Ovum, the analyst and consulting company, has warned governments and regulators against the break-up of incumbents. It argues that in the absence of a proper cost-benefit analysis, such radical restructuring of the telecommunications industry is not justified.
Structural separation would involve dividing the incumbent's local loop network (Loopco) from the rest of its business (Netco) and running it as a regulated, not-for-profit monopoly. Many influential voices within the European telecommunications industry now appear in favour of it.
"Yes, the economies of scale involved in provisioning and maintaining local access networks makes them a natural monopoly," says Julian Hewett, chief analyst at Ovum. "And Loopco is an attractive proposition in theory."
"But governments are not designing a competitive telecommunications industry from scratch. They must thoroughly compare the status quo with various options for a structurally separate local access network and weigh the pros and cons. And as soon as you do this the case for Loopco starts to look feeble," says Hewett.
These are Ovum's fundamental objections to breaking up the incumbents:
Structural separation is no answer to the failure of local loop unbundling (LLU). Less than three percent of DSL services use unbundled local loops in Europe. Ovum research suggests that the main problem is the cost of connecting the unbundled local loop to the competitive carrier's core network, not the unbundling itself.
Competitive carriers currently make a single purchase of the wholesale DSL service from today's incumbent. With a break-up, they would have to buy backhaul from Netco and unbundled local loops from Loopco. Structural separation would therefore make competitive supply of mass broadband more difficult.
Structural separation does not improve competition. It will undoubtedly make access to local loops easier. But Netco will continue to have a substantial competitive advantage over its rivals as its network edge equipment will remain co-located with Loopco equipment.
Defining what Loopco actually is throws up more problems. It is not clear whether Loopco includes all incumbent loops, or just the 'mass market' loop. Does Loopco include the competitive carriers' loops?
Where is the boundary between Loopco and Netco? If Loopco extends too far into the network, technology co-ordination between Loopco and its customers becomes difficult , and innovation is slowed. On the other hand, if Loopco stops close to the edge of the network, then competitive carrier backhaul costs are much higher.
The industry has to face up to the political reality. Governments are committed to rapid broadband rollout which structural separation would almost certainly delay. "The world's best example of broadband rollout - Korea - is based mostly on the incumbent's local loops," says Hewett. "In addition, the process of actually creating Loopco would be hugely complex, time-consuming and costly. What government is going to take these risks?"
"Our view is that governments should try to make the best of the existing infrastructure and create ongoing, tight regulation. They must ensure that incumbents provide an appropriate range of unbundled and wholesale offerings to competitive carriers at reasonable prices."
