Pressure is rising on telcos around the world to roll out next-generation access (NGA) for broadband. In some nations, including Hong Kong, Japan, South Korea and Sweden, high-speed broadband access of 100Mbit/s has been available for some time. In many other countries, little has happened. The press tends to pin the blame for this inactivity on incumbent telecoms operators, but there have been good reasons for their inertia.
Analysys Mason has developed business cases that model an incumbent operator undertaking three types of fibre roll-out (VDSL2 (fibre to the cabinet); GPON; and point-to-point (the latter are fibre to the home)) for four types of geographical area (high-rise, urban, suburban and village). Our models examined the costs of rolling out fibre and the revenue likely if this was used to deliver a triple-play offer. The analysis showed that, while NGA roll-outs in high-rise areas are almost certain to be profitable, a viable business case for roll-out in the suburbs is unlikely.
The difficulty with NGA roll-outs in the suburbs is the low housing density: the costs of civil works ruin the business case – dispatching people and trucks to dig up the streets is expensive. In the other types of area, if existing ducts can be used, there is a reasonably good chance that the investment will pay for itself in five years, based on free cash flow – although this is a long time for shareholders to wait. In the suburbs, NGA roll-out is only feasible if every cable can be put into an existing duct.
Availability of existing ducts is not something upon which a telco can rely. Analysys Mason recently undertook a duct survey for Ofcom, in which it was found that there would only be enough space in 78% of the duct ends inspected to add even a single 25mm cable – and this does not necessarily ensure that a new cable could be inserted, because the duct might have collapsed along the route. A 25mm cable would be sufficient for a point-to-point fibre roll-out in the scenarios we examined. A thinner cable could be used in a GPON roll-out and, hence, this method might be able to make use of a higher proportion of ducts than point-to-point. Even the thinner GPON cables are very unlikely to find space in every single duct.
The high cost of roll-out would not be an issue if subscribers were happy to pay substantial sums for their fibre connections. Some operators in Norway have top-end subscribers who pay as much as EUR200 per month. In this situation, the investment will easily pay for itself. However, there are few countries in which operators can rely on earning such large amounts of revenue. Our calculations assumed an ARPU of EUR75–85 per month, a level achieved in some early FTTH roll-outs, but this represents significantly more than many operators are achieving now.
Even where customers are willing to spend that much, telcos must compare projected NGA revenue with revenue from their existing networks. The additional revenue is likely to be at most EUR30 per month, which is not enough to justify FTTH.
The business case is different in areas where there is a cable operator with a triple-play offering based on DOCSIS3.0, which can offer faster broadband and connect two or more TVs in the home – something not currently possible on a copper network. In this case, the telco might fear that it would lose most of its highest-spending customers to the cableco and, consequently, decide to take into account all of the revenue from these customers to justify the new investment. According to our calculations, this approach would allow the incumbent to justify rolling out any of the technologies in a high-rise area, and VDSL in urban areas. However, none of the other scenarios would pay back in five years (see Figure 1).
Figure 1A–C: Cumulative free cash flow for VDSL, GPON and P2P roll-outs [Source: Analysys Mason, 2009]
An exceptionally long-term investor that was willing to wait for ten years for payback based on free cash flow would be willing to support the roll-out of FTTH in all areas. Reggefiber, which is rolling out FTTH in the Netherlands, is able to do this because its majority shareholder is a family-owned construction company that is investing for the next generation and has no external shareholders pressing for short-term value creation.
Such investors are few and far between. If governments want fibre to be connected to every home, some kind of public support will be needed. Many governments – central and local – have already offered public funds to support fibre roll-outs in areas that would otherwise be unprofitable. Without this, fibre will be a long time coming for many citizens.