One of the most important trends we track at Wireless Watch is the changing pattern of capex investment in mobile networks. Early indications are that operators will deploy 5G more gradually, in most cases, than they did 4G, using it to complement LTE rather than replace it, and spreading the spend over more years – at least a decade, according to China Mobile.
They are also looking for a very different cost base, with increasingly commoditized white box hardware, even at the base station end, and a shift towards software and hosted services.
Another expected change is that operators will be far more open than they were in the past to co-investment with non-traditional partners. In the past decade, where regulators have allowed, MNOs have become more willing to share passive infrastructure such as tower, with their rivals, and to lease backhaul links to reduce capex. However, they have been slow to share any active equipment such as base station – a key factor in the slow roll-out of small cells in many markets – and there have been very few instances of sharing networks and investment with non-MNOs.
But as their profits and growth are squeezed in saturated markets, MNOs will have to think creatively about how they can afford to invest in high performance networks to support rising data usage, while much of the revenue generated by those networks is taken by non-MNOs such as over-the-top and webscale providers. Those companies are becoming more interested in deploying their own hardware to ensure good quality services – to date, that activity has mainly centered on WiFi networks or content delivery networks (CDNs). But if MNOs are moving too slowly to support the kind of performance and quality of service companies like Netflix want, those organizations may step in to accelerate progress in enhancing the network.
This is even more true in parts of the Internet of Things (IoT), where industrial players want networks that are optimized for their particular requirements – such as low latency or critical availability – but are not always convinced the MNOs will prioritize those capabilities.
One answer is for industrial players to go beyond their existing activities in cloud or edge infrastructure, and software platforms, and invest in the connectivity they want too. That could create an alternative 5G power base, focused on specialized edge networks, private small cell deployments and shared spectrum technologies. But it could also join forces with the MNOs’ own plans to invest in a platform that suits everybody’s needs.
Most commonly, the co-investment is with a fixed broadband player. For instance, last week, Orange Belgium said it was interested in forming an “industrial partnership” with Nethys and Brutele, the parent companies of Belgian telco Voo, which offers digital TV and broadband services in Wallonia and parts of Brussels,
“Such a partnership would lead to the creation of a convergent national telecom operator with strong roots in Wallonia and the Brussels region, and to the launch of a major multiyear investment plan for very high speed fixed and mobile networks in the Walloon and Brussels region. Furthermore, it would create a healthy, structural competition between three convergent players on the Belgian market, to the benefit of consumers,” read a statement from Orange.
Interest is rising among non-telcos too. It would be a small step for GE to extend its activities, within its Industrial Internet Consortium, to network investment in tandem with its MNO partners. AT&T is a member of the Consortium, and GE also works with Verizon to support its advanced Industrial IoT services, such as proactive maintenance, both internally and for third party customers.
Some utilities have made moves in this direction, such as Enel in Italy, with its alliance with Wind. The most dramatic example of the kind of co-investment model which could shape a very different 5G model of capex and ownership is seen in China.
There, the government had toyed with merging the second and third operators, China Telecom and China Unicom together, partly to address the latter’s growing financial problems. But the final decision was more innovative.
A 35% stake in Unicom was sold to a group of 14 strategic investors, including well-known Chinese Internet names such as Tencent, Baidu, JD.com and Alibaba Group, all of which have increasingly been extending their search, advertising, ecommerce and social media services to mobile platforms and seeking to drive the mobile experience in the same way that Google does in the west. Baidu and Alibaba have even experimented with their own devices and mobile operating systems, but having a close relationship with a carrier could help them to increase their collective influence over smartphone platforms and services, to a greater degree than with a simple MVNO arrangement alone.
This is part of a broader government plan of injecting new growth and innovation into state companies through outside investment and private capital. It was the biggest recent deal under Beijing’s ‘mixed ownership’ reforms, which mark a significant departure from the Communist managed economy of the past, but stop short of full private ownership and accountability.
In a statement, Unicom said the funds from its new investors will be used to upgrade 4G capabilities, develop 5G technologies and trials, and “develop innovative businesses” to enhance its “core competiveness … and speed up its strategic transformation”. It said the investors, which also include industrial groups like Didi Chuxing and Suning Commerce Group, have very complementary businesses to Unicom’s own, and “strong fundamentals”.
It is right to say this. If the new ownership structure is well managed, Unicom could benefit not just from its backers’ money, but their understanding of the web and industrial worlds. That could help accelerate the process of getting new applications and user experiences out to Chinese consumers, giving Unicom the differentiation to make economic sense of its network expansion programs.
It is already clear that the capex reductions of 2016 and early 2017 have come to an end now it has access to new funds. For 2018, Unicom said it plans to increase capex by 19% to as much as CNY50bn ($7.9bn). At its recent quarterly results announcement, the operator said it had installed 110,000 LTE base stations in 2017, making a total of 850,000, and plans to add a similar number this year for “capacity expansion and experience enhancement”. Spending on mobile infrastructure will account for 37% of this capex bill, similar to 2017’s figure of 38%.
The new financial structure looks set to help Unicom compete more effectively with the larger China Mobile and smaller, but more nimble, China Telecom. Not that the capex increase this year is fully making up for the sharp drop in 2016-17, when the budget fell by 41.6% to CNY42.1bn. Like Mobile, Unicom has said it expects to plan its capital investment in 5G differently to that in LTE, and to spread it over a longer period, while for several years, the main spending will still be on enhancing 4G networks with densification, new spectrum, and selective 5G.
More importantly, the new investment structure could help it make a reality of the idea that 5G networks must support a wide range of vertical market, industrial and IoT services, in order to deliver ROI. That is a nice idea, but in many markets, industries like transport, manufacturing and energy complain that the MNOs do not understand their requirements, or do not build networks which are optimal for enterprise use. In future, network slicing should help provide optimal network connectivity for each sector, but in the near term, it is important that MNOs forge close, cooperative links with industry partners.
That could lead to a situation where there is shared investment in 5G networks. Rather than shouldering the entire burden of each new upgrade, the MNO could be the anchor partner. By investing themselves, industry partners would have a better ability to drive the way networks are planned and ensure their needs are met. Then all players could monetize the network in their particular sectors, whether for external services like MNOs, in-sector B2B services (as GE is doing for other manufacturers with its cloud platform), or purely for internal efficiencies.
Unicom could use its new structure to do just that, and start to win back ground lost to its rivals. It said key areas of cooperation will include big data analytics; payment and internet finance; IoT; content aggregation; and cloud computing. It has already been establishing ecommerce operating centers with Tencent and Alibaba, independently of the change of ownership.
Some of the impact of its new alliances are starting to show already. It almost trebled its net profit for 2017, to CNY1.83bn ($290m), and said its objective is to increase pre-tax profit at a compound annual growth rate of 68.7% over the 2017-20 period. Some of that growth, it said on its earnings call, will come from expanded activities with its partners/investors, such as boosting its cloud computing business, run with Alibaba.
Its rivals have not engaged in such radical change, and while they are also starting to set out their plans to invest in LTE expansion and 5G, they are being more cautious in capex terms. China Mobile, one of the world’s most adventurous operators in developing and experimenting with new technologies, nevertheless said – during its results call last month – that it will reduce capex this year. In 2018, it will spend 6.4% less than in 2017, forecasting a budget of CNY166bn, after reducing its bill by 5.2% in 2016-17, to CNY177.5bn in 2017. As a percentage of revenue, capex was 26.6% in 2017, compared to 30% in 2016.
However, China Mobile may be saying it will only invest in 5G at a pace which can be justified by short term returns, but its scale is still awe-inspiring. It said it ended 2017 with 1.87m 4G base stations and an NB-IoT network that already spans 346 cities, and will extend to all cities at the county level and above in 2018.
Meanwhile, China Telecom may be the smallest of the three national operators, but it reported the most positive financials for 2017, and so looks well positioned to narrow the gap with its rivals in the 5G era, as it already did in 4G. It announced a 3.3% increase in net profits in 2017, at CNY18.62bn, beating analysts’ expectations. Revenues rose by 3.9% to CNY366.2bn.
Other operators which are associated with very advanced technology are, however, being very cautious investors this time around. NTT Docomo, set to launch commercial 5G services in time for the Tokyo Olympics in 2020, and famous for its cutting edge approach to deployment, nevertheless says it will continue to reduce capex. Its CTO, Seizo Onoe, has no doubt given major vendors sleepless nights over the past couple of years, as he has repeatedly said that 5G will be deployed at far lower upfront cost than 4G. So despite the high hopes of Ericsson (see separate item), 5G may not deliver the sharp revenue boost they hope for, but is more likely to provide a slower drip of spending over a decade or more – and with some of that money going to non-traditional suppliers, whether start-ups or IT players offering virtualized platforms and commoditized hardware.
At February’s Mobile World Congress, Onoe was scaring the vendors again, saying that one of the “killer services” for 5G might just be the ability to boost data capacity “with no increasing trend in capex.” That isn’t too far from what Vodafone’s CTO, Johan Wibergh, has said in various forums – that 5G has to justify itself on the basis of cost efficiency alone, before new revenue streams are considered. At MWC, he reiterated his theme, saying: “Revenue is not growing and has been very flat or rising just 1%, and so there is no room for increasing costs.”
A presentation by Onoe at an NGMN Alliance event at MWC showed that Docomo’s capex has been in steady decline since the launch of 3G in 2001, when it peaked at almost JPY1 trillion ($9.3bn), while in 2017 it was JPY600bn ($5.6bn), and the operator is not projecting significant increases for 5G.
“In the past capex has been on a downward trend,” said Onoe. “We can continue this trend.” At the same event, US Cellular’s CTO, Michael Irizarry, said 5G “would have to cost no more than 4G and probably less.”
Docomo is supporting some of the open source initiatives which seek to reshape the mobile network cost base – it is a founder member of ORAN, which was formed recently by combining AT&T’s XRAN group and the Cloud-RAN Alliance. The group aims to develop open source software and white box hardware to support a virtualized, highly flexible RAN which would dramatically reduce 5G capex and opex.
Deutsche Telekom and Telstra are also among the ORAN founders, and the former’s group CTO, Bruno Jacobfeuerborn, believes ORAN could help to halve RAN costs, which currently account for about 70% of networks capex. DT is also active in the Facebook-initiated Telecom Infra Project (TIP), which is supporting start-ups and open source projects to develop low cost, virtualized mobile networks.
So, despite a major plan to densify its fiber and radio network, DT expects to boost capex by a relatively small amount this year – spending €12.5bn ($15.4bn) across fixed and mobile, compared with €12.1bn ($14.9bn) in 2017 – but then to reduce its spending again in 2019 and roll out 5G on a gradual basis.
This cautious view of capex investment came despite strong 2017 fiscal results, which saw DT’s revenues rising 2.5% to €74.9bn ($92bn), and net profit up 29.4% to €3.5bn ($4.3bn), though this was largely down to a tax gain in the US. DT is clearly looking to leverage activities like TIP to achieve its ambitious densification plans while still keeping capex stable or even lower. It has an infrastructure subsidiary called Deutsche Funkturm (DFMG), led by Jacobfeuerborn, which will be responsible for 5G cell sites.
The operator has about 28,000 towers across Germany, and has added about 500 a year since 2014, but expects that rate of expansion to rise to about 2,000 towers per year in the “midterm” to support ubiquitous 4G and the start of 5G – all of this without increasing capex dramatically.
Orange has also warned its suppliers that it is supporting alternative vendors which promise lower cost to deploy dense 5G networks, and forecasts that its capex will rise a little in 2018 and 2019, but after that will start to fall from a 2019 peak €7.4bn.
So the message to equipment vendors is clear – they need to adapt to a new world of white boxes and open source software, and enable their customers to roll out dense, ubiquitous 4G and 5G networks while keeping capex stable. This will be challenging, and vendors will need to consider other ways to grow their revenues – by moving up the stack to offer more services and applications; or moving sideways to sell their wares to new customers such as cloud or video providers. Meanwhile, the operators too will need new partners, because while they may be resisting a big bang capex hike like those of 3G and 4G, they will still be investing heavily over the next 10 years in their infrastructure, and as China Unicom has shown, that may be best done in collaboration with the industries that have a real financial interest in shaping 5G.