Open Access LTE Network: Intriguing but not without Pitfalls

There are arguments with a significant economic and operational basis to justify interest in the idea of national wholesale LTE networks, in the extreme case only one. However, such a network or small number of such networks would necessitate tackling and resolving satisfactorily a number of complex regulatory, policy, and business issues, some of which are reminiscent of the era before market liberalization. The success of such initiatives from the perspective of mobile customers will depend upon the application of intelligent and enforceable incentives and constraints on the behavior of the wholesale and retail players. The best outcome would be realization of the potential benefits of mobile broadband technology deployed in wide channel bandwidths with healthy competition at the services level. The worst case would be a bureaucratically and inefficiently managed wholesale operator, passively or actively colluding with a handful of large services providers, thereby succeeding in stifling innovation at the services layer and leaving customers with no or few competitively differentiated offerings.

The Economic Justification for Open Access, Even Monopoly LTE Network

The idea of an open access or entirely wholesale LTE network is being entertained in a variety of markets from Kenya to the US. This idea is comparable in principle to the structural separation between wholesale and retail that has been instituted in the fixed network arena for BT in the UK and Telecom New Zealand, or perhaps even more closely to proposals for the breakup of major incumbent operators into independent companies for retail and wholesale businesses. Interest in this extreme form of mobile network sharing is being stimulated by the scarcity of spectrum and the notably greater efficiency, i.e. higher capacities of emerging OFDMA-based mobile broadband technologies, of which LTE will be the most widely exploited, when they are deployed in much wider channel widths than previous generations of mobile systems. In other words there is an argument that at the level of mobile network infrastructure there is a natural monopoly or oligopoly in which only a small number of distinct networks can be economically justified. The attractiveness and justification for this finding is reinforced by a focus on ensuring competition at the level of services, which is after all what customers care about more than the details of the platform or platforms on which these services are delivered. In essence a fully open access LTE network is an extreme case of a shared network such as that (the strangely and extremely ambitiously named Everything Everywhere consortium) between Orange and T-Mobile in the UK which includes their 2G and 3G assets. This consortium will presumably include future LTE deployments as well, when frequencies in the 800 MHz “digital dividend” and 2.6 GHz bands are finally awarded.

The benefits in terms of the cost/performance of mobile networks in favor of a monopoly or at least oligopoly in this infrastructure are greater in the emerging LTE than in previous eras of mobile technology, since as noted the performance of LTE deployments is notably improved, and indeed only manifestly superior to HSPA+ when it is deployed in wide channel widths (e.g. 2x20 paired or 20-30 MHz unpaired). In contrast HSPA is only designed for deployment with 5 MHz carriers. Yet the wider the channels or the bandwidths of individual licenses that are allocated in a band, the fewer the number of separate licensees that it can accommodate.

More than LTE Alone

An additional technology development just emerging with substantial further improvements on the horizon that further increases the benefits of an operator having access to large amounts of spectrum (i.e. fewer operators within the spectrum allocated to mobile services) is carrier aggregation, both intra- and inter-band. This technology is just being introduced for HSPA+, and is foreseen for LTE networks and ultimately for combinations of LTE and HSPA+ networks. When – or if, since it faces a number of technical hurdles – carrier aggregation is fully commercialized, it will enable operators to achieve the following benefits:

Maximize the total peak capacity and throughput performance they can deliver by combining the peak capacities and throughput performance available at different frequencies.
Provide a higher and more consistent quality of service to customers thanks to load-balancing across frequencies and systems. A customer encountering congestion in one band and one system can seamlessly access unused capacity available at another frequency or system
Mitigate the relative inefficiencies that are inherent in LTE deployments in some of the narrow non-contiguous channels of 5 MHz or less that many operators have in their spectrum holdings, often with such channels in more than one band.

Thus the combination of LTE and carrier aggregation (the latter also applying to HSPA+), in the context of rapidly rising demands for mobile capacity for broadband services and applications, may create unassailable economic and performance advantages for an operator with access to amounts of spectrum that are substantially larger than its competitors. It may be unreasonably difficult or even impossible for these spectrum-poor competitors to overcome these advantages no matter how superior they are to the spectrum-rich operator in other critical aspects of the business such as customer service and responsiveness. Regulators may therefore try to ensure that any differences in spectrum holdings between mobile competitors are not dangerously or anti-competitively large. However, depending on the number of competitors legally and otherwise able to enter a market perhaps none of them will then acquire access to enough bandwidth to be able to deploy efficient and economical broadband networks and services to their customers. As a result users of mobile services will suffer from lower quality and higher prices than they would and should otherwise enjoy, which would hardly be a desirable outcome in the “public interest”.

Let us suppose for example that an efficient operator needs 150 MHz of spectrum in all to offer broadband services in capacity-limited areas, an amount for example that individual German operators already currently hold after that country’s mega- and multi-band spectrum auction in 2010. The total amount of available spectrum in the 900, 1800, 2100, and 2600 MHz bands, plus the initial European 800 MHz “digital dividend” band is less than 600 MHz. So at the very most it would make no sense with this assumption to have more than four operators in a market (of course the number of services providers using the networks can be larger).

Thus a critical question for both policy makers and regulators, as well as for services providers themselves, is whether and if so how effective competition can be achieved and sustained in the context of an open access LTE network, and/or the growing trend towards network sharing and consolidation. The recent announcement of the proposed acquisition of T-Mobile USA by AT&T presents this dilemma as one to be resolved in the particular (some would say peculiar, others “exceptional” in the positive sense) regulatory, legal and political climate of the US. The resulting controversy or even furore and intense lobbying, which is likely to lead to a decision or finding in 12-18 months, revolves around whether this major acquisition should be approved, or if so under what conditions (e.g. divestment of some assets, coverage obligations, assurance that wholesale and low price retail services will be offered or continued etc.) to ensure that marketplace competition and ultimately consumers’ interests are not harmed as a consequence.

Bases of Competitive Differentiation

The bases on which positive competitive differentiation can be achieved and the scope for the introduction of distinctive innovations have to be clarified for an environment – an open access LTE network - in which possibly all providers will eventually be using the same basic platform on which to build and deliver broadband services. Clues and concerns about whether and how this differentiation may be achieved can be found in the history and experiences of MVNOs (Mobile Virtual Network Operators), which will be the situation of all wholesale customers of an open access LTE network. On a cautionary note, until now there have been many more examples of failures among MVNOs than of obvious successes. However the competitive situation may be sufficiently different that the past need not be prologue to the future. MVNOs have in the past been competing not only against each other, but also against the MNOs (mobile network operators) from whom they acquire capacity, whereas in the open access LTE network concept they would only be competing against each other.

The scope of sharing or common network infrastructure in an open access LTE network may be broader than that of the radio access network (RAN) alone. In principle, which is one of the questions that a decision to deploy an open access LTE network has to address, the wholesale offer can encompass backhaul links, core networks, and even billing and other services in addition to the RAN itself. The wider the extent of sharing, the narrower the scope for differentiation and the more complex or multilateral decisions about investments, upgrades etc. that may affect the financial and operating parameters of all services providers (MVNOs) using the network are bound to become.

The area of core network sharing raises the most potentially troublesome questions about the implications for competitive differentiation. The core network performs several functions that are essential to network performance (e.g. billing) and contains confidential information on an operator’s business, which heighten concerns about the anti-competitive impact of sharing of this type. Fortunately physical separation of switching and service and control functions as in next generation core networks can mitigate these concerns.

Many Possibilities for Competitive Differentiation

There are many possibilities for an MVNO or wholesale customer of an open access LTE network to pursue to establish a viable business model, covering the range from niche to full service, depending on their other assets and capabilities. They may choose a strategy and supporting tactics based on one or several thrusts or foci from among a portfolio (this list is illustrative only and not meant to be exhaustive):

Selected segments of customers, by demographics and/or geography, and/or communities of interest
An overall low cost position
Superior customer service and responsiveness
Value added services for specific vertical markets, such as health care, smart metering
Superior performance in the delivery of valued content
Roaming agreements (national and international)
Innovative pricing
Combinations with other services (based on new operational capabilities as well as simply pricing discounts for “bundles” of services)
Brand recognition
Distribution channels
Mobile device deals

This list (to emphasize it is I am sure far from exhaustive) illustrates that innovators and entrepreneurs, as well as established mobile operators, would have plenty of scope in principle for building and sustaining competitively differentiated positions as wholesale customers of an open access LTE network

Nevertheless the prospect and potential value of an open access LTE network that could be a monopoly or at least one among a small number of modern mobile broadband networks raises fundamental policy and regulatory as well as business issues in a new context, with some “back to the future” features. Key questions include but are not limited to:

Can a wholesale-only operator be profitable, and if not what conditions should be imposed on the relationships between its wholesale and retail operations
On what basis will prices be set if this LTE network is a monopoly, and what incentives will be put in place to stimulate it to operate as efficiently as possible if this basis is cost-plus ( a “back to the future” issue reminiscent of monopoly or utility regulation prior to market liberalization)
What range of services will the “wholesale” network be allowed to offer, given that some of their customers may wish to procure some services from a third party (outsourcing), which could be the underlying wholesale network operator, while others will prefer to operate these same services themselves (such as billing, content delivery optimization)
In practice how can non-discriminatory treatment of all customers of this network regardless of size and other characteristics best be ensured (with respect to pricing, quality of service etc.)
How and by whom will investments in upgrading and expanding the network be decided and prioritized in light of the different needs and circumstances of various customers (in terms of timing, geography, need for more capacity etc.)?

The messages from this brief review of the idea of an open access and possibly monopoly LTE network operating as a wholesale provider are mixed. While such a network may offer a path towards allowing the deployment of LTE technology in the most efficient and economical manner, it would also require tackling a number of difficult regulatory, policy, and operational issues. If these issues are not resolved with the right set of incentives and constraints in place and enforced, then there could be a resurgence of the very same problems such as blocking of innovations and unreasonable pricing that have led to and justified the introduction of market liberalization (competing networks and services providers) during the last 30 years around the world.

View the discussion thread.