Excessive access costs

There has been much discussion on Telecom's latest broadband products, with Russell Brown raising the issue of contention rates and Paul Brislen of Computerworld (the magazine not the shop) calling for "boots and all" regulation.

Keith Ng, writing last week suggests that Xtra sells to "suckers-who-are-unaware-that-other-ISPs-exist". This might be partly true, but I get my connectivity from Xtra and I *am* aware of other ISPs. Why do I stay with Xtra? Basically because I sort out IT problems all day for a living - I don't want to do the same when I get home. I could see buying connectivity from a Telecom reseller (there are no competitive non-Telecom resellers, such as Woosh, that can provide services to my Auckland inner-suburb residence) resulting in endless circular arguments about whether a fault was with Telecom or Xtra - each helpdesk naturally blaming the other.

The "market" at present seems to work by Telecom setting their end-user pricing a little above the cheapest competitor. This means that the price is defined by the competition's cost of sales (not by Telecom's).

The main difference between Telecom and the rest is that they own a network worth around NZ$11 billion (at cost) or NZ$4 billion (depreciated) (estimate based on Telecom's 2005 results).
This network is largely paid for by Telecom's voice customers - broadband is an added value extra. Because the other ISPs don't own a network like this, they have to buy service from Telecom or build a very expensive alternative infrastructure. Hence the Telecom cost of sales is much less than the competition, thus enabling Telecom to offer a lousy service at high cost because no-one else can afford to sell anything better.

I don't believe there can really be fixed-line competition in an economy the size of NZ - at least not with Telecom around in its present form. The current solution as advocated by Paul Brislen and others (and favoured by most of the ISPs) seems to be for Telecom to be forced to lease out dark copper (or dark fibre) from premises to exchange, along with rack space in the exchanges for ISP equipment. I can't see this working that well - Telecom would still dominate the resale market and be the monopoly provider of wholesale infrastructure.

I can think of two possible alternatives:

Option One would be to force Telecom to divest it's local network to a mutual body controlled by a consumer trust. This would then lease circuits back to all ISPs and telcos - Telecom would be just another supplier working on leased capacity. Two problems with this: firstly, the network would have to be bought with public money (or confiscated from the shareholders, including the Super Fund); secondly, prices for basic telephone service might end up higher than they are today in order to finance the "artificial market" (resulting in poor old grannies with one phone subsidising wealthy geeks with 16 megabit circuits).

Option Two, which I think better, is to recognise that Telecom is inevitably going to be the monopoly provider in NZ. As such, it needs to be strictly regulated. I think international comparison is the best approach. Telecom should be required to price their services (on a PPP basis) to be in the cheapest quartile of the OECD for each bandwidth segment. At the same time, the "full bandwidth" offering should offer world-class bandwidth (as well as matching worldwide best practice for features such as contention ratios and bandwidth caps). The details should not be for politicians to bother with - there should be a new telecom regulator with the ability to enforce the price/delivery regime.

The latter option would not necessarily be welcomed by the ISPs - prices might fall too quickly for many of them to compete. But the purpose of regulation isn't to help ISPs make money - it's to enable consumers and businesses to buy broadband at a reasonable price.