Time is running out for Telkom - TechCentral

Time is running out for Telkom

Dirk de Vos

Dirk de Vos

All indications are that Telkom is and will remain implacably opposed to the regulations introducing local-loop unbundling (LLU). This is deeply unfortunate — a drawn out legal process will not solve any of Telkom’s longer term problems. If past legal processes of this nature provide any guide, then by the time the matter is finally determined by a judge, the local loop might well have lost relevance and well on its way to becoming a dead asset.

Broadband access is fast developing the characteristics of a commodity. And new wireless technologies — notably next-generation long-term evolution (LTE) systems — will make for an even tougher competitive environment as LTE exceeds the performance of wire-based digital subscriber line services.

Telkom makes essentially two arguments in opposing LLU. Firstly, it points out that unbundling would amount to a subsidy of competitors who do not want to invest in their own infrastructure. Secondly, it argues that it provides the basic fixed services to subscribers at a loss (known as the “access-line deficit”), only made profitable by voice, data and other value- add services on the top of the basic line rental. Competitors, Telkom says, would then “cherry pick” the best high-paying customers, leaving it with the rest.

Both arguments are poor. South Africans, in particular, have been schooled in the difference between formal equality and substantive equality. Affirmative action, after all, requires positive discrimination to rectify previous advantage. Moreover, the economics of building another, parallel and competitor local loop is just not there.

Telkom built its local loop while it was a state-owned monopoly operator. More importantly, the number of staff accounted for in its financials expensed to the local loop is inflated and Telkom should manage its labour costs efficiently.

While shedding jobs should be avoided, this should not hold back the development of fixed broadband services. As we have seen, the liberalisation of telecommunications has led to an explosion of job opportunities in the sector, if not the type of jobs that once existed.

All this calls out for a creative response by Telkom, not a Stalingrad-style defence.

There are several working business models that could be explored to support a different approach. Happily, all this rests on one firm assumption: an explosion of the amount of data traffic. Research shows that for the foreseeable future:

  • Consumer Internet will remain the primary driver of Internet protocol (IP) traffic.
  • Consumer and business traffic will both be dominated by Internet traffic, though business traffic is more evenly distributed across public Internet and managed IP.
  • Consumer traffic accounts for the majority of IP traffic: consumer traffic is 89% of all Internet traffic and 82% of managed IP traffic.

The problem is that residential customers are not yet ready to pay a premium for super-fast broadband Internet access. So, while many look to fibre and 4G wireless technologies to replace the local loop, copper is not being replaced anywhere near projections.

Fibre penetration all the way to businesses in developed markets is still only between 20% and 30% of the total, and the figure for homes is just more than 10%.

Investments in fibre have tended to be in the core network, focused on dealing with exploding backhaul requirements to support mobile data and video applications. The reality is that the majority of homes and businesses will remain connected with plain old copper and will continue to be so for some time to come.

The economics of fibre do not look good. Increased fibre deployment is part of the solution, but this will remain largely about building fibre to the node rather than fibre to premises.

There are several options available to service providers using copper as the final loop. These include deploying very high-speed digital subscriber line (VDSL) technologies with their broadened frequency band (the latest iteration being VDSL2), shortening the copper loops and using multiple copper pairs (known as bonding). Any one of these can considerably increase access network speeds.

A VDSL2 fibre-to-the-node deployment can be almost three times less expensive than a fibre-to-the-home deployment. The cost savings come mainly from reduced civil works costs.

VDSL technologies provide a solution which is much easier to justify financially as the capital expenditure required is only a fraction of what’s needed for home fibre. It also takes the performance of wire-line services well beyond LTE. All these allow for the leveraging the existing copper infrastructure to stave off the huge investment that fibre to the home and fibre to the business would require.

A problem comes in, however, when some subscribers are satisfied with the existing performance of their existing DSL line and do not want to go through the hassle and cost of a buying a new modem. If some modems are not cable of “vectoring”, it reduces the performance of neighbours who have opted to pay for vectoring. Further, the type of unbundling that should be contemplated is important. Vectoring works best when all the lines are under full control of a single service provider. To maintain the benefits of VDSL2 vectoring, the best solution is to have a single operator deploy it from the street cabinet, control all the lines and offer bitstream access to other operators.

Telkom could, conceivably, turn the whole matter of LLU to its advantage and solve some of its over-staffing problems at the same time. It could convert its street cabinets and the copper wire pairs leading into people’s houses into a series of standalone franchise operations. Existing Telkom employees would have the right to buy into some of these franchises. (Another version could be owner-operated delivery service in which SAB-branded beer trucks, independently owned, are assigned specific areas or customers for the delivery of beer.)

The neighbourhoods served by these cabinets would see such franchises in much the same way as they currently arrange private security or how gated communities or sectional title schemes pay levies to procure a range of services. Depending on how the franchises are structured, neighbourhoods could even buy into them and capitalise the investment needed to upgrade the copper wire infrastructure or even purchase new modems on a bulk-buy basis.


Telkom’s fixed-line subscriber base continues to dwindle

Perhaps the franchises could even be sold or freely traded such as what happens with independently run cellphone outlets serving a particular network. Bitstream unbundling would be the requirement, but even here Telkom, with its huge advantages in its existing links from the cabinet back to the exchange, could be the default option.

Telkom has suffered a decline in its fixed-line copper wired subscriber base for a number of years. Bigger businesses are increasingly using exclusively fibre and residential consumers are going the wireless route, even when they have the option of DSL.

Telkom could turn it around and make this infrastructure work for it and its remaining customer base. Time is running out, though. If Telkom mires itself in another protracted series of legal battles over LLU, it will make the path wide open for new LTE services. Once customers are lost, they are likely lost forever. LLU is then of academic interest only.

1 Comment

  1. I am looking to buy a new router and I am wondering if Telkom has actually decided on VDSL2 as their preferred mechanism for their > 20Mbps roll-out?

© 2009 – 2019 NewsCentral Media