Monday, 19 August 2013

The price of your broadband (4)

Previous posts (herehere and here) covered how the Commerce Commission sets some wholesale prices that play a big part in the retail price you pay for your current copper-based broadband.
Here's a quick diagram (from page 48 of the government's recent discussion paper on broadband pricing options) setting out what these prices are.

By taking the combined UCLL and UBA package, at the Commission's regulated wholesale prices, a telco gets access to the data stream to and from your computer, and can get into the broadband game.


Apologies for the long lead-up to the nitty gritty, but we're finally there, and here it is. This chart comes from page 50 of the discussion paper.

On the left hand bar, you've got what telcos will pay to get into the broadband business using the flash new high-speed high-capacity optical fibre network that's currently being rolled out up and down the country (what's usually called UFB - ultra fast broadband). This price (which goes up over 2015-2019) was agreed between the government and fibre-laying companies as part of the bidding process when companies tendered for the fibre-laying in different parts of the country. All these prices, by the way, are the cost per customer per month.

On the middle bar, you've got what telcos pay today to get into the copper-based broadband business. The total comes out more expensive than the fibre option. As noted in previous posts, the two components of the total $44.98 cost were set by reference to what sort of costs telcos pay for the same services in comparable countries overseas ("benchmarking").

And on the right hand side, you get to the numbers that have caused a great deal of ruckus and angst in the telco business and telco policy worlds.

These are the prices that the Commerce Commission has proposed for future pricing of access to the copper network. They're not only substantially lower than the current cost of copper access, but are also lower than the access prices for fibre.


As you can see, the thing that's changed is the proposed new price for the UBA service - what a telco pays to use Chorus's electronic gear. And it's changed because the Commission was directed to change from one way of benchmarking, to another (under legislation governing the separation of Chorus and Telecom)*.

The current price was benchmarked by looking at overseas prices that were calculated, at the wholesale level, as a discount off the overseas retail price (a process known, unsurprisingly, as "retail minus").

The proposed new price, on the other hand, was also benchmarked, but this time by looking at overseas wholesale prices that were cost-based - ones that had been set by overseas regulators on what they thought was a fair estimate of the actual costs involved.

Why is the cost-based number ($8.93) so much lower than the "retail minus" number ($21.46)?
Mainly because the "retail minus" approach to price setting has big deficiencies.

Suppose, as a hypothetical example, you've got an inefficient, profiteering incumbent with a monopoly or near-monopoly on a particular service. Let's say this dominant company sells something, at retail level, for $100, when an efficient level of costs would be (say) $50, and the remaining $50 is a combination of the excess costs it runs as a slack monopoly business and the excess profit it makes as a monopoly.

Suppose now, as a regulator, you want to let potential competitors to get access to some of this lazy incumbent's infrastructure. "Retail minus" means the new competitors get access, all right, but at a price which is likely to be high, and potentially way above the real costs of efficiently providing it. "Retail minus" takes the incumbent's (excessively high) retail costs, and subtracts what overseas regulators have decided is a fair or efficient level of marketing costs. Let's say they are $15 dollars. The resulting wholesale price ($85) is therefore very likely to be an excessively high estimate of the true cost of providing a wholesale service. It's still way above the $50 an efficient provider would sell it for.

So there's a lot of sense in the Commission's having to use what looks like a better way of benchmarking.

It's also the case that moving to a cost-based form of benchmarking ought to be relatively straightforward (as benchmarking exercises go) in the case of the UBA service, where the costs are largely concerned with the costs of the electronic gear used to handle the data traffic. That's something where you could expect to get a reasonable sighting shot on things like what the gear costs and how much data it can handle.

The Commission's approach, in principle, made sense. But the lower UBA prices it came up with have set off a firestorm.

You've got Chorus complaining. Its revenue (from charging the higher "retail minus" price) is going to be chopped back, its share price has taken a hit as a result, and, it says, it was relying on that higher "retail minus" revenue to fund its fibre-laying programme.

You've got the government complaining. It's put a lot of money into this new fibre network, when it didn't look as if copper would be an attractive alternative, and now the Commission has gone and made the old-fashioned copper a more competitive alternative,  damaging the attractiveness of fibre to consumers and so damaging the likely rates of fibre take-up.

And you've got consumers, you and me, complaining. Because the government's reaction has been, damned if we're going to let the Commission set copper prices lower than fibre. In fact, we're going to make them set higher copper prices, roughly equal to fibre prices, and if they don't, we'll set them ourselves at the fibre level.

You've had the strange outcome that a government which has asked the Commission to be the guardian of consumers' interests, and to set prices at levels that don't allow incumbents to rip us all off, and at levels that allow more competition, is now insisting on higher prices to users than the Commission would have set. The government says that its proposed approach will actually work out in consumers' longer term interests (for example, by helping the new fibre system arrive faster and offer more), but it does seem an odd way to do it.

If you feel that there is an Alice in Wonderland feel to all of this, and that somewhere along the way someone - or several people, or indeed everyone - has lost the plot, well, I sympathise.
The next post will try to make more sense of why the various protagonists have taken the positions they have, and how things are likely to play out from here.

*Original text amended to make it clear that the benchmarking change was a legislative requirement


  1. The Commission didnt decide that it wanted to change the methodology, Parliament specifically changed the law to require it due to the split up of Telecom and Chorus.

  2. Thanks Owen. I've amended the text to make the reason for the change clear.