In a highly competitive market, such as the telecommarket, the tariff set for a new service will have a huge impact on both market share and revenues. Moreover, taking into account the possible impact of customerperception, it is nearly impossible to raise prices of a service in a later stage, which means that setting your initial price too low might either jeopardize your profits or your market share (in case you do raise the tariff later on). On the other hand, setting the initial tariff too high or even postponing the introduction will reduce your competitive advantages over other service providers, at least in the early stage, and might thus impact your market share. In this paper we show that we can derive important information about the sustainability of a tariff through the use of a bottom-up cost calculation and allocation approach. We will describe such an approach and explain why this method should be used as opposed to other methods. We also show the positive effect that the introduction of a new service might have on the profit-margins of other services and how to quantify this effect using the proposed method. Finally we use the proposed method to determine a good pricing scheme in a realistic case where an iDTV service is introduced in a network mainly carrying broadband internet traffic. In this case study we also show in which way the results will vary with the considered forecast-term and indicate that a long-term calculation will give more reliable results. Introduction The process of setting a tariff for a certain service is a difficult problem, especially when considering the introduction of a new service on the market. The tariff set for the service might have a huge impact on the telecom company, since in the highly competitive telecom-market of today, services typically have a low elasticity. This means you should avoid raising your tariff in the future because it would have a significant negative impact on your market share. As such, setting an optimal initial tariff for a new service is crucial yet complex: if it is set too high, this might result in a lower take-up of market share, while setting it too low might result in a unsustainable situation. Even though it might be a strategic decision to take a small loss in order to gain market share very fast, it is still very important to have an idea of a sustainable tariff for the service. Within the process of pricing the service provider has to make several important decisions. First the service provider can decide which pricing scheme he will use for the considered service. Several pricing schemes and variations are detailed in the literature [1], [2], [3]. While the pricing scheme is very important for the marketing of the new product, setting the actual tariff charged, regardless of the pricing scheme used is more important from the company’s profitability point of view. Making this decision relies on a lot of information in which the company tries to model the reactions of the different players in the field. The following players can be identified (non-exclusive): − The customers will buy the new service or a competing service depending on the tariff proposed. Their buying behavior is usually retrieved through the use of market research. − The competition is selling similar services. Accurate information about the competition is hard to get and the interpretation of this information involves knowledge of their strategic background. − The regulator tries to maximize consumer surplus and might, from this point of view, impose regulations on the considered tariff. − Finally, the company tries to maximize their profits and turnover, especially in the long run. In this light, costs and revenues are important as well as the strategy the company pursues. Before turning to game theory or another method for calculating an optimal solution of the problem, it is possible to define indicative margins on the tariff using information of the (forecasted) cost of this new service. Since all services are provided over the same network, the costs per service of using this network are not easily retrieved. The margins which can be calculated without taking this sharing into account, give only limited information. By using a bottom-up cost-allocation approach more indicative margins can be calculated giving additional information about the sustainability of a tariff. It is also important to calculate these costs on a longer term, since short-term effects could influence the resulting margins. Especially when introducing a new service over the network, initial effects on the customer adoption will have a large influence on the costs calculated, and costs should be calculated once the new service moves to a steady growth. The remainder of this paper is structured as follows. Section II gives a background on cost-calculation and allocation for network-services and shows which pricingmargins can be deduced from these costs. In section III these margins are related to each other in the scope of a realistic use case in where an iDTV service is introduced BroadBand Europe Geneva, Switzerland 11-14 December 2006 ISBN10 : 907654607X ISBN13 : 9789076546070 Paper We2B2 – Casier Page 2 of 7 over an existing network in which a broadband internet access service is already provided. Finally Section IV provides a general conclusion and future work. Calculating pricing-margins A service is profitable when the total revenue of all customers covers all costs incurred for delivering this service to these customers. Since the network-resources are shared amongst different services, these costs are the result of a cost calculation and allocation process. From these costs, pricing margins can be calculated, giving an indication of how low a tariff can be set and whether the tariff set is sustainable. Cost calculation and allocation In a converged network, the available bandwidth is shared amongst all different services over this network and several different services might require a bandwidth increment at the same time. This allows major savings due to economy of scale and scope, but requires a costallocation process to determine the cost attributed to each of the different services. Depending on the different cost bases, different costs per service can be revealed. Dependent on the considered starting point of the network modeling process, two approaches can be followed for allocating costs to the different services: top-down and bottom-up cost modeling. The first approach, the top-down method, starts from the existing network infrastructure. In this case, the actual network dimensioning is a result from fluctuations in historic and current demand. The network is therefore less efficient than a new network. The cost of existing equipment is allocated to the elements needed to deliver the service, through the use of cost drivers [4]. Therefore, an accurate identification of real cost drivers is required. In practice, it might be difficult to select the correct driver, leading to less efficient and less fair allocations. Two important cost bases can be distinguished for the top-down valuation of equipment. Historical Cost Accounting (HCA) uses the asset purchase costs as book value, taking depreciation into account. Since this method counts all historical costs, it does not lead to an optimization of the network-resources and can not be used in case of a new service. Current Cost Accounting (CCA) values assets at the current market price. This cost base represents the replacement cost of an asset. However, as a result of the continuous evolution of technology, it is not always possible to find the same equipment on the market as what has been installed in the network previously. A possible solution to this problem is given by the Modern Equivalent Asset (MEA) cost base, where the costs of equipment is valued using the cost of new technology offering the same (or more) functionality as the one currently installed. The second approach, the bottom-up method, requires as starting point the demand for the services. Both a forecasted demand for new services as for existing services can be used here. The network is dimensioned in such way that it is optimal for the current situation: it can serve all customers with the requested services at the proposed quality of service. Service costs are allocated according to their required network equipment and usage. In the bottom-up method, the company’s properties and goods will be evaluated following the forward looking cost (FLC). When considering a new network this means that only new and efficient technology will be used. When modeling an existing network, on the other hand, it might mean that less expensive technology is used in the study. This implies that the current network must be reconsidered and remodeled. There are three approaches for doing so: the scorched earth (green field), scorched node (path dependent) or incremental node approach. In scorched earth, the network is redesigned with as few constraints as possible: a different number of nodes, a changed topology and other technological solutions can be taken into account. Both other possibilities make a more fair compromise between efficiency of new technologies and networks and the existing network structure. The nodes stay at their original positions. In scorched node all equipment in the node can be changed, whereas in incremental node the existing equipment in the node is kept and expanded [5]. With the introduction of a new service in the network only bottom-up cost-allocation can be used and any existing resources used in the network should at least be allocated using current cost (MEA) or even FLC. With an incremental design of the network for the introduction of a new service in the network and for providing the increasing bandwidth-requirements of the existing services, scorched node or incremental node will reflect best the actual situation.
[1]
Peter C. Fishburn,et al.
Dynamic behavior of differential pricing and quality of service options for the Internet
,
1998,
ICE '98.
[2]
K Stordahl,et al.
Models for forecasting cost evolution of components and technologies
,
2004
.
[3]
David Hausheer,et al.
The Cumulus Pricing model as an adaptive framework for feasible, efficient, and user-friendly tariffing of Internet services
,
2003,
Comput. Networks.
[4]
Andreas P Antonopoulos,et al.
Analysis of the potential benefits of OXC-based intelligent optical networks
,
2002
.
[5]
N. Economides.
The Economics of the Internet
,
2007
.
[6]
Kevin C. Almeroth,et al.
A dynamic pricing scheme for e-content at multiple levels-of-service
,
2004,
Comput. Commun..
[7]
Sofie Verbrugge,et al.
A fair cost allocation scheme for CapEx and OpEx for a network service provider
,
2006
.
[8]
Burkhard Stiller,et al.
Charging for packet-switched network communication - motivation and overview
,
2000,
Comput. Commun..