Discussion Paper Series

#2004-4
Science and Technology in Latin America and
the Carribbean: An Overview

Léa Velho

February 2004

United Nations University, Institute for New Technologies, Keizer Karelplein 19, 6211 TC Maastricht, The Netherlands
Tel: (31) (43) 350 6300, Fax: (31) (43) 350 6399, e-mail: postmaster@intech.unu.edu, URL: http://www.intech.unu.edu
 SCIENCE AND TECHNOLOGY IN LATIN AMERICA AND THE
CARRIBEAN: AN OVERVIEW

Léa Velho

Abstract

This paper presents and analyses a series of indicators of the state of S&T in Latin America and
the Caribbean. In doing so it compares LAC countries among themselves as well as with the
industrialised countries in order to identify trends concerning the closing or widening of gaps in
the capacity to produce and utilise knowledge. The findings suggest that some LAC countries
have created significant capacity in research but they have not been able to build virtuous links
among the various relevant social actors involved in knowledge production and use. There are
problems at both the supply and demand sides. Concerning the former, universities and public
research institutes, which together perform almost 70% of R&D, have not created mechanisms
to identify user needs and instead base their research agenda on scientific criteria dictated by
international mainstream science. On the latter, there has not much demand on local R&D given
that TNCs innovate on the basis of R&D conducted in the advanced countries and local private
firms, in a bid to be competitive, also prefer to import foreign technology. Government initiated
schemes to correct this imbalance - including provisions for public-private collaboration,
science parks and firm level research training - have been largely unsuccessful. This state of
affairs is largely due to the enactment of macroeconomic policies aimed at opening up the
country to foreign competition and privatising state enterprises, without putting in place the
necessary measures and incentives to guarantee the investment and to diminish the risks
involved in R&D. Firm and focused government intervention is necessary if such trends are to
be reversed. The most obvious sphere for intervention is creating a more just society in the
region, and thus granting opportunity to education to social groups currently excluded. Other
policies have more direct bearing on R&D and include incentives for private sector investment
in R&D and hiring of researchers. Most of all, measures must be implemented by all relevant
S&T actors in terms of strengthening links among themselves.

Key words: Latin America and the Caribbean, science and technology indicators, science and
technology policy

UNU-INTECH Discussion Papers

ISSN 1564-8370

Copyright International Development Research Centre 2003. Funding for this paper was
provided by the International Development Research Centre, Ottawa, Canada. Any views,
findings or conclusions presented in this paper are solely those of the author(s) and not of IDRC
or its Board of Governors.

Published by the United Nations University Institute for New Technologies (UNU-INTECH).
UNU-INTECH discussion papers intend to disseminate preliminary results of the research
carried out at the institute to attract comments
TABLE OF CONTENTS

1.0 INTRODUCTION 7

2.0 THE R&D SYSTEM IN LAC ACCORDING TO INPUT AND OUTPUT BASIC
INDICATORS 11
2.1 EXPENDITURE IN R&D 11
2.1.1 Intensity of expenditure 11
2.1.2 Expenditure by financing sector 13
2.1.3 Expenditure by socio-economic objective 15
2.2 PERSONNEL 16
2.2.1 Availability and sectors of activity 16
2.2.2 Socio-economic and gender participation 18
2.3 THE OUTPUT OF R&D 21
2.3.1 Publications 21
2.3.2 Patents 26
3.0. S&T AND R&D ORGANISATIONS IN LAC 29
3.1 GOVERNMENT: S&T P OLICY FORMULATION AND RESOURCE ALLOCATION
INSTITUTIONS 29
3.2 INSTITUTIONS OF HIGHER EDUCATION 31
3.3 PUBLIC RESEARCH INSTITUTES 35
3.4 BUSINESS ENTERPRISES 36
3.5 NON-GOVERNMENTAL ORGANISATIONS – NGO S 39
3.6 INFRASTRUCTURE TO ARTICULATE SUPPLY AND DEMAND OF R&D 40
4.0 CONCLUSIONS 43

BIBLIOGRAPHY 47

TABLES 53

THE UNU-INTECH DISCUSSION PAPER SERIES 75

1.0 INTRODUCTION

The Latin America and the Caribbean region (henceforth LAC) comprises 34 independent states
and 12 territories inhabited by 524 million people. The average per capita income in the region
is about (PPP) US$ 7,200 per year. The population is predominantly urban based and is
differentiated by deep wealth inequalities, with almost one-third of people living in poverty
(defined as living on less than US$2 a day). Despite historical, language and cultural
commonalties that have created a measure of regional identity, diversity in LAC is as important
as its homogeneity. Besides Spanish and Portuguese, the main languages spoken in the region,
English, French and some 400 indigenous languages are the only communication means in a
few countries and of a number of social groups in various others. LAC topography and
ecosystems range from tropical islands to high mountains, rainforests, deserts, savannahs and
frozen lakes. LAC includes some of the developing world’s largest countries, such as Brazil
(170 million people) and Mexico (about 98 million), to some of the smallest, such as Grenada
(74,000) and the isles of St. Kitts and Nevis (42,000).

During the 1990s and up to 2002 the annual per capita growth rate of LAC was below 2%, with
an amazing degree of homogeneity across countries (Lopez, 2003). Economic stagnation across
the region has not only increased the development gap between LAC and the traditionally more
advanced countries, but also between LAC and other countries such as South Korea and Taiwan
which, two decades ago, were at a similar development level.

The reason for this state of affairs is complex and manifold. In the realm of knowledge
development, however, there is considerable agreement that in order to enhance their economic
growth prospects LAC countries must improve their ability to produce, select, adapt,
commercialise, and utilise knowledge. Doing so requires the articulation of focused strategies to
address the region’s deficits in skills and technology. As part of this strategic approach there is
need to create and/or strengthen local R&D institutions in the public sector and to stimulate
them to pursue activities that are relevant to development needs. In addition even more effort
has to be put on fostering an environment that encourages the private sector to engage in this
kind of activity.

Increasing the local level and quality of R&D, however important, is not enough. It is generally
accepted today that the countries that successfully evolved into knowledge societies had a well
functioning system of innovation. A system of innovation consists of a network of social actors
– firms, public and private laboratories, universities, professional associations, trade unions,
grassroots organisations, etc - together with the institutions 1 and policies that influence their

7
innovative behaviour and performance (Freeman, 1987; Lundvall, 1992; Nelson, 1993). The
most important element in the system is not so much the strength of the individual actors as the
links between them. The system of innovation may operate more effectively (in terms of
learning and in generating innovations) in a system where individual actors are not particularly
strong but have well developed links between them than in a system comprising some strong
actors but with weak links between them.

While the relevant literature does not necessarily suggest that national governments in the
technologically successful countries have explicitly sought to build innovation systems, some
have clearly done so2 . This suggests that for the countries desiring to become knowledge
societies, the system of innovation concept “carries a normative weight” (Arocena & Sutz,
2000: 58). However, rather than presenting a blueprint for achieving this, the system of
innovation approach is concerned with identifying the participant social actors, mapping the
knowledge flows among them, identifying bottlenecks and suggesting remedial actions. Such an
exercise can be done at different levels: for a whole country, for specific regions or sectors
within a country, or even for particular technological innovations (Clark et. al, 2003).

This paper aims to make a small contribution to the task of devising a strategy to build
innovation systems in LAC countries. Whatever the strategy is, and it clearly is not the same for
all LAC countries, it requires a background picture of the local environment of knowledge
production and utilisation. The objective here, thus, is to provide such a picture. This is
achieved by mapping out the data and trends on S&T and R&D as well as the organisations
involved in such activities. In addition, as this kind of information is more useful for policy
making purposes when it provides a comparative analysis, this exercise compares LAC
countries among themselves as well as with some industrialised (and newly industrialised) ones
and tries to identify trends concerning the closing or the widening of the gap in R&D.

The first section concentrates on the presentation and analysis of basic input and output data
related to S&T and R&D, that is, the supply side of S&T in terms of financial and human
resources as well as results (publications and patents). This is followed by a second section
focusing on the organisations (social actors) that supply and demand S&T and R&D. It is
recognised, however, that a supply/demand framework for the analysis of S&T organisations is
a gross oversimplification, particularly when one has a systemic approach in mind. University
researchers, for example, can be both on the demand side for financial inputs as well as for
research results on which to build up their own work, as well as on the supply side, as
knowledge providers. For this reason, and given the importance of interactions among social
actors highlighted in the systems of innovation framework, an effort is made to identify and
describe such links (and the initiatives to build them) between knowledge producers and users in
LAC. This may be a preliminary indication of whether a system of innovation is in place or,

8
perhaps, being built in the region. The final and concluding section sums up the main issues
identified previously and raises the question of the continuation or not of present S&T and R&D
trends in LAC.

The picture that emerges in the following sections is based on both quantitative and qualitative
information derived from multiple sources with varying degrees of reliability. Considerable care
was taken not to draw strong conclusions from weak data as well as to cross check data sets
from different sources and to balance the analysis with information provided by relevant
empirical research as well as by interviews with key informants.

9
10
2.0 THE R&D SYSTEM IN LAC ACCORDING TO INPUT AND OUTPUT
BASIC INDICATORS

2.1 Expenditure in R&D

2.1.1 Intensity of expenditure

A country’s commitment to becoming a “knowledge society” is partly reflected in the financial
resources devoted to Scientific and Technological Activities (S&T) and, more specifically to
research and development (R&D)3 . Although all LAC countries fund some kind of S&T, not all
of them have a specific budget line for this and some do not keep statistics on this kind of
expenditure. Among those that do, some countries collect information for S&T only, others for
R&D only and, from 1995 onwards, thanks to the efforts of the Iberoamerican Network of S&T
Indicators, RICYT4 , there is a tendency to present data for both categories. The discussion here
will take this distinction into account and the comparisons made are mostly about R&D only.
However, when it involves all scientific and technological activities (S&T) this will be
explicitly pointed out.

S&T investment (R&D included) of the LAC countries amounted to more than US$ 15 billion
in 2000, of which around US$ 11 billion was spent on R&D (see Table 1). This latter volume is
still low when compared to advanced countries and to the Asian NICs. The investments in R&D
in 2000 by just two countries, Canada and South Korea (each investing about US$ 12 billion)
for instance, exceeded those of all LAC countries together in that year (NSB, 2002). This
notwithstanding, the region’s investment in R&D practically doubled during the last decade
(Table 1), while the R&D investment of the European Union increased by 32% in the same
period (Urzúa, 2002).

Although this magnitude of increase in R&D spending is generally accepted (see Hill, 2000;
Hansen et al, 2002; and Urzúa, 2002), a word of caution is necessary: more LAC countries
provided information in 2000 than in 1990, and the overall increase may be attributed, at least in
part, to the aggregation of newcomers to statistics collection. Despite overall increases, some
countries (Brazil, Chile, Cuba and Venezuela) have experienced a number of kinks on their
spending during the 90s. Such fluctuations may be a combination of two things: on the one
hand, the effect of difficult economic and political circumstances, which stretched to the limit
the commitment of governments to S&T activities and were detrimental to the overall
performance of the system. This was the case of Brazil, for example, when resources for the so-
called Centres of Excellence in research were drastically cut in 2000, affecting most notably the

11
hiring of young researchers, purchase and maintenance of equipment and institutional
collaboration, areas that are decisive for the competitiveness of the S&T system (Hansen et al,
2002). On the other hand, kinks in the data may simply reflect problems in data collection, such
as conceptual and methodological changes. Brazil is also a case in point here for as the Ministry
of Science and Technology warns, R&D data for 2000 onwards are not strictly comparable to
those of the previous years5 .

As is to be expected the absolute volume of expenditures in R&D also differs enormously

among LAC countries, given their varying stages of development, size of the economy and
population. Brazil is by far the frontrunner, accounting for 42% of total LAC expenditure in
2000, followed at a considerable distance by Mexico (20%) and Argentina (11%). This position
of Brazil in LAC resembles that of the US vis-à-vis the other OECD countries. Controlling for
population sizes, however, Argentina is in a better position than Brazil (33.7 and 27.1
respectively) and both Chile and Mexico are very close followers in terms of R&D spending per
resident (about 26.0 and 23.3, respectively in 2000, Table 1).

Considering the more usual and relative indicator of R&D expenditures to GDP, the average for
LAC countries was 0.57 in 2000 and it has remained relatively stable through the past decade
(RICYT, 2002 and Table 1). This expenditure rate compares very unfavourably with advanced
countries and Asian NICs like South Korea, which invested from 2.2 to 3% of GDP on R&D
(with the exception of Italy and Canada, which have fluctuated narrowly at 1.0 and 1.6,
respectively in the last five years, NSB, 2002). However, it is also the case for the latter that the
latest R&D/GDP ratio is no higher now than it was at the start of the 1990’s. In both LAC and
OECD countries, the stagnation of R&D/GDP rations seems to have been a result of reduced or
level spending by governments.

A country by country analysis shows that only Brazil had consistent investment rates higher
than the regional R&D/GDP average since 1990 (from 0.76% in 1990 to 1.05% in 2000, which
is at the same current level as Italy), while Chile fluctuated from above average in 1995 to
average in 2000. Cuba is an interesting case in point where R&D/GDP in 1990 was quite high
for regional standards (well above LAC average and comparable to Brazil at 0.70%) which
explains the research capabilities the country was able to achieve particularly in medical
sciences and in biotechnology (Thorsteinsdóttir et al, 2003). These levels dropped consistently
during the 1990’s due to the loss of political and economic support from the former communist
block. None of the other countries reached the regional average for R&D/GDP during the 1990s
up to 2000. Thus, to the extent that the ratio of R&D spending to GDP is a reliable indicator of a
country’s commitment to growth in scientific knowledge and technology development, LAC
countries, perhaps with the exception of Brazil, still have a long way to go. The picture becomes

12
even more unfavourable when one looks at where the financial resources are coming from, as
discussed below.

2.1.2 Expenditure by financing sector

International standards attribute sources of funding to the following sectors: government (all
levels), business enterprises, higher education, private non-profit organisations, and funds from
abroad (NSB, 2002). This classification, which is followed by Ricyt and presented in Table 2,
makes difficult the analysis of public versus private investment in R&D. Nonetheless,
inferences may be drawn on the basis of available knowledge.

The most noteworthy characteristic of R&D expenditures in LAC is the crucial role played by
government: about two thirds of R&D funding in 1990, decreasing and stabilizing around 57%
from 1995 onwards. However, significant public spending is channeled to other sectors such as
‘enterprise’. In the case of Brazil, for example, it has been estimated that public enterprises
provided one third of industrial R&D funding in 1998 (Hansen at al, 2002). If it is reasonable to
apply this estimate to the overall figures for LAC, public spending in 2000 would amount to
57% (government) plus 11% (one-third of enterprise spending). In addition, as most higher
education institutions performing research in LAC tend to be public, a conservative estimate
would be that half of R&D spending in this sector was provided by public funds. As a result,
public funds for R&D in LAC may well account for about 70% of R&D expenditures in LAC,
while private firms would be responsible for no more than 20% (these figures are very close to
those estimated by Urzúa, 2002). This picture is the opposite to the high relative importance of
the business sector in R&D funding in South Korea and Japan (over 70%), the US (69%) and
the European Union (54%) (NSB, 2002).

The share of enterprises engaged in R&D funding has risen gradually in LAC during the 90s -
from 26.1% in 1990 to 32.6% in 2000 (see Table 2). This trend applies to most of the countries
in the region, although in only a few of them (Brazil, Colombia, Cuba and Uruguay) the
business sector seems to play a significant role with a share around 40% of total R&D expenses.
A significant part of this amount comes from public funds, as discussed in the above paragraph.
Although these public enterprises are expected to adopt a business logic in relation
technological innovation and R&D spending, only a few, such as Petrobras (oil industry) in
Brazil, are able to cover their R&D costs from profits and external contracts. Finally, for a
number of LAC countries, such as El Salvador, Panama and Paraguay, the enterprise sector
funds less than 5% of total R&D activities.

The unequal relationship in R&D expenditure between the public and private sector, as
estimated above, is an impediment to further advancement by LAC countries in their
technological development. No country has ever secured advanced technological capabilities

13
without significant private R&D expenditures. Moreover, to become technology-followers
countries need to have a critical mass of business enterprises that are willing to invest in R&D.
The reasons are that R&D units located in firms allow “better and faster diffusion within the
economy of new technologies, lowers the cost of technology transfer and captures more of the
spillovers benefits created by the operation of foreign firms” (Lall, 2002: 3). As a matter of fact,
it is the countries that were able to make the transition to knowledge economies that show a
consistent and marked increase in private sector participation in R&D investment over time
(South Korea and Ireland for instance).6

Another sector of R&D funding in LAC comprises the institutions of higher education. The
latter’s share has increased consistently, albeit slowly, from 5.7% in 1990 to 8.3% in 2000.
Given that government spending on higher education tended to decrease or level off over the
last decade, the increase in R&D spending by this sector is probably due to the aggregation of
new, mostly private, institutions to the system. Nevertheless, it is difficult to work out more
precisely what universities contribute to R&D. In some countries (Brazil, Colombia, Mexico)
government permission for the creation of new universities is dependent on their funding of
research activities. However, the creation of new universities, does not sufficiently explain the
comparatively high share of R&D expenditure by the higher education sector in Uruguay
(50.3% in 1995 decreasing to about 40% in 2000) -the sole public university (Universidad de la
Republica) has an internal research council (Commission for Scientific Research – CSIC) with a
specific budget and is the most important source of funding for academic research in the country
(Davyt & Velho, 1999).

Other sponsors of R&D activities in LAC include non-profit organisations and foreign sources.
Altogether these sources represented no more than 3% of R&D expenditures for LAC as a
whole during the 90’s. However, both sectors are quite significant for poorer LAC countries
such as Bolivia and El Salvador (around 24% for both sectors together), Paraguay (40% from
foreign sources) and even for a few middle-income countries like Panama and Uruguay (Table
2), for reasons worth exploring.

First of all it should be said that the distinction between R&D funding coming from non-profit
organisations and foreign sources is somewhat blurred. Non-profit organisations dedicated to
R&D do not have their own funds in LAC and can hardly raise funds domestically, depending,
therefore, on foreign resources. When foreign money is granted directly to NGOs without local
government mediation, it is classified as R&D expenditure by ‘non-profit organisation’. This
means that a country with a high share of non-profit organisation expenditure actually gets
considerable foreign support for research. This is the case in Bolivia, where non-profit
organisations play a very important part in R&D expenditure, but receive all their support from
foreign sources. Nicaragua (which is not featured in Table 2) shows a similar picture and,

14
together with Bolivia, are the largest development assistance recipients in LAC (WDI, 2002).
Although, as a norm, only a fraction of official development assistance (from 5 to 10%) is spent
on R&D activities 7 , studies carried out in Nicaragua and Bolivia reveal that not only NGOs but
also universities in both countries are heavily dependent on international co-operation and
foreign aid to carry out R&D (Velho, 2000). There is, however, a decreasing trend on the share
of non-profit and foreign expenditure even for the poorer countries of LAC (Table 2, see
especially Bolivia). This reflects a general decrease in development assistance for research
capacity building in the South in general (the so-called donors fatigue), as well as donors
preference to support programmes in Africa, to the detriment of LAC.

Foreign funding for R&D in LAC comes from different sources, most commonly from multi-
and bilateral development agencies and philanthropic organisations. In particular, the World
Bank and the Inter-American Development Bank (IDB) provided substantial resources to
support S&T activities in the region, such as multi-year loans amounting to about US$600
million for Argentina, Brazil and Mexico during the 90’s (Hansen et al, 2002). Since the
intensity and amount of government spending in R&D differs considerably across these
countries, the impact of such loans varies accordingly, being practically negligible in terms of
R&D total expenditures in Brazil but rising to almost 7% in Mexico (Table 2). Two other
countries call attention here, namely Uruguay and Panama. In the former the share of foreign
funding was 12.5% in 1995 and dropped to about 5% in 2000. This reflects the support granted
to the Programme for the Development of Basic Sciences -PEDECIBA- by various international
agencies with a special role of the United Nations Development Programme (UNDP – Barreiro,
1997) 8 . In the case of Panama, foreign funding for R&D (over US$18 million in 2000) is due to
contributions by the Smithsonian Institute of Tropical Research, which is located in that
country, but is part of the Smithsonian Institution of Washington, DC, US, (Ricyt, 2002). This
institution alone spent in R&D almost twice as much as the other three sectors together.

2.1.3 Expenditure by socio-economic objective

Although a significant number of LAC countries provide information concerning their R&D
expenditures by socio-economic objective, as presented in Table 3, a close inspection of such
figures reveals that they are not readily comparable. The low proportion of funding going to
industrial development technology in Brazil (1.5%), for example, as compared with other less
industrialised countries such as Peru (29%), raises doubts on the reliability of the data. The
reason for these variations is that in the Brazilian case, the unpacking of information concerning
expenditure by socio-economic objective is provided on the basis of budget allocated to
different ministries. So it is that the highest proportion of expenditure appears in the ’general
promotion of knowledge’ line, which reflects expenditures made by the Ministry of Science and

15
Technology, irrespective of which objective or programme the funds are channeled to. This
includes all programmes funded by the Research Council (CNPq), as well as by FINEP and
MCT itself that target specific objectives (such as Sectoral Funds, incubators and the software
industry).

In view of such discrepancies, there is little that can be derived from the information in Table 3
unless one has specific knowledge of the methodological procedures used by each country to
produce the data. It is possible to infer, for example, that countries that have a significant
proportion of expenditure in ‘general promotion of knowledge’, are likely to be those that
traditionally support Research Councils and research universities. This is typically the case for
Brazil, Chile and Uruguay. However, the absence of Mexico from this list is noteworthy as this
country also has an operative and well-funded Research Council. This must indicate that
Mexico might unpack the information using other methodology.

It is most unfortunate that this kind of information is not of much use because it could provide
an interesting analysis by confronting supply (in the form of R&D funding) to socio-economic
objectives with specific country’s development needs or demands.

2.2 Personnel

2.2.1 Availability and sectors of activity

There is wide agreement among analysts that LAC countries are short of researchers capable of
performing high quality research. In 2000, there were around 240,000 researchers in the region,
which is equivalent to 0.89 researchers for every 1,000 economically active people. Moreover,
not all are full-time researchers, so that when controlling for time dedicated to research the
figure comes down to the equivalent of 147,019 full-time researchers (Ricyt, 2002).
Corresponding rates for OECD countries are 10 (Spain) to 15 (US) times higher. There are,
however, significant differences among LAC countries: Argentina, Costa Rica and Chile are the
best performers in this respect (2.5; 1.5 and 1.2 researchers per 1,000 residents, respectively),
while Mexico and Brazil do much worse than expected (about LAC average – see Table 4),
given their level of R&D expenditure and their emphasis (particularly Brazil) on post-graduate
training programmes.

This shortage in the science and engineering workforce, however, is better analysed in light of
the size of a country’s economy. It is to be expected that poorer countries simply lack the money
to train researchers either domestically or abroad. Actually, the calculation of the number of
scientists and engineers per 10,000 workers as a function of per capita GDP for some LAC
countries revealed that “most Latin American countries do not have a low stock or flow of

16
scientists and engineers for their income level”(Ferranti et al, 2003, p. 53, emphasis added). In
this exercise, LAC countries such as Brazil, Costa Rica, Uruguay, Bolivia, Nicaragua and Peru
have a scientific workforce commensurate with the sizes of their population and economy that
may indicate (certainly for countries like Brazil, Costa Rica and Uruguay) government
commitment to the training of researchers.

In LAC few scientists and engineers are employed in the business enterprise sector (Table 2).
Two out of 3 researchers are located in universities and only 11% operate in companies (some
of which are state-owned) and this has not changed much since 1995. This compares with
almost 40% in leading OECD countries with a tradition of close firm-university collaboration in
research (OECD Basic S&T Indicators, 2001). Brazil is the only LAC country showing a quite
significant proportion of researchers in the enterprises in 2000. The amazing increase in this
share since 1995, however, (from 7.8% to 31.1% in just five years) suggests that there are
problems with the data, as some authors have already pointed out (Velho & Saenz, 2002) 9 .
Costa Rica’s case is also noteworthy, having reported almost one-fourth of researchers working
in enterprises. For all other LAC countries, the figure is never above 12% and is more often than
not quite insignificant. As a consequence of the low share of business spending in R&D and of
researchers in enterprises, LAC puts much less effort into development research (which is more
often carried out by the private sector). Thus, for all LAC countries that classify their R&D
expenditures into basic research, applied research, or development categories, the share of the
latter has been below 29% (and often much less than that) during the 90s, and shows no
tendency to increase (see Ricyt 2002). Comparatively, the US spends around 62% of its R&D
budget in development and the equivalent figure for South Korea is 61% (NSB 2002, 4-49).

In short, the large majority of LAC researchers work in the public sector, most notably in
universities. A further problem here is the fact that about 60% of the faculty members work
part-time and that they lack formal research qualifications: less than 6% have a PhD degree and
the share with a Masters is less than 26% (García Guadilla, 1998). LAC averages, once again,
hide considerable differences among countries: over 50% of researchers in Brazil, 35% in
Trinidad y Tobago, and about one-fourth in Argentina have a PhD. In Mexico, surprisingly,
64% of researchers have not undertaken any post-graduate training and only 6% has a PhD
degree. Masters degree holders are much more generalised in the region, as the case of El
Salvador indicates (77% - Table 5). Despite these problems it has been estimated that there was
a 70% increase of Masters graduates in the LAC region through the decade. An even faster
growth was experienced in the number of doctorates, which practically doubled in the same
period, albeit from a very small basis (Urzúa, 2002).

An important aspect in analysing research personnel has to do with their distribution in the
various fields of knowledge. Figures in Table 4 show that over 30% of researchers in LAC work

17
in the social sciences and humanities. It has been forcefully argued that an additional problem
with the research workforce in LAC is that there are too many social scientists as compared to
natural scientists and engineers (Schwartzman, 2001). Since there is no established parameter
for what that proportion should be, a good practice is to compare these levels to those in the
technologically advanced countries. The proportion of researchers in the social sciences in the
academic institutions in the US is about 20%, but as part of the whole employed science and
engineering workforce, it is 28% (NSB, 2002: table 5-30). This is very similar to the LAC share
of social scientists. On a country-by-country basis, however, it is striking that about 60% of the
researchers in Mexico and El Salvador are concentrated in the social sciences and humanities.
Of the LAC countries featured in Table 4, those with the lowest share of social scientists are
Bolivia and Ecuador (about 10%). These data are actually confirmed by case studies in Bolivia
as well as in Nicaragua (for which there is no available quantitative data). Research in the social
sciences in Bolivia is believed to be “poor, personalised, isolated, discursive and lacking of
systematic approach and empirical basis” (Souza Paula et al, 2000). For this reason, and
assuming that a critical mass of social scientists is essential to development, some bilateral co-
operation programmes have decided to support initiatives in this direction in Bolivia. In the
same vein, university officials and researchers in Nicaragua have insisted in negotiations with
donors that support is needed to strengthen research capacity in the social sciences in the
country, but have not succeeded so far 10 .

Two countries in LAC, namely Uruguay and Chile have strong participation of natural scientists
among researchers (around 30% - Table 4). For Uruguay this is probably the outcome of strong
support granted to the natural sciences in the framework of the Programme for the Support of
Basic Sciences (PEDECIBA – see note 8). Chile has a strong tradition in the natural sciences
and mathematics fields, in which the best graduate programmes in the country are concentrated
(Krauskopf, 2000). Brazil has a very similar pattern of distribution of researchers in scientific
fields to that of the US, with the exception of the medical sciences where the US has a
significantly higher share (20% in Brazil and 34% in the US). However, a recent estimate of
researchers in the medical-related fields in Brazil argues that their share in the R&D workforce
varies from 25 to 33%, depending on the definition used (Guimarães, 2003). Finally, the share
of researchers in engineering in LAC is also comparable to the one in the US – 17% and 18%,
respectively.

2.2.2 Socio-economic and gender participation

An important dimension of personnel in S&T and in R&D is its composition in terms of ethnic
origin, socio-economic origin and gender. Reliable information concerning the first two aspects
is not readily available for LAC countries, neither is it included in the S&T indicators series

18
traditionally compiled by individual countries or by Ricyt. This is important information for
policy purposes, which may indicate how the composition of the S&T workforce matches that
of society. However, it is possible to shed some light into this aspect by looking at basic
information on education in LAC.

It is widely acknowledged that education is extremely unequally distributed in LAC. The

composition of education in the region is still skewed towards primary education. Thus, 40-50%
of the population reaches primary school but never surpass it, while only 20% make it to the
secondary level. “[This] trend is broken in Chile and Mexico where the proportion of the
population that attained secondary education as the highest level increased to 30-40%. Finally,
only 10% of the Latin Americans attain some extent of tertiary education. Costa Rica and
Argentina are among the regional leaders, where approximately 20% of the population reach the
level of tertiary education” (Hansen et al, 2002:33). In addition, the share of university students
from the lowest third of the population in terms of income distribution is 6% in Peru, 11% in
Chile and and 18% in Uruguay (García Guadilla, 1998). Needless to say these figures reveal a
lot about the extent of social inequity in the region. An important consequence of this situation
is the deleterious effect it has on the countries’ R&D capabilities: a less educated workforce
hampers the process of technological innovation at the firm level. But most important of all,
education inequalities in LAC restrict the population pool from where researchers are recruited;
the result being the automatic exclusion of people from lower income and ethnic groups with
potential vocation for research and the constitution of a socially homogenous ‘class’ of
researchers. Because in LAC the poorer population includes the majority of black11 and native
groups,12 the their cultural perspectives and indigenous knowledge do not get a change to reach
the universities and other locations where formal R&D takes place. This diversity of
perspectives, cosmologies and knowledge could substantially enrich the process through which
formal R&D in LAC looks for solutions to local problems.

On the positive side, there have been significant advances of women relative to men in terms of
education. Gender disparities in education have almost closed in all the countries, and in some
cases women’s educational achievements are higher than those of men. Primary education has
similar enrolment ratios of girls and boys, except for countries with a significant indigenous
population, such as Bolivia, Ecuador and Guatemala, where the ratio is 90% or lower. Although
there is an overall decrease in the enrolment ratio in secondary education, on average girls
continue their education further than boys, as the latter tend to drop out school more often than
girls to take on productive jobs. This is particularly true for Argentina, Brazil, Cuba, Dominican
Republic, El Salvador, Honduras, Jamaica, Nicaragua, Trinidad & Tobago, Uruguay and
Venezuela. Only Barbados, Bolivia and Peru have gender ratios of 90% or lower.13

19
At undergraduate level gender participation is relatively balanced on the whole, even perhaps
more favourable to women in some countries such as Argentina, Uruguay and Brazil. Although
there are some technical and scientific disciplines, such as agricultural sciences or engineering,
where men prevail, women form a majority in chemistry and biological sciences. The ratio in
social sciences is about 50%, and in humanities it exceeds 60%. This gender ratio is maintained
through the masters (50% in the case of Brazil, 41% in Mexico) and doctoral degrees (46% in
Brazil, about 50% in Argentina). These differences widen, however, where scholarships systems
are in place. In Mexico, only 3 out of 10 scholarships are awarded to women,14 while in
Uruguay, women account for only 35% of postgraduate scholarships. 15 In Ecuador the ratio is
27%,16 while in Brazil only 38% of Fulbright award winners are female 17 (calculated from
information available on the Fulbright website).

The participation of female researchers in the S&T systems in most LAC countries ranges from
30 to 50% of the total staff in 2000 (see Table 6). This is well above the norm in more
developed regions such as the EU, where on average more than 60% of researchers in
government institutions and 75% in higher education are males.18 Similarly in the US only one
researcher in five is a female. One possible explanation for the higher participation of women in
research activities in LAC compared to that in more developed countries could be the existence
of a family and social support network enabling the combination of career and family
responsibilities. Another explanation could be the differences in pay and working conditions
between public and private sector, since most research is carried out at universities or public
institutions. The relatively low pay in the public sector may be a disincentive for males, who are
culturally the income providers for the family (Velho & Leon, 1998). On the other hand, the
flexibility in working conditions in public institutions may act as an extra incentive for women,
who still have to balance their culturally determined reproductive and productive roles19 . In
terms of scientific disciplines, the distribution of researchers is similar to that observed in
tertiary education. Women dominate in most humanities, there is a balance in social sciences
and in some ‘hard’ sciences such as biology, chemistry and medicine, while researchers in
exact, agricultural or engineering sciences are mostly male (Sedeño, 2001).

Entry requirements into research institutions do not seem to discriminate along gender lines,
with the exception of Mexico, where some de facto entry barriers seem to exist. However, once
inside, in terms of hierarchy and career progress, women researchers in Latin America face the
same constraints and barriers as their colleagues elsewhere. Hierarchy implies decision-making
power, which is important for the selection of research topics and resource allocation, and it is
here where the position of women deteriorates. In Argentina, of all women researchers at
CONICET, 72% are concentrated in the lower grades compared with 51% of all men. Only
0.4% of women researchers reach the highest level, compared with 4.5% of men, which leaves a

20
proportion of nearly 10 to 1 at the top. The same occurs in Uruguay, where men occupy 80% of
management positions in CONACYT and a negative relationship has been established between
the size of the project and female directors -33% when projects are below $30,000, 19% when
above $150,000 (Kochen et al, 2001). In Brazil too, the gender balance exists only in the early
stages (50% of staff aged below 24 years are female) of the research career; thereafter it
deteriorates to around 40% (only 30% of 60-plussers are female).20 In Ecuador, women have led
only 13% of the university research projects (1983-1996) and 9% of FUNDACYT-BID research
projects (1994-1996). One exception seems to be Cuba, where women account for 58% of
researchers and 45% of research directors at the university (Fernández, 2001).

The data above suggest that increasing the number of women in the work force is not sufficient
to level the playing field. More studies are needed to explore the determinants of this dominance
by male researchers. In the same line, it is to be expected that the long needed removal of entry
barrier for ethnic groups to the world of science will have to be accompanied by measures that
will guarantee their access to the control of resources and rewards.

2.3 The output of R&D

2.3.1 Publications
The output of R&D can take many forms; from the publication of articles in specialised, peer-
reviewed ‘international’ journals (the so-called mainstream publication channels) to research
reports in local languages contracted out by governments and donors and presentations in
various types of meetings (the ‘grey’ literature) to patents, one of the few indicators measuring
successful innovative attainment. How to collect systematic information from such a broad
spectrum of outputs and how to ascertain their ‘scientific quality’ or ‘social and economic
relevance’ has long been a matter of discussion and debate.

Concerning the output of research, there is no publication database to date that is reliable,
systematically updated and that covers all fields and LAC countries. All existing databases
suffer from some form of conceptual or methodological limitation: some cover only
publications related to specific fields (such as Medline for the medical-related fields and
Chemical Abstracts for research using chemistry-based methodology, but include both the
‘mainstream’ and the ‘grey’ literatures); others are intended to cover only specific countries and
institutions (such as the one maintained by the Brazilian Graduate Agency –CAPES- or by the
Chilean CONICYT); and others attempt to develop a comprehensive LAC coverage but lack
the financial and human resources needed to accomplish the project ( a good example here is
Periodica, maintained by the Autonomous University of Mexico -UNAM-, which is admittedly
skewed in its coverage in favour of Mexican publications, neglects a large part of South

21
American output, and does not provide information about countries outside LAC, therefore
making comparability across regions unfeasible (Narvaéz-Berthelemot et al, 1999). Finally, the
most commonly used databases produced by the Institute for Scientific Information (ISI), of
which the Science Citation Index (SCI) is the best known, have long been pointed out as
discriminating against non-English publications coming from developing countries and from
more applied, locally-relevant and interdisciplinary research21 .

In view of the above, when Ricyt set out to publish output indicators for LAC, it organised a
number of meetings with experts and policy-makers and a decision was made to present the
research output through different databases. The rationale was that a number of carefully
targeted databases would be more effective in addressing specific research areas and enable the
analyst to choose according to his or her research objectives. Thus, 10 different databases report
LAC research output in the Science and Technology Indicators series produced by Ricyt. Only
three of these (SCI, INSPEC and Pascal) intend to cover all scientific fields and all countries,
thus allowing us to estimate LAC contribution to world science, which varies from 2.9% of a
total output of 988,156 publications registered in the SCI in 2000; 2.1% of 335,089 publications
compiled by INSPEC in the same year; to 2.6% of 511,617 items registered by Pascal also in
2000. In all 3 databases Brazil is responsible for about 45% of the publications originating in
LAC in 2000 and the publication ranking of other LAC countries is also the same in the three
sources. What these figures show is that, by whatever means publication data for different
countries is obtained (the 3 databases use different collection criteria), the aggregated
22
contribution of LAC to world science is very modest, between 2 and 3%. This
notwithstanding, looking from 1990 onwards, it is also clear that this contribution has improved
significantly and consistently, as shown in Table 7.

Given the convergence of the 3 databases, it seems reasonable to use the SCI for comparative
analysis, as presented in Table 8. On a population basis, the LAC region produces 5.6 articles
per 100,000 inhabitants, while the equivalent figure for the US is 116, 64 for Spain and in
Portugal it is 30 per 100,000 (Ricyt, 2002). Brazil, the biggest producer in LAC, had a ratio of
only 7.6 publications per 100,000 people in 2000, although this was a considerable
improvement from the beginning of the 90s when the ratio of publications per 100,000 people in
Brazil was 2.5. Countries like Argentina, Chile and Costa Rica perform better than Brazil when
ratios of publication/population are considered. Overall, both in terms of number of publications
and publication per population, LAC performance shows a consistent upward trend during the
90s, albeit starting from low levels, as seen in Table 8. This increase in output is considerably
dimmed, however, when compared with the performance of NICs such as South Korea and
Taiwan. In 1986 Brazil had 1,777 publications registered in the SCI and South Korea only 516;
in 1999, the figures for Brazil jumped to 5,655 while the corresponding output of South Korea

22
was 6,675 publications. Similarly, Argentina started off in 1986 with 1,459 publications, which
rose to 2,361 articles in 1999, whereas the corresponding figures for Taiwan are 904 and 5,655
for the same period, a more than six fold rise (NSB 2002, Appendix Table 5-43). This
difference between LAC and South Asian NICs in terms of increment of scientific output
clearly reveals their divergence in policies concerning R&D support, particularly the intensity of
investment in R&D. As a result South Korea and Taiwan jointly now contribute to world
science more than all LAC countries combined.

The SCI database also provides unique information concerning the institutional affiliation
(including country) of all authors of indexed articles. This allows for policy relevant analysis
related to a number of aspects, such as the relative prominence of institutions (and sectors) in
the country and in specific fields of science and - most significantly given our interest in
identifying links among organisations - the identification of research partnerships, including
inter-institutional, inter-sectoral, intra-regional and international. This requires, however, a
further processing of the information into relevant indicators and, although this is traditional
practice in the Science and Engineering Indicators Series of the US National Science
Foundation, this is not the case for Ricyt. Therefore, whatever we know about institutional
origin of LAC papers is based on specific studies, mostly for particular countries and fields of
knowledge. For example, in the case of Brazil, it has been found that over 80% of the SCI-based
articles in 1997 originated in only 15 public universities (out of 150 universities and 22
government research institutions in the country - Indicadores de C&T em SP 2001, chap.6). The
remaining articles are scattered among several institutions, none of which are private firms.
Although it is well known that researchers working in firms are not motivated by publication,
data for the US show that the private sector there produced 21% of the SCI articles in 1999
(NSB 2002, p.5-40).

Concerning collaboration across sectors and nations, the standard indicator is the number of
articles co-authored by researchers coming from different institutions or countries. Cross-
sectoral collaboration is viewed as a vehicle for making research results more useful to practical
application. Consequently, the ratio of papers involving authors from say, the academic and the
industrial sector indicates the existence of linkages between them. To some measure, this ratio
also reflects the impact of government schemes to foster collaboration. Again, such an indicator
has yet to be constructed for LAC, and indeed should, as no relevant information was found in a
review of the literature. By contrast, corresponding information for the US is readily available
and shows that collaboration across institutions is extensive in that country and is closely
associated with government efforts to stimulate this trend. Thus, in 1999, over 77% of the
papers coming from the various sectors (federal and other government, private for profit, private
non-profit) were co-authored with researchers from other sectors. The exception was the

23
academic sector, which had 37% of papers published in collaboration with colleagues in other
sectors (NSB 2002, p. 5-45). No doubt this trend is highly desirable for a well functioning
innovation system.

International collaboration in research is also becoming increasingly significant as it provides

intellectual cross-fertilisation and ready access to work done elsewhere.23 An empirical exercise
was performed specifically for this study to identify the number of co-authored articles
involving at least one researcher with institutional affiliation in LAC and the other in any other
country, for the years 1993 to 1999 and the results are presented in Table 9. For LAC as a
whole, about 30% of the articles in the ISI database are internationally co-authored and this
share has not changed during the period analysed (although this is a 10 point decrease from
1990, according to Narvaez-Berthelemot et al, 1999). For most LAC countries, with the
exception of Cuba, US and Canada are the main international partners. However, for larger
countries like Brazil, Mexico, Chile and Argentina, collaboration with the US is much less
important (around 15% in the 4 years covered by this study) than for the smaller countries of
Central America, like Guatemala (42%), Honduras (53%) and Nicaragua (44%). A group of
countries – Costa Rica, Cuba, Bolivia and Colombia – have notably increased their
collaboration with Europe, which is probably an outcome of the support granted to these LAC
countries by the EC (for example, in the framework of the INCO-DEV programme) and of
bilateral co-operation in the framework of development co-operation programmes.

The most significant finding, however, is the increase in LAC intra-regional collaboration.
Brazil, for example, has maintained the same level of collaboration with US and European
partners during the 90’s, but doubled its collaboration with other LAC countries. This may well
reflect links among researchers established during graduate training in Brazil, as the
strengthening of such programmes in the country is attracting a significant number of Latin
American students (Velloso & Velho, 2001). Uruguay and Argentina show similar trends
concerning intra-regional collaboration, which may be the result of policies implemented since
the creation of the Mercosur (Narvaez-Berthelemot et al, 1999). In sum, the exercise revealed
that LAC countries are not part of the rocketing increase in international collaboration between
the US and European countries among themselves. The improvement in South-South
collaboration in LAC is a good sign, but it certainly does not compensate for the stagnation in
the North-South co-operation.

An important aspect of a country’s research output is its “product mix”, that is, the scientific
fields from which such publications originate. This reflects both historical trends of the
scientific enterprise in each country as well as government policies to strengthen specific
research areas to achieve particular objectives. The Ricyt database derived from the SCI,
however, does not provide this kind of information. For this it is necessary to resort to the

24
Science and Engineering Indicators produced by the US National Science Foundation (NSB
2002). This source is particularly useful because it presents comparable information for all
countries and regions, from which the relevant figures are given in Table 10. The product mix of
LAC as a whole is similar to that of the US: over one fourth of the publications are related to
clinical medicine, followed by physics and math, and further down by biomedical research.
However, looking at particular countries it is clear that some of them put much higher emphasis
than is the LAC norm in clinical medicine (over 40%) – all Central American and Caribbean
countries follow this pattern. It is also the case in these countries that adding the publications in
biology and agriculture to those in clinical medicine raises their total output to around two-
thirds. The reason for this pattern is well known: both clinical medicine, biology and agriculture
are connected with local conditions and natural resources and require local research. Thus, most
of the tropical developing countries have created research institutes in these fields (see Table
14).

LAC countries that have reached higher levels of scientific production like Brazil, Argentina,
Mexico and Chile (at least over 0.1% of world output, see Table 10) have a significant
proportion of articles in physics and math (23 to 25%), even more so than the US. This
production closely resembles the distribution of researchers by fields, as shown in table 4, and
indicates that these countries were able to create a diversified critical mass of productive
researchers. It is noteworthy, however, from Table 10 that LAC countries have low publication
ratios in engineering and technology (5.4%). While this figure is not much lower than that of the
US (6.0%) and Western Europe (5.8%), it is considerably so when compared to ratios for the
Asian NICs like South Korea (20.9%) and Taiwan (19.4%). Since the 1980s the latter countries
concentrated their publication in engineering and technology (for both countries, the figure was
over 20% in 1986) - a reflection of their industrialisation strategy. The portfolios of these two
NICs underwent sizeable shifts between 1986 to 1999, for instance, when increases in physics
and mathematics exceeded 10 points, which also reveals their intention to perform better in the
more “basic” sciences which give support to engineering and technology (NSB 2002, p.5-43). It
seems reasonable to assume that if LAC countries intend to improve their capacity to innovate
they would benefit from following a similar strategy: concentrate resources to train researchers
and fund projects in scientific fields related to the technology areas they select as important
(Bernardes & Albuquerque, 2003). This would be a notable change from the current practices
whereby research fields tend to grow on the basis of the increasing scientific competence of its
practitioners, as judged by their own peers.

Even more difficult and controversial than estimating scientific output is to make any statement
about its scientific quality and social relevance. About the latter it is widely accepted that it can
only be assessed through case studies and with the use of qualitative information. As far as

25
scientific quality is concerned, the conventional approach is to use the proxy measure known as
the ‘impact of publication’. This measure is derived from the SCI database and relates to the
number of times an article has been cited (citation analysis), on the assumption that this reflects
how influential the research results have been in advancing the state of knowledge. Table 11
presents citations data for LAC publications in 1990 and 1999 and what this represents in terms
of the distribution of total citations received by the articles indexed in the ISI database. The
figures reveal that LAC received less than 1% of world-wide citations, which is less than half
the region’s contribution (2.3%) to world publication. This suggests a low impact of papers
published by LAC and, in this specific indicator, no LAC country performs well. It is worth
mentioning, however, that Latin America was the only developing region whose world share of
cited literature increased in the last decade: from 0.58% in 1990 to about 1% in 1999 (Tale 11).
It is significant, therefore, that in spite of their success in increasing the number of scientific
publications, Asian NICs have not increased their share of the world citations, remaining at
0.6% all through the 90s, which may indicate that their international ties have been concentrated
within the US and their own regions. The good news for LAC is that in addition to increasing
their ratio in world citations, self-citation indexes in these countries fell notably. Taken together,
these measures indicate that LAC has improved its access to international science and that its
research effort has enhanced its influence and visibility worldwide (NSB 2002, p. 5-51).

2.3.2 Patents
Patents are one of the few indicators measuring successful innovative attainment, despite the
many limitations noted in the literature (for example, Patel & Pavitt, 1995). In terms of total
number of patent applications, LAC R&D output more than doubled from 1990 to 2000 (Table
12). This, however, does not indicate increased innovative activity within the region as the
resident patent application accounted for the smallest part of patents. In other words, non-
resident patent applications almost tripled during this period, while resident patent applications
increased by less than 25%. The individual position of specific countries is even more
unfavourable: Cuba, Mexico and Uruguay, for example, experienced a notable decrease in the
number of resident patent applications in the period. Only Chile seems to have achieved a
significant increase in this indicator of about 60% (the corresponding figures for Brazil and
Argentina were 25% and 11%, respectively).

Expressed in terms of population, the same picture remains: Brazil is the country with highest
ratio of resident patent applications/100,000 population (5.3) in 2000, followed by Argentina
(3.0) and Chile (2.8), which is probably a result of the higher investment they make in
R&D/GDP (see Table 1). Cuba, however, despite significant investment in R&D/GDP and
number of researchers, has suffered a drastic reduction in its GDP which may account for its

26
decreasing resident patent application output. The performance of Mexico in this indicator is
surprisingly weak, given that the country has a strong research tradition and stands among the
highest R&D investors in the region. As a consequence of this decrease, Mexico’s dependency
rate (number of non-resident patent applications as a ratio to the number of resident
applications) rose significantly during the 90s: from 6.9 in 1990 to 29.3 in 2000, an almost five
fold rise, while the average increase in the dependency rate for LAC was two fold (Ricyt 2002).
This increase for LAC in general, but specially for Mexico is “largely a consequence of the
liberal reforms carried out since the 1980s, which have facilitated a massive transfer of
technology to the region” (Hansen et al, 2002: 68). The rate at which US companies established
assembling plants in Mexico probably explains this country’s dependency figures.

The innovative attainment of countries in terms of patents can also be analysed using data from
the US Patent and Trade Office (USPTO). This is a more reliable source for the purposes of
making international comparisons as it standardises patenting regulations and practices. From
such a database it is possible to derive the number of patents granted in the US to residents from
any other country’s nationals. Figures for LAC are presented in Table 12 and show that, taken
together LAC countries doubled the number of patents granted by the USPTO from 1990 to
2000, when the figure of 365 patents was reached. Brazil, again, was the highest performer (113
patents in 2000), this time closely followed by Mexico (100). This is a growth rate higher than
that of the US itself (40% in the period but from a starting figure of over 50,000 in 1990 – NSB
2002, p.6-21). However, LAC growth rate of patents is completely overshadowed by that of the
Asian NICs: South Korea, for example, raised the number of patents 12 fold - from 290 in 1990
to 3472 in 2000. Taiwan also made an impressive jump from 861 to 5,806 patents in the US
24
during the 90s (a more than six-fold increase). Since 1998, Taiwan and South Korea have
become, respectively, the fifth and sixth most active foreigner inventors in the US, surpassing
Canada (ibid).

The extraordinary growth of both publication and patents for South Korea and Taiwan in the
last 15 years or so raises the question of a likely relation between these two activities. A recent
study has addressed this question by searching for a correlation between data on publication and
patents for these two countries as well as for Brazil. The findings reveal not only that there is a
high correlation between the growth of scientific and technological output but that the
concentration of scientific production, as the Asian NICs did, in disciplines with high relevance
to industrial technology (engineering fields, material sciences, physics) explains a significant
part of the differences in patent growth and impact in economic development between LAC
countries and Asian NICs (Rapini, 2000 as quoted in Bernardes & Albuquerque, 2003). The
difference can further be explained in terms of the much higher levels of investment in R&D by
Asian NICs and by additional factors such as skills of the workforce and capabilities in policy

27
design and implementation. Finally, another study also taking this issue for a much larger
sample of countries, suggests that there may exist a threshold level in scientific production of
around 150 papers per million inhabitants, beyond which the efficiency in the use of the
scientific output by the technological sector increases. No LAC country has yet reached such
threshold: Brazil is in the range of 70 papers per million inhabitants and Argentina in a better
position of about 135, while the Asian NICs have surpassed it since the beginning of the 90s. Of
course, reaching such a threshold is not a sufficient condition to bring about interaction between
the scientific and technological sectors, but below such threshold there can be only weak
interactions, if at all (ibid).

*
From the basic indicators presented and discussed above, a picture of the S&T (more
specifically, R&D) capacity in LAC begins to take shape. First, for the region as a whole,
national governments seem to understand the importance of R&D investment for economic
growth. This is demonstrated by the efforts made by governments in funding the bulk of R&D,
and in increasing R&D spending despite the stagnant economy during the 90s. Nonetheless, the
intensity of expenditure is much lower than that of the technologically advanced countries and
Asian NICs - the bulk of R&D in the latter being paid for by private enterprises. The LAC
pattern of a high share of public sector spending in R&D has had a number of consequences:
most researchers work in the public sector (mostly universities) and they produce almost the
totality of scientific articles. This output tend to be concentrated in fields of knowledge that
reflect historical patterns of scientific development in LAC with little contribution to fields
related to industrial technology. Also, given that private investment in R&D remains small,
patenting activity is also low. Under these circumstances, links between the scientific (mostly
public) sector and the technology-users (mostly private) sector are expected to be weak. This is
indeed confirmed by the indicators - there is minimal funding of universities by enterprises;
there is a lack of collaboration between academic researchers and those in industry in the
authorship of papers; and there is, little connection between publications and patent activities.
Moreover, as the data on the socio-economic background of researchers testify, researchers are
drawn from a very narrow stratum of the LAC social structure. To this extent, it is reasonable to
say that the scientific enterprise has weak links with the larger society.

A word on the availability, reliability and breakdown of data is due here. On the one hand, the
effort made by Ricyt to compile the information, standardise concepts and data gathering
methodology is commendable. However, policy-makers in each country need much more
targeted, and policy relevant data, as pointed out in the analysis above. This requires the
building of capacity at the government level in this kind of activity.

Lets us now turn our attention to the organisations involved in S&T and R&D activities

28
3.0. S&T AND R&D ORGANISATIONS IN LAC

One way of looking at national systems of innovation, particularly for analytical purposes, is to
identify the existing set of functioning organisations, institutions and policies relevant for
innovative activities, as well as the interactions (or knowledge flows) among them. This section
attempts to do just that. Obviously this is not a complete picture as the amount and quality of
information available varies considerably for each country. For this reason, emphasis is placed
on the research supply side, which includes government bodies allocating research resource, and
universities and research institutions performing research. Much less information is provided
about the research demand side, including those organisations or ‘social actors’ with the
capacity, ability and willingness to utilise the outputs of research. Also, evidence about the
existence of linkages or interaction between research “supply and demand” is mostly taken from
common sense statements or, at best, is based on case studies, which can seldom be generalised.
This notwithstanding, the picture below provides the contours of the existence and operation of
some essential elements of innovation systems in LAC.

3.1 Government: S&T Policy Formulation and Resource Allocation Institutions

The idea of an S&T policy apparatus at the government level is not new in LAC. Following the
general trend after WW II, and under the auspices and coordination of UNESCO and the
Organisation of American States (OAS), S&T policy-making bodies were founded early in the
1950s in Brazil and Mexico, followed by Argentina later in the decade. During the 60s and 70s
a significant number of countries had already established some form of systematic policy
thinking on S&T policy matters (see Table 13). At the stage of their creation, most of these
bodies were structured as Research Councils, which were charged with both articulating S&T
policy - normally through a process of negotiation with ministries of other sectors - as well as
implementing the policy through funding and incentive mechanisms.25 The mere creation of
such institutions, however, did not make them operational or dynamic. In a few countries, S&T
plans as well as the so-called Fondos de C&T (S&T funds) exists on paper only. This is the case
of Bolivia - where at least three unsuccessful attempts were made to actually institutionalise the
research council - as well as Paraguay and Nicaragua.26 . In the countries where the policy and
funding mechanisms do work (Brazil, Uruguay, Argentina, Chile, Colombia, Venezuela, Costa
Rica and Mexico), they have been under considerable control of the local scientific community.
Assuming a linear model of innovation, efforts were concentrated on the scientific end - the

29
supply side - the rationale being that a high quality critical mass of researchers, well equipped
labs and strong universities would result in "good science" which, sooner or later, would find its
application in technological development.27 Thus, in LAC, identifying research priorities for
resource allocation has traditionally been a task left to the research community and not
negotiated with potential users. As such, research themes tend to be selected more on the basis
of their scientific importance, taking the lead from international science, than on the basis of
local needs.

It is fair to say, however, that the association of S&T policy with the objectives of economic and
social development is not a new concept for LAC either. As early as the 1970s, a group of
influential Latin American thinkers, some of them imminent researchers, argued that the way in
which the Research Councils were operated was ‘marginalising’ local science from local needs.
They associated this with the character of the industrialisation model adopted - its reliance on
technology transfer – which did not require local R&D activities but only the accumulation of
specific capabilities to operate technology developed elsewhere. As a consequence, local R&D
was shut out or played a strictly circumscribed role, unable to match the advantages of foreign
technologies. With no demand made on them made by the industrial productive sector, the
scientific institutions were themselves alienated or ‘marginalised’ from production activities
(Herrera, 1975). With this picture in mind, S&T policy analysts in Latin America have tried,
over the years, to break up such a "circular-causal" process. They have advocated a strong role
for government policies in articulating the supply and demand sides of R&D, through a model
that became known as the “Sabato Triangle” of government, universities and firms (Sabato,
1975).

Following the growth of local awareness of the need to overhaul the S&T policy framework,
which was well in tune with international trends, LAC governments, during the 1980s, sought to
implement such changes. There was a consistent trend towards the creation of Ministries of
Science and Technology or special divisions for this under ‘strong’ ministries of Economics and
Planning, to which the research councils became subordinated. A division of labour was thus
envisaged whereby the Ministries of S&T would be responsible for articulating the S&T Policy
and the research councils for implementing it through funding mechanisms. To achieve the
latter, a number of policy schemes were devised by research councils in many countries to foster
linkages between public sector research (particularly universities) and firms, as listed in Table
13. Such schemes include: co-operative projects between public and private sectors, support to
incubators and science parks, and support to research training at the level of the firms.

The success of such schemes is difficult to ascertain. On the one hand, there is inadequate
monitoring and evaluation of the individual schemes, not to mention of the policy as whole, as
many analysts have pointed out (Vonortas, 2002; Velho & Saenz, 2002; Mullin et al, 1999;

30
Dagnino et al, 1996; Bastos & Cooper, 1995). Secondly, the existing evaluation studies of co-
operative research involving LAC universities and firms are mostly case studies, selected on the
basis of their having ‘something to show’ and are therefore not helpful for drawing
generalisations. This notwithstanding, the common perception among analysts (even if hard
evidence is thin) is that the objectives of the policy schemes have not been attained, at least not
in a way that can make a difference in the aggregate (although there are isolated stories of
success). .28 As one analyst has put it “[the region] has not managed to ignite the virtuous cycle
of learning, innovation and economic growth” observed in East Asia (Cardoza, 1997: 377). And
part of the blame is clearly put on the lack of an effective innovation (S&T included) policy
framework.

Determining what ingredients a S&T policy framework in LAC should have in order to play an
effective role in innovation policy is also contentious, as different stakeholders have different
views, and so do policy analysts. That R&D expenditures must increase in LAC is probably the
only consensual idea. Where the money should come from and how it should be allocated is a
totally different matter. The research community, for example, lobbies for more public research
funds as well as for retaining control over how resources are allocated. It is an undeniable fact
that since the public sector is the primary sponsor of R&D in LAC, any reduction in government
spending will have a stronger negative effect. Most LAC countries have suffered such effects as
increased private sector participation in the region (from 20% in 1990 to 36% in 2000) did not
compensate for the decrease in public R&D funding (Ferranti et al, 2003). On the other hand,
despite general decreases (or kinks) in public spending, the academic community has been the
main beneficiary and has kept control over whatever was available. What is mostly
controversial in this aspect is the indication by other stakeholders and also by policy analysts
that an important part of such resources has been spent on salary supplements for researchers. In
Chile and Brazil, over 25% of project costs funded by the research councils are dedicated to
salary supplements (Mullin et al, 1999, www.cnpq.br). Other countries like Mexico and
Venezuela have also created mechanisms to supplement the salaries of researchers.29 In
conclusion, there is considerable concern among analysts that despite the shift in the S&T
policy framework now recognising the importance of interactions and articulating supply and
demand, in practice, there has not been much change in the allocation criteria for public R&D
funds.

3.2 Institutions of Higher Education

LAC universities have a strong public tradition and were created after the classic Humboldt
model and the contemporary research universities in the US and Europe. Although in some

31
countries universities were created early in the colonial period it was only in the middle of the
20th century that research started to be an important part of activities in some universities.
Moreover, during this process, these pacesetters suffered important setbacks due to political
events. The military dictatorships, which took one country after the other from mid 60s to the
end of the 80s - from Brazil to Argentina, Uruguay, Paraguay and Bolivia, to mention just a few
- hit the universities, which were the main focus of intellectual resistance to the military
regimes, particularly hard. In some countries, of which Uruguay is perhaps the best example,
this led to the almost complete dismantling of the research groups, which had been formed with
great effort. The return of democracy has led to almost total political freedom, but there is no
denying that in many respects, including in university and research life, the region is still paying
the toll for the ‘black political period.’ The most evident sign of this legacy is the absence of a
‘negotiating culture’ among social groups with differing political interest in the benefit of all,
not only in government, but also in public universities and research institutes, and the
consequent discontinuity of plans and projects. It is a common feature of LAC countries that
changes in government mean a halt on on-going projects and a new starting phase almost from
scratch. The same takes place at LAC universities, which are extremely politicised and have
their higher administrative posts (rectors, deans, directors and department heads) subject to
election by the university community. This process reproduces, inside the universities, the
political party politics taking place in the larger context with all its benefits and problems. In
most public universities reform-oriented rectors are seldom elected as they are perceived as a
threat to the status quo (World Bank 2002). Therefore, LAC countries have a huge challenge
ahead in consolidating their democratic political institutions at all levels. This is compounded by
the accelerated pace of economic globalisation, which adds more variables to the already
complex situation.

In the past 15 years, many LAC countries have experienced an impressive growth of private
tertiary education institutions. The distribution of such universities is presented for most LAC
countries in Table 14. In the Dominican Republic and El Salvador the share of student
enrolment in private tertiary education rose from about 25% in 1970 to about 70% in 1996, and
for the region as a whole the figure is 40% (García Guadilla, 1998). Some countries also have a
significant non-university tertiary sector (79% of total enrolment in Cuba, 43% in Peru, 38% in
Brazil and 35% in Chile); in others – among them El Salvador, Guatemala, Honduras,
Nicaragua and Panama – the non-university sector accounts for less than 5% (World Bank,
2002).

Most of the new private universities and non-universities education institutions concentrate on
‘chalk and blackboard’ disciplines (law, business, accounting, social work etc) and do not carry
out research. There is no dispute among analysts and policy makers that public universities are

32
the main locus of knowledge production all over the region. A look at the figures in Table 2
reveals that higher education institutions performed around 41% of R&D activities in 1995 with
a slight decline to around 38% in 2000. This seems to be a declining trend confirmed by the
drop in the share of researchers in higher education from 62% in 1995 to 59% in 2000 (also
Table 2). This notwithstanding, academic institutions in LAC produced over 80% of the
publications in 2000, as shown in Table 8.

University researchers in LAC tend to be academically oriented, for a number of reasons. The
most obvious is that, in some countries, they are evaluated and promoted on the basis of their
formal qualification and publication record, which traditionally puts a premium on publications
in mainstream journals.30 Another important reason is that university researchers tend to have
very weak linkages with other segments of society, be it government, the productive sector or
civil society organisations. Under these circumstances, the research agenda is more likely to be
influenced by the international mainstream. Of course there is nothing intrinsically wrong with
doing academic research. Many argue that basic research is the actual function of universities.
When research is disconnected from local context, however, it lacks the close contact with other
social segments, which might contribute to the identification of real research problems and the
defining of basic research needed to solve such problems. In other words, there is need for
socially relevant basic research. This general academic orientation of university research in
LAC has been pointed out since the late 60s (Varsavsky, 1969; Herrera, 1975) and subsequently
confirmed by many empirical studies in different countries (Velho, 1995; Casas, 1997; Arocena
& Sutz, 2001). There are, however, some notable differences among scientific fields. In
agriculture, for example, LAC universities have some significant traditions involving strong
links with applications in plant breeding and pests and disease control. However, even in such
an applied field, empirical research has found that university researchers tend to choose their
research topics out of their own assessment of production needs and not from contacts with
farmers and with extension agents (Velho, 1990). There are also links between university
research in tropical diseases and knowledge demands to solve them, but the channels connecting
knowledge producers and users are far from satisfactory (see, for example, the case of the
Medical School in Leon, Nicaragua, in Velho, 2004). The links between LAC universities and
the industrial sector are, in addition, quite limited. They do not normally involve research
activities, consisting mainly of short-term contracts mainly for consultancy, trouble-shooting
and routine analysis.31 A report compiling results for various case studies of university-industry
links in LAC countries concluded that “human capital in universities appears to be poorly
matched to the needs of the private sector”, and that “LAC business leaders do not generally
consider higher education to meet the needs of the economy” (Ferranti et al, 2003: 228). Finally,
the social sciences, by the very nature of their research subject, do tend to focus on the local

33
reality. But once, again, research problems are identified as such by the researchers themselves
and do not arise from social demands. Ironically, one notable exception to this trend is the work
done by the elite social scientists in the best universities, who are often contracted out by
government to carry out studies to inform policy decisions.

One fundamental contribution of universities to knowledge production is the training of new

professionals as well as new researchers. With regards to the former, there is considerable
evidence that successful innovative activities in any sector are dependent on human resources
originating in local universities. As for the formal training of researchers, expansion of post-
graduate education in LAC has been extremely slow. Students enrolled in post-graduate
programmes represented, on average, only 2.4 % of overall tertiary education enrolment in
1997, compared with 13% in the US. Moreover, most of these students were enrolled either in
non-degree specialisation courses or in Masters programmes in the social professions such as
law and business (Schwartzman, 2001). Concerning doctoral training, whereas OECD countries
produce, on average, 1 new PhD per year per 5,000 population, the ratio is 1 PhD per 28,000
population in Brazil, 1 per 140,000 in Chile and 1 per 700,000 in Colombia (Table 15).
Moreover, the training of post-graduate students is highly concentrated in just a few countries in
LAC, with more than two thirds of all Latin American post-graduate students in just two
countries, Brazil and Mexico (World Bank, 2002). Brazilian universities, however, graduate
almost 6 times as many PhDs as Mexican ones (Table 15). In addition, the Brazilian
government also provides scholarships to over 1,500 doctoral candidates to study abroad every
year. Another important feature of the Brazilian post-graduate training programmes is that they
are free of any tuition, provide maintenance stipends to a high proportion of the students
(around 50%) and extend this access to foreign students. For this reason, a large number of
students from other LAC countries are enrolled in graduate programmes in Brazil. 32

Although there is agreement that LAC countries must increase their R&D workforce in order to
meet the needs of a “knowledge society” (Hansen et al, 2002), a clear problem faced by those
who do get trained is a lack of career opportunities. In most LAC countries, PhD programmes
are seen almost exclusively as a necessary step for those willing to pursue academic careers and
not for taking a job in the private sector or other career options, which require independent
thought and analysis of complex problems. In Brazil, for example, an extensive survey of
doctoral students in all scientific fields revealed that over 80% of them aspired to take up a job
at a research university after the degree (Velloso & Velho, 2001). While aspirations may be
interpreted as reflecting a knowledge of the job market - where firms do not seem interested in
hiring PhD holders - the fact remains that research is seen as an activity typical of the public
sector which already faces difficulties in absorbing the very same research work force so needed
by these countries. The best known result is brain drain, or the relocation of highly trained

34
researchers to countries where they can find better working conditions. Although Brazil, Costa
Rica and Chile have, unlike the other LAC countries, been able so far to retain most of the
researchers they have trained, some estimate these countries have now reached a level where
more aggressive retention policies will have to be implemented (Guimarães, 2002). No doubt
LAC countries cannot afford to waste research resources in this way and therefore there is an
urgent need to address this problem of how to make PhD holders more attractive to the
enterprise sector.

3.3 Public Research Institutes

Research activities in most LAC countries started in the 19th century with the creation of
government research institutes in specific fields, and aimed to produce relevant knowledge to
solve practical problems. Therefore it is not surprising that the institutes concentrate in fields
related to natural resources (botanical gardens, agriculture, forestry, geology) and health which,
being locally specific, could not rely only on knowledge from the North (Dantes, 1980). A look
at Table 14 reveals that all LAC countries have established research institutes in such fields –
agriculture, for obvious reasons, is the most common, as confirmed by figures in Table 3, which
show that agriculture receives the highest share of R&D expenditures in many LAC countries.
The significance of natural resources for the region has also led to the establishment of a
number of regional and international research institutes, such as: the Inter-American Institute for
Co-operation in Agriculture and three centres of the Consultative Group on International
Agricultural Research - CGIAR system (for maize in Mexico; for potato in Peru; and for
tropical agriculture in Colombia). Despite the importance of such international institutes in
knowledge generation, the general feeling is that they are not part of the R&D structure of LAC
countries. The international research centres are perceived to be too much oriented towards the
scientific mainstream and far from the needs of the countries where they are physically located.

LAC countries also established industrial technology research institutes to support the process
of industrialisation that was taking place. In addition, some countries created state enterprises in
strategic sectors, such as oil, telecommunications, electricity and space, with dedicated R&D
institutes attached to them. The assessment of the performance of these institutes is mixed, but
there is a tendency to accept that they used to enjoy a disproportionate share of the national
research budgets (data presented in Table 2 show that they perform 24% of R&D) and delivered
proportionately little to the productive sector. There are two reasons for this. First, government
researchers have relatively little understanding of the specific needs of industry and hence the
necessary feedback is absent. Second, it is difficult to provide researchers in public institutes
with strong incentives to be responsive to industrial needs (Ferranti et al, 2003:224).

35
With the fiscal crisis faced by LAC in the 80s and the shift in mainstream political thought
concerning the role of government in economic growth, public research institutes were called
upon to modernise and to source part of their funds from the market, by establishing
partnerships with the productive sector. This required a reorganisation of the institutes in terms
of setting up the research agenda, stimulating interdisciplinarity, acquiring skills in planning,
management, monitoring and evaluation and foresight. 33 On the negative side the institutes now
tend to work for clients who can pay for their services and to neglect the needs of diverse other
potential clients, such as small farmers and SMEs, who cannot afford to contract out S&T
services. Budget cuts have also been achieved at the expense of the little independent and long-
term research and development projects initiated by the institutes. Moreover in the poorer
countries the institutes are increasingly dependent on foreign financial assistance. In addition, a
complete absence of wage, recruitment, promotion and training policies has provoked
professional frustration and low moral among researchers and many have expressed desire to
leave the institutions and the countries (Alcorta and Peres, 1998). Some institutes have been
able to adjust better than others. Many LAC agricultural institutes, for instance, have been able
to reinvent themselves and are often cited as examples of success (the Brazilian EMBRAPA, a/
OECD, 2001). But there are also success stories in industrial fields, as is the case of the Institute
for Rubber and Plastic in Colombia that is an important source of technology transfer to the
revived Colombian plastic industry (Ferranti et al, 2003). In all cases it seems that when the
institutes have strong links with the private sector, they are able to do relevant R&D and thus
raise part of the funds they need. The question remains about the unattended demand or needs of
users who cannot pay.

3.4 Business Enterprises

Business enterprises in LAC contribute 33% to R&D expenditures and perform about the same
share, 36% (Table 2). This has been an increasing trend from past decades, but seems to have
reached a plateau since 1995. Yet, businesses hire only 11% of LAC researchers, which is an
indication that whatever R&D they are performing does not require highly qualified personnel.
Of course, these figures may also be interpreted to mean that enterprise researchers have been
well funded in their activities. While it is a reasonable assumption, the latter does not find
support on the available evidence.

The composition of business enterprises in LAC includes state-owned enterprises (SOEs) and
the private sector, which can be grouped in three categories of firms, namely local subsidiaries
of transnational corporations (TNCs); small and medium-sized enterprises, most of which are
family-owned firms (SMEs); and large domestic conglomerates.

36
SOEs were created in LAC after WW II, as the state took on the production of goods and
services in energy, transport, telecommunications, as well as some industries related to the
defence sector of the economy such as oil and petrochemicals, iron and steel. It soon became
necessary for SOEs to create their own R&D and engineering departments, which were very
intensive in R&D, had strong links with some universities as well as with SMEs in the private
sector (Dagnino & Velho, 1998). A number of studies document the success of such
arrangements, but also recognise that “the innovation system developed was fragmented […]
and failed to serve as a true engine for growth in the domestic environment” (Katz: 2001: 6).
With the privatisation of state-owned enterprises during the 90s, a significant reduction in
enterprise investment in R&D took place (Amann & Baer, 1999). As most of these enterprises
are now foreign-owned, they are operating on the basis of imported capital equipment and
engineering know how, which they bring from their respective headquarters (Katz, 2001). Firms
that have easy technological assets already developed elsewhere do not replicate such
investment under conditions where there are lesser economies of scale and scope (Erber, 2000).
Authors tend to agree that privatisation in LAC “involved the ‘destruction’ of human capital
and domestic technological capabilities and their substitution by capital ‘embodied’ new
technology, as in the case of imported capital goods, or by foreign-produced R&D and
engineering services, as in the case of the telecom and energy companies now operating in the
region” (Cimoli and Katz, 2001:16). 34 In short, privatisation of SOEs has diminished the
technological gap between LAC and advanced countries, but has had a very negative impact on
LAC R&D institutions by excluding them from technological development process.

Together with the privatisation of SOEs, an important macroeconomic policy adopted by all
LAC countries during the 90’s relates to the liberalisation of the market. The main argument
against the market protectionist measures adopted by LAC during the phase of industrialisation
by import substitution (IS) process is that the incentive for firms and governments to act in the
technology creation direction is often provided by market competitive pressures, both domestic
and external. Therefore, opening up the economy to external competition, as LAC countries
increasingly did in the 90s, should have had a beneficial effect on technological innovation.
Actually, TNCs have been rapidly moving into LAC countries as a result of the opening of the
national markets. They have not only moved into the sphere of activities of the privatised SOEs,
but also into SMEs and locally owned conglomerates. LAC countries compare very favourably
to newly industrialised countries in securing Foreign Direct Investment (FDI). In LAC FDI
flows increased from 0.5 % of purchasing power parity (PPP)/GDP in 1989 to 3.2% PPP/GDP
in the region (Chile being the largest receiver: 10.3% PPP/GDP in 1999 - Ferranti at all, 2003).
Given this trend towards increased TNC participation in LAC countries, one immediately asks
whether FDI encourages or discourages local research activities. Most studies that deal with

37
the impact of FDI on the domestic economy agree that recipient firms benefit from foreign
investment. However, results on intra-industry spillovers are more ambiguous. The composition
of FDI matters: in a number of countries, a large component of FDI has been in the service
sector and retail distribution, both low R&D intensive (for example, Brazil and Mexico) where
the level of technology and knowledge transfer is reasonably small. Studies in the
manufacturing sector, among car assemblers and pharmaceuticals in Brazil and Argentina for
instance, found that despite differences in companies’ strategy, the tendency is towards a
downsizing of R&D personnel and infra-structure in Latin America, while increasing Brazil’s
role in production (Quadros & Queiroz, 2000). In pharmaceuticals, not only was R&D excluded
from Brazilian subsidiaries but the local production is regarded as being simpler and of lower
aggregated value products in comparison with the medicaments produced in developed
countries (Queiroz, 2001). In Mexico, most high-tech products are produced in the export
processing zones. Linkages with the domestic economy are weak and even in the case of the
vehicle industry, which is based in mainland Mexico, earlier stronger linkages with local
specialised suppliers seem to be waning (Alcorta and Peres, 1998:876). In short, relatively little
FDI in LAC is in R&D intensive activities; and when it is so, most R&D is done at the parent
firm. Therefore, there seems to be little demand from TNCs for local R&D institutions in LAC
countries, meaning that the liberalisation process in the region determines that such countries
participate more in the globalisation of production and less in the globalisation of S&T, which
remains very poor (Katz, 2002).

The general assessment presented above about the low incentive for R&D investment by
enterprises in LAC is confirmed by innovation surveys conducted in a few LAC countries since
the mid-90s. In Brazil, for example, PAEP, which stands for Survey of Economic Activity in the
State of São Paulo, is a large sample survey of firms in the most industrialised Brazilian federate
state. The first PAEP collected data in 1997, referring to 1996, in more than 10,000 industrial
firms, related to a number of economic variables, following as close as possible the Oslo
Manual. The main conclusion from the survey was that “there has been much innovation and
little knowledge in the innovation process of the industry of São Paulo. […] such innovative
performance has relied little on internal R&D activities carried out by firms. They rely primarily
on sources of information other than R&D to innovate, such as clients and competitors and other
firm departments” (Quadros et al, 2001: 215-216). A comparative analysis of the innovation
surveys conducted in Argentina, Chile, Colombia, Mexico, Venezuela and Uruguay has pointed
out as its main conclusion that “industrial innovation in Latin America is still highly informal”
(Arocena & Sutz, 2000: 64). This implies that even if a fair proportion of firms innovate in
process and product, they do not carry out R&D. Neither do they fund research activities in the
local universities and public research institutes. As a matter of fact, the surveys revealed very

38
few interactions between firms and universities (only 6% of firms surveyed in all the countries
indicated some form of interaction with local universities).

The picture above is also compatible with the low private sector investment in R&D and low
patenting activity that has been discussed previously.

3.5 Non-governmental organisations – NGOs

In most developing countries NGOs mushroomed in the last ten years. In Nicaragua, for
example, during the 1980s, the government registered 114 NGOs. From 1990 to 1997, the
number of NGOs increased to 1615. Between 1990 and 1995, these NGOs channelled US$316
million into the country in development aid (Toni & Velho, 2000). Only a very small proportion
of these NGOs carry out some form of R&D and when they do, it is mostly connected to
problem identification and is action-oriented. The figures in Table 2 show a somewhat
conflicting picture of the share of NGOs in R&D activities in LAC. On the one hand, data on
R&D financing and performing by NGOs has decreased during the 90s; on the other, the
number of researchers in NGOs has more than doubled - to almost 10% in 2000 - a quite
significant share, similar to that of researchers in enterprises. The interpretation of this figure is
not straightforward, however. Not only could there be a problem in data collection but the high
figures may also reflect the fact that young researchers are joining NGOs in increasing numbers,
either because of the difficulty of finding jobs in the formal R&D sector or simply by choice.
There is evidence, though, that the R&D conducted by NGOs is gaining recognition and is
increasingly being published in mainstream journals. This has only been possible because
external donors have elected NGOs a privileged receiver of funds, an entry point to the
community or society donors want to reach, thus by-passing local governments and universities,
which are regarded as corrupt and inefficient (Stiles, 1998). Under these circumstances, the
focus of research conducted by NGOs is much more in tune with donors’ agenda. Donors tend
to have a strong commitment to environmental issues (including biodiversity conservation),
poverty alleviation (which is easily associated with education and support to poor farmers and
SMEs), issues of social justice (including gender and minorities’ rights) as well as to the
maintenance of cultural diversity (bilingual education and indigenous knowledge), themes
which are commonly marginal in the research agenda of universities and research institutes.
Table 14 presents some examples of research-oriented NGOs in a few LAC countries for which
it was possible to gather reliable information. The research fields where such NGOs concentrate
resemble the preferences of donors mentioned above.

Research oriented NGOs play an important role in the recollection, preservation, transmission
and use of indigenous knowledge in LAC. A number of them attempt to record local peoples’

39
knowledge of plants, their medicinal uses, and the ways they are prepared and administered,
thus helping communities control their increasing dependence on external medical support. 35
Others report stories of traditional knowledge being best adapted to local conditions and more
environmentally friendly. One of the most fascinating such cases is about the communities
inhabiting the Pacific region of Colombia where a network of about 140 local organisations
were able to develop what might be described as an alternative political ecology framework that
includes conceptualisations of development, conservation and sustainability (Escobar, 2001).
This and other examples show how transnational development, environmental and cultural
rights networks – involving the linkage of indigenous organisations, the state, universities and
international actors – have helped to reconfigure traditional power relations and increased
indigenous peoples’ access to land, resources and state institutions and thus, preserving their
culture and strengthening their knowledge. In terms of promoting interactions between
stakeholder, government policymaking bodies seem to have a lot to learn from NGOs.

3.6 Infrastructure to articulate supply and demand of R&D

The traditional way of making research results available to users is publishing. LAC scientific
journals are neither well known nor properly distributed; they scarcely reach libraries outside
the region and, since many of them are published in Spanish or Portuguese, they have limited
potential readers in the international scientific community. An initiative to enhance the visibility
of the region’s journals is in place, however. Known as LATINDEX, it is a regional information
system based on a co-ordinated network of national resource centres, each taking responsibility
for the collection of bibliographical information in all knowledge areas from their respective
countries of LAC (Cetto & Gamboa, 1998). Electronic publishing is also growing, albeit at
varying speed across the region: Brazil, Cuba and Mexico are already offering a number of
online titles, while Argentina and Colombia have limited online information about their
journals. Most countries in LAC, however, especially the smaller ones, do not have the
communication infrastructure needed to support these online services. This affects not only the
access of researchers to journals (they are the local actors most likely to have access to ICTs)
but mainly the dissemination of results to potential users in society. The number of Internet
hosts (per 10,000 people) and personal computers (per 1,000 people) in industrial countries are
811 and 353 respectively. In LAC the numbers are 23 and 44, but vary across the region. Costa
Rica has the highest penetration of computers in the region (100 per 1,000 people), while in
Cuba and Bolivia the penetration is only 10 per 1,000 people. Argentina, Chile and Mexico
feature between 40 and 50 hosts per 10,000 people whereas Cuba and Bolivia are again the
countries with the lowest numbers (1 one per 10,000 people) (Hansen et al, 2002).

40
Other forms of articulating supply and demand of R&D, such as extension services in
agriculture, have been dismantled in LAC under the allegation that they were outdated and had
not been effective (Souza, 1988). It is true that such ex-post mechanisms of interaction whereby
research takes place first, followed by dissemination of results to users is generally believed to
be inconsistent with a systemic approach to innovation. It seems to be the case that interactions
among “nodes” of an innovation system need to be fostered ex-ante, that is, included in the
definition of the research agenda.

The evidence presented above suggests that LAC countries have not been able to build virtuous
links among the various relevant social actors involved in S&T production and use. Problems
stand both in the supply and demand sides. Concerning the former, universities and public
research institutes, which together perform almost 70% of R&D, have not created mechanisms
to identify user’s needs and set their research agenda more on the basis of scientific criteria
dictated by international mainstream science. On the demand side, there is not much demand on
local R&D since TNCs innovate on the basis of R&D conducted in the advanced countries and
local private firms, in order to be competitive, also prefer to import foreign technology. Of
course, this supply/demand framework is grossly oversimplified: universities and research
institutes also demand knowledge (codified and tacit) from other social segments and
enterprises and social organisations are important suppliers of knowledge. Of particular note is
the lack of connection between the various agents. Government in some countries have
implemented a few schemes aimed to bring together supply and demand but, with the exception
of a few cases, they have not been successful.

41
42
4.0 CONCLUSIONS

There seems to be a big knowledge divide between LAC countries and the technologically
advanced regions (as well as with the Asian Tigers) in most of the dimensions of the S&T sector
examined here. Although this divide holds for LAC as a whole as well as for individual
countries in the region, there are also significant differences among LAC countries themselves.
As a norm, countries like Argentina, Brazil, Chile, Costa Rica and Mexico tend to perform
better in a number of indicators than the smaller countries of South America, Central America
and Caribbean.

The first group of countries has reached a significant level of R&D expenditure per population,
achieving a critical mass of researchers who have been able to increase considerably their
contribution to mainstream science, through publications. It is also good news that almost half
of those researchers are women (even if most professorships are held by men). The other group
of countries is still struggling to reach this same state. More generally, there has been an
increase in intra-regional collaboration across LAC and most governments are aware of and
committed to implementing schemes to increase the utilisation of research results.

The main issue here is what to make of these trends. An immediate question is whether the
existing divides are being maintained across the region and in relation to the advanced countries
or if there are signs that both gaps are closing?

If the current trends continue, the divide will grow. It is clear that the transition of the current
advanced countries to a knowledge society took place with an enormous growth in R&D
expenditure and performance not from the part of government but mainly from the private
sector. We saw in the data and analysis presented above that the public sector is still responsible
for about 70% of R&D funding and performance in LAC and while there are faint signs that this
may be changing, the current pace is too slow to allow catching up.

An important feature of the technologically advanced countries was their ability to articulate a
national system of innovation that has established strong links among its diverse components
(firms, universities, research institutes, financing institutions, regulation agencies, etc) and
operates under the right incentives provided by the state, in terms of macroeconomic policies
and political climate. Our LAC countries, however, (or at least some of them), concentrated
their (erratic) efforts in creating a strong public sector research system, with variable results, but
did not foster the links between this system and the business enterprises. In the sectors where
such links were established this took place under a protectionist economic policy, which created

43
a research bureaucracy without the incentive to be creative and public firms that did not have to
face competition to survive. With the liberalisation and privatisation measures adopted in the
90’s, LAC countries found themselves ill-equipped to face fierce competition from foreign
firms. If the current trend continues - where local firms (SMEs, large domestic conglomerates
and SOEs) are acquired by TNCs that bring their own technology developed elsewhere and
force the same technology importing pattern on the few local firms that wish to be competitive –
it is highly unlikely that LAC countries will be able to articulate their national innovation
systems. Under these circumstances, local R&D will continue to be disconnected from potential
users.

Such a scenario can only be avoided with government intervention in the form of the right
macroeconomic policies and incentives to innovate, creating the famous links among the
components of a potential system of innovation. Obviously governments cannot do this for the
whole economy, but specific industries where the countries have particular strengths and
interests could well be the focus to start the process.

Finally, one notable trend of the R&D sector in LAC is its distance (or lack of links) with the
needs of the various social groups in civil society. Some of these needs are being addressed by
researchers working with NGOs and funded by external donors. It is extremely important that
the connection is made between research oriented NGOs together with the groups they assist on
one hand and the formal R&D systems on the other. Both NGOs and universities have a lot to
gain from tackling problems together and society is, ultimately, the greatest beneficiary. It is to
reach this goal that public investment in R&D is justifiable.

1
Institutions are understood here as the combined environment of physical organisations and the
practised routines, norms, shared expectations, morals, etc (Edquist & Johnson, 1997).
2
For comprehensive discussion on how different authors have defined the concept of national innovation
systems see Edquist (1997). The current state of the art is presented in Lundvall, Johnson & Dalum
(2002).
3
According to the UNESCO 2001 Report on Science and Technology, Scientific and Technological
Activities (S&T) cover research and development, scientific and technical services and scientific and
technical education and training (for proper definition of each activity see www.unesco.org/statistics).
The conceptual distinction between these three types of activities is not always easy to operationalise. For
example, investment in scientific training should not be included in R&D, unless it involves post-graduate
training, which has a strong research component. However, in practice time spent in class and research is
not easy to separate. Therefore, some countries, like Brazil, do include in R&D salaries of all faculty
employed in universities that perform research, although some may not be very active researchers. Thus,
comparisons across countries may be somewhat misleading, as concepts are open to different
interpretations and collection practices. This notwithstanding, it is always the case that R&D is contained
in S&T.
4
Ricyt is the Iberoamerican Network of S&T Indicators, co-ordinated from Argentina and funded by the
Organisation of American States (OAS) and the Spanish Development Co-operation Agency – CYTED.
Although Ricyt puts an enormous effort into conceptualisation and standardisation for data collection, it
does not collect data itself, but relies on information provided by the relevant organisation in each
country. See: www.ricyt.org

44
5
See http://www.mct.gov.br/estat/52.htm for an explanation of the changes in methodology for data
collection.
6
The evolution of the ratio of private to public R&D expenditures, from 1980 to 1995 in South Korea and
Ireland was, 1.62 to 4.80 and 0.80 to 2.60, respectively (Ferranti et al, 2003).
7
Expressed in terms of GNP, Official Development Assistance (ODA) fell everywhere from 1989 to
1996, but still represents from 1.03% (Denmark) to 0.12% (US) – something from USD 7-8 billion
(Japan, USA, Germany and France) to 2-3 billion (UK, Denmark, the Netherlands), according to Gaillard
(1999, p. 298-301). Although accepting that data on ODA are quite unreliable, the author estimates by
crossing various sources that between 5 and 10% is spent on research (see also Wagner et al, 2001, p.21-
2).
8
The Programme for the Development of Basic Sciences -PEDECIBA- was established in October 1986
by agreement between the Ministry of Education and Culture, on behalf of the Government of Uruguay,
and the University of the Republic, with strong support from UNDP. From 1993 to 1997 research projects
and fellowships were partially administered by CONICYT. Since 1995 PEDECIBA has been
incorporated to the national budget on a permanent basis. This programme had, among others, the
objective of reverting the huge brain drain of natural scientists provoked by the military dictatorship from
1973 to the mid-80’s (http://www.rau.edu.uy/pedeciba ).
9
Data on R&D expenditures by and researchers in the enterprise sector in Brazil were collected in a
survey based on a very limited sample of firms. There is serious controversy in the country on the
representativeness of the sample and the methodology used to generalise the information. Most analysts
believe that the participation of the business sector as it appears in the indicators of S&T produced by the
Ministry of Science and Technology has been grossly overestimated.
10
Bilateral development agencies supporting social sciences research in Bolivia are the Swedish
SIDA/SAREC and the Dutch Directorate for Development Co-operation – DGIS. Paradoxically the same
SIDA/SAREC has not agreed to support social sciences in Nicaragua, despite insistence and justification
from local universities (for Bolivia see Souza Paula et al, 2000; for Nicaragua, personal information).
11
A recent survey in Brazil found out that only 2.3% of black and mulatos in the age group of 18 to 25
years were enrolled in tertiary education, The equivalent figure for the while population of the same age
is 12% (JC e-mail 2305, 24 junho de 2003, www.jornaldaciencia.org.br)
12
It is not surprising that the LAC with significant indigenous populations, such as Guatemala and
Bolivia, the participation of such social segment in S&T activities is dismal. Language barrier plays a part
here as a significant proportion of indigenous population does nor speak English, the language of
instruction, at home (World Bank, 2002).
13
See http://www.unicef.org/sowc01/tables/
14
“El factor género en las estadísticas del CONACYT México”, www.segecyt.org.uy.
15
“Mujeres, Ciencia y Tecnología en el Uruguay: Situación del CONICYT”, www.conicyt.gub.uy.
16
see León (2001).
17
Data for this paragraph was taken from: www.cnpw.br for Brazil; www.segecyt.org.uy for Argentina; e,
www.conicyt.gub.uy for Uruguay.
18
“Women in public research and higher education in Europe”, Eurostat, Statistics in focus, Science and
Technology, Theme 9 – 7/2001.
19
“Participación de la mujer en el sistema de Investigación y Desarrollo en Argentina”,
www.secebyt.org.uy.
20
See http://lattes.cnpq.br/censo2002/sumula/index_sumula.htm
21
In 1984 I published a pioneer article, which had a considerable impact, criticising the use of the ISI
database to analyse the scientific output of LAC (Velho, 1984). In that occasion, as in many others that
followed (Velho, 1994; Velho, 1998), I pointed out that the bulk of the publication of any LAC country,
particularly in the social sciences, agriculture and environment as well as results of more applied research,
appeared in local journals, published in Spanish and in Portuguese and that there was no evidence that
such publications were of lower quality than the ones published in journals indexed by the SCI.
Therefore, my main argument has been that scientific output of a specific LAC country must be based on
local data collection of publication. However, I also pointed out that for comparative studies among
different countries there is no other way but to use international databases.
22
If specific research fields are considered separately, the contribution of LAC to world scientific output
can vary considerably. In the medical field, for example, LAC contribution to world science, according to
Medline is 1.8% of 479,731 publications indexed in 2000; in biology, on the basis of 572,218 itens
compiled by Biosis, LAC contributed 2.8%. In agriculture, however, LAC contribution to the world
output in 2000 was considerably higher: of total 162,507 titles compiled by CAB, almost 7% originated in
LAC countries (www.ricyt.org). The scientific output in the social sciences is particularly difficult to

45
estimate given the absence of reliable databases and this is a universal problem and not one restricted to
LAC. The existing Social Sciences Citation Index is too limited and has too many methodological
problems to be used even for comparative purposes only. In LAC social scientists are the most likely to
publish in the local languages and in domestic journals (Velho, 1994).
23
From 1985 to 1995, European countries have doubled their production of scientific articles
but have tripled the number of articles co-authored with partners of advanced countries in other
continents (Georghiou, 1998). From 1986-88 to 1995-97, the total number of articles in the ISI
databases increased by 12 percent; internationally co-authored articles increased by almost 115
percent in the same period (NSB, 2000: 6-48)
24
See http://www.uspto.gov/web/offices/ac/ido/oeip/taf/apat.pdf
25
Research Councils as both policy and funding organisations for S&T were established in LAC mostly
through the 60’s and 70’s, under advice and support of UNESCO and the Organisation of American
States (OAS). A number of authors have analysed the consequences and impacts of this “blueprint” (see,
among others, Amadeo, 1978; Oteiza, 1992; Bastos & Cooper, 1995).
26
For an analysis of the case of Bolivia, see Escobar (2002); for Paraguay, see Davyt (1997); and for
Nicaragua, Velho (2004).
27
This is what has been called by Latin American authors as an "ofertista" S&T policy, meaning a policy
which takes care of the supply side only (Sagasti, 1980; Avalos, 1991). According to this logic, by
supporting scientific research and strengthening research-training capabilities in universities, government
was, indirectly, contributing to technological development of the firms.
28
Policy analyses of U/I relations in LAC have been done in: Mexico (Casas, 1997), Argentina
(Chudnovsky & López, 1996); Brazil (Velho & Saenz, 2002); and the whole region (Arocena & Sutz,
2001), among many others.
29
A special issue of Interciencia presents and analyses the case of salary supplements as a policy adopted
by research councils in different LAC countries.
30
In some LAC poorer countries such as Nicaragua, Bolivia and Dominican Republic, the public
universities have no research tradition, although most faculty members are hired on a full-time basis.
Academic career does not reward degrees or research publications, but is based on seniority. Faculty
members extra pay when they take on administrative positions – see Velho (2004) for Nicaragua, Souza
Paula et al (2000) for Bolivia and Pimentel (2002) for Dominican Republic.
31
Number of studies present convergent findings in this respect. For a review see Velho & Saenz (2002)
and Sutz (2000).
32
The exact number of foreign graduate students in Brazil is difficult to estimate since there is no
centralised system collecting this data for all universities. Brazilian government provides a special
scholarship for foreign students from some poorer countries of LAC and Africa (see www.capes.gov.br)
33
A number of case studies about the re-orientation and modernisation of the public research institutes in
Brazil can be found in Salles-Filho (2000).
34
For the same argument see also Alcorta and Peres (1998); Katz (2002); Cimoli and Correa (2002)
35
Examples of this kind of programmes: Royal Botanical Gardens, Kew about the Yanomami Indians of
the northern Brazilian Amazon (Milliken 2002), and Project Tramil in the Caribbean
(www.funredes.org/endacaribe/tramil ).

46
BIBLIOGRAPHY

Alcorta, L. and Peres, W. (1998), Innovation Systems and Technological Specialization in Latin
America and the Caribbean, Research Policy 26: 857-881.
Amadeo, Eduardo (1978), Los consejos nacionales de ciencia y tecnología en América Latina.
Éxitos y fracasos del primer decenio, Comercio Exterior, 28 (12): 1439-47.

Amann, Edmund and Baer, Werner (1999), From Technology Absorption to Technology
Production: Industrial Strategy and Technological Capacity in Brazil’s Development
Process, Economia Aplicada, 3 (1): 109-138.
Arocena, Rodrigo & Sutz, Judith (2000), Looking at National Systems of Innovation from the
South, Industry and Innovation, 7(1): 55-75.
Arocena, Rodrigo & Sutz, Judith (2001), Changing Knowledge production and Latin American
universities, Research Policy 30: 1221-1234.
Avalos, I. (1991), La politica technologica Venezolana: de la economia protegida a la economia
abierta’, Espacios, vol. 12 (2).
Barreiro, Adriana (1997), La formación de recursos humanos para investigación en el Uruguay
a partir de la experiencia del PEDECIBA, Montevideo: Ediciones de la Banda Oriental,
S.R.L.
Bastos, Maria Inês & Cooper, Charles (1995): ‘A political approach to science and technology
in Latin America”, in M.I. Bastos & C. Cooper, Politics of Technology in Latin America,
London/New York: Routledge/UNU Press, 1-27.
Bernardes, Américo T. & Albuquerque, Eduardo M. (2003), Cross-over, thresholds, and
interactions between science and technology: lessons for less-developed countries, Research
Policy, 32: 865-885.
Cardoza, G. (1997), Learning, innovation and growth: a comparative policy approach to East
Asia and Latin America. Science and Public Policy, 24(6): 377-393.
Casas, Rosalba (1997) ‘El gobierno: hacia un nuevo paradigma de política para la vinculación’,
in Casas, R. & Luna, M. (eds) Gobierno, Academia y Empresas en México: hacia una
nueva configuración de relaciones, Mexico: Plaza Y Valdés, pp. 71-114.
Cetto, Ana Maria & Gamboa, A. (1998), Scientific periodicals in Latin America and the
Caribbean: a global perspective, Interciencia 23(2): 84-93.
Chudnovsky, D.and Lopez, A. (1996) ‘Política Tecnológica en la Argentina:¿Hay Algo mas que
Laissez Faire?’ REDES, vol. VI: pp. 33-75.
Cimoli, M. and Correa, N. (2002), Trade Openness and Technological Gaps in Latin America: a
'Low Growth Trap', Laboratory of Economics and Management - Working Paper Series,
Sant'Anna School of Advanced Studies.
Cimoli, M. and Katz, J. (2001), Structural Reforms, Innovation and the Technology Gap: a
Latin American Perspective, CEPAL.
Clark, Norman; Hall, A.; Sulaiman, R. & Naik, G. (2003), Research as Capacity Building: the
case of an NGO facilitated post-harvest innovation system for the Himalayan Hills,
World Development 31(11): 1845-1863.

47
Dagnino, Renato and Velho, L. (1998) ‘University-Industry-Governement Relations on the
Periphery: The University of Campinas, Brazil’, Minerva, vol. XXXVI, no. 3, pp. 229-
251.
Dagnino, Renato, Thomas, H. and Davyt, A. (1996), El pensamiento en ciencia, tecnologia y
sociedade en Latinoamérica: una interpretación de su trayectoria, Redes, 3:13-51.
Dantes, Maria A. (1980), Institutos de Pesquisa Científica no Brasil, in: Ferri, M. & Motoyama,
S. (eds), História das Ciências no Brasil, São Paulo: EDUSP.
Davyt, Amilcar (1997), Cooperación en Ciencia y Tecnologia en los países del Mercosur.
Estudio de caso de Paraguay, in Velho, Léa (ed), Cooperação em Ciência e Tecnologia no
Mercosul- Fase I. Síntese Final e Estudos de caso. Brasília: Ministério de Ciência e
Tecnologia, Organização dos Estados Americanos
Davyt, Amilcar & Velho, L. (1999), Excelencia cientifica: la construcción de la ciencia a través
de su evaluación. La Comisión Sectorial de Investigación Científica (CSIC), Uruguay,
REDES 6 (13): 13-48.
Edquist, C. (ed.) (1997), .), Systems of Innovation: Technologies, Institutions and
Organisations. U.K.: Pinter Publishers/Cassell Academic.
Edquist, C. and B. Johnson (1997), Institutions and Organisations in Systems of Innovation, in
C. Edquist (ed.), Systems of Innovation: Technologies, Institutions and Organisations. U.K.:
Pinter Publishers/Cassell Academic.
Erber, F. (2000), “Structural Reforms and Science and Technology Policies in Argentina and
Brazil”, http://www-tecno.inti.gov.ar/erber.htm
Escobar, Arturo (2001), Culture sits in places: reflections on globalism and subaltern strategies
of localization, Political Geography 20: 139-174.
Escobar, Silvia Cristina Pabón (2002), Institucionalização da Política Científica e Tecnológica
na Bolívia: avanços e retrocessos, Master Thesis presented to the University of Campinas,
Brasil, Graduate Programme in Science and Technology Policy.
Fernández Rius, L. (2001), Roles de género y mujeres académicas. El caso de Cuba”, in Pérez
Sedeño, E. (ed.), Las mujeres en el sistema de Ciencia y Tecnología, Cuadernos de
Iberoamérica, Organización de Estados Iberoamericanos, Madrid.
Ferranti, D. d., Perry, G. E., Gill, I., Luis, G. J., Maloney, W. F., Sánchez-Páramo, C. and
Schady, N. (2003). Closing the Gap in Education and Technology. Washington, D.C., The
World Bank - Latin American and Caribbean Studies: 1-301.
Freeman, Christopher (1987), Technology Policy and Economic Performance: Lessons from
Japan, London: Francis Pinter.
Gaillard, Jacques (1999), La Coopération scientifique et technique avec les pays du Sud, Paris:
Éditions Karthala, 340 p.
García Guadilla, Carmen (1998), Situación y principales dinámicas de transformación de la
educación superior en América Latina, Caracas: CRESAL/UNESCO. Colección Respuestas
2.
Georghiou, L. (1998), Global Cooperation in Research, Research Policy 27: 611-626.
Guimarães, Reinaldo (2002), A migração de pesquisadores do Brasil, Ciência Hoje 32(187): 40-
43.

Guimarães, Reinaldo (2003), Pesquisa em Saúde e Reforma Sanitária, JC e-mail 2305, 24 June,
www.jornaldaciencia.org.br

48
Hansen, Thomas N.; Agapitova, N.; Holm-Nielsen, L. & Vukmirovic, O.G. (2002), The
evolution of science and technology: Latin America and the Caribbean in Comparative
Perspective, The World Bank: LCSHD Paper Series 80, December.
Herrera, Amilcar (1975), Los determinantes sociales de la política científica en América Latina.
In Sabato, Jorge (ed.), El pensamento latinoamericano en la problemática ciencia-
tecnología-desarrollo-dependencia , Buenos Aires: Paidos, 1975, pp. 98-112.
Hill, Derek (2000), Latin America: R&D spending jumps in Brazil, Mexico, and Costa Rica,
Tropical Report, Arlington, VA: National Science Foundation, Division of Science
Resources Studies.
Indicadores de C&T em São Paulo (2001), São Paulo: FAPESP.
Katz, J. (2002). “Efficiency and Equity Aspects of the New Latin American Economic Model.”
Economics of Innovation and New Technology 11.
Katz, Jorge (2001), Structural reforms and technological behaviour. The sources and nature of
technological change in Latin America in the 1990s, Research Policy 30:1-19.
Kochen, S., A. Franchi, D. Maffía and J. Atrio (2001), La situación de las mujeres en el sector
científico-tecnológico en América Latina. Principales indicadores de género, in Pérez
Sedeño, E. (ed.), Las mujeres en el sistema de Ciencia y Tecnología, Cuadernos de
Iberoamérica, Organización de Estados Iberoamericanos, Madrid.
Krauskopf, Manuel (2000), Los doctorados en Chile. Perfil y capacidad cientifica de los
programas en ciencias acreditados en Chile, paper presented at the
Lall, S. (2002), Science and technology in Southeast Asia, paper presented at the EC Strata
meeting in Brussels, 22-23 April 2002.
León, Magdalena (2001), Participación femenina en actividades de ciencia y tecnología en la
universidad ecuatoriana. Primeras evidencias, in Pérez Sedeño, E. (ed.), Las mujeres en el
sistema de Ciencia y Tecnología, Cuadernos de Iberoamérica, Organización de Estados
Iberoamericanos, Madrid.
Lopez, Ramon (2003), The policy roots of socio-economic stagnation and environmental
implosion: Latin America 1950-2000, World Development 31(2): 259-280.
Lundvall, Bengt-Ake (1992), National Innovation Systems: Towards a Theory of Innovation
and Interactive Learning. London: Frances Pinter.
Lundvall, Bengt-Ake; B. Andersen & R. Dalum (2002), National Systems of Innovation and
competence building, Research Policy 31(2): 213-231.
Milliken, William (2002), People, plants and publishing, In: Laird, S. Biodiversity and
Tradicional Knowledge, London: Earthscan Publications Ltd, pp. 79-81.
Mullin, James; Adam, R.M.; Halliwell, J.E. & Milligan, L.P. (1999), Science, Technology and
Innovation in Chile, Ottawa, Canada: IDRC.
Narváez-Berthelemot, N.; J.M.Russel & L.Velho (1999), The Scientific Collaboration of the
Mercosur Countries as an Indicator of Latin American Regional Activity, Research
Evaluation 8(2): 83-90.
Nelson, Richard (ed.) (1993) National Innovation Systems A Comparative Analysis, N.Y. &
Oxford: Oxford University Press.
NSB- National Science Board (2000), Science and Engineering Indicators 2000, Washington,
DC: National Science Foundation, www.nsf.gov
NSB- National Science Board (2002), Science and Engineering Indicators 2002, Washington,
DC: National Science Foundation, CD-ROM.

49
OECD (2001), Using Knowledge for Development. The Brazilian Experience, Paris: OECD
Publications.
OECD (2001), Basic Science and Technology Indicators, Paris: OECD Publications.
Oteiza, Enrique (1992): ‘El complejo científíco y tecnológico argentino en la segunda mitad del
siglo XX: la transferencia de modelos institucionales”, in E. Oteiza (coord.), La política de
investigación científíca y tecnológica argentina. Historia y perspectivas, Buenos Aires:
Centro Editor de América Latina, 115-25.
Patel, Pari & Pavitt, Keith (1995), Patterns of technological activity: their measurement and
interpretation. In Stoneman, P. (ed.), Handbook of the Economics of Innovation and
Technological Change, Oxford: Blackwell.
Pimentel, Josefina (2002), Renovation and Reform in Dominican Higher Education,
Academe 88(3), online version:
www.aaup.org/publications/Academe/02MJ/02MJpim.htm
Quadros, Ruy & Queiroz, Sergio (2000), The Implications of Globalisation for the Distribution
of Design Competencies in the Auto Industry in the Mercosur, paper presented at the VIII
International Colloquium of GERPISA, Paris, 8-10 June.
Quadros, Ruy; Furtado, A.; Bernardes, R. & Franco, E. (2001), Technological innovation in
Brazilian industry: an assessment based on the São Paulo innovation survey, Technological
Forecasting and Social Change, 67 (2/3): 203-219.
Queiroz, Sérgio (2001), Internationalization of R&D and the Development of Technological
Capabilities in the Brazilian Pharmaceutical Industry, Campinas: DPCT/IG, Research
Report, Mimeo.
RICYT - Red Latino Americana de Indicadores de Ciencia y Tecnologia – (2002) –
www.ricyt.org
Sabato, Jorge (1975), El pensamento latinoamericano en la problemática ciencia-
tecnología-desarrollo-dependencia, Buenos Aires: Paidos.
Sagasti, F. (1980), Science and Technology for Development, A review of schools of
thought on science, technology, development and technical change’, Ottawa:
International Development Research Centre (mimeo).
Salles-Filho (2000) (coord), Ciência, Tecnologia e Inovação. A reorganização da pesquisa
pública no Brasil, Campinas: Editora Komedi/CAPES.
Schwartzman, Simon (2001), The future of education in Latin America and the Caribbean,
paper presented at the Seventh Meeting of the Intergovernmental Regional Committee of
the Major Project in the Field of Education in Latin America and the Caribbean – ED-
01/PROMEDLAC VII/REF2 – Working Document
Sedeño, E. (ed.) (2001), Las mujeres en el sistema de Ciencia y Tecnología, Cuadernos de
Iberoamérica, Organización de Estados Iberoamericanos, Madrid.
Stiles, Kendall (1998), Civil Society empowerment and multilateral donors: international
institutions and new international norms, Global Governance 4: 199-216.
Souza, Ivan S. F. (1988), A importância do relacionamento pesquisa/extensão para a
agropecuária, Cadernos de Ciência e Tecnologia, 5(1/3): 63-76.
Souza Paula, Maria C.; Velho, L. & Barreiro, A. (2000), Comparative Study of the Impacts of
Donor-Initiated Programmes on Research Capacity in the South. The Case of Bolivia, report
presented to The Hague, NL: Ministry of Development Co-operation (DGIS).
Sutz, Judith (2000), The University-industry-government relations in Latin America, Research
Policy, vol. 29, pp. 279-290
Thorsteinsdóttir, Halla; Daar A. D & Singer, P. A. (2003), Building a Biopharmaceutical
Innovation System in Cuba: Growth Through Linkages, presented at the International
Workshop: Building (Bio)pharmaceutical systems in developing countries, Maastricht, the
Netherlands: UNU/INTECH, February (mimeo not for distribution).

50
Toni, Fabiano & Velho, L. (2000), Comparative Study of the Impacts of Donor-Initiated
Programmes on Research Capacity in the South. The Case of Nicaragua, report presented to
The Hague, NL: Ministry of Development Co-operation (DGIS).
UNESCO (2001), Report on Science and Technology, available at www.unesco.org
Urzúa, Raúl (2002), Regional overview: Latin America and the Caribbean, Higher Education
Policy 15: 277-290.
Varsavsky, Oscar (1969), Ciencia, política y cientificismo, Buenos Aires: Centro Editor de
América Latina.
Velho, Léa & Saenz, T. (2002), Public and Private R&D in Brazil: Complements or
Substitutes?, UNU/INTECH Discussion Paper # 2002-8, July.
Velho, Léa & León, Elena (1998), A Construção Social da Produção Científica por Mulheres,
Cadernos Pagu, 10: 309-344, 1998.

Velho, Léa (2004), Building Research Capacity and Contributing to Development: From Old to
New assumptions, Research Policy (accepted for publication), forthcoming.

Velho, Léa (2000), Comparative Study of the Impacts of Donor-Initiated Programmes on

Research Capacity in the South. Latin America Regional Report, The Hague, NL: Ministry
of Development Cooperation (DGIS).
Velho, Léa, (1998), Indicadores Científicos: aspectos teóricos y metodológicos e impactos en la
política científica. In E. Martínez & M. Albornoz (eds), Indicadores de Ciencia y
Tecnología: estado del arte y perspectivas, Caracas (Venezuela): Nueva Sociedad y
UNESCO, pp.25-51.
Velho, Lea (1995), Fontes de Influência na Construção da Agenda de Pesquisa Acadêmica,
Educação Brasileira,17(35): 87-105.

Velho, Lea (1994), Indicadores Cientificos: Aspectos Teoricos y Metodologicos, in Eduardo

Martinez (org.), Interrelaciones Entre La Ciencia, la Tecnologia y El Desarollo, Caracas,
Venezuela: Nueva Sociedad, pp.307-348.

Velho, Lea (1990), Sources of influence on problem choice in Brazilian Agricultural Research,
Social Studies of Science, 20: 503-517.

Velho, Léa (1984), Publications and citation practices of Brazilian agricultural scientists, Social
Studies of Science, 14: 45-62.

Velloso, Jacques & Velho, Lea (2001), Mestrandos e Doutorandos no País. Trajetórias de
Formação, Brasília: UNESCO & CAPES.

Vonortas, Nicholas S. (2002), Building Competitive Firms: technology policy initiatives in

Latin America, Technology in Society, 24: 433-459.

Wagner, Caroline S.; I. Brahmakulan; B. Jackson; A. Wong; T. Yoda (2001), Science and
Technology Collaboration: Building Capacity in Developing Countries?, Santa Monica,
CA: RAND Science and Technology, MR-1357.0-WB, 90p.

WDI – World Development Indicators (2002), World Bank – CD ROM

World Bank (2002), Constructing Knowledge Societies: New Challenges for Tertiary
Education, A world Bank Report, Washington DC.

51
TABLES

Table 1: Expenditure on S&T and R&D: Volume, % GDP and Expenditure/ 1,000
population
1990 1995* 2000*
millions % millions % millions % Exp./
US$ GDP US$ GDP US$ GDP 1,000 Pop.
Argentina S&T 647.10 0.33% 1 353.00 0.50% 1 430.00 0.50% 38.6
R&D 1 136.20 0.42% 1 247.20 0.44% 33.7
Bolivia S&T 47.00 0.54% 5.6
R&D 24.00 0.37% 24.50 0.28% 2.9
Brazil S&T 7 457.50 1.59% 8 897.71 1.26% 7 157.25 1.35% 42.0
R&D 3 544.07 0.76% 6 134.54 0.87% 4 626.52 1.05% 27.1
Chile R&D 154.93 0.51% 401.08 0.62% 394.96 0.56% 26.0
Colombia S&T 441.91 0.55% 303.40 0.36% 7.2
R&D 236.39 0.29% 153.72 0.18% 10.1
Costa Rica S&T 159.96 1.75% 164.94 1.58% 43.3
R&D 35.28 0.39% 36.23 0.35% 9.5
Cuba S&T 214.20 1.09% 188.70 0.87% 290.60 1.05% 26.0
R&D 136.60 0.70% 101.10 0.47% 146.30 0.53% 13.1
Ecuador S&T 33.03 0.18% 26.30 0.19% 2.1
R&D 14.30 0.08%
El Salvador S&T 28.70 0.30% 99.20 0.84% 15.8
R&D 9.65 0.08% 1.5
Honduras S&T 3.40 0.06% 0.5
R&D 3.20 0.05% 0.5
Mexico R&D 886.00 0.31% 2 283.64 0.40% 23.3
Nicaragua S&T 2.80 0.14% 0.6
R&D 2.60 0.13% 0.5
Panama S&T 33.45 0.63% 60.35 0.76% 101.57 0.91% 35.6
R&D 20.22 0.38% 29.96 0.38% 44.62 0.40% 15.6
Paraguay S&T 71.82 1.00% 13.1
R&D 5.69 0.08% 1.0
Peru S&T 389.82 0.74% 691.21 1.29% 27.0
R&D 58.30 0.11% 2.27
Trin & Tobago S&T 19.30 0.34% 15.9
R&D 7.60 0.13% 6.3
Uruguay R&D 20.62 0.25% 49.65 0.28% 47.75 0.24% 14.3
Venezuela S&T 176.60 0.37% 357.90 0.48% 404.86 0.33% 16.8
LAC S&T 10 395.11 0.90% 13 405.75 0.80% 15 037.73 0.76% 29.2
R&D 5 872.93 0.51% 9 528.11 0.57% 11 137.59 0.57% 21.6
Source: RICYT, 2002
* or closest year for which information was available

53
Table 2: Expenditure on R&D by Financing and by Performing Sector and Researchers by Sector
Financing Performing Researchers
1990 1995 2000 1995 2000 1995 2000
Argentina Government 45.5% 41.9% 40.9% 37.6% 33.8% 31.6%
Enterprises 27.7% 23.4% 25.9% 25.4% 16.1% 12.1%
Higher Education 21.8% 30.8% 31.5% 34.7% 49.1% 54.6%
Non-profit organisations 1.5% 1.9% 1.7% 2.3% 1.5% 1.8%
Foreign 3.5% 1.8%
Bolivia Government 37.0% 22.0% 25.0% 22.0% 17.5%
Enterprises 17.0% 22.0% 25.0% 26.0% 11.3%
Higher Education 12.0% 32.0% 30.0% 46.0% 67.0%
Non-profit organisations 22.0% 15.0% 20.0% 6.0% 4.1%
Foreign 10.0% 9.0%
Brazil Government 71.5% 59.1% 60.2% 12.4% 18.4% 17.1% 10.5%
Enterprises 23.9% 38.2% 38.2% 42.6% 37.4% 7.8% 31.1%
Higher Education 4.7% 2.7% 1.6% 45.1% 43.6% 75.1% 58.0%
Non-profit organisations 0.6% 0.3%
Foreign
Chile Government 46.1% 58.4% 70.3% 51.4% 40.4% 20.5% 19.7%
Enterprises 35.0% 26.5% 23.0% 6.4% 14.9% 5.9% 5.9%
Higher Education 40.9% 43.8% 68.2% 69.4%
Non-profit organisations 13.4% 9.0% 1.9% 1.0% 0.9% 5.5% 5.0%
Foreign 5.5% 6.1% 4.7%
Colombia Government 35.0% 16.6% 5.0% 6.0%
Enterprises 52.8% 48.4% 36.0% 18.0%
Higher Education 10.9% 33.6% 41.0% 57.0%
Non-profit organisations 1.4% 1.4% 18.0% 19.0%
Foreign
Cuba Government 55.1% 50.5% 53.1%
Enterprises 44.9% 49.5% 40.1%
Higher Education
Non-profit organisations
Foreign 6.8%
Costa Government 12.3% 17.0% 11.4%
Rica Enterprises 21.7% 24.8% 24.1%
Higher Education 36.6% 36.1% 60.5%
Non-profit organisations 29.3% 22.2% 4.0%
Foreign
Ecuador Government 39.8% 90.6% 45.1% 61.9%
Enterprises 32.5% 9.1% 4.8%
Higher Education 38.2% 16.1%
Non-profit organisations 4.9% 0.5% 7.8% 17.2%
Foreign 22.9% 8.9%

54
Table 2 Cont.
El Government 51.9% 41.7%
Salvador Enterprises 1.2%
Higher Education 13.2% 43.7%
Non-profit organisations 10.4% 14.6%
Foreign 23.4%
Mexico Government 66.2% 59.1% 63.6% 44.0% 31.0%
Enterprises 17.6% 24.3% 19.0% 26.3% 10.3%
Higher Education 8.4% 10.8% 7.7% 26.2% 57.8%
Non-profit organisations 1.1% 0.1% 0.6% 3.5% 0.9%
Foreign 6.7% 5.6%
Panama Government 45.5% 34.4% 43.0% 62.2% 41.0% 64.3%
Enterprises 0.5% 0.6%
Higher Education 0.9% 0.4% 8.2% 7.1% 45.8% 19.9%
Non-profit organisations 1.1% 0.7% 48.8% 30.6% 13.2% 15.7%
Foreign 52.0% 64.1%
Paraguay Government 51.1% 36.4% 28.5%
Enterprises 3.9%
Higher Education 4.0% 19.3% 46.2%
Non-profit organisations 0.8% 44.4% 25.3%
Foreign 40.1%
Uruguay Government 15.0% 6.1% 20.3% 18.5% 25.0% 5.0%
Enterprises 58.0% 31.1% 39.3% 31.2% 39.3% 5.0%
Higher Education 27.0% 50.3% 35.7% 50.3% 35.7% 90.0%
Non-profit organisations
Foreign 12.5% 4.8%
Venezuela Government 47.5% 46.2% 51.0%
Enterprises 37.3% 30.2% 32.8%
Higher Education 15.2% 23.6% 16.1%
Non-profit organisations
Foreign
Total Latin Government 65.2% 56.3% 56.7% 18.6% 24.6% 22.5 24.3
America Enterprises 26.1% 34.3% 32.6% 37.9% 36.2% 11.2 11
and Higher Education 5.7% 6.9% 8.3% 40.9% 37.5% 62.3 59.4
Caribbean Non-profit organisations 1.3% 0.8% 0.4% 2.6% 1.6% 4.7 9.5
Foreign 1.7% 1.7% 2.0%
Sources: adapted from RICYT, 2002

55
Table 3: Expenditure on R&D by Socio-Economic Objective, 2000
El Trinidad
Argentina Brazil Chile Cuba Ecuador Salvador Mexico Panama Paraguay Peru Tobago Uruguay
Agriculture, forestry 17.0% 11.3% 23.1% 23.9% 44.0% 19.2% 22.0% 39.8% 50.5% 26.4% 73.0% 31.7%
and fishing
Industrial 26.7% 1.5% 6.3% 29.4% 4.8% 5.4% 20.0% 5.1% 3.2% 29.0% 7.7% 6.7%
development
technology
Energy 1.8% 2.1% 3.2% 11.5% 0.3% 0.8% 7.4% 1.3% 0.3% 6.9% 2.4% 0.2%
Infrastructures 2.0% 0.4% 5.8% 2.1% 0.9% 2.8% 5.3% 6.0% 0.7% 13.0%
Environment 4.2% 0.2% 10.9% 16.1% 21.6% 11.8% 9.0% 0.5% 4.2%
Health (excluded 14.0% 9.1% 20.0% 5.3% 7.3% 12.4% 6.1% 2.8% 1.6% 3.9%
pollution)
Social development 5.7% 0.1% 0.9% 6.9% 29.3% 6.1% 10.0% 20.2% 0.0%
and services
Earth and atmosphere 4.7% 1.0% 14.5% 12.8% 2.6% 1.0% 0.6% 1.9% 2.3%
General promotion of 15.6% 71.4% 46.2% 8.8% 11.1% 4.3% 21.2% 18.3% 33.5% 45.1%
knowledge
Civil space 1.7% 2.6% 0.3% 0.2%
Defence 0.8% 0.4% 0.5% 0.2% 0.4%
Not specified 5.8% 1.3% 9.7% 0.6% 3.6% 9.8%
Source: RICYT, 2002

56
Table 4: Number of Researchers and Distribution by Scientific Field, 2000 or nearest year

Argentin Bolivia Brazil Chile Colombia Ecuador El Mexico Panama Paraguay Trin.& Uruguay Venezuel LAC
a Salvador Tobago a
Natural Sciences 27.4% 27.0% 26.0% 30.9% 24.5% 27.3% 14.3% 6.5% 26.7% 8.5% 42.4% 31.3% 21.7% 26.2%
Engineering and 18.7% 20.0% 16.2% 13.6% 21.2% 18.2% 8.0% 17.2% 10.8% 6.3% 17.4% 16.3% 13.1% 16.4%
Tech.
Medical Sciences 12.8% 24.0% 19.7% 14.6% 13.6% 14.6% 13.4% 12.6% 11.6% 26.2% 8.4% 14.3% 34.9% 18.4%
Agr-Sciences 12.6% 15.0% 11.8% 10.8% 7.4% 28.4% 9.7% 2.8% 21.0% 30.9% 22.1% 16.7% 9.0% 16.5%
Social Sciences 15.5% 10.0% 16.4% 20.1% 29.1% 8.3% 47.8% 58.5% 23.4% 26.3% 9.7% 17.5% 21.4% 25.3%
Humanities 13.0% 4.0% 9.8% 10.0% 4.1% 3.3% 6.8% 2.4% 6.5% 1.9% 3.9% 5.5%

N0. researchers 35,015 1,050 77,822 6,105 4,987 1,422 1,172 26,479 446 543 547 2,513 4,756 162,800*
Researc/workforce 2.2% 0.33% 0.98% 1.04% 0.27% 0.25% 0.46% 0.75% 0.4% 0.2% 0.98% 1.65% 0.46% 0.89
Source: RICYT 2002
* The number of researchers for LAC in this table was calculated by adding the number of researchers in the countries included in the table only.
According to Ricyt 2002 (Indicator 12, S&T Personnel) the total number of researchers in LAC is 240,427.

57
Table 5: Formal qualification of researchers, 2000
El Trinidad &
Argentina Bolivia Brazil Chile Colombia Ecuador Salvador Mexico Panama Paraguay Tobago Uruguay Venezuela
PhD 25.8% 20.0% 56.7% 12.1% 5.0% 3.0% 6.0% 4.7% 4.2% 35.1% 17.7% 21.4%
MSc 7.4% 35.0% 29.5% 25.3% 26.0% 77.0% 24.0% 16.1% 26.1% 53.0% 18.1% 12.3%
Univ. degree or equiv 63.9% 40.0% 13.6% 62.6% 69.0% 20.0% 64.0% 28.6% 30.7% 11.9% 17.0% 27.9%
Tertiary not- 5.0% 30.3% 24.0% 12.0% 16.8%
university
Others 2.9% 0.1% 6.0% 20.4% 15.0% 35.3% 21.7%
Source: RICYT 2002

58
Table 6: R&D Personnel by Gender
1995 2000
Argentina Female 41.5% 48.1%
Male 58.5% 51.9%
Bolivia Female 39.0%
Male 61.0%
Brazil Female 38.6%
Male 61.4%
Colombia Female 34.4% 37.6%
Male 65.6% 62.4%
Ecuador Female 25.0% 31.5%
Male 75.0% 68.6%
El Salvador Female 37.3%
Male 62.7%
Honduras Female 33.4%
Male 66.6%
Panama Female 26.5% 39.5%
Male 73.5% 60.5%
Paraguay Female 49.9%
Male 50.1%
Trinidad & Tobago Female 38.4%
Male 61.6%
United States Female 19.0% 20.6%
Male 81.0% 79.4%
Uruguay Female 41.6%
Male 58.4%
Venezuela Female 42.0%
Male 58.0%
Source: Ricyt, 2002

59
Table 7. Number of Publications of LAC (and % World Output), according to three
international databases

SCI PASCAL INSPEC

1990 19955 2000 1990 1995 2000 1990 1995 2000

485171 858970 988156 488344 460586 511617 259781 302220 335089
TOTAL

LAC 11046 17072 28657 6994 7138 13555 2586 4252 6682

% of 1,67 1,98 2,9 1,43 1,55 2,65 0,99 1,41 1,99

World

Source: Ricyt, 2002

60
Table 8: LAC countries number of publications (SCI-SEARCH) and per 100,000
population

1990 1995 2000

n.publ p/100,000p. n.publ p/100,000p. n.publ p/100,000p.
Argentina 2 343 7.3 3 159 9.3 5 101 14.3
Barbados 25 9.6 47 17.8 39 14.6
Bolivia 24 0.4 62 0.8 75 0.9
Brazil 3 885 2.7 6 727 4.3 12 655 7.8
Chile 1 220 9.3 1 629 11.8 2 277 15.5
Colombia 208 0.6 358 0.9 733 1.7
Costa Rica 146 4.9 177 5.3 223 5.9
Cuba 223 2.1 355 3.2 646 5.8
Ecuador 52 0.5 94 0.8 136 1.1
El Salvador 9 0.2 4 0.1 17 0.3
Guatemala 49 0.6 57 0.6 64 0.6
Haiti 11 0.5 7 1.3 11
Honduras 21 0.2 17 0.1 25
Jamaica 344 0.4 288 0.3 255 0.4
Mexico 1 705 14.5 3 261 11.5 5 190
Nicaragua 9 2.1 12 3.6 26 5.4
Panama 148 0.2 126 0.3 161 0.5
Paraguay 13 6.2 17 4.8 30 5.4
Peru 163 0.3 177 0.4 228 0.5
Dominican 22 0.7 14 0.7 49 0.9
Republic
Trin.&Tob 65 0.3 94 0.2 107 0.6

Uruguay 107 5.3 201 7.5 351 8.4

Venezuela 519 3.8 736 7.0 1 170 10.6
LAC 11 046 2.7 17 072 3.4 28 344 4.9
Source: Ricyt 2002

61
Table 9. LAC output in the SCI (Search), co-authored with researchers from the USA & Canada, European Community and other LAC
countries
1993 1995 1997 1999
SCI collaboration with SCI collaboration with SCI collaboration with SCI collaboration with
# total USA & AL&C EC # Total USA & CA AL&C EC # Total USA & CA AL&C EC # Total USA & CA AL&C EC
CA
Argentina 2476 12.1 4.6 11.9 3159 9.2 5.3 12.2 4262 10.7 5.9 12.7 4874 11.4 7.8 15.6
Barbados 42 38.1 2.4 7.1 47 29.8 12.8 8.5 50 34.0 2.0 16.0 46 32.6 4.4 8.7
Bolivia 65 21.5 3.1 33.9 62 27.4 11.3 59.7 86 15.1 17.4 43.0 105 19.1 35.2 60.0
Brazil 4908 15.6 2.5 14.9 6727 15.9 2.8 14.9 8972 13.7 3.8 13.7 11729 14.7 5.0 13.6
Chile 1404 15.9 5.0 14.9 1629 17.2 7.1 17.3 1770 14.3 10.5 19.7 2078 18.1 10.9 22.1
Colombia 237 32.1 15.2 14.8 358 32.7 19.0 21.5 545 32.3 16.3 18.5 608 35.9 19.9 24.5
CostaRica 173 24.9 7.5 9.8 177 31.6 10.2 23.7 281 28.5 12.5 12.8 220 36.8 11.8 25.9
Cuba 284 2.5 3.2 14.8 355 5.4 11.3 18.9 435 5.8 16.3 23.2 682 5.3 20.1 23.0
Ecuador 48 35.4 16.7 31.3 94 31.9 7.5 26.6 115 33.0 9.6 33.0 103 35.0 26.2 31.1
El Salvador 11 27.3 100.0 9.1 4 50.0 0.0 0.0 14 28.6 35.7 7.1 6 16.7 33.3 16.7
Guatemala 74 41.9 35.1 2.7 57 64.9 40.4 1.8 64 53.1 26.6 15.6 69 55.1 46.4 13.0
Honduras 17 52.9 17.7 11.8 17 70.6 17.7 0.0 27 51.9 0.0 14.8 26 30.8 26.9 23.1
Jamaica 326 62.0 2.2 8.3 288 62.9 1.4 11.5 269 57.3 3.4 13.4 292 60.3 1.4 1.7
Mexico 2497 16.1 3.6 11.7 3261 18.1 5.1 12.6 4129 17.0 4.9 12.5 4942 20.6 6.0 15.9
Nicaragua 18 44.4 0.0 22.2 12 66.7 41.7 25.0 30 20.0 20.0 26.7 25 44.0 16.0 12.0
Panama 88 35.2 27.3 10.2 88 77.3 23.9 19.3 106 80.2 15.1 16.0 131 72.5 20.6 28.2
Paraguay 12 33.3 0.0 25.0 17 17.7 5.9 11.8 20 15.0 0.0 25.0 23 39.1 4.4 69.6
Peru 106 53.8 14.2 26.4 177 39.6 13.6 25.4 173 43.9 14.5 20.2 175 41.1 20.6 30.9
TyTobago 99 18.2 3.0 25.3 94 23.4 6.4 11.7 84 21.4 1.2 19.1 98 20.4 2.0 23.5
Uruguay 161 11.2 13.7 11.8 201 11.9 15.9 19.9 293 18.1 24.2 17.4 353 18.4 25.8 24.4
Venezuela 257 56.0 10.1 45.1 736 20.2 3.7 17.7 972 16.4 5.6 17.2 1077 22.2 6.4 17.1
Analysis performed by A.Roa-Atkinson specially for this article
Explanatory notes:
- European Community included: France or Germany or Italy or England or Denmark or Belgium or Netherlands or Portugal or Greece or Luxembourg or Ireland or Spain.
- USA and Canada: Canada or any state of USA (NY or CA or MA or AZ or CO or CT or DC or Fl or Ga or IL or KS or LA or MD or ML or NC or NJ or OH or TX or UT or VA or WA or
WI)
- AL&C included papers co-authored between (X country) and (the another LA countries)
For the purpose of this table, the total output was calculated individually country by country by listing all the papers whose address contained the country
name.

62
Table 10: Regional and LAC countries portfolio of scientific articles, by field: 1999 (Percentages)

All Fields All Fields Clinical Biomedical Biology Chemistry Physics Earth Engin. & S.Sciences Profess.
Region and country Number percent medicine research & Agric. & Math & Space technology & Psyc. Fields
Worldwide 528,643 100.00 30.0 14.7 7.0 12.5 17.3 5.4 6.8 4.7 1.8
North America (US+Canada) 183,211 34.66 33.4 16.9 6.7 7.7 11.9 6.2 6.0 7.6 3.6
Western Europe 188,548 35.67 33.1 14.6 6.8 12.4 17.1 5.5 5.8 3.8 1.0
Asia 86,405 16.34 23.8 12.7 5.8 18.8 23.0 3.2 11.1 1.1 0.4
South Korea 6,675 1.2 16.5 9.1 3.4 20.8 25.2 2.4 20.9 1.0 0.7
Taiwan 5,655 1.0 23.8 8.9 5.3 18.3 17.7 4.3 19.4 1.3 1.1
Near East and North Africa 9,086 1.72 28.7 9.9 7.2 16.0 20.0 3.9 9.3 3.5 1.3
Sub-Saharan Africa 3,632 0.69 34.9 11.0 22.0 6.8 7.2 7.5 2.8 5.8 1.8
Latin America 12,034 2.28 25.1 13.9 13.2 12.2 21.9 6.0 5.4 2.0 0.4
Argentina 2,361 0.45 24.3 13.5 16.1 14.0 20.4 5.2 4.6 1.9 0.1
Barbados 15 0 19.3 2.7 38.7 8.7 20.0 0.0 0.0 10.0 0.0
Bolivia 33 0 45.0 6.1 36.1 0.9 1.8 3.7 1.5 5.2 0.0
Brazil 5,144 0.97 24.6 14.8 10.3 11.9 25.4 4.7 6.2 1.7 0.4
Chile 879 0.17 33.7 13.1 14.2 11.8 11.7 9.9 3.3 1.9 0.4
Colombia 207 0.04 27.1 10.8 19.3 9.6 21.5 3.5 5.3 2.8 0.2
Costa Rica 69 0.01 29.2 8.9 36.8 8.6 8.1 4.7 3.4 0.4 0.0
Cuba 192 0.04 27.3 17.9 10.5 20.1 15.6 2.2 5.7 0.7 0.0
Dominican Republic 6 0.00 55.4 10.7 16.1 0.0 0.0 0.0 0.0 17.9 0.0
Ecuador 20 0.00 49.0 6.6 16.7 1.5 6.1 15.2 1.5 0.0 3.5
El Salvador 0 0 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Guatemala 14 0.00 44.2 28.3 15.9 3.6 0.0 0.0 0.0 7.9 0.0
Guyana 4 0.00 45.2 7.1 11.9 0.0 11.9 23.8 0.0 0.0 0.0
Haiti 1 0.00 100.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
Honduras 110 0.02 62.3 9.4 23.6 0.0 0.0 0.0 0.0 4.7 0.0
Jamaica 44 0.01 33.6 12.6 11.0 17.6 2.3 8.5 2.3 9.6 2.7
Mexico 2,910 0.55 22.6 12.4 13.5 10.7 23.8 8.4 5.8 2.4 0.5
Nicaragua 8 0.00 43.2 21.0 18.5 0.0 0.0 12.3 3.7 0.0 0.0
Panama 370 0.07 9.3 18.3 51.5 0.0 6.0 4.6 2.7 7.6 0.0
Paraguay 4 0.01 80.0 0.0 20.0 0.0 0.0 0.0 0.0 0.0 0.0

63
 Peru 56 0.01 52.1 10.3 14.0 1.8 5.5 8.2 5.0 2.3 0.9
Trinidad-Tobago 37 0.01 41.2 4.3 33.2 9.2 0.0 1.3 5.4 2.7 2.7
Uruguay 144 0.03 28.2 18.9 17.4 12.6 16.9 4.4 1.0 0.7 0.0
Venezuela 448 0.09 18.7 14.3 12.8 18.1 17.8 8.8 7.1 1.9 0.6
Source: NSB 2002, Appendix table 5-43
It is important to note that although data in this Table 9 and in Table 8 are both based on SCI database, they show different figures concerning the number of publications for LAC
countries. The reason is that the NSF uses a fixed set of journals indexed by the SCI since 1985 and Ricyt searches in all journals in the SCI database.

64
Table 11: Citations by world's scientific papers to scientific literature, by region and
country/economy: 1990 and 1999
(All fields)
1990 1999
Cited Number World share Number
region/country/economy (%) World share
(%)
Total 2,098,342 100.00 2,749,022 100.00
North America 1,181,861 56.32 1,360,447 49.49
Western Europe 629,039 29.98 939,901 34.19
Asia 153,294 7.31 263,941 9.60
Pacific 49,881 2.38 64,051 2.33
Eastern Europe 42,145 2.01 56,488 2.05
Near East and North 21,605 1.03 28,854 1.05
Africa
Sub-Saharan Africa 8,303 0.40 8,466 0.31

Latin America 12,214 0.58 26,874 0.98

Brazil 3,437 0.16 10,197 0.37
Argentina 3,136 0.15 5,691 0.21
Mexico 2,243 0.11 5,103 0.19
Chile 1,472 0.07 2,384 0.09
Venezuela 750 0.04 1,010 0.04
Colombia 229 0.01 512 0.02
Costa Rica 117 0.01 245 0.01
Cuba 110 0.01 389 0.01
Jamaica 114 0.01 153 0.01
Panama 124 0.01 217 0.01
Peru 166 0.01 139 0.01
Uruguay 82 0.00 383 0.01

Source: NSB, Science and Engineering Indicators 2002, Appendix table 5-50

65
Table 12. Patents applications in LAC countries and Patents granted in the US to LAC
residents

1990 1995 2000

Argentina residents 955 676 1062
not residents 1955 3588 5574
Total 2910 4264 6636
USPTO 19 32 63
Bolivia residents 21 30 30
not residents 53 95 245
Total 74 125 275
USPTO 0 0 2
Brazil residents 6619 7234 8807
not residents 6125 8559 10518
Total 12744 15793 19325
USPTO 45 70 113
Chile residents 161 315 407
not residents 637 1847 3276
Total 798 2162 3683
USPTO 2 7 16
Colombia residents 0 141 0
not residents 0 1093 0
Total 0 1234 0
USPTO 6 5 11
Costa Rica residents 25 0 0
not residents 41 0 0
Total 66 0 0
USPTO 2 10 8
Cuba residents 187 104 149
not residents 73 33 160
Total 260 137 309
USPTO 0 0 3
Ecuador residents 0 69 54
not residents 0 276 494
Total 0 345 548
USPTO 2 0 0
El Salvador residents 6 20 28
not residents 48 72 217
Total 54 92 245
USPTO 0 1 0
Guatemala residents 27 32 54
not residents 97 77 250
Total 124 109 304
USPTO 2 0 2
Haiti residents 0 0 0
not residents 0 0 0
Total 0 0 0
USPTO 0 0 0
Honduras residents 6 6 7
not residents 17 18 94
Total 23 24 101
USPTO 1 1 1
Jamaica residents 4 7 0
not residents 77 54 0
Total 81 61 0
USPTO 1 2 2
Mexico residents 661 432 431

66
 not residents 4400 4961 12630
Total 5061 5393 13061
USPTO 34 45 100
Nicaragua residents 0 0 0
not residents 0 35 0
Total 0 35 0
USPTO 0 0 0
Panama residents 6 15 0
not residents 62 63 0
Total 68 78 0
USPTO 1 1 2
Paraguay residents 16 12 11
not residents 36 88 207
Total 52 100 218
USPTO 1 0 0
Peru residents 46 23 40
not residents 222 524 1038
Total 268 547 1078
USPTO 5 3 3
Dominican residents 0 3 8
Republic
not residents 0 53 159
Total 0 56 167
USPTO 1 1 5
Trinidad & residents 17 24 6
Tobago
not residents 46 63 148
Total 63 87 154
USPTO 1 0 0
Uruguay residents 169 122 34
not residents 156 281 588
Total 325 403 622
USPTO 3 2 1
Venezuela residents 262 258 175
not residents 1090 1784 2743
Total 1352 2042 2918
USPTO 20 31 32
LAC residents 9324 9560 11510
not residents 15383 23692 40858
Total 24708 33252 52368
USPTO 146 211 364
Source: RICYT 2002 and US PTO (http://www.uspto.gov/web/offices/ac/ido/oeip/taf/apat.pdf)

67
Table 13. Government Institutions for Science and Technology Policy Formulation and Resource Allocation in LAC

S&T Policy Institutions Research Councils Government Policies to foster R&D activities and innovation in the private
Country and Funds sector
•
Argentina Secretariat for Science, Technology National Research Law 23.877 in 1990 with the objective to fund collaborative research between the
and Innovation (linked to the Ministry Council: CONICET public and private sectors
for Education, Science and (ANPCYT; FONTAR • Establishment of the “units of links” (Unidades de vinculación) for the
Technology) FONCYT identification, selection and formulation of R&D collaborative programmes
Bolivia National Secretariat for S&T/Comité National Research • A research fund for Science and Technology as well as an Innovation Law have
Ejecutivo de la Universidad Boliviana Council: (CONACYT) been approved but are not operational, as financial resources have not been
(SICYT-CEUB) allocated
Brazil Ministry of Science &Technology National Research • Programme to support Industrial Technological Capability (PCTI) with various
Council (CNPq), components: Fiscal Incentives to the Scientific and Technological Development of
Main advisory committee: Council of Financier of Projects Industry (PDTI) and the same for Agriculture (PDTA); Support to technological
Science and Technology (CCT) (FINEP), Agency for innovation in micro and small industries: Project Alfa ; Support to
Graduate Education university/industry co-operative projects. Project Omega; National Programme for
(CAPES); State ;
the Support of Enterprise Incubators (PNI) Program of Technological
Foundations (FAPs) Management for Competitiveness (PGTec)
• Sectoral funds
Chile National Research Council (linked to National Research • FONDEF – a scheme to support collaborative research between the public and
National Office of Planning – Council (CONICYT) private sectors (maximum 50% funding from government)
ODEPLAN- which is linked directly • FONTEC – to increase quality and quantity of R&D and the provision of scientific
to the Presidency of the Republic) services with impact on productive activity (at least 20% contribution from private
National Government level firms)
• FIA, FIM, FIP – the same as above, for the agricultural sector, mineral resources
and fishing activities, respectively
Colombia Ministry of Planning (Departamento Secretary of S&T • Fondo Nacional de Garantias (SMEs) to foster technological upgrade and
de Planeacion Nacional) (Colciencias) competitivity of small firms and thus increase their sustainability
National Research Council: CNCyT,
Costa Rica Ministry for Science and Technology National Research • government schemes to foster links between universities and private firms – ex:
– MICIT Council: CONICIT Alliance INTEl + Technological Institute of Costa Rica (ITCR)
• programme of technological incubators – ex: Industrial Park of Cartago
Cuba Ministry for Science and Technology Cuban Academy of No information
and Environment Sciences
Ecuador SENACYT (liked to the Vice- National Research • FUNDACYT – funds one scheme for collaborative research and SMEs
president of Republic) Council: CONICIT • no information
El Salvador Ministry of Economics Department of Funding • no information
National Research Council: for Development of S&T
CONACYT

68
Guatemala National Research Council: National Fund for S&T • No information
CONICYT (FONACYT)
Guyana Council for Science, Technology and National Research • No information
Environment (linked to Presidency of Council: CONICYT
the Republic)
Honduras Council for Science, Technology and National Research • UNIDO facilitating scheme with Dutch Government Funds to support
Environment (linked directly to Council: COHCIT technological upgrade of SMEs
Presidency of the Republic) • CERTEC: a foundation established to institutionalise the earlier UNIDO
programme achievements, pass lessons learned and methodologies
Jamaica National Research Council: NCST • UNDP support to JAMPRO, government development agency with a scheme to
foster competitiveness of SMEs by technological upgrade
Mexico Secretariat for Public Education National Research • Schemes to foster “knowledge spaces” at regional level, integrating Public
Council: CONACYT Research Institutes with local university, SMEs and large corporations
• Establishment of science parks
Nicaragua Nicaraguan Council on Science, National Research • Research council operates with IDB loan to foster technological innovation in
Technology (linked to Ministry of Council: COHCIT agriculture and forestry
Industry and Commerce) • UNIDO facilitating programme to foster technological upgrade of SMEs and to
articulate network strategies (clustering)
Panama National Secretariat for Science and National Research • City of Knowledge: created under Law #6 of 2000, to promote the generation,
Technology - SENACYT (linked Council: CONACYT dissemination and use of knowledge (funds from IDB)
directly to Presidency of the Republic) (manages FONACIT)
Paraguay National Research Council: National Fund for S&T • No information
CONACYT (linked to Presidency of (FONACYT)
the Republic)
Peru National Research Council: National Fund for • No information
CONCYTEC (linked to Ministry of Development of S&T
Education) (FONDECYT)
Trinidad & National Institute of Higher Education, • No information
Tobago Research, S&T- NIHERST (linked to
Presidency of the Republic)
Uruguay National Research Council: DINACYT- OPP (Office • FINTEC – scheme to foster technological innovation at the firms level
CONICYT of Planning and Budget - Sectoral Commission of Scientific Investigation (CSIC) at the University of Republic
(linked to the Ministry of Education National Directorate of funds a scheme for collaborative research between university and the private sector
and Culture) S&T and Innovation)
Venezuela Ministry of Science and Technology National Research • FONACIT: National Fund for Science, Technology and Innovation support for a
Council: CONICIT programme for organisational updating and strengthening of technological
(manages FONACIT capacities of private firms
• PIN Industrial: schemes to insert young researchers in work positions in private
firms
Sources: elaborated by the author from multiple sources: Vonortas (2002); Velho & Saenz (2002), Velho et al (1998), Ricyt (2002), Hansen et al

69
Table 14. Science and Technology Performing Organisations in Latin America and the Caribbean

Country S&T Performing Organisations

Argentina • Higher education: 36 public and 43 private universities
• Public research institutes in: health, atomic energy, agriculture, environment and industrial development (regional and national level).
E.g. INTA with 46 research centres in agriculture distributed in 18 regional centres.
Bolivia • Higher education: 12 public universities
• Public research institutes: 14 (4 in agriculture, 2 environment, 1 education, 1 indigenous knowledge, 1 health)
• NGOs: 5 (3 social science research, 2 indigenous knowledge and 1environment)
Brazil • Higher education: 150 (77 Public, 76 private) research universities; 820 non-university institutions (16% public and 84% private) for professional
education.
• Public research institutes and State-owned enterprises: 22 (1 health, 1 environment, 1 education, 1 indigenous knowledge/Amazonian, 1 agriculture, 2
health, oil, mineral)
• 15 professional and national not-for-profit organisations (promotion, co-ordination and services for agriculture, SMEs and technical training)
Chile • Higher education: 67 universities, 72 professional education and 128 for technical training (23 universities carry out research)
• Public research institutes: 7 (health, environment, education, indigenous knowledge, agriculture, mineralogy)
• NGOs and professional not-for-profit institutions (150)
• Stated-owned enterprises: (1 Defence)
Colombia • Higher education: 70 public, 132 private and 119 non university institutions
• Public research institutes: 18 (environmental; health; biotechnology; education; agro industrial)
• NGOs:17
• private sector: 7; Technological development centres: 4
Costa Rica • 146 institutions dedicated to R&D: 100 in education; 16 in general services; 11 in the private sector; 3 regional and international
• 2 co-operatives and foundations.
Cuba • Higher education: 62 public universities (50 universities are active in R&D)
• 200 research centres (agriculture and livestock; biotechnology and development of vaccines and pharmaceuticals; Health; industrial activities; sugar
plantation; biodiversity and environment; social and economic problems)
• Science and production poles: 14 territorial networks
Ecuador • Higher education: 31 institutions with 150 R&D units
• Public sector = 12 R&D units (5%)
• Private sector = 46 R&D units (20%)
• NGOs = 17 units of R%D (8%)
El Salvador • The institutions with major participation in S&T activities are from the governmental sector. There is little participation from the higher education
sector, private sector or NGOs
Guatemala • Higher education: 1 public and 6 private universities
• Public research institutes: 3 in agriculture, 2 in environment

70
 • Private sector: 1 Indigenous knowledge, 1 SMEs support
Guyana • Higher education: 1 public and 6 private universities
• Public research institute: 1 Guyana's Agriculture Information Network (13 agricultural research centres)
Honduras • Higher education: 3 public universities
• Public research institutes: 3 institutions in agriculture, 1 in Public Health, 1 in water quality and management
Jamaica • Higher education: 1 public university
• Public research institutes: 3 in agriculture, 2 in Public Health, 2 in environment and water, 2 in industry
Mexico • Higher education: 1140 universities ( 35% public and 65% private) and 393 non-university (28% public and 70% private) – most R&D concentrated
in 3 public universities
• research centers linked to public universities.
• Public research institutes: 29 centres SEP-CONACYT
Nicaragua • Higher education: 34 (4 public and 30 private). The public universities carry most R&D activities.
• Public research institutes: 1 environmental, 1 health, 1 education, 1 agriculture
• NGOs: agriculture, forestry, industrial associations
Panama • Higher education: 4 public universities carry most R&D activities.
• Public research institutes: 6 (1 health, 2 agriculture & livestock, 1 environment, 1 biology, 1 industry)
• NGOs: 4 (1 health, 1 economics, 2 social sciences)
Paraguay • Higher education: 6 public universities involved with R&D
• Public research institutes: agriculture, health, education, communications and industry and commerce
• NGOs: 14 (3 on environment)
Peru • Higher education: Total 78 ( 33 public and 45 private)
• Public research institutes: 1 health, 1 environment, 1 indigenous knowledge (Amazonian); 2 agriculture and fisheries
• NGOs: 1 agriculture (E.g. Cipotato)
Trinidad & • 1 public university
Tobago • 1 regional research institute
Uruguay • Higher education: 6 universities (1 public concentrates infrastructure and most research groups; and 5 private of which 3 do not carry out R&D)
• Public research institutes: 3 (agriculture, livestock and fisheries)
• State enterprises: 2 (electricity generation and supply; telecommunications)
Venezuela • Higher education: 37 university institutions (17 public, 20 private) and 94 non- universities (52% public and 48%)
• Public research institutes : 9 (1 petroleum, 2 agriculture, 2 health, 1 geology, 1 astronomy)
• NGOs: 4
Sources: own elaboration from RICyT, 2000. CDRom available at http://www.ricyt.edu.ar
- Brazil: Indicadores de C&T em S.P- 2001, FAPESP and MEC/INEP/
- Chile: http://www.cse.cl/Indices/Estadisticas/fr_estadista.htm
- Colombia: http://www.Icfes.gov.co/ Resumen Estadistico 1990-1999
- Cuba: Fte: http://www.cuba.cu/educacion
- Mexico: Estadísticas de la educación superior – 2000 http://www.anuies.mx.
- Peru: www.concytec.gob.pe http://www.iesalc.unesco.org.ve/documentosenlinea.htm

71
Table 15: Earned S&E doctoral degrees in selected regions and locations: 1999 or latest
available year

Region/location All Natural Math and Biology and Social and Engineering Non-S&E
Fields sciences computer Agriculture behavioural
sciences sciences

China 10,160 2,135 417 444 513 3,269 (32%) 3,382

India 10,408 3,498 NA 968 NA 298 (3%) 5,644
Japan 14,800 1,481 NA 1,094 420 3,580 (24%) 8,225
South Korea 5,586 375 209 180 327 1,393 (25%) 3,102
Taiwan 1,337 150 119 76 65 482 (32%) 445

Australia 3,271 816 188 182 220 398 1,467

Finland 1,708 254 77 40 235 312 (18%) 790

France 10,582 3,924 845 179 1,559 1,852 (17%) 2,223
Germany 24,545 6,271 980 522 1,982 2,229 (9%) 12,561
Spain 5,931 1,517 253 245 229 381 (6%) 3,306
United Kingdom 11,338 3,668 680 326 907 1,805 (16%) 3,952

Canada 3,347 763 204 178 562 544 1,096

United States 41,140 9,989 1,935 965 7,727 5,337 (13) 15,187

Argentina 408 218 8 97 18 41 (10%) 26

Brazil 6,042 788 NA 1,519 1,501 765 (12%) 1,469
Chile 88 63 NA 0 0 5 (6%) 7
Mexico 1,069 345 22 57 227 92 (9%) 326

Source: NSB, Science and Engineering Indicators 2002, appendix table 2-42
(modified)

73
THE UNU-INTECH DISCUSSION PAPER SERIES

# 2004-4 Science and Technology in Latin America and the Carribean: An Overview by Léa Velho
# 2004-3 Coping with Globalisation. An Analysis of innovation capability in Brazilian
telecommunications equipment industry by Sunil Mani
# 2004-2 Learning and Local Knowledge Institutions in African Industry by Banji Oyelaran-Oyeyinka,
January 2004
# 2004-1 Productivity, Exports, Skills and Technological Capabilities: A Study of Foreign and Local
Manufacturing Firms in Uganda by Rajah Rasiah and Henry Tamale, January 2004
# 2003-12 Regulation of Foreign Investment in Historical Perspective by Ha-Joon Chang, December 2003
# 2003-11 Illusory Competitiveness: The Apparel Assembly Model of the Caribbean Basin by Michael
Mortimore, November 2003
# 2003-10 The Role of Market, Trust and Government in the Development of the Information Hardware
Industry in Taiwan By Rajah Rasiah and Yeo Lin, October 2003
# 2003-9 Growth Theories Revisited: Enduring Questions with Changing Answers By C. V. Vaitsos,
October 2003
# 2003-8 Designing National Regimes that Promote Public Health Objectives By Padmashree Gehl
Sampath, September 2003
# 2003-7 FDI-facilitated Development: The Case of the Natural Gas Industry of Trinidad and Tobago.
By Lou Anne A. Barclay, September 2003
# 2003-6 Sources of Training in African Clusters and Awareness of ICTs: A Study of Kenya and Ghana
By Catherine Nyaki Adeya, September 2003
# 2003-5 The Internet Diffusion in Sub-Saharan Africa: A cross-country Analysis By Banji Oyelaran-
Oyeyinka and Kaushalesh Lal, July 2003
# 2003-4 Defining an Intellectual Property Right on Traditional Medicinal Knowledge: A Process-
Oriented Perspective By Padmashree Gehl Sampath, July 2003
# 2003-3 New Wave Technologies: Their Emergence, Diffusion and Impact. The Case of Hydrogen Fuel
Cell Technology and the Developing World By Lynn K. Mytelka, July 2003
# 2003-2 Systems of Innovation and Human Capital in African Development By Banji Oyelaran-
Oyeyinka and Lou Anne Barclay, May 2003
# 2003-1 Deregulation, Entry of MNCs, Public technology procurement and Innovation Capability in
India's Telecommunications Equipment Industry By Sunil Mani, February 2003
# 2002-10 Moving Up Or Going Back The Value Chain an Examination of The Role of Government With
Respect to Promoting Technological Development in The Philippines By Sunil Mani,
November 2002
# 2002-9 Research Capacity Building in Nicaragua:From Partnership with Sweden to Ownership and
Social Accountability By Léa Velho, October 2002
# 2002-8 R&D in the Public and Private Sector in Brazil: Complements or Substitutes? By Lea Velho
and Tirso W. Saenz, July 2002
# 2002-7 Systemic Coordination and Human Capital Development: Knowledge Flows in Malaysia’s
MNC-Driven Electronics Clusters By Rajah Rasiah, June 2002
# 2002-6 What is the ‘Knowledge Economy ’? Knowledge Intensity and Distributed Knowledge Bases
By Keith Smith, June 2002

75
# 2002-5 Internet Access in Africa: an Empirical Exploration By Banji Oyelaran-Oyeyinka and
Catherine Nyaki Adeya, May 2002
# 2002-4 Institutional Support for Investment in Ne w Technologies: The Role of Venture Capital
Institutions in Developing Countries by Sunil Mani and Anthony Bartzokas, May 2002
# 2002-3 Manufacturing Response in a National system of Innovation: Evidence from the Brewing
Firms in Nigeria By Banji Oyelaran-Oyeyinka, April 2002
# 2002-2 Prospects for the Digital Economy in South Africa: Technology, Policy, People, and Strategies
By Derrick L. Cogburn & Catherine Nyaki Adeya, April 2002
# 2002-1 TRIPs and Capability Building in Developing Economies by Rajah Rasiah, March 2002
# 2001-11Government, Innovation and Technology Policy, An Analysis of the Brazilian Experience
During the 1990s By Sunil Mani, December 2001
# 2001-10Innovation Systems, Institutional Change and the New Knowledge Market: Implications for
Third World Agricultural Development By Norman Clark, December 2001
# 2001-9 Working with the Market: The Israeli Experience of Promoting R&D in the Enterprise Sector
and the Lessons for Developing Countries By Sunil Mani, December 2001
# 2001-8 Conditions for Successful Technology Policy in Developing Countries– Learning Rents, State
Structures, and Institutions By Ha-Joon Chang Ali Cheema, December 2001
# 2001-7 Technological Change and Corporate Strategies in the Fertiliser Industry By Anthony
Bartzokas, October 2001
# 2001-6 A Framework for Policy-Oriented Innovation Studies in Industrialising Countries By Anthony
Bartzokas and Morris Teubal, September 2001
# 2001-5 Networks and Linkages in African Manufacturing Cluster: A Nigerian Case Study By Banji
Oyelaran-Oyeyinka, September 2001
# 2001-4Financial Markets and Technological Change: Patterns of Technological and Financial Decisions
by Manufacturing Firms in Southern Europe By Anthony Bartzokas, August 2001
# 2001-3 Role of Government in Promoting Innovation in the Enterprise Sector An Analysis of the
Indian Experience By Sunil Mani, April 2001
# 2001-2 Government and Innovation Policy an Analysis of the South African Experience since 1994
By Sunil Mani, February 2001
# 2001-1 Firm Size, Technological Capabilities and Market-Oriented Policies in Mauritius By Ganeshan
Wignaraja, February 2001

76