Trends in Mobile Telephony Services Tariffing

Trends in Mobile Telephony Services Tariffing

Stefan Zehle
Prodata Partners Ltd., London, UK

Mobile telephony services tariffs and tariff structures are likely to change substantially over the next few years. The differences in mobile tariffs between countries are likely to disappear and tariffs offered by competing operators are likely to become more similar. Currently, mobile telephony services tariffs vary considerably between countries. For example, Finland has one of the lowest mobile tariffs in the world, whereas tariffs in the UK and Germany remain quite high. Many mobile operators have developed tariffs aimed at specific segments, such as the off-peak caller or low user segment. In the future, mobile tariffs must be rebalanced to reflect costs. The drivers that will lead to a change are market forces and regulatory pressure.

Market-driven Changes in Mobile Tariff Structures

Several large-scale market research surveys have been conducted in markets worldwide to ascertain potential mobile telephony penetration. These surveys have consistently shown that the highest penetration rates could not be achieved with a conventional mobile telephone service. A typical survey result shows that 30 percent of non-users say they will become mobile telephony users while an additional 30 percent indicate they will become subscribers to a locally focused fixed-mobile convergence (FMC) service.

The difference between the services is not technology but positioning. The FMC or local mobile service concept highlights the demand for a telephone service that is convenient, accessible and sufficiently affordable to compete with fixed telephony. Convenience and affordability mean that subscribers want to use a mobile telephone in the same way as a cordless telephone. In Europe, some segments of the population are already using mobile telephones as their primary voice telephone. Operators such as Helsinki Telecom have launched local mobile services with tariffs close to fixed-network prices. This is the point where fixed and mobile services converge. However, mobile services cannot realistically be treated as alternatives to fixed networks as long as mobile tariffs for local calls remain at their current level.

FMC services positioning requires that mobile services be tariffed similar to fixed-line services to use a mobile telephone in the same way as a fixed telephone. The emergence of FMC-style services will allow mobile operators to gain a significant slice of the local telephony market. In other words, there is not only a demand pull but also a supply push.

Historically, most new tariffs have focused on reducing the monthly cost of ownership for low volume users. Low user tariffs with a low monthly fixed charge combined with a high per-minute price and prepaid tariffs are designed to attract lower spending segments of the market by reducing the minimum monthly cost of ownership. However, low user tariffs of this type discourage calling because they increase costs dramatically. Fixed-network outbound voice call usage in Europe is usually approximately 400 minutes per line, and in the US it is up to 1000 minutes.

More recently, a trend is emerging that gives mobile users better value for their money in terms of the minutes of use for an agreed-upon bill amount. New 1800 MHz Global System for Mobile communications (GSM) entrants have reduced the cost of calls substantially but offer less coverage. Digital communication system (DCS) 1800 networks in Europe have launched with tariffs at least 18 percent and up to 53 percent below the incumbents. In the US, the new personal communications service (PCS) entrants offered tariffs 10 to 15 percent lower than cellular with 25 percent savings possible for high users. These tariff reductions prompted incumbent cellular operators to reduce their prices. The 180-minute inclusive tariff from French PCS operator Bouygues Telecom was a pioneer of this trend. In 1996, the average GSM customer in France used approximately 90 mobile-originated minutes per month. Bouygues Telecom’s average usage is close to 180 minutes whilst being 53 percent cheaper than the equivalent offer from its competitors.

Other operators have followed this trend. For example, in Ireland, Eircell’s new Eirtime tariffs introduced in November 1997 show how an operator can give customers better value by bundling more minutes into the monthly line rental. Eircell kept all tariffs constant, but increased the amount of bundled minutes by over 50 percent. More recently, Vodafone in the UK also started to bundle minutes into the monthly line rentals rather than reduce tariffs.

These developments are not confined to Europe. PCS operators in the US and Canada are changing pricing for mobile services by trading lower airtime prices for higher usage volume. PCS-bundled per-minute price plans are approximately 20 percent lower than similar analogue and digital cellular services. The average bundled per-minute price for a 60-minute bundled PCS plan is US$0.51, compared to US$0.60 for analogue and digital cellular. These numbers fall to US$0.20 per minute for a 500-minute PCS plan vs. US$0.27 for analogue and US$0.25 for digital cellular. Even more aggressively, Microcell is offering 400 minutes for C$40 (US$29.16) while PrimeCo offers 500 minutes for US$50. Powertel offers a similar per-minute rate in Columbus, Georgia with bundled plans of US$20, $40 and $60 for 200, 400 and 600 minutes, respectively.

In Finland (where mobile penetration is already above 40 percent), Helsinki Telecom launched a local mobile service in late 1997 that offers much lower tariffs for calls and focuses on bundling substantial usage minutes into the monthly line rental. The local mobile service in the Helsinki metropolitan area is aimed purely at local mobility and does not offer roaming capabilities. Tariffs are billed on the fixed-network tariff principle, which includes local, long-distance and international charges. The basic tariff consists of FMk80.30 (US$15.15, including 22 percent value-added tax) per month with FMk0.45 (US$0.085) per minute for local calls regardless of the time of day. Long-distance calls are charged at the same rate as in the fixed network plus the local element (for example, FMk0.45), meaning the tariff is easily understood by customers. Three other options are also available in addition to the basic tariff plan. (The heavy-user tariffs offer even better value.)

A general trend exists in telecoms services pricing to maintain price levels for access or line rental while reducing usage- and distance-related prices. Internet telephony is a case where the cost may not be related to the duration of calls or to the distance between caller and called party. The bundling of minutes into the monthly charge now common in mobile telephony tariffing has the effect of lowering usage-related charges whilst maintaining fixed monthly charges for the dial tone (that is, the facility to make and receive calls).

This trend does not mean that tariffs that focus on usage-related charges will disappear. There will always be segments of the market that prefer variable costs to fixed costs and are prepared to pay a premium for this method (as demonstrated by the phenomenal success of prepaid tariffs in many markets worldwide). It is likely that prepaid tariffs and low user tariffs will continue to be offered, but their relative importance will decrease.

Changes in Mobile Network Cost Structures

Mobile tariffs will decline to the point where they are priced only slightly above fixed services, and mobile communications will be the norm. The conventional view is that mobile operators will not be able to increase revenue per customer for simple voice services. However, additional spending for new applications will counterbalance the decline in average monthly bills. These new applications include data communications and substitutional use and divert spending for fixed-network services. Nevertheless, mobile operators will be forced to deliver evermore usage for less money.

Mobile operators can offer much lower effective per-minute tariffs. As subscriber numbers increase, the marginal cost of network usage declines. The main reason for this decline is that busy-hour traffic (expressed in Erlangs) will not increase as quickly as minutes of use. Busy-hour traffic is a key network design parameter and cost driver. In the early stages of mobile market development, busy-hour traffic accounted for a high percentage of daily calling. Over time, usage has become more distributed. If an existing customer who already uses a mobile telephone in the busy hour makes more use of the telephone as a result of price cuts, these calls are made at times other than the busy hour. The decline in cost is also due to a higher density of customers, leading to better utilisation of and lower prices for infrastructure. As long as an operator makes a net of interconnect margin on mobile-to-fixed calls that is lower than marginal costs, it will be possible to profitably deliver better value for money services.

The design for 1800 or 1900 MHz networks that offer FMC-style services is different from that used by networks with mobile positioning. Mobile positioning implies wide geographic coverage, that is, competition on the basis of breadth of coverage. An FMC positioning implies coverage anywhere in the covered area where subscribers spend 90 percent of their time. The focus is on depth of coverage (inside shopping centres, flats and offices) combined with a high grade of service. A network design that delivers an FMC service is also a high capacity network, meaning a new DCS operator who has built a network to address the FMC market segment in terms of coverage has ample spare capacity. The pricing strategy will be to fill capacity by offering bundled-minute tariffs. As subscriber numbers increase, more revenue becomes available to build an even denser network that provides even better quality and more capacity, thus initiating a virtuous circle.

While it may have been logical for network operators who are capacity constrained to charge high prices for calls to reflect high marginal costs, operators with an FMC focus will seek to cover their fixed costs. Therefore, bundled-minute tariffs are the preferred choice for later market entrants in 1800 or 1900 MHz. There is also a tendency for 1800/1900 MHz operators to compete on the basis of value rather than geographic coverage. The incumbent 800/900 MHz operators have an advantage in terms of geographic coverage, but 1800/1900 MHz operators may have an advantage in terms of delivering capacity in high density urban areas.

Tariff Changes Driven by Interconnect Rates

Currently, mobile tariffs in most countries do not reflect costs. Most operators have low line rental and high per-minute charges, particularly for low user consumer tariffs. This pricing structure is diametrically opposed to fixed-network pricing but is sustainable as long as mobile network termination charges remain high. However, there is an emerging trend towards lower mobile network termination charges. (Regulatory as well as market forces are driving this development.)

The UK regulator Oftel was one of the first regulators in Europe to address the high cost of fixed-to-mobile calls. The regulator came to an initial agreement with the two UK cellular operators Vodafone and Cellnet stating that their termination charges must be reduced by a significant percentage each year to bring them closer to cost. A final agreement has not been reached and Oftel is lobbying for cost-based interconnect pricing not just for fixed but also for mobile operators. The basis for negotiating cost is usually the marginal cost of calling or long-run incremental cost. In February, the European Commission began to evaluate the high cost of calling mobile subscribers. The competition commissioner has requested information on fixed-to-mobile interconnect rates and retail prices from operators.

Good cost data for mobile operators are not readily available because mobile operators have not conducted cost studies in the same way as fixed-network operators. However, assuming that mobile operators do not sell their services below cost, a cost-based interconnect termination charge should be lower than the lowest mobile retail per-minute tariff. In Finland, mobile retail tariffs have declined to US$0.063 per minute. On this basis, an average cost-based mobile interconnect charge could be US$0.04. Assuming that wireless termination charges will decline to approximately US$0.04 and the fixed networks add an origination cost of US$0.05 plus a margin of US$0.01, the cost of calling wireless numbers from fixed lines would be US$0.05 to US$0.06 per minute. This cost is substantially lower than today’s average wireless per-minute call charge and demonstrates that mobile per-minute tariffs in most countries bear little relation to costs.

If interconnect rates are based on costs and tariffs remain well above costs, call charges would be hugely different depending on the direction of the call. This scenario leads to callback behaviour, a form of arbitrage where customers react rationally to lower their costs by exploiting market distortions. The only way to remove this market distortion is to bring retail mobile tariffs closer to interconnect rates so that the cost of calling is the same or similar in either direction. Effectively, this cost structure means termination charges will be the main driver in setting tariffs except where distance-related charges account for a high share of the total cost of the call. This scenario implies that mobile operators will be forced to rebalance their tariffs and increase their monthly line rental charges while simultaneously decreasing their per-minute rates.

By and large, low user and prepaid tariffs operate on the opposite principle, meaning significant call-back activity may result. Given the success of prepaid tariffs in many markets, it is unlikely that mobile operators will want to discontinue them. However, mobile operators may reposition these services as premium rate services, allowing inbound call pricing that is not based on interconnect. Alternatively, they may shorten the inbound call validity period for a minimum recharge value, which effectively assures a minimum consumer cost.

The market trend towards lower per-minute prices for mobile services suggests that mobile operators themselves agree that marginal costs are much lower than current termination charges. Mobile operators who are positioned to deliver FMC-style services also have an interest in reducing their interconnect rates to reduce the cost of calling. High interconnect charges for mobile-terminated calls produce unacceptably high costs to those calling mobile numbers from the fixed network, undermining the FMC positioning since the customer is unlikely to consider full substitution if callers are reluctant to call.

Mobile operators have been favoured by European regulators as a means of introducing competition into the telecoms market. Interconnect regulation and regulation of the unbundling of services have been designed to reduce the dominance of the incumbent fixed-network operator. However, the question of dominance soon may be asked the other way around. For example, in Finland, where mobile penetration exceeds 40 percent, mobile service turnover accounted for 40 percent of the telecoms market by value in 1997 (and local fixed telephony accounted for only 34 percent). In the future, it is likely that regulators will require the unbundling of services not only from fixed-network incumbents but also from mobile operators.

The GSM specification is ideal in this respect. In order to be a GSM operator, a radio network is optional. An operator needs only a switch and a home location register. Fixed-network operators may purchase radio network capacity in bulk and provide the other services themselves. This change means the cost components of mobile networks must be detailed and cost-based interconnect pricing will become inevitable. This scenario where mobile operators sell airtime in bulk is not a distant possibility; the first agreements of this type are already in place.

Tariff Changes Driven by Nonvoice Services

Today’s mobile usage relates almost exclusively to voice telephony. Nonvoice communications will increase substantially and account for an ever-increasing share of total traffic. This condition will be facilitated by an increasing number of wireless smart telephones and other wireless data connections. A Q1 1997 study for the European Commission’s Universal Mobile Telecommunications System forum predicts that by 2005 nonvoice use over mobile communications devices in Europe will exceed voice usage. Analysis by Nokia shows that a GSM operator pursuing an aggressive strategy on wireless data would be able to generate 20 to 30 percent of its revenue from new services by 2000.

With current voice-orientated mobile tariff structures, the forecast growth in nonvoice services is unlikely to materialise. Whilst high speed circuit switched data over cellular telephony will play a significant role, packet-orientated transport may be far more important, meaning that pricing will have to incorporate aspects of bandwidth and quality of service. In a world where voice is only a subset of data, tariffs for mobile service may change in ways that cannot be fully predicted today. However, in terms of structure, the tariffs are likely to have a significant component for service availability. Other components will be usage related to the volume of data and quality of service. Bundled-minute tariffs with a low marginal per-minute cost are closer to this concept than traditional mobile tariffs and, therefore, are likely to form the basis for future tariff structures.

Some mobile operators may be tempted to introduce tariffs that do not have a usage-related component. This method is satisfactory as long as the fixed price is high enough and the service is used for voice only because the saturation point for voice usage is relatively low. However, the use of mobile services for Internet access or other nonvoice applications changes the situation. For example, US local exchange carriers that do not charge for local calls experience network congestion caused by Internet dial-up access. When local networks are used primarily for voice and networks are fully built out, the marginal cost of connecting a voice call is very small. However, the increased demand caused by Internet access means that local exchange carriers must make significant investments in the network without charging for the traffic. This scenario demonstrates that failure to charge for a service that has a cost is as much a market distortion as it is to charge prices at levels well above costs.

Acknowledgment

The data on cellular/PCS pricing in the US are courtesy of The Yankee Group. The pie chart breakdown of the telecommunications services market in Finland is courtesy of the Finland Ministry of Transport & Communications.