The economics of wireless systems are completely different from those of wired systems. To provide a wireless service, an operator must erect a transmitter. Service then is offered to the customers in the coverage area. Those who require the service are provided with a receiver unit to be mounted on the side of their house (as with cable, customers may pay for the unit, but its cost is borne in some fashion by the operator). If a subscriber subsequently no longer requires service, the subscriber unit is removed from the side of the house and installed elsewhere.
The key costs for WLL are the cost of installing a transmitter, the number of subscribers for whom the transmitter provides service, and the cost of a subscriber unit. Those costs vary dramatically across the different WLL technologies, as will be explained in detail in Part IV. For the moment, assume that a transmitter costs $150,000 to buy and install, that it covers an area with a radius of 4 km, and that subscriber units cost The Economics of Wireless Versus Fixed 63
1. Until WLL systems have been in deployment for a number of years, the typical maintenance cost will not be known. Perhaps the best comparison is with cellular systems, which typically have a 1% maintenance cost. However, WLL networks also need to maintain subscriber equipment, which may push the cost higher.
around $400 each. The key variables determining the cost, then, are the following:
■ The number of houses in the coverage area of one transmitter;
■ The costs of the subscriber units.
Suppose, for example, that there are 1,000 houses in a cell with a radius of 4 km, that the transmitter costs $150,000, and that subscriber units are $400 each, including installation. The fraction of the transmitter cost applying to each house, then, is $150, and the total cost per house is
$550. That is slightly more than the $450 per house calculated in the cable case.
Now suppose that penetration is only 20%. The transmitter cost per subscriber rises to $750 per house and the total cost to $1,150. That is significantly less expensive than the $1,650 calculated in the cable case.
The graphs in Figures 5.2, 5.3, and 5.4 show the relative costs of cable access and wireless access for a range of housing density and penetration levels. The advantages of WLL as the housing density falls or the pene- tration falls are obvious.
The figures show WLL consistently providing a less expensive access technology than cable, with the difference decreasing as the penetration increases. The costs of both systems rise as the density of homes falls, with WLL remaining the least expensive alternative.
5,000 4,000 3,000 2,000 1,000 0
5% 10% 15% 20% 25%
Penetration
Cost / subscribers ($)
Cable WLL
Figure 5.2 Relative costs of cable and WLL in a high-density case.
A new operator faced with the choice of using wired or wireless access techniques probably should adopt a wireless approach. Of interest to operators in first world countries will be whether they can compete against the incumbents, particularly where the incumbent already has 100% penetration and has depreciated installation costs. In that case, a comparison is made between the maintenance costs of wired and the installation and maintenance costs of wireless. Taking an average main- tenance figure of 1.25% for wireless and 5% for wired and using the same assumptions as in the earlier example, Figure 5.5 shows the relative costs of an incumbent using depreciated wire and of a new entrant using a WLL solution, considered over a 15-year period and taking into account the cost of capital.
Cost/subscribers($)
8,000 6,000 4,000 2,000 0
5% 10% 15% 20% 25%
Penetration
Cable WLL
Figure 5.3 Relative costs of cable and WLL in a medium-density case.
Cost/subscribers($) 10,000 8,000 6,000 4,000 2,000 0
5% 10% 15% 20% 25%
Penetration
Cable WLL
Figure 5.4 Relative costs of cable and WLL in a low-density case.
The Economics of Wireless Versus Fixed 65
That presents quite a different picture. Although its maintenance costs are higher, the incumbent no longer has to bear the installation costs. According to Figure 5.5, it appears that the new operator should not attempt to compete with the incumbent. However, there are a number of reasons why the incumbent’s costs may be higher. First, the incumbent, as a previous monopoly supplier, is likely to be somewhat inefficient and hence to have a higher overall cost base than is suggested here. Second, the incumbent is likely to have a USO; as discussed in Section 4.5, that typically means that the incumbent subsidizes loss-mak- ing customers through other customers, increasing the cost per customer shown in Figure 5.5. Third, the wireless operator can embark on cream- skimming, in which they cover only those areas where the economics are particularly favorable.
Competing with an incumbent certainly is possible, for example, in the United Kingdom, Ionica is successfully gaining market share and is predicting profitable operation, despite apparently higher costs. How- ever, it requires considerably more skill than the case in which the incumbent is nonexistent or is providing an inadequate service.
0 200 400 600 800 1,000 1,200 1,400
5% 10% 15% 20% 25%
WLL Incumbent
Penetration
Cost per subscriber ($)
Figure 5.5 Through-life costs for incumbent and new operator.