Incumbent Telecoms
Will Never Learn to 'Share'

European telecommunications are in deep crisis. France Telecom and Deutsche Telekom are crumbling under mountains of debt. Their main shareholders, French and German national governments, are looking for innovative financial engineering solutions. The French Treasury has proposed to create a publicly owned and funded investment vehicle that will take over the government's shareholding of France Telecom and then assume the major part of FT debt.

But clever financial engineering will not suffice to resolve the crisis. Operators, whether incumbents or newcomers, need to fundamentally re-evaluate their strategies and decide which areas and sectors they want to focus on. It is not an accident that the best performing and most highly valued European operator, Vodafone, is the one that chooses to specialize in mobile communications and not offer the full range of telecom services. Likewise, France Telecom should go beyond the restructuring it proposed last week. It could go a long way toward resolving its debt crisis and engage a strategic redeployment by transferring to the new investment vehicle its so-called local loop infrastructure that it is currently supposed to be "sharing" with its competitors.

The regulatory environment also needs a fundamental overhaul. This was the message delivered by a group of European telecom CEOs, including Thierry Breton, to the European Commission late last month. Their most highly publicized complaint concerned the policy for the new generation of mobile communications, 3G, which is often considered as the major cause of the parlous state of telecommunications in Europe.

But 3G is not the only area where the European telecommunications policy proved costly and counterproductive. Local loop unbundling, the core element of European policy of promoting competition in telecommunication services, has not worked either. While in mobile telephony, international communications and Internet services, many entrants achieved significant market share, this was not the case for the local communications. In most countries, with exceptions of few Scandinavian countries, incumbents maintain market shares ranging from 70% to 90%-plus.

There have been efforts to create alternative local infrastructure either through cable or the wireless local loop. But so far, these have proved very costly and, for most part, uneconomical. In view of these results, regulatory authorities sought to force incumbents to share their infrastructure through unbundling. Not surprisingly, incumbents resisted in every possible way, from setting interconnection fees at a very high level to dragging their feet to make space available in their local exchanges for the newcomers.

It should shock no one they were unwilling to facilitate access to their core assets. Thus the local loop unbundling became a lose-lose quagmire. Newcomers could not get access. Incumbents were reluctant to invest in upgrading the infrastructure they would have to share with their competitors. Customers suffered from lack of competition, resulting in local-call prices increasing more than other services, as well as from the slow deployment of new services based on the local loop, such as broadband Internet (DSL).

The time has come for a fundamental reappraisal of local loop unbundling policies. They suffer from a deep conceptual inconsistency. Either the local loop is an area in which competition is viable -- and in this case incumbents should not be subject to heavy constraints -- or it is a natural monopoly and therefore should be treated as a common utility. If the latter is the case then keeping them under the control of incumbent operators can only create a serious conflict of interest.

This inconsistency reflects confusion between competitive aspects of infrastructure and services. Services can be provided in a competitive fashion, while the infrastructure, with its high fixed costs and exclusivity, is often subject to monopolistic forces. As telecommunications have become more sophisticated, the relative importance of service component has increased, making competition possible and desirable. However, this was not necessarily the case for infrastructure, leading to complex regulation that put a heavy burden on incumbents without actually facilitating entry to newcomers.

The inescapable fact is that the local loop infrastructure is a "natural" monopoly, comparable to basic rail and water networks. According to Ken Anderson, editor of Wired magazine, local-loop provision is subject not to the network economics of scale but to the butt-crack economics of ditch digging. Duplicating ditches and local switches is as costly and inefficient as laying new tracks for every train operator.

In recognition of natural monopoly of basic networks in railways and water, most European countries sought to separate the infrastructure from services in these areas. The former is entrusted to a regulated utility, which can be owned by public authority, as is the case in France for railways (RFF) or private shareholders (as is the case in the U.K.). A similar approach is now being adopted across Europe for energy.

So why not apply it to telecommunications and divest local-loop infrastructure from incumbent operators by transferring its ownership and management to independent entities? These entities would then lease access to the infrastructure back to incumbents and upstarts alike, free from the conflicts of interest that currently lead incumbents to obstruct their own competitors.

One argument against it is that it is technically difficult. Yet, why it should be more difficult that current unbundling approaches, which are hindered not by technical feasibility but by inherent conflicts of interest?

Detailed analysis, carried out in the case of U.K. in 2001 by a specialized investment bank, Babcock & Brown, has shown that the separation is technically feasible. Moreover, it could be done without affecting the incumbent operators' relationship with the customer, as they (and their competitors) would continue to handle service applications such as numbering, billing and quality of service management. Thus, the competitive ability of the incumbent would not be impaired while that of new entrants would be enhanced.

Another reason was the resistance of incumbent operators. Thus, BT rejected Babcock & Brown proposal, despite the proposed valuation of €13 billion.

In the current situation, it is not certain that France Telecom and Deutsche Telekom, and more importantly, their main shareholders, should adopt a similar attitude. Their local-loop infrastructures are probably worth well in excess of €15 billion each, according to some investment banks. Furthermore, this money could be raised through the markets, as managing local loop infrastructure appears to be a steady business, with profitability closer to that of energy and water utilities than that of railways.

The French government has taken a first step toward this approach by proposing to create a public establishment. It could move further, by affirming the vocation of the establishment to manage local loop infrastructure and by opening its capital to other shareholders.

These approaches will not succeed unless the European Commission take a close look at its local-loop unbundling approach and, instead of another regulatory initiative, create the broad conditions that facilitate divestments of local loop and more broadly, a sharing of telecommunication infrastructure. The commission should align its telecommunication policy with that applied to other basic infrastructure, such as electricity and natural-gas distribution.

Mr. Goldfinger is a Brussels-based international consultant. He has advised telecom operators and regulatory authorities.

Updated December 9, 2002