Document(s) 4 de 12

This book focuses on developing understanding around two key issues:

• the role that different models of shared access ICT (information and communication technology) centres have played in meeting universal access goals and supporting development in South Africa and Uganda;

• the external factors that have facilitated and constrained the activities of these centres.

Shared access ICT centres include cybercafés, telecentres, public phone shops and booths, computer secretarial services, and any centre, for profit or not, that provides shared public access to one or more ICTs.¹

South Africa and Uganda were selected for the study because their political leaders had expressed commitment to the idea of using ICTs for development, and because both countries have had relatively early experience in putting such ideas into action. Each country chose a different path to implement its ideas, with much of Uganda's early experiences sparked by foreign donor interest. Thus, the depth and range of experience within these two countries provide a valuable opportunity for broader learning around these issues. The analysis is based on a review of national policy and implementation strategies, case studies of access centres in five communities (two in South Africa, three in Uganda), and analysis of the links between national policy and strategies and on-the-ground experiences. This analysis is intended to clarify the role that access centres play in larger universal access and development strategies. This is further used to identify and recommend strategies that policy makers, donor agencies and supportive intermediaries (such as national NGOs, networks and associations) can use to support and strengthen shared ICT centres and to increase their developmental impact.

The study had four research objectives:

• describing what universal access means and is intended to achieve, as stated by policy and national leaders in South Africa and Uganda;

• exploring what forms shared access centres have taken and their role in national universal access strategies;

• analysing the roles that different forms of shared access centres have played in development;

• identifying the ways policy makers, donors and supportive intermediaries can expand, sustain and use shared access initiatives to achieve universal access goals and maximize developmental benefits.

Key definitions

Shared access centres

Shared access centres, sometimes referred to as access points, are resources available to either all or a significant segment of the public. They provide direct access to ICTs and related value-added services. Common examples are public telephones, cybercafés, telecentres, computer training centres, and computer secretarial services and business centres.

Telecentres

'Telecentre' is a widely recognized term that encompasses most variants of shared access facility with an explicit development objective. Most telecentres are multipurpose, offering a range of services and ICTs, often including photocopying, computer typesetting, faxing, Internet (although many are beset by connectivity problems), phone and computer training, plus other value-added services that vary from site to site. Most telecentres in Africa have been established by external funding agencies in partnership with local NGOs and/or government structures. They are intended to 'bridge the digital divide' by reaching those who otherwise would be unlikely to access services. As such, telecentres are sometimes located in places where the market is not providing such services, or otherwise attempt to reach a broader segment of the population than the market would.

Information and communication technologies (ICTs)

The term ICTs usually refers to telecommunication-related and digital computer technologies, although with technological convergence, the distinction between telecommunication and broadcasting technologies is decreasing in both clarity and relevance. This research focuses specifically on telephone, fax, computer and Internet/e-mail, although it also includes a project that uses data and voice transmission over satellite broadcasts. While radio is also considered within the scope of the research, it is not considered an ICT but rather a 'close relative', while television and other media are not included. This follows general convention and the definitions used in most national access policies.

Universal access

Universal access is a common policy goal in which 100 per cent of the population is able to make use of a publicly available resource, whether it be basic schooling, health centres, or in this instance, ICTs. Universal access to ICTs is defined by some benchmark that changes from country to country and over time. For example, South Africa defined universal access as the population within 30 minutes of a public telephone, while Uganda defined it as a public telephone at every subcounty or community with a population of 5,000 or more. Once targets are reached, the benchmarks can be raised: for example, from within 30 to within 15 minutes of a telephone. Rapid technological change also requires that these definitions be regularly reviewed. Most policies attempt to be 'technologically neutral,' focusing on the service (ie, point-to-point voice telephony) rather than the means of achieving it. In practice, however, monopoly rights granted through licensing often limit the application of new technologies that could otherwise help to reach access targets.

Universal service

Universal service is a common policy goal whereby 100 per cent of the population is able to receive, by reasonable request and at a reasonable cost, a specific service on an individual or household basis. Traditionally, in the realm of ICTs, the target has been universal telephone service; that is, all households should be able to have, on request, home telephone service at an affordable rate. This definition has expanded in many countries to include value-added phone services such as messaging and voice mail as well as data transmission and Internet capacity. In countries with low average incomes and very low tele-densities, universal service is not normally an immediate short-term policy goal, but may remain as a long-term goal.

Development

In this research, 'development' is understood in two senses. Firstly, it is progress towards whatever a national government has defined within its policy as the priority goals of its people.² Secondly, it refers to the increased capacity to choose. When applied to a community or national collective, this definition further requires that the capacity of one individual to choose does not reduce the range of freedom of others. Thus, it is an increase in the overall range of choice and freedom at the individual and collective level. This latter definition is based primarily on the work of Amartya Sen.

Context and rationale

Brief background on universal access in South Africa and Uganda

In recent years, universal access to ICTs has become a policy goal for many national governments, international development agencies and intergovernmental agencies such as those of the United Nations. ICTs are widely seen as a fundamental element of a newly emerging global information (or knowledge) society. As such, they represent both threat and promise. They may lead to greater opportunities for those who can partake of them; but they may also lead to greater exclusion for those who cannot.

The fear of this rift is made more urgent by the observation that new and deepening forms of exclusion are appearing in our globalizing society. In Africa, the development of nations has often been slow or stagnant and even when national indicators are moving up, close observers are often skeptical as to how much of this growth is reaching the poorest people and communities (eg, Jamal, 1998). In the context of rural Africa, the focus is not only on including people in the information society, but also on what these ICTs can offer individuals and communities who are in immediate need of better services and livelihood opportunities.

South Africa and Uganda are two African nations which made early and strong political commitment to the concept of universal access and whose leaders publicly requested the international community to help them in their efforts. By the mid-1990s, the governments of both countries were engaged in similar processes of developing universal access policies and setting up bodies to regulate telecom operators and implement access initiatives, while beginning to privatize and liberalize their telecommunication sectors. They were influenced and supported by a large number of international players and to a lesser extent by their own private and civil sectors. These two nations had unique experiences with universal access as a whole and with the role of shared access centres in particular. Their stories illustrate the relationship between shared access centres and broader universal access and telecommunication reform strategies, and the kinds of influence that national governments can exert, positively or negatively, on grassroots ICT access efforts.

The rhetoric of universal access is often underlain with self-interested political and economic forces redefining the public interest to coincide with their own. In particular, international telecommunication companies and the old state telecommunication monopolies are infamous for their talents as influential lobbyists on their own behalf. The American Telephone and Telegraph Company (now known simply as AT&T), coined the term 'universal service' in an argument to be granted monopoly rights in the USA in the early twentieth century (Melody, 1998). More recently, large telecommunication operators attempt to influence the international playing field and enter into new markets on terms that will maximize their own profits. 'Universal service' is still an argument that they use to justify monopolistic conditions that they claim are necessary to allow and justify large investment in telecommunication infrastructure.

The WTO in 1998 established rules of telecommunication sector reform in the General Agreement on Trade in Services (GATS). South Africa and Uganda were both signatories to this agreement. It requires signatory nations to set up an independent regulator while allowing them to determine their own definitions and indicators for universal service and access.

National policy makers and regulators play the crucial role of protecting the public interest through policy creation, implementation, monitoring and revision, and through the effective enforcement of regulatory conditions. On the one hand, failure to do so implies that universal access policy is simply rhetoric to sugarcoat the self-interested actions of powerful telecom operators, which are unlikely to coincide with public interest or to achieve universal access (Melody, 1998). On the other hand, success in creating publicly meaningful universal access policies that are integrated into a larger development strategy implies greatly increased opportunities for socio-economic development and improved well-being.

Shared access centres, universal access and development

Telecommunication reform has been undertaken in many African countries, including South Africa and Uganda. These processes follow the same general structure and logic, although heavy lobbying by various interested parties means that they do not always run quickly or smoothly. Interesting dramas are often created along the way.

Both South Africa and Uganda, like most African countries, previously had state-owned telecom monopolies with lacklustre performance, resulting in low teledensities, a large unmet demand for telephone service, low quality of service and little telecom infrastructure outside major urban centres. In addition, South Africa, although it had one of the highest teledensities in Africa at the end of the apartheid era, suffered a huge disparity in service between different racial groups.

The process of reform has generally involved partial privatization of the telecom company. Typically, one or more companies, mainly foreign-owned, buy shares while the government retains a significant number of shares which it is supposed to dispose of over time. The company is granted a period of exclusivity, usually about five years, coupled with licensing obligations related to performance measures and universal service targets. At the end of exclusivity, the market is opened to further competition, which may include one or more licensed operators. During the same period, cellular phone companies have entered the market and been licensed under a separate category. South Africa and Uganda each had three licensed cellular operators in 2003.

Both telecom sector reform and licensing obligations are intended to directly support universal service and access goals. By themselves, however, they are not sufficient to realize the full potential of ICTs for development. Bill

Melody argues that while these changes may increase the supply of telecom infrastructure, they are unlikely, given historical precedent, to fully penetrate rural areas. Even if they do, they will not bridge the accessibility divide unless demand can be built through greater public awareness of ICTs and capacity to use them (Melody, 1998).

This is where the potential role of shared access centres becomes apparent. Telecentres, for example, originated in Scandinavia as a way to make ICTs more economical, but more importantly, to build demand by sensitizing and training previously unexposed segments of the population to the potential utility of the technologies. The collective aspect of the centres also meant they could serve as focal points for community development. In the mid-1990s, many African countries were exposed to the telecentre concept through international symposia, where the International Telecommunication Union (ITU) championed them as valuable development tools. A number of international donor agencies set up pilot telecentre projects in Africa. And the South African government implemented an ambitious telecentre programme as part of its own universal access strategy. At the same time, entrepreneurs in African cities and towns began setting up cybercafés, telephone bureaux and business centres offering ICT services. While these were motivated by perceived market opportunities, not development goals, they often offered similar services, and sometimes competed directly with the telecentres.

Where this research fits in

This research aims to analyse and learn from on-the-ground experience with different forms of shared access centre in the context of overall universal access policy goals and strategies. These centres are generally recognized to be an important component in universal access, and especially in harnessing ICTs for development. However, it is not yet well understood how these centres contribute to, and are affected by, broader universal access strategies in the African context. Because most initiatives have been on a pilot scale, they have not yet been fully employed towards achieving universal access on a national scale. South Africa did make an ambitious attempt to do this, but unfortunately fell well short of the mark, as described later. However, these efforts, together with various pilot projects, stand-alone initiatives and market-driven entrepreneurial centres, provide an important chance to learn what role shared access centres could play, on a larger scale, in promoting universal access in a way that furthers development.

Methodology

The overall goal of the research was to understand, across multiple contexts, phenomena that are inherently dynamic and complex. Policies, markets, communities and projects are continually changing. So it was important to capture trends over time rather than merely take a snapshot of individual situations. Because there was no clear conceptual basis at the outset for deciding which elements should be the focus of attention, the research had to be somewhat open-ended.

The general methodology was to conduct five local case studies nested in two national case studies. Uganda and South Africa, as the national cases, allowed for comparison in terms of policies, market environments, and the apparent effects of these. The selected local communities in each country were analysed both as sub-components of the national studies and as case studies in their own right. These communities were selected based on two main criteria: firstly, they had an access centre that had been previously documented, and secondly, together with the other communities, they were representative of the spectrum of contexts (traditionally underserviced areas, both rural and disadvantaged urban) typically targeted by access initiatives. Within each community, the known access centre was researched with a view to understanding change. In addition, other access centres and ICT services available within the area were identified and researched. These typically included about ten other centres, including cybercafés, computer-based secretarial and printing services, public phone shops, local radio stations and training centres. This research included a semi-structured interview with management or staff and was sometimes augmented by user exit surveys. A small survey of between 65 and 95 households was conducted in each area. In total, 62 access centres were included in the study and 370 households were surveyed.

The national case studies were augmented by primary and secondary research on national policy, regulation, ICT-related service providers, national government programmes and NGOs. Research in each country also included examples of local projects extraneous to the community case studies.

As with any research, the approach was also delimited by available time and resources. Fieldwork was carried out over three and a half months, with locally hired research assistants doing the bulk of the survey work, as well as serving as guides and cultural and linguistic interpreters. Research assistants also contributed to the design of the case research.

Field methods used

Data collection took place over a period of six and a half weeks in South Africa (April to June, 2003) and seven and a half weeks in Uganda (June to July, 2003). Field methods are described in Table 1.1. Appendix 1 is a list of what was included in data collection and from whom.

Data handling and analysis

Interviews were recorded through note-taking. Observations were noted and typed after the event. Questionnaires and surveys were recorded directly onto pre-made sheets. Any translation was done on the spot by the research assistants so that all information was recorded in English. Survey and questionnaire data were entered into an Access database and from there could be imported into Excel and SPSS (Statistical Package for the Social Sciences) for quantitative analysis. Most of the analysis techniques used were qualitative. Where quantitative techniques were used, caution was exercised in their interpretation as in no case were the samples of sufficient size and the sampling of sufficient rigour to be considered representative of the populations sampled within 95 per cent confidence, or to meet the criteria of most statistical tests.

The general approach to analysis has been to build a coherent picture of the cases through consideration of all available data. Thus, triangulation across different collection methods combined with due consideration of the limitations of the methods is the best insurance of rigour. Since this research often traded off rigour for breadth, the pictures that emerge are ones 'painted with broad strokes' rather than highly detailed images.

Conceptual framework

To understand the link between universal access policies and on-the-ground experiences, one must first understand the overall context and analyse the ability of the centres themselves to provide for universal access and contribute to development. This conceptual framework attempts to do both. By abstracting from the immediate situation and identifying key actors, processes and factors that exist in different contexts, it also provides a tool for generalizing across different contexts while remaining sensitive to the differences within these contexts. For example, we will see that the Ugandan government's path to universal access (as depicted in Figure 1.1) involved market stimulation whilst the South African government took a more direct approach to implementation through the purpose-specific Universal Service Agency. The final chapter summarizes some of these similarities and differences and from these draws general lessons.

Universal access policy arena and process

The policy arena

The formation and implementation of universal access policy take place in an 'arena' consisting of various actors from the public, private and civil sectors, operating at the local, national and international level, as illustrated in Figure 1.1. The nature of the relationships and influences between all these actors is a factor in understanding differing national experiences in supporting shared access centres, and in identifying possible future interventions to strengthen these initiatives. Figure 1.1 lists some of the generic actors in both South Africa and Uganda, as well as other countries. The specifics of the national and local situations in each country are further detailed in Chapters 2 (South Africa) and 3 (Uganda).

Figure 1.1 Policy arena

The policy formation and review process

General aspects of formulating, implementing and revising universal access policy are shown in Figure 1.2. While this diagram fairly describes the main components in South Africa and Uganda (though not necessarily those in other countries), the emphasis on different means of implementing such policy, the details of both implementation and policy, and the relationship between mechanisms are unique to each country. This is covered in further detail in Chapters 2 and 3. For each national case study, a small number of community case studies are used to focus on the apparent effects of ICT policy and national-level activities upon ICT access within the communities generally, and shared access centres specifically. These are then compared with the original goals of the relevant policies.

Figure 1.2 Policy formulation and review process

The policy intent behind universal access and service to ICTs occurs within a broader framework of a national development policy. The policy will outline specific implementation mechanisms. Some of these may operate by stimulating and regulating the market. Through licensing telecom operators, governments can impose certain obligations upon them but may also grant them privileges, usually monopoly rights. One obligation is to contribute to a fund that supports access initiatives. Fund resources may be used to subsidize market enterprises or to support government or not-for-profit activities. In addition, non-governmental actors may undertake access-related initiatives or projects independently. The government may wish to support or monitor these initiatives so that they better contribute to overall access targets.

Four success factors for access centres

This research focuses on four factors as criteria for understanding the strengths and limitations of the various models of access centre encountered in the case studies and elsewhere. These factors are scalability, sustainability, reach and use, and development impact.

Scalability

Scalability in the context of this research means the capacity to provide enough access centres to reach the population of an entire country. At the micro-level, the key questions related to scalability are: What are the requirements to set up a centre, and who will actually do it? Market-based initiatives will depend on the capacity of local entrepreneurs, entry requirements and local demand, while non-profit initiatives may depend on local organizational capacity, the relationship between local and external implementers, and an adequate source of funding. Both will require technologies that allow the centre to offer services, and these technologies in turn will likely require infrastructure, especially telephone lines and electricity.

Scalability in the context of universal access doesn't demand that one model fit all situations. Rather, it requires a strategy in which a number of models and broad approaches are intelligently applied and locally adapted across different circumstances, allowing the entire population to be reached. Chapter 4 analyses the scalability of different centre types encountered in the case study areas.

Sustainability

Sustainability, the capacity to continue once begun, is a natural second requirement for access services. From a broad perspective, it doesn't much matter whether the services persist because a single centre persists, or because when that centre fails, another one or more take its place. The continuation of services will normally depend on local demand and the ability of local actors to recognize and respond to that demand. Initial access centres in previously underserved areas are sometimes described as 'market primers' because they are able to expand the local market for ICT services, creating a particular kind of sustainability that may outlast the centre itself (Fuchs, 1997).

Sustainability depends on factors both external and internal to a centre (see Box 1.1). Key external factors include the local market, national market and outside support mechanisms. Internal factors play the ultimate role in determining whether a centre will succeed, since ingenuity and strong management may overcome many obstacles. Yet external factors are important determinants of how tough an environment is and, in most parts of Africa, they add up at least to 'very tough'. All of these factors can be influenced to some degree, and often directly, by national policy. If large numbers of centres are failing within a country, the cause is more likely to be a set of inhibiting external factors than widespread manager incompetence.

Access centre reach and use

To provide universal access, access centres should be priced, distributed and designed such that any person can use the target services if he or she so chooses. This means that the price of the service must be within the reach of prospective users' ability to pay, that the users are aware of the service, and that they have the capacity to use it, or feel able to acquire such capacity. The requirements, then, are based on the characteristics of not only the access centre (supply) but also the potential users (demand).

To understand the contextual relevance of ICT use, this research uses a livelihoods approach to explore why some groups of people use ICTs and others do not, how different groups use them, and the likely impacts of such use. By applying livelihoods approaches across the community case studies, we can gain insights into common motivations for using ICTs and common barriers to use, and how current patterns of use are likely to influence development.

Chapter 5 applies this analysis to different types of access centre, generalizing across contexts where possible.

Development impact

The development impact of universal access initiatives will be the sum of the centres themselves, the process of establishing them, the services they provide over time, and the improvements they engender in people's lives. Some access centres may involve complex processes of community mobilization and these processes in turn may have a development impact, irrespective of the centre's performance. Local employment opportunities, the capacity developed by staff, and the money that flows into, out of, and circulates through the community as a result of the centre also have important implications for local economic development. The services themselves are the focus of most expectations for development impact, and this impact will vary according to the kinds of service, who uses them and the nature of the community.

The dual definition of development given earlier provides the basis for analysing development impact based on two questions. How do the centres support relevant national development goals? And what role do they play in increasing (or decreasing) the overall range of choices of local community members? In assessing the latter aspect, this book focuses primarily on local livelihood activities and priorities, as determined through household surveys and census data.

In Chapters 2 and 3, community experiences with ICTs are compared with development goals set out in policy. Chapter 5 focuses on development impact of access centre services, but also of productive opportunities associated with providing services, through the lens of livelihoods.

Document(s) 4 de 12