The Economic and Social Impact of Electronic Commerce Preliminary

Economic impact of the Internet: study of cybercafes in Gaborone, Â Botswana
Tomas Mauta Sairosse and Stephen M. Mutula
The authors Tomas Mauta Sairosse is Director of Libraries, Universidade Catolica de Mocambique, Beira, Mozambique. Stephen M. Mutula is a Senior Lecturer, Department of Library and Information Studies, University of Botswana, Gaborone, Botswana. Keywords Internet, Economic conditions, Botswana Abstract This paper discusses the findings of a study that was carried out to determine the economic impact of cybercafes in  Gaborone, Botswana. The research design was a survey. The results from the study showed that the cybercafe sector in  Gaborone, Botswana, has grown by 1,300 per cent since 2001. The sector is contributing to the gross domestic product of Botswana largely through employment, government taxation and earning the country foreign exchange through e-commerce and e-business transactions. However all was not going well for cybercafes as they were  faced with problems of competition, high tariffs, low bandwidth, and high equipment costs, among others. The study recommended the need for the coordination of cybercafe sector in order to make its growth less haphazard;  reduction of taxes on computers; enactment of cyber law, modernization and improvement of bandwidth. Electronic access The Emerald Research Register for this journal is available at http://www.emeraldinsight.com/researchregister The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0737-8831.htm
Library Hi Tech Volume 21 . Number 4 . 2003 . pp. 451-462 # MCB UP Limited . ISSN 0737-8831 DOI 10.1108/07378830310509754

Introduction
The Internet is one of the information systems much used in the world today. Modern organizations are increasingly using the Internet virtually in all aspects of society. The Internet consists of millions of interconnected computers scattered around the globe, all linked by fiber, phone lines or other cabling, using a common set of communications protocols that allow the computers to talk to each other (Bounds and Karl, 1996, p. 21). The Internet uses TCP/IP (Transmission Control Protocol/Internet Protocol), which provides a common language for interoperation between networks that use a variety of local protocols, such as Ethernet, NetWare, AppleTalk, Decnet and others. The Internet allows millions of people throughout the world to communicate with each other by transferring computer files, searching databases, exchanging electronic mail, and even chatting with other Internet users. However, due to the relatively small number of people who can afford a phone line, let alone a computer, Internet public access services are more widespread in the urban compared to rural areas in Africa (African Internet Connectivity, 2002). The imbalance in Internet access between rural and urban areas in Botswana, like the rest of Africa is skewed in favor of urban areas. Telecommunication infrastructure in Botswana A sound telecommunication infrastructure is critical for the development of Internet services. Botswana is regarded as having one of the most well developed telephone infrastructures on the Africa continent. And the government is committed to improving the sector further. The government of Botswana embarked on the liberalization of the telecommunication sector in 1996 when a Bill was enacted by Parliament which subsequently established an independent regulatory body, the Botswana Telecommunications Authority (BTA) (BTC, 2002). The liberalization of the telecommunication sector has spurred the growth of the Internet over the last four years, though the Botswana Telecommunications Corporation, remains the sole external gateway

451

Economic impact of the Internet: study of cybercafes in Botswana Â

Tomas Mauta Sairosse and Stephen M. Mutula

Library Hi Tech Volume 21 . Number 4 . 2003 . 451-462
. . . . .

bandwidth provider (Balancing Act, 2002a). Following a privatization policy that was endorsed in the first part of the year 2000, it is expected that the privatization of BTC will be achieved in the future. The telecommunications infrastructure in Botswana is composed of an all-digital microwave and fiber optic system with digital exchanges at the main centers, providing a high quality service. The current network comprises over 7,300km of microwave, radio and fiber optic links between 12 main processors, each serving a specific area of the country. The microwave network links 50 automatic exchanges and provides connections to neighboring countries such as, South Africa, Zambia and Zimbabwe (BTC, 2002). Telecommunication services in Botswana include: . national and international telephony; . data networks; . very small aperture terminals (VSATs); . private wires; . leased circuits; . toll free services; . Internet; . paging; . public telephones; . voice messaging; . telex; . packet switching; and . telegraph. Botswana has nine telephones for every 100 people in the population, consequently the country has one of the highest teledensities on the African continent (BTC, 2002). A frame relay service together with an integrated services digital network (ISDN) is also being introduced in Botswana. ISDN offers enhanced capacity, by utilizing advanced telecommunications technologies capable of carrying multimedia services. In addition to integrating speech, data, and text and video services and offering users end-to-end digital connectivity through a standardized access point, ISDN offers users faster speeds, improved communications quality and reduced operating costs. ISDN has capability to offer: . Internet access; . high quality sound broadcasts; . LAN interconnection/access;

lease line backup; transaction services; interactive multi-media; file transfer/file sharing; and desktop video conferencing (BTC, 2002).

Today there are about one billion Internet users worldwide (Nelson, 2001). Despite this growth, it is estimated that only 5 per cent of Internet growth has been attained. The Next Generation Internet (NGI) that is predicted for the next ten years will, according to (Nelson, 2001), be: . ubiquitous; . seamless; . reliable; . intelligent; and . friendly. Progressive governments worldwide are making efforts to provide a conducive environment to usher in NGI through: . education and research; . partnerships with private sector; . policy and regulatory mechanisms; and . letting the private sector to lead in infrastructure development (Nelson, 2001). Development of the Internet in Botswana In Botswana, the Internet sector was liberalized in 1999 and currently there are 14 ISPs with about 60,000 Internet users majority of them being corporate organizations. In 2001 data service sector was opened up for competition and today there are about (seven) service providers in this area (BTA, 2002). The Botswana Telecommunication Corporation, BTC opened a gateway, Botsgate, to offer a wide range of bandwidths. The BTC operates an advanced network with almost 100 per cent digital exchanges and an optical fibre transmission system throughout the county. There are some developments such as rural electrification initiatives, the modernization of government data networks, and the rolling out telephone services to rural areas, that will likely have positive impact in propelling Botswana towards an information society (Mutula, 2002). In 1996 very few organizations and individuals in Botswana had a presence on the World Wide Web. There were no national Internet service providers (ISP) and as such access was limited and expensive. A few agents

452

Economic impact of the Internet: study of cybercafes in Botswana Â

Tomas Mauta Sairosse and Stephen M. Mutula

Library Hi Tech Volume 21 . Number 4 . 2003 . 451-462

provided connections through South African ISPs. At this stage e-mail was used more than the World Wide Web. The University of Botswana was one major user with a batch connection to Rhodes University in South Africa. The Internet scene has changed drastically since then (Balancing Act, 2002a). The Internet infrastructure in Botswana has grown rapidly in the last four years. Although private Internet access is limited, many people are increasingly gaining access to the Internet at work, school or through cybercafes.  History of cybercafe Âs The first cybercafe known as Cyberia cafe   opened in London in 1994 (BBC, 2001). The term cybercafe refers to combining computer  connectivity with somewhere to have a cup of coffee or tea in a public place. The cybercafe  culture has been successfully exported to parts of the UK, USA, Canada, and Australia. Cybercafes remain places for people to meet  their friends, read newspapers, discuss important events of the day, read and answer their e-mail correspondence. Today, people communicate with their friends and workmates by means of e-mail, from their offices, or from the privacy of their own homes and now  increasingly from cybercafes (Bleecker, 2002). Around the world, cybercafes are providing  people with opportunity to gain access to the World Wide Web for different purposes. Within Africa, countries with the highest concentration of cybercafes include:  . Ghana; . Senegal; . Malawi; . Tanzania; . Kenya; and . Mozambique. In Botswana cybercafes are concentrated in  Gaborone with the exact number difficult to estimate with accuracy, because there is no properly kept record by government about the registered cybercafes (Balancing Act, 2002b;  Shoki, 2002; Mburu, 2002; Ahiabenu, 2002; Tinega, 2002). Cybercafes industry in Africa  Cybercafes have grown consistently in Africa  during 2001. In two reports, one from Ghana

and the other from Kenya, Balancing Act observes that there simply are too many cybercafes chasing too few customers (The  Telecommunication Initiative, 2002a, b). cybercafes are the most important and popular  options for access to the Internet for majority of users, since individuals and some organizations can not afford dedicated access and have to use shared access. In Ghana for example there are over 150 cybercafes with 90 per cent of these  cafes found in the national capital Accra. On  the other hand, in Kenya, Africa Online was one of the first large-scale operators to get into the cybercafes market with the e-Touch  product that offered a franchise to any company that was already operating a communications center. Africa online gave them an Internet connection, marketing services and a 50 per cent share of all revenues generated. The product was originally rolled out in six countries: (1) Zimbabwe; (2) Tanzania; (3) Uganda; (4) Cote d'Ivoire (52); (5) Kenya (160); and (6) Ghana (64) (Balancing Act, 2002a). There are currently around 300 cybercafes in  Kenya and most of which are in Nairobi the capital city. Some of the cybercafes in Kenya  are offering training (both online and offline) as a way of expanding their own markets. Others are offering training in Web design programs or software. A few others are offering CD recording, and digital imaging. In Kenya, the core users of cybercafes are aged between 23-35  with 55 per cent men and 45 per cent women. The use of e-mail is especially popular amongst this group and constitutes 60 per cent of all access time (Balancing Act, 2002a). Cybercafes in Botswana  Botswana has experienced a remarkable growth in cybercafe sector in the last two years. In  2001, the country had only one cybercafe listed  by the Global Cybercafe Directory (Chees.com,  2002). During November 2002, a pilot study that was curried out by the authors of this article in Gaborone, Botswana's capital, revealed that 13 cybercafes served on average Â

453

Economic impact of the Internet: study of cybercafes in Botswana Â

Tomas Mauta Sairosse and Stephen M. Mutula

Library Hi Tech Volume 21 . Number 4 . 2003 . 451-462

42,480 customers per year. The growth of cybercafe sector in Botswana has been spurred  by both economic and social factors according to the research findings of the pilot study. The cybercafes service providers are motivated to  fulfill a need in the market and also make money from it. On the other hand, users are motivated by both economic and social imperatives in using cybercafes. Users of  cybercafes revealed that the factors for visiting  cybercafes include:  . accessing e-mail; . conducting business; . looking for employment; . searching for education opportunities; . chatting on the net; and . generally socializing with friends within the cybercafes and on the Internet.  Economic impact assessment models Nicholas and Rowlands (2000) define, impact as a variety of phenomena along a continuum that change a behavior or transform the individual or social level. Impact assessments can be measured in various ways. Rush (2002) observes that the Tourism Economic Assessment Model (TEAM) of the Canadian Tourism Research Institution is a useful standard for economic assessment of both demand and supply side of an event. TEAM is a computer-based model designed to assess the impact of events on the economy. The model utilizes supplicated input/output methodology and econometric modeling techniques. The model also uses input/output matrices to ensure reliable, conservative analysis and reporting. The demand side (expenditure) analysis measures the impact that a dollar spent (input) in one sector of the economy impacts the output of all other sectors of the economy. The supply side (the operating expenditure) measures the impact relating to expenses incurred by the business entity, and other bodies, relating to the business or operating expenses of a business, including one-time capital expenditures. It generates a variety of economic impact measurements. Rush (2002) further, categorizes the economic benefits into three main elements namely: (1) direct effects which are the initial user's expenditures on purchases;

(2) indirect effects which are the subsequent purchases made by the service provider to sustain the direct expenditures; and (3) the induced effects that emerge when workers in the sector simulated by direct and indirect expenditures spend their additional incomes on goods and services. Schenck (2003) observes that in evaluating impact assessment, it is important to evaluate over time how the service has directly or indirectly led to socio-economic progress at the personal, community, organizational or national level. He poses some of the key questions that need to be asked such as: . What are the benefits that have accrued to the community of users since the introduction of the service? . What are the patterns of use of the service statistically? . What is the usefulness of the service to the users? . What should be done to increase the value of the service? . Who are the users of the service and what are the tangible benefits? This study relied upon TEAM, and Rush's and Schenck's models to assess the economic impact of cybercafe s. Borrowing from TEAM  model, the demand side was represented by cybercafe users' expenditures in the sector.  Similarly the supply side was represented by expenditures and costs made by the cybercafe  service providers in supplying the service. The study used Rush's (2002) models applying the three measuring dimensions of his model namely: the direct impact, which refers to the direct effects such as initial use expenditures and purchases of goods and services; indirect effects, which arise from subsequent purchases made by the service provider to sustain the direct expenditures; and induced impacts, which emerge when workers in the sector stimulated by direct and indirect expenditures spend their additional incomes on goods and services. Schenck model was useful in providing impact assessment indicators.

454

Economic impact of the Internet: study of cybercafes in Botswana Â

Tomas Mauta Sairosse and Stephen M. Mutula

Library Hi Tech Volume 21 . Number 4 . 2003 . 451-462

Research questions
This study sought to answer the following questions: . What is the demographic status of cybercafe managers and users?  . What are the users occupations? . What is the economic impact of cybercafes  (from users' and service providers' points of view)? . What are the costs of providing cybercafes  services? . What are the costs incurred by users of cybercafes?  . What are the challenges of providing cybercafes services?  . What are the prospects for the growth of cybercafe sector in Botswana?  Methodology This study used a survey method. The populations of study consisted of cybercafe  users and service providers (managers) in Gaborone, Botswana. The population was stratified into: cybercafes users and cybercafes   service providers. The users' stratum consisted of 42,480 members; 0.5 per cent of the users' population (212) was selected for study and questionnaires administered to them. A total of 207 questionnaires were completed and returned giving response rate of 97.6 per cent. Disproportionate systematic sampling was thereafter used in each cybercafe. The service  providers' stratum consisted of (13) service providers. Within the cybercafe service  providers' stratum, all the 13 managers were given questionnaires to complete, but only (11) filled the questionnaires and returned them giving response rate of 84.6 per cent. The study  used the list of cybercafes that was generated during the pilot study in November 2002 as the sampling frame. Data were collected using questionnaires, interviews and observations. The interviews and observations complemented the results that were obtained through questionnaires. Data collected were analyzed using the Statistical Package for Social Sciences (SPSS) and results were represented by both inferential and descriptive statistics.

Research findings
Demographic status of cybercafe  managers Managers were asked to state their age and gender. The purpose of this question was to find out whether there were gender related factors in providing cybercafe services. The  number of subjects who responded to this question was 11. The results showed that: . six (54.5 per cent) of the respondents were male managers aged 21-30; . two (18.2 per cent) were male or female aged 31-40; and . three (27.3 per cent) were males or females aged 41-50. The results show that most of the managers ± six (54.5 per cent) were males aged 21-30. This outcome may suggest that young people have the propensity for the Internet, presumably because of the awareness they acquired while in schools, colleges or universities. Demographic status of users The users were asked to state their gender and age in order to inform the study whether demographic status influenced the use of cybercafes. The number of respondents who  answered this question was 203. The results showed that: . 32 (15.8 per cent) were aged 11-20; . 125 (61.6 per cent) were aged 21-30; . 39 (19.2 per cent) were aged 31-40; . three (1.5 per cent) were aged 41-50; and . four (2.0 per cent) were aged 51 or more. The results further showed that 141 (69.4 per cent) were males and 64 (31.5 per cent) were females. The results showed that majority of users of cybercafes were males who belong to age range  21-30. In addition, the results showed that the cybercafes were heavily used by males ± 138  (67.9 per cent) than females 65 (32.0 per cent). The large percentage associated with males compared to females could probably be associated with the widely held perception in Africa that IT (information technology) is more a man's area rather than for women. The results are further summarized in Table I.

455

Economic impact of the Internet: study of cybercafes in Botswana Â

Tomas Mauta Sairosse and Stephen M. Mutula

Library Hi Tech Volume 21 . Number 4 . 2003 . 451-462

Table I Age and gender factor is cybercafe use  Gender Male Female 15 36 10 2 2 32.0 65 Total Male/female 32 (15.8 per cent) 125 (61.6 per cent) 39 (19.2 per cent) 3 (1.5 per cent) 4 (2.0 per cent) 100.0 203

Table II Occupations of respondents Categories Worker Marketing executive Voluntary worker Photographer Academic Missionary Graphic designer Teacher Lecturer Student Business people Diplomat Unemployed Foreign HIV/Aids researcher Artist/musician Free-lance artist Engineer Non-respondents Total Frequency 71 1 1 1 1 1 1 2 1 80 35 1 2 1 1 1 2 3 204 Per cent 34.8 0.5 0.5 0.5 0.5 0.5 0.5 1.0 0.5 39.2 17.2 0.5 1.0 0.5 0.5 0.5 1.0 1.5 100.0

Scale of years 11-20 21-30 31-40 41-50 51 or more Per cent Total

17 89 29 1 2 68.0 138

Users' occupations
Respondents were asked to state their occupations. The purpose of this question was to show whether occupation influenced pattern of cybercafe use. The number of subjects who  responded to this question was 204. The results showed that: . 71 (34.8 per cent) of the respondents were employed; . one (0.5 per cent) was a housewife; . one (0.5 per cent) was a marketing executive; . one (0.5 per cent) was a voluntary worker; . one (0.5 per cent) was a photographer; . one (0.5 per cent) was an academic; . one (0.5 per cent) was a missionary; . one (0.5 per cent) was a graphic designer; . two (1.0 per cent) were teachers; . one (0.5 per cent) was a lecturer; . 80 (39.2 per cent) were students; . 35 (17.2 per cent) were business people; . one (0.5 per cent) was a diplomat; . two (1.0 per cent) were unemployed; . one (0.5 per cent) was a foreign HIV/AIDS researcher; . one (0.5 per cent) was an artist or musician; . one (0.5 per cent) was a freelance artist; and . two (1.0 per cent) were engineers. The result showed that people using cybercafes  were from different backgrounds and profession with the majority being students 80 (39.2 per cent) followed by working and business people. This result could suggest that those groups of users in the majority have necessary awareness and skills, about the Internet and can afford to pay for the service. The results are further tabulated in Table II.

Economic impact of cybercafes Â
Expenditure by users in cybercafes  Users were asked to state how much money they spend in cybercafes. The purpose of the  question was to determine the economic dimension of cybercafes from the users' points  of view. The number of subjects who responded to this question was 201. The findings used the Botswana Pula (US$1 = P6.00) and showed that: . 87 (43.3 per cent) of the users spend P01.00-10.00; . 76 (37.8 per cent) spend P11.00-20.00; . 20 (10.0 per cent) spend P21.00-30.00; . eight (3.9 per cent) spend P31.00-40.00; and . ten (5.0 per cent) spend P41.00 or more. The results show that most of the cybercafes  users 87 (69.6 per cent) spend on average P01.00-10.00 and 76 (60.8 per cent) users spends about P11.00-20.00. Most of the respondents in the expenditure range P21.00-30.00 were conducting e-business, e-commerce, accessing e-learning facility, carried out research for academic work and entertainment, which required more time. Majority of the users who spent on average P01.00-10.00 mainly used cybercafes for e-mail Â

456

Economic impact of the Internet: study of cybercafes in Botswana Â

Tomas Mauta Sairosse and Stephen M. Mutula

Library Hi Tech Volume 21 . Number 4 . 2003 . 451-462

communications. Majority of users in the expenditure range P01.00-10.00 were probably constrained financially from carrying out prolonged searches. The results are further summarized in Table III. Reason for using cybercafes  The respondents were asked to state the purpose(s) for which they visited cybercafes.  The respondents were free to choose more than one purpose. The purpose of the question was to determine the economic dimension of cybercafe to the users. The number of  respondents to this question was 201. The results showed that: . 187 (93.0 per cent) of the respondents used cybercafes for e-mail communication;  . 45 (22.4 per cent) cited e-learning; . 44 (21.9 per cent) cited e-business; . 23 (11.4 per cent) money transfer; . 18 (8.9 per cent) e-commerce; . 18 (8.9 per cent) advertisement; and . one (0.5 per cent) for job searching. The results tend to suggest that cybercafes have  strong economic orientation to the users. The results further summarized in Table IV. Affordability of using cybercafes  Respondents were asked to state whether they found the charges levied on them by cybercafes Â
Table III Costs incurred by users in cybercafes (in  Botswana Pula) Amount spent in BWP 1.00-10.00 11.00-20.00 21.00-30.00 31.00-40.00 41.00 or more Total Table IV Cybercafe use  Cybercafe use  E-mail communications E-learning E-business Money transfer E-commerce Advertisement Job searching Frequency 187 45 44 23 18 18 1 Per cent 90.3 22.4 21.9 11.4 8.9 8.9 0.5 Frequency 87 76 20 8 10 201 Per cent 43.3 37.8 10.0 3.9 5.0 100.00

affordable. Those who responded to this question were 206. The findings indicated that: . 113 (54.8 per cent) of the users felt that the costs were affordable; . 59 (28.6 per cent) felt that the costs were high; . 17 (8.3 per cent) felt that the costs were low; and . 17 (8.3 per cent) felt that the costs were very high. The findings suggest that most of the respondents 113 (54.8 per cent) considered the use of cybercafes affordable. This finding tends  to correlate with generally observed view that cybercafe industry is so competitive that most  service providers are reducing costs just to survive rather than make profit (Balancing Act, 2002b). The results are further summarized in Table V. Economic benefits of cybercafe Âs The respondents were asked to state what they perceived as economic benefits of cybercafes.  The number of respondents were 117. Respondents were allowed to state more than one option, if it was applicable. The results show that: . 29 (24.7 per cent) of the respondents were getting access to the global business market; . six (5.1 per cent) were selling or buying goods online; . 51 (43.6 per cent) were saving money and time through Internet communication, rather than using post or telephone to contact friends and families; . 19 (16.2 per cent) were banking and transferring money using Internet facilities; . two (1.7 per cent) were saving money by importing goods online; . four (3.4 per cent) gained communication skills online;
Table V Affordability of cybercafe services  Affordability Costs Costs Costs Costs Total affordable high low very high Frequency 113 59 17 17 206 Per cent 54.8 28.6 8.3 8.3 100.00

457

Economic impact of the Internet: study of cybercafes in Botswana Â

Tomas Mauta Sairosse and Stephen M. Mutula
.

Library Hi Tech Volume 21 . Number 4 . 2003 . 451-462

.

.

two (1.7 per cent) were accessing employment opportunities online; three (2.6 per cent) indicated that they were not making any profits; and one (0.9 per cent) was using the Internet for information searches.

The results generally suggest that cybercafes are  being used for communication and business transactions. The results further underline the economic utility of cybercafes. The results are  further depicted in Table VI. Managers' perspective on the economic impact of cybercafe Âs Managers were asked to state the economic  impact of cybercafes. The purpose of the question was to determine the economic dimension of cybercafes from the managers'  points of view. Five subjects responded to this question. For those who responded to the question, the results show that: . two (40 per cent) of the respondents cited contributions to e-commerce; . one (20 per cent) cited savings compared to using the home connection; . one (20 per cent) cited business exchange (import and export); and . one (20 per cent) cited gambling online. The results suggest overall that cybercafes are  being used for communication and commercial purposes. Annual average turnover (in Botswana Pula) Respondents were asked to state their average annual turnover. The number of respondents to

this question was seven (36.4 per cent). Results showed that: . one (14.3 per cent) of the cybercafe realized  P95,125.00; . one (14.3 per cent) made P75,000.00; . one (14.3 per cent) made P30,480.00; . one (14.3 per cent) made P1,000.00; . one (14.3 per cent) made ``below P100,000.00''; . one (14.3 per cent) realized P100,000.00; and . one (14.3 per cent) made P400,000.00. The results show that the annual turnover of the most of the cybercafes was on average  P114,504.71. The highest was P400,000.00 and lowest was P1,000.00. The turnover tended to depend on geographical location with cybercafes in busy malls recording higher  returns. The results are further displayed in Table VII. Level of business growth in cybercafes  The respondents were asked to state the level of business growth in the cybercafe sector. The  number of respondents to this question was 11. The results showed that: . three (27.2 per cent) of service providers felt that the level of business growth was low; . five (45.5 per cent) felt that the business growth was fairly high; . two (18.2 per cent) felt that business growth was high; and . one (9.1 per cent) felt that business growth was highest.

In general the service providers who were making good business felt that the prospects for Table VI Economic benefits of cybercafes  cybercafes business growth were good, and  Benefit Frequency Per cent those whose business were not doing well were
Access to global market Selling and buying on the Net Save money compared with posting Banking and receiving money Saving costs on imports Gain communication skills Employment opportunity No benefit Information searches Total 29 6 51 19 2 4 2 3 1 207 24.7 25.1 43.6 16.2 1.7 3.4 1.7 2.6 0.9 Table VII Annual average turnover (in Botswana Pula) Amount BWP 95,125.00 75,000.00 30,000.00 1,000.00 100,000.00 400,000.00