Internet in Developing Countries: The Case of Brazil

Claudio Pinhanez (*)

Since the U.S. government announced the plans for the National Informational Infrastructure [1], a world-wide race for the future economic hegemony seems to have been started. While the governments of some other developed nations have already launched similar plans (although fundamentally different in terms of actual government support), the governments of developing countries are facing the issue in a quite different perspective.

My objective in this work is to analyze some of the aspects of the "informational superhighway" from the perspective of countries struggling to evolve economically. The basic question is to determine whether an "information super-highway" should be a governmental priority in developing countries .

Although it is clear that most developing countries have still basic problems such as education, poverty, housing, and health, which are sure priorities for any government, it is important to determine whether an advanced informational infrastructure provides critical economic advantages as a tool for economic leapfrogging and social development. If that is the actual case, then early investments in the provision of that infrastructure can potentially bring fast and radical changes in the future of the nation's economy.

The future use of Internet in the developed countries is certainly still undetermined (I am using the term "Internet" in this paper to define the broad range of services provided by an advanced, nationwide informational infrastructure [2]). Thus, in order to discuss the importance of Internet for those countries' economic future, I limited my investigation on how the present usage of Internet in developed countries -- and particularly in the U.S.-- is likely to be reflected in the social and economic fabric of the 3rd world.

This methodology is justified considering that there is still a long way for the developing countries to match the present infrastructural conditions of the U.S. Thus, the near term impacts of Internet provision can be reasonably expected to be similar to the present effects in the U.S. economy. Moreover, it seems to be safer to use a concrete scenario as a starting point for comparisons instead of futuristic previsions, what would be the case if we really wanted to access the future relations between economy and informational infrastructure.

I also intend to see economic development in a broad sense, meaning both the increase in business, market and jobs, and also as the improvement in the quality of life of most of the population. In developing countries, it is pointless to have economic growth without a corresponding decrease in social problems [3]. In other words, I want to investigate if the use of Internet can either expand businesses in a socially fair way, or, alternatively, can contribute to improve the equity of a 3rd World society.

The situation of the developing countries can not be easily accessed in a generic way without referring to particular elements of each society. In order to make the discussion more concrete I restricted my analysis to the case of Brazil, although most of the conclusions are believed to be valid for most countries in similar situations: specific differences will be pointed out whenever is necessary.

The choice of Brazil was governed mainly by three factors. First, the country has an interesting mixture of degrees of industrialization. Second, the society in Brazil is highly divided configuring a typical situation where 1st world's and 3rd world's conditions live together. And third, the involvement of Brazilian government in the provision of Internet services has recently been the subject of a national debate [4].

1. The Meaning of the Poverty

It is necessary to start this analysis by providing some relevant data about the economic and social conditions in Brazil. As I commented earlier, Brazil has a quite exquisite mixture of 1st and 3rd world features. Far from being an exception, the apparent contradiction between high technology and endemic poverty is an important characteristic of the developing countries, where poor children beg for money to BMW '95s at street lights [5].

Notwithstanding the low GDP/capita of US$ 2,740, the income is poorly distributed among the population: the top 1% receive 15% of the total, and the bottom 25% share 12% of the GDP [6]. Two thirds of the houses have no sewage system, although 80% of the population lives in urban areas (somewhat high for developing countries).

A major problem is literacy: 20% of the population declare themselves as completely illiterate, unable to sign their own name. Half of the population has less than 4 years of schooling and can be classified as functionally illiterate. The future is also jeopardized by an almost broken basic public school system. Starting by the bad physical conditions of most public schools, the school system also suffers from poor quality of teaching due to the low wages received by teachers, normally under US$ 4,200/year or the equivalent of a bus driver. Teaching positions are thus unable to attract qualified people and in most rural and poor neighborhoods areas the teachers have at most high school education.

The upper classes do not receive their basic education from the public system. A network of well staffed, reasonably rich private schools provide the education up to high school for those who can afford tuitions which can reach US$ 10,000 a year. College education is provided free of charge by the governmental universities which select their students in nation-wide entrance exams. The public universities get the lion's share of the government's education budget and their quality is not matched by most of the private colleges. However, the nature of the entrance exams makes extremely difficult to a high-school student from a public school to get into a public university. In practice, Brazilian's government heavily subsidizes college education for the upper classes.

A similar situation is found in the case of foreign languages learning, although in this case both the private and the public basic school systems provide poor language training. Most Brazilians who can read or speak a foreign language learned it in private language schools.

Brazilian economy can be labeled as a ``wild'' capitalism, where markets are either completely free (with no provision of anti-trust laws), or completely dominated by the government. State-owned monopolistic companies control critical infrastructure areas such as oil, electric energy and telecommunications.

Particularly important for my analysis is the situation of the telecommunication infrastructure. Brazil's telecommunications are controlled by state-owned companies since early 70's [7] . During their first decade those companies were quite successful in modernizing the telecommunications system, but in the 80's most of their investment capital was used by the government to pay for the nation's debt. The result is a situation where the provision of telephone lines is severely constrained, presently with an average of 6.8 lines for each 100 inhabitants (compared to 56 in the U.S.A.).

The service itself is of low quality. Most of the lines are old with digital switching responsible for less than 22%. A typical downtown line can be easily corrupted by noise, making impossible the connection of computers by modem even at very low rates of about 1200 bps. The reduced size of the middle classes makes the provision of wired cable TV economically unfeasible, and the few "cable" services use satellite transmission shared by small networks of up to 50 users.

The new federal government inaugurated in 1995 established the improvement of telecommunications as a priority. The shortage of funds makes the state unable to invest in the sector and forced the opening of the telecommunications market to private companies. Interestingly, the services provided by the private companies are going to be taxed, and the generated resources are destined to be invest in the expansion of telephony services to low-income families.

Finally, the dissemination of computers is sharply mirrored by the income distribution. Most high and upper middle class families have computers at home, as well as some of the best private basic schools. In contrast, the use of computers is absent in the vast majority of public basic schools (except at public universities), and computers are virtually a mystery, depicted only on TV, for most of the population.

2. Present Uses of the Internet

Since its start in the 70's the Internet has been used in a multitude of ways and for very different purposes. However, some applications clearly have been more used than others and it is helpful to display the evolution of the most popular uses of the Internet in a chronological line:

E-mail -> Newsgroups -> MOOs -> WWW -> audio-services (?) -> video-services (?)

Here I am considering possible future uses as the provision of audio-services (music-on-demand, telephony, audio-conferencing) and video-services (TV-on-demand, videophony, video-conferencing).

We can see easily that bandwidth and the need for real-time response increases from the left to the right, what is not surprising since large bandwidth applications could not be developed in the early days simply because there was no hardware to support them.

What is interesting to notice is that the uses of Internet which evolved in the primordial bandwidth-starved environment rely heavily in the written language. The early Internet users traded off bandwidth with the expressive power of the written language. Therefore it is not surprising that the first group to be "wired" was the academy for whom it is essential to have a good command of the written language.

Recently Internet applications are moving away from the written language as witnessed by the popularity brought by WWW-browsers which require shallower linguistic skills than e-mail and newsgroups. If audio and video services become available in the Internet then the ability to read and write clearly becomes much less important. However, this is possible only with a corresponding increase in bandwidth and computing power. In a few words,

The worse the hardware, the more literate is the user.

Another important aspect to be considered in this paper is the role of the Internet in the economy. In the way it is presently used the Internet is not an essential part of the economic infrastructure . Unlike telephones and roads, Internet is not used for standard business purposes. Of course this can be just a current state of affairs due to the present limitations in infrastructure. For instance, electronic catalogs have a clear role in the communications between companies and their customers; and video-conferencing and groupware applications can potentially change the way some businesses are carried out. Part of the reasons why Internet has not yet become a medium for business is related to the lack of security mechanisms. Until this problem is successfully solved the risks of using the Internet for advertising, communications, and buying are likely to prevent serious businesses to be conducted through the net.

Finally, it is very likely that the information and entertainment industries are going to be significantly affected by the existence of the Internet. But considering that most developing countries are not seriously involved in those industries, the Internet effects in those business are not likely to impact in the economic development. It is interesting that there is at least one developing country, Australia, which is viewing Internet as a way to enter the business of entertainment provision, and investing heavily in Internet infrastructure [8].

3. Using Internet Services in Brazil

In the first two sections of this paper I tried to clarify the context where Internet can be used in a developing country like Brazil. Essential to my following analysis is the idea that there are two very different Brazils which I call Brazil's 1st and 3rd worlds, and that the provision of Internet services has very different impacts in each of these worlds.

It is also important to analyze separatedly the effects of Internet in each stratum of Brazilian society because I believe that improving the living conditions of the poorers must be the priority of the government. Facing the present situation of limited resources, the Brazilian government must decide to invest on Internet provision only if economic development, in its broad, social sense, is attainable in the medium range run.

Internet In Brazil's 1st World

There is reason to believe that Brazil's 1st world is likely to use Internet in the same ways as it is used in the developed countries. Some limitations may be expected due to the conditions of the telecommunication infrastructure, driving the use, at least in the initial stages, towards the left of the applications spectrum, centered around written language.

Executives and top managers are potential beneficiaries from the Internet. They have already most of the computer and language skills needed, and good communication technology can reduce the need of personal meetings, a real problem in cities with heavy traffic such as Sao Paulo and Rio de Janeiro.

Professionals such as physicians, lawyers, and journalists are in a similar condition. The existence of professional on-line forums may also enable the sharing of experiences between professionals from different urban areas, contributing to reduce the gap between the more advanced urban centers of the southeast and of the northeast of Brazil.

Many private basic schools have already computer classrooms which are likely to be linked to the Internet. If not because of learning reasons, at least for the hype of being "modern", of being able to provide their upper class students access to the world. In the case of public universities, most of them are already connected to the Internet through research computer facilities.

High income households can be expect to join the Internet, though in a smaller degree. Video-on-demand has a particular potential in a country with high indices of street violence as Brazil has. That's the basic appeal by which cable TV is being marketed, and Internet-based entertainment services will probably follow the trend.

In Brazil's 3rd World

Mapping uses for the Internet in Brazil's 3rd world is a quite more complex task because in this case the impacts of the poor infrastructure and the illiteracy are much stronger, creating a scenario without comparison to previous usage models of the Internet which are based on the reality of the developed countries.

Although I previously included the public universities in Brazil's 1st world, I must say that universities are divided in two realities, where high quality undergraduate degrees live with low levels of academic research. The academic research in Brazil suffers from a chronicle lack of resources and industrial partnerships. In practice, the conditions of academic research in Brazil are far from similar to those found in developed countries.

However, as in most countries, the research community was the first user of Internet in Brazil. Starting in 1988 with the first BITNET nodes in Sao Paulo, the use of Internet spread through most public universities reaching about 50,000 users. Networking enabled researchers to overcome some barriers, from the cost of international calls to the access to up-to-date libraries and reference material. Although hard to measure, it is likely that Internet has been a driving force towards improving the quality of Brazilian research [9].

Paradoxically, research in Brazil is concentrated in 1st world themes, partially because international academic journals are centered around developed countries' issues. For instance, there is more research in AIDS than in malaria in Brazil because getting an international publication on tropical diseases is harder than in AIDS. This is just a small piece in the complex feedback system that keeps countries poor, which Internet seems to be unable to break. Therefore, since most of the research is not directly tied to Brazilian's 3rd world reality, we can expect few influences in the economic development of the nation even if research centers were fully connected to the Internet.

Wiring up the poor classes is certainly a fantasy in a reality of slums and hunger. However, Internet access can provide a reasonably cheap way to empower non-governmental organizations (NGO), grass roots movements, and rural and urban unions. In a context where media access is heavily controlled by the elites, Internet's low cost of information publishing is certainly appealing.

There is a whole theory about the importance of interconnecting NGOs, mainly originally from the experience of the Earth Summit'92 [10]. Some authors go further and consider that the communication among NGOs spread throughout the world is creating a "global civil society" and a "world public opinion"[11], which can potentially increase the power of the NGOs' movements. Through the Internet information can flow easier and less susceptibly to government and mass media restrictions and interests. A typical example was a killing of indigenous people in the extreme north of Brazil which was reported in a newsgroup 3 days before in the press [12].

It is also interesting that till recently the only Internet provider in Brazil (outside academic world) was a foundation involved with eradication of hunger, IBASE. This foundation was providing Internet access, through FIDONET local nodes, to Brazilian NGOs involved with housing, hunger eradication, environment, land reform, children care, and human rights.

There are, however, two big risks in this process of "wiring up" NGOs which are both related to representativeness issues. First, the ability to post information world-wide is orthogonal to the actual representativeness of the NGO. For instance, a small organization with a good hacker can create very effective WWW-pages, while another organization with more members and more popular participation may have considerably less visibility just because it lacks a member with computer skills. The existence of a reliable global civil society requires stable, representative, and credible NGOs, and relying too much in Internet exchanges can jeopardize the foundations of this structure.

A similar problem may develop inside the organizations through a process of empowering too much the middle class, literate leaders in those movements. If the NGOs start to depend too much on their Internet connections, there is a tendency of giving priority to more "academic" type of leadership and broadcasting the "1st world" versions of the solutions for the "3rd world".

Another possible use of Internet is related to the above mentioned problems in the public basic school in Brazil. I don't believe that providing Internet access to students in poor schools is feasible or useful. Most schools can not afford even to provide books for their students, and in some areas the perspective of receiving a meal is still the most powerful attractor for students. Putting computers connected to telephone lines, under these conditions, is completely unreal.

Instead we can imagine using Internet to promote teacher training. Some experiences in remote teaching using standard Internet tools have already been reported [13, 14] and can provide some ideas about likely results. Especially in the case of rural areas, networking might be one of the few feasible options for retraining the teachers. Many programs have been proposed in Brazil to solve the retraining problem, such as summer schools and regional intensive training weeks, without any success. By using a network-based training system, teachers might be able to receive support from educators based in the best universities, as well as share experiences with other teachers in similar realities but living in different regions of the country.

Teachers can be expected to have reasonable written language skills, thus overcoming the barrier of cheap hardware and narrow bandwidth. A possible problem is the very low wages received by most teachers: with the basic computer training required to network they can become very valuable workers, and might be attracted by better paid jobs in private companies and schools.

4. Conclusions

The analysis above leads to some conclusions about the role of Internet in the economic development of a developing country. The first important observation is that Internet is used by an intellectual (literate) elite . This is especially true in developing countries because the poor telecommunication infrastructure drives the use towards electronic mail and newsgroups which require a fair command of the written language. The term literate is important here because it also means that other intelligent and/or sensitive people lacking written language skills is potentially excluded, as it can be the case of grass roots leaders and artists. There is more room for a bureaucrat than for an artist in cyberspace.

Since literacy is almost an exclusive item of the upper classes, we can predict that Internet is going to be used mostly by the developing country's 1st world . The elites possess both the language skills and the buying power to be the primary users of Internet while the poorer classes, and even their representative movements, normally lack some of the requisites for networking.

This conclusion is not necessarily valid in the case of developed countries in terms of their lower classes. In more developed nations literacy and written language skills are more fairly spread among all the social classes, and the telecommunications infrastructure enables less dependency in non-verbal interaction through the net. The combination of poor infrastructure and illiteracy, concentrated in the same social segment, is the critical differentiating point.

If we add this last conclusion with the fact that Internet has not yet proved to be a basic requirement for business development, we can see that in developing countries Internet is not a vector of social or economic development . This is especially true if we consider the resolution of the social situation as an essential part of the economic development.

5. Recommendations

Considering the debate about the role of Brazilian government in the provision of Internet services, which reached national news proportions during April and May of 1995, I decided to include in this work some recommendations for developing countries' governments.

First, governments in developing countries should not invest directly in the provision of Internet services for the general public . Using as a basic assumption the fact that the real target for those governments is the developing of the whole nation and particularly the decreasing of the poverty levels, Internet can not be considered an effective tool according to the reasons contained in the previous sections.

However, considering that there is a demand for Internet services from the upper classes, governments in developing countries should incentive (by de-regulation) the provision of Internet services by private companies . Considering that most of the present market (Brazil's 1st world) has reasonable buying power, it is likely that private companies can get interested in investing in the area. In the best tradition of the Brazilian "wild" capitalism, it is also probable that those services are going to be over-priced due to lack of competition or cartelization. But since Internet activities are not essential neither for the economy nor for the eradication of poverty, high prices for Internet use are not likely to affect the present or the future prospects of the nation.

There are, though, two issues which must receive attention from the governments. First, governments in developing countries should invest in networking research institutes . The main reason behind this recommendation is that it is still a simple and cheap way to increase the quality of academic research, as the experience at some Brazilian universities suggest.

Finally, governments in developing countries should devise mechanisms to promote the access to Internet for NGOs, grass roots movements and public basic schools . The objective here is to increase democratic participation by empowering the voices of the lowest strata of the society.

My proposal to implement this last recommendation is to create a tax on Internet use, collected into a fund to support connection of NGOs and public schools . Instead of subsidizing universal access at the personal level as it is necessary in telephony, it seems to be more productive to connect to the Internet the organizations which deal with social development.

Just as an example, the fund could establish a program which would provide to registered organizations a laptop computer with modem and an Internet account (from one of the private Internet providers). The organization should be responsible for putting on-air a WWW-page and to participate in pertinent newsgroups.


*. O autor:
Prof. Instituto de Matemática e Estatística da USP (1987-93).
Doutorando no MIT — The Media Laboratory.
WWW page: Claudio Pinhanez
  1. See IITF. 1993. "The National Information Infrastructure: Agenda for Action." Information Infrastructure Task Force. (September 15, 1993).
  2. Using the term "Internet" to refer to the services provided by the NII is reasonable in this paper if we consider that: first, the Internet model seems to be winning increasing popularity among NII designers; second, the Internet is the only presently working model from which inferences can be made; and third, in the context of a developing country the prospects for a network with more bandwidth are still reduced, favoring the adoption of the present Internet scenario as a working model where observations about usage patterns can be withdraw.
  3. See Heinz Arndt, "The Rise and Fall of Economic Growth", The Univ. of Chicago Press, Chicago, 1978.
  4. Some references about this debate, in Portuguese, can be found at
  5. "Pixote", directed by Hector Babenco, is a good movie to obtain a clearer impression about the reality of Brazil and its 1st and 3rd worlds. "Once were Warriors", a movie by Lee Tamahori, can also provide interesting ideas about the nature of poverty in the New Zealand's context (for more information, check The Internet Movie Database or your local video store).
  6. The April 29, 1995 issue of "The Economist" includes a survey about Brazil from which most of the data presented in this paper was collected.
  7. An extensive evaluation of the past, present, and future situation of Brazilian telecommunications is presented in Russell Neuman et. al. "Brazilian Telecommunications in Transition: A New Strategy for Competitiveness", White Paper, published by MIT, Tufts University, and Fundacao Getulio Vargas. (Also available in the reader of MAS-962, Rotch Library, MIT).
  8. Max Lenoy's final project for MAS-962 focuses on the Australian governmental decision to make the country an information provider in the Internet.
  9. See Beryl Bellmand et. al., Technology Transfer in Global Networking: Capacity Building in Africa and Latin America", in Linda Harasim (ed.) "Global Networks: Computers and International Communication", The MIT Press, 1993, pgs. 237--254.
  10. Refer to the chapter "Computers and Developing Countries" in John E. Young, "Global Network: Computers in a Sustainable Society", World Watcher Paper 115, The Worldwatch Institute, Sep. 1993, pgs. 29--36.
  11. See Howard Frederick, "Computer Networks and the Emergence of Global Civil Society", in Linda Harasim (ed.) "Global Networks: Computers and International Communication", The MIT Press, 1993, pgs. 283--295.
  12. Gabriela Nava-Campos's final project for MAS-962 describes another interesting example, analyzing the Internet usage by the Zapatista army in the Chiapas region, Mexico.
  13. See Andrew Feenberg, "Building a Global Network: The WBSI Experience", in Linda Harasim (ed.) "Global Networks: Computers and International Communication", The MIT Press, 1993, pgs. 186--197.
  14. See Margaret Riel, "Global Education through Learning Cycles", in Linda Harasim (ed.) "Global Networks: Computers and International Communication", The MIT Press, 1993, pgs. 222--236.

retorna ao índice dos anais