Just back from my trip to Europe where I followed some of the key international telecoms developments. Because of the highly politicised situation around the NBN in Australia it is important to look beyond the sometimes parochially-focussed NBN discussion in our country.
At the joint forum of the Long-Term Infrastructure Investments Association (LTIIA) and the OECD, which looked at Policy Dialogue to Develop Infrastructure as an asset class, telecoms was one of the key elements on their agenda.
They highlighted three key disruptive developments:
- The trend of change from vertically-integrated telco models to those based on structural separation.
- The disruptive effects of fibre optics and 5G.
- The risk of inefficient duplication/multiplication of investments in both fixed and fibre next-generation networks (NGNs) – the move towards a shared telecoms economy.
Regulators in all major markets have long recognised the discriminatory behaviour of telcos operating in vertical markets – but doing something about it has been far more difficult. Nevertheless progress has been made in New Zealand, Singapore, Australia, Italy, Ireland, Sweden, UK and the Czech Republic.
Lessons from them indicate a better allocation of risks among investors, thus yielding a better value generation for their different owners. New Zealand, Sweden and the Czech Republic in particular have been good examples.
This has led to a rapid development of wholesale-only providers who in general are building fibre-only networks aimed at the emerging market of the gigabit society. Australia is the odd one out, where the wholesale-only provider (nbn) is not developing fibre-only networks, and where the incumbent provider Telstra still runs a vertically-integrated operation.
After separation the service-only companies market grows significantly, as these operators gain more flexibility because of reduced regulation. They are able to pursue their objective of vertical integration without regulation, also making it possible to compete better with OTT providers such as Google, Facebook, Amazon and Apple and to move into new e-markets (healthcare, automotive, energy, smart cities, etc).
In structural separated markets, long-term investors have a much better view on the assets they are interested in, and the lower risks involved in these assets, while other investors can concentrate on the shorter service model cycle and the different risks (and potential rewards) linked to those models.
Looking at the 5G developments we see that many more mobile stations (antennas) will be needed –one in almost every street while at the same time – in order to provide the capacity needed for the massive increase in traffic – they will require fibre connections to these stations. With a structurally separate model, where wholesale-only operators are focused on fibre deployments, a much more efficient overall infrastructure business model becomes achievable.
In the emerging sharing economy we see that the largest e-commerce companies have very few infrastructure assets (warehouses, cars, hotels, networks). The same will increasingly apply to the telecoms world. Telcos don’t need to own physical networks, but instead operate their own software-defined networks, buying infrastructure from the most appropriate providers. Their infrastructure requirements might have little to do with speed; they are based on their specific service business model for the markets they operate in. For this they might look at higher levels of efficiency and reliability of the networks they would want to use – other requirements could include: reduced latency, high capacity and security, as well as speed and fibre is the only backbone infrastructure that can deliver this.
The current fibre-based wholesale-only networks are doing brisk business in those countries where such models have been developed. While current national fibre penetration in many of the developed markets is currently between 20 and 25%, the trend is looking towards J-curve growth. We also see a slow trend towards national monopoly-based fibre networks, as duplication does not make economic sense. This is leading to mergers and acquisitions in this market – for example, in the Netherlands – but here as elswhere it is still hampered by a lack of proper structural separation.
Developments in 5G make full mobile infrastructure competition economically unviable. By far the largest part of these ‘mobile’ networks will be fibre-based and duplication/multiplication of infrastructure here is highly inefficient. The availability of wholesale-only fibre optic infrastructure is essential for efficient and effective 5G networks.
Also 5G goes far beyond mobile telephony and mobile broadband. It will also be the core infrastructure for e-health, automated transport/mobility, energy, environment, smart cities and so on. A structurally-separated infrastructure model will see an explosion of innovation and new services from a range of virtual service operators in these fields. Most, if not all, of the new e-services will also require the availability of fibre optic backbones – copper-based networks will simply not be reliable enough.
Many of these new services will require ubiquitous coverage and quality, and competing and interconnected infrastructure operated by different providers will be highly inefficient and unsustainable for efficient and effective 5G networks. Such complicated models will simply not be able to deliver the performance and the levels of low latency that is required for 5G applications.
A wholesale-only infrastructure will make it far easier for service providers to build the right products and services that will be needed for the many different 5G applications required by the many different vertical markets.
Italy is one of the first countries where mobile operators are looking at sharing fibre infrastructure while still using their own spectrum. They have come to the conclusion that it doesn’t make sense for each operator to deploy its own network to all households in all regions, given the risk of their service not being taken-up sufficiently to build a viable business model around it.
After decades of discussion about fibre optic networks, structural separation and sharing infrastructure it is good to see that long-term investors are now supporting these developments. I vividly remember the criticism that I received from financial institutions and from telcos such as Telstra when I started to talk about these developments in my analyses back in 2001/2002.
Australia is well-positioned to tap into these trends. We do have a wholesale-only operator and we do have structural separation regulation; both now need to be adopted and fine-tuned for the requirements of a gigabit society and the applications that can be achieved through the deployment of 5G.
Paul Budde