In some ways, little has changed in Indonesian telecoms over the past year. The country still has eight operators and average revenue per user (ARPU) is still among the lowest in the world because the market is so crowded. In other ways, everything has changed. Mergers are in the works and the market is poised for consolidation. Once the number of providers is reduced to seven, it is expected that pricing will stabilise, investment will rise and quality will improve, resulting in more usage and better margins. “This is the most bullish period in 10 years,” said Riaz Hyder, a research analyst at Macquarie Securities. “We are seeing evidence of consolidation.”
Key Deal
The XL Axiata and Axis Telekom Indonesia merger was made official in 2014. The former is 66.5% owned by Malaysia’s Axiata Group and under the agreement the firm bought a 95% stake in Axis, which was owned by the Saudi Telecom Company. The deal received bureaucratic support, as the government wanted to reduce the number of players in the market. In July 2013 the Ministry of Communications and Information Technology (MoCI) gave conditional approval, requiring that the new entity relinquished a number of blocks of mobile frequency spectrum. Satisfied that the conditions would be met, the MoCI gave the deal the go ahead in late 2013. After pricing and specifics regarding integration were worked out, the merger was completed in early April 2014. The combined entity is now the second-largest player in Indonesia. XL Axiata has 49m subscribers, while Axis has an estimated 17m. Together, the two will probably have more than Indosat’s 59m subscribers. “Competition will likely ease due to the Axis deal,” said Leonardo Henry Gavaza, a research analyst at Bahana Securities.
Now that the deal is concluded momentum should build, with further transactions helping to bring prices into alignment. Hutchison 3 Indonesia is one of the most aggressive on data pricing, despite its relatively low market share, which was estimated at 7% in early 2013. It keeps the top three from controlling more than 70% of the market, and its prices keep the competitors off balance. There has been speculation over a possible Hutchison 3 acquisition of Indosat, which could give the latter the financial and technological strength needed to compete with Telekomunikasi Selular (Telkomsel), although there was no news on this at time of press.
The market is looking to end a free-for-all that has resulted in low ARPU and underinvestment in critical network infrastructure. It is generally believed that the ideal size of a telecoms market is two or three major operators. “Hutch has been the most aggressive on data pricing,” said Hyder. “Only when it exits will we move to a more stable oligopolistic structure.
Until then pricing will not move up.”
Breaking Away
Along with talk of a turnaround comes an increasing sense that the market is evolving and becoming more segmented, with increased differentiation among players. This is also helping the industry to move beyond the commodity stage, as customers begin to pick and choose and are no longer seeing all carriers as being the same.
Most importantly, Telkomsel – a unit of Telkom Indonesia that is 35% owned by Singapore Telecommunications – is breaking away from the pack. While it has always been well ahead of the crowd, given its strong shareholding and financial position, and superior network, its leadership is now recognised, as much of the rest of the sector struggles under debt or operating losses. The primacy of Telkomsel could help stabilise the market. With a stable market leader, other companies could fall into line and take pricing and services cues from the top company.
Long Time Coming
The light at the end of the tunnel has been a long time coming. Since XL initiated a price war in 2007, when it broke the Rp1/minute barrier, the market has struggled to regain its footing. Telkomsel almost doubled its subscriber numbers over four years – from 65m in 2008 to 125m in 2012 – but ARPU has dropped by more than a third, from Rp59,000 ($5.90) to Rp39,000 ($3.90). At Indosat, ARPU fell from Rp38,639 ($3.90) to Rp27,400 ($2.70) over the same period, as subscriber numbers rose from 37m to 59m. The penetration rate has passed 100%, standing at 117% in early 2013. Even by 2011 there was a feeling that the market was saturated, with little growth possible by adding new call and text subscribers. With price wars the rule and ARPU low, service suffered, with reports of dropped or missed calls and poor signal quality.
The number of telecoms companies has remained high, despite heavy losses for some. Bakrie Telecom is a good example. In 2012 the company lost Rp3.1trn ($310m) on declining revenues. While it recovered somewhat in 2013, it continued losing money: Rp293bn ($29.3m) in the first half of 2013. Axis reported a net loss of Rp5.5trn ($550m) in 2012.
Mobile phone businesses bring prestige to their owners, which may be a contributory factor in some being retained despite losses. Yet there are also practical considerations at work. Merging is not as simple as it may seem, with integration proving to be a problem. The merger of Smart Telecom and Mobile 8, for example, was complicated and costly. According to ZTE, which assisted with the integration, the two companies were working off two very different platforms that offered different features. The challenge was to combine the systems seamlessly into one without loss of functionality, and do so without existing vendor support, as past vendors generally refuse to help once they know that they are being replaced. Mergers between carriers may bring as many expenses as benefits, with the potential for service interruption. “Network integration can be a headache,” said Chandra Pasaribu, head of equity research at Danareksa Sekuritas.
Spectrum
The business of a smaller carrier may not be worth much given the trouble involved in combining entities. It may be more expensive to consume the assets than they are worth. Still, mergers will take place, if not for the networks, customer bases and technology then for the spectrum. Indonesia has a shortage of spectrum and mergers are the best way for a company to increase its service.
Hutchison 3 and Axis are particularly attractive in this respect. In a March 2013 report, CIMB said the two companies had the lowest number of SIM cards per unit of bandwidth, with the former at 1.05m SIM cards per MHz and the latter at 680,000 SIMs per MHz. That compares with 2.68m SIMs per MHz at Telkomsel. While additional spectrum has been auctioned since the report and refarming could boost capacity, spectrum remains at a premium and will be needed for firms seeking to offer stable data products in the future. It is also important because it reduces the number of base transceiver stations (BTS) needed, reducing capital costs. The physical assets may not be of interest to buyers, but the spectrum is. This has resulted in deadlock, whereby sellers believe their business is worth more than what is offered, while buyers are only willing to pay for the spectrum. “Why is consolidation not happening? They want spectrum and nothing else,” said Pasaribu. “With more spectrum you do not need more BTS.”
Auction
New spectrum has been added, but slowly and in a way that leaves the market in need of reorganisation. In late 2012, after four postponements, the MoCI auctioned two blocks in the 2100 MHz frequency, bringing the total at that frequency to 12 blocks. Telkomsel and XL Axiata were announced as the winning bidders in March 2013. While this will help, it does not solve the long-term spectrum problems. The 2100 MHz frequency, for example, is messy despite being relatively new. Five operators have spectrum there, but their channels tend to be separated. For example, Hutch 3’s two channels are split by Telkomsel and Axis channels. Concerns have also been raised about interference from Smart Telecom’s CDMA frequency. It is hoped that spectrum will be rearranged so that all operators at 2100 MHz have contiguous channels. The worry is that rearranging channels will be costly and disruptive.
Between 478 MHz and 806 MHz is used for analogue television, and a dividend of 112 MHz is expected when digital TV arrives in 2015 in the large cities and nationwide in 2018. At 2600 MHz, the spectrum is used by satellite TV and broadband wireless. At 1800 MHz, operators have non-contiguous blocks – Telkomsel has three blocks and Indosat two – and it is not known whether refarming could be done to support newer technologies. At 850 MHz, four operators are providing CDMA services, each with only 5 MHz, too narrow for newer technologies. The resources exist, but they are not properly arranged.
However, changes are afoot. In August 2012 Indosat received permission to use the 900 MHz frequency on a technology-neutral basis. The company had been providing 2G services at 900 MHz, where it has 10 MHz of bandwidth. By October 2013 it was testing 3G services at that frequency in Bukittinggi, West Sumatra, and planning to roll out the service in Jakarta, Depok, Tangerang, Bekasi, West Java and Bali, according to Indonesia Finance Today. Meanwhile, Telkomsel has asked that its 22.5 MHz of bandwidth at 1800 MHz be made technology neutral. The plan is to go straight from 2G to 4G service at that frequency. XL also has bandwidth at 900 MHz. The Axis-XL Axiata merger should change the landscape. The former has bandwidth at 2100 MHz and 1800 MHz and the latter at 2100, 1800 and 900. But the MoCI has said that spectrum is the government’s property and that decisions on usage are for the government to make, not companies.
Data
Development of data will take time and involve breaking key thresholds. The quality of data is currently low and the experience for smartphone users is inconsistent at best, with slow download speeds, low capacity in congested areas and spotty coverage outside city centres. Operators want to get data services to more people, so their fees are low for the relevant services. In Indonesia, a megabyte of mobile broadband costs about 8 cents. In Singapore it is about 15 cents and in Malaysia about 30 cents. The result is a vicious circle. As customers have a less-than-optimal experience, they tend not to use data services much. This makes it more difficult for operators to justify investment in their data networks, which contributes to weak service.
However, analysts say the data market is starting to achieve the critical mass needed to justify significant investment. Venture capital firm Kleiner Perkins Caufield & Byers reports that Indonesia has 36m smartphone subscribers, and that the total number of users is estimated to have grown by 34% over 2012. It is predicted that the tipping point will come when the price of a smartphone drops to $50 a handset. It is now at around $70. “Pricing is quite low, and service is poor. Once the service is good, people will use it more,” Stifanus Sulistyo, a research analyst at Bahana Securities, told OBG.
Consumer behaviour partly explains the lack of demand for fast networks. As the existing connections are limited in terms of bandwidth, users have become accustomed to using applications and services that do not require high data transfer volumes.
The market has evolved around the offerings. Providing more bandwidth may not result in much take-up at this point, as it would require people to dump applications they like and learn new ones. “That is why providers are taking it slowly. People are only using it for chatting and surfing. If we rush into 4G, it will only be used for streaming and social media,” said Aditya Eka Prakasa, a research analyst at Bahana.
Bureaucracy vs Technology
The lack of the development in 4G services is partly a result of bureaucratic indecision. The government has been slow to publish specifications. In 2012 the MoCI said it would take at least two years to get everything settled and standards published. This would leave the country as one of the few in the region without 4G.
The government backed WiMAX in 2011, betting on a system that has been overtaken globally by Long-Term Evolution (LTE). It is now treading carefully while trying to rearrange the spectrum for LTE.
The government would like to go with the 700 MHZ frequency for LTE, as it is far more efficient than 2300 MHz (which was reserved in 2010 for 4G), but as 700 MHz is currently used for television, it could be a number of years before bandwidth is available.
That said, momentum is building. In early 2013, Indosat held LTE trials in Bandung and Surabaya, and XL Axiata and Telkomsel offered LTE services on Bali Island during the APEC summit in October 2013.
Telkomsel also said that it would be conducting LTE trials in Medan and Manado at the end of 2013.
At the same time, the market seems to be maturing, and it may no longer make sense to hold back on faster services requiring large investments. There are signs that a turnaround in pricing is on the horizon due to data demand, and the commercial case is now quite strong. According to analysis by Australian digital media consultancy Venture Consulting, voice ARPU will continue dropping over the next five years, as will SMS ARPU, but the company sees data ARPU firming in 2014 and rising after 2015.
The sense is that the market is transforming from one that is purely about the sale of minutes into one that is more about providing a valuable service.
“What we see now is innovation, not just pricing,” said Norico Gaman, head of research, BNI Securities.
Outlook
The prospects for the sector are the best in years. Mergers, additional spectrum and new technologies are helping to stop the decline in ARPU that defined the market for so long. The vicious circle is now giving way to a virtuous circle, with demand encouraging investment, which should further increase demand. Much remains to be done. The government needs to sort out the issue of spectrum, chart a course for 4G and give the market direction. The operators have to allow mergers to go ahead. Larger companies need to offer a fair price to reduce the number of players and the smaller firms need to sell. Nonetheless, once a few transactions take place, conditions should greatly improve.
Read More from OBG
