator strategies, all targeting subscriber acquisitions There are 3 main tariff plans in the Egyptian market apart from the recently introduced per second billing. The plans say much about the strategies of the different operators. Mobinil’s plans speak of its recent change of strategy in 3Q10. Previously focusing on enhancing usage with plans characterized by a higher first‐minute rate, call set‐up and subscription fees, Mobinil has clearly signaled its return to the subscriber acquisition war with its newest offer, Bedoun Sheroot (Without Conditions), introduced on 21 October to match VFE and Etisalat’s flat rate on off‐net (19pt) and offering a lower on‐net minute rate (14pt). 5
Sector note Telecoms | Egypt 14 December 2010
Current prepaid tariff plans A)
Mobinil
Al Masry
Bedoun Sheroot
Kol El Masryeen
Kol El Nas
18pt 8pt 28pt 18pt Third min. (on net), Second min. (off net) EGP13.2/month
19pt
19pt
NA
NA
NA
NA
Analysis: Encourages usage beyond (1)
Targeted ARPU Cheapest on‐net rate B) Preferred on‐net numbers
Targeting areas outside Greater Cairo
Analysis: Encourages usage beyond Targeted ARPU(1) Cheapest on net outside Greater Cairo Cheapest off net outside Greater Cairo Cheapest on net inside Greater Cairo Cheapest off net inside Greater Cairo
19pt
Analysis: Encourages usage beyond Targeted ARPU(1) Cheapest rate
Rates for customers outside Greater Cairo On net Off net Rates for customers inside Greater Cairo On net Off net Call set up fee
14pt
Number of preferred numbers Preferred number rate On‐net rate Off‐net rate
C)
Etisalat
Flat Rate
On‐net rates First 2 minutes Starting the third minute Off‐net rates First minute Starting the second minute
Vodafone
Ahsan Nas
Hakawy Friends
Yes
4 5pt(2) 20pt 30pt
3 5pt(2) 20pt 30pt
3 5pt 15pt 25pt
3 minutes/day EGP18.0/Month
3 minutes/day EGP18.0/month
NA NA
Hakawy Regional
Baladna
8pt 14pt
Mohafazat
14pt 24pt
15pt
22pt 35pt 11pt
19pt 24pt NA
First minute of the call EGP8.1/month
NA NA
20pt NA
NA NA
Source: Mobinil, VFE, Etisalat Egypt Note: (1) Assuming usage is enhanced beyond targeted minutes and multiplied by 30 (2) VFE and Mobinil: Customers can make calls to preferred numbers for free (12am–6pm for no more than 120 minutes/day) on the condition that they use three charged minutes per day
6
Sector note Telecoms | Egypt 14 December 2010
Some 60% of subscribers earn EGP548/month, have an average ARPU of EGP9.0… We conducted an exercise to estimate average call duration. The aim was to define the separate subscriber groups that contribute to overall prepaid ARPU differently. For VFE, we assumed all the A+ class and half the A class are postpaid subscribers to stay in line with the company’s reported percentage of postpaid subscribers (4%). For Mobinil, we assumed only the A+ class are postpaid subscribers as Mobinil’s postpaid base represents 2.7% of its total base. We assume that half of the D class does not have mobile phones at all. Since we know Mobinil’s postpaid ARPU is EGP192 and the income of that postpaid base is EGP12.5k/month, we derived the ARPU of the remaining classes based on their respective income. We have done the same for VFE. Some 60% of total subscribers are classified under the C and D classes and have a weighted average income of EGP548/month. The exercise yielded a weighted average ARPU of EGP10.0 for VFE and EGP8.8 for Mobinil. We then divided the ARPU of each class by each company’s respective 3Q10 effective price per minute of 18pt for Mobinil and 20pt for VFE to reach the number of minutes a subscriber uses per month and then again by 30.5 to arrive at an estimate for the number of minutes a subscriber uses per day. … and make calls lasting less than 1 minute and 38 seconds a day The estimated weighted average minutes per day a prepaid subscriber uses is 4 minutes 48 seconds. The number is however inflated by the lower percentage of the prepaid subscriber base that make calls lasting an average of 16 minutes per day (an average 10.5% consisting of the A and B+ class). VFE’s C and D class subscribers, which make up 60% of the total base, are using an average 1 minute and 38 seconds a day, and Mobinil’s 1 minute and 31 seconds. Income distribution in Egypt (EGP) 45%
14,000
40%
Monthly income
12,000
35% 10,000
30%
8,000
25%
6,000
20% 15%
4,000
10% 2,000
5%
‐
0% A+
A
B+ Income
B
B‐ % of population
C
D
Source: CAPMAS, AlembicHC
7
Sector note Telecoms | Egypt 14 December 2010
Calculation for Mobinil’s ARPU by subscriber group and implied call duration Income/ month (EGP)
Implied ARPU (EGP) (a)
Address. pop. (‘000)(2)
% of total subs(3) (b)
% of prepaid subs (c)
Postpaid subscribers A+ 2.0% 12,500 Prepaid Subs A 3.0% 7,500 B+ 7.0% 5,000 B 15.0% 3,000 B‐ 8.0% 2,000 C 40.0% 625 D 25.0% 300 Weighted average prepaid ARPU = (a2)*(c1)+(a3)*(c2)… Weighted average blended ARPU = (a1)*(b1)+(a2)*(b2)… Average prepaid min/ day = (d2)*(c1)+(d3)*(c2)… Average blended min/ day = (d1)*(b1)+(d2)*(b2)…
192 115 77 46 31 10 5
1,592 2,388 5,571 11,938 6,367 31,834 9,948
2.3% 3% 8% 17% 9% 46% 14%
3.5% 8.2% 18% 9% 47% 15%
Class
% of pop.
26 30
Min/ (4) month
Min/day (d)
1,070 644 429 257 172 54 26
35.08 21.1 14.1 8.4 5.6 1.8 0.8 4.8 5.5
Source: AlembicHC Note: (1) Average market postpaid ARPU (2) After subtracting 50% of the D class (3) Includes prepaid and postpaid subscribers (4) Assuming a customer makes only a single call per day and arrived at by dividing the implied ARPU by a flat rate of 18pt
Calculation for VFE’s ARPU by subscriber group and implied call duration Income/ month (EGP)
Implied ARPU (EGP) (a)
Address. pop. (‘000)(2)
% of total subs(3) (b)
% of prepaid subs (c)
Postpaid subscribers A+ 2.0% 12,500 A 1.5% 7,500 Prepaid subscribers A 1.5% 7,500 B+ 7.0% 5,000 B 15.0% 3,000 B‐ 8.0% 2,000 C 40.0% 625 D 25.0% 300 Weighted average prepaid ARPU = (a2)*(c1)+(a3)*(c2) Weighted average blended ARPU = (a1)*(b1)+(a2)*(b2) Average prepaid min/day = (d2)*(c1)+(d3)*(c2) Average blended min/day = (d1)*(b1)+(d2)*(b2)
1,592 1,194 1,194 5,571 11,938 6,367 31,834 9,948
2.3% 1.7% 2% 8% 17% 9% 46% 14%
1.8% 8.3% 18% 10% 48% 15%
Class
% of pop.
188 136 91 55 36 11 5 29 36
Min/ (4) month
Min/day (d)
941
30.9
681 454 273 182 57 27
22.3 14.9 8.9 6.0 1.9 0.9 4.8 5.9
Source: AlembicHC Note: (1) Average market postpaid ARPU (2) After subtracting 50% of the D class (3) Includes prepaid and postpaid subscribers (4) Assuming a customer makes only a single call per day and arrived at by dividing the implied ARPU by a flat rate of 20pt
We believe our estimates are reasonable since Mobinil’s Al Masry offer is pushing beyond 2 minutes per call and VFE and Mobinil’s Preferred Number offers push beyond 3 minutes per day. Our weighted average prepaid and blended ARPUs are also in line with VFE and Mobinil’s, prepaid ARPU of EGP29 and EGP26 and blended ARPU of EGP36 and EGP30. Based on our calculations, implied MOUs are 178 for VFE and 171 for Mobinil, also consistent with the companies’ reported 3Q10 MOUs of 178 for VFE and 173 for Mobinil. 8
Sector note Telecoms | Egypt 14 December 2010
ARPU call duration among subscriber groups for Mobinil(1) 250
Calc. prepaid ARPU = EGP26
200
2% Weighted avg. ARPU = EGP8
ARPU (EGP)
150
3%
100
Calc. blended ARPU = EGP30
8% 50
17%
9% 14% 46%
‐ (2)
‐
2
4
6
8
10
12
14
16
18
20
22
24
(50)
26
28
30
32
34
% of Total Subs % of total subscribers
Call duration (minutes)
Source: IMF, AlembicHC Note: (1) Assuming subscribers make a single call per day
ARPU call duration among subscriber groups for VFE(1) 250
Calc. prepaid ARPU = EGP29
200
4%
ARPU (EGP)
150
Weighted avg. ARPU = EGP10
2%
100
8% 9% 14% 46%
‐ (2)
Calc. blended ARPU = EGP36
17%
50
‐
2
4
6
8
10
12
14
16
18
(50) Call duration (minutes)
20
22
24
26
28
30
32
34
Series1 % of total subscribers
Source: IMF, AlembicHC Note: (1) Assuming subscribers make a single call per day
Mobinil targeting net additions by cutting its off‐net rate with Bedoun Sheroot Mobinil’s newest tariff plan cannibalizes its older plan, Al Masry, offering the lowest on‐ net minute rate in Egypt of 8pt (see page 6 for tariff plans). With Bedoun Sheroot, a 2 minute call costs 28pt and a 3 minute call costs 42pt. With Al Masry, a 2 minute call costs 36pt and a 3 minute call costs 44pt. Bedoun Sheroot is cheaper for those making calls less than 4 minutes, yet more expensive for calls longer than that. Since we calculated that 60% of Mobinil’s subscribers make calls lasting less than 1 minute and 31 seconds per day, we believe once customers migrate to the plan it will be ARPU dilutive for Mobinil unless subscribers make calls longer than 4 minutes. We find this unlikely since Mobinil tried to push beyond the second minute with Al Masry, which seems to have failed, causing the company to offer a plan that cannibalizes its previous one. The new plan offers a lower off‐net rate of 19pt, matching VFE and Etisalat’s flat rates. While this actually stimulates cross‐net usage, it will compress margins due to interconnect costs. 9
Sector note Telecoms | Egypt 14 December 2010
Higher price cuts lead to higher subscriber growth We believe operators will continue to engage in price cuts as this is the only way to ensure net adds in a new reality where VFE and Mobinil are fighting for market leadership. VFE is cutting its effective price per minute more aggressively each quarter and Mobinil has experienced healthy net adds in 3Q10 as a result of cutting its effective price per minute by 30% y‐o‐y (see charts below). Mobinil’s 3Q10 bigger price cut leads to higher subscriber growth 10%
15%
20%
Subs growth y‐o‐y 25% 30%
35%
VFE maintaining subscriber growth through higher price cuts 40%
10%
‐16% ‐20%
20%
Subs growth y‐o‐y 25% 30%
35%
40%
‐12%
2Q10a
1Q10a
4Q09a
3Q09a
‐24% 1Q09a ‐28% 3Q10a
2Q09a
‐32%
% change in effective price/ min
% change in effective price/ min
‐12%
15%
‐20% ‐24% ‐28% 1Q10a ‐32%
‐36% ‐40%
‐40%
3Q09a 4Q09a
‐36%
Source: Mobinil, AlembicHC
2Q09a 1Q09a
‐16%
3Q10a 2Q10a
Source: VFE, AlembicHC
We expect moderate elasticity to remain The question is then whether MOUs will respond to price cuts. Mobinil experienced positive elasticity in 3Q10 of 0.61 after 4 quarters of negative or zero elasticity. MOUs increased 18% in response to the 30% drop in effective price per minute. While this could be regarded as a positive indication, we believe it is mainly a function of the free minutes promotion, which was valid throughout Ramadan (30 days). Since the company does not report billable minutes, the increase in minutes doesn’t necessarily reflect higher spending. Free minutes promotions typically do not last as long as those in Ramadan, hence we don’t expect the higher MOUs to be sustainable. However, our bearish ARPU forecasts (an average 12% drop in 2011e and 5% drop in 2012e) still result in positive elasticity averaging 0.45 for Mobinil and 0.5 for VFE. 10
Sector note Telecoms | Egypt 14 December 2010
Mobinil: Usage vs effective price/minute
Vodafone: Usage vs effective price/minute
30%
‐30%
‐30%
‐40%
‐40% Change in usage y‐o‐y
Dec‐11
Sep‐11
Jun‐11
Mar‐11
Dec‐10
Sep‐10
Jun‐10
% Change in effective price/ min y‐o‐y
Change in effective price/ min y‐o‐y
Source: Mobinil, AlembicHC
Mar‐10
‐20%
Dec‐09
‐10%
Sep‐09
Jun‐09
Dec‐11
Sep‐11
Jun‐11
Mar‐11
Dec‐10
Sep‐10
Jun‐10
Jun‐09
Dec‐09
‐20%
Mar‐10
0%
‐10%
Sep‐09
10%
0% Mar‐09
20%
10%
Mar‐09
30%
20%
% Change in usage y‐o‐y
Source: VFE, AlembicHC
Mobinil and VFE’s forecasted (4Q10e–4Q11e) MOUs, effective price/minute (EGP), and elasticity
Mobinil
MOUs Ef. price/min Elasticity
VFE
1Q10
2Q10 3Q10 4Q10 1Q11 2Q11 3Q11 4Q11 1Q10 2Q10 3Q10 4Q10 1Q11 2Q11 3Q11 4Q11
141 0.22 ‐0.21
159 0.19 ‐0.23
173 0.18 0.61
161 0.18 0.63
165 0.16 0.67
167 0.16 0.28
188 0.14 0.41
168 0.15 0.25
159 0.24 0.31
179 0.20 0.60
178 0.20 0.51
172 0.20 0.51
181 0.17 0.48
201 0.16 0.51
200 0.15 0.51
192 0.16 0.51
Source: Mobinil, Vodafone Group, AlembicHC
Tariffs to drop as SIM penetration increases
Egypt’s mobile tariffs are relatively high compared to other emerging markets. We carried out an exercise using ITU ICT basket data and concluded that Egypt’s tariff basket as measured by ITU is 23% higher than it should be compared to other countries. The mobile tariff basket represents the price of standard monthly usage in USD for 25 outgoing calls per month (on net, off net, and to a fixed line, and for peak, off‐peak, and weekends according to predetermined ratios) and 30 SMS messages. The basket is based on 2H09 prepaid tariffs since they represent the dominant payment method. The basket is then divided by GNI per capita to ensure comparability between countries. We use the chart below only as an indication for tariff trend.
Relative mobile tariff baskets (USD) vs penetration rates for emerging markets 6 Brazil
ITU basket/ GNI
5 4
Philippines
3
Egypt in 2009a
Turkey
Peru
2 Indonesia 1
South Africa Colombia 2011e
2010e India
China
Mexico
2012e Hungary Thailand Malaysia Korea Slovenia Chile
Czech Repupblic Poland Israel
Russia
Bahrain
0 40%
60%
80%
100%
120%
140%
160%
180%
200%
Source: ITU, AlembicHC
11
Sector note Telecoms | Egypt 14 December 2010
Our exercise implies an inverse relationship between penetration rates and tariffs, which is consistent with lower incremental ARPU as a result of acquiring more lower‐income subscribers. Applying our forecasted penetration rate for Egypt of 92% in 2010e, 102% in 2011e, and 106% in 2012e yields a basket/GNI per capita of 1.9%, 1.8%, and 1.7% respectively, implying a drop of 8% in 2011e and 4% in 2012e. VFE and Etisalat launched per second billing, Mobinil not competing
VFE and Etisalat recently launched per second billing. Mobinil had launched per second billing over a year ago but opted not to update the offer after competitors launched their new plans, confirming its strategy of trying to preserve value until 2Q10. As expected, VFE matched Etisalat’s offer the day after.
Per second billing Type of call Peak time Off‐peak time Peak time rates First 30 seconds Second 30 seconds Starting second minute Off‐peak rates First minute Starting second minute
Mobinil On‐net to other mobiles
Vodafone
Etisalat
To landline
Flat rate
Sunday: 8pm–12am Monday to Thursday: 8am–1am Sunday: 12am–8pm Monday to Thursday: 1am–8am All Friday and Saturday
No differentiation
20pt 1.3pt/2 seconds (0.65pt/second or 39pt/minute) 2pt/3 seconds (0.67pt/second or 40pt/minute)
2pt/2 seconds (1pt/second or 60pt/minute)
0.4pt/second (24pt/minute)
20pt 1pt/3 seconds (0.33pt/second or 20pt/minute)
2pt/2 seconds (1pt/second or 60pt/minute)
0.3pt/second (18pt/minute)
20pt
20pt
NA
Min call charge
6pm–12am 12am–6pm
Source: Mobinil, VFE, Etisalat Egypt
Per second billing is cheaper for calls shorter than 2 minutes and 45 seconds
The new per second plans have 2 different rates for peak (0.4pt) and off‐peak (0.3pt) times. We calculated a normalized rate (0.34pt/second) based on the typical Egyptian’s day (starting at 9am and ending at midnight), aggregating both peak and off‐peak rates, and we concluded that while a per second billed call may cost close to a per minute billed call and becomes more expensive the longer the duration of a call, per second billing is cheaper for calls shorter than 2 minutes and 45 seconds as such, we believe the per second billing offer would dilute ARPU since the majority of subscribers make calls shorter than 1minutes and 45 seconds, in line with VFE and Etisalat’s strategy of acquiring subscribers as both are targeting the cost conscious majority.
12
Sector note Telecoms | Egypt 14 December 2010
VFE and Etisalat’s per‐second billing for a 20 minute call 5.0 4.5
Call cost (EGP)
4.0 3.5 3.0
Peak rate starts becoming more expensive
2.5 2.0 1.5 1.0 0.5 0.0 1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Call duration (min) Per‐second peak rate
Per‐second off peak rate
Per‐second normalized rate (1)
Minute rate
Source: Mobinil, VFE, Etisalat Egypt, AlembicHC Note: (1) Assumes a day constitutes 9 off‐peak and 7 peak hours
VFE and Etisalat’s per‐second billing for a 5 minute call 120
Call cost (pt)
100 Before this point the normalized rate is always cheaper
80 60 40 20 0 1
2 Per‐second peak rate
3
Call duration (min)
Per‐second off peak rate
4 Per‐second normalized rate (1)
5 Minute rate
Source: Mobinil, VFE, Etisalat Egypt, AlembicHC Note: (1) Assumes a day constitutes 9 off‐peak and 7 peak hours
13
Sector note Telecoms | Egypt 14 December 2010
Market revenue expected to grow at 2011–14e CAGR of 2.8% Egypt’s telecom spending to GDP ratio is high compared to other emerging and developed countries, in our view. As such, we see limited growth in telecom revenue going forward. We believe revenue will grow at par with real GDP growth in the long term but at a lower rate in the short term due to competition resulting in ARPU dilution. We forecast a 2011–14e CAGR of 2.8% for revenue. Given that we do not expect a continuation of the high historical revenue growth levels, we believe Egyptian telecoms now became a yield rather than a growth play. Etisalat, which investors do not have access to, is the only telco that is expected to show revenue growth above that of the market. Successful margin enhancement remains the only feature to look for when evaluating companies’ future performance.
Telecom spending as percentage of GDP vs GDP per capita (USD ‘000)(1) Emerging market average (2) = 2.8%
6% Telecom spending / GDP
5%
Developed market average = 2.4%
Morocco Malaysia
Tunisia
4%
Egypt
3%
India
2% 1%
Croatia
Spain
Chile Russia Turkey
Algeria China
Switzerland Sweden
Japan Germany Finalnd
Denmark
Australia Ireland Canada Austria
France
Indonesia
Netherlands
0% 0
5
10
15
20
25
30
35
40
45
GDP (PPP) per capita
Source: IMF, Several Telecom Regulatory Bodies, AlembicHC Note: (1) All figures as of 2009 (2) Emerging markets average excludes Malaysia and Morocco
Telecom market revenue forecast (EGPm) 60%
45 48%
40
50%
35 36%
30
40% 28%
25 24%
20
27%
30%
23% 28%
15
20%
23%
10
11% 18%
17%
5%
5 0 2004
2005 Etisalat
2006a TE
2007a VFE
2008a
5%
5%
2009a
2010e
Mobinil
3% 2011e
3% 2012e
Market revenue growth
3%
2%
2%
3% 2013e
3% 2014e
3% 2015e
Market subscriber growth
10% 0%
Source: Mobinil, VFG, TE, AlembicHC
14
Mobinil Telecoms | Egypt 14 December 2010
Mobinil
Subscriber growth at a high price Mobinil’s return to the subscriber acquisition game comes at the expense of profitability, subscribers forecast to grow at a 2011–14e CAGR of 4.4% Reconsider Mobinil by the end of next year, EPS expected to drop 30% y‐o‐y in 4Q10e and to remain flat in 2011e, while we forecast a dividend yield of 4.3% in 2010e and 5.8% in 2011e
We cut our TP 17% to EGP179.6/share and maintain our Neutral recommendation Limited subscriber growth going forward We believe subscriber growth should be limited going forward (2011–14e CAGR of 4.4%) given our expectations of limited market growth (real penetration of 95% in 2010e). ARPU dilution is expected to continue (2011–14e CAGR of ‐4%) as Mobinil is back in the subscriber acquisition fight, with operators competing for existing subscribers leading to increasing dual SIMs. The full consolidation of LINKdotNET will take place in 2011e and as such we expect revenue growth of 5.0% versus only 2.4% for the mobile segment. Profitability is expected to come under pressure until pricing stabilizes in 2012e. We forecast a 2011–14e CAGR of 2.7% for revenue and 3.0% for EBITDA. Operational pressures in 2010e and high expected CAPEX in 2011e lead to lower yields Mobinil’s EPS is expected to drop 30% in 2010e as a result of margin compression. Improvement in revenue is not expected to filter through to the bottom line in 2011e due to 11% higher interest cost leading to a flat EPS. FCF is expected to be negative in 2010e as a result of EBITDA dropping 15% and license payments, while 2011e FCF is expected to be pressured by increased CAPEX. We therefore assume a 50% dividend payout for 2010e, a yield of 4.3%, and expect the payout to increase to 70% in 2011e due to lower operational pressures, which would result in a 5.8% yield.
Paradigm shift confirmed Our view on Mobinil in our previous note Signs of a paradigm shift (29 April 2010) has been confirmed by 2Q10 and 3Q10. We cut our TP to reflect slower market growth, delaying market stability to 2012e, factoring in margin compression and higher interest expense. The stock is trading at a 16% premium on 2011e PE although it trades at a 16% discount to peers on 2011e EV/EBITDA. We still don’t view the discount as sufficiently attractive.
15
Mobinil Telecoms | Egypt 14 December 2010
3Q10 results (EGPm)
3Q10a
2Q10a
%Δ q‐o‐q
3Q10e
%Δ dev.
3Q09a
%∆ y‐o‐y
3Q10c
%Δ dev.
Subscribers (000) Net adds (000) Reported ARPU (EGP) Calculated ARPU (EGP) Effective price/min (EGP) MOUs Revenue EBITDA EBITDA margin Operating profit OPM Net income NPM
28,401 2,253 32.0 30.6 0.18 173.4 2,707 1,086 40.1% 583 21.5% 289 10.7%
26,148 27 32.0 30.1 0.19 159.2 2,530 1,099 43.4% 613 24.2% 381 15.0%
8.6% NM 0.0% 1.4% ‐6.9% 8.9% 7.0% ‐1.2% ‐3.3% ‐5.0% ‐2.7% ‐24.1% ‐4.3%
27,357 1,209 30.6 29.0 NA NA 2,501 1,120 44.8% 645 25.8% 342 13.7%
3.8% 86.4% 4.6% 5.4% NA NA 8.2% ‐3.0% ‐4.7% ‐9.6% ‐4.3% ‐15.5% ‐3.0%
24,625 1,772 38.0 36.9 0.25 146.9 2,793 1,268 45.4% 795 28.5% 491 17.6%
15.3% 27.1% ‐15.8% ‐17.1% ‐29.7% 18.0% ‐3.1% ‐14.4% ‐5.3% ‐26.8% ‐7.0% ‐41.1% ‐6.9%
26,965 817 31 NA NA NA 2,649 1,136 42.9% NA NA 393 14.8%
5.3% 175.8% 2.6% NA NA NA 2.2% ‐4.4% ‐2.8% NA NA ‐26.5% ‐4.1%
Source: Mobinil, AlembicHC
Back in the subscriber acquisition war, but subscriber growth still limited
Operators have the same strategy, limited market growth and regulatory risk We are forecasting a 2011–14e CAGR of 4.4% for Mobinil’s subscribers. We see 3 factors as preventing historical growth rates: (1) limited market growth, (2) subscriber acquisition strategies of the 3 operators, and (3) regulatory risk in the form of a shortage of dials. Mobinil signaled its return to the subscriber fight in 3Q10 with a 27.1% y‐o‐y increase in net adds only possible because of 1m in new dials, driving its utilization rate back to 84% at the beginning of 3Q10. A shortage of dials remains a regulatory risk for Mobinil since management is not able to foresee when the NTRA will extend new dials and on what conditions. Share of net adds 60% 50% 40% 30% 20% 10% 0% 2Q09a
3Q09a Mobinil
4Q09
1Q10a Vodafone
2Q10a
3Q10a Etisalat
4Q10e
Source: Mobinil, VFE , AlembicHC
16
Mobinil Telecoms | Egypt 14 December 2010
Operational efficiency to determine the success of operators in the short term
We believe Mobinil and VFE will most probably lose market share to Etisalat equally in the long term as we believe Mobinil and VFE should reach about the same utilization rate (89%, 90%) and would therefore put both in the same competitive position. It will be a matter of which operator has the operational efficiency to manage its existing dials, recycle old dials in a timely manner, make use of special algorithms, and manage dials allocated to roaming partners. Mobinil was able to increase its operational efficiency in 3Q10 – in 1Q10 it had the same utilization rate and added only 767,000 subscribers. We expect the company to add another 1.9m in 4Q10e with the activation of 1m dials in October and an additional 2m to activate this year. This way of looking at the market puts VFE in the lead but only slightly with a market share just 0.5% higher than Mobinil. Market share and subscriber growth (quarterly)
Market share and subscriber growth (annual)
100%
70%
90%
60%
80%
50%
70% 60%
40%
50%
40% 30%
30% 20%
20% 10%
10% 0%
0% 2Q09a
3Q09a
Marke share
4Q09
1Q10a
Utilization rate
Source: Mobinil press releases, AlembicHC
2Q10a
3Q10a
4Q10e
Subs growth y‐o‐y
2007a 2008a 2009a 2010e 2011e 2012e 2013e 2014e Marke share
Subs growth (y‐o‐y)
Source: Mobinil press releases, AlembicHC
ARPU dilution driven by collective acquisition strategy and increase in dual SIMs in light of market maturity
Subscriber growth remains the source of revenue growth in Egypt as usage does not respond well to moderate price cuts. This was demonstrated by Mobinil in 1H10 when it cut its effective price per minute more cautiously and, as a result, saw 2 quarters of inelasticity. Realizing that usage only responds to aggressive cuts and having more dials on hand, Mobinil cut its effective price per minute 29.7% y‐o‐y in 3Q10, up from an average 18.6% in the previous year, and was consequently able to stimulate usage (MOUs increased an impressive 18% y‐o‐y). Its Bedoun Sheroot offer also indicates that it plans to continue competing on pricing. VFE on the other hand has been experiencing 3 quarters of positive elasticity as it is increasing its price cut each quarter. We believe that at current market maturity level, operators will continue cutting prices as they vie for each others’ subscribers. The increase in dual SIMs is going to put additional pressure on ARPU as subscribers divide their spending wallet on different operators. 17
Mobinil Telecoms | Egypt 14 December 2010
We expect ARPU dilution to continue as long as Mobinil and VFE’s market shares remain close and Etisalat is aggressively pushing its own. Operators’ market share forecast
ARPU forecast (EGP)
50%
40
45% 40%
35
35% 30%
30
25% 20%
25
15% 10%
20 1Q10a
2Q10a
3Q10a
Mobinil
Source: Mobinil, VFE, AlembicHC
4Q10e
2011e
Vodafone
2012e
2013e
2014e
1Q10a 2Q10a 3Q10a 4Q10e 2010e 2011e 2012e 2013e 2014e
Etisalat
Mobinil
VFE
Source: Mobinil, VFE, AlembicHC
Revenue to grow at a 2011–14e CAGR of 2.7% driven by LINKdotNET and some limited subscriber growth
We believe limited subscriber growth and continuing competition leading to ARPU dilution will result in flat mobile revenue in 4Q10e, but we expect some recovery in 2011e when normal levels of net adds return. Factoring LINKdotNET in our forecasts, overall revenue is expected to grow 5.0% next year versus 2.4% for the mobile segment. Normal growth rates should resume thereafter (2011–14e CAGR is 2.7%). We believe subscriber growth in 2011e is going to come at the expense of margins as competition is expected to remain in place into 2011e. The consolidation of LINKdotNET is also expected to put some pressure on margins (2010–11e CAGR of 3.0%) – we expect an average EBITDA margin for the internet segment to be 14%. 18
Mobinil Telecoms | Egypt 14 December 2010
Aggressive acquisition efforts compress margins until market stability in 2012e. Revenue, EBITDA margin forecast by segment (EGPm) 47.4%
12,500
50%
42.4% 41.1%
40.8%
12,000
42.1%
41.6%
45% 40%
11,500
35%
11,000
30% 25%
10,500 10,000
11.8%
13.5%
14.0%
2011e
2012e
20%
14.4%
14.2%
15%
9,500
10% 2009a
2010e
Mobile revenue
Internet revenue
Mobile EBITDA margin
2013e
2014e
Internet EBITDA margin
Overall EBITDA margin
Source: AlembicHC
Mobinil revenue and EBITDA growth Full effect of LINKdotNET consolidation is the driver of revenue growth in 2011
40% 30%
2011–14e CAGR for revenue of 2.7% and EBITDA of 3.0%
20% 10% 0% ‐10% ‐20% 1Q09a
2Q09a
3Q09a
4Q09a
1Q10a
2Q10a
3Q10a
4Q10e
Revenue growth y‐o‐y
2006a
2007a
2008a
2009a
2010e
2011e
2012e
2013e
2014e
EBITDA growth y‐o‐y
Source: AlembicHC
Pressures on yields to persist into next year EPS to drop 30% in 2010e and remain flat in 2011e We believe revenue and margin compression will cause this year’s EPS to fall 30%. The improvement in revenue (5.0%) in 2011e is not expected to filter through to the bottom line as interest expense is expected to increase 11% as management has plans to continue leveraging its balance sheet in 2011e by an additional EGP2bn. Therefore, EPS is expected to remain flat in 2011e.
19
Mobinil Telecoms | Egypt 14 December 2010
Debt accumulation until 2011e will put additional pressure on the bottom line (EGPm) Exceptionally high due to USD60m of forex losses
12,000
35%
32% 30%
10,000
Total debt
8,000
25%
25%
24%
22%
6,000
26%
25%
25%
30%
28%
20%
20%
17%
25%
24%
22%
20% 15%
4,000
10% 6%
2,000
5%
5%
0
0% 1Q08a 2Q08a 3Q08a 4Q08a 1Q10a 2Q10a 3Q10a 4Q10e
2006a 2007a 2008a 2009a 2010e 2011e 2012e 2013e 2014e
Total debt
Interest expense as % of EBIT
Source: Mobinil, AlembicHC
EPS (EGP) and EPS growth 20
35%
18
25%
16 14
15%
12%
12
10% 5%
10%
6%
5%
10
5%
‐2%
8
‐5%
6
‐15%
4
‐25%
2
‐30%
0
‐35% 2007a
2008a
2009a
2010e EPS
2011e
2012e
2013e
2014e
EPS growth
Source: Mobinil, AlembicHC
Unattractive FCF and dividend yield in 2010e and 2011e FCF is expected to be negative this year given our forecast of 15% drop in EBITDA along with EGP1.85bn in license payments (of which EGP1.1bn expected to be made in 4Q10e). FCF in 2011e is expected to be pressured (FCF yield of 3.0%) by additional CAPEX (20% of sales versus 16% in 2010e) as we expect normal levels of CAPEX to resume after the company saved on CAPEX in 2010e to be able to make license payments and purchase LINKdotNET. As such, we assumed a dividend payout of 50% for 2010e (in line with 2009), leading to a dividend yield of 4.3%, and expect the payout to increase to 70% in 2011e as operational pressures recede, which will however still result in a 5.8% yield. We believe investors should consider Mobinil again by the end of next year when we expect operational recovery, and according to our estimates FCF yield reaches 9.6%, with dividend yield of 6.6% by 2012e. 20
Mobinil Telecoms | Egypt 14 December 2010
Dividend yield and FCF yield pressured until 2011e (EGP) 15% 10.3%
9.6% 10%
8.4%
7.0%
5.7%
3.9%
7.6%
EPS
5%
11.6%
5.6%
‐0.1% 4.3%
2009a
2010e
‐2.4%
0% ‐5%
8.3%
3.0% 5.8%
6.6%
2011e
2012e
9.2%
‐7.7%
‐10% 2006a
2007a
2008a
Dividend yield
2013e
2014e
FCF yield
Source: AlembicHC
Mobinil FCF calculation (EGPm) EBITDA CAPEX Interest expense Taxes Change in WC FCF FCF yield
2009a 5,122 (2,776) (688) (536) (188) 935 5.6%
2010e 4,371 (3,541) (596) (401) 143 (23) ‐0.1%
2011e 4,518 (2,921) (692) (369) (31) 507 3.0%
2012e 4,637 (2,087) (679) (387) 120 1,604 9.6%
2013e 4,769 (2,012) (603) (428) (3) 1,723 10.3%
2014e 4,914 (1,967) (534) (473) (2) 1,939 11.6%
Source: Mobinil, AlembicHC
Revised estimates (EGPm) Subscribers ('000) Net adds ('000) ARPU (EGP) Revenue EBITDA EBITDA margin Net income EPS (EGP) EPS growth EV/EBITDA (x)
New 2010e
Old 2010e
∆ Dev.
New 2011e
Old 2011e
∆ Dev.
30,339 4,985 29.4 10,712 4,371 40.8% 1,423 13.2 ‐30.1% 5.1
27,966 2,612 29.5 10,098 4,402 43.6% 1,508 14.0 ‐25.9% 5.5
8.5% 90.9% ‐0.3% 6.1% ‐0.7% ‐5.6% ‐5.6% ‐6.5%
33,298 2,959 26.5 11,246 4,518 40.2% 1,391 12.9 ‐2.3% 5.0
30,721 2,756 27.1 10,223 4,499 44.0% 1,284 11.9 ‐14.8% 5.2
8.4% 7.4% ‐2.4% 10.0% 0.4% 8.3% 8.3% ‐3.9%
Source: AlembicHC
Our valuation We cut our TP to reflect slower market growth, delaying market stability to 2012e, adding LINKdotNET to our numbers, factoring in margin compression, and higher interest expense. Upside risks to our valuation would be lower ARPU dilution and Orascom Telecom selling its stake in Mobinil by exercising its put option for EGP221.7/share, which will force France Telecom to launch a mandatory tender offer to minorities.
21
Mobinil Telecoms | Egypt 14 December 2010
Mobinil is trading at a 16% discount to its peers on 2011e EV/EBITDA yet at a 16% premium on 2011e PE 2011e PE vs EPS 2011–13e CAGR
2011e EV/EBITDA vs EBITDA 2011–13e CAGR 10
20 18
EV/EBITDA 2011e
PER 2011e
Relaince 16 Zain
14 Mobinil 12
Turkcell MTN
Cellcom 10 Partner Omantel 8
Etisalat Qtel
‐10%
Cellcom
10%
Source: Mobinil, Bloomberg , AlembicHC
Turkcell
6
Wataniya Qtel
5 Mobinil Omantel 4 Etisalat TE
MTN
3 ‐4%
20%
EPS 2011‐2013e CAGR
7 Partner
Wataniya TE
0%
Reliance Maroc Tel
8
Batelco
6 ‐20%
9
Zain
Bharti
‐2%
0%
2%
4%
EBITDA 2011‐2013e CAGR
6%
8%
10%
Source: Mobinil, Bloomberg , AlembicHC
22
Mobinil Telecoms | Egypt 14 December 2010
Financial statements and ratios (EGPm)
2009a
Income statement Revenue Mobile revenue Mobile revenue growth (%) Internet revenue Total revenue Total revenue growth (%) EBITDA EBITDA margin Depreciation, amortization Operating profit OPM Other non‐operating income Net interest expense PBT Taxes Minority interest Net profit Net profit margin Appropriations Net profit after appropriations EPS Growth in EPS (%) Dividends/share Dividends payout (%) Dividend yield FCF yield Balance sheet Intangible assets, other Tangible assets and investments Total fixed assets Total current assets Total current liabilities Total long‐term liabilities Shareholder equity, minorities Cash flow statement Cash flow from operations Interest Taxes CAPEX + investment Dividends paid Net cash flow pre financing Financing Change in cash Key ratios Net debt/EBITDA CAPEX to sales
Source: Mobinil, AlembicHC
2010e
10,807 8.0% 10,807 8.0% 5,122 47.4% ‐1,907 3,215 29.7% 47 ‐688 2,574 ‐536 0 2,038 18.9% ‐145 1,893 20.4 3.9% 9.5 46.6% 5.7% 5.6% 2,970 9,800 12,770 1,869 5,597 5,364 3,679 5,061 ‐654 ‐331 ‐2,398 ‐932 747 ‐767 ‐21 0.8 24.2%
2011e
10,586 ‐2.0% 126 10,712 ‐0.9% 4,371 40.8% ‐1,973 2,398 22.4% 22 ‐596 1,824 ‐401 0 1,423 13.3% ‐100 1,324 14.2 ‐30.2% 7.1 50.0% 4.3% ‐0.1% 4,272 9,888 14,160 1,968 5,034 6,909 4,184 4,090 ‐543 ‐420 ‐3,371 ‐889 ‐1,601 1,744 119 1.3 16.0%
2012e
10,838 2.4% 408 11,246 5.0% 4,518 40.2% ‐2,067 2,451 21.8% 0 ‐692 1,759 ‐369 0 1,391 12.4% ‐97 1,293 13.9 ‐2.3% 9.7 70.0% 5.8% 3.0% 4,722 10,291 15,013 1,922 4,397 6,964 5,575 4,640 ‐692 ‐371 ‐2,921 ‐712 ‐55 55 0 1.3 20.0%
2013e
10,986 1.4% 458 11,444 1.8% 4,637 40.5% ‐2,095 2,542 22.2% 0 ‐679 1,863 ‐387 0 1,475 12.9% ‐103 1,372 14.8 6.1% 11.1 75.0% 6.6% 9.6% 4,422 10,584 15,005 1,785 4,565 6,281 5,944 4,791 ‐679 ‐369 ‐2,087 ‐974 683 ‐683 0 1.1 19.0%
2015e
11,404 2.0% 528 11,932 2.3% 4,914 41.2% ‐2,109 2,806 23.5% 0 ‐534 2,272 ‐473 0 1,799 15.1% ‐126 1,673 18.0 10.4% 15.3 85.0% 9.2% 11.6% 3,821 10,947 14,768 1,836 5,026 5,120 6,458 4,861 ‐534 ‐428 ‐1,967 ‐1,386 546 ‐546 0 0.8 17.3%
11,737 2.9% 560 12,297 3.1% 5,105 41.5% ‐2,106 2,999 24.4% 0 ‐473 2,526 ‐525 0 2,001 16.3% ‐140 1,861 20.0 11.2% 17.0 85.0% 10.2% 12.8% 3,521 11,099 14,620 1,875 5,220 4,518 6,758 5,036 ‐473 ‐473 ‐1,959 ‐1,529 603 ‐603 0 0.7 16.7%
11,175 1.7% 488 11,663 1.9% 4,769 40.9% ‐2,107 2,661 22.8% 0 ‐603 2,058 ‐428 0 1,630 14.0% ‐114 1,516 16.3 10.5% 13.9 85.0% 8.3% 10.3% 4,122 10,788 14,910 1,808 4,864 5,666 6,188 4,723 ‐603 ‐387 ‐2,012 ‐1,107 615 ‐615 0 1.0 18.0%
2014e
23
Telecom Egypt Telecoms | Egypt 14 December 2010
Telecom Egypt
Strong yields
Revenue expected to remain flat in the medium term due to wholesale, but TE still offers high FCF and dividend yield of 16.6% and 8.1% expected in 2010e
In our view, concerns about interconnect dispute and MVNO are minimal
Reiterate Overweight view and lower TP 2.0% to EGP21.1/share
We forecast a flat top line growth
On the retail front, broadband revenue (2011–14e CAGR of 18.5%) is expected to offset the drop in voice and access (CAGR of ‐3.1%). Domestic wholesale is expected to grow at a CAGR of 5.6%, offsetting the drop in international revenue (CAGR of ‐2.4%) now that Etisalat also has an international gateway. We expect the EBITDA margin to remain volatile as TE recognizes different types of cable revenue. We expect continued pressure due to competition with mobile operators until stability kicks in 2012e, but EBITDA is expected to follow its normal trend going forward with a 2011–14e CAGR of 0.8%.
VFE to continue cutting prices
We expect VFE to continue cutting prices aggressively to maintain its market leadership position, which we believe will lead to considerable ARPU dilution (2011–14e CAGR of ‐4.9%). Compared to Mobinil, VFE might seem better poised, but we expect revenue and EBITDA to see increasing pressure compared to historical levels. We expect revenue to remain flat in 2010e and EBITDA to drop 4.7%. We believe revenue will grow at a 2011– 14e CAGR of 2.3% and EBITDA 2.5%. We lower our TP for VFE to EGP8.5 per TE share based on revised estimates in line with our bearish view on the market and reflecting margin compression as a result of competition. Two concerns with minimal downside risk
On the interconnect dispute, we believe that if the old regime is restored it will bring down the EBITDA margin to 45.4% from 47.1% in 2011e and to 46.2% from 48.2% in 2015e, push the dividend yield down to 8.3% from 8.7%, and dilute our TP by 3.9% to EGP20.3. The other concern is the Mobile Virtual Network Operator (MVNO) setup, which we believe TE will take as a short‐term solution to penetrate the mobile market, which is also its cheapest option. TE’s plans to establish a full MVNO would require a level of investment (already factored in our numbers), which we believe would have a minimal effect on its dividend distribution although we did not forecast any potential upside in our model.
We cut our TP 2.0% to EGP21.1/share and maintain Overweight
We lower our TP 2.0% to EGP21.1/share and maintain our Overweight rating on the stock due to (1) TE’s strong balance sheet (cash balance to rise from 8% in 2010e to 24% of total assets in 2015e), (2) a sustainable high FCF yield averaging 14.2%, (3) an EPS 2011–14e CAGR of 5.6% fuelled by higher EBITDA and interest income, and (4) an expected dividend yield of 8.1% in 2010e and 8.7% in 2011e. We believe that dividend distribution can increase further given the strong cash accumulation. The stock trades at a 36% and 10% discount to peers on 2011e EV/EBITDA and PE, respectively. 24
Telecom Egypt Telecoms | Egypt 14 December 2010
Revenue expected to remain flat in the medium term
Retail revenue to remain under pressure (2010–14e CAGR of 1.5%) saved by the data segment (CAGR of 18.5%)
We expect voice revenue to continue dropping (2011–14e CAGR of ‐2.5%) as a result of TE’s efforts to compete on pricing with mobile operators as was the case in 4Q09 when it offered a fixed‐to‐mobile rate of 15pt per minute. We expect some improvement in the voice segment once pricing stabilizes by 2012e and average growth starting 2013e of 1.8% due to lower ARPU dilution and the resumption of moderate subscriber growth (2011–14e CAGR of 1.1%). We postponed the second fixed‐line license to 2013 (to be consistent with statements from the Telecom Ministry) and expect penetration to reach 13.2% by 2014e as a result of competition. Access revenue is also expected to remain under pressure (2011–14e CAGR of ‐3.8%) as result of promotions eliminating installation costs and lower net adds. Growth in internet and data revenue (2011–14e CAGR 18.5%) on the back of subscriber growth (CAGR of 24.9%) is expected to offset the decline in both voice and access revenue and drive overall retail revenue to grow at a 2011–14e CAGR of 1.5% on our estimates. Fixed‐line penetration vs GDP (PPP) per capita
Broadband penetration vs GDP (PPP) per capita 10%
35%
Turkey
Russia
25%
China
Brazil
Hungary Poland Czech Rep.
Turkey Chile
20% Colombia
Indonesia
15%
Broadband penetration (%)
Fixed‐line penetration (%)
30%
9%
Mexico Malaysia
Thailand
Morocco
10%
Egypt
Peru Philippines
5%
South Africa
0
2
4
8%
China
8
10
12
14
16
18
20
22
5%
Colombia
4% Morocco
3%
Philippines
2%
India Pakistan 0
24
2
Peru
Egypt
Sri Lanka 4
South Africa
Thailand 6
8
10
12
14
GDP(PPP) per capita ('000)
GDP (PPP) per capita ('000)
Source: ITU, IMF, AlembicHC
Russia
Malaysia
6%
0% 6
Chile
Brazil
7%
1%
India
0%
Mexico
Source: ITU, IMF, AlembicHC
Wholesale revenue to remain flat at a 2011–14e CAGR of ‐0.2%
Domestic wholesale has seen an impressive pickup in 9M10 of 16.2% y‐o‐y mainly driven by the increased mobile traffic. With increased usage from mobile subscribers this trend is expected to continue and the segment is expected to grow at a 2011–14e CAGR of 5.6 %. International wholesale on the other hand, which includes revenue from the cable system business, is expected to decline at a CAGR of ‐2.4% as additional revenue from the cable system is not expected to compensate for the drop in the remaining revenue since Etisalat now has its own international gateway and is competing aggressively on this front. International wholesale revenue excluding the cable system business is expected to drop at a 2011–14e CAGR of ‐2.6%. Cable system revenue, according to management guidance, is expected to be around USD500m–USD600m for the entire lifetime of the cable business, which we divided equally over the years following 2011 amounting to EGP523m each year.
25
Telecom Egypt Telecoms | Egypt 14 December 2010
Retail revenue by segment (EGPm)
Wholesale revenue by segment (EGPm) 6
14
7
2011–14e CAGR of ‐0.2%
2011–14e CAGR of 1.5%
5
12
5
10
4
4 8 3 6
2 1 0 2008a
2009a
Access Others
2010e
2011e
2012e
2013e
Voice Access lines
Access lines (m)
6
3 2
4
1
2
0 2008a
2014e
2009a
2010e
2011e
2012e
2013e
2014e
Internet and data Domestic
Source: TE, AlembicHC
International
Cable systems
Source: TE, AlembicHC
EBITDA to remain flat despite volatility caused by cable revenue
Volatility to remain with recognition of different types of cable revenue
The EBITDA margin has been volatile over the past 3 quarters due to the recognition of different types of cable system revenue. The margin was exceptionally high in 1Q10 due to the recognition of the dry part of the transit corridor revenue, which has high margin revenue of around 90%, according to management as discussed in the earnings conference call. On the other hand, 2Q10 and 3Q10 EBITDA included cEGP135m and EGP99m (respectively) in costs related to the wet part of the transit corridor that has a margin that ranges from 10% to 15%. The effect of the volatility is apparent in other operating costs that saw a hike of 44% q‐o‐q in 2Q10 yet remained almost flat in 3Q10 given the recognition of the same type of revenue. This quarter, TE changed the method for calculating EBITDA, taking out provisions and impairments, which provides a more consistent view on the EBITDA margin and eliminates some volatility.
Operating costs and EBITDA margin (EGPm)(1) 750 650
60% 53%
54%
54%
54%
Operatng costs
55%
52%
550 450
50%
48%
350
44%
43%
250
45%
150
40%
50 ‐50
1Q09a
2Q09a
3Q09a
Salaries and wages
4Q09a Other expenses
1Q10a
2Q10a
Interconnection fees
3Q10e
4Q10e
35%
EBITDA margin
Source: TE, AlembicHC Note: (1) For consistency, the figures above are all according to EAS
26
Telecom Egypt Telecoms | Egypt 14 December 2010
EBITDA to grow at a 2011–14e CAGR of 0.8% We believe EBITDA will decrease only 2.6% in 2010e (lower drop than in 9M10 of 5.0%) as a result of high margin revenue from the cable system business. EBITDA excluding this segment is expected to fall 5.8% in 2010e along with a 3.7% decline in revenue versus a growth of 1.7% caused by cable revenue. We believe revenue and EBITDA will remain under pressure as long as competition is rigorous in the mobile market and will experience some recovery starting 2012e. We expect revenue to grow at a 2011–14e CAGR of 0.7% and EBITDA at 0.8%.
Revenue and EBITDA growth including cable
Revenue and EBITDA growth excluding cable
6%
6% 4%
4%
2% 0%
2%
‐2% 0%
‐4% ‐6%
‐2%
‐8% ‐4%
‐10% ‐12%
‐6%
‐14% 2007a
2008a
2009a
2010e
Revenue growth (y‐o‐y)
Source: TE, AlembicHC
2011e
2012e
2013e
2014e
2007a
2008a
2009a
2010e
Revenue growth (y‐o‐y)
EBITDA growth (y‐o‐y)
2011e
2012e
2013e
2014e
EBITDA growth (y‐o‐y)
Source: TE, AlembicHC
27
Telecom Egypt Telecoms | Egypt 14 December 2010
More than a traditional MVNO: Differentiation through convergence Entering the mobile market through an MVNO setup is a first stage The Telecom Ministry decided to postpone offering a fourth mobile license beyond 2013 and discussions between TE and VG to purchase the latter’s stake in VFE were not concluded, meaning that TE’s only option to venture into the mobile space would be buying Egypt’s first MVNO license. TE is currently studying the viability of being a virtual mobile operator and is expected to reach a final decision by 1Q11. We believe TE will continue to eye VG’s stake in VFE or bid for Egypt’s fourth (4G) mobile license therefore the MVNO setup would provide TE with a short term solution (the cheapest as well) to access Egypt’s mobile market. We previously argued in favor of fixed‐to‐mobile convergence (FMC) In our last note Not a doomsday scenario (6 April 2010) we argued in favor of a fourth mobile license as we are confident in the opportunities FMC can offer. TE’s MVNO venture will still play on convergence but with an alternative technology deploying femtocells. Femtocells are small cellular base stations (very similar to Wi‐Fi routers) that allow a customer to connect to the broadband network once indoors and be charged lower rates in line with the fixed‐line rate. In essence, TE will be able to compete with mobile operators by utilizing femtocells to capture mobile users when at home (60% of voice and 70% of data usage occurs indoors) and divert their mobile originated voice and data calls to the fixed‐line network, like with VoIP. The femtocell value proposition Femtocells offer customers six main benefits: (1) improved in‐home mobile coverage, (2) enhancing video and audio downloading, (3) cheaper in‐home call/ data rates, (4) higher mobile internet speed, (5) improved phone battery life (especially for Wi‐Fi users), and (6) various value‐added services, of which the most appealing would be allowing customers to have a virtual home number, meaning they would require only one handset instead of two. As such, we see more to the story than the traditional minute reseller as we believe in the value proposition convergence could offer. Reasoning behind TE and VFE’s willingness to agree on an MVNO setup TE is likely to opt for an MVNO setup on VFE’s network since TE would probably receive a better wholesale price than from mobile operators. VFE’s CEO already expressed his willingness to the Egyptian press. We believe VFE would benefit from (1) delaying the issuance of a fourth mobile license in the market and ensuring that they get a share of the forth entrant’s revenue and (2) delaying TE’s acquisition talks regarding VG’s stake in VFE. Full MVNO: Relative independence of MNO and various synergies TE plans to launch a full (also called heavy) MVNO to capitalize on the synergies it can extract from its fixed‐line business including a large distribution network in all Egyptian governorates, a customer service call center, experienced employees in various functional areas, an international gateway, and the fixed‐line infrastructure (saving on
28
Telecom Egypt Telecoms | Egypt 14 December 2010
interconnect costs and offering convergence to the broadband network). In doing so the only thing it would be missing is the spectrum. Full MVNO is the most capital intensive type, yet downside to TE’s cash flow is minimal There are several types of MVNO players, each with different technical requirements and costs. In general, MVNOs don’t have a spectrum of their own for access service but can provide it to customers through an agreement with a mobile network operator (MNO). Just as MNOs currently lease TE’s network for their ISPs and international calls, TE would be indirectly leasing VFE’s to offer 2G and 3G wireless access services. TE is aiming to become a full MVNO, which means there will be investment requirements and therefore cash flow will see some pressure. We decided to incorporate potential investments in our cash flow forecasts to take into account a worst case scenario for dividends. We factored in an MVNO license fee of EGP550m and CAPEX of EGP150m in 2011e onwards. Despite these assumptions we still get a dividend yield of 8.7% in 2011e and 9.1% in 2012e. MVNO operations could start in less than a year According to management, it will take TE 6‐9 months to launch its services or the company could opt to involve a Mobile Virtual Network Enabler (MVNe), which would allow operations to commence immediately. An MVNe provides infrastructure and services to enable MVNOs to offer services and focus on their relationship with customers. In TE’s case an MVNe would probably provide services from the mobile switching center to billing. MVNO business models: Full MVNO = MNO – radio access Radio access
Mobile switching center
Network services
Application services
MNO
Billing
Customer care
Distribution
Marketing and branding
Sales
Full MVNO
MNO
Intermediate/ hybrid MVNO
MNO
Thin MVNO/ enhanced service provider
MNO
Reseller
Source: TRAI, AlembicHC
More than just another bundling strategy We believe TE aims to provide its customers with a “total telecom offering.” To put it simply, it wants to bundle its triple play services at discounted rates (especially in‐home zone rate). We understand management’s point of view that mobile services could serve as a means to stop the churn on its fixed‐line network and see a potential competitive advantage in convergence over its MNO competitors. Femtocells might potentially persuade families with multiple operator accounts to consolidate service plans.
29
Telecom Egypt Telecoms | Egypt 14 December 2010
MVNO strategies
Cost leadership
• Definition: Focusing on providing simple, no‐extra‐costs mobile services mainly to the prepaid segment at prices undercutting the incumbents' offerings. • Comment: Given the current level of pricing we believe VFE will limit TE's retail price as aggressive pricing would cannibalize VFE's customer base.
• Definition: Focusing on utilizing brands with high sentimental value to create service features to persuade consumers that the service is superior to competitors'offerings. • Comment: TE's brand name is in Egypt associated with the idea of a government‐owned entity. Hence, we don't believe there is much to capitalize Differentiation on in this regard.
• Definition: A cost leadership or diffrentiation strategy applied to a specific segment of a market such as youth, an ethnic group, etc. • Comment: MNOs have been successful in reaching all segments of the market, especially now with a real penetration rate expected to reach 95% by the end of 2010. The Egyptian population is very homogineous making it hard to classify in this manner. At best, the population can be Market segmentation segmented by income distribution, which brings us back to cost leadership.
Bundling
• Definition: Bundling, also referred to as differentaited cost leadership, is a strategy of bundling services (i.e. triple play at reduced rates) as a differentiation point to raise the customer's willingness to pay. • Comment: We believe that TE will go beyond the traditional bundling. With a 62% market share of the broadband market there is definitely an element to capitalize on. TE will also be the first to introduce femtocells to the market.
Source: TRAI, AlembicHC
Femtocell potential customers
Femtocells are likely to appeal to high ARPU households and consumers with mobile data plans or 3G smartphones. We believe these two characteristics apply mostly to A+ and A class subscribers, 4.5m who have an ARPU above EGP115 (see page 8) and 70% of whom are postpaid subscribers. This is in direct contrast with TE’s fixed‐line service, of which a high percentage are lower‐income customers. As such, we see ARPU enhancement as a potential upside as the service would also limit fixed to mobile substitution. TE Data currently has 819,000 broadband connections to households consisting of 5.5 residents on average. We forecast 1.1m broadband subscribers by the end of 2011e. Assuming there are 4 adults with mobile handsets in each household, we believe TE could potentially add 4.6m subscribers in 2011e or 2012e (in line with our calculation of 4.5m of A+ and A class subscribers).
The potential of femtocell subscribers (‘000)
Some 60% of voice, 70% of data usage occurs indoors
12,000
120%
10,000
100%
1,589
8,000
80%
1,454 1,327
6,000
45%
60%
1,214
24%
4,000 2,000
36%
4,649
20%
‐
25%
40% 40%
30%
0% 2011e
Femtocell subscribers
Source: AlembicHC
2012e
2013e
Femtocell net adds
2014e
2015e
Broadband subscribers
Voice Elsewhere
Data Office
Home
Source: Informa Telecoms & Media, Mobile Broadband Access at Home
30
Telecom Egypt Telecoms | Egypt 14 December 2010
Concerns surrounding the MVNO endeavor While we believe in the opportunities the service has to offer, its implementation and customer appeal remain questionable. The technology is new, and although 17 operators in 11 countries have already adopted it, subscriber numbers are not disclosed and marketing initiatives have not been aggressive. The only exception to this has been Vodafone UK, which has slashed femtocell prices 70% since their launch in 2009 and rebranded the service to Sure Signal in an attempt to market it for areas with limited coverage. Pricing is also a potential issue since a femtocell’s selling price is cUSD200, and operators are still testing retail pricing schemes to offset the cost of service and equipment while stimulating adoption. Pricing ranges from a complete subsidy (KDDI and Softbank in Japan) to monthly installments (ranging from USD9 to USD32). Our bearish outlook on the mobile market is based on real penetration reaching 95% in 2010e, which indicates that there is no room for another market entrant, and means high churn for other operators if the service gains customer appeal. Therefore, we chose not to reflect any upside potential in our model but only the worst case scenario (CAPEX cash out flows) on dividend distribution. MVNO setup with VFE is a double‐edged sword for TE Cannibalizing VFE’s market share would affect its revenue, which will eventually reflect on TE’s investment income booked from VFE. On the other hand, VFE would be adding a new revenue segment that would reflect positively on TE’s investment income. Whether these 2 phenomena would offset each other and how they will eventually affect investment income from VFE (an important element to TE’s bottom line) is uncertain in our view as it will be dependent on the wholesale agreement and the MVNO’s rate of success.
31
Telecom Egypt Telecoms | Egypt 14 December 2010
MVNO implications on TE–VFE relationship in specific markets and the telecom market as a whole TE’s potential mobile subscribers
VFE’s subscribers
2 Competition
Other mobile operators
Other mobile subscribers
5
Unserved market segments
Fixed‐line subscribers
Mobile pricing
3
1
Network
45% ownership translating into investment income to TE
VFE
TE New revenue segment: Wholesale and lower OPEX
Wholesale agreement
New revenue segment: Mobile and higher OPEX
4
1: Effects on TE’s investment income can’t be assessed as there is a cannibalization effect on VFE’s subscribers. At the same time, VFE will be adding a new revenue segment. 2: The higher TE’s subscribers become the more CAPEX VFE must spend to maintain the network; poses a risk of network congestion. 3: Retail price is limited by the wholesale agreement. 4: The higher TE’s subscribers become the lower the rate of growth of its wholesale domestic revenue from mobile to fixed termination. 5: Taking off VFE’s market share: cannibalization effect, greater customer churn.
Source: AlembicHC
Interconnect dispute effects
Another cash flow concern is the effect of the interconnect dispute if the court ruling to cancel the new interconnect regime is implemented. There are 4 main elements that we need to highlight to explain the basis of our exercise:
(1) The earliest such a ruling could come into effect is 1Q11.
(2) The total claims mobile operators owe to TE are EGP573m as of 3Q10, 75% of which are already provisioned for (EGP430m), according to management, and the remainder will be charged to the income statement if the dispute is lost.
(3) Adding the claims to the interconnect fees drives up the ratio of interconnect fees to revenue to 11.4% from 9.7% in 2009 and 2010e – while we know that this ratio decreases over time, we remain conservative and keep it stable throughout our forecast.
(4) Additional claims for 4Q10e are estimated at EGP21m and will also be charged to the income statement. 32
Telecom Egypt Telecoms | Egypt 14 December 2010
Putting these elements in our model would lead to an EBITDA margin down to 45.4% from 47.1% in 2011e and to 46.2% from 48.2% in 2015e, diluting our fair value per share 3.9% to EGP20.3/share. Our FCF yield would drop to 10.7% versus 11.2% and dividend yield would drop to 8.3% versus 8.7%). Effect of reinstating the old interconnect regime on EBITDA margin 54% 52%
52%
50%
49%
49% 48%
47%
46%
47%
47%
47%
48%
45%
45%
46%
46%
2011e
2012e
2013e
2014e
44% 42% 2007a
2008a
2009a
2010e
EBITDA margin
EBITDA margin (adjusted for old interconnect regime)
Source: TE, AlembicHC
A strong yield story
Despite the overhang on the stock caused by the interconnect dispute and the MVNO, TE continues to be a strong FCF generator and has reinstated its promise to investors to remain a dividend yield story. As a result of the deleveraging, TE’s EPS is expected to grow at a 2011–14e CAGR of 5.6%, its dividend yield should increase to 11.6% in 2015e from 8.1% in 2010e and its FCF yield would average around 14.2% (see chart below), according to our forecasts.
EPS, dividend yield, and FCF stronger each year (EGP) 2.8 21% 2.4
19%
EPS
2.0 1.6
16.8%
17%
16.6% 14.9%
13.9%
13.3%
1.2
13.9%
11.2% 0.8 0.4 0.0
7.4%
7.4%
8.1%
8.7%
9.1%
9.9%
14.6%
15.4%
13% 10.7%
11.6%
11% 9% 7%
5.7% 2007a
15%
5% 2008a EPS
2009a
2010e
2011e
2012e Dividend yield
2013e
2014e
2015 FCF yield
Source: TE, AlembicHC
33
Telecom Egypt Telecoms | Egypt 14 December 2010
TE FCF calculation (EGPm) EBITDA CAPEX Interest expense Taxes Change in WC Dividends received from VFE FCF FCF yield
2009a 5,014 (981) 11 (453) 171 703 4,465 14.9%
2010e 4,956 (993) 206 (524) (25) 1,354 4,974 16.6%
2011e 4,797 (2,146) 206 (539) 62 979 3,359 11.2%
2012e 4,811 (1,504) 271 (570) (13) 1,007 4,002 13.3%
2013e 4,918 (1,497) 397 (637) (46) 1,041 4,175 13.9%
2014e 5,060 (1,487) 532 (712) (66) 1,068 4,394 14.6%
Source: AlembicHC
2010e dividend yield
2010e FCF yield 18%
10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0%
Source: TE, Bloomberg, AlembicHC
MTN
Mobinil
Zain
MarocTel
Wataniya
Partner
OmanTel
Qtel
Cellcom
Etisalat
TE
MTN
Wataniya
Mobinil
Qtel
Etisalat
Zain
MarocTel
OmanTel
TE
Cellcom
Partner
16% 14% 12% 10% 8% 6% 4% 2% 0%
Source: TE, Bloomberg, AlembicHC
Extra dividend and a potential upward adjustment to TE’s market cap
TE’s balance sheet is underpinned by a strong cash balance expected to grow at a 2011‐ 14e CAGR of 30% to represent 24% of total assets in 2015e from only 8% in 2010e. We believe the company may decide to increase its dividend distribution as cash balance is expected to reach EGP5.9bn in 2013e (assuming an 85% dividend payout).
TE’s adjusted EV/EBITDA calculation (EGPm) TE EBITDA VFE proportionate EBITDA Total EBITDA TE Cash Cash % of total assets Total EV EV/EBITDA
2010e 4,798 2,581 7,379
2011e 4,697 2,596 7,293
2012e 4,711 2,660 7,371
2013e 4,818 2,747 7,564
2014e 4,960 2,847 7,807
2015e 5,113 2,956 8,068
2011–14e CAGR 0.8% 2.5% 1.4%
2,544 7.7%
3,162 9.4%
4,488 13.1%
5,903 16.8%
7,316 20.3%
8,714 23.5%
30.2%
27,471 3.7
26,655 3.7
25,329 3.4
23,914 3.2
22,501 2.9
21,102 2.6
‐4.9%
Source: AlembicHC
34
Telecom Egypt Telecoms | Egypt 14 December 2010
VFE: From value proposition to subscriber acquisition It’s harder to maintain a leadership position than to get to it While it has become increasingly more difficult to acquire subscribers in light of high real penetration and impeding regulatory pressures, competition in the mobile market has become more rigorous and is expected to last until 2011e (see Egyptian Telecom Sector section). VFE changed its strategy from offering value‐added propositions to being the market leader, which it was able to achieve last quarter with a slightly bigger subscriber base than Mobinil’s. Now that Mobinil is back in the game, competition will be even higher. VFE’s subscriber base is expected to grow at the same rate as Mobinil’s (2011–14e CAGR of 4.4%) as we believe that since both companies have the same utilization rate, operational efficiency will determine success. VFE market share, subscriber growth (quarterly trends) VFE market share, subscriber growth (annual trends) 100%
50%
90% 80%
40%
70% 60%
30%
50% 40%
20%
30% 20%
10%
10% 0%
0% 2Q09a
3Q09a
Market share
Source: VG, TE, AlembicHC
4Q09a
1Q10a
Utilization rate
2Q10a
3Q10a
4Q10e
2008a
Subs growthy‐o‐y
2009a
2010e
2011e
Market share
2012e
2013e
2014e
Subs growth y‐o‐y
Source: VG, TE, AlembicHC
ARPU dilution expected at a 2011–14e CAGR of 4.9% While the new strategy ensures subscriber growth through stable strong net adds, in our view, it also causes ARPU dilution, averaging 23% each quarter over the last year. VFE’s effective price per minute decreased 30% y‐o‐y in 1Q10, 37% in 2Q10, and 33% in 3Q10, up from an average 18% y‐o‐y over 2009. The aggressive cuts stimulated usage successfully, which increased 9.3% in 1Q10, 22% in 2Q10, and 17% in 3Q10, yielding 3 quarters of positive elasticity after 4 consecutive quarters of negative elasticity. VFE is expected to maintain its strategy of cutting prices in an attempt to capture new subscribers throughout 2011e.
35
Telecom Egypt Telecoms | Egypt 14 December 2010
VFE usage vs effective price per minute
VFE net adds vs % change in ARPU
30%
1Q09a 2Q09a 3Q09a 4Q09a 1Q10a 2Q10a 3Q10a 4Q10e
0%
20%
2,000 1,800
‐5%
Sep‐10
Jun‐10
Mar‐10
Dec‐09
Sep‐09
Jun‐09
‐10%
Mar‐09
Dec‐08
0%
‐20% ‐30%
1,600 1,400
‐10%
1,200 ‐15% ‐20%
1,000 800
‐19%
‐20% ‐22%
‐25%
‐40%
‐23% ‐25%
‐24%
‐22% ‐22%
400 200
‐30% % Change in effective price/ min y‐o‐y
0
% Change in usage y‐o‐y
Source: VG, AlembicHC
Net adds
600
Net adds ('000)
% change in ARPU
10%
%∆ in ARPU
Source: VG, AlembicHC
Revenue to grow at a 2011–14e CAGR of 2.3% and EBITDA 2.5% VFE’s strategy seems to be more successful than Mobinil’s in maintaining the top line, which has declined 0.4% y‐o‐y in 9M10 as opposed to Mobinil’s, which dropped 2.4% y‐o‐ y. We expect revenue to remain flat in 2010e since we are assuming lower ARPU dilution, but still robust as a result of continuing competition.
Forecasted revenue, EBITDA, and EBITDA margin (EGPm) 14,000
13,248
12,916
12,668
12,496
12,112
80% 70%
12,000
60%
10,000 8,000
50%
47%
47%
46%
2,000
46%
5,742
6,000 4,000
47%
5,774
47% 5,917
48%
47% 6,110
6,334
50% 40% 30%
1,443
3,214
3,068
2,952
2,878
1,380
1,434
20% 1,485
10%
0
0% 1Q10a
2Q10a
3Q10e
4Q10e Revenue
2010e EBITDA
2011e
2012e
2013e
2014e
EBITDA margin
Source: VG, TE, AlembicHC
36
Telecom Egypt Telecoms | Egypt 14 December 2010
Revenue and EBITDA growth 20% 15% 2011–14e CAGR for revenue of 2.3% and EBITDA of 2.5%
10% 5% 0% ‐5% ‐10% 1Q09a
2Q09a
3Q09a
4Q09a
1Q10a
2Q10a
3Q10a
4Q10e
2008a
Revenue growth y‐o‐y
2009a
2010e
2011e
2012e
2013e
2014e
2015e
EBITDA growth y‐o‐y
Source: Vodafone Group, TE, AlembicHC
Strong dividend yield all the more positive for TE VFE paid dividends in July of EGP12.5/share (amounting to total dividends of EGP3.0bn) after it withheld them in 2009 as a result of the interconnect dispute. We expect its dividend yield to continue to be strong going forward. VFE revised estimates (EGPm)
New 2010e
Old 2010e
∆ Dev.
New 2011e
Old 2011e
∆ Dev.
30,734 7,409 37.3 12,112 5,742 47.4% 2,554 10.6 ‐18.6% 6.1
26,336 3,011 41.5 12,618 6,057 48.0% 3,533 14.7 12.6% 5.1
16.7% 146.0% ‐10.1% ‐4.0% ‐5.2% ‐27.7% ‐27.7%
33,742 3,009 32.3 12,496 5,774 46.2% 2,562 10.7 0.3% 6.0
28,925 2,589 41.5 13,765 6,566 47.7% 3,441 14.3 ‐2.6% 4.7
16.7% 16.2% ‐22.2% ‐9.2% ‐12.1% ‐25.6% ‐25.6%
Subscribers ('000) Net adds ('000) ARPU (EGP) Revenue EBITDA EBITDA margin Net income EPS (EGP) EPS growth (1) EV/EBITDA (x)
Source: AlembicHC Note: (1) Based on our estimated equity value for VFE of EGP32.3bn
VFE FCF calculation (EGPm) EBITDA CAPEX Interest expense Taxes Change in WC FCF FCF yield(1)
2009a 6,028 (2,199) (90) (600) (240) 2,900 9.0%
2010e 5,742 (2,127) (136) (606) (242) 2,631 8.1%
2011e 5,774 (1,999) (275) (625) (250) 2,625 8.1%
2012e 5,917 (1,774) (275) (633) (253) 2,982 9.2%
2013e 6,110 (1,679) (275) (646) (258) 3,252 10.1%
2014e 6,334 (1,656) (275) (662) (265) 3,475 10.8%
Source: TE, AlembicHC Note: (1) Based on our estimated equity value for VFE of EGP32.3bn
37
Telecom Egypt Telecoms | Egypt 14 December 2010
Our valuation Revised estimates Our updated TP of EGP21.1 per share reflects a more optimistic view of the fixed‐line business and a bearish outlook on the current situation of the mobile market in Egypt and its implications on VFE and TE’s voice segment. We revised our revenue and EBITDA estimates for the fixed‐line business upwards to reflect the strong results we saw in the last 2 quarters: (1) revenue recognition of the cable business along with updated guidance for that segment from management, (2) a new method for calculating EBITDA applied by management, and (3) improved performance starting 2012e when we expect stability to kick in the mobile market. We have not included cable system EBITDA in calculating terminal value since it is comprised of 2 segments: asset sale, which we regard as a nonrecurring item, and capacity sale, which gives the customer the right of use for 15 years while all revenue is recorded upfront. On the other hand, we lowered our revenue and EBITDA estimates for VFE as the market is still seeing intense competition, which is expected to continue diluting ARPU and compressing margins until 2011e. Changes to WACC We lowered VFE’s value 14% to EGP8.5 per TE share from EGP9.9 per TE share as we cut our estimates in line with our bearish view of the market and reflecting margin compression. We lowered our WACC for TE’s fixed‐line business to 13.4% from 14.6% as a result of lowering our beta to 0.81 from 1.0 to reflect the defensive nature of the stock. The combined effect of all our changes yielded a slightly lower TP of EGP21.1/share. TE trades at a discount to peers on PE and EV/EBITDA for 2011e TE trades at a 2011e PE of 9.8x and an EV/EBITDA multiple of 3.7x (adjusted to TE’s stake in VFE), which represents discounts of 10% and 36% to peers on 2011e PE and EV/EBITDA multiples, respectively.
2011f PE vs EPS 2011–13e CAGR
2011f EV/EBITDA vs EBITDA 2011–13e CAGR
20
10
18
Bharti
PER 2011e
Relaince 16 14 Cellcom 12 Partner
‐10%
Mobinil
‐5%
Turkcell Wataniya Etisalat 10 TE Omantel 8 Qtel Batelco 6 0%
5%
10%
MTN
EV/EBITDA 2011e
Zain
9 Reliance
8
Maroc Tel Turkcell
7 Partner Cellcom
6
4
Wataniya Qtel
Mobinil
5 Omantel
Etisalat
MTN
TE
3 15%
20%
‐4%
EPS 2011‐2013e CAGR
‐2%
0%
2%
4%
6%
8%
10%
EBITDA 2011‐2013e CAGR
Source: TE, Bloomberg, AlembicHC
Source: TE, Bloomberg, AlembicHC
38
Telecom Egypt Telecoms | Egypt 14 December 2010
Financial statements and ratios (EGPm)
2009a
Income statement Retail revenue Wholesale revenue Total revenue Revenue growth EBITDA EBITDA margin D&A Operating profit OPM Investment income Net interest PBT Taxes Minority interest Net profit NPM EPS Growth in EPS DPS Dividend payout Dividend yield FCF yield Balance sheet Intangible assets Tangible assets Investments Total fixed assets Total current assets Total current liabilities Total LT liabilities Minority interest Shareholder equity Cash flow statement Cash flow from operations Dividends received Interest Taxes CAPEX and investments Dividends paid Net cash flow pre‐financing Liquid resources Financing Change in cash Key ratios Net debt/EBITDA (x) CAPEX to sales
Source: TE, AlembicHC
2010e
5,764 4,197 9,960 ‐1.5% 4,925 49.4% (2,642) 2,396 24.1% 1,309 11 3,370 (453) (5) 2,911 29.2% 1.71 4.0% 1.30 76.2% 7.4% 14.9% 128 17,036 7,644 24,808 7,222 3,930 1,586 41 26,474 4,986 703 (209) (532) (922) (2,219) 1,720 157 (1,736) 141 (0.2) 9.8%
2011e
5,255 4,875 10,130 1.7% 4,798 47.4% (2,483) 2,492 24.6% 1,226 206 3,562 (524) (2) 3,036 30.0% 1.78 4.3% 1.42 80.0% 8.1% 16.6% 110 15,691 7,576 23,377 9,602 3,864 1,361 20 27,734 5,060 1,354 8 (481) (992) (2,045) 2,590 (2,325) (97) 168 (0.3) 9.8%
2012e
5,097 4,876 9,973 ‐1.5% 4,697 47.1% (2,391) 2,481 24.9% 1,151 206 3,604 (539) (2) 3,062 30.7% 1.79 0.8% 1.52 85.0% 8.7% 11.2% 585 14,970 7,748 23,304 10,173 4,038 1,223 22 28,194 4,730 979 206 (524) (2,146) (2,429) 816 0 (198) 618 (0.5) 16.0%
2013e
5,146 4,879 10,025 0.5% 4,711 47.0% (2,330) 2,557 25.5% 1,184 271 3,777 (570) (3) 3,204 32.0% 1.88 4.6% 1.60 85.0% 9.1% 13.3% 510 14,219 7,926 22,655 11,514 4,188 1,283 25 28,674 4,694 1,007 271 (539) (1,504) (2,603) 1,326 0 0 1,326 (0.8) 15.0%
2014e
5,324 4,861 10,185 1.6% 4,818 47.3% (2,259) 2,733 26.8% 1,225 397 4,120 (637) (3) 3,480 34.2% 2.04 8.6% 1.73 85.0% 9.9% 13.9% 435 13,532 8,110 22,077 12,977 4,487 1,343 28 29,196 4,768 1,041 397 (570) (1,497) (2,723) 1,415 0 0 1,415 (1.0) 14.7%
2015e
5,574 4,827 10,401 2.1% 4,960 47.7% (2,194) 2,940 28.3% 1,256 532 4,493 (712) (4) 3,778 36.3% 2.21 8.5% 1.88 85.0% 10.7% 14.6% 360 12,900 8,298 21,558 14,455 4,816 1,403 31 29,763 4,896 1,068 532 (637) (1,487) (2,958) 1,413 0 0 1,413 (1.3) 14.3%
5,845 4,773 10,617 2.1% 5,113 48.2% (2,134) 3,154 29.7% 1,300 666 4,884 (789) (4) 4,091 38.5% 2.40 8.3% 2.04 85.0% 11.6% 15.4% 285 12,338 8,493 21,116 15,919 5,160 1,463 35 30,377 5,048 1,105 666 (712) (1,497) (3,211) 1,399 0 0 1,399 (1.5) 14.1%
39
Sector note Telecoms | Egypt 14 December 2010
Rating Scale Recommendation Overweight Neutral Underweight
Potential Return Greater than 20% 0% to 20% Less than 0%
Disclaimer This document was issued by HC Brokerage, which is an affiliate of HC Securities and Investments (henceforth referred to as “HC”) – a fully fledged investment bank providing investment banking, asset management, securities brokerage, research, and custody services – and Alembic Global Advisors, which is registered with US‐based broker dealer Pulse Trading Inc. (collectively the “Firms”). The information used to produce this document is based on sources that the Firms believe to be reliable and accurate. This information has not been independently verified and may be condensed or incomplete. The Firms do not make any guarantee, representation, or warranty and accept no responsibility or liability to the accuracy and completeness of such information. Expression of opinion contained herein is based on certain assumptions and with the use of specific financial techniques that reflect the personal opinion of the authors of the commentary and is subject to change without notice. The information in these materials reflects the Firms equity rating on a particular stock. The Firms, their affiliates, and/or their employees may publish or otherwise express other viewpoints or trading strategies that may conflict with the views included in this report. Please be aware that the Firms and/or their affiliates and the investment funds and managed accounts they manage may take positions contrary to the included equity rating. This material is for informational purposes only and is not an offer to sell or the solicitation of an offer to buy. Ratings and general guidance are not personal recommendations for any particular investor or client and do not take into account the financial, investment, or other objectives or needs of, and may not be suitable for any particular investor or client. Investors and clients should consider this only a single factor in making their investment decision while taking into account the current market environment. Foreign currency‐denominated securities are subject to fluctuations in exchange rates that could have an adverse effect on the value or price of, or income derived from, the investment. Investors in securities such as ADRs, the values of which are influenced by foreign currencies, effectively assume currency risk. Neither HC nor any officer or employee of HC accepts liability for any direct, indirect, or consequential damages or losses arising from any use of this report or its contents. Disclosures We, Karim Khadr and Sarah Shabayek, certify that the views expressed in this document accurately reflect our personal views about the subject securities and companies. We also certify that we do not hold a beneficial interest in the securities traded. The Firms are not a market maker in the securities of the subject company. The Firms, their affiliates, and/or directors and employees may own or have positions in and effect transactions of companies mentioned in this document. The firms and their affiliates may also seek to perform or have performed investment‐banking services for companies mentioned in this memorandum. Copyright No part or excerpt of its content may be redistributed, reproduced, or conveyed in any form, written or oral, to any third party without prior written consent of the Firms. The information within this research report must not be disclosed to any other person if and until The Firms have made their information publicly available. Issuer of report: US distributor of report: Pulse Trading HC Brokerage Alembic Global Advisors 2 Liberty Square, 2nd Floor 780 Third Avenue, 8th Floor Building F15‐B224, Smart Village Boston, MA 02109 New York, NY 10017 KM28 Cairo‐Alexandria Desert Road Telephone: +1 212 359 8292 Telephone: +1 617 316 5620 6 October 12577, Egypt Website: www.alembicglobal.com Website: www.pulsetrading.com Telephone: +202 3535 7666 Fax: +202 3535 7665 Website: www.hc‐si.com
40
Sector note Telecoms | Egypt 14 December 2010 Research
research@hc‐si.com
Karim Khadr Sarah Shabayek Hassan Ahmed
Head of Research (ME)/Telecoms Telecoms Head of Research (US)/Chemicals
karim.khadr@hc‐si.com sarah.shabayek@hc‐si.com
[email protected]
+971 4 293 5381 +971 4 293 5389 +1 212 359 8291
Registered with US‐based broker dealer Pulse Trading Inc.
Lovetesh Singh Jaap Meijer, CFA Janany Vamadeva Majed Azzam Ankur Khetawat Nermeen Abdel Gawad Sapna Sharma Hatem Alaa, CFA Mennatallah El Hefnawy Mai Nehad Tudor Allin‐Khan, CFA Amr Abdel Khalek Rehaam Romero Danielle Bouchard Mohamed El Saiid, MFTA Sameh Khalil, CFTe
Chemicals Financials Financials Real Estate & Construction Real Estate & Construction Real Estate & Construction Real Estate & Construction Diversified Diversified Diversified Chief Economist Economist Editor Editor Head of TA Research Technical Analyst
+971 4 293 5385 +971 4 293 5387 +20 2 3535 7362 +971 4 293 5382 +20 2 3535 7354 +20 2 3535 7360 +20 2 3535 7356 +971 4 293 5386 +20 2 3535 7368 +20 2 3535 7364 +20 2 3535 7370 +20 2 3535 7390 +20 2 3535 7392
salesandtrading@hc‐si.com
Shawkat El‐Maraghy Mohamed Helmy Aboubakr Shaaban Hossam Wahid Hassan Kenawi Nihal Hany Ahmed Nabil
Managing Director Sales Sales Sales Trading Sales Trading Sales Trading Sales Trading
selmaraghy@hc‐si.com mohamed.helmy@hc‐si.com aboubakr.shaaban@hc‐si.com hwahid@hc‐si.com hkenawi@hc‐si.com nhany@hc‐si.com anabil@hc‐si.com
+20 2 3535 7500 +20 2 3535 7502 +20 2 3535 7518 +20 2 3535 7522 +20 2 3535 7528 +20 2 3535 7532 +20 2 3535 7516
Sales and Trading – Dubai, UAE
Hassan Aly Choucri Nadia Kabbani Anne Marie Browne Samer Azzam Mohamed Galal Wael Atta
Managing Director/Sales Trading Sales Sales Sales Sales Trading Sales Trading
Stephen Matthews
+91 9772 755 777 +971 4 293 5383 +971 4 293 5384
melsaiid@hc‐si.com skhalil@hc‐si.com
Sales and Trading – Cairo, Egypt
Sales and Trading – New York, US
lovetesh.singh@hc‐si.com jaap.meijer@hc‐si.com janany.vamadeva@hc‐si.com majed.azzam@hc‐si.com ankur.khetawat@hc‐si.com nermeen.abdelgawad@hc‐si.com sapna.sharma@hc‐si.com hatem.alaa@hc‐si.com menna.elhefnawy@hc‐si.com mai.nehad@hc‐si.com tudor.allin‐khan@hc‐si.com amr.abdelkhalek@hc‐si.com rehaam.romero@hc‐si.com danielle.bouchard@hc‐si.com
Sales
hassan.choucri@hc‐si.com nadia.kabbani@hc‐si.com annemarie.browne@hc‐si.com samer.azzam@hc‐si.com mohammed.galal@af‐hc.com wael.atta@hc‐si.com
+971 4 293 5305 +971 4 293 5365 +971 4 293 5301 +971 4 293 5302 +971 4 293 5309 +971 4 293 5388
[email protected]
+1 212 359 8292
Registered with US‐based broker dealer Pulse Trading Inc.
41