Friday, 2 April 2010

Making up a cost

Ofcom, the UK's communications regulator, has decided that mobile phone companies must cut the termination fees they charge each other from 4.3p/minute to 0.5p/minute.

This is the cost paid by, say, Vodafone to T-Mobile when I call a T-Mobile customer using my Vodafone handset.

According to Robert Peston, these fees represent 16% of revenues for British mobile phone companies. Therefore, he says, they will try to make it up by increasing charges in other areas. There are a couple of problems with this view, however.

The first is that for every Vodafone customer calling T-Mobile, there's likely to be a T-Mobile customer calling Vodafone in return. The numbers won't be exactly the same - perhaps T-Mobile customers are more stingy - but a large part of that 16% of revenues is cancelled out by costs that the telcos incur. These costs will disappear at the same time as the revenue.

So instead of charging 15p/minute and paying 4p to another company, then receiving 4p later from the same company, they'll be charging 15p a minute and paying out almost nothing and receiving nothing either. Therefore, there's no revenue loss to "make up".

This view actually is a bit too simple, and there are two caveats. There is some revenue from landline-to-mobile calls which is not cancelled out, because mobile-to-landline calls do not incur the same level of charges. My guess is this would represent perhaps a quarter of that 16%.

The other caveat is that this will change the nature of competition in the industry in the slightly longer run, and this is where revenue may well decline. You may have noticed that you can get a package with (essentially) unlimited calls to UK landlines while having a limited number (say 500) minutes to other mobiles. This kind of option may emerge for calling between mobiles too. The insidious aspect of termination fees is that the fees cannot be negotiated by the person who causes them to be incurred. Vodafone sets a termination fee of 4.3p/minute but it's the T-Mobile user, who can influence no meaningful competitive influence over Vodafone, who has to pay it. This someone-else's-money problem (famous from the US healthcare debate) is certain to make prices higher than they would otherwise be.

The removal of this effect will stimulate greater competition in the market, so it's likely that the revenues of mobile phone companies will decline after all. Standard economic theory predicts that, if competition operates, the fall in price will indeed be around that 16% mark.

But as for making it up. If the mobile companies could be making more revenue on (say) text messages or data, don't you think they'd be doing so already? There's little doubt that they all have hundreds of people working on revenue maximisation and those people won't suddenly start working harder now that revenue is falling in an unrelated area.

The only way in which this revenue can be made up is if the market becomes less competitive in data or texts at the same time as it has become more competitive in voice. Perhaps the companies will arrange this by devising an internal charging mechanism from their "network infrastructure department" to their text or data departments. The infrastructure department could claim that their costs have in effect risen because the voice department can no longer make as large a contribution to the running and rental costs of the infrastructure. This would raise the cost base of sending data or texts, and change the competitive dynamics of that service, increasing its prices.

However, given that the network is almost entirely a sunk cost, this position will be hard to justify for the companies and will be sustainable only if they (implicitly or explicitly) collude to set some kind of infrastructure price.

Perhaps Ofcom can model what would happen if each company spun off its infrastructure assets into a separate entity and allowed the companies providing the handset contracts to competitively divert traffic over each separate set of infrastructure. I can only imagine that competition between the asset owning companies would drive down prices again, countering any attempt to sustain a higher charge for data services.

Ultimately, you only have to look at the profits made by phone networks to see how uncompetitive the market is. It's a bit hard to find detailed figures, but in 2007 Vodafone had a 26.6% profit margin on its UK revenue, and the combined Orange and T-Mobile businesses are expected to make 22% profit next year. (This article says 20% overall, though that is a figure from Deutsche Telecom trying to justify why they should be allowed to merge T-Mobile with Orange).

For a commodity business these figures are quite high. The only way they are likely to be maintained is by reducing competition through mechanisms such as termination fees, price obfuscation and by anchoring users to headline contract prices so they don't notice the cost of texts and other services. If termination fees are removed, it will be a step in the right direction.

Interesting question for extra credit: Can a business which relies on high fixed investment and low variable costs ever generate a long-term return on capital in a competitive market? Imagine all five mobile phone operators have each spent £20 billion building their network. That's now a sunk cost, meaning that - if the market is competitive - it should not be reflected in the prices they charge. What mechanisms, other than an illegal cartel, could the companies use to ensure they can make a long-term return on their investment?

1 comment:

PunditusMaximus said...

The only thing I can see allowing for profits is oligopoly results -- which requires a kind of bubble, where a few companies emerge from the fray and can jack prices up to pay down all those debts they've incurred until now.

Or maybe the government can buy the infrastructure once it's obvious that it's needed, I dunno.