Comparison of the cost of providing WiMAX network coverage
The need to address the widening gap between those with easy access to ICT and those without is recognised beyond the telecoms industry. The UN has included in its Millennium Development Goals (a series of targets for 2015 agreed at the Millennium Summit of world leaders), which include the eradication of extreme hunger and poverty, achievement of universal primary education and combating HIV/AIDS and malaria, the target: "In cooperation with the private sector, make available benefits of new technologies, especially information and communications technologies."1 Broadband has been singled out for particular attention in this context; a UN press release on its recent progress report highlighted the finding that developing countries "…have unprecedented access to new information and communication technologies, with over 77 per cent of the population able to receive a mobile cellular telephone signal… But the digital divide between developed and developing countries continues to widen for technologies that drive modern information sharing (such as broadband internet connection)."2
A key question for the telecoms industry and for policy makers, therefore, is how to promote greater availability and higher take-up of broadband services in developing countries. This is not an easy question to answer, and no single set of solutions fits every country's situation, but, as a first step, it is necessary to identify the obstacles to universal broadband provision. On the demand side, the barriers to take-up of broadband in developing countries are: the fact that broadband services and computing equipment, such as PCs, are not affordable, unreliable electricity supplies and the paucity of local content. On the supply side, some of the limiting factors are: the shortage of and high price of international bandwidth, lack of incentives for operators to invest in rural roll-outs, limitations on access to spectrum and the lack of a supportive, reliable regulatory environment.
Some of these issues are beyond the sphere of influence of the telecoms industry, but policy can address most of them, or at least mitigate their negative impact. Wireless broadband technologies, such as HSPA and WiMAX, have a key role in provision of broadband access in developing markets, even if this is to provide access to a fixed location. Given their dependence on the use of spectrum, the roll-out of wireless networks will inevitably involve regulatory intervention. Aside from the often significant sums of money involved, government policy on spectrum may be considered a dry subject by policy makers, but is crucial to operators' business plans and has a direct impact on the ability of operators to provide the country's citizens with broadband coverage.
A recent Analysys Mason case study of the Kenyan WiMAX operator KDN highlights this point.3 KDN paid USD25 000 in 2003 for a 2×28MHz spectrum allocation in the 3.5GHz range. Since that time, other WiMAX players, including AccessKenya Group, UUNET Kenya and OneCom, have been awarded spectrum, but none has received an allocation of more than 7MHz. This means that KDN is the only WiMAX operator in the country with sufficient spectrum to deploy a WiMAX network of significant size. In addition, despite receiving requests for spectrum in the 2.5GHz range to be released in subsequent awards, the regulator released spectrum in the 3.3GHz band.
Both the amount of spectrum an operator holds and the band within which the spectrum lies are crucial to the operator's plans. In general, the higher the frequency of the spectrum, the smaller the area that can be covered. Using spectrum in the 2.5GHz band, for example, the area that can be covered by a single cell (using a given amount of spectrum, and assuming equal population density and coverage depth) can be over twice as large as is possible using spectrum in the 3.5GHz band. The cost implications of this for a network roll-out are dramatic. Figure 1 compares typical costs of providing WiMAX coverage for a large population using 2.5GHz or 3.5GHz spectrum: to provide good outdoor coverage using 3.5GHz, rather than 2.5GHz, spectrum results in a substantial 33% increase in associated costs. If indoor coverage is added to the equation, the cost of using 3.5GHz spectrum is 73% higher than if using 2.5GHz spectrum. Clearly, the business case for extending a roll-out to less-densely populated areas, or even for rolling out a network at all, can depend on factors such as spectrum availability.
This example clearly indicates that policy makers need to take into account how regulation can affect business cases for, and hence investment in, wireless broadband networks. Policy that accommodates commercial considerations and regulates accordingly can act as a key enabler of private-sector investment in developing markets. It has an impact not only on operator's bottom lines, but also on the services available to the population and on the pricing of services. A well-designed policy framework promotes win-win situations in which operators can provide broadband access profitably and end users can take advantage of this access and the benefits it can bring.
Clearly, spectrum is not the only issue affecting the implementation of effective broadband policy. A forthcoming research report from Analysys Mason, Creating a successful broadband policy in developing countries, will explore the relevant issues in more depth.