We reiterate our NEUTRAL view on the telecommunication sector as the on-going spectrum reallocation exercise coupled with the uncertainty of spectrum fees are expected to continue to cap the incumbents’ near-term performance. On top of that, the current price & data quota-centric competition is expected to prolong with the new entrance of Webe in 2H16. Our recent spectrum study suggested that Cellcos’ FY16E core PATAMI could potentially be lowered by 0.7% to 13.0% (on a full-year basis) should we value the 2x10MHz 900MHz band at RM605m and the 2x20MHz 1800MHz at RM1.07b price tag. Valuation-wise, while we make no changes to all our telco companies’ FY16-FY17 earnings estimates, we trimmed Maxis and Digi’s targeted EV/forward EBITDA levels to account for the challenges ahead and shifting our valuation base year to FY17. Telekom Malaysia (TM) and Axiata’s valuations remain unchanged as we had rolled over the valuation base year previously. We reiterate our OUTPERFORM call on TM (TP: RM7.20) while keeping MARKET PERFORM rating on Axiata (TP: RM5.81); Maxis (TP: RM5.90 ↓ from RM6.44 previously) and Digi (TP: RM4.84 ↓ from RM5.05 previously). TM remains our favourite pick for the sector given: (i) less competition in its fixed-line broadband business, and (ii) potential better-than-expected synergies from Webe. OCK, on the other hand, is maintained at OUTPERFORM with an unchanged target price at RM0.95.

Spectrums’ pricing. The uncertainty relating to the quantum of the 900/1800 spectrum fees has been a major overhang to the local telco sector YTD performance. While the authority is set to announce spectrum fees in the mid-year, the quantum of the charges as well as the payment schedule remained vague at this juncture. Having said that, we believe the upcoming spectrum fees may mirror the reserve prices recommended in Thailand and Singapore’s spectrum auctions. Based on our observation, the reserve price for the 900 MHz band ranges between RM0.92m-RM1.05m per MHz per population while the 1800MHz band will cost approximately RM0.84m-RM0.91m. This suggested that the value of Malaysia’s 2x10MHz for the 900MHz band may be worth RM605m while the 2x20MHz 1800Mhz frequency may fetch a price tag of RM1.07b. Meanwhile, should we measure the valuation base on the International Telecommunication Union recommendation (where the auction reserve price for each spectrum is recommended to be equivalent to 70% of the spectrum’s ‘full value’), the full value price tags for the 2x10MHz 900MHz and 2x20Mhz 1800MHz bands could be up to RM865m and RM1.53b, respectively.

Potential financial impact on the new spectrum fee. All the big three mobile incumbents are currently paying a combination of RM70m-RM80m annual spectrum fee for their respective 900MHz and 1800MHz bands (under the current Apparatus Assignment structure). Based on our sensitivity analysis, should the projected spectrums' fee stagger into 5-15 years, we estimated that it would lower Cellcos’ FY16 core PATAMI by c.-0.7% to -13.1% on a full-year basis or -3.0% to - 20.5% if we were to use the ‘full value’ of spectrum price. Meanwhile, Digi will receive the least earnings impact due to its relatively smaller spectrum allocation (Please refer to figure 6 for details).

Landscaping keeps changing. The recent conclusion of the 900MHz and 1800MHz spectrum reallocation plan (where the frequency assignments are expected to be issued in August 2016 for full implementation by 1 July 2017) is expected to keep Cellcos busy in the remaining quarters to re-modernise/re-configure their new networks. On top of that, we also understand that MCMC intends to optimise usage of other relevant spectrum bands such as the 700MHz, 2300MHz, 2600MHz bands and etc., by the 4Q16. All these coupled with the uncertainty of the spectrum fees are expected to cap the telecom sector performance in the remaining quarters.

Lower our expectation on Cellcos. The current price & data quotas-centric competition in the mobile space has yet to see any signs of abating and is likely to continue in the 2H16 following the entrance of Webe. Similar to Celcom, both Maxis and Digi’s communication services are expected to be interrupted temporary when the network re-configuration process takes place (to cater for the new 900MHz and 1800MHz spectrum allocation). All in all, while we are making no changes to our Maxis and Digi’s FY16-FY17 earnings’ estimate, we have lowered our targeted EV/forward EBITDA level after rolling over the valuation base year to FY17. Our new Maxis target price is now RM5.90, based on a lower targeted FY17 EV/forward EBITDA of 12.4x (from RM6.44, 12.9x EV/forward EBITDA previously). Similarly, our new Digi target price is also reduced to RM4.84 (from RM5.05 previously) after lowering the targeted FY17 EV/forward EBITDA to 12.9x (vs. 13.6x previously). Our Axiata target price, however, remains unchanged at RM5.81 (based on FY17 EV/forward EBITDA of 7.1x) as we have earlier factored the above concerns into our consideration.

1QCY16 results’ snapshot. The sector incumbents’ 1QCY16 report cards were generally lackluster, with TM being the only player that delivered solid results despite its Webe continuing to suffer losses at the EBIT level. On the mobile operators front, the sector incumbents’ aggregate service revenue contracted by 4.5% YoY to RM5.2b in 1Q16 (vs. -2.6% YoY in 4Q15) with normalised EBITDA margin (to service revenue) narrowing to 48.8% (vs. 49.6% in the preceding quarter). The tepid service revenue growth was mainly attributed to the heightened price & data quotas-centric competition and the poor Celcom performance (as a result of heightened competition, particularly in the postpaid and overseas foreign workers segments. Meanwhile, the sharp drop in Celcom’s value-added-services revenue (due to a temporary halt following customer complaints on spam charges) also aggravated the deterioration). Moving forward, despite all Cellcos maintaining their FY16 KPIs post the results released, we believe the mobile operators may face some challenging time ahead to achieve the targets. Our mid-cap telco – OCK’s 1Q16 results were aligned with expectations and it is set to complete its Myanmar project by end-CY16. We continue to like OCK for its healthy cash flow on the back of escalating recurring income trend.

Price of the spectrums. The uncertainty relating to the quantum of the 900/1800 spectrum fees has been a major overhang for the local telco sector YTD performance. While the authority is set to release the spectrum fees in the mid-year, the quantum of the charges as well as the payment schedule remains vague at this juncture. Having said that, we believe the upcoming spectrum fees may mirror the spectrum rights auctions in both Thailand and Singapore. Based on our observations, the reserve price for the 900 MHz band ranges between RM0.92m-RM1.05m per MHz per population while the 1800MHz band will cost approximately RM0.84m-RM0.91m per MHz for each population (figure 5).

Both Maxis and Celcom have been awarded 2x10Mhz each in the 900MHz band, suggesting the spectrum fee could come in at c.RM605m as compared to the RM303m range in Digi and UMobile due to their respective lower spectrum allocations. Similarly, on the 1800MHz front, all the big three mobile incumbents have been allocated 2x20MHz each, suggesting that the respective spectrum fee could be RM1.07b.

Having said that, TeleGeography reported earlier that the International Telecommunication Union (ITU) had recommended the reserve auction price for a 900MHz should be equivalent to 70% of the spectrum’s ‘full value’ of THB768.5m per MHz when Thailand’s National Broadcasting & Telecommunications Commission (NBTC) launched the spectrum auction in 2015. Should the same methodology is applied to Malaysia’s spectrums, it suggested that the 900MHz spectrum’s ‘full value’ is worth RM43.2m per MHz per population while the 1800MHz frequency cost RM38.3m per MHz for each population

Financial impact. All the big three mobile incumbents are currently paying a combination of RM70m-RM80m annual license fee (under the current Apparatus Assignment structure which comprised of fixed and variable fees (i.e. network rollout, bandwidth/site utilisation)) for the 900Mhz and 1800MHz spectrum usage. MCMC has earlier highlighted that these two bands of spectrums will be assigned for a fee for a period of 15-year usage. While the new spectrum fee is yet to be determined, MCMC had confirmed earlier that a similar system (to the previous practice for 3G frequencies) will be adopted where payments will be made in phases and allow operators to roll out services without passing through the cost to consumers.

Based on our sensitivity analysis, should the projected spectrums' fee stagger into 5-15 years, we estimated that it would lower Cellcos’ FY16 core PATAMI by c.-0.7% to -13.1%, assuming a full-year impact. Similarly, should the projected spectrum’s fee is based on the ‘full value’ of the spectrums' price, it will dampen Cellcos’ FY16 core PATAMI by -3.0% to - 20.5%, if we were to assume a full-year basis.

A crowded battleground in the mobile space. The entry of the upcoming Webe is expected to post a credible threat to the existing incumbents given its parent – TM’s dominant position in the fixed line and broadband services. While TM is keeping no secrets in positioning its Webe’s services to complement the existing fixed line and HSBB services, it will ultimately post a credible threat to existing cellcos should the group launch comparable mobile packages. The group has set to pre-launch its Webe service in the coming weeks, but the official product launching will only ensue in August and aims to be on par with the mobile incumbents (in terms of the LTE population coverage) by 2019. As of end 1QCY16, Maxis, Celcom and Digi have achieved 74% (or 81% on a comparable basis to peers), 61% and 73% 4G population coverage, respectively.

Meanwhile, the current battleground in the mobile space has yet to see any signs of abating with the incumbents continuing to enrich plans (by giving more Internet quota and bundle services) at low or similar subscription prices. Celcom, on the other hand, has announced to initiate a nationwide 4G network modernisation plan in 3Q16 following the recently concluded 900MHz & 1800MHz spectrum reallocation that is set to be assigned in August 2016 for full implementation by July 2017. During the re-tuning and optimisation spectrum process, Celcom foresees that subscribers may experience weaker connections or service interruptions, and even zero coverage in some instances. Thus, we do not discount that the group may adopt aggressive marketing campaigns to strengthen or regain its position upon the completion. All in all, we believe the current ‘irrational’ price war will likely to continue in 2H16 in view of the heightened competition.

Fund raising capability. On fund raising, Axiata’s latest gearing is close to the optimal capital structure which suggested limited room to gear further moving forward (figure 7). Digi and Maxis, on the other hand, appeared to have additional buffers to raise fund. Nevertheless, should the 900MHz/1800MHz spectrum fee come in way above expectations, we do not discount that operators may need to consider asset rationalisation or lower their respective dividend pay-out to preserve cash. TM, meanwhile, appears to have a better financial position. However, in view of the additional capex required by P1, HSBB2 and SUBB projects, the group’s fund raising ability could be limited, in our view.

Prefer Fixed-Line than Cellcos. While competitions among the mobile incumbents are expected to persist into 2H16, TM being the dominant fixed-line operator, appears to be a safer bet in the local telecommunication sector. Besides UMobile is posing an increasing threat to the big three players, TM-Webe is set to launch its LTE mobile services in mid-2016, which could provide a credible threat to the incumbents in view of the former’s valuable spectrum assets (850Mhz, 2.3Ghz and 2.6Ghz) and potential leverage on TM extensive network. We continue to favour the fixed-line operator TM than the mobile players (given the persistent price war, heightened competition and spectrums reallocation) due to less competition in its fixed-line broadband business as well as it entering the lucrative mobile segment through Webe in the coming months. Having said that, we do not discount that P1 may require a gestation period of a few quarters in view of its higher-thanexpected losses incurred in the previous financial year.

Source: Kenanga Research - 15 Jun 2016