Connectivity and the Digital Divide – Technology, Policy, and Design Tradeoffs for Developing Regions

This paper examines some of the causes of poor connectivity in developing countries, and suggests options for overcoming the so-termed digital divide. Based on a techno-economic analysis of connectivity technologies and design, I show that technical limitations per se are not the bottleneck for widespread connectivity; rather, design, policy, and regulatory challenges are the reasons for high costs (which determines penetration to a large extent). The first part of this paper deals with the “digital divide” and a brief look at existing measures and difficulties with such measures, such as issues of granularity. I present a starting point for a new framework for the digital divide, based on what we call the 4Cs of Information and Communications Technology (ICT): Computers, Connectivity, Content, and (human) Capacity. The next section focuses on the Connectivity aspect of ICT, and examines cost components of different technologies. While there are obvious differences between technologies in terms of speeds, reach, and infrastructure, there are non-technical implications of such choices, such as legacy requirements, competition, leapfrogging, etc. Once we formulate a basic, indicative, network design for broadband, using estimated costs and some limited real-world emerging economy data, we can see what reasons there are for higher costs (and prices) than the model would indicate. We identify over a dozen policy add-ons (such as ISP licensing, spectrum fees, taxes, limits on applications such as VoIP, etc.) that raise the costs to end-users. Expanding from the idealized model, I also present a new design for broadband across Africa (“FiberAfrica”) based on GIS modeling and a bottom-up design. This is not an idealized model but a first attempt at a hypothetical continental network based on available data. It not only includes appropriate technologies (optical fibers and fixed broadband wireless) but also innovative business models such as open access networks. The analysis shows such a design could bring affordable, accessible broadband to the majority of Africans, for a one-time capital cost of only a little over $1/capita.