_~\":

 

' Afi-OI'IMVLCQ,‘ _. «—-agu--c:.vuv~
- V

 

n1»

. 5figam%~‘ac

?.:£

 

’ r
—-.--. ,.
..‘...,,,..

H...‘

aunt I‘-

'1‘

~~~.t..r.~ V“. a....v-v
.5). , , —

 

OUU‘IV"VII
NE

Itiiiiii’l‘l’ililiifiliil Liliiiiufiiiiiiiii

3 1293 00778 6936

 

 

'\

LIBRARY
“101113311 State
University

 

 

This is to certify that the

thesis entitled

Science and Technology Planning and

Implications for the Development of
Indigenous High Technology Capabilities
in LDCs: An Analysis of the Brazilian
Computer Industry Pfifimfifiiby

Ritin Singh

has been accepted towards fulfillment
of the requirements for

 

 

 

Masters clegreeinCommunication
\ W
. - '
Majo ofessor
Date—BLLAQQLJB 9 2

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE ll RETURN BOX to remove this checkout from your record.
TO AVOID FINES Mum on or More due due.

II DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

flT—i .

MSU Is An Affirmdlve Action/Equal Opportunlty Instituton
emu”!

 

SCIENCE AND TECHNOLOGY PLANNING AND IMPLICATIONS FOR THE
DEVELOPMENT OF INDIGENOUS HIGH-TECHNOLOGY CAPABILITIES IN
LDC'S: AN ANALYSIS OF THE BRAZILIAN COMPUTER INDUSTRY

By

Ritin Singh

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Connunication and Urban Studies

1992

ABSTACT
SCIENCE AND TECHNOLOGY PLANNING AND IMPLICATIONS FOR THE

DEVELOPMENT OF INDIGENOUS HIGH-TECH CAPABILITIES IN LDCI: A
CASE STUDY OF THE BRAZILIAN COMPUTER ININJSTRY

By
Ritin Singh

LDCs share a growing concern of the value of new
communication technologies for their economic growth.
Therefore, the strategies they adopt to acquire and/or
develop technologies becomes a key issue. In this context,
the role of the state in stimulating, shaping and inhibiting
technological development is believed to be crucial.

The Brazilian computer industry underscores problems
which emerge as a direct result of new technologies.
Problems such as the acquisition of foreign technology, and
the need to develop a domestic high-tech industry.

This thesis investigates two closely interrelated
propositions. rings; that indigenous capabilities can be
fostered by pragmatic science and technology policies. Such
policies focus primarily on the selective application of
regulatory mechanisms. figggng‘ domestic policies and
technology transfer via TNCs is not necessarily
contradictory but can be complementary. Establishing
strategic relationships with TNCs can help in acquiring the
more sophisticated technologies.

On analyzing the Brazilian computer industry, both

these propositions are reconfirmed.

ACKNOWLEDGMENTS

I would like to acknowledge my committee members, not
only for their academic input but also on the often ignored,
though equally important, ‘behind the scene' support.

Dr. David Johnson, for his patience and criticizms
while reading endless drafts, out of which emerged this
document. Dr. Joseph Straubhaar, for his constant
willingness and ‘open-door' policy, when it came to
discussing Brazil. And most importantly, for opening my eyes
to the world south of the 0.8 border. Dr. Bella Hedy, for
allowing me to handle thorny Third World issues on my own.
Not to mention her support, without which I would not have
been at Michigan State University.

Last, but by no means least, I must acknowledge the
support and sacrifice of Mary Rolston, for whom another
deadline had become almost as hazardous as a dentist

appointment.

TABLE 01" CONTENTS

Introduction

Literature Review

Case Study: Brazilian Computer Industry

Implications for Science and Technology
Planning in LDCs

Conclusion

Bibliography

iv

15

51

96

115

124

C BIA P T I R I

INTRODUCTION
1. Problem

Recent advances in communications technology have been
described heralding the advent of a new technological
revolution that will drastically transform the shape of
contemporary society. The high speed information processing
potential of these technologies in combination with their
flexibility and wide ranging applicability, can lead to
significant improvements in productivity in industry,
agriculture, office management, education and health
(Parker, 1984). New communication technologies include
satellite based TV broadcasting and long distance
telephony: interactive computer technologies : VCRs: fibre
optic cables: etc.

With the integration of computers and communications-
these new technologies have also been called 'Telematics”.
These two elements of telematics, namely, telecommunication
networks (communication) and informatics (computer hardware
and software) have been heralded as the future base for
advanced economic development (Sweeney, 1981). Host of the
interest in these technologies has, predictably, been in the
industrialized world--i.e. the United States, western Europe

2
and Japan. Nevertheless, increasing attention is being
given to the potential applications of these technologies in
the lesser developed countries (LDCs). Though some
cautionary voices have been raised, the majority of the
reports and studies have focused on the benefits that could
accrue to these countries (often treated as one homogenous
entity) through the adoption of the new technologies.

The LDCs too share a growing concern and awareness of
the value of these communication technologies for their
future. It is important to recognize that what is sought is
not only the acquisition of foreign modern technology but,
equally important, development of indigenous capabilities.
The LDCs are cognizant of the major transformation taking
place worldwide towards an information-based environment, as
well as the advantages that accrue from use of modern
technology to spur their development process. Policy makers
do comprehend the importance of telematics in offering the
means to integrate national resources with international,
and in supplying the required inputs for domestic
technological development. They largely realize that it
would be foolhardy and impractical to opt out of the
networks of satellites, computers and digitalized
telephones, if one wants to participate in the global
economic order (Jayaweera, 1987).

LDCs, however, fear that their lack of resources and

infrastructure to develop all aspects of telematics

3

(informatics and communication technologies) might lead them
to 'miss the boat'. The idea of developing some parts of it
domestically, by focusing on particular sectors, does seem
appealing and a part of the solution (eg: computer industry
in Brazil). This brings us to the crux of the problem which
this thesis deals with, namely, what strategies can LDCs
adopt to acquire and/or develop advanced communication
technologies and yet maintain or develop their autonomy?

The desire to be autonomous and self-reliant, in these
technologies, has been a goal for some of these newly
industrializing countries (files), such as Brazil, Mexico,
India and South Korea. And they have struggled hard to
achieve this difficult goal in various industrial sectors,
failing in some and achieving success in others. Complete
national self-reliance is elusive for all countries today.
But this is particularly the case for LDCs in light of the
fact that control over new technologies is becoming
increasingly concentrated and the prospects of complete
technological autonomy ,even for the industrialized world,
disappearing (wad, 1982). We are still at the stage where
the options and possibilities of these new technologies are
being articulated. What does seem relevant however, at this
point in time, for national policy makers is to understand
the reality and design strategies for the building of
indigenous technological capabilities (which also entails
the effective acquisition of foreign technology).

Unfortunately there is no universal model for

4
technological development, and it does seem a
multidimensional and dialectical process changing from
society to society. Therefore each country should learn to
discover and execute its own strategy--one which is well
within the framework of its economic, industrial, political
and social capabilities.

Even though over generalization is dangerous, some
points do emerge as central in all this. Most importantly,
the role of the state in planning or even controlling the
development of new communication technologies is seen as
_crucial for almost every country. The way in which goals
are set up now will influence all future developments, and
many think only a coordinated, national policy can attack a
problem of this magnitude.

The related common problem is that transnational
corporations (TNCs) are major actors in most LDC markets and
economies. And it is primarily in the context of TNCs that
transfer of technology can take place (Hobday, 1986).
Unfortunately, the propensity of these corporations is to
decentralize commercial and industrial activities, but hold
tight control over research, development, investment, and
marketing of technologies. The immediate policy response to
this situation might be to impose trade barriers and tighter
control on the TNCs activities, but this is likely to
jeopardize technological development of LDCs.

The early 1960s to the late 1970s had, in fact, been the
period when TNCs were increasingly being regulated in LDCs.

5
The objective of such comprehensive policy measures was to
achieve Import Substitution Industrialization (181).
However, with passage of time, and the increasing failure of
this policy, governments have been shifting away from rigid
controls to more open economic policies. Simultaneously,
industrial policy has become the focal point in national
efforts to develop indigenous scientific and technological
capabilities.

To this end, this thesis addresses the issue of
technological development through the dual process of
indigenous technology development and strategic technology
transfer.

2. Thesis Statement and lethod

This thesis deals with the issue of the
adoption/creation of new communication technologies in LDCs.
I would like to argue that a holistic approach to indigenous
technology development is not always to be found, and a
dynamic field such as technology transfer studies is
somewhat compartmentalized. Concern over technology has
been normally confined to technological and exogenous
forces. Other internal/international non-technical issues
are usually ignored. One of the main benefits of this focus
on the issue of technology development is that a systematic
view is being taken on a highly syncretic field. To this
end, I have tried to incorporate analysis, theories, and
explanations from literature that deals with communication,

economics, channels of transfer, and the role of the state

6
in science and technology planning. The issues highlighted
explain the phenomenon of the development of indigenous
technological capability and technology transfer (TT).

At the outset, it is important to highlight the inherent
nature of TT (domestic or international) as a communicative
process. It is the communication of information (in the form
of knowledge) between people within organizations and
outside which enables this process. Industry-government,
university-industry, and TNCs-local government linkages, to
name a few, are patterns of communication and exchange. and
normally these linkages between universities, government
laboratories, and private industry are established to
encourage TT. This definition is, however, very inclusive,
since it includes informal as well as formal structured
relationships. Whatever the form it takes the key to the
process is that people transfer technology, and in turn form
a multitude of relationships.

This analysis further lays the explanatory groundwork
for the introduction and analysis of the Brazilian computer
industry, which provides an example of the above stated
ideas. The significance of the case study is that it helps
in understanding the problems which emerge as a direct
result of development of new technologies on a global scale.
Problems such as dependency, the acquisition of foreign
technology, and the need to develop a domestic high-tech
industry capable of producing new technologies appropriate

for its use.

7

The choice of Brazil and its domestic Informatics
policy/ computer industry for the study is particularly
appropriate. Brazil, with the eighth largest economy in the
non-socialist world, is one of the so-called
industrializing, or ‘semi-peripheral' LDCs and boasts a
rapidly expanding industrial capability. According to a
report published in 1986 by the Office of the U.S Trade
Representative, “Brazil is less dependent on imports than
nearly any other nation in the world" (New York Times,
1986). This industrial capacity includes a rapidly growing
advanced technology sector which has received special
attention from the most powerful levels of the government
for several decades.

The development of the computer industry was yet another
link in Brazil's already operational communication system.
No new social development applications were planned. As a
matter of fact, the entire policy to create a new industry
was geared towards commercial and economic interests. This
does raise the question of who benefits from these
technologies: and does the benefit spread equally among all
social and economic groups. It is obvious, on analysis of
the computer industry, that the state did not try to solve
the "old problems", like the spread of social services more
equitably, and rural-urban integration. It has instead
worked within the framework of industrial development
(Nettleton, 1986).

By focusing exclusively on the Brazilian context, this

8
study does not mean to imply that the international patterns
of technology dependence are irrelevant to the issue.
Rather, as the ‘rigid' dependency paradigm begins to give
way in the face of incontrovertible evidence of increasing
technological and economic capabilities of industrializing
nations, such as Brazil, 8. Korea, India and Mexico, new
approaches must begin to explore the dynamics of this
transformation (Pransman, 1985). Clear out foreign control
which in the past dominated Brazil's technology
infrastructure has evolved into a complex interplay of
forces between government, foreign and national industry,
and civil society. And if developing domestic technological
capabilities is the objective, then our understanding of
this new LDC technological environment is crucial.

The case study can contribute uniquely to our knowledge
of organizational, social, economic, and political
phenomena. It enables one to investigate structures of a
given industry (for example the computer industry in
Brazil). In this situation, the distinctive need for a case
study arises out of a desire to understand complex and
interrelated phenomena, holistically (Yin, 1989, pp.14-22).
Case studies also allows the investigator to ask the 'how'
and "why" questions of a given event. As Schramm so
appropriately stated:

“The essence of a case study is that it tries to illuminate
a set of decisions: why they were taken, how they were
implemented, and with what result" (1971).

These are the primary questions which will be investigated

9
and then explained by the propositions laid out later in the
chapter on regulations and technology planning in Brazil.

By highlighting the complexity of the Brazilian computer
industry, and the various factors which effected its
development, one can justify the case study method. Yin
claims that a case study:

'* Investigates a contemporary phenomena within its real
life context: when

* the boundaries between the phenomenon and context are not
clearly evident" (1981, p.59).

Chapter 3 explains why and how contexts and phenomena in
Brazil were interrelated in the development of the computer
industry. A

The model developed for this study is wholly
qualitative. Documentation analysis, utilizing academic
literature, is the primary source for the study. It is
important to note here that most of the research on the
development and performance of the Brazilian computer
industry was undertaken by relatively few scholars. I have
in turn utilized their studies to further evaluate them for
investigating and inferring my propositions.

It is important to note that in analyzing the evidence,
by focusing attention on certain data and ignoring the rest,
the propositions have helped in the organizing the entire
case study. These propositions have been useful in guiding
and supporting the case study analysis, and in turn the
implications for other LDCs.

3. Propositions

10

The objective of this thesis is to investigate two
primary, and closely interrelated propositions, through a
review of literature and the analysis of the Brazilian
computer industry. These propositions deal exclusively with
science and technology planning policies within high
technology industries in LDCs.

The first proposition claims that the development of
indigenous technological capabilities can be fostered by
introducing pragmatic science and technology policies. Such
policies enable the acquisition, as well as the
assimilation, of new technology. Pragmatic policies are
those which do not get bogged down by ideological
considerations, but in fact are a reflection of the
infrastructural realities at hand. A primary focus of such
policies is the selective application of regulatory
mechanisms which enhance domestic capabilities.

This selectivity comes in the form of decision making on the
levels of technology to be developed, and the relevant
regulations to be introduced to develop such capabilities.

The figggnn proposition is an offshoot of the first one.
This posits that domestic technology development policies,
and technology transfer via TNCs are not necessarily
contradictory but can be complementary. Developing a
strategic working relationship with TNCs is more likely to
help int the acquisition of the more sophisticated
technologies, since the possibility of creating it

indigenously is very difficult.

11

On analyzing the Brazilian computer industry both these
propositions are reconfirmed. The formalization of the
Informatics policy strengthened the science and technology
objectives of developing indigenous capabilities in the
computer industry. This policy clearly laid out the
foundation in terms of infrastructural development, such as
R a D, human resources, component supplies, etc. The policy
makers were quite pragmatic in their decisions regarding
regulations. The market reserve for the lower end of the
industry highlights their focus on keeping certain selective
levels of technology for indigenous development.

On the other hand, the Brazilians had initially
misperceived their technological capabilities. Having closed
the microcomputer market to TNCs, they had plans of
achieving domestic capabilities at the more sophisticated
end too. On realizing their incapacity to do so, technology
was acquired by utilizing a variety of technology transfer
channels.

These channels and relationships with TNCs were
strategically selected depending upon the sophistication of
the technology. For example, technology for minicomputers
was licensed from TNCs. And in the case of superminis, it
was direct foreign investment (DPI), and the creation of
local subsidiaries by TNCs, which allowed for technology to
be transferred. There were, however, regulations laid down
for the TNCs to export a certain number of minicomputers.

Since the mainframes were intrinsically too sophisticated

12

for local manufacturers the TNCs were the sole manufacturers
and suppliers. It is relevant to note at this point that as
the level of sophistication increased, which also meant that
it was beyond the capabilities of domestic manufacturers,
relationships with TNCs increased.
4. Chapter Arrangement

Chapter 1 is an introduction to the thesis. Chapter 2
is a review of the comparative literature on the mainstream
and critical approaches to communication technology
development. It lays the theoretical foundation of the
contrasting viewpoints on the significance of new
technologies in LDC development. I also review the concerns
of development economists whose primary focus is on the
indirect costs that LDCs bear when transferring technology.
Their suggestions on the possible technology strategies that
industrializing countries can pursue is explored to
underscore the one which Brazil chose. Finally, the over
riding concern with the role that the state can play in
stimulating, shaping and inhibiting technological change is
analyzed. This activity comes in the form of direct
regulations and the building of indigenous technological
capabilities. Following that, I review the channels through
which technology can be transferred, such as, joint ventures
and licensing agreements (Arm's length) and Direct Foreign
Investments (DPI), through TNC subsidiaries. The current
trends in technology regulation towards greater flexibility

and ‘new pragmatism' as pursued by governments ends the

13
chapter.

The rational for incorporating this analysis is that it
lays down the explanatory framework for the analysis of the
Brazilian computer case. There is a marked difference
between the two main sections in chapter 2. The first
section, dealing with communication theorists, clarifies on
the conceptual issues of technology and development. Whereas
the second part, bringing forth the viewpoints of
economists, stresses the application oriented issues of
technology transfer, strategies, and planning.

Chapter 3 deals exclusively with the Brazilian case
study. It basically gives one answers to the question of
what Science and Technology strategies a government can
adopt to improve indigenous capabilities, chiefly through
regulations. This chapter is in four sections. Part 1
details the technology policies of the Brazilian government
since the 1960s. This allows the computer policy to be
viewed in light of other technology policies and highlights
its technological advances. Part 2 gives a brief review of
the Informatics Policy. The analysis of the development of
the computer industry, which underscores the role of the
state in providing conditions for its development is
examined in Part 3. The major influencing factors for the
need to develop nationalistic policies (market reserve) for
indigenous technological development, are the key issues
analyzed. Part 4 evaluates the success of the policy in

increasing domestic production capacity. The key areas

14
assessed are the process of technology assimilation,
development of indigenous capabilities, costs and effects on
users, production capacities (economies of scale), and human
resources development.

Even though Chapter 4 and 5 utilize the Brazilian
policies for the computer industry as a foundation, they
incorporate different levels of analysis. While chapter four
clearly lays out the practical implications of pursuing
strategic science and technology strategies for other LDCs,
chapter five concludes on the theoretical ramifications of
technology and development.

More specifically, Chapter 4 reviews the successes and
failures of the Brazilian policy and reiterates the premise
on which the thesis is based, namely, that technology
policies for the development of local capabilities and
technology transfer through TNCs is possible. It further
discusses the implications and the role the state can play
in science and technology planning and in industrial
technical change, while accepting foreign technology
transfer. The strategies and policies which can not only
enhance indigenous capabilities but also the acquisition and
absorption of foreign technology is the central theme of
this section.

Chapter 5 sums up the discussion and suggestions made in

the thesis and theorizes on its implications.

mu

LITERATURE "VIII

The first section of this chapter reviews the
literature on new communication technologies in LDCs from
the opposing theoretical perspectives of “mainstream! and
'critical' communication scholars. The appropriateness of
new technologies and technological development in LDCs is
analyzed. This is followed by the concerns of development
economists who primarily focus on the role that the state
can play in stimulating, shaping, and inhibiting
technological change. Their suggestions on possible
technology strategies that industrializing countries can
pursue underscores the one that Brazil chose. And, finally
their concern with the high indirect costs LDCs pay when
transferring technology from TNCs, is analysed. The third
section discusses the various modes of transferring
technology, ranging from Direct Foreign Investments (DPI) to
Joint ventures 8 Licensing. The current trends in technology
regulations towards greater flexibility and pragmatism, as a
result of economic liberalization is also reviewed.

It is important to note here that the first section, on

the communication perspective, deals specifically with

15

16
theoretical issues. The other three parts highlight the
practical and applied aspects of science and technology
planning within LDCs. The ideas discussed, though distinct,
heve the value and importance of being applicable in LDCs.
Last, but not the least, elaboration on these perspectives
lay the groundwork for the terms, definitions, and state
role in science and technology planning, for analyzing the
Brazilian computer industry. By understanding the various
viewpoints it becomes easier to comprehend the dynamics
under which LDCs have to plan and address science and
technology issues.
1. Gem-unioatioa Perspective
The mainstream approach emphasizes the theme of
political development, nationalism and development, and
modernization through assistance in capital and technology.
These studies tend to be empirical analyses of discrete
projects, and emphasize attention to economic
rationalization as a basic route towards improving
efficiency. In contrast, the critical approaches focus on
themes of underdevelopment, dependency and imperialism.
This approach starts from a historical perspective, often
analyzing topics according to the dialectical process of
transformation caused by foreign penetration, and questions
basic assumptions that societal structures are a fixed given
(Nattos, 1982). It also assumes the power of the

transnational corporate system to be overwhelming, and its

11

ideological control so hegemonic, that there seems little
possibility of technological or economic change (Tsui,
1991) .
W

The informally labelled sub-specialty, development
communications, had its start in the late 1950's when North
American communications researchers began to look at ways
their expertise could be used to help bring about progress
and modernization in LDCs. They believed that the
mechanical application of the economic, technological, and
political systems of the west could solve he problem of
backwardness. They theorized that Western-style
modernization was a benefit and could be brought to the
underdeveloped via communications technology and planning.

With roots deep in the bedrock of nee-classical
economics, they assume that the basic challenge to be faced
in economic development is increasing economic productivity.
Schramm (1968) considered the most productive sector of
modern society to be the industrial sector, which could
benefit from communication technologies, thus contributing
to modernization and industrial growth. however, since the
LDCs lack the necessary communication infrastructure to
implement such large scale development projects, Schramm
suggested a massive transfer of communication technology,
e.g. radio, television, computers, and satellites, from.the
West .

This emphasis on economic growth has been the benchmark

I.
which “mainstream” scholars have utilized to influence their
adoption of new co-unicatien technologies. This has also
been extremely influential in national (for example the
Brazilian computer industry) and international planning
programs. The contexts they look at are those organizational
and infrastructural constraints that obstruct the full
realisation of the potential of these technologies.
Constraints include lack of expertise, finance and
technology (Budson, 1985).

Research has often focused on the planning and
evaluation of specific communications projects, for social
applications as well as commercial ones. Inspite of this
project oriented research, there has been a trend in recent
years to recognize the importance of social, political and
economic contexts, which has been linked to the growing
awareness that technology is not politically neutral. There
is also a growing awareness of the need to study financial
and infrastructural problems (Bruce, Cunard & Director,
198:).

The point to note here is that relatively few studies
have been undertaken to evaluate the role of new
technologies in development. {Much of the research has been
in favor of the mass media. The little research that was
done in the past had usually been sponsored by the private
sector or government agencies, who often assessed benefits
by utilizing direct return on investment (ROI) as the
exclusive criteria. A few studies focused on the study of

19
indirect benefits and effects, to a developing country,
where the researchers stressed the importance of technology
as an essential tool to development. Their conclusion was
that the positive effects of these technologies spread
beyond the users to benefit society and the economy of LDCs
in general, although a certain threshold level of
infrastructure was required to be truly effective (Parker,
1971: 1982: 19s4)

This obviously calls for a better understanding of
technological change, and the contribution of new
communication technologies to economic development. This
can lead to an increase of investments and help eliminate
the felt and real risks which LDC policy makers feel when
putting money into a sector which does not really seem as a
priority (Budson, 1984: 1985). Others now stress the
importance of integrating communication technologies within
general economic development programs which will facilitate
economic growth and ruraldmetropclitan integration.
National integration is seen as an important precondition
for a successful integration of a country into the global
economic system. And rural-urban integration is essential
to national integration. Consequently improvement of the
communication infrastructure becomes the single most
powerful engine of such integration and sustained industrial
development (Parker, 1977: 1982: 1984).

Parker further stresses the point that the question of
appropriate technologies, and the ability of LDCs to procure

8.
equipment is more of an institutional problem.than
technical. Problems can be solved if consensus among LDC
borrowers, international lenders and equipment suppliers is
achieved. Not only will this result in the installation of
appropriate technology but also a resolution of foreign
exchange problems, which LDCs more often than not face
(Parker, 1985).

Pool has argued that LDCs should not restrict the flow
of information and technology transfer. One of his main
beliefs has been that dependence occurs when LDCs are unable
to acquire know-how from advanced nations, which usually is
the case because of the restricted flow of information. he
has also been extremely confident in the potential
applications of new communications technologies in LDCs,
firmly believing that technology is the response to social
needs (1979). Be is basically a 'free' market advocate,
objecting to any form of control or intervention by the
state or elites. Accordingly to this stalwart of mainstream
communication research, privatization and foreign
investments, as communication policy planning tools, were
key to modernization and development (Pool, 1983).

Rogers has been one of the most active mainstream
scholars for modernization and the “diffusion“ of
communication technologies in LDCs. Bis belief has been
firmly grounded in the utility of these technologies (Rogers
1988: 1989). The core of his argument is that the efficient
distribution and feedback of information in society is

31
facilitated by new communication technologies. These new
technologies also have the potential to enable LDCs to
leapfrog the industrial era, and generate sufficient
economic activity in terms of employment, manufacturing and
productivity, thus resulting in an information society.

The mainstreaa scholars have build an explicit
integration model, which basically maintains that technology
produced in the West is appropriate to the needs of the
LDCs. And the most efficient and adequate mechanism for
transferring the technology is commercially, through TNCs.
The rational for this strategy is based on the proposition
that (a) commercial technology transfer (TT) through TNCs
offer all necessary combinations of technology components:
and, (b) TT needed to establish any new productive facility
is a complicated process, requiring special knowledge and
skills for each stage of its development (assuming LDC
ineptitude). horeover, their belief is that the success
TNCs have in transferring technology is due not only to
their mastery of the whole ”technology package“, but also to
the fact that this package is closely integrated in
management, marketing and financing skills. Consequently,
the argument follows that any attempt to separate elements
of technology transfer will either fail or result in
considerably higher costs (Pool, 1983).

MW
Not only did the modernization-inspired communications

88
projects run into practical operational difficulties, some
theoretical deficiencies were discovered in the mainstream
approach as well. Strongly influenced by Narxist and neo-
Narxist thinking, the critical approach to development
communications was born in response.

This approach rejects the premises of the integration
model. Critical scholars believe that the social and
economic structures and problems of LDCs are different, so
western technologies are not appropriate for the
industrially weak countries. LDCs that have accepted
technology imports, are today more dependent upon the
outside world, since indigenous science and technology
capabilities were never developed. The way out of this
problematic is to limit technology imports and turn to
import substitution and internal technology development
programs.

For marxist scholars like Schiller economic development
is viewed from a global context. he argues that the poverty
stricken state of the LDCs is a consequence of their
economic exploitation by the industrialized countries (and
the elites in LDCs). This has created an inequitable
situation, in which the West is growing wealthier at the
expense of the LDCs. While these countries grow wealthier,
the poorer countries become all the more dependent, making
it difficult to pull them out of their poverty (Lee, 1980,
pp. 31-35).

he sees culture/communications as absolutely critical

I)
in the future struggle against capitalism, external cultural
domination, technological dependency, and the implementation
of new technologies. Be is negative about new communication
technologies because he feels that they only reinforce
existing capitalist structures, while social applications
and underlying inequitable structures are ignored (1975).
Using a political economy approach Schiller focuses on how
TNCs subject LDCs to economic and cultural dependence on
industrialized countries. Technology plays a vital role in
the emerging scheme of things. It serves dually: first, to
integrate the transnational corporate system, and second, to
deepen the dependence of the peripheral world on hardware,
software, training and administration supplied by that
system. Any transfer of technology from this system to LDCs
would only exacerbate the vicious cycle and increase the
existing dependence of the latter on the former and thus
further heighten underdevelopment.

According to Schiller, the new communication
technologies have been developed out of the felt needs of
the governments and corporations of the West and are hardly
likely to be appropriate for the LDCs in any case (1981).
Analyzing Schiller's research Tsui (1990) suggests that:

...his proposed strategy is to delay, postpone,
and defer. he calls for a deceleration in the

rush to adopt new technologies, and a resistance
to an ideology that development is a "race,” and
that participation cannot be delayed. Ne claims
that protecting national sovereignty and cultural
integrity does not mean supporting traditionalism.

W-fls

84

WW (elem-1- added) -
Iamelink has been probably one of the more radical

scholars. Advocating a dissociation strategy for LDCs, he
calls for indigenous and self-reliant technological
capability completely free from transnational dominance. he
believes that the concentration of knowledge, exclusive
right over technical innovations and restrictive business
practices are intrinsically related to western interests
since they consolidate transnational activity. Formulating
the principles of a national information policy together
with the dissociation strategy he suggests that the LDCs,
should be suspicious of assistance and cooperation from the
TNCs. If the LDCs are to survive in the “information age”,
its program for survival will need to be based on the
building of a strong, autonomous information technology
infrastructure. And only when such an infrastructure has
been installed can meaningful technological negotiations
with the TNCs take place (1984). (This was the initial
strategy Brazil has tried pursuing).

This leads Bamelink to stress the importance of not
having any North-South discussion since any form of
cooperation would not meet the interest of the LDCs.
Increasing regional and interregional joint ventures could
probably improve local capability and enhance collective
bargaining power. This present state of weak bargaining
power has been the primary reason why LDCs have not managed

IS
te'improve their lot vis a vis the TNCs.

On studying the technology transfer issue, lamelink
analysed the sources, forms and the financial terms of the
transfer. The most common form of transfer was through
direct foreign investment wherein TNCs established
subsidiaries in LDCs for production. Turnkey projects and
training of technical staff were other forms of technology
transfer, but in all these activities Bamelink discovered
some form of LDC dependence or transnational exploitation.
The capital intensiveness of these technologies also result
in serious balance of payment problems for the LDCs (1983,
1995) .

Nattelart and Schmucler (1985) have refined the
somewhat simplistic view of the rigid dependency theory that
new technologies are imperialistic because of their ties to
multinationals. Locking more deeply at the problem they
make the point that information systems that are being
installed, include in their methods of operation and
implicit in their very architecture new forms of economic
control. however, the key to their opinion is that they do
not accept the Luddite attitude sometimes present in Latin
American critical communications research that imported high
technology is inherently exploitive and should be resisted.
They feel that since new communication technologies are
already an established presence in LDCs, they should be
dealt with to extract the most advantages. Prom a policy
standpoint, it would be a mistake to ignore an existing

3‘
process, when it is already tied in with the rest of the
economy. Noreover, their practical and realistic analysis
also accepts the fact that if a country does not enjoy a
communications system, with adequate capacity and up-to-date
technology, it cannot function as a qualified, capable
economy.

This conditional acceptance of technology, if it can be
designed and installed on Latin American terms, also
stresses the fact that progress is not a natural force but
results from a series of decisions made by local and
transnational policydmakers. They propose that science and
technology can be effective tools of human liberation, but
have not achieved that condition yet because of the
inequitable nature of society and the decisiondmaking
elite.

Servaes and Jayaweera on the other hand, approach the
problem of new communication technologies from a realistic
and pragmatic point of view. Their analyses is based on the
basic premise that no nation is completely autonosous and
there is no universal path to development. While Servaes
posits that these technologies bring with it western values
and ways of thinking, this can be rectified by balancing it
with apprOpriate indigenous technologies (1988). Jayaweera
on the other hand stresses the fact that the demand for
technology is also from within LDCs (1987). The Brazilian
case study of the computer industry clearly demonstrates
this.

11

The primary rationale both these scholars work with is
that transnationalization and the introduction of new
communications technologies are a given factor within the
world capitalist system. Disassociation or delinking from
the world economic system is a fantasy with little relevance
to the realities at hand for the LDCs. Considering the
inevitability of this process, what policies LDCs can work
with to harness this condition for their own benefits?

They suggest that these technologies should be utilised to
improve these sectors of the economy where productivity can
be increased. This also leads to the question of national
policies with regard to the incorporation of TNCs in the
domestic arena. Servaes suggests that TWCs be invited to
participate provided certain conditions are fulfilled:

One such condition is that transnational

technology transfer is directed and controlled by

the host countries so as to contribute to the

creation of domestic scientific and technological

capability. Another condition should be that

these transnational technology imports are not

considered the exclusive or the most important

channel of transfer. (p. 1987, p. 8)

An interesting point to note from.Jayaweera's point of
view is his opinion on capitalism and the use of new
communication technologies. he looks at the technological
problems of the LDCs as part of the workings of the
capitalist economy. Since the development of capitalism is
related to the use of technology, to minimize cost and
increase profit (through the International Division of

Labor), new communication technologies have made it easy to

8.
invest and manage capital in LDCs. This has resulted in
'the failure of any indigenous technological base, and
increasing dependence on the massive k a D base of the core“
(1986, p. 22).

On analyzing the diverse viewpoints posited by these
scholars, one can summarize that no nation is completely
autonomous. There also seems no universal path to
technological development. An interesting point to note, and
one that is often ignored by critical scholars, is that
demand for new technologies can often come from within a
country and is not necessarily imposed externally. The
Brazilian case is a good example (Adler, 1987).

Technology transfer from the TNCs has however allowed
some LDCs (particularly the more privileged groups within
them) to benefit from technological advances in the
developed world. The acquisition of modern technology has
enabled these countries to use it without often going
through the costly, time consuming and difficult process of
creating it. Nainstream scholars have often given this as a
primary reason for technology transfer. Yet there are also
no long term benefits derived from this dependency and
concentration of technology in the hands of a few. A small
number of rich countries are still the source of almost all
applied technology, and the TNCs are almost the sole
suppliers (Barton, 1986). Technology transfer may expand
the industrial output in LDCs, but it does not necessarily
increase their ability to produce new technologies

39
appropriate for their needs. This will not occur unless the
state undertakes explicit science and technology planning to
create indigenous capabilities.

The following three sections elaborate on the whole
gamut of science and technological planning, which
incorporates regulations, technology strategies, and
approprite channels of technology transfer. It is important
to note at this point that the economic perspectives
analyzed below enables one to comprehend the plethora of
practical issues development planners have to deal with in
order to develop technological capabilities. On the other
hand understanding the communication perspective helps in
laying the basic theoretical framework for the presence of
technology within LDCs. These are two very distinct ideas.

a. Development looaomdcs Perspective

Previously economists had focused attention largely on
the questions of choice of technique and the direct cost of
transferring foreign technology. lore recently their focus
has shifted to a consideration of three complex
conceptualizations of technological change in LDCs. This
section tries to elaborate on these, by dividing it up into
three parts. Part 1 deals with the role that the state plays
in shaping, stimulating, and inhibiting technical change.
And how it can intervene, through selective direct
regulations and science and technology planning, to build
indigenous technological capabilities as well as in the

30
acquisition of foreign technologies. Two ideal technology
strategies that LDCs can pursue to achieve their goals are
explored in part 2. This helps underscore the one Brazil
chose for its computer industry. Part 3 deals primarily with
the indirect costs LDCs hear when they acquire technology,
in the form of “packaged“ transfers. This helps in
understanding the variety of problems LDCs have to go
through to develop their infrastructure in order to
assimilate new technologies.
1I__Tha_atate_and_Teshnisal_Chansa

Pew would disagree that in a number of LDCs the state
plays a central role in shaping, stimulating and inhibiting
various forms of technical change and assimilation.

Devising appropriate policies to accelerate the absorption
of technology is an important consideration for policy
makers concerned with economic and industrial developaent.
This means that policy makers need to be continuously aware
of the ways in which the country's economic development
goals can be best achieved, through the infusion of
technologies.

All this calls for purposeful state intervention to
bring about science and technology development. This also
implies a rejection of the concept that market forces on
their own will lead to the development of technological
capabilities in LDCs. The organization of state intervention
requires consciousness and political commitment to the use
of a variety of government policy instruments. These include

31
those aimed explicitly at the development of science and
technology, and those affecting other development policies
that have relevant technological implications (example:
education, finance, industry, etc.) . The development of
technological capability also includes complex activities
including training, experience, local RSD development, and
most importantly the diffusion of technology throughout the
economy (Sagasti, 1979). The crux of the matter is that any
form of state planning should have a technological component
to it. With the resulting effect of an industrializing
strategy that imbeds a technological strategy. The Brazilian
science and technology planning since the 19608 has been
following this model.

Given the centrality of state involvement in technology
areas, some of the more commonly found activities are direct
regulations. Such regulations have the effect of increasing
the bargaining power of the LDCs. In today's market, which
is basically imperfect and oligcpolistic, regulations have
been used to reduce costs (thereby saving scarce foreign
exchange). They have also been used to protect local
technological efforts as the government policies for the
Brazilian computer industry clearly shows.

A common form of intervention has been control
exercised by the state over foreign technology agreements.
This relates primarily to the price of imported technology
and, removal or addition of restrictive clauses from the

contracts. A good example of this would be the Brazilian

32
agency's (SEI) decision to allow IBM and Burroughs to
manufacture medium-sized computers in the country. The
agreement called for a certain percentage of the subsidiary
manufactured computers to be exported, thus providing
significant foreign exchange, besides the advanced
technology.

Related to the governments' attempts to reduce the
price of imported technology have been attempts to influence
forms of such imports (Joint Ventures/Licensing/DFI). The
form of technology import may to some extent influence the
price, but it may also have a bearing on the net social
benefits. Many governments have intervened, where feasible,
to limit the control by foreign capital, particularly
wholly-owned direct foreign investment. In addition, the
state has also used science and technology policy to effect
changes in the structure of industry. Measures designed to
influence industrial structure while simultaneously building
up local technological capabilities has also included
placing restrictions on technology imports (Amsden & Kim,
1982: Dahlman & Sercovich, 1984). The market reserve of the
lower end of the Brazilian computer industry for local
manufacturers completely changed the structure of the

industry, which had previously been dominated by TNCs.

Wise
The explanation of this particular point gains

relevance when I elaborate on the Brazilian informatics

33
industry. The government had explicitly followed a ‘catch-
up' strategy in pursuit of the state-cf-the—art technologies
(as laid out in the Informatics policy). Consequently, their
failure to achieve this objective resulted in the more
realistic pursuit of ‘technclogy-following' strategy. It
may be illuminating to distinguish between the two different
”ideal type" technology strategies a LDC can pursue in their
effort to develop a high-technology industry. a) Attempting
to reach the industrial world technology frontier, that is
the ”catch-up" strategy: and b) using, but not attempting,
even in the long run, to generate frontier technology, that
is, the "frontier following" strategy. These are elaborated
on in the following paragraphs.

However, there are extremely difficult problems that
are confronted in following these strategies. A major
problem in the ”catch-up" strategy is the initial selection,
by private corporations and/or by the state, of the
technologies that are to be developed. Not all technologies
will merit attention and development, since in some cases
the long-run costs will outweigh the associated long-run
benefits. Besides the industry chosen will in all
likelihood be inefficient in the short run. This brings up
the question of the "right" paths or "sequencing" that has
to be followed, in the long run, to develop "deeper levels
of knowledge" for catching-up. (Should the industry
establish a components base first and then move on to

manufacturing complete products or should it start with

34
products?) However, the "sequencing" problem is complex
since in some instances "deeper knowledge may be required in
order to facilitate just routine production at “lower"
levels of knowledge (for example the manufacture of computer
components) (Rosenberg, 1982).

The "technology following" strategy differs from the
"catch-up” strategy in that it attempts to use foreign
knowledge efficiently without building up the capabilities
to challenge the frontier leaders, in the short or long run.
The selection of technologies takes place keeping this in
mind (example: 8-bit instead of 16-bit computers). The
strategy does not preclude the selection of high technology
industries that, it is hoped, will become internationally
efficient with the passage of time (through success in
exports, as hoped by the Brazilian government). Investments
in technology transfer and basic infrastructure (example--
high level of education) is undertaken to increase
productivity and efficiency. Foreign technology is
"imported" in various forms utilizing both market and non-
market processes, such as DFI, licensing agreements,
imitation, scanning of foreign trade journals, learning-by-
exporting, reverse engineering, etc. Technology following
often involves situations where the frontier is shifting
slowly, or where earlier-vintage (relatively speaking)
technology is used, that is not currently employed in the
more highly industrialized countries. The case becomes more

complicated when the frontier is moving rapidly and the

35
problem is one of "keep up" rather than "following." This
process of trying to keep up with rapidly shifting
technology obviously results in a tremendous strain on the
infrastructural capabilities of LDCs (Hoffman a Rush, 1980).

There is an emerging consensus in the literature that
there are important advantages in a country being a
”technology follower". Most importantly, the costs of
knowledge and risk are reduced (since the technology is
already established), thus facilitating increases in
productivity at relatively little cost. This, suggests that
only "relatively shallow" levels of knowledge are required
in adopting a technology-following position behind the world
technology leaders.

Besides this general belief, it does matter where a
country is, relative to the frontier of technology.
Countries not on the frontier can ”play catch-up" without
excessive R&D spending provided their rates of "physical
investment" are high. This seems to be the key for
technology followers. On the other hand, "catching-up" with
foreign technology is not a passive process. Perhaps the
most relevant and distinctive single factor determining the
success of technology transfer is the creation of indigenous
technological capacity, which seems to be a difficult task
to achieve without the requisite basic infrastructure
(Westphal, Rhee 8 Pursell, 1981).

It is relevant to comprehend here that in following

either strategy, an analytical problem will arise, when

36
evaluating the long-run social costs and benefits of the
technological self-reliance, given the alternative of
importing technology. Scholars have always justified self-
reliance on the rational premise that the long-run social
benefits outweigh the associated costs. But is this
criteria met? Lall (1984) suggests that in some areas
"self-reliant" technological change can result in
internationally inefficient costly products, even with the
passage of long periods of time. A good case in point are
the Brazilian micros, which are still 2.5 to 3 times more
expensive than those from international markets (Adler,
1987). Since the possibility of importing technology always
exists, a good policy question to ask is: whether it is more
efficient/productive, to utilize domestic resources to
generate indigenous technology (as India has tried), or to
transform such resources into exports to earn foreign
exchange, and consequently import the technology (the path
chosen by Mexico). It must also be kept in mind, that
importing technology, and generating indigenous technology,
are usually working in tandem and are not mutually exclusive
alternatives.

Touching upon the reality of the international
diffusion of technology, scholars such as Kaplinsky (1984a &
1984b) and Hoffman (1981) have argued that recent "leaps" in
the technology frontier as a result of microelectronics
applications have made both "catch-up" or ”technology

following” strategies more difficult, if not impossible to

37

pursue. The complexity of microelectronics-based systems,
it is argued, makes it hard for LDCs to follow suit.

As Kaplinsky puts it

The assertion that LDCs can continue to assimilate

DC technology at an unchanged rate, must be open

to question. In contrast I offer a view that

suggests that the gap between DC and LDC

technology is reopening, but at the same time DC

technology is becoming increasingly inappropriate
for LDCs. (1984b, p. 154).

l; £9§t_9f_1ran§fsf

A good deal of attention has been paid, to the direct
and indirect costs of foreign technology transfer. Direct
costs are either overt or hidden. Overt costs appear in the
contracted price (for example, the price of a turnkey
project) while hidden costs are implicit. Both overt and
hidden costs are direct in that they refer to the costs of
purchasing of the knowledge itself. Indirect costs are
considered below, the most important of which is the
influence of the imported technology on local technology-
creating capabilities. These costs seem to be of greater
relevance to the LDCs, because of the long lasting effects
such costs have, than the immediate ‘one-time' costs (Enos,
1982).

The literature on formal technology transfers, which
involve an active role for foreigners, has considered the
process from the point of view of both the seller and buyer.
The seller, attempting to get the highest price for the
technology being sold, will try as far as possible to
monopolize that knowledge. Simply by "packaging" the

38
technology, that is, by combining all the elements of the
transfer (such as the engineering, supply of machinery and
equipment, and training) into one "package," the seller may
be able to obtain a higher price than if elements were sold
separately. Sellers in a very strong position may only be
willing to transfer technology when they retain the maximum
control over its use, as in the case of wholly-owned
subsidiaries. A good example of such a scenario is the
mainframe computer market. With IBM Brazil controlling a
majority share, it has had no trouble in starting up
manufacture there. Had it been absent from the Brazilian
market, it is unlikely that technology would have been
transferred. Conversely, the buyer may be attempting to
reduce the seller's monopoly through negotiations with
several suppliers simultaneously and trying to "unpackage"
the technologies (Enos, 1982: Vernon, 1990).

Account must also be taken of the quality of the
technology transferred. In the case of direct foreign
investment (DFI), Lall (1984) has shown that sophisticated
processes and products may be transferred without the
ability to design or substantially change the technology.
These capabilities are frequently centralized in developed
countries. In such cases, know-how rather than know-why is
transferred. However, it would be relevant to stress here
that it is not obvious that in all circumstances know-why is
more desirable than know-how. To begin with know-why may be

more costly to purchase, and secondly, for some purposes it

39
may not be necessary (Lall, 1987, p.116).

Since the late 19703, increasing focus has been paid to
the importance of local technology creating capabilities.
The rationale for this view is that local "knowledge" plays
an important role in any transfer of foreign technology. An
important insight in this context came from Nelson (1978)
who stressed the "implicitness' and "tacitness" inherent in
any technology transfer. He pointed out that high
technology, as a result of its complexity, cannot be
transferred in its entirety. The result is that the LDCs
always receive less complete information than possessed by
the sellers, despite the transfer of blue prints,
instructions, etc. This obviously means that any technology
transfer requires a certain degree of technological
capabilities on the part of the LDCs. It further suggests
that such capabilities also facilitates the appropriate
choice and working of any technology. An increasing
understanding of knowing how something works, and then
progressing on to knowing why it works in the way it does,
would definitely enhance the assimilation. The implication
is that in moving from the former to the later there is a
qualitative increase in the "depth of knowledge." Adapting
foreign technologies to the local environment and gradually
building up the stock of proprietary knowledge and know-how
becomes relevant. It is here where domestic technological
capabilities actually appear and develop. Such capabilities

have strong economic repercussions for the LDCs, who have to

40
pay the price through the development costs in
infrastructure, scientific training, human resources
development, education, etc. (Katz, 1984).

However these costs are dependent to a large extent on
the channels via which technology is transferred. Channels
which allow for technology assimilation and complete
transfer would obviously result in a higher indirect cost.
Similarly those which do not diffuse technology within LDCs
have the effect of lowering costs (Vernon, 1990: Stewart,
1990). The next section elaborates on the various modes and
levels of technology transfer, and the often implicit

repercussions on development costs for LDCs.

3. CELINBLB OF TECHNOLOGY TRANSFER

In the last two decades, there has been a rapid increase
in technology transfer agreements, largely through the
activities of TNCs. Industrial technology has increasingly
become a marketable commodity, both within the operations of
TNCs and through various contractual arrangements for the
provision of technology and services to unaffiliated
companies. The commercialization of technology, mostly in
the form of fees and royalties for technical know-how and
services, has increased considerably. This rapid increase
has been influenced by several major factors, including the
fast pace of technological innovation: the growing use of
technology licensing: and other contractual agreements and:

most importantly, the higher importance accorded to

41
technological development in the national policies of both
LDCs and industrialized nations (Marton, 1986).

Although most technology transactions have taken place
between corporations in industrialized countries, payments
for technology by LDCs have also increased significantly in
recent years. By the early 1980s, annual payments for
technology and services from these countries rose to around
$2 billion, from $230 million in 1965. By the early 19908
such payments, could exceed $6 billion (Singh, 1981). Over
a period of these two decades, the patterns and conditions
under which technology transactions have taken place has
also undergone major changes. Technology transfer has
increasingly become subject to well-defined regulatory norms
and measures, pursued by governments of LDCs. Consequently,
technology transfer to LDCs has taken place under very
different conditions, than that prevailing in industrialized
countries (where the terms of transfer are normally
determined by the private parties themselves). These
differences stem not only from the very different techno-
economic characteristics of the LDCs but, more particularly,
from the role governments have played over these
transactions (Marton, 1986).

Obviously, different governments have adopted different
approaches towards foreign technology, which to a large
extent has also been influenced by the development model
adopted. Import Substitution Industrialization (ISI)

normally resulted in more restrictions, whereas export

42
promotion led to a more open economy. Though it seems
difficult to generalize, an interesting point can be made on
the relationship between the mode and level of technology
transfer and the development path chosen by a LDC. 181 is
normally more restrictive of foreign technology entering the
country depending upon protectionist policies, whereas
export specialization has tended to rely substantially on
international technology, through DFI or licensing. Since
the objective in the latter case is to manufacture products
of international quality (e.g. Korean PCs) at competitive
costs, DFI strategies enable this (Amsden & Kim, 1982).

In order to place technology transfer developments in
perspective, the pattern of transfer during the 1960s and
19708 is briefly reviewed. Although the major factors and
elements which shaped technology transfer during these
decades continue to operate in the 19808, new trends have
emerged. Technology transfer arrangements have undoubtedly
become among the most important relationships between
enterprises in industrialized countries and LDCs.
W

The choice of the form of TT involves a varying degree
of packaging of it. At one end of the range is
transnational investment, in which ownership and control
remains with the technology seller. This was the dominant
mode until the 19608. Over a period of time, new forms
emerged as LDCs pressed for more control: the development of

international capital markets made it possible for finance

43
and technology to be separated: technologies evolved: and
specialist high technology firms emerged which were prepared
to sell and service technologies on an arms-length basis.
For example, the availability of relatively inexpensive
integrated circuits, along with the possibility of obtaining
technology under license from small new firms, helped Brazil
shift its technological dependence from the established
computer hardware market dominated by market giants (Adler,
1987).

Consequently, less packaged forms of TT developed,
including joint-ventures (shared ownership and control):
licensing (ownership and management responsibilities lie
with the host country, but the conditions laid down during
licensing often introduce constraints): franchising:
management contracts: marketing and technical service
contracts: turnkey contracts: and subcontracting. In
general, the more monopolistic the industry and the more
complex the technology, the more difficult it is to acquire
technology, except through direct foreign investment
(Stewart, 1990).

W

DFI is traditionally the most important channel of TT by
TNCs based in developed countries. The growth of corporate
transnationalization has been accompanied by technology
flows from parent companies to their foreign subsidiaries
and affiliates. What distinguishes the behavior of the
wholly owned subsidiary is the function awarded to it. The

44
subsidiary is an entity in a transnational network, which
draws on the financial, managerial, and technological
resources of the network to carry out its work--which is to
promote the interests of the network as a whole (Vernon,
1990).

For example, in the mid-19708, the Brazilian government
was planning to develop an indigenous minicomputer industry,
through market reserve policies. IBM do Brasil, a subsidiary
of IBM International's primary role was to lobby the
government against any such initiative. This strategy was
based on their desire to utilize the Brazilian subsidiary a8
a manufacturer and exporter of minicomputers into the world
market (Tigre, 1983).

Due to the close relationship between TT and foreign
investments, the flow and stock of foreign investments in
LDCs serve as an indicator of the magnitude and rate of
growth of foreign-technology inflow. DFI increased from
$1.8 billion in 1970 to $7.6 billion in 1980. More than half
of these investments have been from the United States. The
share of DFI in LDCs from Japan and the Federal Republic of
Germany, however, has tended to grow more rapidly in recent
years. These investments have been accompanied by an inflow
of complex and advanced technologies. Investment in high
technology sectors by TNCs based in the United States,
Britain and Japan constituted an important share of total
investments, ranging from 70 percent in Nigeria to 50

percent in Brazil, Ghana, Malaysia and Peru, 25-30 percent

45
in Argentina, Mexico and Singapore and 5-10 percent in the
Republic of Korea, India and Thailand (Marton, 1986).
W:

Although TT in conjunction with DFI has remained the
most important channel in most LDCs, the ownership structure
of foreign operations has been changing. As a result of
Third world regulatory measures on foreign investments,
joint ventures have become an important form of TNC
participation, which have established affiliates with
varying levels of local share holding and responsibilities.

At one extreme are cases in which the local partner,
although owning a substantial share of the equity, is no
more than a figurehead in its operations. Their main
function is to satisfy the mandatory regulations of the host
government. Frequently, however, a joint venture pools the
different skills and resources of the partners and divides
the responsibilities of management between them. Apart from
enjoying a favorable position with government authorities,
the local partner is likely to have a comparative advantage
in selling the product in the local market, while the
foreign partner is best suited for production and
development (Manser & Webley, 1979).

Was

Policy makers in LDCs usually look on the subsidiaries
of TNCs, a8 distinctly inferior to a joint venture that
links foreign enterprises with local firms, and they often

look on a joint venture as inferior to a licensing

46
agreement. The implicit assumption is that the TNCs will be
unable to optimally utilize technology if there is a
reduction in its share of ownership (Vernon, 1990, p.258).

Licensing activities between TNCs and affiliated
companies has been growing rapidly. This trend has been
influenced by industrial policies of governments (who share
the above mentioned fear) as well as by the increased
technological capabilities of the domestic industrial
sector. In several LDCs the public sector companies are
involved in a number of priority industries (such as Cobra
in Brazilian computers). These enterprises are often the
largest companies in the country and have major stakes in TT
agreements. They therefore develop an extensive network of
licensing agreements with TNC. Besides, the acquisition of
foreign technology by these public sector enterprises has
been facilitated by the government, which has often
allocated foreign exchange for this purpose, on a
preferential basis (Ramamurti, 1987).

Though these arms-length licensing agreements are a
result of nationalistic policies, they often entail higher
costs and greater risks to the LDCs. The interests of TNCs
in the actual effectiveness of the TT is relatively limited.
Probably the largest single factor determining the
efficiency of a licensing agreement, is the same as that of
a joint venture, namely the extent to which managers and
technicians in the licensee enterprise are prepared to

absorb the technology and adopt it to local conditions .

47
This obviously is greatly influenced by the national
environment in which they operate (UNCTC, 1983: Vernon,
1990).

There have been a number of developments in the various
channels of technology transfer over the last two decades.
Some have been popular with the LDCs because of the control
it gave the governments over the TNCs, whereas there are
others which have been particularly effective for the
former. However, new trends have emerged which directly

influence the LDC policies to becoming more flexible.

4. TREIDB IN TECHNOLOGY REGULATION

There have been two trends in the technology acquisition
process from the 19608 to date, where LDC policies have
moved from regulation to liberalization.

In the earlier phases, regulation aimed to reduce the
costs of foreign technology by establishing maximum payment
levels. Considerable attention was given to the terms and
conditions in technology agreements: by regulatory agencies
prescribing norms and guidelines (e.g. CAPRE/SEI for
Brazilian Informatics). Varying degrees of regulation has
existed in different LDCs, though government control has
been fairly extensive. In some countries, as in India,
foreign-technology inflow in non-essential sectors was
prohibited. While in other countries, like Brazil, transfer
of know-how without patent rights, was strongly discouraged

(Marton, 1986).

48

Several countries which adopted technology regulation
neasures in the 19708, significantly modified them in the
latter half of the 19708 and the early 19808. With growing
industrialization and liberalization, development plans had
increasingly involved diversification and production in
technologically advanced fields (such as computers). Here
the availability of technology is limited to a few TNCs. In
such cases, these TNCs have not always been willing to
transfer their technology under the rigorous regulations,
particularly if this is accompanied by controls over foreign
investments. In several instances, the inflow of advanced
technologies has been promoted by relaxations in regulation.
Thus in countries where joint ventures are generally the
only permissible means for capital participation by TNCs,
governments have permitted the entry of wholly foreign-owned
subsidiaries if the inflow of complex and sophisticated
technologies has been involved (example: IBM in Mexico).
Besides, the effective absorption of the more complex and
advanced technologies require a longer period, which often
exceeds the duration prescribed in guidelines (UNCTC, 1984).
This has been reason enough for LDC governments to becoming
flexible in their TT policies.

The trend towards liberalization in regulation has been
accompanied by differentiated policies for different kinds
of technology. The Brazilian computer industry amply
signifies this trend. With the introduction of the

Informatics policy, the government hoped to manufacture

49
computers for the lower-end of the market. This required
relatively less sophisticated technology. At the same time,
they left the more sophisticated and of the market to TNCs,
who were encouraged to bring in their technology to
manufacture the mainframes, super-minis, and medium-sized
computers. This relaxation in regulations, within the same
industry but at different technology levels, underscores the
pragmatism of the policy makers.

This new orientation in government outlook, which has
been referred to as the "new pragmatism", reflects the
realization that the adoption of strict measures may be
counter-productive for the acquisition of complex and
advanced technologies. Thus, in Brazil, which was amongst
the first to regulate the activities of TNCs, both as
investors and technology suppliers, policies have become
differentiated and more flexible.

BHIISII-

It is obvious from the above review that no nation is
completely autonomous or free from the international
economy. As mentioned by Jayaweera, transnationalization and
pressures of new technologies are a given factor in the
world economy of which the LDCs are a part (1987). Therefore
trying to dissociate or delink the national economy from it
is a fantasy with little relevance to the realities at hand.
"Critical" scholars, such as Hamelink and Schiller, who
suggest these options seem to have completely failed to

apply the internal dynamics within LDCs, who demand these

50
technologies, to their theoretical assumptions. A balance
has to struck between the position taken by the ”mainstream"
scholars, pushing for a total adoption and assimilation of
TNCs within LDC economies: and the one taken by the
"critical” scholars, who push for dissociation from the TNC
dominated world.

Considering the inevitability of new technologies and
TNCs, what are the science and technology policies that LDCs
can work with to improve their lot? There also seem to be
no universal paths to development. The analysis of ideal
technology strategies, possible state policies for technical
change, and the various channels of technology transfer
illuminate the variety of paths LDCs can choose from, to
establish technological capabilities. Obviously national
policies, which incorporate the TNC presence yet pursue the
simultaneous development of indigenous capabilities, is the
key. The Brazilian example of the computer industry helps
illustrate this point. They were willing to allow TNCs
within certain segments of the industry while developing
technological capabilities in others. Chapter 3 elaborates

this process.

CHAPTER III

CASE BTUDY: BRAZILIRI OOIPUTER INDUSTRY

The Brazilian government's nationalistic policy towards
science and technology development has become a focal point
for economic and political trade discussions. Brazil's
passage of the 1984 Informatics Law, which formalized the
government's policy towards high-technology infant
industries led other LDC policy-makers to examine the
replicability of the Brazilian strategy. This chapter tries
to analyze the nature and success of the Brazilian
Informatics policy.

This allows one to understand the variety of technology
strategies and channels utilized for industrial development.
And more importantly, highlights the flexible and pragmatic
nature of the government when it realized the difficulties
it faced in achieving their objectives. In addressing these
topics, I first present in Part 1 the technology policies of
the Brazilian government since the 19608. This allows the
computer policy to be viewed in light of other technology
policies and highlights the interdependency in policy
making. Part 2 turns to detailing the Informatics policy.
The analysis of the development of the Brazilian computer

industry in which the role of the government in providing

51

52
conditions necessary for its development is explored in Part
3. The successes and failures of the policy for foreign
technology assimilation and development of indigenous
capability along with the effects on users is evaluated in
Part 4.

LWLLMH

The state has played a very active role in science and
technology planning. At no time in its history did the
government feel the need to forsake its presence in this
area. As a matter of fact planners have always presupposed
that direct policies, effecting regulations and costs, are a
prerequisite to effective technological development (Evans,
1986). Even though the Informatics case specifically
highlights state presence, a historical analysis of
Brazilian technology policies only reestablishes the point
being made.

Technology policy in Brazil has gone through dramatic
changes in the last thirty years. From a minor aspect of
industrial policy, science and technology has moved to the
forefront of Brazilian policy debates, forming a central
component of economic development strategies. Brazil has
also given more explicit attention to the development of
technological capabilities than other Latin American
countries. The two primary objectives for this attention

has been their desire for steady economic development and

53
development of indigenous capabilities (Baranson, 1981,
pp.15-22).

Prior to 1262, industrialization involved extensive
reliance on DFI and foreign borrowing. There were a number
of important by-products of the early industrialization for
the later informatics/computer industry. The fairly rapid
rate of early industrialization led to the creation of
infrastructure and experience in manufacturing and
management (which also distinguishes Brazil from many other
LDCs). The availability of institutional and physical
infrastructure, and professional, managerial, and skilled
industrial workers would contribute to the growth of high-
technology manufacturing in the 19708 and 19808 (Wirth et
a1, 1987).

The first half of the 19608 was a major transitional
period in Brazilian economic and political history, which
indirectly affected the technology policy. Industrial
growth was slowing and social protests were on the increase.
It was during this period that policies began to address the
problems of technology transfer. Laws were formulated which
attempted to examine and regulate the terms of transfer,
foreign investments and the relations between Brazilian
affiliates and TNCs. Mechanisms to control abuses in TT
contracts, with the aim of making conditions favorable to
domestic firms, were put into place (Mattos, 1982).

In 1261, in response to the political and economic
crises, the military, with the backing of industrial and

54
urban elites, took control of the government. This period
marked an important transition in government approaches to
technological development. The military leadership pursued
an active and explicit science and technology policy in both
commercial and defense sectors. There was a high level of
state involvement in directing policy, providing credit,
ensuring qualified labor, and encouraging TT. This
increased intervention was reflected in the growth of
technology bases of both economic and defense activities
(Adler, 1987). Policies to assure "real" technology
transfer and to aid industry in acquisition of foreign
technology were implemented between 12§A_§ng_12§1. Laws
were passed which regulated remittances between TNCs and
affiliates: capitalization of research costs: and imports of
machinery and equipment. These policies were to provide
”incentives for foreign direct investors to encourage local
production” by requiring domestic reinvestment of profits
(UNCTC, 1984, p. 13).

The period from 12§§_thrgugn_1214 marked a further
increase of government interest in and support for science
and technology development. The first explicit policies for
technology and R 8 D were detailed in 1968. Numerous policy
statements were issued, all affirming the importance of
technological advancement to national autonomy and economic
power. Programs to initiate TT and promote domestic
innovative capabilities were started up. One indication of

the growing importance of science and technology to

55
national development is its increasing share of the national
budget. In 1970, the share was 0.84 percent, by 1982, it
had risen to 3.64 percent (Tigre,1983).

The Brazilian government relied on four approaches to
support technology policy: subsidized financing to local
firms: the development of physical and human infrastructure:
regulation of contracts for acquisition of foreign
technology: and the protection of the market for capital
goods. While the existing reliance on foreign technology
was maintained, the terms of TT were increasingly regulated,
to encourage "real” transfer. The National Institute of
Industrial Property (INPI) was created in 1970 to administer
TT, patents and trademarks. In 1971 modifications in
regulations of TT were carried out with the dual objectives
of speeding up the transfer process, and increasing the
bargaining power of the government vis a vis technology
suppliers (Frischtak, 1986).

The Second National Plan for 1215:1212, reaffirmed past
goals, but also called for improvements in technological
capabilities and increasing sophistication of Brazilian
industries. Support for high technology industries (such as
computers, aeronautics, petrochemicals) was explicitly
targeted. The year 1974 marked a transition to more
aggressive bargaining with TNCs and a greater reliance on
national firms. The key objective was to develop local
technological capability. Special attention was paid to

regulatory mechanisms which screened TNCs for their

56
technological inputs into Brazil. This transition was, in
part, fueled by the deceleration of economic growth, both
nationally and internationally. By 1974, Brazil had to face
economic repercussions of the first oil crises, a balance of
payment problem, and recession in the global economy. This
turned the government from state spending strategies to
protection and market reserve mechanisms. Thus enabling
further support for indigenous technological development
which helped reduce imports and increase national autonomy.
It increased support for national industries, and
implemented requirements for local equity control and
capabilities (Adler, 1987). The commercial successes of
this period--in the production of internationally
competitive weaponry, passenger aircraft, and single-engine
planes--appear to confirm the wisdom of this strategy.

The national plan for the period between 1229:25
detailed a broad set of policies and priorities rather than
specific programs and projects. These are the policies
formalized in the Informatics Law, discussed in the next
section. The policies highlight present goals as gradual
increases in industrial R 8 D: increases in the
technological sophistication in industrial processes: and
decreases in the protection of domestic production as

competitiveness is achieved (UNCTC, 1984).

II- W
The 1984 Informatics Law clearly delineates the elements

57
of the current computer policy. It basically formalizes the
policies articulated in the late 19708 and early 19808 by
establishing an overarching policy for the development of
high-technology industries and technologies. This law
reaffirms the government's commitment to the protection of
an infant computer industry by reserving the Brazilian
market for domestic producers of minicomputers,
microcomputers and peripherals. It specifies when
industries will be classified as internationally competitive
and when they qualify for protection. It is this law that
had sparked the 0.8. claims of an unfair trade practice and
has drawn additional international attention to the
successes and failures of Brazilian science and technology
policies (Felder 8 Hurrell, 1988).

The elements of the informatics policy can be identified
in four stages. In the first stage, the technology is
beyond the capabilities of Brazilian producers. Foreign
producers are encouraged to supply the Brazilian market with
products and technology. The second stage is reached when
Brazilian firms have the capability to produce
technologically sophisticated products, but do not possess
the ability to generate the technologies domestically. In
this case, the market is protected and these infant firms
are supported through subsidies and government assistance in
obtaining technologies from abroad. The technology strategy
at this stage requires domestic firms to follow the

international frontier.

58

When firms possess both domestic financial and
technological foundations for production, they have reached
the third stage. The government begins to introduce export
incentives and encourage increased competitiveness. The
market in this third stage is still protected from foreign
competition. In the fourth stage, when firms are
internationally competitive as well as have indigenous
capital and technology, market protection mechanisms are
removed (UNCTC, 1984 p. 64).

The Informatics Law additionally specifies that the
market reserve will be enforced for 8 years, until 1992. It
further defines that a Brazilian firm is one where
Brazilians have majority control over stock ownership:
managerial decision-making and technological development.

In 1985, policy developments included the adoption of
the first National Informatics and Automation Plan. It
called for an increased commitment to human resource
development. It additionally highlighted the importance of
strengthening R 8 D activities which would integrate
programs involving research centers, universities and
industry (Frischtak, 1986).

In one sense, informatics has been accorded a privileged
position in Brazilian policy. The sector has been
consistently supported politically and financially (despite
0.8. retaliatory trade policies). It has also provided a
model and rationale for similar protection and support to

other strategic sectors. In another sense, however, the

59
policy is not a radical departure from past industrial
practices of import substitution, which drew upon a strong
tradition of nationalism, where policies reflected
commitment to industrial growth and economic development
(Adler, 1987).

One can claim that the informatics policy in Brazil has
struggled with conflicting requirements since its emergence.
The objective of promoting domestic technological
capabilities has had to balance with accessing foreign
technologies, the latter being necessary to the long-term
technological and competitive needs of Brazil's industries.
Reliance on foreign technologies can also undermine attempts
to foster local capabilities and innovation (Evans, 1986).
An analysis of the development of the computer industry and
the related technology policy illustrates this dilemma, and
how the government tried dealing it.

This analysis will attempt to reconfirm the propositions
laid out in the introductory chapter, namely:

1) Pragmatic Science and Technology policies, which
regulates a high-technology industry, can lead to the
development of indigenous technological capabilities. The
Informatics policy is a suitable example.

2) Domestic technology policies and technology acquisition
through TNCs is a complementary process, as highlighted by
the indigenous manufacturing of micros: joint

ventures/licensing of minis: and DFI in mainframes.

60
In. W191

The development of the computer industry in Brazil has
been clearly shaped by government policy. The policy
approach has been to regulate the computer industry by
reserving the lower-end of the market, the micro and mini-
computer market, for local producers while permitting TNCs
to supply the more sophisticated end of the market
(mainframes). The modalities of acquiring technology has
been differentiated, depending upon the complexity of the
technologies. For relatively "simple" systems, the
Brazilians preferred licensing and joint ventures, and
allowed TNCs to invest in technologies which were more
“difficult" to manufacture. This is elaborated upon later in
the chapter. These policies have promoted claims of unfair
trade practice by international governments and firms. But
many have argued that Brazil's approach to the industry is
consistent with both Brazil's and other nations treatment of
infant industries (Botelho, 1987). This section describes
the computer industry, the development of the related
informatics policy and the major influences which affected
the two. It also highlights the consequences of this
policy on industrial development, indigenous technological
capabilities and technology transfer from TNCs. This leads
to an attempt in answering the question: To what extent has
the state succeeded in fostering an indigenous high

technology industry?
Th8_£9meufer_1ndu§frx

61

The computer industry in Brazil got off to a relatively
late start. It was non-existent to the end of the 19608 and
when it did emerge in the early 19708, the expansion was
based on imports accompanied by neither government
regulations nor domestic production facilities. The only
computer manufacturing to take place in Brazil was some
final assembly-type operations by an IBM affiliate. There
were also minimal efforts to develop human resources or R 6
D activities (Tigre, 1983).

The computer market grew quickly. Excluding micro—
computers, the number of installed computers increased by
673 percent between 1973 and 1982. The primary consumer of
electronic data processing equipment was, not surprisingly,
the government, which accounted for 45.6 percent of the
market in 1980. Micro-computers were introduced into the
Brazilian market in 1973 and accounted for a growing share
of the market. Even though the micro- and mini-computer
markets was growing rapidly, it accounted for a small
proportion of the total market share (Tigre, 1983).

There are five major minicomputer manufacturers--Cobra,
Edisa, Labo, SID, and Sisco--and three major micro-computer
manufacturers--Cobra, sisco, and Polymax. Hewlett-Packard
is the only foreign firm with local manufacturing of
microcomputers and this is permitted only because HP
specializes in scientific micro-computers. The 1980 mini-
computer market was divided as follows: Cobra, the national

company controlled 40 percent of the market: Labo 20

62
percent: SID 15 percent: and Sisco and Medidata each
controlled 2.2 percent. In 1980, the distribution of shares
of micro-computer production was: Cobra with 63.3 percent:
Sisco with 2 percent: and Polymax controlling 34.7 percent
(Frischtak, 1986).

As compared to the characteristics of the market in the
early 19708, the Brazilian policies have had some success.
In 1984, sales of computers and peripherals manufactured in
Brazil totalled US $1,700 million. Of that total, national
production accounted for $881 million. This was based on
the sale of 1,082 mini-computers: 61,680 micro-computers:
11,218 business micro-computers: 25,857 serial printers:
1,114 parallel printers: 10,267 video terminals: 35,273
financial terminals: 1,824 cartridge disk drives: 2,348
Winchester disk drives: 20,965 floppy disk units: 439
magnetic tape units: and 20,021 modems.

Brazil now produces all the computers for the low-end of
the market. Local content increased for many products: the
import content of total sales by national producers
decreased from 28 percent to 7.5 percent between 1979 and
1982. These accomplishments are completely a result of the
market reserve established under Brazilian informatics
policy (UNCTC, 1984, p.99: Frischtak, 1986, p. 41).

The industry, moreover, seems dynamic. In 1977, only
nine informatics companies operated in Brazil and six were
foreign. By 1985, there were 274 registered companies in

the sector and 247 were national. There are now a number of

63
software and consulting companies in Brazil which account
for both additional economic growth and employment. Efforts
to strengthen R 8 D infrastructure have had some success.
By 1978, there were 208 scientists engaged in R a D per
million population while in 1974 the corresponding number
was 75 per million. There are now "an estimated twenty-five
to thirty thousand scientific and industrial researchers and
over one thousand graduate programs“ (Frischtak, 1986 p.
34).

Despite the successful technological and marketing
advances of the Brazilian firms, domestic firms still
control a small share of the overall computer market and
rely heavily upon foreign technology.
W

A description of the Brazilian computer industry does
not, however, explain why the government intervened when and
in the manner it did. This section specifically addresses
why the intervention was so nationalistic and why it
targeted the lower end of the computer market. A number of
factors contributed to the growth of support for
nationalistic approaches to industrial development. The
underlying reason for the antidependency approach was the
growing recognition of the political and economic costs of
technological dependency. Different groups had different
motivations and ultimately the confluence of national
security, industrial, technological, commercial, economic

and political pressures led towards policies for

64
technological autonomy.

A significant factor contributing to the nationalist
sentiment was the n11itg:y_1gaggr§hip;§ increasing sense of
vulnerability to economic and technological dependence. It
had been buffeted by the external shocks of the oil crises
and global economic crises of the early and mid-19708. The
military responded with attempts to nationalize the arms
industry in order to reap the economic benefits of
increasing arms sales. Restrictions on imports continued as
did subsidies to domestic producers. Many were interested
in the possibilities of a nuclear program, whereas the Navy
was particularly interested in a domestic computer industry
to incorporate systems in their vessels (Adler, 1987).

There was also a growing sentiment among 1ndng§ria11§§§
an§_;hg_m1gg1g;g1g§§ that the country should have modern

industries. In part this was a continuation of earlier
nationalist sentiment--the image of a modern nation with
modern industries. It was also a pragmatic reaction to the
changing technological basis of production. There was a
widespread recognition of the problems of dependence on
foreign technologies and equipment (such as obtaining
outdated equipment and processes, lack of spare parts or
customized equipment). Furthermore, the importance of
global communications to economic activities was growing.
Industrialists thus expressed interest in increasing
indigenous technological capabilities (House, 1987).

There was support within the state for increased

65
technological autonomy. BnIgangrg;§_an§_§§g§g_gggngig§
realized that technological dependency reduced the autonomy
of the state to make its own policies. Government agencies
also requires specialized customdmade data processing
systems, which was not easily available from international
suppliers.

Additionally, the size of the Brazilian market had been
a source of leverage for the government vis a vis the TNCs.
Experience in other industrial areas, such as in arms and
petroleum, had demonstrated that local technological
capabilities strengthened bargaining power.

Another concern was expressed by nniyg:§1;y_§;gg§n;§_ang
2191:5822: in the sciences. They were frustrated by the
lack of professional options. Many had graduate training in
Brazil or abroad and wanted the opportunity to implement
their prototypes and designs but could not interest the TNCs
in local R 8 D. They too desired the commercial
opportunities as characterized by the Silicon valley. This
group (labelled as "anti-dependency technocrats") was
particularly instrumental in the formulation of Brazil's
technological policy, directing and participating in Capre
and in many other agencies in the science technology
institutional network (Adler, 1987: Evans, 1986).

ua319n§1_gggngnig_ggnggxn§ provided an additional
rationale for controlling imports and encouraging local
production. First, payments for technology to TNCs, by

their subsidiaries, was a major means of avoiding

66
regulations on levels of profit remittances. For example,
these payments were characteristically much higher than the
cost of technology were it purchased by an unaffiliated
buyer. TNCs could thus overcharge for the technology used
by their affiliates.

Second, licensing, whether or not by affiliates, is a
relatively expensive way of obtaining technical know-how.
Despite the many regulations the Brazilian government had
been imposing on technology transfer, the acquisition of
technological capabilities is a lasting way of avoiding the
problem (Erber, 1981).

Summarizing the concerns of the various groups one
discovers that three predominant forces had influenced the
state to form the Informatics policy. The first was the
tradition of nationalism, protectionism, and state
intervention which had characterized Brazilian economic and
industrial development plans. Second, there was a growing
recognition of the political and economic costs of
technological dependency. A third force was the changes in
the international economy, and the prominent role TNCs plays
in it.

W

The development of government policy towards the
computer industry clearly demonstrates how the above
described characteristics of the state, technological
change, and the international economy interacted to push

Brazil's policy into its present protectionist form. This

67
section explores why the government intervened when and in
the manner it did.
Ear1y_§gmputgr_zg1igy. In 1972, when computers were just
entering the Brazilian market, the government created Capre
to coordinate the purchase and use of computers in
government agencies. As the use and purchase of electronic
data-processing equipment increased in the early 19708,
Capre grew in power and authority. It was the only official
agency with responsibility for the review of computer
purchases. Capre's explicit objectives were to achieve
economies of scale, reduce imports, promote the development
of a national industry, prolong the life of equipment, and
help R 8 D centers obtain technology. These goals were
consistent with the general industrial policy and not
specifically nationalistic for Brazil's economy. (Ramamurti,
1987). However, it is important to note here that the
government was strategically targeting an industry and had
specific plans for its development.

Early interest in the development of computer
technologies was expressed by the Navy and the Ministry of
Planning. In 1971 a contract was signed between the Navy
and BNDE establishing Funtec 111, a project to encourage the
development of minicomputers. The model pursued, although
not explicitly, was to foster a national champion in the
industry, like those that characterized the defence and
aeronautics industries.

In 1972, a holding company was formed, the Electronica

68
Digital Brasileira (EDB), through which the banks and
government could establish computer companies. The EDB
sought to foster a tzing (tripod) arrangement, a
partnership of the state, private industry, and TNCs. A
partnership with Ferranti Limited, a British firm already
supplying computer-based ship and submarine systems to the
Navy, was favored by the military. The BNDE preferred a
partnership with Fujitsu.

The EDB decided to create two computer companies, one
with military orientations and the other with civilian ones.
The civilian version did not materialize and consequently
the 1:1:92 venture with Ferranti, called Cobra, was expanded
to cover both military and civilian markets. EDB,
meanwhile, was renamed Digibras and given responsibility for
the national computer project.

Ferranti was one of the few specialized high technology
firms which had emerged internationally, and was prepared to
sell and service technologies on an arms-length basis. It
was also the first computer company to be willing to license
its minicomputer technology to the Brazilians for
manufacture. It was allowed very limited participation,
being relied upon to provide the technology for only the
first computer (Adler, 1987). As described by Adler, "the
[zr1ng] arrangement reflected a pragmatic position. Those
involved thought that domestic technology would progress
most rapidly if Cobra used foreign technology to develop the
national computer, but only if the foreign firms committed

69
to full technology transfer, with technology then to be
absorbed by Cobra." He further contends that Ferranti was
only chosen because it was receptive to Cobra's technology
transfer requirements and not because of its product quality
(1987, p. 246).

Digibras indeed found that they could not easily obtain
access to foreign technologies on Brazilian terms.
Negotiations with Data General, Fujitsu, and Nixdorf for the
development of a business computer stalemated on the
Brazilian insistence that patents, blueprints, and other
know-how be transferred to Cobra at the end of the license
period. Digibras eventually contracted with a small
American company, Sycor Inc., which was willing to accept
the technology transfer terms. The license resulted in the
production of two series of business and accounting
minicomputers. Inspite of the pressure by Data General on
the American government, for an embargo on this technology
transfer, Sycor managed to sell its know-how (Mattelart 8
Schmucler, 1985).

In both cases, that of Sycor and Ferranti, the chosen
foreign partner was selected because of its willingness to
accede to Brazilian technology transfer requirements and not
because of the sophistication or suitability of the
technology. This reiterates the point made in chapter 2 on
technology strategies. The Brazilians were willing to follow
the frontier of cutting edge technology rather than play
‘catch-up'. Their objective at this stage was to assimilate

70
new technologies rather than be on the threshold of the
latest.

With the deficit problems of 1974 and 1975, Capre was
given additional power to control imports since government
purchases of electronic data processors accounted for a
major outflow of foreign exchange. Once they had the power
to review and control imports, the individuals in Capre
began to formulate a general Brazilian policy for the
informatics sector (Tigre,1983).

The next critical moment for Brazil's policy development
occurred in 1976 when IBM announced plans to manufacture a
mini-computer--the system 32--in Brazil. Rather than accept
IBM's proposal, Capre, and then the National Economic
Development Council (CDE), countered with a new policy
regarding computer imports and production. The policy
detailed by Capre in July 1976 subdivided the computer
market--large computer systems were to be imported, while
national production of smaller computers was to be
encouraged (although without market reserve or state
intervention). CDE also outlined the criteria for fiscal
incentives: degree of nationalization, export potential,
extent of technology transfer, viability of enterprises in
market, and domestic capital majority.

Although formulated by the strongly nationalistic anti-
dependency members of Capre, the policy also reflected the
concerns of the Navy. The Navy, and thus others in the

military leadership, had continued their involvement in

71
Cobra and had pinpointed the computer industry as an area
for local technological and industrial development due to
the need for computerized guidance systems (westman, 1985).

Capre was made up of engineers and computer scientists
who recognized that much of the computer industry was beyond
the reach of Brazilian technological capabilities. The
lower end of the computer market (then the mini-computers)
was consequently targeted for governmental support and
encouragement. Due to the growing recognition within the
government of the costs of technological dependence: the
needs of the Navy: and the existing low technological
capacity in microelectronics in Brazil, the government
supported the Capre computer policy.

When Capre announced an open-bid for minicomputer
production in 1977, fourteen proposals were submitted, seven
MNCs submitted independent proposals, one was a joint
venture, and six were independent proposals by Brazilian
consortiums. Apparently the MNCs were relying on the
greater sophistication of their products in the bid
competition and did not propose joint ventures entailing
real technology transfer. Capre responded by approving four
Brazilian proposals-~Cobra and three domestic consortium8--
and rejecting the independent TNC bids and the joint
ventures. Capre later approved the entry of a fifth domestic
company (Westman, 1985, p. 35).

This was a clear indication of the governments' desire

to create a computer industry through explicit science and

72
technology planning. Though regulations were yet to be
formalized, bureaucrats felt the need to enter the less
sophisticated end of the computer market to facilitate
technology development. This was a reconfirmation of the
technology strategy the government wanted to pursue. A
frontier following strategy seemed more appropriate at this
stage since the costs of knowledge and risks would be
relatively reduced, and the potential for technology
assimilation high.

W- The changes in
policies, as described below for the medium-sized and super-
mini computers, very clearly demonstrates the pragmatism of
the government. It also supports the proposition laid out in
the first chapter that domestic technology development
policies, and TT through TNCs are not necessarily
contradictory, but can complement each other. Having
realized that it was not possible to indigenously develop or
manufacture the more sophisticated technologies, the
Brazilians went about allowing the TNCs to produce the
medium sized computers. A working relationship with TNCs
was more likely to help in the acquisition of the more
advanced technologies. Inspite of the change in plans, the
government still regulated production in order that the
domestic economy could benefit from it.

In the years that followed, the Brazilian bargained with
the TNCs about what could and had to be produced in Brazil,

what would be an acceptable degree of nationalization, and

73
what would be the export/domestic market sales balance. The
consistency of the policy, the power of the government to
implement the policy, and the growing domestic production
capabilities all served to strengthen the Brazilian
government's bargaining power vis-advis the TNCs.
Furthermore, the initial rejection by the TNCs of the
government's priorities led to a super-nationalistic policy,
in which the level of confrontation between the parties was
heightened and the conditions for production much debated
(Evans 8 Tigre, 1989).

This policy was further adjusted under the Special
Secretariat for Informatics (SEI), the agency which replaced
Capre in 1979. In 1980 IBM was given permission to produce
a medium-sized computer, the 4331 M62. This production was
strictly regulated: the minimum memory capacity was held to
two megabytes, the minimum national content raised to 85
percent, and the number sold on the Brazilian market limited
to two for every three units exported with a maximum of 242
units sold domestically by 1983. IBM was also allowed to
manufacture magnetic disks, for export only, which generated
income for both the Brazilians and IBM.

Burroughs was similarly allowed to manufacture computers
in Brazil only if subject to export/domestic market
requirements. The domestic manufacture and sale of the
B6900, a large computer, was approved at 1,950 units with
the requirement that an additional 6,020 units be

manufactured in Brazil and exported (Adler, 1987).

74

These agreements highlight the change in the
relationship between Brazil and the TNCs. The size and
potential value of the Brazilian market, coupled with the
possibility of exclusion from the market entirely, forced
the TNCs into accepting conditions that five years
previously they had rejected (the 1977 mini computer bids).
Brazil, faced with a worsening balance-of-trade, was also
forced to grant concessions, foregoing some long-term self-
sufficiency for the income of TNC subsidiaries' exports. The
state had also realized its incapability to manufacture the
more advanced computers. Taking this into consideration
makes it clearer why a highly nationalistic policy turned
pragmatic.

The increased strength of Brazil's bargaining position
did not obviate the difficulties when obtaining relatively
more sophisticated technology. In 1984, for example, the
market reserve policy was expanded to allow Brazilian
companies to produce "super-minis." Eight Brazilian firms
submitted proposals. Three proposals planned to develop
these computers with indigenous technology, while another
five companies proposed to rely on imported technology. SEI
approved the three-proposals-submitted by Labo, SID and
Cobra-relying on domestically developed technologies and
hoped that some of the other five firms would merge. This
did not occur. SEI had intended these indigenous
manufacturers to be on the frontier of super-mini

technology. The local manufacturers on the other hand had

75
also been confident on being able to catch up with
international technology. This too did not occur resulting
in the slower development of the superdmini technology. In
June 1984, SEI decided that licensing foreign technology was
the only way to resolve the ”technology gap" and approved
all supermini projects relying on foreign technology. In
order to remain competitive, Cobra, Labo and SID were forced
to abandon their indigenous efforts and purchase foreign
technologies (Evans 8 Tigre, 1989). Thus, indigenous
technological development in this case seemed too costly:
competitiveness ranked above self-sufficiency. In
retrospect, one can claim that SEI took a tough yet
pragmatic position: tough because it ruled out joint
ventures, and pragmatic because it understood domestic
industry's need for foreign technology to allow it to stay
abreast of developments abroad. This clearly underscores the
complementarity of TT to the development of indigenous
capabilities.

Another area of the policy's evolution is the place of
the computer secretariat in the government's structure.
Initially, when Capre was responsible for the policy, it was
a separate, insulated agency. In 1979, SEI replaced Capre,
with direct links to the National Security Council and the
President: this development reflected the growing importance
of the sector to national security interests. In 1984, with
the opening up of the government and the first moves towards

re-establishing civilian rule, the government passed the

76
National Informatics Law (reaffirming past regulatory
policies) and created CONIN, which is on “equal footing"
with the National Security Council instead of under its
control. Despite these changes in institutional structure,
the informatics policy has remained largely consistent
(Evans, 1986).

In summary, the lack of initial success in bargaining
with the TNCs led in part to the implementation of a market
reserve policy. The case of the computer industry largely
followed the government's broader approach to the
acquisition of technology and local managerial and equity
control, when the policy was originally formulated in the
early 19708. The refusal by the TNCs to include local firms
in the 1977 bids was a complete rejection of the priorities
expressed by the Brazilian government. The Brazilian
reaction was to leave them out of the future market,
especially when the local firms had alternative sources of
technology.

More recently, the Brazilian government has made some
concessions to international producers. The necessity of
keeping channels open for continued technology transfer
required that Brazil adopt some conventions of the industry
such as copyright laws and the number of years, after
licensing or purchase, that technology would remain
proprietary (House, 1987). The need to remain competitive,
as in the case of super-minicomputers, has also pushed the

government to modify the informatics policy (clearly

77
demonstrating their pragmatism). When continued access to
technology was threatened through the actions of
transnational firms, the Brazilian policy has consistently
been to modify regulations. Whenever the structure of the
technology and the industry made it feasible, the Brazilian
government has maintained its commitment to local industry
and technological development. The Brazilian policy, in
this light, appears both dynamic and pragmatic.

The Brazilian firms have followed a strategy of
technological development that combined several features:
formal licensing contracts, copying and adaption without
contracts (reverse engineering), local development, and
imports of components. Formal licensing linked to systematic
R 8 D effort was more frequent with the minicomputer
producers: on the other hand, microcomputer producers use a
copy-and-adapt approach, coupled with the import of
components (Erber, 1985).

There are also a number of characteristics of the
computer market that contributed to the feasibility of this
technology strategy. First, the Brazilian policy-makers
concentrated on the low-end of the computer market (minis)
because it required the least technological sophistication
to enter the market. The presence of small high technology
companies in the industry, such as Ferranti and Sycor, who
were willing to sell technologies played another important
role.

The subsequent development of micro-computers proved

78
fortuitous for the Brazilians. As a new product, micro-
computers had no established producers or importers
supplying the Brazilian market. The micro-computers
involved an even lower technological level to assemble.
Besides the global micro market was characterized by
relatively 'unpackaged' forms of production. In contrast to
the large computers, where the technology and components had
to be purchased together, micro-computer components could be
purchased independently (Evans, 1986).

Therefore, producers of microsystems have had to invest
less than makers of minisystems, because of their ability to
perform reverse engineering (in itself an indication of
technical capability), and to import the necessary
electronic components. As a matter of fact, Brazil still
relies on the import of standard components. Mereover, in
addition to soaring sales worldwide, microcomputers have
become an integral part of communication systems thus making
it a necessity for all productive activity. Ironically it is
the lowest end of the computer market which is now critical
to the development of the more sophisticated technological
systems.

Summary. Through an examination of the historical
evolution of Brazil's technology policy, it appears that the
computer case is similar to many of Brazil's other
industrial sectors. An ISI (Import Substitution
Industrialization) strategy, complete with import controls
and support for domestic industries, largely describes the

 

79
Brazilian government's approach to the computer industry.
However, the importance of the computer industry
specifically, and technological advancement generally,
.militated for a more nationalistic and protectionist
strategy. The fear of not gaining technological
sophistication, and the resulting difficulties in
establishing the infrastructure for future technological
advancement, provide two explanations for Brazil's
commitment to its computer strategy. Moreover, government
policy appears to be as much a consequence of the attitude
of the TNCs to fostering local Brazilian production as to
nationalist sentiment within Brazil: the two have been
mutually reinforcing. It is the confluence of numerous
factors, local support for technological autonomy and
domestic production, the general logic of the state,
strategic bargaining with the TNCs, and the structure of the
international computer market, which provides an explanation
for the emergence of the Brazilian policy. To the extent
that the policy has had some success, it is likely that it
will prove the model for other high-technology sectors.

At this point, it would be useful to evaluate the
success and failures of the policy, in technology
assimilation and the development of indigenous capabilities.
The following section also analyzes the effect of the policy
on the various user categories.

“.18W3112me
As stated earlier, Brazilian technology policy has been

80
most recently and most publicly formulated in the 1984
Informatics Law. This law reserves the Brazilian market for
domestic producers of minicomputers, microcomputers, and
peripherals. This law also establishes an overall policy
for the development of high-technology industries, with the
objective of fostering technological self-sufficiency.

Critics claim that Brazil's market-reserve policies have
resulted in higher costs, lower quality and decreased
diversity and, most importantly, the unavailability of
computer goods for the Brazilian consumers. Others claim
that limited technology has been transferred and
assimilated.

The data presented here suggests that the Informatics
policies have led to the creation of a strong computer
industry. Yet while sales and industry data point to some
commercial success because of the protectionist strategy, it
does little to clarify other areas of concern (Archer,
1986). Questions which become relevant now are: To what
extent has the policy succeeded in transferring
technological know-how and providing the foundations for
future indigenous development? To what extent has the
policy influenced the competitiveness of firms within the
sector? To what extent have the Brazilians achieved self-
sufficiency ? What are the costs to the users?

In addressing these questions, I examine the
accomplishments and barriers to the future growth of the
industry. The Brazilian policy and strategy has had some

81
success in transferring technological know-how and providing
the foundations for future indigenous development. It has
encouraged the development of human resources: R 8 D
facilities: domestically controlled industries: and ”real"
technology transfer. Yet, there still exist barriers to the
further development of the sector, such as abnormally high
product prices: low economies of scale: dependence on
imported components and user dissatisfaction. The extent of
these accomplishments and barriers are reviewed below.

Human_ngsgnzgg§LB_§_n. In the area of human resource
development, the Brazilian policy has been largely
successful. There are over 30,000 researchers and
scientists in the informatics sector. There are five
institutions providing doctoral degrees in computer science
for approximately 10 people each year. The total number of
researchers with Ph.D.s in computer science and engineering
is 108. There are 15 institutions offering Masters degrees.
Estimates of the total number of Brazilian researchers
engaged in R 8 D range from 750 to 1,100. Frischtak
estimates that approximately two-thirds, or 500 people, are
working on software development (1986, p. 64).

Proportional Brazilian investment in R 8 D compares
favorably with that of the U.S. In 1980 Brazilian firms
spent 8.7 percent of total sales on R 8 D. American
computer and peripheral companies spent an average of 6.1
percent of total sales that same year. R 8 D expenditures

per employee were $4730 for the Brazilian firms and $3265

82
for the U.S. firms. In absolute terms, however, the
limitations of Brazil's R 8 D are striking. Brazilian
public and private R 8 D spending is low in comparison to
the global leaders in the informatics industry. This may be
a further area of concern for the Brazilian industry.
Nationally, only 0.54 percent of Brazil's GNP is invested in
the R 8 D of new technologies. Between 1979 and 1982,
Brazilian national and private computer firms spent US $160
million on R 8 D. IBM, in comparison, spent more than US $2
billion, equivalent to 6% of IBM's global sales, on R 8 D in
1982 alone (Tigre, p. 96). The difference in the financial
resources available to the international industry leaders
and those available to the Brazilian firms highlights their
difficulty in reducing the existing technological gap.

These comparisons suggest that Brazil has considerable
barriers to cross to further technological development. It
does not mean, however, that the Brazilian industry cannot
continue to gain in market size and technological
capabilities. Instead, the differences in financial and
human capital resources suggest that Brazil's strategy must
not be to compete directly with the TNCs for either markets
or products, nor to strive for state-of-the-art
technological capabilities in many product areas (technology
following strategies), such as mainframes. Instead, these
differences suggests that Brazilian policy must focus on
enlarging the market for Brazilian firms (by promoting
exports as well as the growth of a domestic market),

83
encouraging cooperative R 8 D among Brazilian firms, and
ensuring continued financing of R 8 D and human resource
development.

This also has strong implications for other LDCs wanting
to pursue the cutting edge of technology by following the
catch up strategy. Since such LDCs have low R 8 D
expenditure, it behooves them to pursue the frontier
following strategies, which is more in line with these low
budgets.

1ndnstzia1_§zgwtn. The growth of production can be best
traced by analyzing data for the total revenue of the
industry, which rose from $830 million in 1979 to $2.3
billion in 1985 (Frischtak, p. 45). The share of the
national industry in production has risen rapidly relative
to that of the TNCs. In 1979, the shares were 23 percent
and 77 percent, respectively. By 1985, the share of the
national firms surpassed that of the TNCs, at 50.7 percent
versus 49.3 percent. Production of micro-computers exceeded
70,000 annually. Considering that in 1985, the production
of 16-bit microcomputers reached 7,000 units, it would
appear that approximately 90 percent of the micros produced
are of the older 8-bit type (Cline, 1987, pp. 39-43).

1ndgs;rial_§§rngturg. By 1986, there were 250
registered informatics firms in Brazil. In spite of such
large numbers there still exists a significant
concentration. Cobra, remains the largest firm, although

its share has fallen from 59 percent in 1979 to 17 percent

84
in 1983. The five largest firms accounted for 89 percent in
sales in 1979, and 46 percent in 1983 (Tigre, 1983, p. 63:
Cline, 1987, pp. 43-45). While such dynamism is probably
caused by the changing market shares, a fundamental question
remains (as discussed later), concerning the need for large
scale efficient production, and the scope for a large number
of firms to exist in a small market.

WWW- Despite these
achievements, the discussion presented in the previous
section suggests that Brazilian firms rely heavily on
foreign technologies. While the informatics policy clearly
outlines increasing self-sufficiency in technological
innovation, the pattern of R 8 D expenditure in Brazil,
changes in products and the increasing pace of international
technological change suggest otherwise. The Brazilian firms
will find it difficult to decrease the technological gap at
its present status unless they to depend on foreign
technology.

At this stage, it becomes relevant to elaborate on an
important point made in chapter 2. When the government laid
out the Informatics policy, it explicitly tried following a
‘catch-up' strategy in pursuit of leading edge technology.
They were not really successful in reaching the
international frontier of computer technology, either in
developing a state-of-the-art computer or an indigenous
component base at par with world standards. Consequently,

the policy had to shift to a ‘technology following'

85
strategy. The result has been that the industry now
manufactures more 8-bit than 16-bit micros: it has had to go
into licensing agreements for the super-minis: and the
medium-sized computers are completely manufactured by the
TNC subsidiaries.

The last few years have noted a technological shift
towards a closed architecture in micro-computers (those
using 16 and 32 bit micro-chips) rather than open an
architecture. Closed-architecture systems require
customized semi-conductors and proprietary operating
systems, and are more difficult to copy or reverse engineer.
There has also been a global shift to automation and
vertical disintegration in the electronics industry. Larger
production volumes and shortened product cycles now
characterize the industry (Mody, 1989).

This combination of shortened product cycles,
automation, and closed architecture, means that, as
Frischtak notes, 'efforts...to design, manufacture as well
as copy...newer generations of machines becomes increasingly
expensive (pp. 55)“. Brazil may have great difficulty in
achieving international competitiveness as well as the
technological state-of-the-art.

2zigggzzggngm1g3_gfi_2gg1g. Measured by price
performance, the Brazilian industry appears to be relatively
inefficient. Although domestically produced low-end
microcomputers (8-bit technology) have fallen in price to

international levels: for the 16-bit micros, prices are in

86
the range of 2.5 to 3 times the international level. The
high cost of micro-computers is related to two factors.
First, the attempt to reproduce technology, rather
reinventing the wheel, which is available internationally,
tends to increase costs. And more importantly, the scale of
production is very low. Technology is not the dominant
factor here, considering that the key microprocessor is
available on the international market, making the micro more
of a “commodity." In 1985, only 7,000 (16-bit) micros were
produced in Brazil. Frischtak suggests that efficient
production requires 50,000 to 100,000 machines per year for
an efficient plant size (p. 64). On comparing the markets
and economies of scale achieved by foreign and Brazilian
manufacturers, he found that the Brazilian producers of 16-
bit IBM clones manufacture on an average 200 units per month
(or 2400 per year). In contrast, Apple shipped 250-300,000
units of Mackintosh, Compaq over 130,00 units, and IBM over
two million PCs. Not being able to achieve these high
production volumes, and the resulting economies of scale,
means that it is very difficult for the Brazilian firms to
be price competitive (Cline, 1987).

Besides, the relatively small size of the Brazilian
market (by international standards) and the plethora of
producers involved in manufacturing will not allow for this
scale efficiency. This orientation of producers to the
domestic market also imposes high prices. Therefore the

inability of the Brazilian firms to increase production

87

seems a consequence both of unrestricted entry of domestic
firms and limited domestic demand (because of high prices).
This barrier could be eliminated, if exports are encouraged
or firms merge and consolidate.

anpgngntg. An additional difficulty derives from
Brazil's continued dependence on importation of components.
The extent of links between the component and final product
manufactures is increasingly becoming a critical determinant
of international competitiveness. Leading computer
manufacturers increasingly rely on in-house production of
components. Control over the design and manufacture of
components seems necessary to the development of the
customized systems, which now characterize the global
computer industry. As Tigre explains,

A high degree of dependence on foreign supply of

components can inhibit the future of the local

industry. If Brazil does not develop semiconductor

design and production capability the local computer

industry would have little control over supply and

costs or the major computer hardware component and

would be reduced to a passive role with respect to

technical change. This can constrain the long-term

competitiveness of local computer products (p. 118).

The technological requirements for semiconductor
manufacture are much higher and entails a greater depth of
knowledge than for the manufacture and assembly of other
computer products. Brazil has very limited capabilities in
high precision mechanical components (used in computer
peripheral equipment) and microelectronic components (for

integrated circuits). Tigre examined semiconductor

production in Brazil and found that there were 13

88
manufacturers of semiconductors in 1981, and only one,
Transit was nationally owned. The foreign firms merely
assembled imported chips. Transit had a more integrated
production system but halted its operations in 1981 due to
financial and technical problems (1983 pp. 116-119).

The Brazilian government thus faces a dilemma not unlike
that faced in 1976 with respect to minicomputer manufacture.
Even if the government were to invest heavily in the
foundation of R 8 D (thereby increasing their indirect
costs) and manufacturing facilities for semiconductor
production, the quality and reliability would be lower than
current imports. The use of domestically produced
semiconductors would harm domestic manufacturers of micro
and mini computers. On the positive side, the investment in
semiconductor production capabilities would benefit the
long-term autonomy of the Brazilian informatics industry.

Meanwhile, the existing policy has had some success in
decreasing imports of components (UNCTC, 1984, pp. 99-100).
There are strict government controls over the import of
components. Firms are allocated a fixed annual quota for
imports. This quota varies for the different product lines,
depending on feasible minimum sources of components. The
import content of national firms' production declined
dramatically from 1979 to 1981--from 29 percent to 8
percent. In contrast, the import content rose during the
same period for foreign affiliates, from 28 percent to 40

percent. Tigre found, however, that there was growing

89
interest among foreign manufacturers in the nationalization
of their products: this was due to both the government
controls and the bureaucratic problems associated with

regulations (p. 120).

nserzManufasturer_nieeatisfastien- Potentially one of
the most significant effects of the informatics policy is

its adverse impact on users (or prospective users) of
computers, relative to what would occur under "normal"
prices. Though the price barrier is more an issue with
micro-computers, when it comes to minis and super-minis
technological disadvantages play a greater role (Cline,
1987, pp. 60-65). Since the higher end computers are not at
an international technological level, corporate users have
displayed a high degree of dissatisfaction.

“A survey of 217 of the major Brazilian corporate
consumers by Arthur Anderson, in June 1988, indicated a high
level of dissatisfaction with the current situation: 85% of
those surveyed said the major reason for their
dissatisfaction with the industry was the high price they
had to pay for what was obsolete equipment in comparison
with that offered by foreign manufacturers (Straubhaar,
1990)".

On the other hand, manufacturers have been displaying a
growing tendency towards questioning the validity of a
”rigid“ regulatory policy. The desire to be competitive in
technology (and the resulting increase in market shares) has

led a number of local manufacturers to calling for greater

90
flexibility in acquiring technology via DFI and joint
ventures. Interestingly, with the change in administration
there has also been a more liberalized and open economy.
Perhaps most significantly, joint ventures are now being
allowed in microcomputers-an area which had previously been
totally protected. The most dramatic of these has been an
agreement between SID Informatica and IBM to manufacture the
PS/2 series micros in Brazil. This should result in
computers being produced at costs considerably lower than
the completely national machines, and obviously, at more
current technologies (Straubhaar, 1990).

Market reserve still remains a highly charged
nationalistic issue today, with strong overtones of
technological independence. In economic terms, however, the
direct and indirect costs of current policy appear to be
substantial. Although the Brazilian economy is large by
comparison with those of most LDCs, it may be too small to
sustain production at an efficient scale. The evidence does
suggest that costs are well above international levels.
Moreover, informatics is a sector driven by dynamic
technological change. It is unclear whether Brazil can keep
pace with technical change in the absence of cooperation
with TNCs. The proposition laid out in chapter 1 that
technology transfer (TT) via the TNCs, and domestic
technology development policies should be complementary
simply highlights this dilemma. There is an obvious need for

a greater working relationship with TNCs.

91

The central issue for the future is whether Brazil will
pursue an increasingly comprehensive market reserve policy
(for example, reaching up into the mainframe computer
market) at higher costs: or will policy makers begin to prod
the domestic industry towards international competitiveness
(through a gradual reopening of the market). The economic
liberalization being pushed by the new president, Collor,
seems to lean towards the latter. He has clearly stated that
the Informatics Law will be interpreted in the manner best
suited for more open markets and fewer regulations
(Straubhaar, 1990, p.16).

fiummury. The above review of the impact of the
Brazilian policy with respect to human resources, R 8 D,
industrial growth, industrial dynamism, technological
parity, indigenous technological and industrial
capabilities, price performance and effects on users
suggests only partial answers to the questions presented at
the opening of the section. The Brazilian policy has had
some success in establishing the foundations for further
growth in the industry. Yet the local firms also lack the
technological and industrial basis for semiconductor
manufacture, one of the key inputs in the sector and of
increasing strategic importance.

The Brazilian policy, based largely on traditional
approaches to industrial development, has succeeded in
achieving the explicit policy goals regarding human

resources and increased output. The requirements of the

92
international informatics sector, however, means that these
traditional strategies are inadequate for competing in the
world markets. The resources and mechanisms of traditional
development policies (market reserve) cannot begin to cope
with the pace of international technological change, the
direct and indirect costs of technological advances, and the
dynamics of implementing the technologies in a domestic
market.

Although Brazil's policy has succeeded in creating the
basis for an informatics industry, there are indications, as
summarized above, that Brazilian firms may have difficulties
in reaching stages three (technological self-sufficiency)
and four (international competitiveness) of the informatics
policy program. Alternatively, there is the promise held
out by Brazil's success in other technologically
sephisticated industries such as petroleum exploration,
aeronautics, and weaponry.

There is no doubt that the Brazilian policy objectives
of fostering a national industry have been met. It has
created locally controlled manufacturing capacity in a
technologically complex industry. It has also shifted its
reliance on imports from final products to components. But
it would be foolhardy to believe that these transformations
have led to overcoming dependence or autonomy. The role of
international technology is still pervasive throughout the
industry. The imported microprocessors dominate the micros:

licensing agreements influence the minis: and finally the

93

larger end of the market is completely dominated by foreign
technology, either in the form of imports or production
through subsidiaries. This leads one to believe that there
are significant barriers to further consolidation of the
Brazilian industry. Barriers arising from changes in the
global technological frontier as well from the structure of
both the international and Brazilian informatics industry.

melanin
Technology policy in Brazil has struggled with a number

of conflicting requirements since its emergence in the late
19608. It has had the explicit objective of promoting
domestic technological capabilities, and simultaneously
access foreign technologies. The acquisition of
technological capabilities was portrayed as critical to
continued economic growth, competitive functioning of
industries, and to military power. The two distinct, and
often contradictory, goals were linked through the promotion
of pragmatic and strategic nationalist programs, such as the
Informatics policy.

There exists a dilemma regarding the source of
technology. Access to foreign technologies is necessary for
the long-term technological and competitive needs of
Brazil's computer industry. At the same time reliance of
foreign technologies can undermine attempts to foster local
capabilities and innovation. Brazilian policy has tried to

strike a balance between these two constraints by addressing

94
the terms of technology transfer (through regulations) and
the introduction of market protection mechanisms. It has
tried to resolve the political contradictions of such a
policy through pragmatic bargaining with the TNCs and the
implementation of increasingly nationalistic policy. The
rhetoric of nationalism, the ideology of anti-dependency,
and the resulting regulation of technology transfer, foreign
direct investment, capital remittances and market production
levels (domestic versus export) allowed the Brazilian
government to create new areas for negotiations with the
suppliers of technology.

The costs of not gaining technological sophistication,
and the difficulties in establishing the infrastructure for
technological advancement provide two explanations for
Brazil's commitment to its computer strategy. It is the
confluence of numerous factors, local support for
technological autonomy and local production, the general
logic of the state, strategic bargaining with the TNCs, and
the structure of the international computer market, which
provides an explanation for the emergence of the Brazilian
policy.

The review of the impact of the Brazilian policy with
respect to human resources, R 8 D, industrial growth,
industrial dynamism, technological parity, indigenous
technological and industrial capabilities, price performance
and technological diffusion suggests that the Brazilian
policy has succeeded in achieving its explicit policy goals

95
regarding these issues. But, the requirements of the
international informatics sector, however, mean that these
goals are insufficient to ensure further technological
development in Brazil. The resources and mechanisms of
traditional development policies, while producing some
significant results, are inadequate to respond to the pace
and costs of international technological change.

On the other hand, one of the significant results that
the policy has achieved is strengthening Brazil's bargaining
position vis-a-vis the TNCs. Nevertheless, dependence
especially with respect to required imports of components
and technologies appears to have simultaneously increased.
This dependence on continued access to technological
advances has forced Brazil to make some concessions to
international firms.

One of the critical questions at this juncture is
whether Brazil will maintain its current balance between
autonomous development and dependence: will it move towards
technological self sufficiency: or will it become
increasingly reliant on foreign technologies and components.

The following chapter elaborates on the implications of
the Brazilian experience for other LDCs, specially with
regard to science and technology planning, regulations, TT,
and the role that TNCs play in it. The central theme is that
strategies and policies can over time not only result in the
development of domestic capabilities, but also in the
acquisition and assimilation of foreign technologies.

CEIPTER IV

INPLICRTIONB I'OR SCIENCE AND ammo! PLANNING IN LDC!

The affluence in communication technologies in the
industrialized West and poverty in the LDCs, is a condition
few planners find acceptable. What the LDCs need is the
means to be self-reliant, and the resources and information
to develop their society and economy on their own terms.
While appropriateness of communication technology for their
own particular needs is relevant, the means to develop their
own communication infrastructure, produce their own software
and manufacture their own basic equipment, is equally
important.

This presents them with a difficulty. The LDCs seek
independence and self-reliance. It is clear, however, that
no country in the contemporary world can be fully
independent from all others. Besides history has shown that
more frequently than not, efforts to industrialize has led
to increased technological dependence. The LDCs'
desire for complete independence is therefore not possible.
Obviously, most of them will not have the resources (in the
foreseeable future) to launch their own communication

satellite, manufacture their own computers, and produce all

96

97
the equipment necessary for modern information exchange.
The objective, therefore, cannot be total independence.
Rather, it is self-reliance to the extent possible, and a
lessening of the one-sided dependence which LDCs experience
in relation to industrialized states.

This thesis deals with these issues of adoption and
creation of new communication technologies within LDCs.
Obviously this phenomena touches upon the simultaneous
development of indigenous capabilities and technology
transfer. The Brazilian case study helped in understanding
the problems a country goes through in the acquisition of
foreign technology, and in developing a domestic high
technology industry capable of producing new technologies.
The complex interplay of forces between government, foreign
and national industries bring out the dynamics of technology
development.

It also investigates the propositions laid out in
chapter 1, that local technological capabilities in a high
technology industry can be developed by pragmatic and
scientific policies which encompass regulatory mechanisms.
Similarly, technology transfer via TNCs can play a
complementary role to domestic technology development
policies. A working relationship with TNCs can help in
acquiring the more sophisticated teChnologies. To this end,
this chapter seeks to analyze these problems. The issues
discussed here have practical implications for LDCs, and

have been set up in three sections.

98

Part 1 discusses the success and failures of the
Brazilian Informatics policies, and the implications it has,
if any, for other LDCs. The second part lays out the
strategic and pragmatic policies LDCs can pursue (example:
regulations and creation of indigenous technological
capabilities) to achieve technological development. And the
final part suggests that selective involvement and
interaction with TNCs can be a reasonable way of acquiring
technology, provided it is pursued in tandem with the
creation of domestic capabilities. The chapter also reflects
on the preferable TT channels and the technology strategies

useful for LDCs to pursue.

MW:

One of the central concerns of those examining Brazil's
policy is to determine to what extent the protectionist
strategy can provide a model for the development of high-
technology industries in other LDCs. However, this strategy
is of limited use as a blueprint for the development of
other LDCs. First, the market reserve policy is but one of
many factors contributing to the growth of the micro-and
mini-computer industry. An understanding of the other
factors--political, institutional, economic, and
ideological--suggests that many government
mechanisms/policies, as well as the characteristics of the
global computer market, contributed to the emergence of the

Brazilian industry.

99

Besides, Brazil had previous manufacturing and
engineering experience which contributed to the feasibility
of the sector's development. The size of the local market
also allowed Brazil greater bargaining power with TNCs in
obtaining technology. Since few others share these
advantages, Brazil's policy cannot act as a blueprint. The
policy experience may, however, suggest the utility of
particular measures, such as market reserve, preferred
channels of TT, etc. An assessment of just how applicable
different measures can be, for other industrializing
nations, is one question for further analysis.

Advocates of the computer policy have praised the
development of a national industry, specially a high-
technology industry. They focus on expected decreases in
technological dependence, improvements in the skill levels
of the workforce, and the improved bargaining position of
the country. They look to the policies, and the indirect
costs borne, which have promoted the industry-the
development of human resources, the promotion of R8D, the
regulation of technology transfer and capital, the channels
utilized, and the protection of the market-to assess the
adequacy of each measure, and the applicability to other
countries.

However, a few important points need to be made in terms
of where the industry stands today and what it has or has
not achieved, through regulations of the market.

At the outset, it is quite clear that the market reserve

100
policy has not created complete technological autonomy, as
was desired by the Brazilians (even at the lower end of the
market). But one has to look at it from the perspective that
domestic technological capabilities in a new high-technology
industry has been created. The point being made here is that
even if they can reproduce the state-of-the-art technology,
instead of creating it, it is good. This simply reiterates
the point made earlier that a “frontier following' strategy
is useful for technology assimilation. The downstream
benefit of not being completely autonomous and self-reliant
is that the local industry invests relatively less capital
in R8D (thereby reducing indirect costs). On the other hand,
if Brazil pursues the “catch-up“ strategy to create cutting
edge technology for the industry to move up the technology-
ladder, TNC participation becomes important. A case in point
was the super-mini project where access to international
technology became necessary. The government immediately set
up licensing agreements with TNCs to achieve this
capability.

One must realize that the market reserve policy for
microcomputers was only part of the whole strategy, which
also included a wide variety of associations with foreign
companies at different technology levels. This is a
reflection on the pragmatic nature of Brazilian policy
makers. A complementary role for TNCs was more successful in
the technological development of the industry rather than

complete protectionism. The regulations were able to develop

101
national competence and stimulated TNCs to behave in ways
they would not have otherwise. Because of this policy, and
because there were sophisticated manufacturers in Brazil,
TNCs were willing to talk about joint ventures and
technology transfer.

The other area of concern is the orientation of
Brazilian output towards the domestic market. This precludes
the manufacturing of the sophisticated microcomputers (16-
bit) at adequate scales of efficiency. The high cost/price
experience, as a result of low production, tends to confirm
the significance of export orientation-to the extent that
economies of scale is achieved. The basic point is that the
policy should be dynamic so that, by increasing pressure for
efficiency, the ”infant industry” grows up.

There would be little gained from forcing market reserve
higher towards the mainframe range, and much to be lost in
terms of increased costs. In this regard the original
perception of the architects of the regulations appeared
pragmatic and accurate: Brazil could aspire to manufacture
the smaller, less sophisticated computers, but an attempt to
produce the sophisticated mainframes without participation
by the TNCs would be a mistake. At a macro level, the
Informatics policy should not be restricted to just market
reserve where the primary goal is the achievement of
technological autonomy. In order to make it a truly
effective policy it should strive for a higher level of
technological capability, as well as try to incorporate it

102
in other areas of the economy. It should also integrate
sectorial policies from education, human resources training,
industry, telecommunications, and employment. (Straubhaar,
1991).

For those LDCs contemplating the promotion of indigenous
high-technology industries (such as computers), key policy
considerations culled from the preceding chapter can be
useful.

Even those LDCs with the technical resources to make an
entry into high-technology production feasible, need to
select carefully the appropriate point of entry. Few, if
any, LDCs can ever expect to become over-the-board producers
of all the technologies involved, ranging from the
relatively simple ones to the very sophisticated end. For
example, not many countries can hope to manufacture all
sizes and classes of computers. Due to lower capital
requirements, and the lesser level of complexity involved,
products at lower technology levels holds somewhat greater
potential for entry by LDCs (such as microcomputers).

The second important consideration is the influence of a
pragmatic regulatory environment, which accepts the
influence of TNCs as a major source of high-end
technologies. And finally, science and technology planning,
which encompasses all the relevant sectors of the economy
effected by technology, should be prioritized.

The policies which LDC governments can pursue to achieve

their technological development capabilities are elaborated

103
in the following section. These objectives can be achieved
through systematic TT (by introducing regulations): and by
enhancing the efficiency of transfer (by improving domestic
capabilities).

MW

It is imperative that coherent and comprehensive
national science and technology policies are created to
improve economic and technological conditions within LDCs.
The technological means to achieve their objectives must be
acquired by the LDCs if development is to have the chance of
success. Technology should play a central role in national
development plans. Acquisition of technology in itself,
however, will not lead to economic growth, but the ability
to utilize and adapt it will. It is important to understand
at the outset, the major distinction between acquiring
technology and acquiring technological capability. One
requires the strategic use of TT channels, and the other a
pragmatic blend of science and technology planning. It is
possible, for example, to acquire foreign technology through
DFI, licenses, know-how agreements etc. Technological
capability, however, is acquired only through
industrialization, formal education, on-the-job training,
experience and specific efforts to obtain, assimilate,
adopt, or create technology (Dahlman, 1984).

Moreover there are levels of assimilation, ranging from

equipment being dumped in LDCs where it is operated at low

104
capacity and high costs, to full technological capability
being transferred, where it is operated efficiently. The
various possible stages of TT are shown in the figure below.
How far up the ladder a country wants to go depends on the
technology in question and capability of the country.

In the process of development, most countries strive to
mount the ladder of full assimilation for three reasons:
firstly, higher stages permit more efficient operation of
the technology: secondly, the ability to operate at a higher
stage indicates indigenous capability (both technologically
and organizationally) and therefore increased ability to
acquire and bargain over new transfers: and thirdly, in the
higher stages LDCs can take control over the direction of

technical change (Lall, 1987).

Lexels_9f_Testhles¥_Assuisifien
INNOVATIVE Ability to generate next
(Advanced design) generation system
ADAPTIVE Ability to adapt product
(Technological design 8 reengineer
self-reliance) production process
DUPLICATIVI Ability to expand output
(Intermediate without further foreign
level) assistance
OPERATIONAL Ability to manage/operate
(Basic level) production facility

designed 8 built by
foreign partner

(Lall, 1937)

105

All this calls for purposeful state intervention to
bring about science and technology development. This also
implies a rejection of the concept that market forces on
their own will lead to the development of technological
capabilities in LDCs. The organization of state
intervention requires a consciousness and political
commitment to the use of a variety of government policy
instruments. These instruments include those aimed
explicitly at the development of science and technology, and
those affecting other development policies that have
relevant technological implications (e.g. education,
industry, finance, etc.) (Sagasti, 1979). The development
of technological capability also includes complex activities
such as training, experience, local R8D development, and
most importantly diffusion of technology throughout the
economy. The crux of the matter is that any form of state
planning in economic development should have a technological
component to it. With the resulting effect of an
industrialization strategy that imbeds a technological
strategy.

Given the objectives of economic growth and the
development of technological capability, LDCs' policies
should therefore be designed to improve the terms of TT
(through regulations), enhance the efficiency of the
transfer (by improving local technological capacity), and
increase the appropriateness of the technology in use.

There may be some trade off between these aims, and

106
individual countries would have to decide where their
priorities lie. The recommendations made here are highly
selective and do not pretend to “cover the waterfront."
There are a whole mountain of issues that have not been
breached, including technology policies of TNCs, the poor R
8 D performance of most of the LDCs' private sector, etc.
On analysis of the previous section, two types of policies
are relevant: direct regulation of the form and terms of
technology transfer and policies to build up local
technological resources (science and technology planning).
These will be discussed in turn.
Win:

An important issue is how governments should intervene
in the form of technology acquisition, either by providing
special incentives for DFI or by limiting the areas in which
it may operate: requiring TNCs to participate in joint
ventures or permitting only arms-length TT. Different
countries have adopted each of these regulatory approaches:
for example, many countries provide generous tax incentives
to encourage DFI, while others only permit minority
ownership in industries. Singapore is a successful example
of the first approach and South Korea (until recently) has
encouraged arms-length TT. The most appropriate approach
partly depends on the stage of development of a particular
country. In the early stages, LDCs may lack the capability
to organize TT and therefore may need to rely on DFI. But

at intermediate stages of development, there is considerable

107
evidence that arms-length TT is associated with greater
technological learning than DFI (Stewart, 1990).

A variety of restrictive measures can be employed by
LDCs to promote their domestic technological capabilities.
These measures-which include tariff protection, ban on
imports, export performance requirements, and local content
requirements - are clearly not unique. Their effectiveness
depends on a number of factors, including the economic
characteristics of the particular country. A country with a
large internal market is in a more advantageous position to
impose conditions than one with a tiny market.

The nature of the industry is also relevant. On the one
hand it may be impossible to secure sophisticated technology
(e.g.telecommunication switching systems) without DFI. On
the other hand, the more sophisticated the technology, the
ability to put it all together without DFI could be less.
It may also be difficult for LDCs to get access to the
latest technological developments. This probably is the
reason why South Korea encouraged DFI for the more
sophisticated technologies and export industries (computers)
(Kim, 1988) .

Such regulations generally have the effect of increasing
the bargaining power of the LDCs. In today's market, which
is basically imperfect and oligopolist, regulations are
effective in reducing direct costs (thereby saving scarce
foreign exchange). They also protect local technological

efforts, as the government policies for the Brazilian

108
computer industry clearly shows. However, whatever
mechanisms LDCs utilize it is imperative that selective and
temporary protection of the industry is undertaken.
Government intervention is essential, but what form it takes
is equally relevant. Consistency among the various elements
of a national policy is important but this should not be
synonymous with rigidity. The state should leave open a
variety of avenues for the acquisition of technology.

There seems to be strong reasons for preferring arms-
length 'unpackaged' TT, especially in relation to building
up local technological capability. But this may not be a
realistic possibility for the more sophisticated
communication technologies, nor an efficient alternative for
LDCs at an early stage of industrialization/development,
with weak or no domestic technological capability.
anildins_Le8al_Tsshnolesisal_£anabilitx=

Similarly domestic ownership and control of technology
is not sufficient to ensure successful technology transfer.
Local efforts are definitely needed for effective and
efficient assimilation of technology because of two
important factors. Firstly, it enables adaption and
absorption of the technology transferred, and secondly, a
higher level of appropriateness to the local conditions can
be achieved.

High levels of indirect costs, such as human capital,
are essential for this process. Government policy to build

up education in general, and scientific and technical

109
education in particular, has been a vital element in the
success of technological developments in Taiwan and South
Korea, and its absence one of the major missing factors in
Brazil and many African countries.

Any LDC government with serious ambitions of developing
a high-technology industry must recognize the
indispensability of a strong technological base to support
such an industry. Even more important perhaps than capital
is the availability of highly trained and experienced
scientific, engineering, and technical personnel. Such
personnel are the key asset in the development of a high
technology industry. since the educational system plays an
important role in the preparation of the labor force, the
state has a crucial role to play in laying the foundation
for the establishment of the infrastructure.

In addition to support for general education government
aid is needed for R 8 D institutions. Local R 8 D is needed
to adapt products and technologies to local conditions,
especially to generate appropriate technologies. A close
tie with end users: a strong commercial focus: a clear,
defined, mandate and high-class personnel are essential.

Some general recommendations can be drawn from the
Brazilian case study which can aid in the acquisition and
assimilation of technology in LDCs. First, the existence of
technically qualified manpower is an essential prerequisite.
Second, old style formal R 8 D in laboratories is

ineffective by itself: research must have strong links with

110
industrial and productive activity. Third, no country can
aim for complete independence: international trade in
technology will continue to be an essential aspect of
industrial and economic development. Even though the line
dividing dependence from independence can be unavoidably
arbitrary, there is obviously a difference between a country
that imports most of its technology and lacks the capacity
to create its own, and a country that enjoys a two-way trade
in technology and could rely on its own technological
resources for a large part. The balancing act between
excessive dependence and unrealistic independence is a
pursuit which more often than not calls for a selective

involvement with TNCs, and some degree of self reliance.

WW

LDCs understand the importance of communication
technology for their development, but they are hindered from
exercising real choices in choosing strategies for
technological transformation. The international economic
system has numerous mechanisms and institutions which keep
LDCs dependent and widen the technological gap. A handful
of corporations and government agencies control most sources
of new technology. For example, in the electronics
production industry, ten TNCs based in five developed
countries manufacture most communication equipment and they
account for much of the world's R8D for information

technology (Stafford, Dunning 8 Haverich, 1980). The result

111
of this high level of concentration is a strong oligopoly on
new technology. These TNCs determine which technological
resources will be available and how much the technology will
cost.

Hence, it becomes imperative that LDCs set goals to
establish a technological base to overcome this TNC
dependence. This reduces the excessive dependency by
building capacity to partially produce their own hardware
and software. Autonomy in the production of technology does
not mean technological independence. It just means that
selective involvement and interaction with TNCs takes place
when the technology is beyond the means of LDC production.
Self-reliance in communication technology also does not mean
that LDCs should reinvent the wheel. It simply means that
they should aim to have the capability and capacity to do
so. It is also the capacity to identify national needs and
to select and apply both imported and domestic hardware to
meet these needs. All this requires political will and
national self-confidence for progressive change.

The world has become not only economically
interdependent but technologically as well. And is the case,
this interdependence is asymmetrical. Thus even countries
with strong nationalist policies, such as Brazil, have had
to articulate their development efforts with the process of
technology transfer from TNCs. Even policies as
controversial as the Informatics Law did not preclude

intimate connection with the TNCs.

112
As a matter of fact, the overall informatics policy could
not have proceeded without the cooperation of some of these
corporations.

This results in a dilemma for the LDCs regarding the
source of technology for the development of their high-tech
industries. Access to cutting edge foreign technologies is
necessary for the long-term technological and competitive
needs of the industry. At the same time this reliance can
also undermine attempts to foster local capabilities and
innovation. The state policy should try to strike a balance
between these two constraints by addressing the terms of
technology transfer (through regulations) and the
introduction of protection mechanisms. It should try to
resolve the contradictions of such a policy through
pragmatic bargaining with the TNCs and the implementation of
increasingly nationalistic policies. Nationalism, anti-
dependency, and the resulting regulation of technology
transfer: DFI: profit remittances: and market production
levels (domestic versus exports) allow governments to create
new areas of negotiations and cooperation, with the
suppliers of the technology. This addresses the proposition
put forth at the beginning of the study: that domestic
technology development policies and TT through TNCs can
complement each other.

The ability of a country to obtain better conditions in
such international bargaining largely depends on the success

of its policies in fostering some level of self-reliance, in

113
some technological field. On the other hand, TNCs also
recognize the fact that an active state, capable of
developing an infrastructure is a better economic partner
than an inefficient one. Thus nationalist technological
policies and the processes of technology acquisition through
TNCs are not contradictory but complementary, and the
conflicts arising from such interactions should be accepted
as expressions of an endless bargaining process.

Given the current interdependency of the world economic
system, the LDCs should not shy way or be defensive about
being at the receiving end of technology. If they want to
further this balance in their favor, it can only be done by
being aggressive, competitive and smart in pursuing their
self-interests. TT is an interactive process between
suppliers and users. LDCs do have a difficult balance to
strike in choosing which technologies to import and how to
import them. On the one hand, they need the efficiency and
productivity that modern communication technologies offer.
On the other hand, they do not want to pay an excessive
price (which TNCs demand): they want to avoid the
inappropriate characteristics of imported technologies: and
they want to build up their own technological capacity,
which can be inhibited by excessive dependence on imports.
9211211181211

The main arguments of this chapter are that given the
present circumstances and future possibilities, the

alternatives before the LDCs is either to stagnate or invest

114

in building advanced technological capabilities. This,
however, cannot be done under a blanket formula. The Third
World is not a monolith. There are wide variations within
it, in terms of history, social structure, culture, and
scientific and technological competence. But there is a
uniformity in terms of needs, though levels may vary.

Technology planning and choices must be governed by a
pragmatic assessment of conditions and circumstances in each
particular case, not by blanket formulas or ideological
considerations. Technological competence can only be built
by "learning-by-doing" , which requires strategic and
pragmatic science and technology planning. This opportunity
exists for a number of LDCs today, through locally available
knowledge as well as through international technology
transfer (as the case of the Brazilian computer industry so
clearly emphasizes). Both sources are equally important and
must be utilized for building advanced capabilities. There
are certain things the LDCs can do on their own. For
others, they must borrow technologies from outside, master
them, and adapt them to suit their own needs and
circumstances. Practical difficulties apart, the intense
debate between the North and the South, which borders on the
philosophical, the ideological and the polemical, often mars
the judgement of policy makers and planners.

The final chapter reflects on the theoretical issues
which have arisen because of the debates by scholars

analyzing communication technology and development.

CHAPTER V

CONCLUG ION

This concluding chapter reflects on the theoretical
issues which were raised in chapter 2. The realities at
hand, for the LDCs, are grounded to the various approaches
communication scholars have taken to technology and
development. In order to understand this, it is important to
realize that the rapidly advancing technological frontier
will inevitably have major consequences for the LDCs. Even
if, for the moment, we discount applications of new evolving
technologies within LDCs, their spread and employment in
industrial nations is likely to have significant impact on
world patterns of trade, production, investment and
consumption. On the other hand, we simply cannot discount
the fact that new technologies are a response to social and
economic needs, be it in the industrialized world or within
LDCs (Pool, 1983: Jayaweera, 1987).

Equally predictable is that modern communication
technologies will be deployed in LDCs. Governments feel the
need for economic rationalization and efficiency, and they
do believe that the introduction of new technologies will

achieve these objectives. There is also an inherent belief

115

116
that technology applications will solve the problems of
economic and social backwardness. Indeed microelectronics
and information technologies are already prevalent in the
newly industrializing countries (NICs). Other technologies
will follow, and the dissemination is likely to “trickle
down“, to some extent, to even the poorest nations. The
diffusion of modern technologies to and within LDCs will
result in even more profound changes in world, regional and
local social, political and economic changes and their
relationships (Rogers, 1989).

The literature review in chapter 2, which dealt with
technology strategies, regulations and costs, presupposed to
an extent that the most efficient mechanism for TT is via
TNCs. The Brazilian case study also partly proved that TNC
presence is an important factor for the development of a
high technology industry, such as computers. These are
points which need to be highlighted to explain what the
“mainstream" scholars have pushed for all along. Pool (1983)
had suggested that open markets, which do not restrict TT,
is the key to development and modernization. He had argued
that the complete technology package should be purchased
from the TNCs, because of the intrinsic benefits one obtains
from a 'package' deal. This was obviously based on the
assumption that domestic development of the state-of-the-art
technology was hard, if not impossible. The Brazilian
problems with minis and mainframes proved his point. On the

117
other hand, his view that new communication technology was a
response to domestic social and economic needs was also
true.

Governments do realize the significant relationship
between a favorable economy and the introduction of new
communication technologies. Schramm (1968), Hudson (1985),
and Parker (1984) had also posited that the application of
new communication technologies would increase productivity
and lead to the modernization of the economy. The creation
of the computer industry in Brazil led not only to
infrastructural development, in terms of human resources and
R 8 D, but a whole new industry developed. Though the
positive influence of computers in the Brazilian economy has
not been analyzed in this thesis, one can safely assume some
increase in business productivity and modernization.
Moreover, technological change and use of new communication
technologies does reflect a certain increase in LDC
capability to handle their economic development
independently.

Rigid controls does not seem to be the answer. Far less
predictable, however, are the future mechanisms through
which technologies will be introduced in poorer countries:
the level of geographic and sectorial penetration: the
interface between technology imports and domestic
generation: and the distribution of costs and benefits

associated with technological advance.

118

This raises one of the most fundamental problems of
development, and that is, the integration of technology into
traditional societies. LDCs are trying to move into the
twenty-first century with strong economies to serve their
people. To do so they adopt eclectic strategies of
development, using ideas drawn from socialist, capitalist
and mixed models. In a majority of LDCs, state control over
major resources is a necessity in the early stage of
development, whatever may take place later on. It is only
the state, within the LDCs, which has the power and
resources to acquire and integrate new communication
technologies into society. The Brazilian case clearly
demonstrates this trend.

In practice, the transfer of communication technology is
largely in the hands of TNCs who serve the needs of LDCs
only if it is profitable to do so. Implicit in this is the
acceptance of the truth that TT is not an altruistic
phenomenon. No country or TNC can be expected to share
their know-how with any LDC on terms that are inconsistent
with their own self-interest. They will give only if they
receive proper compensation in return. To put it
succinctly, LDCs have to recognize that technology export
has a clear profit motive (Lall, 1987, p.72).

This acceptance of transnationalization, and the
international economy is the key to LDCs planning their
technology development. Jayaweera (1986: 1987), Mattelart
and Schmucler (1985), and Servaes (1987) are probably the

119
only “critical“ scholars who appreciate and realize the
realities at hand. No matter what development strategies
LDCs choose it is this fact, that no LDC can be completely
autonomous, that becomes relevant.

Moreover a Luddite type attitude that imported
technology is inherently exploitive becomes impractical. New
communication technologies are already an established
presence in LDCs, so would it not make more sense to extract
the most out of it? The policy should not be to ignore an
existing process when it is already tied into the rest of
the economy. There is a need for advanced technology, for a
country to function as a capable and qualified economy. All
three suggest that the solution to the TNC presence does not
lie in delinking or dissociation from the international
economy, but rather from utilizing existing technologies to
improve those sectors-where productivity can be increased.
Moreover, if there is a technological dependence, this can
be rectified by balancing it with appropriate science and
technology policies which improve domestic capabilities. The
preceding two chapters illustrated how to achieve these
objectives.

National programs for improving domestic capabilities in
acquiring and using technology vary significantly due to
differences among governmental policies, and local economic
and environmental factors. The success of South Korea in
receiving and adapting foreign technology resulted in part
from state policies encouraging technology imports, by both

120
the private and public sectors, and private/public sector
cooperation in all stages of production (from education and
training personnel to research and export of products)
(Evans 8 Tigre, 1989). The Brazilian government has also
supported technological infrastructure development and has,
for example, encouraged the use of public training
institutes by private organizations. Despite continued
technology import regulations that are fairly stringent,
resulting from balance of payment problems, Brazil continues
to attract new technologies because of its strong
infrastructure and market potential.

Technology never comes cheap. It is quite wrong to
think that acquiring technology is a costless process. Every
new application of a given technology requires adaptive
engineering work. And the costs entailed are usually higher
in LDCs, where the absorptive capacity for new technologies
is weak (Teitel, 1984). Add to this the consideration that
no frontier type technology is static. All technologies
change, some faster than others, and some at greater costs
than others. Any LDC going in for the acquisition of modern
technology must have the wherewithal to keep abreast of
technological frontiers (or just behind it). It must also be
accepted that TNC investment can promote an efficient
technology transfer to LDCs in a variety of circumstances.
Where the line is drawn between autonomy and "dependence,"
must be a very pragmatic issue, not ideological. It would
appear that the industrialization experience of leading LDCs

121
has made them pragmatic in this sense. Their need for
advanced communication technologies, the realisation of
their handicaps and the pace of technological change, has
led even those NICs adverse to TNC entry (Korea among the
dynamic ones, India among the laggards) to believe they must
be liberal to an extent (Lall, 1985).

The question which still remains unanswered is whether
technological applications will solve problems of
backwardness.

Hamelink and Schiller argue that LDCs should isolate or
delink themselves from western science and technology, that
culture should be protected from technological onslaughts.
In some situations, cultural and technological disengagement
11gb; stimulate local technological capability. But this is
not a viable option. The use of satellites and computers
requires interaction with foreign technology. LDCs can not
ignore the international economic system. Besides, the
belief that LDCs are being exploited by the TNCs, through
the export of technology, leads to the assumption that
industrialization within LDCs is possible uithgu; the
infusion of technologies. And that such imports are
beneficial 9n1y when under stiff control by the
bureaucracies. This belief diverts the attention of the
LDCs from their own failures, to build on their indigenous
potential for producing technology, and to make a valid
assessment of alternative foreign technologies. There is

little evidence that the inflow of foreign technologies

122
ultimately damages the emerging technological capability of
LDCs. It is more likely to help LDCs in upgrading their
technological levels than to cause harm.

These substantive and very real issues were ignored by
the ”critical” scholars for a long period of time. Today
Schiller agrees that total exclusion of TNCs is impossible
and insists on a “cautious" choice (Tsui, 1990). Similarly,
Hamelink's push for a total dissociation of all relations
between LDCs and TNCs, does seem rather unrealistic. Not
only does they ignore the internal social and economic
dynamics of LDCs, but also the fact that the demand comes
from within.

We must not reject communication technologies, much less
wage an ideological war against them and their creators, the
TNCs (since it is their job to pursue a healthy bottom
line). My argument is that the assumptions and beliefs
underlying the use of these technologies (such as computers)
in the LDCs are wrong. Technologies cannot provide
solutions to problems that are primarily political, economic
or sociological. And if they are used as an alternative to
domestic reforms (which usually is the case), they are more
likely to consolidate and perpetuate these problems.

The problems of underdevelopment are usually not because
of technological backwardness. It is the result of a lack of
political will to bring established knowledge, be it
domestic or from TNCs, to bear on questions of social,

economic and technological relevance. It is the and goals

123
of individuals that make science and technology a problem
issue for society. The crux of the matter does not always
lie in providing more funds, more personnel and new
laboratories for the science and technology sector. The
more pressing problems for the efficient use of technology
lie outside the sector itself. They concern matters such as
the organization of the country's economic sector and
improving the quality of life of its people by developing

the basic infrastructure.

BIBLIOGRAPHY

BIBLIOGRAPHY

Adler. E. (1987). The_nexer_of_ideolos!1__The_snss§_for_
teshnolesisal_aut9noax_in_Arsentina_and_Brazili_
Berkley, CA: University of California Press.

Amsden, A. 8 Kim, L. (1982). Korea's technology exports
and acquisition of technological capability. flgr1g__
Bank: Development Research Department. Washington,
D.C.

Archer, R. (1986). Brazil's Informatics policy achieves
success.
9:10, October.

Barancon. L- (1981) Nerth:Seuth_teshnelesx_traneferi_
finansinsins_and_instifutien_huildins. Mt- Airy. MD:
Lomond.

Bhalla, S.A. 8 James, D. (1988). Some conceptual and
policy issues. In A.S. Bhalla 8 D. James (Eds.), Hey

fe8hn9l9siss_and_dexelenment1__zxnsriense_in__
lisghnglggz_hlend1nu'. Boulder CO: Lynne Rienner.

Botehlo, J. (1987). Brazil's independent computer strategy.
Teshnolesxnnexiex. 90:4. Hay/June-

Bruce, R. R., Cunard, J. P. 8 Director, D.M. (1988). Ing_

 

Lendon: Butterworth Scientific.

Cline. R.W. (1987)- Inf9rmatiss_and_dexslonmenti__1rade__

Washington D.C.: Economic International, Inc.

Dahlman, J.C. (1984). Foreign technology and indigenous
technological capability in Brazil. In M. Fransman 8
K. King (Eds.),
E9111 (pp.317-335). London: MacMillan.

Dahlman, C.J. 8 Sercovich, F. (1984). Exports of
technology from semi-industrial economies and local

technological develop-ant. lenrnal.of_nexelonment_
124

125
lggngm1ug, V01. 16, Nos. 1/2, Sept.-Oct, 63-99.

Enos, J.L. (1982). The choice of technique vs. the choice
of beneficiary: What the Third World chooses. In F.
The_esenemiss_ef_new_

Stewart 8 J. James (Eds.),
(pp.69-83). London:

Frances Pinter.

Erber, F. (1981). Science and technology policy: A view
from the periphery. In J. Szyliowicz (Ed.), Igunnu1ggy_
in_Intsrnatienal

Altair: (pp.190-191). New York:
Praeger.

Erber, F. (1985). The development of the ‘Electronics
Complex' and government policies in Brazil. World

ngu1gpmung, vo1.13, No. 3, 293-309.

Evans, P. (1986). State, capital, and the transformation of
dependence: The Brazilian computer case. fig;1g_

DEYEIQRIQDE: 1437: 791-803-

Evans, P. 8 Tigre, P.B. (1989). Going beyond clones in
Brazil and Korea: A comparative analysis of NIC
strategies in the computer industry. figm1g_
nuuu1gpmunt, Vol. 17, No. 11, 1751-1768.

Folder, A.E. 8 Hurrel, A. (1988). Th§_nufiuznrazilian__
Informatics_diznnte. washinqton DC: Foreign Policy
Institute, John Hopkins University.

Fransman, M. (1984). Conceptualizing technical change in
the Third WOrld in the 1980s: An interpretive survey.

Iourna1_ef_nszslonment_§tudies. Vol-21. No.4. 572-652.

Frischtak, C. (1986). Brazil. In F. Rushing 8 C.G. Brown

(Eds.),
. Boulder, CO: Westview Press.

Halolink: C- (1983)- cnltural_autonomx_in_slohal_
ugmmun1uu§19ng. New York: Longman.

Hamelink, C. (1984). Information technology and

international relations: Perspectives for North and
South. In J. Becker (Eds.),

Infornatien_teshnelos¥_and_
a_Naw_1ntgrnut12nu1_9rger. Student- Litteratur: Lund.
Hanolink. C- (1984)- Transnafienal.data_flexs_in_the

Infigzmatign_Agg. Studentlitteratur: Lund.
Hamlink, C. (1985). High technology transfer: Selling the

126

canoe without the paddle. ngyg1gm-unt, 28-37.

Hobday, M; (1986). Telecommunications-A leading edge in
the accumulation of digital technology? Evidence from

the can of Brazil. WW. 110- 103.
23-43e

Hoffman, K. 8 Rush, H. (1980). Microelectronics industry
and the Third World. zugumgu, Aug, 289-301.

Hoffman, K. (1981). Third World industrialization
strategies in a restructuring world economy: The role
of technical change and innovation policy. Paper
submitted to UNIDO Global and Conceptual Studies
Branch (mimeo) .

House, R. (1987). Counting the cost of the micro boom.
5.211131: 81: Jul-Y-

Hudson. 8- E- (1984)- WWW.
Norwood, NJ: Ablex.

Hudson, H. E. (1985). Risks and rewards: Why have'nt
telephones reached the village? In H.E. Hudson (Ed.),

fuz_flgr§h_un§_§gu§h. Dedham, MA: Artech House.
Jayaweera, N. (1986). MW
WW

Occasional Paper No. 20, Singapore: AMIC.

Jayaweera, N. (1987). Rethinking development
communication: A holistic view. In N.Jayaweera 8 S.
Amunugama (Eds.),

(pp. 76-94). Singapore: AMIC.

Kaplinsky, R. (1984a). Trade in technology-who, what,
where 8 when? In M. Fransman 8 K. King (Eds.),

(pp.139-
161). London: Macmillan.

Kaolin-Ry. R. (1984b). WWW
minty. London: Macmillan.

Kim, K. (1988). Korea in the 19908: Making the transition

to a developed economy. flgz1§_ngy§1ummgnt, Vol. 16,
No. 1, January.

Katz, J. (1984). Domestic technological innovations and
dynamic comparative advantages: Further regulations

on a comparative case study program. 12urna1_91_
' V01 e 16 ’ NO 0 1/2 ' sept‘OCt e

127

Lall, S. (1984). India's technological capacity: Effects
of trade, industrial science and technology policies.

In M. Fransman and K. King (Eds.), Igghnglggignl_
(pp.225-245). London:

Macmillan.

Lall. 8- (1985). Mnltinafienala1_technolesx_and_exnerts.
New York: St. Martin's Press.

Lall. 8. (1987). ALearnins_te_industrializei__The_
London: Macmillan.

Lee, c.c. (1980). -
. Beverly Hills:

Sage.

Manser, W.A.P. & Webley, S. (1979)- Teshnolesx_transferito_
duyg1gp1ng_ugun§r1§§. London: Chatam House Papers.

Merton, K. (1986).
. Lexington, MA: Lexington Books.

Mattelart, A. 8 Schmucler. (1985). gummuniuutiun_und_
America? New York: Ablex.

nattos, s. (1982).
San Antonio, TX: Klingen-Smith.

Molnany. 8- (1980)- cemmunisation_in_the_rural;Third_

° . New
York: Praeger.

Mody, A. (1989). Information industries in the newly
industrializing countries. In Crandell and Flamm

(368-). changing_the_rnlss (pp- 298-327)- Brookings
Institute.

Nelson, R.R. (1978). Innovation and economic development:

theoretical retrospect and prospect. firudig§_1n_
. Buenos

Aries: IDB/CEPAL.
Nottlcton. 6- (1986)- Brazills_telesemmunicationis_
satellite_axatem1_The_nelitisizatien_ef_

flgr1u. M.A. Thesis, University of Texas, Austin,
Dept. of Radio/Television/Film.

New York Times. (1985). 18th November. Section D, pp. 9.

128

Parker, E. (1977). Planning communication technologies and
institutions for development. In S.A. Rahim 8 J.

Middleton (Eds.),
(pp.43-76). Honolulu: East-West Center.

Parker E. (1982). Communications satellites for rural
development. In J.R. Schement, F.G. Guttierez 8 M.A.

Sirbu (Eds.),
(pp.3-9). New York: Praeger.

Parker, E. (1984). Appropriate telecommunications for
cconolic dcvolop-ont. Tsleeesmunisatien8_zelisx. Sept.
173-180.

Parker, E. (1985). Financing and other issues in
telecommunications development. In H.E. Hudson (ed.),

220). Dedham, MA: Artech.

Pool, I. De Sola. (1979). Direct broadcast satellites and
the integrity of national cultures. In H.I. Schiller 8

K. Nordenstreng (Eds.),
' (pp. 120-153). Norwood,

NJ: Ablex.

Pool. 1- Dc Sola- (1983)- Technologiee_f2r_freedem.
Cambridge : Belknap.

Rana-urti. 8- (1987)- afat8:ewned_entsrn:iaes_in_hish_
feshnelesx_industriesl__Stndies_in_1ndia_and_nrazil

New York: Praeger.

Rogers, E., Fulk, J. 8 Glinow, M. (1988). Managing change
through communication technologies in Third World
countries. I9nrnal_ef_Qrsanizafienal_£hanse_and_

Management. 1:2: 219-250-

Roqerc. B- 5 sinqhal. A- (1939). Insials_information_
ruyg1urign. New Delhi: Sage.

Rogers, E. 8 Dearing, J.W. (1990). University-industry
technology transfer in the Information Society. In
B.D. Ruben 8 L.A. Lievrouw (Ed8.) Mediurigny

bghay1ur_yg1_2. New Brunswick, NJ: Transaction

Publishers.
Rosenberg, N. (1982). '
ugungl1gu. Cambridge: Cambridge Univ. Press.

Rushing, F.W. 8 Brown, C.G. (1986).
. Boulder CO:

129

Westview Press.

 

Secesti. 1L!- (1979)- 19.81111919511an
WWW. New York:
Praeger.

Schramm, w. (1968). W
W- Paric: UNESCO

Schramm, W. (1971).
working paper for the Academy for
Educational Development. Washington D.C.

Schiller, H.I. (1975). The appearance of national
communications policies: A new arena for social
struggle. 9:13:33, Vol XXI, No. 2.

Schiller. H I. (1981) WWW
Igrtung_§gg. Norwood NJ: Ablex.

Serve“. 51- (1987) WWW
. Paper presented at UNESCO International

Seminar, Kaula Lumpur, Malaysia.

Serve“. J. (1988)- W- Paper
presented at the XVI Conference of the IAMCR,
Barcelona, Spain.

Servaes, J. (1989). Apres le deluge. In B. Dervine, L.P
Grossberg, B.J. O'Keefe 8 E. Wartella (Eds.),
(pp.214-218), vol. 1,
Paradigm Issues. Newbury Park, CA: Sage.

Singh, K. D. N. (1981). Long term needs and expectations of
developing countries in technology licensing.
, Atlanta, Georgia, Annual

Meeting, Sept.

Stewart. P- (1977). W.
London: Macmillan.

Stewart, F. (1990). Technology transfer for development.
In G. Ranis 8 R.E. Evenson (Eds.),
° . Boulder,

CO: Westview Press.

Straubhaar, J. 8 Sanger, E. . (1990). arguking_

W
W. Paper presented to ICA. Dublin.
1990.

Straubhaar, J. (1991). Communication, culture, and

130

Informatics in Brazil: The current challenges. In G.

Sussman 5 J. Lent (Eds.), Transnatignn1_
eaImuni2ati9ns1_Ei:ing_tbs_1hird_uozldl Sage: CA-

Stapford, M.J, Dunning, B.J 8 Haberrich, 0.1:. (1980). 1119.
. New

York: Facts on File Vol.2.

Sweeney, G.P. (1981). Telematics and development. The
informatinn_soeietx. V011. No.2-

Teitel, S. (1984). Technology creation in semi-industrial
economies. I2nrnal_91.nexelonment_zcgnenics. 16. 39-61-

Tigre, B.P. (1983).
. New York: St. Martin's
Press.

Tsui, L.S. (1991). An interview with Professor Herbert
Schiller: Media_ne!elenment. 1. 50-52-

UNCTC- (1984)- IIansb9rder_data_fleus_and_nrazil- Prepared
by SEI, New York: United Nations #175.

UNCTC. (1983).
° . New York: UNCTC pp. 138-

187.

Vernon, R. (1990). Trade and technology in developing
countries. In G. Ranis & R.R. Evenson (Eds.), figigngg_

(pp.255-271). Boulder, CO: Westview Press.

Wad, A. (1982). Microelectronics: Implications and

strategies for the Third World. Ih1:g_flgglg_gugztgzly,
Vol. 4, No. 4, Oct.

Westman, J. (1985). Modern dependency: A crucial case study
of Brazilian government policy in the microcomputer
industry. ' '

WI Summer, 25-47 -

Westphal, L.E., Rhee, Y.W. & Pursell, G. (1984). fig:ggn_
° . World Bank

Staff WOrking Paper, No. 469, washington, D.C.

Wionczek, S.M. (1981). On the viability of a policy for
science and technology in Mexico. Lg§1n_5nexigan__
Bessarsh_ne!ie!. 16. Ne- 1. 57-78-

Wirth, J., Oliveira de Nunes, B. & Bogenschild, T.
° Boulder,

13 1
CO: Westview.

Yin, R.R. (1981). The case study crises: Some answers.
, 15 (March), 58-65.

Yin. mm (1989)- W
Newbury Park, CA: Sage.

"IT'EflfiflfljflfliflWfl’Lfliiflfil‘T