Fibre: a new bottleneck?

Spain was on summer holidays, but on the hot morning of friday 1st August, CMT (the Spanish NRA) ruled out that Telefonica will not be regulated in wholesale fibre market, opening up the launch in September of their first commercial offer. See here.

It was my first summer in Madrid, and I thought it was a joke when one of my colleagues explained to me about the perfect move of the Spanish regulator. In Telecom regulation slang this practice is called "regulatory holidays", and CMT couldn't find a better time for it. But,...

...as the new manager of the Irish Football Team ,Giovanni Trapattoni, once said "don't say cat, if you don't have it in the bag",....

... in the meantime, during the summer, Vodafone and Orange's regulatory experts and lawyers have been called back to work on preparing their positions and propositions to the press and finally the court. They are questioning why Telefonica's investments in new technologies are protected, and why theirs are not.

After critics from the EU commission and a preliminary hold ruled by a national court, (La Audiencia Nacional) Telefonica had to stop its plans to commercialise FTTH in Spain.

Is this decision Bad or Good for the market? Let's try to evaluate the possible impact.

Bad for Innovation in the short term: In general terms we can say that with this move the Audiencia National has postponed FTTH in Spain for at least one year.

Good for service competition: It is certain that giving absolute rights on the fibre network would have reduced competition on the very high speed internet market. It is most unlikely that competitors, like Jazztel, would have been able to provide a comparable offer to Telefonica's.

Bad for infrastructure competition: It is also true that by launching new services competitors would have felt the pressure to combine their resources with other investors, like public utilities for example, and invest in open access models. Moreover, communities can promote open system networks, instead of closed ones.

Neutral for municipalities: It is also true that in some Municipalities the incentives to protect Telefonica (a major contributor to the national welfare) might be very hight. However, in distant locations and autonomous regions local partnerships have more possibilities to build FTTH networks than the Spanish giant.

Good for new entrants and investors: New entrants, such as utility companies that do not suffer from economic downturns, cable companies, that do not have a modern network infrastructure that can compete with FTTH, and financial investors, have the opportunity to profit from a polical/burocratical/legal still-stand and invest in the lucrative areas.

For example, if compared to other European cities Madrid has bad coverage of the ADSL and Fibre access network: there are many areas of the city where the maximum speed offered by Telefonica for DSL is still 3MB!
__________________________________________________

I personally like the French approach, of having only one operator cabling the building, while the others can access on an equally first come first basis of the house infrastructures.

__________________________________________________

Finally, what are the options policy makers have?

- the deep pockets firm takes all model: we let operator X with larger capital build the network and maintain a monopoly for following years. Operator X will invest in the most lucrative areas, and postpone the entrance in rural areas. The entrance of a second large operator is unlikely. A small regional operator, with regional financial aid, might be able to build regional FTTH networks.

- the sharing model: operators that invest in FTTH are obligated to deploy extra fibre to rent to competitors and permit access to the houses and to other firms. It might be possible also to introduce obligatory network coordination. All operators have the same chance of entering the market and investing.

- the utility-type model: network providers build regional networks and offer dark fiber or wholesale services. This model would probably promote service competition in rural areas.

- a combination of the three: there is a first mover (investor) advantage of 2 years, but it has to build a network that supports the sharing model and open access. Moreover, ADSL and VDSL wholesale prices will be massively reduced, to increase competition from alternative operators in areas where FTTH is deployed.

The best model? Drop me a line and we can discuss about it.

The national and European policy on this subject is still wide open. Many European countries these days have policy makers, regulators, stakeholders and investors who are discussing how to bring Europe into the next century of communications services. Interests are very strong for one and the other model, and the threat of another era of high prices/low quality monopoly is back!

-----
The Audiencia Nacional has given right, for the moment, to CMT on 26/09/2008. The case is still to be decided in court, probably before December.
-----

www.regulation.tk

See also my previous posts on NGN.

NGN Regulatory and Commercial Aspects

Commercial aspects of NGN interconnection should be not underestimated when setting wholesale rates.

We know that in wholesale there are two main groups of services that will be interconnected: regulated and not regulated. We can include also two other categories more common in the retail market, services that are perfectly homogeneous and 'status' goods.

What I intend with this?

Perfectly homogeneous a.k.a. commodities: The firm in this case is a passive actor with absolutely no power over the market for its products. This is a key feature of the ‘perfect competition’ model of the market place, in which the individual firm is a price-taker (e.g. voice services).

Premium (Veblen goods): The firm has a power on the market of this product. Elasticity for these products can be positive or neutral.(e.g. Euro2008).

We should consider all these elements when we decide the pricing mechanism we want to choose for our wholesale product.

Analyzing the issue I realized that there are a multitude of pricing mechanisms that are not always compatibles with retail. If we run a compatibility test between pricing and charging principles we see that they are almost always only directly compatible (for example, QoS at wholesale with QoS at retail).

In my study I have identified the implications at retail level of applying indirect relashionships. It might seems absurd but the less destructive methodology remains the per minute charging. I am not saying that it should be always used, but that there should be a direct relationships between wholesale charging and retail if we want to reduce market's distortion.

This would suggest that ….

1- there isn’t a best way to price products where there are a multitude of diverse services, but there’s a multitude of options.
2- Therefore it depends on several environmental factors e.g. product, market and regulation.
3- However combinations of models are not infinite or without constrains.
4- the importance of costs of wholesale products to assure innovation, fair competition and flexibility in the market.

Net Neutrality: Who Pays for it?

In the last two days I had the pleasure to meet and discuss with Martin Cave, Director of the Centre for Management under Regulation, Warwick Business School, Coventry. He came to Switzerland to present his views on the economics of the ICT market at the annual Comdays. A must be event of the Swiss ICT industry.

Martin highlighted a contradiction of the Net Neutrality principle. Capacity doesn't come for free. Someone has to pay for extra capacity used for new services and the investments operators do for providing a good QOS.

I personally agree with the basic concept of any-to-any communication of the bit and net neutrality. I also respect the decision of some governments to block inappropriate content, such as porn and offensive material. I personally disagree that providers can decide what kind of services I - customer - can have (e.g. block my Skype or Zatoo packets).

As Martin said, someone has to pay for the 'neutrality'.

I agree with this principle, let's look together what the options are:

- operators, can recover investments from the monthly broadband subscription or from advertising: a mix of paid and advertised support model might be the solution.

- Operators, can charge a premium for valued content (efficient pricing) to recover these costs.

- Service providers (e.g. google) can support the extra costs with a revenue sharing model.

- Finally, governments, communities and philanthropic associations.

I know, it's not so easy to pick the best option. However, we should be aware that we will be confronted with this dilemma soon.

As already argued in this Blog there is another risk, the formation of new bottlenecks from network / content to distribution. The distributor of the 'information' (everything that can be digitized) might be the next real monopolist. See for example, the rights for sport events, such as Euro08!.

More Info:

6. Biel-Bienne Kommunikationstage, 25. / 26.10.2007, im Kongresshaus, Biel

NGN Interconnection: far from an agreement

There is no agreement or even a consensus on how NGN interconnection in UK will look like. The current debate is so far from reality that many operators (someone says also BT) do not know even what the problems are. Part of the challenges arises from a general lack of experience in interconnection of a so large and extended set of products and services.

Consultants, regulators and academics have different and divergent views on how to set interconnect rates for NGN. Views essentially spread from bill&keep innovative models to old-fashioned per minute or per service charging.

A possible solution might be to charge per 'contended capacity' and per 'quality of service'. However, even here there is no agreement whether voice needs more or less priority.

Some observers argue that voice do not need more quality as it can be accepted and priced at low quality (see Skype). Therefore regulated voice services will be almost free of charge and priced according to capacity. Others instead argue that voice needs a high prioritisation and should cost more than other services. What is clear from the above discussion is that there is still lot to say and argue about this topic in the future.

Interconnection of NGN can remain billed per minute for quite some time during the transitional phase, even if minutes are not a cost driver for NGN. However, prices probably should be adapted to the costs of an efficient operator (e.g. overcapacity should be reduced) in order to avoid irrelevant costs.

Simplicity and continuity in pricing models (for retail customers at least) are and should remain a must in the near future in order to reduce cases of anti-competitive behaviors and the effect of bottlenecks. Operators should look for a solution that combine the need for stable rates and the request of innovation and profitability from shareholders. Two kings for the same throne!