Africa Region Working Paper Series Number 20 Information and Communication Technology, P2overty, and Development in sub-5aharan Africa and South Asia Miria Pigato August 2001 twr Information and Communication Technology, Poverty, and Development in sub-Saharan Africa and South Asia August 2001 Africa Region Working Paper Series No. 20 Abstract The objectives of the paper are: to examine patterns of utilization, ownership and affordability of Information and Communication Technologies (ICTs) in sub-Saharan Africa (SSA) and South Asia; to discuss applications of ICTs to the poor by the private sector, the Government and external donors; and to suggest ways in which ICTs can best be used in poverty alleviation strategies. The paper finds that SSA and South Asia have the lowest access to ICTs resources. There are two critical access gaps, between urban and rural areas, and between the poorest and the richest 20% of the population. For their infornation needs, the poor rely mostly on the informal networks that they trust, such as family, friends and local leaders. By contrast, formal sources of infonnation, such as NGOs, newspapers or politicians, are less trusted and used. The new forms of ICTs, including Internet, fax and computers, have touched only some 2% of low-income households, mostly in urban areas. Evidence indicate that in a sufficiently competitive and liberalized market the private sector is most effective in providing commercially viable communications services, even in rural or less viable areas; Government-led initiatives have mixed results; externally funded initiatives have potential benefits but also several drawbacks in terms of financial as well as technological sustainability. The paper proposes the following principles for using ICTs in poverty alleviation strategies: i) information is a dynamic process of acquisition and use; ii) the effective use of information by the poor may be constrained by lack of skills, financial resources and the existence of urban/rural, gender and other inequities; iii) the effectiveness of ICTs must be assessed along with existing information systems; iv) ICTs must focus on areas where complementary investment has already been built and encourage the participation of the communities and of the poor; v) long term strategies for ICT diffusion must be centered around integrating ICTs into wider educational programs. Authors' Affiliation and Sponsorship Miria A. Pigato, Senior Economist, AFTM2, Africa Region, The World Bank Email: mpigato@worldbank.org THE WORKING PAPER SERIES The Africa Region Working Paper Series expedites dissemination of applied research and policy studies with potential for improving economic performance and social conditions in Sub-Saharan Africa. The Series publishes papers at preliminary stages to stimulate timely discussion within the Region and among client countries, donors, and the policy research community. The editorial board for the Series consists of representatives from professional Families appointed by the Region's Sector Directors. Editor in charge of the series: Antoine Waldburger, AFTM3, Email: awaldburger*worldbank.org, who may be contacted for hard copies. For additional information visit the Web site hftp://www.worldbank.ora/afr/wDs/index.htm, where copies are available in pdf format. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s). They do not necessarily represent the views of the World Bank Group, its Executive Directors, or the countries that they represent and should not be atfributed to them. Information and Communication Technology, Poverty, and Development in sub-Saharan Africa and South Asia Miria Pigato Senior Economist, AFTM2, Africa Region, The World Bank Email: mpigato@worldbank.org August 2001 Special thanks to Christopher D. Scott and Alan Gelb for useful comments and to Kim Murrell for editorial assistance. The paper incorporates comments received at a TechNet/InfoDEv seminar in March 2001 at the Economists' Forum in May 2001; and it draws on inputs provided by Subhash Bhatnagar, Richard Duncombe, Reena Marwah, Maura Liberatori and Emmanuel Mungunasi. Financial support from the Netherlands Partnership Program for the background papers is gratefully acknowledged. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s). They do not necessarily represent the views of the World Bank Group, its Executive Directors, or the countries that they represent and should not be attributed to them. CONTENTS EXECUTIVE SUMMARY .................................................................... i 1. INFORMATION COMMUNICATIONS TECHNOLOGY IN POVERTY ALLEVIATION. I 1. The widening gap in access and use of ICTs ................................................................ 1 2. The role of ICTs in Poverty Alleviation and Development .................................................2 3. Constraints to the use of ICTs by the poor ................................................................ 6 4. Conclusion .................................................................8 2. ACCESS TO ICTS IN SUB-SAHARAN AFRICA AND SOUTH ASIA: AN EMPIRICAL ANALYSIS ................................................................9 1. Access to telephony - fixed and mobile lines - and to computer based information ..........9 2. Access to non digital communication resources ............................................................... 13 3. Ownership of information and communication assets by rural/urban households and by wealth quintiles ............................................................... 14 4. Affordability of Information and Communication resources ............................................. 16 5. User skills and capacities for access ................................................................ 17 6. Conclusions ................................................................ 18 3. LEARNING FROM CASE STUDIES AND CONCLUSIONS .20 1. Surveys of information needs and access to ICTs in poor households .20 2. Studies of information needs and ICT impact in small and medium enterprises .21 3. Findings from private sector involvement in extending access to ICTs .21 4. Findings from Government-led ICT applications for rural development in India. 22 5. Externally funded initiatives to improve access to information and ICTs by the poor..,,24 6. Policy Implications of the Information Revolution for Poverty and Development. 26 ANNEXES Annex 1: Information needs and access to ICTs by poor urban and rural households .................... 29 Nepal .............................................................. 29 India .............................................................. 31 Annex I1: Information needs and access to ICTs by small, micro and medium enterprises ........... 35 Botswana .............................................................. 35 Tanzania .............................................................. 37 Annex III: Communal Access Models .............................................................. 42 South Africa ................................ 42 Annex IV: Private sector led ICT development ................................ 45 Ghana ................................ 45 Annex V: ICT applications for rural development in India ............................................................ 47 Government led initiatives improve the administration of rural programs ............................. 47 Public-private partnership: the application of information technology in Milk Collection Cooperatives .................................................................. 50 DonorlNGO-financed initiatives: the Village Information Shops in Pondicheri .................... 51 Annex VI: The Grameen Village Pay Phone Initiative in Bangladesh ........................................... 53 Annex VII: Comparative Indicators ................................................................. 57 ANNEX VIII: LIST OF SSA AND SOUTH ASIAN COUNTRIES ACCORDING TO GDP PER CAPITA GROUPINGS ................................................................. 62 ANNEX IX: THE DIFFUSION OF ICT AMONG THE RICH AND THE POOR ....................... 63 References ................................................................. 65 Executive Summary The current wave of globalization - the trend towards worldwide integration of markets- is spurred by the development of information and communication technologies (ICTs), including the Internet, mobile phones and satellite networks. Rapid growth of ICT usage in high-income countries is raising concerns about a "digital divide" emerging between rich and poor nations. Lack of access to ICTs in developing countries has not traditionally been viewed as a deprivation in the way that lack of food, basic health care and shelter have been. However, there is increasing evidence that access to ICTs can have a direct impact on raising living standard and quality of life of the poor. The indirect impact on poverty alleviation, through growth and productivity, has long been recognized. This paper focuses on the relationship between ICTs and poverty. Empirical evidence is taken from the two poorest regions of the world - Sub-Saharan Africa (SSA) and South Asia. Several sources of information are used: i) statistical data from international sources; ii) Demographic and Health Surveys (DHS) from 26 low income SSA and South Asian countries during the period 1991-99; iii) two surveys of poor urban and rural households in Nepal and India (conducted in 2000); and two surveys of small, micro and medium enterprises in Tanzania (2000) and Botswana (1999); and iv) case studies of applications of ICT in rural areas from both SSA and South Asia from published sources. The objectives of the paper are to: * examine patterns of utilization, ownership and affordability of ICTs within countries in SSA and South Asia; * discuss experiences of extension of access and applications of information technology by the private sector, the Government and external donors; * suggest ways in which information and ICTs can best be used in poverty alleviation strategies. Summary of findings SSA and South Asia have lowest ICT access - SSA and South Asia are the poorest regions in the world and have the lowest access to information and communication resources. With respect to telephone lines, for example, there are only 14 and 19 telephone per 1000 inhabitants respectively in the two regions, compared with 69 in low and middle-income countries. However, since the mid-1990s, countries that have adopted deregulation and privatization policies have experienced rapid growth in the expansion of communication networks. The Internet, for instance, is beginning to be utilized in urban areas but the constraints to expansion are considerable and include lack of basic connectivity to telephone networks, high connection charges, and low computer skills. The prices for fixed line network access in many SSA and South Asia countries are still well ahead of the world average for low-income countries. Taken overall, fixed line telecommunication tariffs are about 112% of GDP per capita in very low-income economies compared with 8.25% and 2% of GDP per capita in middle and high-income economies. By contrast, many countries have experienced a real revolution in mobile networks particularly where fixed wire networks are underdeveloped. But tariffs for mobile cellular access represent a considerably higher proportion of GDP per capita at present. - and within countries there is urban/rural and rich/poor divide There are significant differences in access to ICTs within SSA and South Asian countries. The first is between urban and rural areas. Data from DHS in 26 countries during 1991-99 reveal that ICT ownership varies greatly, with radio, television and telephone ownership registering 65%, 28% and 5% respectively for urban areas, compared with 38%, 5% and .18% i respectively for rural areas. The second access gap is, not surprisingly, between the poorest and the richest population quintiles. Telephone ownership is confined to only some 10% of the richest quintile. Television ownership is high among the better off but almost non- existent for the lowest 40% of the population. By contrast, radio ownership shows the highest overall level of ownership, reaching about 85% among the wealthiest, and falling to below 10% for the poorest in Africa. Lower rates of ownership are observed in South Asia throughout all quintiles. An unmet demandfor information Surveys of urban and rural households in Nepal and India suggest that the poor favor informal networks of trusted family, friends and local leaders over formal sources of information, such as NGOs, newspapers or politicians. Advanced ICTs (Internet, fax and computers) are used by only 2% of low-income households, mostly in urban areas in both countries. The surveys found however, that while the poor rely on informal networks, these networks do not adequately satisfy their information needs. The urban poor would like more information regarding employment and job training opportunities, while the rural poor desire additional information regarding agricultural practices, markets and prices. Surveys of SMEEs in Tanzania and Botswana reveal a similar dependence on informal networks for information needs. Access to advanced ICTs (Internet, fax, and computers) is very low except among firms in the services and export sectors. And even among firms that own a computer, usage levels are low. There are a number of reasons for this. First, with respect to the Internet, the available content is often not relevant to the information needs of local firms. Second, off-the-shelf software is not well suited to the requirements of small local enterprises. Finally, firms often to not have staff capable of maintaining hardware. ,Private sector initiatives are most successful Findings from country studies indicate that in a sufficiently competitive and liberalized market (i.e. with a transparent legal framework, liberal entry and pricing policies) the private sector is effective in providing commercially viable communication services, even in many rural or less viable areas. And in areas where providing services is not financially viable, regulators can use a number of instruments such as service requirements, special financing mechanisms and investment subsidies to provide incentives to the private sector to promote public access. For example, privately run pay phones, phones shops and even telecenters have proved to be successful, and financially sustainable, in remote areas in many countries, including India, Ghana and South Africa. Experience shows that the poor are often willing to pay at least a portion of the commercial cost in view of the social and economic benefits of local access. Government led IT initiatives for rural development have mixed results There have been a number of Governments initiatives, particularly in South Asia, that have used information technology to enhance rural development programs and improve the delivery of public services. India, for example, has a wide range of experience including Government computerization schemes, satellite communications and distance education and training via the Internet. Some of these projects have been quite successful suggesting that the potential impact of IT on development can be enormous, particularly in terms of improved health, hygiene, nutrition and education. However, replicating success on a large scale often proves elusive for a number of reasons. First, some projects succeed in large part because of the enthusiasm and competence of the initiators, factors that are not always present. Second, the motivation to improve administration is often lacking. The third reason is that there are often inadequate financial resources to scale up pilot projects. Finally, there is often unwillingness on the part of Government agencies to identify and develop suitable ICT applications and/or to reengineer bureaucratic work accordingly. For example, an initiative to computerize district Rural ii Development Agencies (see Annex V) failed to replicate the success of the initial pilot, resulting in only a marginal administrative efficiency gain. The lesson from India's experience is that technology alone does not create change. It takes commitment by people and requires a long implementation period. Even when initiatives are successful (i.e. brought measurable benefits), there remains the issue of financial sustainability and the problem of how to ascertain what value citizens attach to services provided, given that they are often initially free. Externallyfunded initiatives have numerous drawbacks In recent years, externally funded initiatives (NGOs, bilateral donors, etc.) have mushroomed throughout SSA and South Asia aimed at extending ICT access to the very poor who would otherwise be unable to pay. An example would be the Swaminathan Research Foundation in South India (details provided in Annex). The potential benefits, in terms of bringing information regarding healthcare or educational opportunities to the very poor, are obvious. However, there are also a number of potential drawbacks such as financial sustainability and lack of local demand for ICT-based information. With respect to the former, donor funding usually dries up after the initial investment and local groups usually have difficulty keeping up with the follow-on costs. In terms of the latter, it takes time for people to comprehend the potential benefits of ICT-mediated information, or be willing to trust information that does not derive from personal networks. Towards an integratedframeworkfor using ICTs in poverty alleviation strategies This paper advocates the need for an integrated framework to develop appropriate policies of access and diffusion of ICT within developing countries. Evidence shows that technology is not a goal in itself, but a means of achieving development goals. In this context, it should be recognized that even within low and very low income countries there are divergent needs for information and technology. The modem sector and the educated elites can probably benefit in the very short run from the rapid introduction of ICTs. In this context the Government has a key role in establishing a liberalized and competitive market for communications within which the private sector can flourish. As already noted and discussed in other papers', the Government can use a number of incentives/requirements to catalyze private investment and to assure that the private sector extend public access to communications services even in rural or disadvantaged areas. International agencies and donors may also leverage resources through direct support in areas that are bypassed by the private sector. Information strategies should become a crucial component of poverty alleviation strategies. First, there should be recognition that information is not a static, equitably available commodity, but a dynamic process of acquisition, application and use. Second, the Government should recognize that the effective use of ICTs by the poor may be constrained by lack of skills, financial resources and the existence of urban/rural, gender and other inequalities. Moreover, the effectiveness of ICTs must be assessed alongside existing information systems, and in the context of interaction within pre-existing organic information environments. Third, ICTs must focus on strategic benefits in areas where complementary investments have already been built. In this sense it should encourage the active participation of community-based intermediaries, helping to ensure two way transmission of information, thus giving the poor a voice. Finally long-term strategies for ICT diffusion among the wider population must be centered around revised education policies such as integrating ICT skills into programs of vocationally based training which emphasize wider information handling skills, business and management skills and enterpreneurship training. X It is beyond the scope of this paper to introduce telecom regulatory issues. This topic is covered comprehensively in: Dymond et al (2000). iii 1. INFORMATION COMMUNICATIONS TECHNOLOGY IN POVERTY ALLEVIATION 1. This chapter explores the relationship between information communications technologies (ICTs) and poverty alleviation. ICTs include both the information infrastructure -wires, transmitters, computers - and the information technology, i.e. the applications and content that travel through these infrastructures.2 The chapter looks at the direct and indirect impacts of information technologies on growth and the way in which benefits from ICT adoption can be distributed to the poor. It presents the 'optimistic view' of ICT in development and suggests that this focuses too heavily on technology potential, rather than on adapting appropriate information systems to meet the real needs in different developing environments. It analyses the constraints for accessing and applying information technologies by the poor and suggests the need for a more "integrated framework" for understanding both information and ICT in relation to poverty alleviation and development. 1. The widening gap in access and use of ICTs 2. There are currently 1.2 billion people living under one dollar a day, of which about 70 per cent are located in SSA and South Asia (see Table 1.1). The two poorest regions of the world are also those with the lowest access to information and communication technologies. The gap in access to and use of ICTs, the so-called 'digital divide' often follows and reinforces existing inequality and poverty pattems. The digital divide has emerged both within countries - where the rich, educated and young (often male) are most likely to use ICTs and between developed and developing countries. For examnple, in March 2000 out of the 304 million people with Internet access, only 2.6 million were in Africa while 137 million were located in the U.S. and Canada. By way of another illustration, the average OECD country has 11 times the per capita income of a South Asian country, but 40 times as many computers. Table 1.1 - Comparative indicators of poverty and information Share of the People living on Urban Electricity Radios Mobile High population living less than $1 a day population (% cons.Per per 1000 phones technology on less than $1 a of total) capita people per 1000 exports (% day (Kilo/hour) people mfg. exports) 1987 1998 1987 1998 1999 1997 1997 1998 1998 EastAsiaandPacific 26.6 15.3 417.5 278.3 34 771 302 25 28 EuropeandcentralAsia 0.2 5.1 1.1 24.0 67 2693 442 23 8 LatinAmnericaandthe 15.3 15.6 63.7 78.2 75 1402 420 45 12 Caribbean Middle East North 4.3 1.9 9.3 5.5 58 1159 274 8 1 Africa South Asia 44.9 40.0 474.4 522.0 28 324 112 1 4 Sub-Saharan Africa 46.6 46.3 217.2 290.9 34 446 198 5 Source: World Bank Live Database 2 See World Bank Group (2000), ICT Sector Strategy Paper: Bridging the Digital Divide, November, Draft. 1 3. In 1998 Sub-Saharan Africa and South Asia had 14 and 19 telephone lines per 1000 inhabitants respectively (see Fig. 1.1) as compared with an average of 69 in low and middle income countries and 146 in the world. The same picture emerges from the analysis of other indicators, such as television sets or computers per 1000 people. Moreover, for many poor countries, the rate at which ICTs are developing is slower than in developed countries. The gap between those who have access to new technologies and those who do not is widening. Recent research (Rodriguez and Wilson, 2000) found that there is increased divergence in the technological progress Fig. 1.1 Access to ICTs in different regions, 1995-1998 (an index composed of Telephone Mainlines per 1000 people measures of personal computers, Internet Wad hosts, fax machine, Low& Middle Income mobile phones and Sub-Saharha Afica television) across South Asia F V975, countries. The Middle East & NAfrica llt inequality in access to Latin Ameica & Carib. ICTs is larger than Europe & Centr Asias income inequality. East Asia & Pacific __ _ Evidence shows that 0 no 200 300 400 new 'digital' sources of Television sets per 1000 people information and knowledge, while Wold benefiting the minority Low & Middle Income of the well off and Sub-Sahwwn Africa educated, are bypassing South Asia the less educated and Middle East& NAfric& the poor (UNDP, 1999; Latin America & Carib. Wilson, 1999)3. Euroe&Centrat Asia East Asia & Pacific 2. TheroleofICTsin 0 Do 200 300 400 500 600 700 Poverty Alleviation and Personal Computers per 1000 people Development W__d 4. Is the widening Wome gap in the access to and LoSw&aMiddlelncorne provision of ICTs reason Sub-Saharan an_ for concern? And is it SouthAsia U 205 relevant for the poor? Middle East & NArica b 1.89 Lack of access to Lan America & Carib information and ICTs is Europe & Central Asia hardly a characteristic East Asia & Pacific that identifies the poor in 0 20 40 60 80 100 20 the way that lack of food, ___ or basic health care and shelter do. The poor suffer from material deprivation, as well as low levels of education and health; they are often powerlessness vis a vis political and social institutions4; and they have a limited ability to make choices and to lead the life that they value. By contrast, the better off have a range of choices, resulting from better education and healthcare, more work opportunities and access to global information, to name just a few. 3See Annex IX for an illustration of the model of diffusion of ICT among the rich and the poor. 4 See World Bank (2000), World Development Report, chapter 1. 2 5. The poor also produce, receive and share information, but most of it typically draws from informal and unrecorded interpersonal communications that are more qualitative in nature. Reliance on informal information sources has many limitations both for households and for small enterprises in that it is often constrained and insular particularly in remote areas that lack a basic communication infrastructure and transport links (Barton, 1997). Much of the information and knowledge5 that would help the poor to live healthier lives, to improve their educational standards and employment/business opportunities is simply not available to them. The technological revolution provides an opportunity to get information to the poor in a way that can be used by them to improve their lives6. The new information technologies have the potential to process and disseminate vast amounts of information and therefore can have a far greater impact on the lives of the poor than informal information networks. By contrast, unequal access to modem information technology exacerbates the existing gaps in material well being; it implies a lack of access to the global pool of knowledge, and to the global digital commerce. Global knowledge is becoming increasingly important in promoting economic growth through higher competitiveness, and to enhancing human development and reducing poverty. The following section discusses the direct impact that access to ICTs can have on a number of areas that are central for poverty alleviation and human development and indirect effects on economic growth. Direct impact of ICTs on developnent 6. Direct impacts of ICT on poverty reduction take several routes. First, there is the educational aspect. Information and knowledge enable the poor to understand their own circumstances and to voice their own opinions and needs. ICTs such as radio and television have been effectively used in many countries to reach students in poor rural areas (see Annex V). The Intemet provides a virtual classroom in which intense interactivity and the sharing of resources and information take place. The second important impact is on health. The ITU report on Internet for Development (ITU, 1999, Chapter 4) observes that the most immediate and visible impact that the internet can have in the poorer parts of the world is on the volume and flow of medical information. The third area of impact is productivity and income generation. ICTs give micro and small enterprises access to market information (faster and cheaper than printed material), input prices and output markets and it may strengthen forward linkages to the market (Van Crowder, 1997). Indirect impact of ICTs on development 7. Studies that have considered the indirect impact of ICTs on development have drawn a distinction between ICT consumption and ICT production (investment in information infrastructures, including telecommunications, hardware and software production). 8. Production of ICT goods and services has contributed significantly to the economic success (Mansell and Wehn, 1998) of many Asian countries. Evidence of the benefits of investments in information infrastructure appears conclusive, particularly for telecommunications. The economic benefits of telecommunications - especially voice and fax communication for business - are considerable for individual small and micro-scale businesses in LDC market environments. (Grameen Bank, 1999; Duncombe, 1999). Conversely, studies suggest that lack of accessible and affordable communication services are inhibiting the transmission of market information, preventing interaction between producers 5 Broadly, information becomes knowledge when it has been processed and understood by the recipient in a way that is useful and it is adapted to the existing knowledge base. 6 The understanding of technology as part of a wider socio-economic system is particularly important when the technology is adapted within environments very different from those for which it was originally intended (Schumacher, 1973; Anderson, 1985). When technology is transferred from one society to another it reflects the social values, institutional forms and culture of the transferring country. Contexts in recipient countries will vary enormously. 3 and consumers and the promotion of enterprises (Mgombelo and Wemer, 1997; International Telecom Review, 1998). 9. Overall, research has concluded that for developing countries benefits from ICTs are more likely to accrue from consumption rather than production (Kraemer and Dedrick, 1998). This is particularly true for poorer countries where the ability to use ICTs can dramatically improve, even in the short run, the capability of firms to face competition from developed countries. Conversely, there are thought to be fewer advantages for 'latecomer' countries to develop indigenous ICT production, which requires large capital investments and specialized skills (Pohjola, 1998). Recent research within the Asia-Pacific region (Mody and Dalman, 1992; Kraemer and Dedrick, 1998) indicates that countries with higher levels of annual growth in ICT consumption, i.e. Thailand, South Korea, India and Taiwan exhibited the highest levels of growth in productivity and GDP. Research in India shows a strong correlation between growth in ICT consumption, productivity and GDP between 1984 and 1990.7 Significant productivity benefits for individual firms and economic sectors through investment in ICTs have been demonstrated for exporting firms in less developed countries and for other modem sector enterprises that do business across national borders (Brynjilfsson and Hitt, 1993, UNU/WIDER, 1999). 10. Recent research in the U.S." has concluded that the production and use of information technology has contributed half or more of the acceleration in U.S. productivity growth in the second half of the 1990s. However, earlier studies, based on the experience of the 1970s and 1980s, failed to establish any relationship between ICT consumption investment and growth in productivity - giving rise to the so-called 'productivity paradox'. These studies indicated that although considerable amounts of ICT investment took place over the period, growth in productivity actually slowed down (Baily, 1986; Baily and Gordon, 1988). This was found to be particularly the case in the service sector (Roach, 1988) both at the national level and within groups of enterprises (Loveman, 1988). Although this research has now largely been superseded, it did give rise to three important observations concerning the impact of ICT diffusion, which may still be relevant for countries in a nascent stage of technological development today: * First, there is a significant time lag necessary for benefits to accrue - possibly spanning many years or decades. Diffusion of IT must achieve a 'critical mass' in terms of coverage, organizational adaptation and learning by doing before widespread productivity gains become observable (Freeman, 1988, 1996). * Second organizational and management changes, including the redesigning of wider business processes and the development of new business or organizational cultures, are important. Productivity gains do not arise directly from the technology, but from the associated improvements in worker productivity that come from changes in systems, procedures, skills and attitudes. The most recent firm level evidence in the U.S., indicating that IT investments are very effective only when coupled with complementary organizational and managerial changes, has confirmed these finding. The constraints to productivity gains from ICTs have been shown to be more substantial in LDCs (Moussa and Scwhare, 1992; Hanna, Guy and Arnold, 1996). * Third, a large proportion of the benefits of ICTs is not directly measurable - including product and service improvement factors relating to quality and choice. 7 India achieved an average growth rate in IT-related investment of 22.21% per annum between 1984 and 1990. It is suggested that India is now seeing the benefits of such high levels of technological investment in terms of increasing productivity and through the creation of new 'information-based' industries and services (Kraemer and Dedrick, 1998). See US Department, 'Digital Economy 2000', June 2000 4 11. These non-economic variables in the process of ICT diffusion are likely to be of greater importance for countries in a nascent stage of technological development where practitioners are at an early stage of learning by doing. Distribution of benefits 12. While there is a consensus that more effective and efficient use of information and of ICTs can stimulate economic growth and development, concerns remain on how such benefits can be shared among the minority ICT users and the majority non-ICT users. Because of the wide access gap between rich and poor, it is indeed possible that ICTs may reinforce, or widen, existing social and economic inequalities in developing countries. Within this context, there is a consensus that the Government should play a key role in establishing a competitive, private sector led communication market, and by promoting supportive measures to enhance the capabilities for accessing ICTs. The former would involve the liberalization of the telecommunications sector, the privatization of public monopolies, and the creation of an independent regulator with the mandate of reaching universal access in basic communication services (Kenny et al, 2000). Complementary measures, such as reducing taxes, tariffs and trade barriers on imported ICT goods and services would also help improving access and reducing telecommunication costs. The second important role that the Government could play is in promoting ICTs through investments in 'intangibles' such as local training, consulting services, software adaptation, content providers and building the capacity of local importers and suppliers of IT goods and services. These measures are particularly important for micro and small enterprises to benefit from ICTs. 13. It has been suggested that developing country governments can play the role of a technological leader by developing advanced applications and becoming a role model for the more backward 'local' private sector (Pohjola, 1999). Evidence suggests, however, that large bureaucratic institutional settings may not be the best environments within which the benefits of technology can, or should, be demonstrated (Moussa and Schware, 1992; Heeks and Bhatnagar; 1999). Government-led IT projects often suffer from institutional constraints (weak planning and management commitment etc.), human resource constraints, lack of capital and problems of technological adaptation. As discussed in Annex V, which reports the experience of the Government of India, ICTs need to be integrated into wider processes of institutional reform and organizational change. In this respect a more fundamental human resource based transformation is needed to be associated to the technological transformations to increase productivity and performance. The 'optimistic' view of ICTs in development 14. The desire to close the 'digital divide' between developed and developing countries- has led, in recent years, to a number of initiatives by international organizations to harness information and knowledge in less developed countries. The successful experience of East Asian countries has lent support to an 'ICT-led' development thesis implying that poor countries can adopt 'leapfrogging' strategies. According to this view ICTs represent a 'window of opportunity' for developing countries (Hobday, 1997) to progress from a situation of 'zero' or 'limited' technology to widespread adoption of 'sophisticated' technologies, without going through the stages of technological adaptation and learning experienced in developed countries. In Africa, this optimistic view is promoted by the African Information Society Initiative (AISI)9, which in 1996, proposed a 15-year program to provide connectivity 9 It is estimated that there are well over 100 separate agencies and organisations actively promoting ICTs within Africa through funding ICT-based development projects in all areas of development. For a full list of these projects see: http:/www3.sn.apc.oreafrica/proiect.htm 5 and electronic access to information for all citizens of Africa.'0 The program is to be achieved through partnership arrangements between national governments and a wide range of external investors and donors and international development organizations. It recognizes that ICTs can bring wide ranging efficiency gains through the free-flow of information, improved delivery of public services and the development of indigenous human resource capacity to handle ICTs." 15. The optimistic view may be tempered by evidence showing that learning is a critical feature of technological change. The importance of accumulated knowledge and expertise for the growth of firms has been emphasized by a number of studies (Arrow, 1962; Lall, 1980, Bell, 1985). These studies suggest that internal information, and hence knowledge, gained through learning by doing over time, represents the most significant factor in a firm's ability to adapt and change to a new technological/market environment. Within the "optimistic view", the importance of ICTs for addressing the real information needs of the poor has tended to be overstated and there is a danger of the policy debate becoming too strongly focused around the capabilities of emerging technologies. The next section discusses some of the constraints that in many developing countries are making difficult for the poor to access, assess and apply information through ICTs. These include lack of human capabilities, urban and rural location, gender inequalities, affordability and information content. 3. Constraints to the use of ICTs by the poor Lack of human capablities 16. Much of the information exchanged by the poor is organic, transmitted through interpersonal communication, which does not even require literacy (see Table 1.2). Some modem means of communications, such as the telephone, can facilitate interpersonal communication, albeit at a distance. Some others, like radio or television do not require particular abilities for access. Table 1.2 Information Systems Information-based Mode of Means of Technology Storage/ Communication Transmission Electronic Digital Computer Networking Fixed line/mobile telephony Intermediate Analogue TV/Radio Signals Fixed-line telephony Literate Written Word Printed Sources Organic Oral Interpersonal Communication 17. Telephone, television, radio and the printed media require few formal skills and only mother tongue literacy for effective usage. Thus, they are likely to be in greatest demand from the majority of illiterate and semi-literate users for the foreseeable future. By contrast, effective use of e-mail and Internet requires not only literacy but language skills, predominantly the use of English. It also requires technical and computer literacy, i.e. the ability to operate and interact with a computer-based information system (see Table 1.3). 10 The Economic Commission adopted the African Infornation Society Initiative (AISI) for Africa (ECA) Conference of Ministers of Development Planning in 1996 through Resolution 812 (XXXI), as an action framework to build Africa's information and communication infrastructure. For detailed information, see: http://www.bellanet.org/ partners/ aisi/ I The overall aims of the AISI, are also supplemented by specific aims directed at increasing access to inforrnation and ICTs for poor and disenfranchised populations. They specifically support facilitating the establishment of locally-based, low cost and accessible internet services and information content; adopting policies and strategies to increase access to information and communication facilities with priorities in servicing rural and other disenfranchised groups, particularly women. (AISI Goals). 6 Table 1.3. Capability levels required for operation of contrasting user technologies No Basic High literacy/ Computer Technical Literacy Literacy Language skills Literacy competence Oral Communication * Radio * Television * Fixed line Telephone Mobile telephone * Public phone * Newspapers and * Printed sources Fax machine * E-mail * * * Intemet * * * 18. Table 1.3 only indicates the user skills that allow people to interact with the technology. A wider range of 'facilitating skills' (Mansell and Wehn, 1998) are required for the design, installation and maintenance of equipment, such as telecommunications infrastructure and computer networks, for example. These industry skills are in extremely short supply in many developing countries. Direct access to computer-mediated information for the populations of most low development countries is likely to remain with the educated elite unless literacy can be considerably raised through greater participation in formal schooling at post-primary level. The lack of skills and human resources may be the greatest barrier for diffusion of ICTs among the poor. Urbanlrural inequalities 19. For most people in the developing world, the probability of having access to ICTs depends on location. The disparities in access to ICTs between urban and rural communities are indeed startling. The average imbalance, in terms of telephone penetration in most low income countries, for example, is over 10 to I and often is as high as 20 to 1. As described in ChapteT 2, the situation is more acute in most African countries (Kayani & Dymond, 1997). For example, in 1996 it was estimated that in Uganda rural mainlines accounted for only 3% of all mainline (ITU, 1998). Urban centers, likewise, represent an overwhelmingly disproportionate share of Internet users in developing countries (ITU, 1999a). However, the situation is changing: urbanization has been growing significantly, and urban populations are forecasted to expand significantly over the next 15 years. 12 Gender inequalities 20. Gender inequality for access derives from the relatively higher levels of illiteracy among the female population and from the lower level of female participation in the formal economy. In most developing countries the vast majority of the female labor force remains confined to rural areas partaking in predominantly subsistence agriculture, while men tend to dominate in industrial and service-based employment. Women have a number of relative disadvantages compared to men that inhibit their access to ICTs including the competing demands on their time both as homemakers and workers. Affordability 21. For ICTs to become relevant for poverty reduction they have to be affordable. Needed financial resources include those necessary for the 'supply' of a technical infrastructure and those necessary to create 'demand' for user technologies, information and communication services. Over the long term, ICTs are expected to become cheaper, as a 12 The proportion of the total population living in urban areas, within sub-Saharan Africa, has grown from 21% in 1975 to 32.4% in 1997, and is predicted to grow to 43.2% in 2015 (Human Development Report, 1999, p200). 7 result of market liberalization and competition and the introduction of new technology. New developments in the field of software, likewise, may contribute to overcoming some of the problems that poorer countries face in terms of infrastructure"3. In the short run, however, the inability of the poor to raise their standard of living and afford to purchase either collectively or individually ICTs services is a major barrier. Their ability to access new user technologies will depend on the level of growth that can be sustained within individual economies to raise incomes among consumers and generate the necessary investment (public and private) in national networks. Content 22. Surveys of urban and rural households and of micro and small enterprises (see Annex I and II) show that the poor favor and trust information sources close to home and those that are applicable to their existing knowledge base. For small entrepreneurs as well, the most valued sources of information are friends, family, and business networks. By contrast, ICT- mediated information often lacks proximity and the element of 'trust, confidence and security' that is gained through business networking and personal contacts. ICT-mediated information is also not easily applicable to the existing local knowledge base, and small-scale entrepreneurs will generally lack the wider knowledge and experience that is necessary to effectively assess, apply and act upon ICT-mediated information. 4. Conclusion 23. While information and the new communications technologies have a potentially large impact on growth and poverty alleviation, their effective use may be constrained by lack of skills, financial resources and the existence of urban/rural, gender and other inequalities. Thus, policies of access and diffusion of ICT in the context of poverty alleviation strategies should be conceived within the specific environment of institutions (organizations, groups and markets) and political, economic, socio-cultural factors of each country. Information is not a static, equitably available commodity, but a dynamic process of acquisition, application and use. Increased access to ICTs should be considered together with the resources for action that are necessary for making use of information - for example, computer skills and financial resources. The effectiveness of ICTs must be assessed alongside existing information systems, and on how they interact within the pre-existing organic information environments. The information needs of households, communities and enterprises should be understood in advance of proposing ICT-based solutions. Finally, information provision is a two way process, and ICTs may play an equally important role in transmitting and making available information from the poor and disadvantaged (households/ communities/ enterprises) to the wider national or international community. 3 An example is the 'free software developmenf or 'open source' movement, described as Linux after its originator, Linus Torvald of Helsinki. The operating system is a Unix clone which can run on a variety of platforms and supports a range of software most of which is available free of charge. The contribution of software comes from the community of individual software developers. Some of the large companies, such as Sun Microsystems, give the software away because they wish to create an extensive user base for their own software. The users who wish to install Linux do not have to lay out a huge investment Since Linux could be run on a variety of platforms the old and cheap hardware could be put to use (Hunt, 1999), However, it does have a steep learning curve. To gain insight into the system a large amount of time is needed to gain an understanding of Unix. 8 2. ACCESS TO ICTS IN SUB-SAHARAN AFRICA AND SOUTH ASIA: AN EMPIRICAL ANALYSIS 24. This chapter presents an empirical analysis of the level of access and affordability to ICTs resources - telephony, television, radio, computer-based technology14 in the two poorest regions of the world, SSA and South Asia. It shows that ICTs are playing an increasingly important role in the development of modem/urban sectors in many countries in these two regions. The rate of technological diffusion, however, has been extremely uneven. While a minority of higher GDP per capita economies is achieving rapid growth of ICT consumption, the majority of low-income economies are failing to keep pace. Within all countries there is also an extreme divide in the level of ICT usage between rural and urban areas and between different socio-economic groups. 1. Access to telephony - fixed and mobile lines - and to computer based information 25. International comparisons indicate an overall negative picture of access to fixed telecommunications lines for SSA and South Asia, though considerable progress has been achieved in many individual countries, often from a low level. Access to fixed line networks 26.Figure 2.1 shows overall access to fixed line networks, or tele-density (telephone lines per 100 inhabitants), also known as direct exchange lines per 100 inhabitants (DELs), in relation to income per capita."5 SSA and South Asian countries have lower levels of access than comparable low-income economies. Only a few SSA countries have achieved tele-density levels greater than the average for low middle income countries, the most significant being South Africa (11.46 DELs). Main telephone lines in India (a country with one billion people) doubled from 11 million in 1995 to 22 million in 1998 (a number only slightly less than Italy, a country with 55 million people). Tele-density, however, remains at approximately 2 households in 100, the same level as in Pakistan. The situation is worse in rural areas. In India, tele-density in rural areas is estimated to be roughly only 0.3 lines per hundred people. Very low-income countries, typically, have I household in 100 connected to telecommunication networks. Only a few countries in SSA - such as Ghana and Kenya - are expected to reach the benchmark of 1 DEL by the end of the year 2000 (ITU Projections, 1999). 14 Data for selected SSA and South Asia countries are provided in Annex VII. IS GDP per capita is the most reliable 'proxy indicator' of access to ICT resources - particularly fixed line telecommunications. This, more than any other indicator, reflects the ability of an economy to pay for investments in infrastructures, as well as the proportion of individual consumers able to afford ICTs. Other economic and non- economic factors have been introduced in a number of studies (Rodriguez and Wilson, 2000; Mansell and Wehn, 1998), such as distribution of income (a proxy of the overall pattern of demand for services between socio-economic groups and between geographic areas); physical and geographical conditions - which indicates the constraints in extending network access; pre-existing infrastructure - the quality of which will contribute toward the efficiency and effectiveness of service delivery; and market liberalisation - determining the level and nature of competition within the telecommunications market. 9 Flg 2.1 GDP per capita and Main Tolephono Lines per 100 inhab for sub-Saharan African and South Asian Countrlis (1098) 1.5 *Sub-Saharan African tS R*0.o477 Countries 0.6 ~~~~~~~~~~~~~~~m South Asian Countries 02 4 --- Low Income Countries (World Average) -0.5 ) Lower Middle Income (World Average) 0 / -b~~~~~~~~~~~~~~~~~~~inear (Sub-Saharan -1 African Countries) -1.5 -2 GOP per capita, natural log Paterns of frced line network growth during the 1990s 27.During 1995-98 SSA achieved a compound annual growth rate (CAGR) for new main telephone lines of 10.8%. This compares poorly, however, with an average CAGR of 17.1% for all low-income countries. Annual network growth has ranged from 0.4% for Zambia to 31.7% for Ghana. The economies with higher GDP growth experienced a more rapid increase in main telephone lines added, particularly since the mid 90s. A number of low-income countries that in recent years have initiated far-reaching network liberalization policies 16 - Uganda, Guinea, Ghana and Mauritania- have experienced high GDP growth and an upsurge in network growth. These trends have also been mirrored in South Asian economies, with India, Nepal and Sri Lanka all achieving compound annual growth rates in excess of 20%, accompanied by average annual percentage growth rates in GDP of about 5% between 1995 and 1998. However, in all countries the telecommunications sector has not been able to cope with the rising demand for new connections. Fig 2.2 Averaga Annual Growth In GOP and Compound Annusl Growth Rate In Telephone Lines Added. sub-Saharan Africa and South Asia (195-199I) so - z ~~ ~ ~~~~40 U * ~~~~30Sub-Saharan Africa | to -5 5 10~~~~n i 1s 20 ,10 -20 .30 A,r,.#. A....lOrIltb Ia ODP 16 Up until 1998, 20 African countries had installed independent regulators. Shares in the national operators have been privatised in 17 countries, and new independent operators have been issued with licences to provide private networks (primarily mobile cellular). This process of liberalisation and privatisation has continued to expand across the African continent in the past two years. (African Telecommunication Indicators, 1998; ITU, 2000) 10 Access in residential areas 28.Aggregate indicators of teledensity hide the fact that many new lines are added for business and government. More meaningful indicators of households' access are the number of residential mainlines per 100 householdsl7. Fig 2.3 Residential Main Lines per 100 Households according to GDP per capita groupings In Sub-Saharan African Countries (1995 and 1998) 1l 25.t 25.0 25.0 2 .0 1;:: 21.0 flOSSs21*SO S51100Mr ha SrO 101 2.a 3 1 10 a .1 0.0 [h..So USS2S0 005251 to 500 0SS501-1000 M-, th-n US$IO00 GDP per o.pIta Grouping. 29. Data on residential main lines are only available for SSA. Figure 2.3 shows that SSA economies with GDP per capita over US$1000 have a household penetration rate approximately twenty five times greater than economies earning less than US$250 per capita. Moreover, the rate of increase in residential mainlines has been larger in higher income countries than in lower where there has been little overall change between 1995 and 1998. However, there are great variations that reflect geographical conditions, location of populations and the status of nature of infrastructures. Botswana and Mauritius, countries with roughly equivalent GDP per capita, had 14.1 and 77 residential mainlines per 100 households respectively in 1998. Countries with lower, but comparable levels of GDP per capita, also show wide variations. In all low-income countries typical rates of household residential mainline penetration vary between 0.1% and 15.8%. Access to mobile networks 30. The most significant recent area of growth has been the 'mobile revolution'. In South Africa there were 2.5 million cellular subscribers in 1998, or 33% of total telephone subscribers. The total number of mobile cellular subscribers in Africa increased from 652,200 in 1995 to 3,381,300 in 1998 - a CAGR of 73.1% per annum. Mobile subscribers now represent 16.7% of total telephone subscribers across the African continent. Equally dramatic rises have been experienced in the South Asian countries. In Sri Lanka, cellular subscribers represent 25% of total telephone subscribers in 1998. This level of growth has been achieved despite a five times greater cost in monthly subscription rates between a mobile and fixed line connection. India has seen rapid growth reaching 1.2 million mobile subscribers in 1998, or 5.2% of total subscribers (a CAGR of 150% during 1995-98). The mobile revolution in India has been accompanied by some of the cheapest access charges worldwide (less than 50% of the world average, and less than 30% of that charged in South Africa for a monthly basket of calls in US$)1 . Again, growth has been fastest in poorer countries. Bangladesh witnessed a CAGR of 211% between 1995 and 1998 - the highest growth rate of mobile take up of any country in 17 Another important indicator would be access to public payphones. In most low income sub-Saharan Africa and South Asia countries there are less than 5 public telephones per 10,000 people, and the vast majority of these are located in urban areas. IS See ITU, 1999. 11 the world - apart from Senegal (466%). At present growth rates the number of 'global' mobile subscribers is expected to reach 'crossover' and gradually exceed fixed line subscribers some time in the year 2001 (ITU, 1999). 31. The growth of mobile cellular phones has been most rapid in countries with low populations and/or poor existing infrastructure and difficult geographical conditions'9, despite tariffs considerably higher than those for fixed line connections. Mobile phones offer many advantages (ITU, 1999). First, they can be installed rapidly allowing customers to bypass lengthy waiting lists for fixed line connections. And second, pre-paid mobile services offer more flexible payment systems that are suitable for low income users who may not be able to afford regular monthly charges for fixed line services. Overall, the evidence suggests little relationship between the level of mobile penetration and per capita income for African and South Asian countries, unlike the strong relationship exhibited for fixed line network services. Access to computer-based information 32. Intemet penetration shows a pattern similar to that of fixed lines - very little penetration has been experienced within the majority of low-income economies, but extremely rapid take-off within middle income economies. South Africa has an Intemet penetration rate higher than the average for upper middle income countries. The Intemet was introduced into India in 198620. The number of Internet subscribers and users is now estimated at 1 and 2 million respectively. An International Data Corporation (IDC) study estimates that the subscriber base will grow to about 13 million by 2003. As in the case of telephone access, the urban-rural digital divide is acute. Twenty-five state capitals account for more than 80% of Internet access. Estimated Intemet subscribers are only 80,000 in Bangladesh and 100,000 in Pakistan. Internet development is constrained by many factors: the dependence on fixed line network access, which creates an in- built advantage for those countries with better network coverage; the high entry costs for the purchase of equipment, leased lines and the creation and upkeep of professional quality web sites; the inability to have sufficient bandwidth to support fast and efficient up and downloading of information; a low educational level and the lack of computer skills among the population. Table 2.1 Access to the Internet for sub-Saharan Africa and South Asia Countries (1998) Internet Hosts Internet Hosts per 10,000 inhab. Sub-Saharan Africa (Excluding South Africa) Totals Averages USS1000 GDP per capita 4171 7.03 South Africa 144,445 32.6 South Asia (Excluding India) 3788 0.19 India 13,253 0.13 Low Income countries 26,309 0.13 Low-Middle Income countries 382,118 1.64 Upper-Middle Income countries 1,068,356 18.24 High Income countries 42,009,239 450.43 '9 The top ten countries in Africa for mobile penetration are: Rwanda, Guinea, Cote d'lvoire, Lesotho, Uganda, Congo Democratic Republic, Tanzania, Malawi, Madagascar and South Africa - all of which had penetration rates above 2O0/o in 1998. Apart from South Africa, all these countries are classified as low-income, have poor pre-existing infrastructure and difficult geographical conditions. 20 Through the Education and Research Network which links academic institutions. 12 2. Access to non digital communication resources 33. The majority of SSA and South Asian populations depend overwhelmingly on radio and television media, newspapers, as well as face-to-face meetings and word of mouth, for communication and information exchange. Cross country comparisons (Figures 2.4 and 2.5) illustrate the extent to which television and radio usage have grown in poorer SSA and South Asian countries, largely irrespective of growth in GDP per capita. High levels of radio usage are found in all low-income countries, and provide considerable information access benefits, particularly for illiterate populations. The success of the radio is due to affordability and coverage, two criteria that computer-based ICTs are not likely to meet for many years. For example, since the ending of apartheid in South Africa, 100 new local radio stations have been established, while the South African Broadcasting Corporation has itself implemented a new network of local stations.2" These have been particularly effective in promoting the use of African languages within the information media (as opposed to the Internet with its predominantly English content). From 1993 to 1996 Ghana saw an increase from 265,000 to 800,000 television receivers nation-wide, and there are now approximately 5 televisions per 100 inhabitants (ITU, 1998). Television, however, almost everywhere suffers from lack of local content and is increasingly dominated by international programming. Fig 2.4. Access to Information end Communicatlon Resources According to GDP per capita Groupings for sub4taheran African Countries (19K-1)98) 11... T..,.0.. U ...P., ~,0 Lru¢u wunsoo uaso spoo w.rou1ox~loo I 0 -. - R N..p. ... -.70 erOP PlO OlPS arOlwpIn. 34. Overall, levels of access in South Asia to radio and television are lower than in SSA (Fig. 2.5). In India, Bangladesh and Pakistan the number of radios per 100 inhabitants is about half that of SAA countries with equivalent GDP per capita. In India, there are about 10.5 radios per 100 persons. But both radio and television have a wide geographic coverage. Fig 2.5 Access to I -formtton and CRoueatlon Resow ... A ..rdi.g to GDP p.r cepitO Go-pIne. ft0R8.Ut0 A.i. Co.nt.-ts. tl1OO-1055 OR.i,- 00P0., 21 The SABC national Radio network comprises 19 local and national stations that have a combined national adult audience of 15.4 million people. Listeners have been growing substantially in the late 1990s, and output reflects a wide range of issues and interests, not only music and entertainment. 13 3. Ownership of information and communication assets by rural/urban households and by wealth quintiles 35. Fig 2.6 gives an indication of the distribution of information and communication assets between urban and rural areas for 26 low-income SSA and South Asian countries, based on data from Demographic and Health Surveys carried out between 1991 and 1999.22 The figures in Fig. 2.6 represent averages across all the countries surveyed (transportation devices are included for comparative purposes as they also aid communication). Radio is found to be present in 65% of urban households and 38% of rural households. Television is present in 28% of urban households and only 3% of rural ones. Telephone penetration is low, particularly in rural areas. Within the countries surveyed there are differing rates of adoption of user technologies. Radio shows the lowest variation ranging from a low of 30% in Chad to a high of 67% in Senegal. The proportion of households with televisions ranges from 1.6% in Chad and Tanzania to 22% in Ghana and Senegal. Rural ownership of televisions is much lower - between 0.1% in Chad and 9.7% in Ghana. In India 42% of households own radios and in Bangladesh and Pakistan the figure is 35% for each country. About 22%, 13% and 28% of households own television in the three countries respectively. Fig 2.6 Ownership of Communication Assets in sub-Saharan African and South Asian Low Income Countries (1991-1998) 100.00 90.00 80.00 70.00 65.40 5 60.00 i so.oo * i Urban 40.00 37.96 ERural 30.00 28.02 23.73 20.00 16.96 10.31 10.00 3.267NM .4 2094 0MMM II02.8 0.00 ME R.d. TO.1"0o Tefrphon BRJoye MO9o.d. Pst. mr Auass Ownership of information and communication assets by wealth quintiles 36. Figures 2.7 and 2.8 report ownership of assets in the two regions by wealth quintile. In SSA telephone ownership is confined exclusively to the top 10-12% of the richest quintile (Figure 2.7). There is also evidence that many households give higher priority to the ownership of other information and communication tools such as television or means of transport such as bicycles. Radio ownership varies from 10.7% to 88% from the poorest to the richest quintiles. Television ownership occurs across all groups but at a lower level than radio ownership. 22 The DHS were carried out in 26 African and South Asian low-income countries (GDP per capita less than US$1000: 1997) between 1991 and 1999. African countries are: Mozambique, Chad, Niger, Burkina Faso, Madagascar, Mali, Malawi, Tanzaia, Central African Republic, Uganda, Togo, Kenya, Ghana, Benin, Comoros, Zambia, Senegal, Nigeria, Carneroon, Zimbabwe, Cote d'lvoire, Namilbia. South Asian countries are: India, Bangladesh, Nepal, Pakistan. DHS data are supplied by the HNPlPoverty Thematic Group of the World Bank httr)-://www.worldbank.org/inoverty/health/data/index.htm 14 Fig 2. 7 Ownership of Information and Communication Assets by Wealth Quintiles In Low Income sub-Saha ran African CountrIes (1990-1998) 100.0 900.0 $0.0 l OTie.phonm Ownerhip 70.0, eCarOwnwship l 80.0 U Motorcyvls Ownrsuhip so.0 - 0 BaWdi Ownership 40.0 *Electrct Supply (% 200 1 1 1 J Si s _ 1 *~~~~~~~~~~~~Rodioown nhp Poorest Second Middle Fourth Richest Quintib 37. For South Asian countries (Figure 2.8), telephone ownership was unavailable within the data set. For radio, the distribution of ownership between wealth quintiles is similar to SSA, but at a lower level within all quintiles. For television, ownership is high amongst the richest 20%, but non-existent amongst the bottom 40%. The most noticeable contrast between SSA and South Asia relates to the far higher level of electricity supply to households in South Asia - a necessary 23 service provision for utilizing computer-based systems' . Fig 2.1 Ownerhip of Infornmaon and Comwunicaion Assets by Wealth Quintire for Low oconw South Asian Countres (1991-1997) 100.0 soo.0 8000 iCarOwnershlip 70.0 *Mftoryde0OwnrwsNp co.oICp(io Owrrshlp 50 IRsdo wnership 40.0 0'rlTet.*0l Owneship 30.0 BEMl SLq (% 20 0 Poost Second Modb FoeSt Riwot 38. Overall, DHS surveys suggest that the poorest 20% of households, which are mostly located in rural areas, have low or no access to telephone lines and even to electricity and own only the cheapest communication devices, such as radio. This emphasizes the importance of not overlooking oral information and personal networking. Oral tradition and informal communication systems for many centuries have been part of the traditional information structure - particularly in rural areas of countries in SSA and South Asia for both households and small enterprises. Informal 'verbally transmitted' information is seen to play a greater role in less 23 On average, some 46% of households surveyed in South Asia have electricity, as compared with less than 17% in SSA. 15 developed economies because of a lack of a modem communications infrastructure; an overemphasis on confidentiality and secrecy, particularly within state-run institutions; and the lack of capacity (education and skills) to assimilate formalized information. 4. Affordability of Information and Communication resources 39. At present the costs of fixed line access in all SSA and South Asian countries are much higher than those charged in developed countries, and within the poorest countries, the charges tend to be the greatest. Figure 2.9 shows total fixed line telephone tariffs for a single year's usage according to GDP per capita for SSA Africa and South Asia countries.24 The data indicates that many very low-income countries in these two regions are paying charges far in excess of the average cost for lower and even upper middle income countries. Fig 2.5 Total Fixed line Telophone Taitff (1st Year Usage) end OD P per capita, sub- Sahara,, Africa and South Asia (1555) 3 Ioo *a 'Sub-Sah.r- Africa C utnile. 1:t U05 * South Aol. Co unt.i.. Lo..nor C-outri. (Worh s z a R .rs U pp or AM Id dlo Iom. (W orld 211 ~~~~~~~~~~*U P:r.r M I1ddl lut,. (World *:sO+; * vC-untrer ubS)hrn f 't **~ sop p.,.. pIt.Li.. (.b5..,. M 40. Figure 2.10 compares 'fixed line' and 'cellular' telephone tariffs as a percentage of GDP per capita across the specified country income groupings for SSA. The average total telephone tariff for a year's fixed line usage ranges from an average of 97% of GDP per capita for very low- income countries, to an average of 7.2% for countries with GDP per capita greater than US$1000. This figure is still high, however, in comparison with the average for upper middle income countries (5.5% of GDP per capita) and high-income countries (1.4% of GDP per capita). Mobile access (based on 1999 figures) is substantially more expensive. In very low-income countries, usage over a year would typically cost an amount more than double the GDP per capita. Fig 2.10 Teicommunication Tariffs as % of GDP per capita In Sub-Saharan Africa (1998) 35F L0 _ 100.00 800 #,ffi00o _~~~~~~~O.0 21b000 _ b *3S~~~~~~~~~~~~~~~~~~~~~~~~~040 L.-szr U8110 uszS2 H usssa-oaa looo uO.uxl G0P P., .p.P. or-pl,n. 24 A total cost for a single year's usage of a fixed line residential service (total telephone tariff) is obtained by adding the residential connection charge, the monthly subscription charges and local call charges based on a conservative usage rate of two local (3 minute) calls per day. 16 41. Typical monthly subscriptions in less developed countries range from US$5.0 in India ($14.30 for a 100 minute basket of local calls) to US$10 in South Africa ($48.46 for a 100 minute basket of local calls). In most less developed African countries, prices levied tend to be in the higher range. This is particularly the case in countries that lack competition, have licensed single operators, or have small markets with low numbers of subscribers. Dial up Intemet access cost varies greatly among countries. Table 2.2 Comparative ICT pricing structures for selected SSA and South Asian countries 1998) Country Residential Monthly Cbarge for Subscription Dial up Telephone Line Local Call As a % of Internet Access Connection Subscription (USS per 3 min) GDP per capita Cost Charge((US$) (US$) US$ per Month SAfrica 35 10.0 0.07 4.3 10-20 Swaziland 29 2.5 - 2.4 14 Botswana 55 4.2 0.03 1.6 10-20 Ghana 173 1.1 0.09 3.9 30 Kenya 36 4.3 0.06 15.0 15-20 Sudan 51 2.1 0.02 8.4 200+ Zambia 30 2.7 0.06 2.0 25 Uganda 137 8.1 0.18 21.7 50 Mali 86 3.2 0.14 16.8 500+ Mozambique 87 5.1 0.04 71.1 25 Burundi 5 0.9 0.03 9.0 150 Burkina Faso 37 3.5 0.10 21.6 13* India 22 5.0 0.02 Nepal 35 3.0 0.02 Bangladesh 256 4.0 0.04 Pakistan 92 3.0 0.03 Sri Lanka 220 2.0 0.03 Middle Income Countries (Average) 112 6.5 0.07 2.8 High Income Countries (Av.) 112 11.6 0.13 0.7 Source: http://www3.sn.apc.org/africa/ World Telecommunication Development Report, ITU, 1999. Ensuring universal access 42. Evidence suggest that in developing countries, the proportion of income that the poor can afford to allocate to telecommunication services vary between 0.5% and 2% of their income, which in most countries is below the entry level tariff25 (Dymond et al., 2000)). Rural poor households in most African and South Asian countries will not have the capacity to fully pay for communication services for many years. 5. User skills and capacities for access 43. As discussed in Chapter 1, user technologies require different levels of education and literacy, and specific user skills. Statistics relating to levels of illiteracy give an approximate guide to the ability of populations to successfully interface with information and communication systems. In Africa, (see table VIII.2 in Annex VII.) illiteracy among youths (age 15-24) has dropped considerably in recent years, to below 10% in some southem African countries. In countries such as Namibia, Botswana and Lesotho, female youth illiteracy has fallen to a considerably lower level than male illiteracy. For SSA as a whole, the female youth illiteracy rate is estimated at 29%, having fallen from 55% in 1970. Although this is an achievement, the fact remains that between 70 and 100 million young women in SSA presently have no potential to interact with information systems requiring literacy skills. 25 Which includes one year's monthly rental charges, a percentage of connection fee and a typical number of local calls. 17 44. In South Asian countries illiteracy levels are considerably greater. Youth illiteracy rates in Pakistan, Bangladesh and Nepal are 53% or higher for females and 25% or higher for males. In India in 1998 illiteracy rate for young females was 37% and for young males it was 22%. Figure 2.8 reports findings from the DHS data. It shows that less than 25% of non-educated women in the 26 surveyed countries in SSA and South Asia neither listen to radio nor watch television, while newsprint is of virtually no relevance. The relatively high level of access to television and radio by those with no education reinforces the importance of not ignoring these information tools in poverty alleviation strategies. The statistics also shows how levels of access to newspapers rise in line with higher levels of education. Fig 2.8 Level of Access of Female Head of Household to Mass Media in sub-Saharan African and South Asian Low Income Countries by Educational Level (1991-1998) 100.00- .00.00- I2 90.00- 78000 U' 70,00 e7.13 E noon *No education K 50,00 9 ~~~~~~~~~~~~~42.80 *Prlfmary -40.00 OSecondary or 30.00 286129 7827.93 Ngher 20.00 - 18.21 1A a. 10.OD 0.00 Reeds ranoep.po Wetdles tAt-s L.8t0ensS edi dfty JOw We to -e My0. Acess to Mass Media 45. Higher level skills (language and computer literacy) are only likely to be acquired by those who have completed a formal education, primarily those who have attained literate and numerical skills at a secondary level. Those with a primary education will still be able to interact with computer-based information systems, but their level of interaction is likely to be considerably reduced. Surveys suggest, not unsurprisingly, that at present Internetle-mail users in sub-Saharan Africa and South Asia belong to the educational elite, are male, young and earn the highest incomes. 6. Conclusions 46. SSA and South Asia are the regions with the lowest access to ICT resources. As in the rest of the world, penetration rates of ICTs that depend upon fixed line networks tend to have a strong relationship with overall levels of income. Thus, within the two regions, countries with higher GDP per capita are those with more extended fixed line networks. However, the highest growth rates in mainlines are in countries - both low and middle income - which have adopted policies of liberalization and privatization of their networks. Internet penetration, which depends on fixed line networks, shows a similar pattern of growth. Very little penetration has been experienced within the majority of very low and low-income economies, but extremely rapid take-off within middle income economies. 47. Information and communication technologies based on wireless networks, have expanded rapidly across SSA and South Asia, largely irrespective of per capita income level. Very low and low-income countries display the highest penetration rates of mobile cellular subscribers, despite the considerably higher tariffs than those levied on fixed line users. Thus far, new networks have 18 filled a backlog of demand from better-off - predominantly urban-based business users - who have been prepared to pay a premium price for immediate network access, greater functionality and efficiency of communication systems. Penetration of radio and TV shows less association with income. Radio continues to be the primary 'information technology' for both urban and rural populations in all SSA and South Asian countries. 48. Cross-country comparisons give an overview of the digital divide in SSA and South Asia. Within each economy there exists a wide gulf of access and affordability for differing groups. The first critical access-gap exists between urban and rural areas, with rural mainlines accounting for less than 10% of all mainlines in most low income countries. Data from DHS taken during the period 1991-99 from 26 low income countries in SSA and South Asia indicates that about 5% of urban households have telephone ownership compared with 0.18% of rural - or 3.4% of the total. By contrast, radio ownership reaches 65% of all urban households and 40% of rural. TV is present in 28% of urban households and 3% of rural. But the overall coverage of radio and TV is much higher than the figures suggest, given their suitability for communal usage. The second access gap exist between the poor and the better off. Again, DHS data suggests that telephone ownership is confined to only some 12% of the richest quintile in SSA and South Asia. The picture is different for radio ownership. Radio is owned by about 10 % of the 20% poorest households and by about 40% of households in the second poorest quintile. Radio ownership is almost 90% among the richest. Television ownership rates are much lower at all levels of wealth. 49. Evidence concerning affordability suggests that liberalization policies carried out in the majority of countries surveyed have yet to create the necessary market conditions for affordable telecommunications network access. The prices for fixed line network access in many SSA and south Asia countries are still well ahead of the world average for low-income countries. Taken overall, fixed line telecommunication tariffs (based on the total cost of 1 year usage) would represent approximately 112% of GDP per capita in very low-income economies (compared with 8.25% of GDP per capita in middle-income economies). In high-income economies the figure stands at approximately 2%. Tariffs for mobile cellular access represent a considerably higher proportion of GDP per capita at present. 19 3. LEARNING FROM CASE STUDIES AND CONCLUSIONS 50. This chapter summarizes the empirical evidence presented in the Annexes I-VI on the adoption and utilization of ICTs in a number of South Asian and African countries. It brings together the lessons from each individual case to suggest some policy principles to improve the development impact of ICT applications. The empirical evidence includes three surveys commissioned expressly for this paper. Two of these surveys explore the information needs and access to ICTs by urban and rural poor households in Nepal and India respectively. The third surveys small, micro and medium enterprises (SMEEs) in Tanzania. The remaining evidence is taken from a selection of published sources that explores access to ICTs in rural areas. It looks at commercially oriented networks (South Africa and Ghana, India and Bangladesh), externally funded initiatives (Africa and India), private-public partnerships (India), and at the role of ICTs in improving Government's development impact in rural areas in India. 1. Surveys of information needs and access to ICTs in poor households 51. To investigate information needs and access to ICTs by poor urban and rural households, two surveys were undertaken in 2000 in Nepal and India (Annex I). In Nepal, interviews were conducted with 150 urban households located in Katmandu and 100 rural households located in villages at the outskirts of the capital. In India interviews were conducted with 250 low-income households in the slums of Delhi and Mumbai and 150 poor households in villages located in the states of Uttar Pradesh, West Bengal and Andhra Pradesh. The villages were chosen in areas where, according to official information, poverty rates are high or higher than the national average and where access to basic services is low. None of the rural villages in Nepal and India have access to electricity. Households were asked about the sources of infornation, the type of information received and whether this infonnation was needed and effective. 52. Results indicate that the poor rely mostly on informal networks for their information needs. Family, friends, and village and local leaders are the most trusted sources of information. By contrast, formal sources such as NGOs, newspapers or schoolteachers, are rated low and in both countries politicians are the least trusted. The poor have no access to digital information and the new forms of ICTs (internet, fax and computers) are accessible to some degree by only 2% of low income households, mainly in urban areas in both countries. By contrast, the poor have access to radio, and increasingly, to television in urban areas. Interviews indicated that keeping abreast of current events and entertainment are important priorities for radio and TV listeners/viewers. But the interviews revealed an unmet demand for additional information regarding employment opportunities and training (in urban areas) and agricultural practices, markets and prices (in rural areas). Between a quarter and half of the households surveyed claim to have experienced a change in their consumption pattern, incomes, or health by applying information received through both informal and formal sources. Overall, the poor are optimistic about the potential of ICTs for improving their socio-economic condition. 20 2. Studies of information needs and ICT impact in small and medium enterprises 53. Evidence is presented on the findings from two studies, the first conducted in 1999 in Botswana2' and the second in 2000 in Tanzania (Annex II). About 100 SMMEs were interviewed in each country on information needs and access to ICTs using the same methodology and questionnaires. Findings are remarkably similar, the main difference being that ICT use is more widespread in Botswana than in Tanzania. In both countries computer-based information systems (including e-mail and Internet access) are utilized mainly by enterprises in the tourism and services sectors and in foreign-owned enterprises. Firms catering to the domestic market tend to utilize more basic ICTs. 54. The evidence shows that in order to satisfy their information needs all SMEEs rely greatly on informal networks of family and friends, the local business community and own knowledge and experience. Direct contact with customers through face-to-face meetings is considered the single most effective method of business communication. The surveys reveal that SMEEs believe that there is a substantial information gap between desired and actual information. There is a desire among these firms for more information related to the supply of labor and other inputs, market opportunities and availability and sources of finance. The latter appears to be the key information need of SMMEs, indicating either an effective lack of sources of finance, or the inability of entrepreneurs to access critical information regarding alternative sources of finance and markets. 55. Would the availability of more ICTs help SMEEs meet their information needs and improve productivity? The findings from these surveys would suggest that more technology is not what firms need. Indeed the level of computer usage is often very low among firms that own a computer. What is most needed in both countries is raising the skills and overall capacity of SMEEs to access and use new technologies, as well as to upgrade their managerial capabilities and skills, marketing, financial services and credit arrangements. Many entrepreneurs simply do not know how to make proper use of computers. In turn, this may be the result of a gap that exists between the design conceptions of much business software and the organizational realities of many small enterprises (Heeks and Bhatnagar, 1999). This design-reality gap is a key priority that needs to be addressed, together with issues of related software skills within the country. 56. There is also a question of the relevance of much of the 'content' currently posted on the Internet for local firms. This would seem to indicate the potential need for information centers or intermediaries, whether private or public, to fill the 'content gap'. They would not only provide relevant information for SMEES, but could also provide training with respect to computer usage and information access. 3. Findings from private sector involvement in extending access to ICTs 57. Findings from many countries (examples from South Africa, Ghana, Bangladesh and India are provided in Annexes III & IV) shows the effectiveness of the private sector in extending telecommunications services to rural populations and to the poor. There are many reasons for this development. First, new technologies such as low cost wireless solutions (cellular phones, multi- access radios, etc.) and satellite systems, are becoming commercially feasible options for rural areas and distant regions. Second, pro-private sector policy environments, recently adopted by many countries, have introduced liberalization and competition and have removed the barriers to 26 R. Duncombe (1999), 'The role of Information and Communication Technology for SMMEs Development in Botswana', Institute for Development Policy and Management, University of Manchaster. 21 market participation27. Within this environment, many countries have introduced specific interventions to attain universal access through the private sector. This includes service requirements ensuring a minimum level of access throughout the country through payphones, phone shops and, more recently, telecenters. 58. After deregulating its telecoms markets, Ghana experienced a rapid increase in the provision of payphones, digital cellular phone networks, and new Internet access services. In addition, more than a thousand small/micro businesses spread throughout the country offering basic communications such as telephone, fax, typing and other, predominantly, business-related services. Private sector run payphones and phone shops have multiplied in South Africa, the result of built-in obligations to operators, and are financially sustainable. In India, a large number (approximately 300,000) of small-scale and minuscule entrepreneurs (many are women) have successfully set up businesses in the last five years running telephone booths for collective use. These small street shops offer public facilities for long-distance calls and many are now being upgraded to offer email and internet facilities. 59. Private sector experiences in extending communication services suggest two lessons. First, in a deregulated and sufficiently competitive market, services in rural areas can be effectively targeted to attract a wide range of users (e.g., local government, NGOs, public health and education services and tourist facilities that have the ability to pay. Second, even in situations where private sector services are not commercially viable, often a small subsidy is sufficient to make them financially sustainable. Poor potential users in rural areas are often willing to pay a certain price to reap the social and economic benefits of local access. 60. Annex VI covers the Grameen Phone experiment in Bangladesh at length, perhaps the most single most successful experience of the private sector in reaching the poor through telecom services. The experiment consisted in selling (or leasing) cellular phones to women members of the Grameen Bank who would then re-sell the telephone services to villagers. Formal evaluation studies showed that economic factors were most common reason for the poor to make calls (one study reported that 42% of calls were related to remittances from abroad or large domestic cities). Phone users were prepared to spend between 2% and 12% of household income on calls related to remittances. The experiment is relevant for several reasons. First, it shows that providing telecommunication services to rural communities can be profitable. Second, it indicates that the poor can reap real economic benefits from access to information regarding prices, markets, and business opportunities. And finally, it suggests that there are added benefits if the women are targeted (in terms of an improved social status). 4. Findings from Government-led ICT applications for rural development in India 61. Annex V discusses a variety of Government led initiatives aimed at rural development. They are all from India - a country with a long experience in applying information technology to rural development. Interventions range from computerization schemes to installation of satellite communications and can be broadly categorized into the following types: i) improving planning and monitoring; ii) improving services to citizens; and iii) empowering citizens to access information and knowledge. 27 See A. Dymond, N. Juntunen and J. Navas-Sabater (2000), 'Telecommunications and information services for the poor', World Bank Working Paper. 22 ICT initiatives to improve planning and monitoring 62. Initiatives in India have included the computerization of district offices and the provision of PCs in the rural development agencies to improve the monitoring of the Integrated Rural Developmental Program. These experiences were partially successful. While computerization was achieved in the majority of districts, there was only a marginal impact on the efficiency of the administration, as well as on the efficiency of the rural programs, showing that the effort required to push administrative reforms through the use of information technology was grossly underestimated. This is partly because the approach was completely centralized (including the provision of software). As a result, in many districts computers were not utilized or under- utilized. Second, while the motivation for change was strong at the central level, this motivation was not shared at the district level, nor was there an interest in improving the administration of development programs. 63. In recent years several state governments have become active in promoting the use of information technology and the administrative culture is changing. The recent introduction of 600 multipurpose electronic and computer centers (MEXs) at various levels of administration with the purpose of training personnel and servicing computers (hardware and software) has been very successful. About two thirds of MEXs are located in rural areas, and a large variety of training and servicing modules have been prepared. Most centers are financially viable and the program has proved to be particularly useful in supporting rural centers. Improving services delivery to citizens 64. Information technology has been used by the government of India to improve the delivery of public services to citizens. One of the most successful projects is the Computer-aided Administration of Registration Department (CARD), implemented in two hundred locations in Andhra Pradesh. The CARD project has made the system of registering property deals more transparent and it has reduced both the time for registration (from a few weeks to one day) and the number of clerical staff needed (sale documents are now scanned rather than copied by hand). Much of the success was due to the strong leadership provided by the commissioner. The CARD application exemplifies the kind of positive impact on service delivery that can be achieved through the use of information technology. Providing information and training to rural areas 65. Satellite communications for broadcast and interactive training have been successfully carried out in India since the mid-1970s. Initiatives have shown that community television can be successfully deployed, operated and maintained even in remote areas. Within this context, satellite communication is probably the most cost effective solution compared with school outreach, training institutions and government agencies though large investments are required at the outset to create the infrastructure. Evaluation reports show substantial social benefits, in the areas of health, hygiene and nutrition, particularly for female viewers. Interactive training has also been effective, particularly when field staff are scattered over large areas. The recently launched Warana Project covering 70 villages around the river Warana in Maharashtra is another example of how to use infornation technology to provide agricultural, medical, and other educational information to remote villages. State of the art infrastructure is used to introduce Internet access in the existing co-operative societies by establishing networked "facilitation booths" in the villages. 66. These projects have reached their objectives and evaluation studies have shown that there have been social benefits, though a quantiftcation of these benefits has not taken place. However, 23 they require large up frdnt financial investment and as many of the services are free it is difficult to estimate a monetary value that users would accept. Therefore, financial sustainability may become an issue in the future. Exploiting Public-private partnership 67. India also provides successful examples of partnership between the public and private sectors. One example is the application of information technology to the management of milk collection of the cooperative societies in India. A system for testing and weighing milk was created and tested in a national, public laboratory and was subsequently marketed by two private firms. The automation of milk collection and payment procedures to farmers has improved the efficiency of the process and introduced transparency. Producers trust the objectivity of the weighing and quality determination procedures, feel they are getting a fair price for their milk, and have an incentive to produce more. Providing dairy cooperatives Internet access has also improved the delivery of veterinary services (insemination and disease control, for example) and improved breeding programs (Bhatnagar, 2000). 5. Externally funded initiatives to improve access to information and ICTs by the poor 68. In many poor countries the market approach to ICT provision would exclude the very poor who are unable to pay for services. In recent years many non-commercial community-based initiatives supported through external funding (i.e., NGOs, international development centers, or bilateral donors) have flourished throughout Africa28 and South Asia with the purpose of extending ICT access to the poor. In many instances external funds have supported government, non-government and private organizations working with the poor such as enterprise development institutions, educational establishments, clinics and health centers, agricultural extension services, etc. 69. There are many potential benefits arising from the availability of free financing and expertise. For instance even those who cannot afford it can have access to information that is likely to improve their health, education and productivity. The impact can be greater if access to information is accompanied by some basic training in ICTs and it may also generate employment in electronic repair centers and information handling services. Local needs can be served with a variety of communication technologies (radio, second hand computers, etc.) by providing information regarding availability of medicines, credit, transport services, weather, water risks and other educational materials. A successful example is represented by the work of the Swaminathan Research Foundation in South India (see Annex V.) which provides locally relevant information on health, education prices and markets. Externally funded initiatives present several limitations. * Cost effectiveness. Most projects for extending connectivity have to overcome considerable technical obstacles, utilizing for example satellite technology to establish effective communications links. These tend to be relatively expensive technological solutions for providing relatively limited access facilities. * Sustainability. Donor-funded programs have tended to provide initial investrnent capital rather than maintenance and running costs. For local organizations with 28 In recent years a large amount of funding for increased networking through ICTs has been provided through the Africa Link Programme, funded by USAID to agricultural extension and research organizations. Evaluation reports of these initiatives made in Uganda, Zambia, Tanzania and Ethiopia, from 1997 to 1998 echo many of the concerns reported in the section. Further details are available at: http://www.info.usaid.gov/regions/afr/alnk/reports/ 24 limited finances, the follow-on investment costs for updating and expanding the technology, and the recurrent costs associated with maintenance have remained largely unaffordable. * Technical support. Many locations outside capital cities lack any technical support that can often result in a significant amount of downtime when technical problems arise. * Financial opportunity costs. These are of special concern in light of the high costs of access experienced within African and South Asian countries, and are a consequence of overpriced equipment, higher financing costs, institutional inefficiencies and the extra infrastructure costs associated with extending access to remote areas. (International Telecoms Review, 1998). * Information and technology opportunity costs. Rapid promotion of ICTs may supplant other non-digital technologies that offer greater coverage and accessibility to the poor, such as radio broadcasting which has a proven and successful record for information dissemination. This risk should, however, be balanced against the potential benefits that, the Internet for example, offers to existing information providers. A final issue concerns the ability of intermediaries to ensure that the benefits associated with ICTs reach the poor - and therefore the efficacy of donor funded initiatives to support them. A case study of an NGO operating in Botswana serves to illustrate a number of concerns. Box 3.1: Communications for Rural Enterprise Support - Case Study of CORDE (Botswana) CORDE (Cooperation for Research, Development and Education) is an NGO offering rural enterprise support in Botswana. It offers primarily enterprise development services and training programs for micro/small-scale manufacturing businesses. It has a central headquarters in Gaborone that supports a membership organization of 47 small enterprises and coops. Members are supported through a network of full-time trainers that are stationed in different regions. The outreach workers are themselves acting as intermediaries between the members, the central organization, and often the market (customers and suppliers, for example), government agencies and the wider donor community. For such an organization, operating in a country the size of France, communication has been a major issue and constraint. Recently all officers were issued with mobile phones to link into the rapidly developing cellular network that covers most of the densely populated eastern part of the country. The use of the mobile phones created considerably improved communication between the headquarters and the outreach workers. Communications to the members, however, continued as it always had. Only 10% of members could be contacted by phone, and messages delivered through payphones (which were often out of order) were too unreliable. Members (mostly remote small enterprises) continued to prefer, and to depend upon, face-to- face communications and would only respond to requests for information or attendance at meetings/workshops if they were informed personally. 70. The experience of CORDE in Botswana illustrates the difficulties in reaching the poor. Not only were poor micro-business owners unable to afford telephone connections, but they were mistrustful of information from outside their personal network. This reflects a growing lack of compatibility between the information systems used by intermediaries and those used by member organizations. The compatibility gap in information systems between NGOs and recipient populations is widespread and has both positive and negative aspects. On the one hand utilization of emerging technologies is causing local organizations to become more remote from recipient populations. On the other hand, these organizations are improving their effectiveness in 25 interacting with the modem sector. Connectivity through ICTs (Internet, e-mail, mobile communications, etc.) may enhance networking at an intermediary and support level, but may inhibit network linkages between intermediaries and recipient populations. 6. Policy Implications of the Information Revolution for Poverty and Development 71. Case studies in this report suggest the need to consider information technology as a key resource essential to development goals. However, the principal lesson is that technology is not a goal in itself, but a means of achieving development goals. Thus, the modernization of information systems, and the introduction of ICTs, should be fully integrated into the process of organizational and societal change, driven not by the technology itself, but by real needs for economic, social and institutional development. Within this approach, a number of principles are suggested to make policy towards ICTs more relevant to the needs of the poor. Balance Policy Goals 72. There is a need to recognize the divergent interests of different sections of society. For instance, commercial and large organizations' desire for fast and effective information and communication systems, has to be balanced with the requirements of the rural poor and the small/micro enterprises sector for appropriate information systems that reflect the available resource endowments (financial, human and technical) of the user population (Stewart, 1978; Clark, 1985). This might mean favoring basic communication tools such as telephony, newsprint and radio, over digital computer-based systems. The latter should be introduced where they are likely to be cost-effective, alongside or supporting existing means of communicating information. Maximize Benefts - Minimize Risks 73. Significant 'financial opportunity costs' for poverty alleviation strategies may arise if large amounts of limited resources are diverted to ICT-related investments that might not have a direct impact on the needs of the poor. The risk may be compounded by the use of inappropriate applications or the inability to effectively manage technological projects, leading to investment resources being wasted. In poor countries, studies have highlighted the extent to which an 'enabling environment for successful ICT diffusion' is presently lacking. Areas of deficiency include local IT supply industries, domestic demand/user involvement, and technical and managerial capabilities (Moussa and Schware, 1992). Thus, there is a danger of 'technological opportunity costs.' Where funds are limited, introducing digital information systems may supplant existing systems that may be more appropriate and more cost effective in terms of access and coverage. 74. Over-concentration on technological solutions can have the effect of drawing attention away from the underlying determinants of poverty such as inadequate health and educational facilities. It is therefore important to minimize the 'opportunity cost' risks associated with ICT investments. This can be achieved, for example, by facilitating the early involvement of users in policy planning, as well as the participation and consensus from users and testing of new information systems in advance of widespread application. Building User-CenteredStrategies 75. Programs for ICT adoption and implementation in many developing countries have tended to take a 'top-down' approach. For example, the emphasis on projects implementing ICT systems for administrative purposes within central government, and the associated establishment of training facilities and programs for civil servants and government officers. The poor will benefit from improved service provision arising from more effective government information 26 systems or from the economic and employmnent benefits resulting from technology-led economic growth and development. Evidence shows, however, that there may be a considerable time lag before such productivity and efficiency gains can be realized, given widespread institutional, organizational and managerial constraints. It is essential, therefore, that policy encourages a wider community of users to adapt ICTs for the benefit of their enterprises and organizations, where they are deemed appropriate. Building ICT Capabilities 76. Much ICT supply industry, in most developing countries (with the exception of India), is dominated by foreign-owned companies and subsidiaries, catering primarily for the corporate and government sectors. Policies should encourage local small-scale businesses to both provide information and to build capacity and skills as ICT resource providers. Locally based private sector support structures are more likely to create the conditions that foster interaction between community members, and also encourage the production of local content. Supporting Community-based Intermediaries - Giving the Poor a Voice 77. The expansion of private sector ICT capability is unlikely to sufficiently assist the very poor. NGOs and other donor funded support structures will continue to play a role in reaching the poor and giving the poor a voice. However, for ICTs to be made relevant to peoples' lives, initiatives should become more community based. In this respect, small community-run organizations may provide an appropriate vehicle for extending ICT utilization into the wider community. Such organizations provide the advantages of source proximity, trust and locally- contextualised knowledge. Creation of local content should be promoted through local institutions such as schools, local government or businesses. The cost effectiveness and potential for coverage, of local digital content must, however, be weighed against the advantages of pre- existing information systems and communication tools. Building Local Capacity through Education and Training 78. Building ICT resources over the long term will require school-based education, and vocational training for young people. An integrated policy for increased ICT investment within education (from primary, secondary, tertiary to higher) including teacher training and technical support, can ensure effective long term diffusion and local adaptation of ICTs within developing countries. New skills among the young will feed through into all sectors of society, including the wider enterprise sector, teachers and trainers and the creation of IT professionals and enthusiasts. This will produce the necessary ICT support skills that will be essential in the wider economy. 79. Building ICT skills from below, through education and training, would form part of a long-term strategy for transforming the skills base of developing countries. This will include greater financial resources, adaptation of the curricula, language constraints, and effective management of technology based programs - both technical and pedagogical. Computer-based learning should represent an integral part of a wider educational experience, involving extra resources and time. 27 ANNEXES Annex I. Information needs and access to ICTs by poor urban and rural households * Nepal * India Annex II. Information needs and access to ICTs by small, micro and medium enterprises * Botswana * Tanzania Annex III. Communal access models * South Africa Annex IV. Private sector led ICT development in rural areas * Ghana Annex V. ICT applications for rural development in India - Government led initiatives to improve the administration of rural programs - Government led initiatives to increase communication access, education and training in rural areas * Public private partnership: the application of information technology in Milk Collection Cooperatives * Donor/NGOs financed initiatives: the Village Information Shops in Pondicheri Annex VI. The Grameen Village Pay Phone Initiative in Bangladesh Annex VII. Comparative indicators Annex VIII. List of SSA and South Asian countries according to GDP per capita groupings Annex IX. The diffusion of ICT among the rich and the poor 28 Annex I Information needs and access to ICTs by poor urban and rural households29 Nepal Nepal is a small country in South Asia with an extremely high level of poverty. In rural areas about 44% of the population is classified as poor and in urban areas the figure is 23%. Social indicators are also low, not only in comparison with other countries of South Asia, but also with respect to the rest of the developing world. About 77% of the land area in Nepal is covered by hills and mountains and many of the villages (including the ones surveyed) are very difficult to reach. In some instances the only access is via a non-vehicular suspension bridge. Not surprising for a country with this magnitude of unmet basic needs, the majority of the poor have extremely limited contact with "modem" information transmitters such as phone, television, fax and computers. Within this context, a program of empirical research was undertaken during the first half of 2000 to try to understand what kinds of information could benefit the poor as part of an overall strategy to help alleviate poverty. Research A total of 250 randomly chosen urban and rural households were surveyed by questionnaire and then subsequently interviewed during the first half of 2000. The sample households were chosen randomly from areas where the incidence of poverty is very high, based on Living Standard Survey criteria and Survey Reports of the Government of Nepal. Characteristics of the Rural Poor For the rural poor, 100 households in the villages of Adamtar, Kumpura, Hatisada, were surveyed in the Dhading district of Nepal. These villages are very remote and do not have electricity or health facilities, although water is available. Almost 90% of households surveyed had access to radio, whereas the majority did not have access to telephone. There is a telephone booth on the main road, but the terrain is hilly and access is difficult. The survey revealed that the populations of these villages relied on traditional means of communication: friends, family, village heads. With respect to modern communication transmitters, only radio was routinely utilised, with 71% of respondents believing it was an effective method of communication. Effectiveness of computers, on the other hand, was limited to just 1% of respondents. In terms of the types of information most frequently received, family welfare/women' development, news and politics and agricultural development were sited in that order. When questioned about the types of information they would like to have, agricultural development was placed first, followed by family welfare and basic education. Given the high degree of reliance on subsistence agriculture this reordering of information needs is not surprising. 29 Adapted from R. Marwah (2000), "Impact of Information technology on the poor in Nepal" May, and "The impact of Information Technology on the Poor in India", August. 29 Table 1.1 Sources of Information and Communication for the Rural Poor Source Effective Not effective Not used Newspapers 24 35 41 Radio 71 18 11 Television 2 39 59 Telephone 1 18 81 FAX 100 Computer I 99 Village Head 70 19 11 Friends 83 16 1 Political Leaders 78 18 4 Family 81 18 1 Schools 21 49 30 The rural poor indicated that information can have an impact on various aspects of economic and social life and thus bring about a qualitative change. For instance, a majority of those surveyed believe that there has been a considerable improvement in agricultural practices as a result of information obtained through radio programming and through training programs on farming techniques that are occasionally held in the villages. Similarly, over two thirds believe that there has been considerable improvement in health services as a result of new information. With respect to other types of information such as improving employment or education opportunities, households thought the impact had been less great and were not as positive about the future impact of information about these topics. Characteristics of the Urban Poor 150 households were selected from the urban areas of Kathmandu. Compared with their poor rural countrymen, the poor urban Nepalese had much wider access to a range of "modem" information transmitters. In terms of asset ownership, for example, 68% of households owned a radio, 55% a television, and 19% a telephone. Computers were considered effective by 4% of those surveyed Another difference is that nearly twice as many urban households consider the newspaper an effective communication tool, versus just 24% of the rural poor surveyed Nevertheless, as with their rural compatriots, there is still a strong reliance on family and friends for information as can be seen from Table 1.2. Table 1.2 Sources of Information and Communication for the Urban Poor Source Effective Not effective Not used Newspapers 43 31 26 Radio 71 17 12 Television 57 25 18 Telephone 13 20 67 FAX 0 8 92 Computer 4 4 92 Local Leader 19 45 36 Friends 73 21 6 Political Leaders 12 65 23 Family 83 11 6 Schools 16 39 45 With respect to the categories of information currently received by the urban poor, 'entertainment' is the most widely available, followed by family welfare/womens' development and market information However, in terms of what the urban poor say they need, there is a gap between what they get and what they want The urban poor (who have little or no formal education) place basic education at the top of the list Next in order of importance is market 30 information Because of the dearth of employment opportunities in the organized sector, most people in the urban areas in Kathmandu are self-employed or have small businesses in the informal sector, thus explaining the crucial need for market related information The urban poor, like their rural compatriots, believe that information can have a positive impact on their economic and social well being Benefits from applying information (received through various sources) have been translated into better healthcare services, which they believe has led to an improvement in the quality of life With respect to the future, only in the area of improving delivery of social services was there a majority who believed information could make a positive impact. India India, a country with one billion people, has experienced a rapid GDP growth (5-6%) since the mid 90s But it is still a country with over 300 million poor, twice as many as in sub-Saharan Africa The recent growth performance has stemmed from deep structural transformations, the reduction of the economic role of the State, the deregulation of markets and increased extemal and intemal competition. Despite a strong and fast growing IT industry (see Annex V), access to telecommunications and ICTs remain very low, particularly in rural areas. Research As in Nepal, in the first half of 2000 a study was undertaken in India to understand the information needs of poor rural and urban households and their access to ICTs This study utilized desk research to develop the main trends and field surveys to provide the primary data in selected States, districts and villages The field surveys utilized two questionnaires - one for urban households and another for rural households. In rural areas, fieldwork was undertaken with the help of field workers and local people in the States of Uttar Pradesh, West Bengal and Andhra Pradesh In the State of Uttar Pradesh, three villages in three different districts were included in the sample In the States of West Bengal and Andhra Pradesh, one village in each State was included in the sample In each village, a sample of 30 households belonging to low-income groups was surveyed for a total of 150 families. In urban areas, fieldwork was undertaken in the cities of Delhi (100 households surveyed) and Mumbai (150 households surveyed) to assess the impact of ICT on the urban poor. In both cities the sample households were identified on the basis of their declared income and corroborated by assessing their standard of living and ownership of assets. Findingsfrom Rural Areas In the state of Uttar Pradesh the villages of Bhasoli (district Bulandshahr), Panihavar (district Aligarh) and Jait (district Mathura) were surveyed In West Bengal the village of Aagadiya (district Midnapore) was surveyed and in the State of Andhra Pradesh the village of Vaikunthapuram (district Guntur) was surveyed These villages exhibit several common features First, a high percentage of inhabitants are engaged in farming Second, a majority of inhabitants are very poor with income levels ranging from Rs 1,001/- to Rs.3,000/ per month Finally, the majority of households have limited access to "modem" information transmitters such as phone, television, fax and computers In fact, in most instances, these villages are not directly supplied with electricity, but pilfer it from neighboring supplies for a limited number of hours per day. Public telephone access is not available within four of the villages, although public pay phones are generally available a mile or so away. However, respondents report using telephones only in emergencies because of the relatively high cost. In the three villages of Uttar Pradesh there are no community centers/libraries with access to information. 31 In the village of Aagadiya in West Bengal, the state government is opening a number of information centers that will make television and other media available to the public. This village has benefited from its location just 39 km away from the district headquarters and has made progress in the agricultural sector through improved irrigation, banking facilities, and development of markets. Finally, in the village of Vaikunthapuram in Andhra Pradesh. There is public access to a television, public address system and a public pay telephone facility. The surveys revealed that the populations of these villages rely to a large extent on traditional means of communications: friends, family, village heads and politicians (Table 1.3). Dependence on local leaders and panchayat head was considerable in the villages of West Bengal and Andhra Pradesh in particular. There was a very limited level of activity by NGOs in the villages surveyed. Table 1.3 Dependability on the Different Segments of the Societ by the Rural Poor (%) Local leaders Educated people Relatives and Government N.G.Os. (Religious leaders I friends officiais 50.7 64.0 88.7 6.7 2.0 In terms of the effectiveness of the different sources of information and communications, family and friends are viewed as the most effective followed by radio. Newspapers are also considered an effective source of information in some of the villages though not in Aagadiya and Vaikunthapuram. Television is also an effective source of information (particularly in the villages of Uttar Pradesh). Only radios are owned by a majority of households (77%). Televisions, telephones and newspapers, on the other hand, are available to the majority of households on a shared basis. There were a few poor households who had never viewed television, read a newspaper or used a telephone. Very few families had shared access to a computer/Internet connection. Table 1.4 Major Sources of Information and Communications for the Rural Poor Informiation/Advice Personal Shared/communal Not at all ownership Radio 77.3 22.7 0 Television 9.3 84.0 6.7 Telephone 0 63.3 36.7 Fax machine 0 0.7 99.3 Newspapers/printed sources 11.3 80.0 8.7 Computer/intemet,etc. 0 12.0 88 Family/Friends 100 0 0 Village head/community and political leaders 0 100 0 In terms of the types of information most frequently received, family welfare, womens' development, political news, and entertainment programs were most common. Local political news was obtained from newspapers, and local leaders, while information related to education and training was received, to a limited extent, from educated neighbors. With respect to health and family welfare, information was frequently received through health workers. Table 1.5 Frequency of Information by the Rural Poor (%) Informaton/Advice Very Often Quite Often Seldom Not at AlU News-Political/Govt. Policies 57.3 28.7 10.7 3.3 Entertainment 51.4 32.7 12.7 3.3 EducationallTraining programs 17.3 42.0 28.7 12.0 Economic information (Agriculture, markets, business development) 13.3 46.0 26.0 14.7 Employment opportunities 10.7 25.3 38.0 26 HealtWleducation/family welfare and Women's Development 41.3 46.7 10.7 1.3 Welfare Programs of Govemment 11.3 31.3 41.3 16.0 32 When respondents were asked to rank their information needs, not surprisingly, households assigned a high importance to information related to health and family welfare. Most of the villages surveyed had no proper medical facilities, as hospitals were located mainly near urban centers. They also assigned importance to information related to government poverty alleviation programs and educational/training programs that would enhance their skills. Findings from Urban Areas In New Delhi, the sample respondents were drawn from the slums in Kathputali Colony, Pandav Nagar, Garib Basti, Jhuggis and in the Zakhira area. This area of West Delhi has a concentration of poor families - mainly migratory labor from the states of Uttar Pradesh and Bihar. A total of 100 families were surveyed in these areas. The respondents were mainly employed as casual labor in the construction sector or were self-employed in the informal sector. A third of the family heads were also employed in the private sector. The majority of poor households declared having income levels ranging from Rs. 1,00l/- to Rs.3,000/- per month. Mumbai, India's largest industrial and commercial center, has vast communities of slums varying in size from a few huts to settlements with an estimated half million residents. A total of 150 households were surveyed in the slum of Dharavi. Approximately a third of the respondents interviewed were casual workers, about a quarter were employed in the private sector and the rest were domestic laborers, vendors and petty traders. About 15 % of the families belonged to the very low-income strata, and subsisted on a day to day basis. Conditions in Dharavi, as in many Indian slums, are dismal. Some State government schooling is provided. Government health facilities are mostly inaccessible to the poor, however, NGOs, such as Nargis Sanstha, provide health facilities and mobile health vans for the poor. Electricity is scarce, but is pilfered from the main lines. Telephone facilities (public pay phones) are available, but infrequently used. Table 1.6 Major Sources of Information and Communication for Urban Poor Information/Advice Personal ownership Shared/communal Not at all Radio 58.8 41.2 0 Television 46.8 53.2 0 Telephone 0 82.4 17.6 Fax machine 0 3.2 96.8 Newspapers/printed sources 21.6 58.8 19.6 Computer/Intemet, etc. 0 9.2 90.8 Family/Friends 100 0 0 Community and political 0 100 0 leaders The survey found that urban respondents also relied to a great extent on traditional sources of information such as family and friends in addition to utilizing radio and television. In New Delhi television was frequently used by 55% of respondents, whereas in Mumbai the figure was 25%. While televisions, telephones and newspapers are available to the majority on a shared basis, very few families had shared access to a computer/Internet connection (9% in both New Delhi and Mumbai). The results are tabulated in Table 1.6. With respect to the "most often" used sources of Information by the poor households, television is ranked first followed by the radio, family and friends and newspapers. Community and political leaders were seldom relied upon. In New Delhi the urban poor were particularly interested in entertainment programs and political news received via radio and television. Information about health and family welfare was received quite frequently through government dispensaries, secular and non-secular charitable institutions, wall writings and newspapers. Information related to education and training programs was received to a lesser extent, and appears to be largely a function of literacy levels. Economic information related to business and market developments was received frequently by only one fifth of households. 33 In Mumbai, households received information related to local political developments through newspapers, family and friends and local leaders. Information related to educational and training programs was received to a limited extent, mainly through educated persons and NGO representatives, but there is a paucity of information related to employment opportunities. With respect to health and family welfare, information was often received via health workers and medical dispensaries of NGOs such as Nargis Sanstha. Another NGO (Youth for Unity and Voluntary Action) had also initiated educational programs for slum dwellers and had thus served as a significant source of information regarding the information needs of the poor. The government's involvement was reported to be mainly through its Malaria Treatment programs during the monsoon season. Table L7 Frequency of Information Received (% Very Often Quite Often Seldom Not at All News - Political/Govt. Policies 67.2 28.8 4.0 0 Entertainment 47.2 28.8 24.0 0 Educational/training programs 33.6 34.4 13.6 18.4 Economic information/business 21.2 17.6 19.6 41.6 development ._____ _____ ______ _____ ___ _ 23.2 20.0 34.4 22.4 Health/education/fanily welfare and 58.4 34.0 7.6 0 women's development 7.6 Welfare Programmes of Government 23.6 26.0 31.2 19.2 Any other 39.6 34.0 16.8 9.6 When questioned about the types of information they would like to have, employment opportunities were placed first, followed by family welfare, education and training programs. In view of the trend towards self-employment, it is not surprising that the urban poor assigned high priority to information about employment and training. Potential of ICTS to help the poor As a result of increased awareness about new information and communication technologies, and training/educational opportunities, the poor in New Delhi believe that there could be a qualitative change in their living standards, if they had greater access and exposure to new means of communications and information. In New Delhi households were confident that greater access to radio, T.V, telephone and computers would enhance their opportunities for learning (63%), improve their quality of life through entertainment (45%), as well as enable them to raise their income earning capacities (37%). With respect to marketing strategies, national integration and export opportunities, the responses were mainly "may be" or "not at all". Nearly 13% of households that had small self- employment ventures believe that access to ICTs could help them to export their goods/services. 34 Annex II Information needs and access to ICTs by small, micro and medium enterprises Botswana With a per capita GNP of US$3260, Botswana is one of the few middle-income countries in Africa. Its GDP growth rate - averaging 7.3% a year between 1970 and 1995 - was the highest in the developing world for that period. The Government has prudently managed the country's immense natural resources - particularly diamonds - reinvesting the benefits in social and physical infrastructures. This has resulted in rapid improvements in social indicators, particularly in education30. Botswana has a relatively open trade and investment policy and encourages the full participation of the private sector in the economy. The Government is strongly committed to the development of competitive enterprises through use of modem technology. As a result of the 1996 liberalization of telecommunications markets and operations, Botswana experienced one of the fastest rates of growth in the adaptation of new technologies such as mobile telephony, the Internet and e-mail. The research A study of the role of information and communication technology for SMEEs was conducted in 199931. The objectives of the research included: to describe the information needs of SMEEs; to understand the information systems used by entrepreneurs and assess the quality of information being provided; and to establish the potential for ICTs to bridge possible gaps between needs and provision. The sample included approximately 100 enterprises located in predominantly urban areas, which were reached through postal questionnaires and interviews. It covered firms in manufacturing (involved in domestic and export markets) and services (including tourism). About 60% of enterprises depended solely on domestic markets and 17 % exported more than 50% (the figure included tourism enterprises). Slightly more than half of business owners were citizens; the vast majority of respondents had been educated to senior secondary level and beyond. Information needs and channels The survey tried to establish enterprises' information needs (related to business inputs, output markets and business environment) and to investigate whether firms were making use of informal or formal sources of information (journals, internet, government organizations, etc.). Information needs, as stated by the firms themselves, included accessing finance, solving internal fnancial problems, accessing information relative to trained personnel, management training, new customers and new markets. Responses indicated a deeply felt information gap - that is the existence of a large demand for information that remains unmet across all sectors. How did enterprises attempt to satisfy their information needs? High reliance was given to information obtained through the knowledge and experience of the business owner, or received through informal local networking within the business community as well as family and friends and information accumulated through enterprise-specific learning. The exchange of informal 30 Youth illiteracy fell from 32% in 1980 to 17% in 1997 for young males (15-24), and from 25% to 9% for young females. (Source: World Development Indicators, 1999) 31 See Duncombe (1999). The reseach methodology was developed by R. Duncombe and Richard Heeks, IDPM, University of Manchester with the support of a grant from the UK Department for International Development. 35 information was common across all areas of business activity32 and was rated highly. This was true in both the traditional manufacturing sectors and in services, such as tourism. Formal sources of information such as those provided by joumals and the Internet were important only for export-oriented service sector enterprises. The vast majority of enterprises in all categories regarded direct contact with customers through face-to-face meetings as the single most effective method of business communication. ICT utilization and communication channels Access to and use of computers and the Internet varied across sectors. Non exporting manufacturers (predominantly citizen-owned) were overwhelmingly not Internet users (though they made regular use of fixed line telephones and fax and mobile communications). By contrast, computer based activities were relatively spread among manufacturing exporters and service enterprises. However, levels of computer use were typically low: 39% of 'users' actually made no use of their computers for everyday business activities. Amongst the remainder, by far the main use was word processing. Only 20% of actual users had computerized basic business functions such as customer invoicing and internal accounting. It was not uncommon to find computers out of use or consigned to a back room, after initial failed attempts to adapt unfamiliar software to the needs of the business. Service enterprises showed the highest rate of computer use, particularly for e-mail and Internet. Approximately half of respondents had Internet access with 43% using it very often, and 30% rating it very important as a source of business information. However, there was far less use of the Internet as a means of disseminating information about enterprise products and services. Only 13% of Internet users regarded it as very effective as a product/service promotion tool, and a further 47% had not yet used the Internet for business promotion. Conclusions Overall, the study suggests that in Botswana there is little penetration of ICTs in SMEEs. Only in specific sectors, which require regular access to regional or world-wide information such as travel and tourism, are ICTs widely diffused. The survey also suggests that there is a large unmet demand for information. However, the further introduction of ICT is unlikely to satisfy this demand. For the 'non-ICT users' (mainly manufactures selling in local markets), lack of fnance and skills are the major constraints to the introduction or expansion of computers and other technologies. The study suggests that there isn't much scope for introducing these enterprises to ICTs, unless basic skills and/or financial stability are significantly improved. By contrast, service enterprises, which are already intensive users of ICTs would probably gain from continuous upgrading of their information systems and access to emerging technologies. For enterprises with access to, but low use of computers, the priority would be to enhance technical skills and managerial/employee capacity, rather than to expand ownership of ICTs. Failure to use computers often arises from the gap that exists between the design conceptions of much business software and the organizational realities of many small enterprises (Heeks and Bhatnagar, 1999). Closing the design-reality gap will be a key priority, which includes issues of IT production strategy and software production skills within the country. The study suggests that 32 For example, information about loans often comes from family and friends (as do the loans themselves). On-the-job training allows information to pass directly from experienced to less experienced employees. Business associates will provide information in the form of informal advice or technical assistance and market information, leading to new business opportunities. 36 until the amount of local content is expanded in terms of both volume and quality, it is unlikely that the benefits of access to the Internet will justify the costs for most small firms. Moreover, substantial long-term financial, socio-cultural, and knowledge barriers may need to be overcome before many SMEEs can effectively use ICTs. Raising the overall capacity of the SMME sector as a whole will require long-term integrated policies that focus not only on access to new technologies, but also on upgrading managerial capabilities and skills. In addition improving in the quality of marketing and access to local markets and better financial services and credit arrangements will also be required. Tanzania Tanzania has a GNP per capita of $240 (in 1999) and about half of the population lives under the poverty line. The economy is dominated by agriculture (contributing about 60% of private sector GDP), while manufacturing accounts for only 7-8% of GDP. During the last four years macroeconomic stability and growth have re-emerged with growth averaging 4% a year- despite adverse weather conditions. The Government has engaged in deep structural and institutional reforms and has initiated the restructuring and privatization of large parastatals and public monopolies. Regulation and development of the postal and telecommunication industry is under the responsibility of the Tanzania Telecommunication Commission (TCC), established in 1993. TCC has licensed two basic telecommunication operators, five Mobile Cellular Telephone operators and four Public Data Communication Operators. There are also ten Internet Service Providers (ISPs), four Radio Paging and 80 Radio Communication operators and more than 70 licensed telecom equipment dealers. Indicators show that in 1998 Tanzania had only 3.8 telephone main lines, 21 television sets, and 1.6 computers per 1000 inhabitants. The Research33 A survey was carried out in Tanzania during the summer of 2000 to explore information needs and availability/access to information sources by SMMEs (Mungunasi, 2000). The research methodology, questionnaire and reporting structure adopted in this study are the same as those in the Botswana study discussed above. The sample covered 107 SMMEs34 located in the cities of Dar es Salaam (70%) and Arusha (30%). The sample covered both formal (71%) and informal (29%) enterprises in the manufacturing, services and agriculture sectors. Within the service- based enterprises a number of tourism-based firms were examined independently as they showed specific characteristics with respect to the use of new communication technologies (particularly internet and e-mail services). The questionnaires were completed by entrepreneurs/managers, the majority of which have secondary, or higher, levels of education. Sample analysis revealed that 91% of respondents operated SMEEs with the number of employees ranging from 1-29. Finance and information aboutfinance is needed In order to identify the information needs of firms surveyed, respondents were requested to identify a list of factors that would contribute to the success of the their business over the next two/three years. Access to additional finance was mentioned most frequently as critically 33 Adapted from Mungunasi E. A. (2000), "Survey of Information Communication Technologies within Small, Micro and Medium Enterprises in Tanzania. The research methodology, questionnaire, and reporting structure adopted in this paper were developed by Richard Duncombe (1999) and Richard Heeks, IDPM, University of Manchester supported by a grant from the UK Department for International Development. 34 Questionnaires were distributed to a random sample of enterprises based on the weight of private sector activities in urban and peri-urban areas: manufacturing (30%), services (40%) and agriculture (30). 150 questionnaires went to Dar es Salaam (with a response of 78) and 100 questionnaires went to Arusha. (with a response of 29). 37 important by approximately 65% of enterprises while an additional 26% of the enterprises considered it very important (Table II.1). These information needs strongly correlate with the factors considered critical for business success. Manufacturing and service based enterprises placed additional financing at the top of their priority list, while the majority of tourism based enterprises listed improvements in marketing/promotion as the most critical factor for success. All enterprises placed increasing sales/diversifying market base as the second priority. Table II.1 Prioritized List of Information Needed Most Urgently within Enterprises Prioritized Success Factors % of enterprises regarding infornation needs as 'essential' or 'very important' Essential Very Important Sources of Finance 67.3% 22.4% Existing Customers 63.6% 22.4% New Local Customers 57.0% 29.9% New Technology/Equipment 53.3% 31.8% Export markets 19.6% 14.0% Management/Staff Training 18.7% 35.5% Laws and Regulations 17.8% 37.4% Land/Premises 15.9% 13.1% New Staff 9.3% 12.1% Quality information is hard to obtain Respondents were also asked to indicate difficulties encountered in obtaining good quality information on a range of factor considered to be critical for the success of their businesses. (Fig.II. 1). The results indicated a wide gap between information demand and supply. Respondents stated that they were not able to obtain information about sources of finance, new local customers, export markets, new technology and equipment, and management/staff training. Information on sources of finance was difficult to obtain or not obtainable by three-quarters of manufacturing and service-based firms. The situation was even worse for tourism-based enterprises where about 89% of the respondents found it difficult to obtain informnation on sources of finance. Manufacturing sector enterprises mentioned information on sources of finance, new local customers, new technology and equipment, export markets and land/premises as their greatest problem areas. Tourism-based enterprises found it difficult to obtain information on export markets, new technology/equipment, management/staff training, land/premises and sources of finance. Access to ICTs What kind of information sources and channels do enterprises use to satisfy their information needs in the daily running of their businesses? Table 11.2 shows information access channels for the surveyed firms. Results suggest that 92% of enterprises have postal address, 82% telephone (fixed/mobile), 37% fax and 34% computers. Of the 34% of enterprises having computers, 47% had their computers networked. Table 1L2 Percentage of enterprises having access to information Enterprises Percentae (%) Postal address 98 91.6 Telephone fixed line/mobile 88 82.2 Fax 40 37.4 Computers 36 33.6 Networked 17 47.2 Not networked 19 52.8 38 Fig.II.1 Ease of Obtaining Good Quality Information by Respondents (Overall sample) 80.0 70.0 13 L~~~~~~~~~~~~~~~~~~~~asy to Obtain 60.0 50.0 *~~~~~~~~~~~~~~~~~~~~Not Easy to o 40.0 - ~ * g > fi * < g Obtain 40.0 13~~~~~~~~~~~~~~~~~~~Not able to Obtain 2ilm0.0 XE Don't Know E * I ,I 0 E Q e 10.0 k I U C -o E C/ O S Firmn rely on personal networks- With respect to information sources utilized, the results suggest that 98% entrepreneurs rely, to a very large extent, on their own knowledge and experience. Other sources of information were as follows: family and friends (87%) and informal business networks of customers (82%) and suppliers (62%). Information from formal sources such as banks, consultants and business support agencies are rated low. Access to information from Internet and Databases, however, was rated very or quite important by only 30% of entrepreneurs. With respect to tourism-based enterprises, own knowledge and experience, and contacts outside Tanzania were the most important sources of information (78% of the respondents in tourism- based enterprises). The greater dependency on contacts outside Tanzania by the tourism sector indicates the importance of foreign customers to these enterprises. On the other hand, low dependency on contacts from outside Tanzania by both manufacturing-based and service-based enterprises indicate the low importance placed on export markets for the products/services of these enterprises. -andface toface meetings Fig 11.2 presents the different communication channels utilized most frequently by entrepreneurs/managers within formal sector enterprises. Face to face meetings are by far the most common channel, utilized by approximately 94% of respondents within service-based and manufacturing-based enterprises. Other communication channels such as fixed line telephone, e- mail, mobile telephone and postal services are not utilized very often, although the rate of frequency is appreciably higher among service-based enterprises. Respondents from the tourism- based enterprises regarded e-mail/internet and fax as the most often used means of communication. 39 Fig 11.2 Percentage of Respondents using Communication Channel 'Very Often' in their Business Dealings. (Overall sample) 100 90_0 X 80-X 1 l _ .70 [' lVery Often c60 *Quite Often s o ONot Very Often 30 F3 tNotUsed 30 ~-20 10 0 0.; C ~Cmmniaton Channe I~~ ~~~ a 202_Ili_fl_11|I|iglE|1 g planin pakaes CA/A an Intemetemil aces Wor prcssn is use by abu one1 Comute-bae Aciite on : a ula Bais iti sub-sectors) _ o -_, - -_ ~~~~Comuicto Chn.e tow~~~~ ~~~ leel of compxterizati Respondents were a t a the e no whi0c 'a use cu a d_ b qarvier Enterprises 27d by 44% of 4r 4tris. TbeI3PretgofSrieadManufacturing 24se 12errie 6ain Use 12of % Low tevelseo o ctivitisona ouaervis Enufctring and sri enterprises ad 1b % 1ti% 1 t% 1e4% 6s%e About 30% of survey respondents had internet access, but only I11% reported using it frequently and only 10% stated that it was very important to business. For tourism-based enterprises, 67% reported that internet access was an important channel for information dissemination to and from suppliers and customers. For tourism-based enterprises e-mail is an important and cheap way of communicating with customers outside Tanzania. However, in terms of the effectiveness of the iCrntet as a method of promoting products and services, only 9% of intenet users regarded it as very effective, and a further 70% had not used the internet as a business promotion tool. With respect to computer storage and processing of information (Table 23.4), this is particularly low for manufacturing enterprises but higher in service based and tourism enterprises. The results tend to indicate that at present the Intewet is used as a method of accessing information, rather that as a of business promotion tool. 40 Table 11.4 Percentage of Respondents Using Computer-Based Storage Processing Systems (Within sub-sectors) o _ Conclusions~nc b . r C 0~~~~~~~ 0 cd = 0 0~~~~c)i Service Enterprises 21% 18% 19% 22% 19% 23% 32% 18% 14% Manufacturing 12% 15% 12% 15% 12% 12% 9% 12% 6% Enterprises Iemai unmet.I Tourism enterprises 33% 22% 144% 1a1% I11% 133% 144% 144% 11% Conclusions The results from the Tanzania survey are remarkably similar to those of Botswana. First, SMEEs have access to a variety of ICTs, fixed lines and mobile telephone and fax. About one third of enterprises in the sample have access to computers. Second, their perceived information needs very closely match what they consider to be the main factors for business success. Particularly important is the desire to know more about sources of finance. This is a very common fnding in studies of information needs in developing countries. In many cases, however, it indicates not lack of information per se, but the inability of entrepreneurs to access critical information regarding altemative sources of fi ance and markets. Third, many of the infor sation needs of the enterprises remain unmet. Results suggest that this might be because enterprises mostly use and trust informal networks of information. Formal methods of media-based marketing are rated low by the small business sector. The postal and telecommunication infrastructure is thought to be insufficiently developed and not reliable enough for effective direct telephone sales or direct mail marketing/sales. Marketing and advertising services are at an early stage of development and cater predominantly for the corporate sector. Such services would generally not be affordable by SMMEs. The study confirms that having access to technology is not sufficient to make information a factor in increasing firm productivity. Many business owners/managers of SMEEs simply lack the necessary skills to access and assess information provided by the Intemnet. A second reason is that there is likely to be a 'lack of useful content' in the information that can be accessed through the Intemnet. The information that is provided through the Intemnet may be irrelevant to the specific circumstances faced by local firms. 41 Annex III Communal Access Models South Africa35 With a per capita income of about $3000, South Africa is classified as a middle-income country. However, despite its wealth and abundant mineral and energy resources, more than half of the population lives in conditions of poverty. Income disparities are among the worst in the world. In the post-apartheid era the Government has been striving to improve living standards but economic growth has remained slow and employment stagnant. Telecommunications and access to information are regarded as integral to the overall strategy for reducing poverty and inequality.36 The Government's plan is to extend universal access to telecom services through increased investment by both the national operator and the private sector.37 In comparison with other sub-Saharan countries, national telecommunications networks in South Africa are well-developed and the communication industry capability is able to extend networks, utilizing existing (fixed line) or emerging (cellular/radio/satellite-based) technology to under-served areas. However, coverage is extremely uneven, both geographically and between socio-economic and racial groups. Thus, 85% of white households have a telephone line, compared with 14% of African households. Across all races, 1% of rural households have a telephone line, compared with 32% of urban households.38 South Africa already has considerable experience in the provision of communal access to ICTs and can offer many lessons to other developing countries. Communal access is being provided not only through payphones, but also through the provision of subsidized phone shops and telecenters, serving communities in rural disadvantaged areas. South Africa has encouraged communal access through different means. First, service requirements (within the license agreement), have been used by the authorities to ensure the attainment of service targets. Telcom is required to provide 120,000 new payphones, lines to 20,000 priority customers and 3,204 villages by the year 2002. A further 1.7 million lines to 'disadvantaged' areas during the 5 year monopoly period 1997- 2002 are also required (Benjamin, 1999). Privately run phone, most of which offer voice telecom services, fax and 35 Fro R. Duncombe (2000), "Information and Communication Technology, Poverty and Development in sub-Saharan Africa" Mimeo. 36 See 'Poverty and Inequality in South African', a report prepared for the office of the Executive deputy President and the Inter-ministerial committee for poverty and inequality, 13'h May, 1998. Available at: http://www.Rolitv.org.za/govdocs/reports/vovertv.html 37 The state provider, Telcom, is 30% owned by Thintana Communications, which is jointly owned by USA and Malaysian-based telecom companies. Telcom still has a monopoly on the provision of local, national and international calls and public phones, as well as infrastructure for Value-added Networks (VANS) and cellular provision. Two cellular providers operate under licence from the regulator (Vodacom and MTN). Telcom also has shares in the main satellite providers. Telcom will retain its strong (monopolistic) position in the partially de-regulated market until 2003 when the government intends to issue a second fixed line operator licence in line with plans for increased de-regulation of the market for telecom services. 3S 'Poverty and Inequality in South Africa' (May, 1998), p26. Households with a telephone service are defined as having ownership of a fixed-line or mobile telephone in the dwelling. Those with access are defined as having 'self- defined' access either in the dwelling or in a neighbor's house, a communal place a shop or a clinic. South Africa's Household Survey (1997) defined only 14% of Black South Africans as having household ownership, but 59.4% as having (self-defined) access. (Benjamin, 1999). 42 photocopying services, have proved to be financially viable.39 Second, the government is encouraging community-based information provision through the establishment of telecenters. Many of these telecenters (offering telephone, Internet and other computer-based services) are hosted in existing Multi-Purpose Community Centers (MPCCs). MPCCs provide a wide range of community-based services, including education and training, citizen legal advice, health information, business support.40 The National Information Technology Forum (NITF) recently gave an assessment of the extent to which MPCCs are able to mediate ICT-based services. A questionnaire was sent to approximately 8,000 community organizations (both MPCC and non-MPCC) and the results were compiled on the basis of 441 responses, 235 of which were MPCCs. Of these, 87% were found to have telephone connections, 66% had access to computers on the premises and 30% had access to dial-up e-mail facilities. In rural areas, 70% had telephones, 44% computers and 18% e-mail. Within the survey, however, there was no allowance for non-response bias, and the results were likely to favor ICT-users over non-users.41 Many such centers were well established and had only recently begun to offer communications access and other ICT-based services. Financial sustainability is a problem Almost all MPCCs rely on external funding through government bodies, donors and NGOs, and most regard lack of financial resources as their main constraint for development. The few financially viable telecenters have generally been those able to offer a wider range of fee-paying services - usually computer or other training - to local communities. For the majority, however, there has been under-utilization of resources (particularly computer equipment), lack of financial control and record keeping and no clear understanding of responsibilities and obligations.42 Though not financially autonomous, many telecenters represent a successful model of community access, particularly those serving poor/disadvantaged communities (see the example of the Mamelodi Community Information center in box 111.1). MPCCS are controlled by local community stakeholders, and are thus accessible and trusted information sources. The communication services provided are driven not by technology but by the real needs of communities. They often introduce poorer communities and groups to new ways of learning, communicating and doing business. The critical issue within South Africa at present, therefore, is how to balance the need for financial sustainability (possibly through increased private-sector involvement) and the need to satisfy the information needs of the community and the poor. 39 Vodacom in South Africa set up 1,400 phone shops in part to meet their target of 22,000 extra lines to poor and disadvantaged areas. A phone shop costs about R20,000 to set up, and all are franchised to private owners. Phone shops are located overwhelmingly in previously under-served townships and it is estimated that they are earning a total of R8 million per month (Benjamin, 1999). 40 These were a selection of activities undertaken by a sample of 54 MPCCs operating in the Western Cape Province of South Africa. 26 were rural-based and 27 in towns or townships. Of these 76% offered access to telephones, 61% offered access to computer-based facilities and 80% offered access to e-mail. 41 Thus, the results of the survey did point toward a far higher level of ICT usage than had been anticipated. This was particularly so in some of the township and rural areas, where such community-based organizations were found to be the only resource centres in the area. The research was funded by the IDRC and carried out by a group of researchers led by Peter Benjamin from the National Information Technology Forum. Further details can be accessed at: htto://www.sn.avc.org/nitf/mpcc/results.html 42 These observations were made during the course of the initial evaluation (1998) of 6 telecenter initiatives established by the Universal Service Agency, and have been summarized by Khumalo (1998). Further details are available at: http://www2.worlbank.ore/hm/afrcomm/0010.html 43 Box I11.1: The Mamelodi Community Center The Mamelodi Community Information Service (Macis) is a non-profit center initiated in 1995 to serve the needs of a township community close to Pretoria. It is heavily used by the community (up to 50 people per day) and offers information and assistance mostly to young people and small-scale business owners. The services offered include typing, photocopying and access to e- mail/Internet. The center also acts as an employment service, assisting in writing CVs and application letters, and as an advice center. It charges minimal fees for services and is supported largely by donations. The center gets many referrals from local community organizations as it has good local linkages and is trusted and well known in the township. The management of the center is responsible to a board of local trustees, who are community leaders. Access to ICTs, within the center, therefore, forms part of a wider community-based service. Local people without experience of accessing computer-mediated information can be assisted by local community workers who have detailed knowledge of the real information needs of individuals and groups within the community. Source: MPCC Case Studies: http://www.sn.apc.org/nitf/mpcc 44 Annex IV Private sector led ICT development Ghana43 Ghana enjoyed a high rate of economic growth during the 1990s, resulting in a strong reduction in poverty. Since the mid-1980s the Govemment has been implementing a progran of economic and structural reforms that has been transforming the economy. A rapidly expanding 'modem' urban sector exists alongside a 'backward' rural sector with little or no infrastructure. Ghana's technical infrastructure has traditionally been weak. In 1995 the Govemment initiated a development program for telecommunications, including rapid de-regulation of the market and a commitment to introduce private sector operators.4 This has included the support of a duopoly of telecom operators, and the licensing of additional suppliers and service providers in under-served areas. The policy framework also makes a commitment to allow open competition in data transmission, telex, facsimile, Internet, fixed satellite, teleconferencing, radio, telecommunications equipment sales and rentals, the provision or mobile cellular and global satellite services. The new operators - Western Wireless and Telcom Malaysia - have 'build out obligations' that include the provision of 225,000 new telephone lines in Ghana over a 5 year period. This will effectively triple total network coverage for the country as a whole by 2002 and open up access for many under-served areas. The benefits of the liberalization have been considerable. Ghana Telecom had installed 77,000 lines over a 22 year period (3,500 per year) before the liberalization: 70,000 new lines were installed in the two years following private sector involvemente5. Other developments include a rapid increase in the provision of payphones, competition in the provision of analogue and digital cellular (mobile) phone networks, the introduction of VSAT-based networks, and new Internet access services offered by competing private sector ISPs, each with their own direct international backbone connections.46 What is the evidence, therefore, of benefits to rural areas through enhanced private sector participation in the information market? In Ghana, as in other countries, modern sector activities (tourist facilities, small and medium enterprises, commercial activities) only involve a minority of the population in rural areas. They do, however, constitute an important pool of demand for sustaining telecommunications provision. Extension of rural networks in Ghana has initially concentrated on providing individual connections to the minority who were able to afford connections, while offering communal access opportunities to the majority through communication centers, and more recently, payphones. The private sector has stimulated an extremely large number of small organizations, located outside the main urban centers, involved in transmitting information to and from poor/disadvantaged households, enterprises and communities in Ghana. It has been estimated 43 R. Duncombe (2000), " Information and Communication Technology, Poverty and Development in sub-Saharan Africa" Mimeo. 44A 30% stake in the national operator (Ghana Telecom) was sold to a Malaysian consortium, and a second national operator's licence has been awarded to another national consortium with external investment provided by Western Wireless (US) Western Wireless intends to deploy a VSAT satellite-based telephone network around the country that will provide the potential to significantly enhance rural connectivity. 45 From a speech made by: John Mahama, Minister of Communications, Accra, 2 March, 99. Full text available on: http://www. 2hanaweb.com/GhanaHomePage/ghana/comm/news.html 45 there are over 1000 small/micro businesses, spread throughout Ghana's 90 districts, which offer basic communications such as telephone, fax, typing and other predominantly business-related services. The preponderance of this type of private sector provision reflects a strong entrepreneurial ethos that exists amongst Ghanaians, and their potential to take advantage of opportunities offered through liberalization of communication services. A growing number of 'Communication centers' offer public access telephone/fax services to local communities in rural areas (Jensen, 1999). These 'commercially-driven' centers typically take the form of a single room establishment offering a basic communication service (telephone, fax and photocopying) on a fee-paying basis. As yet very few of these have been able to offer Internet, e-mail, CD-ROM-based information or other computer-based services, other than those located in the capital city - Accra. (Owen and Darkwa, 1999). These services are not yet financially viable in light of high comparative investment costs, high connection charges and overall lack of demand. 46 Annex V ICT applications for rural development in India47 Introduction India has one of the most dynamic IT industries in the world, growing at an annual rate of over 40% p.a. over the last decade. The software sector has been the star performer with growth of 60% p.a. The sector now employs around 2.8 million people and its exports represent nearly 10% of India's total exports. It is expected that by 2008 the IT industry will contribute 7.5% of overall GDP growth, 35% of all exports and create 2.2 million high skilled jobs. As described in Chapter 2, access to telecom services in India is low, particularly in rural areas. Out of 600,000 villages in India, just over half have access to telephone facilities (see Table V. 1). Extension of coverage to another 25% of villages is being factored into plans for the coming year, but this goal is unlikely to be mete. Table V.1: Current status of village level access to telephone Sou Circles Total Status as of April 2000 DOT's Targets for 2000-01 and DOT's Villages 2001-02 Targets for _____ ____ ___ _____ _ __ ____ ____ __ __ ____ ____ ____ 2002 VPT Vilages 2000-2001 2001-2002 ViDages with no I Andaman& Nicobar 282 274 8 8 O 8 | _2 Andhra Pradesh 29460 23379 6081 0 0 0 3 Assam 22224 14181 8043 5000 3043 8043 4 | Bihar 79208 24923 54285 22000 32285 54285 | 5_ | Gujarat 18125 13923 4202 0 0 0 6 | Haryana 6850 6807 43 43 0 43 | 7 | Himachal & Pradesh 16997 10364 6633 3000 3633 6633 8 1 Jammu & Kashmir 6764 3793 2971 2000 971 2971 9 Kamataka 27066 25801 1265 1265 0 1265 10 Kerala 1530 1530 0 0 0 0 11 Madhya Pradesh 71526 46498 25028 4500 1360 5860 12 Maharashtra 42467 31541 10926 0 0 0 13 North East 14446 4336 10110 5000 5110 10110 14 Onssa 46989 22928 24061 12000 12061 24061 15 Punjab 12687 12123 564 0 0 0 1 6 Rajasthan 38634 23727 14907 0 0 0 17 Tamil Nadu 17991 17845 146 146 0 146 18 Uttar Pradesh(E) 75698 46492 29206 15500 13706 29206 Uttar Pradesh(W) 39551 23531 16020 6000 10020 1 6020 20 West Bengal 38337 19997 18340 9000 9340 18340 21 Calcutta 468 421 47 47 0 47 22 Delhi 191 191 0 0 0 0 TOTAL 607491 374605 232886 85509 91529 177038 Government led initiatives improve the administration of rural programs Initiatives to spread the use of ICTs at various levels of Government date back to 1985-86, when the National Informatics Centre (NIC) began implementing a national program the District Information System of National Informatics Centre (DISNIC) - to computerize selected district offices. By 1990, each district computer was connected to a state computer through a local dish antenna and a satellite communications network. The state computer in turn was connected to a 47 Subhash Bhatnagar (2000)"Information and Communication Technologies, Poverty and Development in South- Asia", Mimeo. 48 Even in the urban areas the private service providers have not been able to meet their commitments. Six licenses for providing basic services in Gujarat, Madhya Pradesh, Andhra Pradesh, Maharashtra, Punjab and Rajasthan were issued 1997-98. As of April 2000 service had started only in three States - Madhya Pradesh, Andhra Pradesh and Maharashtra, 47 computer in New Delhi. The Computerized Rural Information Systems Project (CRISP), initiated in 1987, developed software for the planning and monitoring of the Integrated Rural Development Program (IRDP). Each District Rural Development Agency (DRDA) was provided with a PC to run the software. State govermments were asked to purchase equipment and provide training for their district level officials, but the NIC district computerization program was completely centralized. District information needs were assessed and standardized across districts. Software design and database specifications were also standardized and originated in New Delhi. In the CRISP program, a few elements, such as purchases of hardware and training of district level functionaries were left to State initiative. But the design of the software including the assessment of required information specifying the type of databases, and the reporting system, were all centralized. Impact of DISNIC and CRISP on rural developmentprograms The initiatives were successful in creating a countrywide network of 500 computer centers and in maintaining and supporting the system. While there has been no formal evaluation of the DISNIC and CRISP programs, partial analysis [Bhatnagar 1991; Madon 1992] indicates that computer utilization was effective in only a limited number of districts. On the basis of fieldwork done in 1989-90, Madon reports that the CRISP system is grossly under-utilized and functions are still overwhelmingly carried out manually in all 19 districts of Gujarat. The study found that in most districts computerization is proceeding at a slower pace than was anticipated and in others computers are not being used at all. Furthermore, the software supplied by DISNIC and CRISP is not being utilized suggesting that the objectives and priorities of the program were not in tune with the perceived needs of local administrators. Overall, the impact on daily administrative functions has been marginal and the administrative culture remains unchanged. The experience suggests that unless district administrators are motivated or held accountable for improving performance of rural development programs, the introduction of new technology will not be sufficient to improve performance. The effort required to implement administrative IT reforms should not be underestimated. The few reported successes in a number of districts were mainly attributed to the strong motivation on the part of participants. The Multipurpose Electronics and Computer Centers By contrast, the establishment of multipurpose electronic and computer centers (MEXs) at the block, watershed and panchayat level49 has been successful. These centers service computers and electronic equipment, provide training, maintain servicing manuals, production and testing documents and provide other support services (Choubey 2000). The project is jointly funded by the Department of Electronics and the Department of Science and Technology of the Government of India and is being implemented in ten States. Over 600 MEXs have been setup under this program, of which two thirds are located in rural areas, and a large variety of training and servicing modules have been prepared. Each center directly employs to 4 to 5 persons (as trainers, technicians, etc.) and indirectly employs another 2 to 3. The great majority of these centers have proved to be financial viable and proved to be particularly useful in supporting rural centers. Government led initiatives to increase communication access, education and training in rural areas Since 1975-76 the Indian Space Research Organization has been using satellite communication technology to provide functional education in rural areas through community television. Direct Reception Sets (DRS) were installed in 2400 villages in six States, and transmit 4 hours of rural 49 In India there are 500 districts, 5000 blocks, 3000 watersheds, 100,000 panchayats and 700,000 villages. 48 development oriented programs daily via one way video and two-way audio teleconferencing interactive networks.50 The program was widely researched and found to be effective in terms of usage and popularity (Bhatia, 2000). The network has been used in many States to organize training programs for government officials, teachers, rural women, and elected representatives. During 1998, a total of 97 training programs were organized with approximately 1,000 participants trained in each program. For example, 40,000 primary school teachers were trained at 2400 receiving sites and a follow-up evaluation found that there was a statistically significant gain in the science knowledge content of the trainees as a result. Another successful experiment is the Jhabua Development Communications Project, undertaken in a predominantly tribal district of Madhya Pradesh. This program, initiated in 1996, installed satellite based interactive classrooms at 12 block headquarters and direct-reception television sets in 150 villages. Every evening two hours of development oriented programs are broadcast over the network and the interactive terminals. An evaluation of the project was done utilizing a large scale survey with experimental and control villages. Data were collected before, during and after project completion. The program reached about 35% of the rural population of which 17% were regular viewers and average attendance per set per day was approximately 40 persons. Some of the findings are: i) the majority of viewers discussed the programs after viewing it (although 7% did not understand the language or the topic); ii) over 75% of viewers enjoyed watching the programs, between 44-59% found them to be informative and about 22-28% found them useful; iii) a high percentage of the respondents believed it had a positive impact (for example, alcohol consumption was reportedly diminished); iv) 37% of respondents mentioned farm improvement as a result of watching the programs and 28% reported increased awareness of health issues. The study concluded that there were a number of benefits from using satellite communications. First is the capability to train a large number of geographically dispersed people in a short period of time, utilizing the best available learning resources and providing uniformity of training content. Second is a multiplier effect when training teachers. Third are significant savings in expenditure on travel, logistics and replication of teaching infrastructure. This is particularly relevant in a vast country like India. The Warana project - Taking IT to an economically prosperous rural cluster The Warana Project (see Vijayaditya 2000) is jointly carried out by the National Informatics Centre, the Government of Maharashtra and the Warana Education Department51. The project aims to establish a state-of-the-art computer communications network in cooperative enterprises in a rural area which is relatively well off. The objective is to: provide agricultural, medical, and educational information to villagers through the Internet; create a database of villages based on various criteria; provide tele-education to both primary and higher educational institutes; and establish a Geographic Information System (GIS) of the 70 villages surrounding Varana to improve administration of land records52. Over fifty software professionals have been involved 50 Teaching is done via a small studio and an uplink earth station. It originates the training material either in the form of "live" lectures or recorded video programs from a small studio and uplinks these television signals to a geo stationary communication satellite (INSAT-2C) by means of an "uplink" earth station located at the teaching end. One extended C-band transponder on INSAT-2C has been earmarked for this purpose. The classroom has the DRS/Rx- terminal capable of receiving a signal in extended C-band from the satellite. The talk back is possible through normal STD telephone and fax connection provided in the classroom. In case the location does not have STD connection a satellite talkback terminal can be used. 51 The estimated cost of the project is around $ 600,000, of which 50% is being bome by the Central Government, 40% by the Government of Maharashtra and the remaining 10 % by the Warana Vibhag Shikshan Mandal. 52 A three tier structure was used to interlink the above institutions and to connect them to six IT centres (to provide computing and communication services) and facilitation booths in 70 villages. A wireless LAN of 2 mbps bandwidth connects the six institutions. The Sugar Administrative Building is the hub for the business centres, the Engineering College is the hub for the intranet of all village booths. The Warana Web-server installed at the college hosts the web- 49 full-time in application development since June 1998. All the applications have been installed at Warana, after approval and acceptance of the users. Facilitation booths have been set-up in 70 villages (more than 30 booths are currently operational) with dial-up facilities for linking to the central hub. The Warana network has already benefited various user groups. For example, computerization of land and administrative records has introduced transparency and improved delivery of these services. The computerization of several activities in the dairy and sugar industries has permitted efficiency gains. The provision of agricultural and marketing information as well as education and health information has benefitted the villagers. The facilitation booths and IT and business centers has also generated employment opportunities for the local population as computers operators. But with a starting cost of half a million dollars, the project is expensive and there has been no formal effort to quantify benefits. Given that many services are free, financial sustainability is an issue that will have to be addressed in the future. Public-private partnership: the application of information technology in Milk Collection Cooperatives This case study relates to the utilization of information technology by milk collection cooperatives in India (Chakravarty 2000). The idea was first conceived at the National Dairy Development Board (NDDB), an organization providing financial services to cooperatives53. The machine, testing the fat content of the milk, was developed in a state owned R&D electronics laboratory and commissioned by NDDB. Two private enterprises marketed the machine over a period of 10 years, and provided free installation and intensive training to officers of the milk societies and maintenance by motorcycle borne service engineers. The experiment took some time to take off, but is now succeeding on a large scale. Nearly 600 such systems are in operation in the Kheda district in Gujarat. A few are also in operation in other centers where dairy is a big business. Milk collection centers are located within 5-10 kilometers of the villages supplying the milk. The number of milk farmers selling milk to these centers varies from 100 to 1000 and the daily collection varies from 1000 litres to 10,000 litres. Farmers bring their milk in a variety of containers and cans and are issued identification cards. The farmer drops the card into a box that reads it electronically and transmits the identification number to a PC. The milk is then emptied into a steel trough and instantly weighed, displayed to the farmer and communicated to a PC. The trough is connected to a pump that sends the milk to a chiller. One operator fills cans and another takes 5-ml. milk samples that are fed into a fat testing machine. The fat content is displayed to the farmer and communicated to the PC that then calculates the amount due to the farmer. The farmer can collect payment from an adjoining window. The payment to the farmer is automatically rounded to the closest rupee and the balance change due to the farmer is stored so that it can be added to the farmer's payoff for the next day. In many centers the whole transaction takes 20 seconds. Benefits The application of IT for strengthening the management of milk collection, assessing the quality of milk, and calculating payments to farmers has provided many benefits to all involved based applications. A bank of 10 dial-up lines with modems is interfaced through a router. A proxy server which acts as a firewall also functions as an e-mail server for all users at the facilitation booths and business centres. 53 Earlier efforts in automating milk collection were directed at installing fully automatic machines such as those utilized in Europe at a cost of $8000. The machines were found to be prohibitively expensive and did not work well in the dusty environment of rural India. Subsequently a semni-automatic machine was developed indigenously at 20% of the cost and has proved to be very popular. 50 stakeholders. First, the widespread use of the system in over 700 locations in India has represented a triumph of public-private partnership and a remarkable business success for private entrepreneurs. Second, farmers have benefited because their earnings are now based on an accurate measurement of fat content and weight. In the earlier system fat content was calculated a few hours after the milk was received because the measurement process was cumbersome. This led to malpractice and under-payment. Also payments to farmers were made every ten days because of the inability of the collection centers to calculate the payment immediately. The IT system enables prompt, accurate and immediate payment. Moreover, because the system is fast, queues at the milk collection centers are short despite the high volume of milk sales. Finally, cooperative societies have benefited in a number of ways. First, they need fewer employees to manage daily accounts. Second, automatic daily printing of payment slips provides a way to communicate with the farmers. For example, a database is maintained in the computer for each farmer indicating the number of milk cattle and other details. When cattle requires inoculation (information provided by a veterinary service database) the payment slip can contain a reminder. Moreover, detailed data history on milk production by farmer is now stored in the milk collection center. This data can be utilized to forecast milk collection and to incorporate seasonal variations in fat content. Donor/NGO-financed initiatives: the Village Information Shops in Pondicheri This case study describes the work of the Swaminathan Research Foundation, a non- profit organization located in South India, which set up six Village Information Shops to enable rural families to access modem information and communication technologies. The project was financed by an NGO, the IDRC (see www.mssrf.org). A Value Addition Centre (VAC) was established at Villianur, Pondicheri in 1998 to act as the hub of a communications network. Village Information Shops were then established in six villages, all of them linked to VAC. Each shop is equipped with a Pentium PC with multimedia, VHF transceiver and a desk jet printer. The PC is connected to the wireless network through a modem and a specially designed interface. Each shop also has a board to display email from the Value Addition Center. Visitors can make voice calls within the region and a circulating library of educational CD-ROMs is maintained by VAC for use in the shops. Meetings were held within the communities in each village to choose volunteers54 to operate the Information shops and to identify community information needs. These volunteers, paid by the community, were trained on organizing and maintaining Information Shops, operating PCs, and handling e-mail. This experiment proved successful, because of community involvement from the early stage of development. Through the shops, rural communities have access to information related to health, credit, input price and availability, transportation, pest surveillance and agronomic practices. Anecdotal evidence indicates, for example, that some farmers started cultivating a hybrid variety of paddy, while others managed to obtain information on seed prices and availability. Women 54 For example, in the Kizhur village, volunteers were chosen by the general body of the council, which also nominated a 23-member group of 14 men and 9 women to guide the shop operations. In the Embalam village, all volunteers are women. Each of them spends half-a-day at the shop and takes turns to attend work. The shop is located in the premises of the village temple owned by the community. In addition to bulletins by e-mail this shop receives Real Audio files containing the same information to facilitate its use among illiterate women. 51 workers also managed to negotiate better wages (partly paid in grain), as a result of having obtained infonnation on grain prices. 52 Annex VI The Grameen Village Pay Phone Initiative in Bangladesh Introduction With a per capita income of $350, Bangladesh is one of the poorest countries in Asia and has an economy that is still largely agrarian. Real GDP growth has been around 4-5% a year during the 1990s, not high enough to reduce poverty significantly. Major impediments to growth include frequent cyclones and floods, the inefficiency of state- owned enterprises, a rapidly growing labor force that cannot be absorbed by agriculture, delays in exploiting energy resources (natural gas), inadequate power supplies and slow implementation of economic reforms. Bangladesh invests only 0.25% of its GDP in the Telecommunication Sector compared with 4-6% in neighboring countries, resulting in less than 5 telephones per 1000 population. In rural areas of Bangladesh where more than 80% of the population live, there are only 3 telephones per 1000 inhabitants (644 in U.S.). In 1999, there were over 4 million telephone subscribers (out of a population of 109 million). The average waiting time for a telephone is over ten years. Overloading and lack of maintenance of the public power sector infrastructure cause frequent outages and necessary planned blackouts. The Bangladesh Telegraph and Telephone Board (BTTB) licenses the satellite links that all of ISPs use. Private companies were allowed to enter the market in 1997, and email and ISP providers are now available in Dhaka. The Internet was introduced in Bangladesh only in 1993. Today there are 22,000 accounts and an estimated 100,000 users.[Rao, M et al, 1999]. Bangladesh has 9 active ISPs in Dhaka, the capital city, and 3 others in Chittagong, the second largest city. Initial efforts to develop networking were pioneered in Bangladesh by private firms and NGOs. Four of the larger ISPs are today affiliated with NGOs. The Granteen Phone experience The Grameen Phone is a commercial operation started in 199655 with the purpose of providing cellular services in both urban and rural areas of Bangladesh. It serves approximately 40,000 customers. In a pilot project covering 950 villages and 65,000 people, Grameen Phone (through the Grameen Bank (GB) and the Grameen Telecom) introduced Village Pay Phones (VPPs). This project is unique due to the successful marrying of communal telecommunication facilities (using cellular wireless technology) and a financing scheme through the GB. The GB is able to finance the loan of telecommunications equipment to its members (mostly women) who then re-sell telephone services to the local population. These VPP operators were chosen because of their excellent repayment record for GB loans and because they operated a good business (preferably a village grocery store). They were required to be able to read and write and their residences had to be located near the center of their villages. 55 Grameen Phone Ltd. (GPL), is ajoint venture of 4 partners which was awarded a nation-wide license for GSM 900 cellular mobile phones on November 11, 1996. The partners and their shares are: Grameen Bank (3 5%/), Telenor Invest AS of Norway (5 1%), Marubeni Corporation of Japan (9.5%) and Gonofone Development Corporation of the USA (4.5%/o). GPL is responsible for network operations throughout Bangladesh and for providing service to urban subscribers in particular. There is a bifurcation in the roles of the 2 GB companies-- GTC is the service provider buying airtime in bulk from GPL which is the network operator. GTC provides cellular mobile phones to GB members under a leasing arrangement. 53 The Bayes, Braun and Akhter evaluation In 1998, after about two years of operations, Bayes, Braun and Akhter analyzed a survey of both VPP operators and buyers of services. Fifty VPP operators were chosen from 50 different villages located 40-50 kms away from the capital, Dhaka. This survey represented 60% of all VPP owners at that time. A total of 406 users were interviewed. A breakdown of the 1060 phone calls made in a week indicates that 85% of them were classified as "important", 12% "less important" and the remaining 3% "could have been avoided". Table VI.1: purposes of Phone Calls Made by Sample Users Economic Status Purpose Extremely Poor Moderately Poor All Poor Non Poor Economic: 34 (53.9) 33 (39.8) 67 (45.9) 429 (46.9) a) Market prices of 3 (4.8) 2 (2.4) 5 (3.4) 50 (5.5) commodities b) Employment 4 (6.4) 3 (15.7) 17 (11.6) 50 (5.5) opportunities c) Land Transactions 14 (22.2) 11(13.2) 25 (17.1) 65 (7.1) d) Business-related 13 (20.5) 5 (6.0) 18 (12.4) 231 (25.3) e) Remittances 2(2.5) 2(1.4) 33 (3.5) Family/personal 16 (25.4) 30 (36.1) 46 (31.5) 323 (35.3) Health-related 11(17.5) 15 (18.1) 26 (17.8) 94 (10.3) Other 2 (3.2) 5 (6.0) 7 (4.8) 68 Totals 63 (100.0) 83 (100.0) 146 (100.0) 914 (100.0) Note: Figures in brackets show percentages of all calls made by each group. Source: JU / ZEF Field Survey, 1998. On the basis of respondents' own characterization of reasons for telephone conversations, the study categorized the calls into several groups: economically related, health (emergency and advice), and social/personal (family and office-related). The poor56 and non-poor groups accounted for about the same proportion of economic/finance-related calls (46% and 47%, respectively). However, the extremely poor seem to use phones chiefly for economic purposes, making about 54% of all their calls with these purposes in mind. The poor also make relatively more phone calls for health-related purposes than the non-poor (18% and 10%, respectively). The non-poor, on the other hand, make relatively more phone calls for family/personal considerations (35% vs. 32%), remittances (4% vs. 1%) and also for business-related purposes (25% vs. 12%). Even the extremely poor group indicates that about 21% of their calls are made for business-related purposes. This goes to show that even the poorest segment of the village, involved in production of eggs, vegetables, puffed rice, poultry rearing, etc., make phone calls in order to keep informed on their needs. The study assessed that the VPP operators made relatively large profits. Half of them reap a net profit of about $5-6 /week and 10% earn more than $10/week. This income is about 24% of the household income on average - and in some cases is as high as 40% of household income. This contributed to social and economic empowerment of the Village Phone operators, which were mostly women. 56 The poor were defined as having an income per capita between $100 and $147 a year; the extreme poor were defined as those with a yearly income per capita of less than $ 100. 54 Table VI.2: Assessment of Selected Benefits Variable Target village Control village (N=50, averages) (N=10, averages) Prices: Paddy (% of consumer prices) 70-75% 65-70% Eggs Tk.. 13/hali Tk. 12/hali Exchangerate Tk. 12.50/Ryal Tk. 11.50/Ryal Cost of information/knowledge Tk. 17 Tk. 72 Chicken/ducks Higher Lower Chick feeds Lower Higher Supply of inputs: Diesel Stable Fluctuating at times Fertilizer Regular Occasional problems Others: Poultry mortality rate Lower Higher Law and order situation Improved Same Communication during disasters Quick, effective Slow, less effective Communication with relatives home Any where, any time, Any day Any where, but fixed time, fixed day and abroad Transmission of new ideas Improved Same Mobility of people Higher Lower Spoilage of perishable products Less More Access to health-care services? Faster/effective Slower/less effective Source: case studies and discussions with local people during JU i ZEF Field Survey, 1998. Notice that 1 tk=$ 0.5. Benefits to the poor were significant. For example, farmers in the target village received 70-75% of the paddy prices paid by the final consumers, discernibly more than the 65-70% of the prices received by control villagers. The Richardson et aL evaluation In a study undertaken by Richardson et al. (2000)57, about 300 interviews were conducted with VP operators and local Grameen Bank borrower members. The study found that the Village Phone program yielded significant social and economic impact. The availability of telephone calls replaced the need for physical trips to the city, which saved between $2.7 and $10. Grameen Bank members found that telephone usage was for: financial matters with family such as remittances (42%) and social calls to family and friends (44%). Bangladesh is a labor-exporting country with many rural villagers working in the Gulf States. The Village Phone assists the villagers in obtaining accurate information about foreign currency exchange rates, status of money transfers, etc. Social calls to family and friends frequently involve exchange of information about market prices, market trends and currency exchange rates, enabling household enterprises to take advantage of market information to increase profits and reduce expenses. The study found that the income that Village Phone operators derived from the Village Phone was about 24% of the household income on average - and in some cases it was as high as 40%. As a consequence village phone operators become socially and economically empowered. 57 See: http://www.telecommons.com/villagephone/section4.html 55 Conclusions The success of the Grameen Phone is due to many factors. First, most important are current regulatory and commercial circumstances58 that prevent existing telecom operators from meeting the demand for telephone services. Within this context, GSM cellular phones appear to be the best available technical solution for rural universal access, though they are a high-cost solution59. Once the regulatory environment improves, cellular phone technology may not be the most viable and efficient means of providing universal service. A second factor is the link to a micro-credit organization. Loans to creditworthy customers enabled them to start profitable businesses providing much needed telecommunication services to villagers. The final success factor is accurate planning, detailed guidelines for implementation, and close supervision. 58 The Bangladesh Telegraph and Telephone Board (BTTB) operates a monopoly on interconnection services (which allow users to place calls to any other phone user, regardless of who operates the lines). BTTB has granted rural licences for 25 year monopoly 'fixed line' service to rural districts to the Bangladesh Rural Telephone Authority (BRTA) and Sheba Telecom. Both operators are using wireless local loop technology, which is more cost effective solution for universal access in rural areas than the GSM cellular. However, these operators cannot deploy extensive rural coverage because BTTB has not provided them with sufficient interconnection to meet demand. Grameen Phone is able to provide rural service (despite the monopoly licences held by BRTA and Sheba) because it holds a licence for mobile service and has managed to obtain a better interconnection arrangement with the BTTB. This arrangement, together with the ability to operate within the corridors provided by the network of cellular towers of the Grameen Telecom has allowed Grameen Phone to establish rural telephone coverage with the more expensive GSM cellular phones. 59 At dhe time the phones were introduced, the cost of the handset was $360 (with interest the sum that accrues on a 2-3 year lease the sum to be repaid by the VPP operators within three years totals $ 475. The weekly installment ranges from $ 4.4 to dollars 4. The cost of handsets has fallen by one third since then. The second issue is to know whether the costs are high in comparison to revenues generated. In the Bayes report it was indicated that the average net profits were $5.6 per week (ranging from 70 cents to $ 14 across operators). This surplus represented 20% of the income of the households which run this business. 56 Annex VII Comparative Indicators Table VU.1: Penetration of information and communication user technologies in SAA and South Asian countries Literatc * Analogue - pigital Daily News Radios Television Residential Public Mobile Personal Internet papers Sets telephone Telephones Phones Computers Hosts Main Lines Per Per Per Per Per Per Per 1000 1000 1000 Per 1,000 1000 1000 10,000 Inhab Inhab Inhab 100 Inhab Inhab Inhab Inhab (1996) (1996) (1998) Households (1998) (1998) (1998) (1998) I__ _ _ _ _ _ _ _ _ _ _ __ __ (1998) 1 I Mediam Develo ment(HDI Rank, 1999) Mauritius 75 368 228 77.0 2.09 52.7 87.1 5.01 S.Africa 32 316 125 29.1 3.46 56.4 47.4 32.6 Swaziland - - - 11.4 0.87 4.9 - 2.92 Namibia 19 143 32 16.0 1.3 11.7 18.6 15.99 Botswana 27 155 27 14.1 1.28 14.6 25.5 4.19 Lesotho 8 48 24 1.8 0.07 4.8 - 0.09 Zimbabwe 19 96 29 4.7 0.2 4.3 9.0 0.81 Ghana 14 238 109 0.9 0.09 1.2 1.6 0.10 Cameroont 7 162 81 2.6 0.02 0.3 1.5 - Congo 8 124 8 - - 1.2 - Kenya 9 108 19 1.4 0.23 0.2 2.5 0.24 Sri Lanka 29 209 92 _ _ 9 4.1 0.4 India .. 121 69 I 2.7 0.18 Pakistan 23 98 88 I 3.9 0.22 Low Development(HDI Rank, 1999I Congo. Dem.Rep 3 98 43 0.1 0.01 0.2 . Sudan 27 270 143 1.6 0.1 0.3 1.9 - Nigeria 24 197 61 1.0 - 0.1 5.7 0.04 Zambia 12 121 80 1.9 0.06 0.6 - 0.35 Tanzania 4 278 21 1.1 0.02 1.2 1.6 0.04 Uganda 2 123 26 0.3 0.06 1.5 1.5 0.05 Malawi 3 256 2 1.2 0.05 1.0 - - Angola l 54 91 2.3 | 0.02 0.8 0.8 Rwanda 0 102 - 0.5 0.05 2.3 - Mali 1 49 10 0.5 0.09 0.4 0.7 - Mozambique 3 | 39 4 0.8 0.02 0.4 1.6 0.07 Burundi 3 68 10 1.0 0.01 0.1 - - Burkina Faso 1 _ 32 6 - 0.08 0.2 0.7 0.16 Ethiopia 1 194 5 0.8 0.03 . 0.01 Bangladesh 9 1 50 6 1 . 1 0.00 Nepal 1.1 38 6 t 1 :: 0.07 Medium development countries (Average) 75 | 244 272 37.15 2.24 35.3 26.9 9.94 High development Countries. (Average) 286 1,300 664 104.8 5.55 266.3 303.6 450.43 Source: 1999 World Development Indicators World Telecommunication Indicators, ITU, 1998/1999. 57 Table VII.2: Proxy indicators of human capabilities in SSA and South Asian countries. (HDI Rank, 1999) Youth Adults Youth Primary Secondary Adult Illbteracy net net Illiteracy Rate (%) enrolment enrolment Rate (%) (1998) % of age % of age (1997) .group group M t F ~~1980 1996 1980 i1996 M F ________ __ Medim DevelomnDI Ran4 1999) Mauritius 7 6 79 98 _ - 13 20 S.Africa 9 9 - - 51 15 16 Swaziland 80 91 - 23 Namibia 11 7 76 91 36 18 20 Botswana 17 8 76 81 14 45 27 22 Lesotho 18 2 67 70 13 17 29 7 Zimbabwe 2 5 - - - 8 17 Ghana 7 14 __ . 22 40 Cameroon 7 8 15 20 33 Congo 2 4 96 _ - 14 29 Kenya 5 7 91 12 27 India 22 37 77.2 59.7 33 57 Pakistan 25 53 _ 42 71 Sri Lanka 99.9 76 12.0 Low Ni'donuuent(HOI Rank, 1999) Congo. Dem.Rep 12 28 _ 29 53 Sudan 18 32 32 57 Nigeria 12 19 30 4.8 Zambia 10 16 77 75 17 16 31 Tanzania 7 13 68 48 _ 17 36 Uganda 15 30 - - 24 46 Malawi 20 41 43 68 27 56 Angola - - = Rwanda 16 21 59 - 29 43 Mali 31 44 20 28 - - 54 69 Mozambigue 26 57 - 40 - 6 42 73 Bun=di 37 42 20 - - 45 63 Burkina Faso 56 79 15 31 68 87 Ethiopia 47 50 - 28 58 70 Bangladesh 40 61 | 75.1 21.6 49 71 Nepal 26 61 78.4 54.6 43 78 Medium development countries (Average) 4 6 - 94 - 10 _ High development Countries. (Averge) - 97 91 Source: World Development Indicators (1999) UNESCO Education Yearbook (1999) 58 TableVII.3: Proxy Indicators of Comparative Purchasing Power of Populations in Saharan African and South Asian Countries Population GNP Growth in National Poverty line International Per capita private Poverty Line (Dollars) consumption % of population living % of population 1998 per capita below nationally defined living below $2 per poverty line day. Avg. annual Average annual Growth rate growth rate (%1) SPPP (%) 1980-1997 1997-98 1 1985 rices) 1997 l 2015 19 Urb Rur Country % Year _](Est) _ _ _ l (yr)l Medium n eveio ent (HDI Rank, 1999) _ Mauritius ___ 1.3 3,730 4.0 5.1 _ _10.6 1992 S.Africa 41.4 49.4 3,210 -1.3 -0.1 _ _ 50.2 1993 Swaziland 0.9 1.5 1,400 -1.3 Namibia 1.7 2.2 1,940 -1L2 -3.0 Botswana 1.6 1.8 3,070 1.8 3.0 . 61 1986 Lesotho 2.1 2.7 570 -5.3 -2.7 27.8 53.9 49.2 (93) 74.1 1987 Zimbabwe 11.7 14.1 620 -1.4 0.4 10.0 25.5 (91) 68.2 1991 Ghana 18.5 26.8 390 1.9 0.2 34.3 31.0 31.4 (92) Cameroon 14.3 20.3 610 3.8 -1.3 32.4 26.7 40.0(84) Congo 2.8 4.3 680 8.4 -0.1 44.4 _ Kenya 29.3 39.1 350 0.3 0.4 46.4 29.3 42.0 (92) 78.1 1992 Sri Lanka 30.5 _ __ India 979.7 440 4.3 2.7 28.0 36.7 35.0 (94) 86.2 1997 Pakistan 131.6 470 0.5 2.0 36.9 34.0 (93) 84.7 1996 Low Devslopment (HDI Rank, 1999) Congo. Dern.Rep 4S.2 791 110 0.7 -4.5 Sudan 8.9 8.6 290 2.7 _ Nigeria 120.8 184.7 300 -1.5 -4.2 49.5 31.7 43.0 (93 59.9 1993 Zambia 9.7 13.0 330 -4.1 -3.6 88.0 46.0 68.0 (91) 98.1 1993 Tanzania 32.1 44.8 220 3.8 0.0 51.1 (91) Uganda 20.9 30.7 310 2.8 1.9 _ 55.0 (93) Malawi 10.5 15.3 210 -1.0 0.8 54.0 (91) Angola 12.0 19.4 380 16.3 -6.5 _ Rwanda 8.1 11.8 230 7.1 -1.0 51.2(93) 88.7 1985 Mali 10.6 16.7 250 1.3 -1.0 = = Mozambique 16.9 23.8 210 9.7 -0.9 . Burundi 6.5 9.2 140 2.6 -0.9 _ _ 36.2 (90) Burkina Faso 10.7 15.9 240 3.8 0.4 Ethiopia 61.3 87.6 100 -4.2 -0.4 89.0 1982 Bangladesh 125.6 161.8 350 4.2 2.1 14.3 39.8 35.6(95) 77.8 Nepal 22.9 32.5 210 0.3 2.0 23.0 44.0 42.0(95) 82.5 Medium development countries (Average) _ _ 2,950 -0.4 4.7 High development Countries. (Average) | | 25,510 1.6 2.4 _ Source: World Development Indicators (1999), Human Development Report (1999) 59 Table VIIA4: Indicators of Urban/Rural communications infrastructure provision and Employment by Economic activity (Agriculture, Industry and Services) Indicators of inequalities of fixed line network access for urban Indicators of proportion of labour force in _________________ ~~~~~and rural areas. Agricultre, Industry and services. Urban Telephone Telephone Agnicatuire Inidustry Services Population Main lines Main lines (Teledensity) (Teledensity) Male/Female Male/Female Male/Female %of total % of labour % of labour % of labour Pop. Perl100 Ihab Perl100 Thab force force force 1995 1998 1992-97 1992-97 1992-97 1997 2015 Urban jRural Main City 1Rest of Male Femn Male Fern Male Fern ________________ _____(Est) Country ________________ _____ MediumDevelopmenit(HDI Rank, 1999) Mauritius 41 48.6 14.43 ____ 30.58 20.07 15 34 24 19 46 68 S.Affica 53 46,3 6.78 ____ 41.52 7.8 16 10 42 15 42 76 Swaziland 33.0 47.2 0.03 10.88 2.37 Namnibia 30 53.2 ____ 31.73 4.46 46 54 21 8 33 39 Botswana 49 88.7 7.81 2.05 17.95 3.1 3 2 38 18 60 80 Lesotho 26 38.9 5.84 0.49 29 59 41 5 30 36 Zimbabwe 34 45.9 3.66 0.61 -7.52 0.87 23 38 32 10 46 52 Ghana 37 47.8 5.43 0.25 64 55 12 14 25 3 1 Catneroon 47 58.9 1.05 0.04 3.0 0.29 62 83 12 3 26 14 Congo 61 70.1 ____ __ 33 69 23 4 44 27 Kenya 31 44.5 3.53 0.14 7.11 0.47 19 20 2 India 28 __ _ _ _ _ _ _ _ _ _ 10.39 1.4 .,. Pakistan 36 ________ __ 6.61 1.68 44 67 2 11 36 22 _______________ _____ ______ Low ~~Develo vjint aH k,i 1999 _ _ _ _ _ Congo. Dem.Rep 30 39.3 0.46 0.04 58 81 20 5 23 14 Sudan 34 48.7 ________ 3.29 0.11 64 84 10 5 26 it Nigeria 42 55.4 1.31 0.25 43 44 9 __3 49 53 Zatmbia 39 51.5 1.66 0.2 2.3 0.59 68 83 13 __3 19 14 Tanzania 31 38.3 1.06 0.05 2.79 0.20 78 91 8 2 14 7 Uganda 14 20.7 1.48 0.04 3.74 0.08 81 88 7 2 12 10 Malawi 22 22.7 1 1.75 0.09 6.18 0.05 50 73 25 __7 25 20 Angola 33 44.1 ____ 1.74 0.31 65 86 14 2 21 13 Rwanda 6 8.9 ________ 3.97 0.03 86 98 6 1 8 2 Mali 29 40.1 1.78 0.08 83 89 2 __2 15 9 Mozambique 38 51.5 1.0 0.01 2.4 0.02 70 96 15 i 15 3 Bwrundi 8 14.5 2.45 6.16 0.02 86 98 4 1 10 I Burkina Faso 17 27.4 3.18 0.15 91 94 2 2 7 S Ethiopia 17 25.8 1.92 4.16 0.12 89 88 2 2 9 11 Bangladesh 23 161.8 1.96 _____ 54 78 11 8 34 11 Nepai 11 32.5 ____ __ Medium development 41.2 51.6 50 56 23 1 7 26 27 Countries (Average) __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 23.8 10.18 _ _ _ _ _ _ _ _ High development 77.8 81.6 5 4 34 17 53 73 Countries. (A ea e _ _ _ _ _ _ _ _ _ _ _ _ _ _ 60.2 47.76 _ _ _ _ _ _ _ _ _ _ Source: World Development Indicators (1999) African Telecommunication Indicators, ITUJ, 1998. World Telecommunication Indicators, ITU, 1999. 60 Table VIL5: Penetration of telephones and internet in SSA and South Asian countries. Telephone Local Mobile Internet Internet Hosts Internet Main lines Telephone Cellublar Service Per 10,000 Users (Teledensity) Network Subscribers Providers Inhab Per 10,000 Per 100 Inbab % 1998 % of 1998 1998 Inbab Digital (k) Total 1998 phone 1998 1998 Sub- scribers Medim Development (HDI Rank, 1999) Mauritius 21.37 100 60.5 19.8 6 5.01 108.85 S.Africa 11.46 82 2,500 33 120 32.6 285.75 Swaziland 3.05 95 4.7 13.9 3 2.92 10.5 Namibia 6.86 98 19.5 14.6 6 15.99 30.12 Botswana 5.64 100 23 21.2 6 4.19 63.69 Lesotho 0.97 99 9.8 32.8 1 0.09 0.97 Zimbabwe 1.72 50 55 20.6 0.81 7.88 Ghana 0.75 70 21.9 17.2 9 0.10 3.13 Cameroon 0.54 68.3 4.2 5.3 4 - 1.4 Congo 0.79 91.6 3.4 13.4 1 _ _0.36 Kenya 0.92 56 5.3 1.9 16 0.24 5.17 India 2.19 100 1,195 6.4 na 0.18 Pakistan 1.93 92.63 202 9.9 na 0.22 Sri Lanka 28 9 I 0.6 _ Low Development I Rank, 199) Congo. Dem.Rep 0.04 56.4 8.9 29.8 - _ _0.04 Sudan 0.57 90 8.6 5.0 1 - 0.18 Nigeria 0.40 42.5 15.0 3.5 15 0.04 0.38 Zambia 0.88 73.1 5.2 6.2 3 0.35 3.42 Tanzania 0.38 82.3 37.9 23.8 14 0.04 0.93 Uganda 0.28 90.6 30.0 34.5 4 0.05 1.95 Malawi 0.35 58.8 10.5 21.9 3 1.86 Angola 0.60 55 9.8 12.0 5 2.07 Rwanda 0.16 100 15.0 58.1 1 1.21 Mali 0.25 93 4.5 14.3 5 - 0.94 Mozamnbique 0.40 99 6.7 8.2 6 0.07 1.85 Burundi 0.29 89 0.6 3.2 1 - 0.23 Burkina Faso 0.36 89.5 2.7 6.2 3 0.16 0.88 Ethiopia 0.28 40.2 - - 4 0.01 1.01 Bangladesh 0.26 . 75 .. na 0.00 0.00 Nepal 0.84 100 0.00 0.0 na 0.007 Medium development countries (Average) 12.35 77.35 - 20.45 9.94 83.21 High development Countries. (Average) 56.06 93.5 32.2 450.43 1,396.45 Source: httD://www3.sn.apc.org1africal African Telecommunication Indicators, ITU, 1998, World Telecommunication Indicators, ITU, 1999. 61 Annex VIII List Of SSA and south Asian Countries According to GDP per Capita Groupings GDP per Capita less GDP per Capita GDP per capita US$501- GDP per capita than US$250 US$251-500 1000 US$1000 or more Africa__ _ _ _ _ _ _ _ *Mozarnbique (97) *CAR (94) *Senegal (97) Swaziland Ethiopia The Gambia Lesotho Angola Congo Dem Rep *Rwanda (92) *Nigeria (90) Namibia *Chad (97) *Uganda (95) *Cameroon (91) South Africa Burundi *Togo (98) *Guinea (99) Botswana *Niger (98) *Kenya (98) *Zimbabwe (94) Mauritius Eritrea Sao Tome & Princ' Cote d'lvoire *Burkina Faso (92) Sudan Congo Rep Sierra Leone *Ghana (98) Djibouti *Madagascar (97) *Benin (96) Cape Verde *Mali (96) *Comoros (96) *Malawi (92) Mauritania *Tanzania (96) *Zambia (96) Guinea Bissau Liberia (No Data) Somalia (No Data) South Asia *Nepal (96) *Bangladesh (97) Sri Lanka *Pakistan (90) *India (92) All data relating to information technology and communications indicators discussed in Chapter 2 is from the ITU (Intemational Telecommunications Union) STARS database, and from published sources including the 'World Telecommunications Development Report (1999) and African Telecommunication Indicators (1998). * Low Income Countries for which Demographic and Health Survey Data (DHS) data are available together with the survey year. 62 Annex IX The diffusion of ICT among the rich and the poor This Annex presents a model60 to illustrate through the use of a single diagram the following issues: (i) Why the poor and the rich use different communications techniques; (ii) Why the nature of technical change in ICT has hitherto been biased towards the rich; (iii) How the consequences of (ii) has been a widening of the digital divide; and (iv) Policy implications of (i)- (iii). 1. In the initial period, ICT consists of three fixed-coefficient techniques (oral, written word, fixed line telephony). Each technique requires different amounts of user time combined with different amounts of 'capital' (hard/software, operating skills) to produce a given amount of information. Since each technique is technically efficient, an information isoquant (qo) can be constructed as a convex combination of techniques. 2. The ratio of the hourly value of user time to the hourly user cost of capital varies between rich & poor. The value of time to the poor is low due to underemployment & low productivity, while the user cost of ICT capital to the poor is high due to liquidity constraints, imperfect capital markets, etc. Consequently, the relative price of capital faced by the poor is high (PP in diagram below). By contrast, the value of time to the rich is high as they are more likely to be employed & at a higher wage than the poor. The user cost of ICT capital to the rich is lower as they are more likely to live & work in an infrastructure-rich environment, and borrow at lower rates than the poor. Consequently, the relative price of capital faced by the rich is low (RR in diagram). 3. The implication of [2] is that the rich and the poor choose different least cost IC techniques, even if they face the same choice set (isoquant). The rich choose to communicate by fixed line telephony, while the poor choose to communicate orally. 4. At the end of the initial period, two new techniques become available (mobile telephony & the Internet). Mobile telephones save on significant amounts of associated infrastructure (transmission towers replace overhead/underground cables), but require the same amount of user time per unit of information communicated as fixed telephony. Communication over the Internet is very fast, thereby saving user time, but requires much more capital per unit of information communicated than any of the existing techniques. This pattern of technical change implies that only the relatively capital-intensive segments of the isoquant shift in towards the origin. Two of the initial techniques remain unaffected (oral, written word), while one (fixed line telephony) becomes technically inefficient (obsolete). 5. The distributional consequences of this pattern of technical change are profound, because only the segment of the isoquant (economically) relevant to the rich is affected. So, in the second period, the rich switch from communicating by fixed telephony to using the Internet, while the poor remain communicating orally. 60 The model was prepared by Christopher Scott, PREMPO, World Bank. 63 6. The effect of [5] is to widen the digital divide as measured by the angle between the capital/labour ratio of the technique used by the rich (y) & that used by the poor (a). The magnitude of the digital divide in the first & second periods is shown in the diagram by the black & red arrows. 7. The implications for a pro-poor ICT policy are clear at a general level: * Reduce the relative price of capital to the poor (improve access to training, extend electricity grid to low income areas, selective & temporary subsidies to poor users) * Shift focus of R&D in ICT to poor-user friendly techniques, ie. shift in the south- eastern segment of the information isoquant. Inforrnation isoquant (q'0) R' ty R. Information isoquant \ Intemnet \ (o Capital input I ' \ \ (hard/software, 13 skills) X I -'\ I ', \s< gFixed-line telephony '> / + Mobile telephony p1 Wttnword P ~~~~~~~~~P'~~~~P R'- R Input of user's time (clock-hours) 64 REFERENCES Barton, C (1997) "Microenterprise Business Development Services." Microenterprise Best Practices, Bethesda, MD. Bhatia, B.S., (2000)"Satcom for Extension Training", in Bhatnagar and Schware (ed.), Information and Communication Technology in Development: Cases from India, Sage Publications, India. Bhatnagar, S.C., (2000)"Enhancing Telecom Access in Rural India: Some Options", Emerging Market Forum, Asia Pacific Center, Stanford, 28-29 August. Bhatnagar, S. (2000)"Information and Communication Technologies, Poverty and Development in South-Asia", Mimeo. Brynjolfsson, E and Hitt, L (1993) 'Is Information Systems Spending Productive? New evidence and New Results.' Proceedings from the 14'h Intemational conference on Information Systems (Orlando, FL: December, 1993, pp47-64) Chakravarty, R.,(2000) " IT at Milk Collection Centers in Cooperative Diaries: The National Diary Development Board Experience", in Bhatnagar and Schware (ed.), Information and Communication Technology in Development: Cases from India, Sage Publications, India. Duncombe, R. (1999) 'The Role of Information and Communication Technology for Small, Medium and Micro Enterprise Development in Botswana' Interim results of a research project sponsored by the UK Department for International development (DFID). Institute for Development Policy and Management, University of Manchester. Duncombe, R. (2000), " Information and Communication Technology, Poverty and Development in sub-Saharan Africa" Mimeo. Duncombe, R. (2000)"An Empirical Analysis of Access and Distribution of Benefits of Information and Communications Technologies in sub-Saharan Africa and South Asia" Mimeo. Freeman, C (1988) 'The Factory of the Future: The Productivity Paradox, Japanese Just in Time and Information Technology. PICT Policy Research Paper No 3. (Uxbridge, PICT, Brunel University) Hanna, N. Guy, K. and Arnold, E. (1995) 'The Diffusion of Information Technology. Experience of Industrialized Countries, and Lessons for Developing Countries' World Bank Discussion Paper No. 281, World Bank, Washington, D.C. Heeks, R. and Bhatnagar, S.C. (1999) "Understanding success and failure in information age reform", in R.B.Heeks (ed) Reinventing Government in the Information Age, Routledge, London. ITU (1998a), World Telecommunication Development Report: Universal Access, (Executive Summary) ITU (1998b), African Telecommunication Indicators, Africa Telecom 98 65 ITU (1999a), Challenges for the Network: Internet for Development ITU (1999b), World Telecommunication Development Report, Mobile Cellular ITU, (1999)"Internet for Development". A Report published by International Telecommunication Union (ITU), Geneva. Kayani, R. and Dymond, A. (1997), "Options for Rural Telecommunications Development", World Bank Technical Paper No.3 59 Loveman, G.W (1998) 'An Assessment of the Productivity Impact of Information Technologies' Working Paper No 88-054 (Cambridge, MA: Management in the 1990s Project, MIT, 1998) i Madanmohan, R. Rashid Imran, Rizvi Hasan, Subba Rajib, (1999)"Online Content in South Asia, Opportunities and Realities, Economic and Political Weekly, November 20. Madon, S., (1992)"The Computerized Rural Information Systems Project, in Subhash Bhatnagar (ed.) Manpower and Training Needs, Information Technology Manpower: Key Issues for DCs; Tata Mc Graw Hill, New Delhi, pp. 171-79. Mansell, R and Wehn, U. (eds) (1998) Knowledge Societies. Oxford University Press for UNCSTD, Oxford. Marwah, R. (2000), "Impact of Information Technology on the Poor in Nepal" May, Marwah, R. (2000), "The Impact of Information Technology on the Poor in India", August, Mimeo. Mgombelo, H.R. and Werner, M.C.M (eds) (1997) 'Telecommunications for Business in Africa' lOS Press, Ohmsha. Mody, A and Dahnan, C (1992) 'Performance and Potential of Information Technology: An International Perspective' World Development, Vol. 20, No.12. (1992) pp1703-1719. Moussa, A., and Schware, R. (1992) 'Informatics in Africa: Lessons from World Bank Experience' World Development, Vol. 20, No 12, pp. 1737-1752, 1992. Mungunasi E. (2000)," Survey of Information Technologies within Small, Micro and Medium Enterprises in Tanzania", Mimeo. Pohjola, M (1998) 'Information Technology and Economic Development: An Introduction to the Research Issues' Working Paper No.153., The United Nations University, WIDER, November 1998. Roach, S (1988) 'White Collar Productivity: A Glimmer of Hope?' (New York, Morgan Stanley. Rodriguez, F. and Wilson, E.J. (2000) 'Are Poor Countries Losing the Information Revolution' InfoDev Working Paper, May 2000. University of Maryland at College Park. Schware, R. (1998) Thailand: Tax Computerization Project Case Study, Indian Institute of Management, Ahmedabad, Working Paper. 66 UNDP (1999), Human Development Report Van Crowder, L.(1997) 'Marketing information systems for small-scale farmers' Information Development, Vol. 13, No.4., pp179-83. Vijayaditya, N., "A Wired Village: (2000) The Warana Experiment" in Bhatnagar and Schware (ed.), Information and Communication Technology in Development: Cases from India, Sage Publications, India. World Development Indicators (2000) The World Bank. 67 Africa Region Working Paper Series ARWPS 1 Progress in Public Expenditure Management in Africa: Evidence from World Bank Surveys, January 1999, by Christos Kostopoulos ARWPS 2 Toward Inclusive and Sustainable Development in the Democratic Republic of the Congo, March 1999, by Markus Kostner (out of stock) ARWPS 3 Business Taxation in a Low-Revenue Economy: A Study on Uganda in Comparison with Neighboring Countries, June 1999, by Ritva Reinikka and Duanjie Chen ARWPS 4 Pensions and Social Security in Sub-Saharan Africa: Issues and Options, October 1999, by Luca Barbone and Luis-Alvaro Sanchez B.(out of stock) ARWPS 5 Forest Taxes, Government Revenues and the Sustainable Exploitation of Tropical Forests, January 2000, by Luca Barbone and Juan Zalduendo ARWPS 6 The Cost of Doing Business: Firms' Experience with Corruption in Uganda, June 2000, by Jacob Svensson ARWPS 7 On the Recent Trade Performance of Sub-Saharan African Countries: Cause for Hope or More of the Same, August 2000, by Francis Ng and Alexander J. Yeats ARWPS 8 Foreign Direct Investment in Africa: Old Tales and New Evidence, November 2000, by Miria Pigato ARWPS 9 The Macro Implications of HIV/AIDS in South Africa: A Preliminary Assessment, November 2000, by Channing Arndt and Jeffrey D. Lewis ARWPS 10 Revisiting Growth and Convergence: is Africa Catching Up?, December 2000, by Charalambos G. Tsangarides ARWPS 11 Spending on Safety Nets for the Poor: How Much, for How Many? The Case of Malawi, January 2001, by William James Smith ARWPS 12 Tourism in Africa, February 2001, by lain T. Christie, Doreen E. Crompton ARWPS 13 Conflict Diamonds, February 2001, by Louis Goreux ARWPS 14 Reform and Opportunity: The Changing Role and Patterns of Trade in South Africa and SADC, March 2001, by Jeffrey D. Lewis ARWPS 15 The Foreign Direct Investment Environment in Africa, March 2001, by Miria Pigato ARWPS 16 Choice of Exchange Rate Regimes for Developing Countries, April 2001, by Fahrettin Yagci ARWPS 17 Export Processing Zones: Has Africa Missed the Boat? Not yet!, May 2001, by Peter L. Watson ARWPS 18 Rural Infrastructure in Africa: Policy Directions, June 2001, by Robert Fishbein. ARWPS 19 Headquarters 1818 H Street, N.W. Washington, D.C. 20433, USA Telephone: (202) 477-1234 Facsimile: (202) 477-6391 RCA 248423 WORLDBK Cable Addr.: INTBAFRAD WAHINGTONDC Working Paper Series web address: http://www.worldbank.org/afr/wps/index.htm European Office 66, avenue d'1ena 75116 Paris, France Telephone: (1) 40.69.30.00 Facsimile: (1) 40.69.30.66 Telex: 640651 Tokyo Office Fukoku Seimei Building, 10th Floor 2-2-2 Uchisaiwai-Cho Chiyoda-Ku, Tokyo 100-0011 Japan Telephone: (81-3) 3597-6650 Facsimile: (81-3) 3597-6695 Telex: 26838 The World Bank