Tuesday, September 14, 2010

How can the quality of your connection service be based on vague "up to" promises?

These Next Generation Networks (NGNs) have compelling performance characteristics in terms of flexible functionality, quality and quantity of service. The underlying costs are attractive when compared to the aggregate benefits that they can deliver. It is clear that the wireless sector, smart sensor-based technologies, Software as a Service and Cloud Computing all depend on ubiquitous fibre grids. The Internet has been a major commercial force for more than five years and is clearly the communications foundation for the future. The core services provided by these NGNs are various types of bandwidth services or as we call them – transport and connectivity services. In the provision of these critical services how can we still be having debates around “net neutrality”, “open access”, “do not compete with your customer”? How can the dominant description of the performance of your connection service be defined by vague theoretical maximums such as “up to 5Mbps” as opposed to the minimum service you will receive?

The answer is that the dominant provider of Internet service is still the incumbent and the incumbent still sees it as an “add on service” not a core service. Can you think of another purchase that you make that is defined as you get “up to xxx”? I can’t. If there was any competition in the Internet transport and connectivity sector there would be actual commitments made to the customers that would differentiate the product or service from the competitor.

The net neutrality debate exists because users of the network see the network owner as having an advantage in the delivery of their content or other services that is inappropriate and needs to be reined in by regulators. The evidence is clear as all one has to do is look at the way the transport and connectivity services are priced and described when they are not bundled with other services. No industry that depends solely on the success of these services would avoid making meaningful commitments in respect of the performance of the service sold. So, from the customer of transport connectivity services perspective it is critical that the owner of the effective monopoly network not be allowed to compete with its customers in the web services sector. This would ensure sustained functional competition would exist in the web services sector.

At this point, I feel compelled to make a couple of rather basic points about business because I am often confronted by leaders in public policy or leaders in the telecom industry saying in one way or another that “telecom is different.” For instance, they can conclude that two vertically integrated competitors is actually competition in telecom industry for backhaul, when they would never come to a comparable conclusion in any other industry.

My first point is that customers will only pay the lesser of what they can afford to pay and what they have to pay based on their choice of alternatives. The second business fundamental point is the strategic imperative for any business to maintain margins is to come up with a service or product that the customer needs but which the competition has barriers to match. Neither of these are new or groundbreaking concepts but legacy telcos have been making extraordinary profits by marrying these two fundamentals while stiff arming regulators and limiting competition at the transport and connectivity layer. Current telecommunications regulations are simply the wrong base from which to start the rate setting framework for NGNs that are focused on end user outcomes.

The regulator can only deliver in the interests of the NGN’s customers after eliminating the fundamental conflict of the network owner competing with its customers in the web services sector. This focuses the rate setting issue on the transport and connectivity rates. Competition will work for price setting for web services. So, what is the appropriate rate setting approach for transportation and connectivity services? Net neutrality and open access both come down to “who pays what price for which transport services.” We have been successfully providing neutral, open access pricing of transport/connectivity services in North America, Europe and Asia. In each case we develop prices within the context of public policy objectives. Sound public policy effectively links any rate setting approaches to desirable public policy outcomes while minimizing undesirable implications. Some of the rate principles that are lively debated are:

• Should costs be allocated to end users based on the specific network elements that they use?
This is the “distance from the global gateway spread over the size of the market” debate. Most governments are enlightened enough to decide that from a public policy point of view this basic communication infrastructure should not cost more to access in a small distant market than in a big market. The minimum is to accomplish this at the “community interconnect” level so that the differences are only related to local access costs within the community as opposed to where the community is relative to the global gateway.

• Should big retailers get wholesale volume discounts over small retailers?
Since big retailers do not cost any less to provide the same services to than small retailers, there should not be any volume discount because of size. From a competition point of view why would one make small niche or local retailers less competitive than the market muscle players? They should compete on a level playing field in respect of access costs to basic infrastructure.

• Should the NGN owners be able to choose to charge different rates based on who the customer is?
The opportunity to differentiate pricing to different Retail Service Providers (RSPs) based on who they are (as opposed to what they are purchasing) appears to be the essence of the net neutrality debate in the USA. Obviously if the owner of the network can choose to charge users more, directly or indirectly simply based on who they are, then the premium is like a royalty which would only be sustainable if that party and their customers had no alternative. This would not fit any effective market-based competitive theory which should be the foundation of regulation focused on creating a competitive market for the benefit of end users. On the other hand, government policy could determine that the public interest is delivered by having a special reduced access rate for targeting a low income group and have the shortfall made up by higher rates for all others. This would be workable if the special interest group was small relative to the others and the rates were still accretive.

• Should everyone pay the same price to have access to the Internet?
While it is often argued by net neutrality supporters that somehow everyone should have access to the World Wide Web on the same basis irrespective of the demands they put on the Internet, the use of the Internet has outgrown such simple concepts. End users have a wide divergence in the demands that they put on the network. These diverging demands drive markedly different costs. In thinking through rate structures, one must weigh the value of simplicity with the value of differentiation in both cases from the end users’ perspective. While no one is able to forecast the future in respect of how NGNs will be used, there are some trends that already demonstrate that different rates for different services are in the end users’ interests as long as everyone has equal access to choose which service they need. Appropriate service differentiators include:
  1. Differentiation of the quality of service in the context of the reliability of real time network performance. The network delivery costs are much less if an end user does not care whether they face a 1 - 5 second delay instead of needing a 5 - 20 millisecond delay. Appropriate rate structuring would provide different rates to enable end users to choose their consumption approach and the network to optimize the effective use of capacity, thereby delivering more effective capacity for the same cost. This approach is differentiation by offering various levels of quality of bandwidth (i.e. standard, interactive, real time, etc.)
  2. Differentiation of rates for throughput as between peak time demand vs. the “graveyard shift.” Network capacity is driven primarily by peak time demand but the engineering approach creates capacity that is essentially 24 hour in nature. The result is that some differentiation in rates between peak and off peak would result in fewer costs to customers that given some commercial incentive would shift their demand from peak to off peak. This grows the capacity of the network at peak time at no cost overall.
  3. Differentiation of rates based on physical redundancy of network paths. Some services are seen as essential by end users such that they need smart switched redundant paths so that they are not exposed to a fibre cut on one path. Most customers are not interested in incurring the incremental cost for the marginal improvement in service reliability.
  4. Differentiation of rates by functionality being used. The ICT sector has proven that where consumers are mass adopters the economies of scale of both software and hardware are powerful. These economies of scale only happen with standards based products and services. Video products and services standards will evolve for this reason. As they evolve, NGNs should offer video transport services that enable the delivery of standards based video services. Customers would then be able to purchase the video functionality that they need.
There are a number of other public policy rate questions. In my experience, they are all handled well when the party owning the fibre grid is a special purpose entity that is only involved in that business and its revenues are regulated on an approved eligible cost with efficiency incentives basis. Transparency combined with the single purpose gives the regulator the capacity to effectively determine the public interest. The shareholder interests are appropriately covered by the eligibility of recovering cost of investment and earning incentives. The structure focuses both the NGN owner and the regulator on the interests of the end users.

If rates derived from the recovery of efficiently incurred costs are too high to function in the market then the public policy choice can be to subsidize the NGN. In rural economies these subsidies can be focused on strategically targeted one time capital grant funds to interconnect small distant markets to global gateways after which the economies of scale offered by the NGN can fund ongoing operational costs. The regulatory framework ensures that the government and the end user get value for money on a lasting basis.

No comments:

Post a Comment