I have already expressed here all the reasons why I think Structural Separation would be a good thing for the UK infrastructure, for competition and for BT itself. If you haven’t read my views on this, you can watch this short speech which neatly summarizes the key arguments. It seems I’m not alone in thinking that this would be a good move since over the week-end a report backed by 121 MPs recommended the same.
BT’s CEO says it would create huge uncertainty in the market. I think that argument is exaggerated provided BT (and especially Openreach) understand the benefits for them. One important benefit could be a promise to revamp the regulatory regime with lighter regulation of BT on the retail layer and long-term regulation of Openreach in exchange for committed FTTH deployment targets.
But there’s one aspect that I don’t hear discussed and that worries me. It seems like the whole debate is focused around the following choice: structural separation or business as usual. I think that’s wrong. I have no way of assessing the likelihood of Ofcom coming on the side of structural separation, and there could be a lot of reasons why they might not want to do that (a few good, most bad). But should they decide not to push for separation, they should offer an alternative plan.
There’s only two ways long term infrastructure gets built in any given market, and I’ve studied my fair share over the years: infrastructure competition or structural separation. If the latter is discarded, the former should be fully operational. The problem is that in Britain it’s not. Sure you have two infrastructure players (Openreach and Virgin) but Virgin is mostly static, upgrading its existing plant just as far as it needs to stay ahead of Openreach in terms of performance. More importantly, everywhere Virgin is not is a monopoly, and that’s roughly 60% of the market.
Some European countries like Lithuania and Portugal have successfully implemented infrastructure competition regimes. They did so in a very simple way (at least conceptually): by opening ducts at regulated rates. In Portugal, PT’s competitor can lease PT’s ducts at very cheap rates, which lowers the cost of fiber deployment dramatically. Similar model in Lithuania where in some areas you have three or four competitors serving the same buildings.
And guess what: in theory this should work in the UK too. BT has an obligation to resell access to its ducts. In practice though they do very little of it. Here’s why: they have no inventory. Most often when a competitor turns to Openreach for access to their ducts from point A to point B, Openreach will respond that they have no idea which ducts run from A to B, how accessible they are or how full, and they don’t know exactly where the cabinets and other access points are located either. And the cost of surveying to find out is on the purchaser. Money and (especially) time required to actually lease the ducts is such that it’s done very sparingly.
Does Openreach genuinely not know ? It’s very possible, but if it is my guess is that it’s an engineered ignorance. And one that could easily be remedied.
A few months ago I met with a Malaysia based company called ZDSL. Amongst many other things, they offer very clever inventory services to solve exactly the type of problem that we’re talking about here. I asked their CEO Peter Macauley for an estimate of how much it would cost to inventory the whole Openreach duct system. He could only give me a ballpark figure, obviously, but says they’re currently working with a national incumbent in Asia of a country half the size of Britain and the value of the project is $500k. Which means that doing it for Britain should probably not be too far off $1m. If that is the cost to build an accurate inventory, it’s no big deal to get it financed. Government could do it. In fact, competitors might even be willing to pitch in provided the information was available to them and managed by a government agency if Openreach is reluctant (or doesn’t play ball.)
All this to say that assuming Ofcom doesn’t fall on the side of separation, it should fall on the side of infrastructure competition instead of the weak current status quo. And forcing the opening of ducts not just on paper but in practice should be a key part of making that happen.
Now if I was BT, I’d be a lot more worried about an effective infrastructure competition regime than I would about structural separation: in a infrastructure competition regime Openreach’s current infrastructure will be quickly overbuilt by superior FTTH in all dense areas. That will force them to invest in FTTH much faster than they currently plan to (if indeed they plan to) and do so facing at least one competitor (and likely more in all profitable areas). Not a good place to be in. If I was a BT shareholder and I understood that that was the choice BT faced, I’d either sell my shares or push for BT to separate.
But maybe that’s juste me.