8/3/2019 DeBrito 2010

1/26

University Industry relations in Iberoamerican countries

Carlos Henrique de Brito Cruz

Scientific Director, FAPESP

Article for Seminario ADI 2010Educacin para el Desarrollo en Amrica LatinaPalacio de Pedralbes, Barcelona

November 5-6, 2010

Documento elaborado para el Seminario ADI Educacin para el desarrollo en Amrica Latina, organizadopor CIDOB, CEPAL y SEGIB los das 5 y 6 de noviembre de 2010 en Barcelona. Este trabajo no ha sidosometido a revisin editorial por lo que se prohbe su citacin sin la autorizacin expresa del autor.

8/3/2019 DeBrito 2010

2/26

1 The R&D landscape in IberoamericaIberoamerica1 is home to 9.6% of the world population (Table 1) and produces 10.8% of the

global gross domestic product (GNP). However, the region accounts for only 4.8% of the

global expenditure on research and development (R&D). In Iberoamerica, the intensity of the

R&D expenditure is only 0.8% of the regional GNP, which translates to one third of the

intensity verified in the developed countries and slightly less than half of the 1.8% observed

in the European Union.

Table 1. Population and GNP data for world regions and some countries of interest for this article.( Data fromUnesco UIS, Oct 2010)

Population (millions) GNP

Quantity (Millions) % world Billion PPP % world

2002 2007 2002 2007 2002 2007 2002 2007

World 6.281,2 6.671,2 100% 100% 46.161 65.435 100% 100%

Developed countries 1.196,0 1.216,7 19,0% 18,2% 29.277 38.124 63,4% 58,3%

Developing countries (excl. less dev. countr.) 4.374,3 4.652,6 69,6% 69,7% 16.283 26.344 35,3% 40,3%

Less-developed countries 711,0 802,0 11,3% 12,0% 601 967 1,3% 1,5%

Americas 856,8 908,2 13,6% 13,6% 15.119 20.462 32,8% 31,3%

North America 321,1 337,7 5,1% 5,1% 11.415 15.060 24,7% 23,0%

United Stated of America 290,8 302,8 4,6% 4,5% 10.418 13.811 22,6% 21,1%

Latin America and the Caribbean 535,7 570,5 8,5% 8,6% 3.704 5.402 8,0% 8,3%

Brazil 179,1 190,1 2,9% 2,8% 1.329 1.840 2,9% 2,8%

Mexico 102,1 107,5 1,6% 1,6% 948 1.513 2,1% 2,3%Argentina 37,7 39,5 0,6% 0,6% 297 521 0,6% 0,8%

Chile 15,8 16,1 0,3% 0,2% 157 231 0,3% 0,4%

Colombia 41,0 44,2 0,7% 0,7% 246 375 0,5% 0,6%

Venezuela 25,2 27,6 0,4% 0,4% 202 334 0,4% 0,5%

Other LAC 134,9 145,6 2,1% 2,2% 524,1 587,7 1,1% 0,9%

Europe 794,9 802,0 12,7% 12,0% 14.374 18.861 31,1% 28,8%

European Union 483,6 491,2 7,7% 7,4% 11.681 14.613 25,3% 22,3%

Portugal 10,4 10,6 0,2% 0,2% 191 240 0,4% 0,4%

Spain 41,3 44,0 0,7% 0,7% 991 1.409 2,1% 2,2%

Germany 82,5 82,6 1,3% 1,2% 2.275 2.728 4,9% 4,2%

France 59,8 61,3 1,0% 0,9% 1.711 2.062 3,7% 3,2%

United Kingdom 59,4 60,8 0,9% 0,9% 1.671 2.047 3,6% 3,1%

Africa 857,9 962,2 13,7% 14,4% 1.680 2.527 3,6% 3,9%

Asia 3.739,7 3.964,7 59,5% 59,4% 14.330 22.717 31,0% 34,7%

Japan 127,5 128,0 2,0% 1,9% 3.417 4.284 7,4% 6,5%

China 1.288,4 1.328,6 20,5% 19,9% 3.686 7.055 8,0% 10,8%

India 1.046,2 1.134,4 16,7% 17,0% 1.760 3.092 3,8% 4,7%

Oceania 31,9 34,1 0,5% 0,5% 658 868 1,4% 1,3%

Iberoamerica 587,3 625,2 9,4% 9,4% 4.885,7 7.051,1 10,6% 10,8%

1

Iberoamerica is defined here as the sum of Latin America, the Caribbean, and Portugal and Spain. Otherdefinitions exist but the exact definition will not change substantially the main conclusions of this paper nor thenumber discussed here

8/3/2019 DeBrito 2010

3/26

8/3/2019 DeBrito 2010

4/26

8/3/2019 DeBrito 2010

5/26

4

As can be seen in Figure 2, the trend in the principal countries of the region shows two

different behaviors. For the Latin American countries the trend is towards stagnation or slight

growth, in the case of Brazil and Argentina. For Portugal and Spain the trend demonstrates a

strong growth, remarkable for the case of earlier.

The composition of the R&D expenditures according to the source of funds in some

Iberoamerican countries is shown in Table 3.

Table 3. Relative composition of R&D expenditures by funding sector in selected Iberoamerican countries. (Source:http://www.ricyt.org/(all data refer to 2008, with the exception of those related to Chile, which refers to 2004)

By source

Higher

Education

Business Foreign Government Non-Profit

Portugal 3,6% 48,1% 3,0% 43,7% 1,7%

Chile 0,8% 45,8% 8,7% 44,4% 0,3%

Spain 0,0% 45,5% 7,0% 47,0% 0,5%

Mxico 3,2% 44,6% 1,4% 50,7% 0,1%

Brasil 2,2% 43,9% 0,0% 54,0% 0,0%

Colmbia 26,5% 27,1% 4,4% 39,3% 2,8%

Argentina 4,4% 26,5% 0,6% 67,6% 0,9%

As a comparison, consider that the average for OECD countries is that 65% of the R&D

expenditures in 2008 were financed by Business. In Korea and Japan this percentage was as

high as 73 and 78%, respectively, while in the U.S. it was 67%

Figure 2. Recent trends in research and development (R&D) expenditure in the principal LatinAmerican economies. (Source: http://www.ricyt.org)

8/3/2019 DeBrito 2010

6/26

5

The limited industry participation in the funding of R&D in Iberoamerica is reflected in the

weakness of the technology sector in the region, as evidenced by the national trade

imbalances, as well as by other indicators of technological competitiveness, such as the

annual number of patents granted. The stagnation of the overall national levels of R&D

investment is at least partially attributable to the lackluster efforts of the corporate sector. In

most countries, in other regions, for which the expenditure on R&D has grown, the growth

has been driven by industry investment in research aimed at increasing competitiveness in

international markets. In Latin America, a subset of Iberoamerica, the limited push to

penetrate global markets has decreased the incentive for entrepreneurial innovation. Other

factors that have had a significant influence in Latin America, include economic instability

and changes in tax regulations, as well as limited protection of intellectual property rights

(unless there is a development strategy that justifies that).

The stylized facts regarding the funding of R&D show that the number of active researchers is

comparatively low in Iberoamerica. The number of researchers per million inhabitants is 652

for Iberoamerica as a whole, the high points being Spain and Portugal, with around 2,700

researcher per million, numbers that fall quickly for the Latin American part of Iberoamerica:

980 for Argentina, 657 for Brazil, and 305 for Mexico. These should be compared with 4,654

for North America, 2,727 for the European Union, and 1,071 for China.

Table 4. Number of patents granted by the USPTO in 2000 and in 2008 to inventors in Iberoamerica, bycountry.(Source: Science and Engineering Indicators 2010, Appendix Table 6-57).

Patents USPTO 2000 2008

Region/Country Qty % Qty %

World 157.494 100% 157.772 100%

Iberoamerica 609 0,39% 574 0,36%

Spain 270 0,17% 303 0,19%

Portugal 11 0,01% 11 0,01%

Latin America 328 0,21% 260 0,16%

Brazil 98 0,06% 101 0,06%

Mexico 76 0,05% 54 0,03%

Argentina 54 0,03% 32 0,02%

Venezuela 27 0,02% 14 0,01%

Chile 15 0,01% 13 0,01%

Colombia 8 0,01% 12 0,01%

Cuba 3 0,00% 6 0,00%

Costa Rica 7 0,00% 4 0,00%

Bahamas 13 0,01% 3 0,00%

Uruguay 1 0,00% 2 0,00%

All others 26 0,02% 19 0,01%

One of the most important indicators of technological advancement, the number of patents

filed with the U.S. Patent and Trademark Office (USPTO), provides evidence that

8/3/2019 DeBrito 2010

7/26

6

Iberoamerica faces important challenges for its competitiveness in a knowledge-based

economy. The data in Table 4 shows that in 2008 only 574 were granted by USPTO to

inventors in Iberoamerica. This quantity corresponds to 0.4% of the global total.

Still worse, the Iberoamerican countries total share of the world patents remained stagnated

from 2000 to 2008, while the share of India grew 5-fold, that of China ten-fold and South

Korea 2.3 times

Table 5. Scientific articles in S&E fields authored in Iberoamerica and some selected Iberoamerican countries.(Source: Science and Engineering Indicators 2010, Appendix Table 5-25).

S&E articles ISI 1995 2007

Region/Country Qty % Qty %

World 564.644,7 100,0% 758.142,0 100,0%

Iberoamerica 21.827,2 3,87% 47.738,9 6,30%Spain 11.315,9 2,00% 20.980,7 2,77%

Portugal 990,3 0,18% 3.424,2 0,45%

Central/South America 9.521,0 1,69% 23.334,0 3,08%

Brazil 3.436,1 0,61% 11.884,9 1,57%

Mexico 1.937,3 0,34% 4.222,9 0,56%

Argentina 1.967,2 0,35% 3.362,1 0,44%

Chile 888,6 0,16% 1.740,2 0,23%

Venezuela 437,3 0,08% 497,2 0,07%

Colombia 162,3 0,03% 489,3 0,06%

Cuba 167,8 0,03% 243,5 0,03%

Uruguay 103,4 0,02% 215,1 0,03%

Peru 65,1 0,01% 153,4 0,02%

Costa Rica 65,8 0,01% 100,2 0,01%

Panama 35,3 0,01% 77,7 0,01%

Trinidad and Tobago 48,3 0,01% 67,0 0,01%

Ecuador 34,3 0,01% 66,0 0,01%

Bolivia 23,9 0,00% 50,6 0,01%

Jamaica 70,0 0,01% 49,4 0,01%

Guatemala 19,5 0,00% 22,4 0,00%

Barbados 16,5 0,00% 15,0 0,00%

Other C./So. America 42,1 0,01% 77,2 0,01%

In terms of scientific articles, Iberoamerica fares somewhat better than it does for the other

indicators, as shown in Table 5. As a region, its share in the world total of S&E articles grew

from 3.87% to 6.30% in the 12 years between 1995 and 2007, while the total of articles grew

by 119%, almost four times higher than the world average of 34% for the period. In this

period the number of articles originated from Brazil and Portugal grew by 246%, those from

Mexico by 118% and from Spain by 85%.

8/3/2019 DeBrito 2010

8/26

7

This is consistent with the composition of R&D investment in the region, which is strongly

biased toward expenditure at the state level, typically supporting basic research in academic

institutions.

There are some noteworthy exceptions within the profile described above. Industry R&D has

been responsible for some impressive achievements in Spain in the area of

telecommunications and energy, and in Brazil, especially in oil exploration and production, as

well as in aeronautics, agribusiness, and bioenergy. In Chile, there have been major findings

in biotechnology and aquaculture, whereas scientists in Argentina have made important

discoveries in the field of information technology. Still, one of the main challenges for the

countries of Iberoamerica appears to be the creation a macroeconomic environment that

would allow the private sector to become more fully engaged in technological innovation.

2 Higher Education in IberoamericaIn 2008 higher education institutions in Iberoamerican countries graduated 1.9 million

students in all fields (Table 6). Seven countries in Iberoamerica respond for 83% of the

graduates: Brasil, Mxico, Venezuela, Colombia, Argentina, Spain and Portugal, with Brazil

alone responding for 41%.

For the graduate studies, four countries dominate the formation of doctoral scientists: Brazil

(41%), Spain (28%), Mexico (16%) and Portugal (6%), totaling 91% of the doctorates

awarded in Iberoamerica.

A measure of the coverage of graduates across society is given by the ratio between the

number of graduates and the total population. This indicator varies broadly across

Iberoamerican countries. The highest values are the ones for Costa Rica, Cuba and Portugal,

for which the ration is above 6,000 graduates per million inhabitants. For Venezuela, Ecuador,

Brazil, Chile and Mexico the ratio is between 3,000 and 4,000, while for the other countries it

is below 2,400 graduates per million inhabitants.

8/3/2019 DeBrito 2010

9/26

8

Table 6. Summary data about higher education in Iberoamerican countries (Source: RICYT)

Pop

(Millions)

Graduates Grad/

Mill. Pop

%

Graduates

Masters

Awarded

%

Masters

Doctorates

Awarded

%

Doctorates

Argentina 39.75 94,909 2,388 4.9% 2.427 1,2% 746 3%

Barbados 0.28 n,d, - 0.0% 0,0% n.d. 0%

Bolivia 10.03 8,105 808 0.4% 710 0,3% 10 0%

Brazil 189.60 779,664 4,112 40.5% 36.014 17,5% 10,611 41%

Chile 16.80 57,944 3,449 3.0% 5.601 2,7% 395 2%

Colombia 44.45 96,450 2,170 5.0% 2.911 1,4% 98 0%

Costa Rica 4.40 33,201 7,546 1.7% 3.349 1,6% 80 0%

Cuba 11.20 71,475 6,382 3.7% 0 0,0% 600 2%

Dominicana 9.63 n,d, - 0.0% 0,0% n.d. 0%

Ecuador 13.90 48,911 3,519 2.5% 4.570 2,2% 53 0%

El Salvador 7.22 10,249 1,420 0.5% 468 0,2% 1 0%

Guatemala 13.70 8,542 624 0.4% 1.196 0,6% 10 0%

Guyana 0.77 n,d, - 0.0% 0,0% n.d. 0%

Haiti 8.79 n,d, - 0.0% 0,0% n.d. 0%

Honduras 7.67 9,293 1,212 0.5% 375 0,2% 0 0%

Jamaica 2.70 2,865 1,062 0.1% 0 0,0% 0 0%

Mxico 106.70 341,311 3,199 17.8% 42.477 20,7% 4,144 16%

Nicaragua 6.19 7,713 1,245 0.4% 527 0,3% 0 0%

Panam 3.40 13,169 3,873 0.7% 711 0,3% 5 0%

Paraguay 6.20 5,909 953 0.3% 1.013 0,5% 71 0%

Per 28.66 38,453 1,342 2.0% 0 0,0% 0 0%

Portugal 10.60 71,619 6,757 3.7% 11.105 5,4% 1,484 6%

Spain 46.16 90,453 1,960 4.7% 90.849 44,3% 7,302 28%

Trinidad y Tobago 1.30 2,248 1,729 0.1% 434 0,2% 19 0%

Uruguay 3.30 5,591 1,694 0.3% 397 0,2% 19 0%

Venezuela 27.93 124,706 4,464 6.5% 152 0,1% 19 0%

Total Iberoamerica 621.31 1,922,780 3,095 100.0% 205.286 100,0% 25,667 100%

One of the challenges for Iberoamerican universities is represented in the fact that none of

them appears among the top 100 of the Academic Ranking of World Universities (ARWU)

prepared annually by the Shanghai Shiao Tong University (Table 7). Only three countries -

Brazil (University of So Paulo), Argentina (University of Buenos Aires) and Mexico

(Universidad Autonoma de Mexico) have institutions ranked in the top 200. Spain has a fair

number of institutions listed, however all in the Top 300 or below.

Table 7. Classification of Iberoamerican countries according to their universities ranked among the 500 best in theAcademic Ranking of World Universities by the Shiao Tong University. (Source: http://www.arwu.org/)

Co Rk Country Top 20Top

100

Top

200

Top

300

Top

400

Top

500Region

19 Spain 4 7 10 IA

22 Brazil 1 2 5 6 IA

24 Argentina 1 1 1 1 IA

25 Mexico 1 1 1 1 IA

34 Chile 2 IA

35 Portugal 2 IA

8/3/2019 DeBrito 2010

10/26

9

In 1975 the total enrollment in universities in Iberoamerica was 3.981 million students. In

2008 this value had climbed to 15.606 million, a growth by a factor of almost four1. The

extraordinary expansion of higher education in Iberoamerica creates large opportunities for

research and for university-industry research cooperation.

One of the main limitations for the quality of higher education in Iberoamerica is the

relatively low fraction of the faculty that is active in research, shown in Table 8.

Table 8. Percentage of faculty that is active in research for some Iberoamerican contries. (Source: UnescoInstitute for Statistics)

Teaching Staff FTE Researchers in

Higher Education

% Res

Argentina 176.810 16.825 10%

Brazil 367.638 70.917 19%

Chile 54.649 0 0%

Colombia 88.337 4.964 6%

Mexico 274.618 13.569 5%

Portugal 36.069 13.114 36%

Spain 144.091 58.813 41%

Venezuela 122.525 4.015 3%

While in Spain and Portugal close to 40% of the teaching staff comprises researchers, for the

other countries in Iberoamerica this percentage is below 20%.

In Iberoamerican countries universities respond for 65% to 95% of the scientific articles

published in international journals and the 20 universities with the largest number of articles

published from 2003 to 2008 are shown in Figure 3. The 20 universities include 8 from Spain,

7 from Brazil, 2 from Portugal and one from each of Mexico, Argentina, Chile.

8/3/2019 DeBrito 2010

11/26

10

3 The case of University-Industry interactions in BrazilBrazilian policy for S&T development has been successful in many instances, even though

research as an academic endeavor is a relatively recent activity in the country. The first

research university in Brazil was created in 1934, the University of So Paulo (USP). Through

steady policies for the last 70 years the country developed a competitive system of public

universities which have a strong graduate system. The country forms ten thousand doctors

yearly. Brazilian scientists publish 17 thousand scientific articles in international journals

every year. Milestones in the state policy for S&T were the creation of the National Research

Figure 3. Number of scientific articles published by the 20 most prolific universities in Iberoamerica (Source:SCIMAGO Institutional Rankings, http://www.scimagoir.com/pdf/ranking_iberoamericano_2010.pdf).

8/3/2019 DeBrito 2010

12/26

11

Council (CNPq) and the Coordination for Training Higher Education Professors (CAPES),

both in 1950; of the Foundation for the Support of Research in So Paulo (FAPESP) in 1962

and of Studies and Projects Funding Company (FINEP) in 1970. Two Brazilian universities

appeared in the 2007 THES ranking of the 200 best world universities: University of So

Paulo (USP) and the University of Campinas (Unicamp). These two universities formed in

2009, respectively, 2,244 and 851 doctorate students, numbers hardly found anywhere around

the world.

The development of this academic infrastructure did not have a parallel in industrial

development. Although Brazil became an industrialized country in the last sixty years, the

strategy was mostly one of substituting imports through the acquisition of existing

technology. It must be said that there are some highly technological and knowledge based

companies in Brazil. The point, however, is that the idea that industrial strategy must include

intense R&D is not widespread in Brazilian industry. Less than 25% of the scientists in Brazil

work in industry, a percentage which is usually higher than 50% in developed countries.

Today, the main restriction to the intensification of university-industry cooperation in

research is due to the weak industrial R&D performance. As is well known, collaborative

R&D requires a dialogue between scientists from universities and those from industry. As the

latter do not exist in sufficient quantity, industry in Brazil, many times cannot even formulate

its challenges in terms that will lead to a productive dialogue with academic scientists.

The unbalance between academic and industrial scientists tends to create a special type of

tension in university-industry relations: industry expects to substitute its non-existent

scientists with those from universities. This strategy frequently ends in bitter disillusionment

both for industry and for universities.

On the other hand, university-industry cooperation for research (as opposed to cooperation fortechnology development) has shown some very good results in Brazil. Telecommunications,

both optical and by wire were two of the premier cases, which started still in the mid-

seventies of last century. Oil extraction in deep oceanic water is another example: Petrobrs,

the Brazilian state owned oil company developed competitive technology for drilling at 2

thousand meters under the sea with the help of a network of Brazilian universities. These and

many other good results represent cases in which the company had their own scientists and

used universities to complement their capabilities, not as a substitute. Embraer, the thirdlargest airplane manufacturer in the world was born out of the best engineering school in

8/3/2019 DeBrito 2010

13/26

12

Brazil, the Aeronautics Technology Institute (ITA) and the competitiveness of Brazilian

agriculture was built by graduates from Brazilian universities working for the state owned

Company for Agricultural Research, EMBRAPA.

It must also be considered that university-industry relations happen in a multidimensional way

and joint R&D is only one among many realizations of this interaction. Universities educate

persons who make industry successful, industry poses challenges for academic curricula,

industrial and academic scientists meet at conferences and workshops and read each others

articles and seee each others presentations.

The main change in Brazilian strategy for science and technology policy in recent years

started in 1999. After a long period focusing almost exclusively academia, the policy was

changed to include industry as a relevant actor not only as a user of knowledge but as a

creator of knowledge. The Ministry of Science and Technology was careful to legitimate this

change in policy, through a National Conference on S&T&I (Science, Technology and

Innovation). The addition of the word Innovation to the traditionally used term S&T

reflected the tendency for change. The Ministry was also smart enough to create new revenues

for funding both academic and industrial R&D, in the form of Sectoral Funds. These are

state revenues created through special legislation which taxes certain economic sectors,

mostly those that were subject to privatization of previously state owned companies2 in the

1990s. The sectoral funds have become the most important instrument for delivering direct

government support for innovation. There are currently 16 such funds in operation, which

altogether mobilize close to one billion dollars every year.

Following the National S&T&I Conference and the creation of the Sectoral Funds the federal

government prepared a proposal for the Innovation Law, which was presented to Congress in

2002. Changes in federal government after the national elections delayed the processing of the

Innovation Law by Congress, but it was finally approved by the end of 2004.

The Innovation Law (Law 10.973/2004) made possible the creation of important instruments

to support industrial R&D, the main one being the economic subvention which allows federal

money to be invested in R&D projects of private companies as a subsidy to internal R&D. For

federal universities, the Innovation Law brought the possibility of getting into agreements

with industry for joint R&D and sharing intellectual property. It is interesting to note that state

universities in the State of So Paulo (USP, Unicamp and Unesp) had already access to this

2 OECD, OECD Economic Surveys: Brazil, pp. 106 (OECD Publishing, 2006).

8/3/2019 DeBrito 2010

14/26

13

kind of relationship with industry because these institutions have a special regime of academic

and administrative autonomy granted only by the State of So Paulo to its universities. One of

the results of the Innovation Law is the formation of Technology Transfer Offices in many

universities in Brazil. Box 1 illustrates the profile of one of the first of these Agencies for

Innovation, the one created at Unicamp in 2003. A recent report2 by the National Forum of

Technology Managers (FORTEC) finds that there were in 2008 120 Technology Transfer

Offices in Brazil (the Brazilian name for those is Nucleus for technology Innovation NIT).

A poll made with 65 of the 120 NITs found that in the period 2005 to 2007 these NITs had

licensed 121 patents, at an average rate of 40 per year. The distribution is quite uneven as the

University of Campinas Agency for Innovation (Inova) licensed 28 patents in the same

period.

On the academic side, one of the effects of the Innovation Law was to foster the discussion

about the valorization of academic intellectual property. The legislation mandates universities

to organize IP offices and allows them to license and sell their IP, through publicized

procedures.

BOX 1: Inova, Unicamp Agency for Innovation, a model Technology Transfer Office for

Brazil

The Unicamp Agency for Innovation, Inova, was created in 2003, with the mission of

fostering university industry linkages through cooperative R&D, consulting and intellectual

property licensing. With a staff of 30 the Agency has already licensed more than 50 patents

and 5 non-proprietary technologies. Prior to Inova's foundation the university had as few as 8

licensed patents. Unicamp has 1,800 professors, and 30,000 students. In 2009 the university

awarded 850 doctorates, published 1,800 articles in international journals and in 2010 was

ranked 248th in the THES University Ranking.

In 2004-2009, Inova helped to close 297 cooperative R&D contracts with private companies.

Inova has applied 40 patents to the Patent Cooperation Treaty (PCT), where none were

applied before and patents applied at the Brazilian Instituto Nacional de Propriedade

Industrial (INPI) in the period 2004-2009 were 328, half of all patents deposit in Unicamp's

42 years existence. Licensing contracts include, mainly, environmental, pharmaceuticals, food

processing and nanotechnology-incorporated products.

Inova was called by FINEP, the Brazilian Innovation Agency, to help Brazilian institutions ofScience and Technology to create their own technology transfer office under the InovaNIT

8/3/2019 DeBrito 2010

15/26

14

project. In two years, 222 institutions and 645 professionals were attended in 34 courses

offered nationwide. Inova coordinates a network of seven So Paulo State institutions,

responsible for almost 50% of the Brazilian academic research, to share best practices in

technology commercialization and early stage technology assessment.

Inova also works closely with the 197 companies that spun-off from Unicamp in the last 20

years , creating an innovation ecosystem around the university and coordinating the creation

of the Unicamp Science Park.

More information at www.inova.unicamp.br.

The opening of the Brazilian economy since the 1990s and the stabilization of the economy

since 1994, the legislation offering fiscal incentives and subsidies and the recent tendency

towards economic growth created the conditions for industry to intensify its R&D investments

and this is causing an increase in the intensity of university-industry joint R&D.

As an example, the Foundation for the Support of Research in the State of So Paulo

(FAPESP) that has a program for co-funding collaborative R&D between industry and

academia (Partnerships in Research for Technology Innovation) received 28 proposals for

university-industry joint R&D in 2007 a threefold increase as compared to the yearly average

for the previous 10 years which was 14. In 2008 FAPESP received 76 proposals, however in

2009, possibly because of the economic downturn, the number of proposals submitted

dropped to 29. By July 2010 34 proposals have been received, signaling a recovery.

Representative themes for these projects include hydrolysis of lignocellulosic materials to

produce ethanol, green polymers, terabit per second optical communications, composite

materials for aeronautics, and software developments for learning and teaching. Qualitatively

speaking there is another important evolution: the projects are more exploratory and related to

basic science than they used to be five or ten years ago.

The increase in the number of proposals and the change in their nature results from a

combination of two factors: one, was the increased interest of business for joint research with

universities; the other was FAPESPs initiative to reorganize the program so that the

foundation enters into agreements with companies to publish Calls for Proposals for a period

of 5 years (one CFP per year) on a selection of themes agreed upon by FAPESP and the

company. For each agreement a Steering Committee is formed, composed by members

nominated by FAPESP and by the partner company that updates the themes list and oversees

8/3/2019 DeBrito 2010

16/26

15

the selection process, which includes the use of reviewers nominated by the partner company

among the reviewers chosen by the foundations panels. There are 13 agreements of this type

in place at FAPESP.

A difficulty that exists in Brazil, and from the data available at RICYT also seems to exist in

many other Iberoamerican countries with the possible exception of Spain and Portugal, is that

universities tend not to have a clear accounting system to demonstrate what fraction of their

income for R&D comes from university-industry research contracts. Government databases

also lack the assignment of sources and performers.

An exception is the data in which shows the funding sources for R&D at the University of

Campinas (Unicamp). Funding for research from business has ranged from 4% to 8% from

2005 to 2009, percentages that compare well with the average for U.S. universities which was

5.5% for 2008.

Table 9. Composition of the sources of funding for R&D at the University of Campinas (Unicamp)3.

Source 2005 2006 2007 2008 2009

State Government (GUF) 68% 70% 67% 71% 70%

Business 4% 4% 8% 6% 6%

Governmental Research Agencies 28% 25% 25% 22% 24%

CAPES 4% 4% 4% 4% 5%

CNPq 8% 8% 8% 6% 7%

FINEP 3% 2% 1% 1% 1%FAPESP 13% 12% 12% 11% 12%

Total 100% 100% 100% 100% 100%

4 R&D in universities and in companiesIn 1776, Adam Smith observed that the primary sources of innovation and technological

advancement were the men who worked with the machines and discovered ingenious ways in

which to improve upon them, as well as the makers of those machines, who found also ways

in which to improve their products3. The world has changed considerably since then but still

the words of Smith are echoed in the words of Joseph Miller, former Vice-President of

Research at DuPont: . . . DuPont invests more than a billion dollars a year in research and

development, employing more than 3,000 engineers and scientists, as well as 2,000 support

technicians. Two thirds of those employees work at our Experimental Station in Wilmington,

Delaware. That is where nearly all of our major discoveries are made. This incredible history

of achievement is a tribute to the political will of the company in pursuing a goal that is

3 Data from Unicamp, Anurio Estatstico 2010. Available athttp://www.aeplan.unicamp.br/anuario_estatistico_2010/index_arquivos/index.htm, Nov 30, 2010.

8/3/2019 DeBrito 2010

17/26

16

inherently unpredictable and inevitably involves a high degree of risk.4The investment made

by DuPont into research and development (R&D) corresponds to 3% of the gross income (as

opposed to the net profit) of the company.

Edwin Mansfield, of the University of Pennsylvania, conducted a study evaluating the sources

of ideas that lead to technological innovations.5 Mansfield found that academic research made

an essential and immediate contribution to less than 10% of the new products or processes

introduced by companies in the United States. In other words, 9 out of 10 innovations are

born within company walls. Mansfield stated, . . . of the new products or processes that

Figure 4. Percentage of Higher Education R&D financed by industry in some OECDcountries (Source: OECD Science, Technology and Industry: Outlook 2008).

8/3/2019 DeBrito 2010

18/26

17

could not have been developed without the aid of academic research, very few were invented

at universities; on the contrary, the majority were invented as by-products of the new

theoretical and empirical discoveries, as well as the new types of instrumentation employed in

the making of those discoveries, that came out of academic research, rather than being

invented specifically through such research itself. This seems unlikely to change. The

successful development of products or processes demands an intimate knowledge of market

details and production techniques, as well as the ability to recognize and weigh the technical

and commercial risks, a skill that comes only with direct experience in the company. This

expertise does not exist within universities, and it is unrealistic to expect that it will arise

there.6

The idea that the applied R&D necessary to the creation of technological innovation and

competitiveness must occur within companies is a concept that is still in its infancy in Brazil.

Because most of the research in Iberoamerica is conducted in the academic setting, common

sense would dictate that only universities should engage in R&D. Nevertheless, this is an

error in thinking that tends to sidetrack universities from the task that only they can do, which

is to educate the professionals that will create technology in the company environment, if

given the chance. This error also sidetracks public policies for R&D, reducing or eliminating

the policy focus on enterprise R&D and innovation and exaggerating the academic side of

those policies.

Figure 4 shows the percentage of Higher Education R&D financed by industry is some OECD

countries. This set of data is useful in setting a range of values for this fraction, which is

between 1.1% and 14% of the total expenditures for R&D at the higher education sector of

those countries. The OECD total is at 6% and the two Iberoamerican countries listed in the

data, Portugal and Spain, are, respectively, at 1.2% and 7.9%.

In many countries in Iberoamerica, much emphasis has been put on the question of university-

industry interaction, which is viewed as a magic solution that could be a substitute for the

R&D most companies do not have. Most of the time the debate seems to ignore the small

percentages shown in Figure 4, in exchange for an idealized view that expects private

financing to substitute for most of the public expenditures for higher education R&D.

University-Industry interaction has been mythologized as being a source of funding for

universities, in lieu of governmental funding, and the experience of American universities is

typically invoked. The data shown in Table 10 belie this expectation which has become

ensconced in the mythos and narrative related to science and technology in many countries.

8/3/2019 DeBrito 2010

19/26

18

Of the 51.9 billion dollars in research contracts awarded to all American universities in 2008,

only 2.87 billion (5.5% of the total) was provided by industry sources. Only four universities

listed receive more than 10% of their research funds from industry: Duke, Ohio State, Penn

State and MIT. On the other hand, the 2.87 billion that companies in the United States

invested in academic research in 2008 is 1.1% of the total that those companies invested in

R&D in the same year (nearly 268 billion). This small proportion indicates that truly essential

industry research is carried out in-house, by company scientists and engineers. The situation

depicted in Table 10 appears to be related to the organizational differences that are intrinsic to

the nature of either universities or companies.

Table 10. Total value and industry share of research contracts awarded to American universities in 2008.(Source:Science and Engineering Indicators, 2010)

Rank/academic institution

(values in MUS$)

All

sources

Federal

gov.

State/

local

gov.

Industry

Academ

ic

instituti

ons

All

other

sources

%

Industry

All institutions 51.909 31.231 3.418 2.870 10.435 3.954 5,5%

1 Johns Hopkins University, The 1.681 1.425 7 39 87 123 2,3%

2 University of CaliforniaSan Francisco 885 473 28 46 158 180 5,2%

3 University of WisconsinMadison 882 474 37 21 270 79 2,4%

4 University of Michigan, all campuses 876 593 5 41 193 45 4,7%

5 University of CaliforniaLos Angeles 871 472 23 47 186 144 5,4%

6 University of CaliforniaSan Diego 842 491 31 50 141 129 5,9%7 Duke University 767 451 19 152 91 53 19,8%

8 University of Washington 765 614 11 74 41 25 9,7%

9 University of Pennsylvania 708 482 17 51 61 97 7,2%

10 Ohio State University, all campuses 703 335 99 128 103 38 18,2%

11 Pennsylvania State Univ., all campuses 701 407 68 99 126 2 14,1%

12 Stanford University 688 509 14 60 53 52 8,7%

13 University of Minnesota, all campuses 683 364 62 29 123 104 4,2%

14 Massachusetts Institute of Technology 660 495 1 87 11 66 13,2%

15 Cornell University, all campuses 654 359 75 26 123 71 4,0%

16 University of CaliforniaDavis 643 269 56 34 215 69 5,3%

17 University of Pittsburgh, all campuses 596 456 15 12 83 29 2,0%

18 University of CaliforniaBerkeley 592 249 44 42 168 89 7,1%19 University of Florida 584 231 111 32 175 35 5,5%

20 Texas A&M University 582 246 124 43 157 12 7,4%

In society, the fundamental role of a company is production and the direct generation of

wealth, whereas the fundamental and singular mission of a university is to produce qualified

individuals. A research project will help accomplish this mission only if it contributes to the

education of students, which restricts the number of projects that will be attractive to

universities. Mansfield stated the following7: As various industry leaders have repeatedly

emphasized, one of the principal roles of the university in the process of technological change

8/3/2019 DeBrito 2010

20/26

19

is to provide well-prepared graduates. One such industry leader, the former Dean of

Research at Stanford University and renowned scientist in the area de lasers and non-linear

optics, when asked about the role that Stanford University had played in the success of Silicon

Valley, had this to say8: The myth is that the technology of Stanford was responsible for the

success of Silicon Valley.However, a survey of 3,000 small businesses found that only 20

were using technology that came, directly or indirectly, out of Stanford. What Stanford

contributed to Silicon Valley was a pool of talented and highly educated graduates.

There are some important and natural differences between research in the academic and

business setting:

Undertaking a project with the involvement of student trainees dramatically alters the timescale for the completion of that project, and speed of completion is an essential factor

from the business point of view. Doing research as a means for a better education is

completely different from doing research at the fastest possible pace.

Confidentiality is essential in a business project, whereas the open debate of results is, and

must be, the norm in an academic project.

At a university, the quest for knowledge is much less motivated by self-interest than it is

in a company. Therefore, basic research is more often conducted in an academic setting,

whereas applied research and technological development are more often carried out in a

business setting.

All of that notwithstanding, it should be borne in mind that university-industry interaction is

important for a university in that it can broaden the education of the students, and that is

reason enough to seek its expansion. Conversely, this interaction can result in the culture of

prizing knowledge being exported from the academic world to the business world.

Nevertheless, it is essential to avoid the illusion that this interaction is the solution to the

funding problems of universities and the technology problems of companies. The truth is that

the principal mechanism of interaction between a university and a company is the hiring of

university-educated professionals by the latter.

The data for the U.S. and for OECD shown in Figure 4 and Table 10 indicate the intrinsic

limitations to the depth and breadth of university-industry interaction, in terms of industry-

sponsored research contracts awarded to universities.

8/3/2019 DeBrito 2010

21/26

20

The challenge in developing university-industry relations is in identifying the overlapping

R&D interests, as shown schematically in Figure 5. However, it should be borne in mind that

there are various other modalities of interaction that can and should to be further explored.

The more widespread and least noted is the hiring in industry of professionals graduated from

universities. Somehow the debate about university-industry relations in Iberoamerica seems to

repeatedly put aside this topic, which is probably the strongest way of interaction. One might

ask how could the agribusiness or the automobile industry install and operate in Brazil, or the

oil industry in Argentina, or banking and telecommunications in Spain, to cite just a few

examples, without a steady flow of graduates from local universities.

The role of students in university-industry relations is highlighted when one considers the

creation of start-up companies by university graduates, as a result of ideas they learned or

have imagined during their time as students. It is not common for universities in Latin

America to value this kind of result and very few universities in the region have databases or

even weaker forms of register about the companies created by former students. One exception

is the University of Campinas in Brazil which promotes frequent meetings with former

students who are company creators and leaders of what they call Offsping Companies.

Figure 5. Schematic representation of the different foci for R&D activities in universities and industry,illustrating a possible overlap area. The box refers to references that assessed the origins of ideasused by industry and found that most of the time ideas come from internal sources in industry (CNIrefers to the Brazilian Confederation of Industry, NSF the U.S. National Science Foundation). In

general, universities tend to create more scientific articles than patents, while in industry thesituation reverses itself.

8/3/2019 DeBrito 2010

22/26

21

Their website informs that there are now 197 related companies to the university4. In the

European part of Iberoamerica Spanish universities show 100 spin-off companies were

created by all universities in 2008, according to the 2008 report by RedOTRI. An important

difference in policy for start-ups is that in Latin America it is infrequent for universities to

have equity in the start-ups, while in Spain and Portugal this happens more frequently.

Another type of university-industry interaction, as those of us who have worked in industrial

research will know, happens through meetings at scientific conferences, panels and

workshops. This of course requires that industry have a number of active scientists in its

payroll, and, as shall be discussed in the following sections, consists in a serious limitation for

university-industry interactions in many countries in Iberoamerica, for which the number of

scientists working for industry is very small or even negligible.

One area of special relevance is consulting, in which professors (or universities) sell part of

their work time to a company, often engaging in activities within the company walls.

Consulting by professors in Brazilian universities has been increasing for the last 10 years. Of

course, hiring a consultant only makes sense for a company that is already engaged in R&D

and requires complementary or specific knowledgefor companies without an R&D

department, consulting tends to be ineffective.

Licensing of patents is another important activity that makes university knowledge available

for use by companies. Universities in Spain licensed 95 patents in 2008. The total revenue

from licenses for those universities is close to 2.4 million Euros9. In Brazil, the total revenue

from licenses in 2008 was close to 2.1 million Euros2.

At a higher level of institutional complexity are initiatives in which universities organize

Science and Technology Parks in regions adjacent to its campi. There are numerous initiatives

of this type in Spain and Brazil. In Brazil there are 25 Science and/or Technology Parks,while other 17 are being developed. In Spain the Association of Technology/Science Parks

has 80 associates. Science Parks help to intensify university-industry interactions, facilitating

the presence of students in the companies laboratories and the presence of industry scientists

in the university departments.

A good development in some Iberoamerican countries has been the formation of networks of

TTOs, aiming at sharing their solutions and procedures. In Brazil the National Forum of

Technology Managers (FORTEC) was created in 2006 and now congregates managers from

4 Unicamp Agency for Innovation, Inova, at http://www.inova.unicamp.br/paginas/empresas_filhas.php.

8/3/2019 DeBrito 2010

23/26

22

120 TTOs. In Spain there is a RedOTRI created in 1997 by the Conference of Rectors from

Spanish Universities, congregating 65 OTRIs. In Portugal there is a Rede OTIC (Oficinas de

Transferncia de Tecnologia e Conhecimento) created in 2005 that congregates the OTICs

from all 22 Portuguese universities10.

A recent study about IP policies5 including 18 Latin-American and 3 European universities

found that IP policies at universities should consider the following points:

a) Create an environment that fosters and facilitates the dissemination of its inventions

and knowledge to benefit the public interest;

b) Protect the rights of researchers inventors;

c) Assure that any benefits accruing from the commercialization of university IP will be

shared in a way that recognizes the contributions of all stakeholders;

d) Assure that the research results are put to public use through effective and fast

technology transfer processes;

e) Foster, preserve and support research;

f) Establish standards and means to determine the rights and obligations of the

university, as well as those of the researchers-inventors and of other contributing

institutions;

g) Foster, assist and promote rewards that are mutually beneficial for the institution and

for its members;

h) Assure the enforcement of applicable norms and regulations, to guarantee the funding

at all stages of the research process.

The first point above is especially relevant: commercialization of academic IP should benefit

the public interest. It is relevant that among a large number of TTOs in Latin-America and

Europe such an agreement can exist.

5 ConclusionHigher education has experienced strong growth in Iberoamerica for the past 20 years, leading

to an increase in enrollment and graduates. However the development of post-graduate level

5 Rede PILA, Gestin de Propiedad Intelectual e Industrial en Instituciones de Educacin Superior, (Unicamp,Campinas, 2009). Available at http://www.pila-network.org/public_documents/Pila_Good%20Practice%20IP%20Management.pdf, October 31, 2010.

8/3/2019 DeBrito 2010

24/26

23

higher education has been much more uneven among Iberoamerican countries. The

innovation systems of some countries in Iberoamerica, mostly Brazil, Spain, Portugal,

Mexico, Chile and Argentina, are going through important and positive changes that tend to

associate the competitive academic base with a growing interest of industry in advanced

R&D. Universities are willing to collaborating with industry and industry understands what it

can expect from university, in both cases much more than they did ten years ago. The

challenge for universities is to stick to high academic standards and relentlessly pursue their

two track mission of education and advancement of knowledge.

There are some challenges and opportunities worth summarizing.

1) Universities must foster relations with business and society in order to better fulfill its

mission of providing education for the students and advancing knowledge. With this in

mind, good universities will find a myriad of opportunities for interaction with industry.

2) Universities are expected to take good care of the IP they generate, and make all efforts

towards getting such IP to be put to use in the public interest.

a) Requires substantial improvements in establishing incentives and reducing red-tape in

the disclosure process by faculty as well as in the evaluation of the IP portfolio and in

licensing efforts.

3) Start-up companies are an important channel through which universities can serve the

public interest disseminating its knowledge.

a) Requires better entrepreneurial education for students and easier access to seed and

venture capital.

4) Joint university-business research projects are one of the important ways for disseminating

university created knowledge and to foster new research and these projects must alsocontribute to the education of the students.

a) Requires high academic standards at universities and a sizable portion of the faculty

with a Doctoral degree; and

b) Requires intensification of business R&D in most of the Iberoamerican countries and a

simplification of processes for the preparation, negotiation and signature of joint

research contracts.

5) On the academic side the pressure for funding, especially in times of fiscal difficulties,

tends to lead to an overemphasis on obtaining funding from industry. This is a tendency

8/3/2019 DeBrito 2010

25/26

24

commonly noticed in Latin America and it creates false and thus impossible to be

fulfilled - expectations about university industry relations.

6) On the industry side, the mirage to be avoided is that of using universities as substitutes

for an R&D organization that the company wont have. The only way for industry to

obtain good results from cooperation with universities is to have its own researchers that

must be able to specify and formulate the problems they expect university researchers to

contribute to, and they must be able to interact and follow up on their universitys partner

work.

Education and the advancement of knowledge can benefit from university-industry

interactions as long as the university has a sense of purpose and a clear vision about its

objectives and about its role in society. Industry can benefit from interactions with

universities as long as they complement internal R&D capabilities that the companies have

developed, which can internalize, create and stock knowledge and connect it to the market

driven opportunities.

The fundamental challenge in university business relations is how one can complement the

other, not how one can substitute for the other. There is a role for academic research in

creating new ideas from apparently non-connected (what practical persons would callspeculative, or non-focussed) observations and a role for industrial research in solving

immediate problems, and this was beautifully summarized by Adam Smith 234 years ago:

All the improvements in machinery, however, have by no means been the inventions of those

who had occasion to use the machines. Many improvements have been made by the ingenuity

of the makers of the machines, when to make them became the business of a peculiar trade;

and some by that of those who are called philosophers or men of speculation, whose trade it is

not to do anything, but to observe everything; and who, upon that account, are often capable

of combining together the powers of the most distant and dissimilar objects6.

6 Adam Smith, "The Wealth of Nations", Book 1, Chapter 1(http://www.bibliomania.com/NonFiction/Smith/Wealth/index.html)

8/3/2019 DeBrito 2010

26/26

6 References1 CINDA, Educacin Superior en Iberoamrica - Informe 2007, http://www.cinda.cl/htm/p0607.htm

consulted on October 24, 2010.2 M.E. Ritter dos Santos, P.T.M. de Toledo and R.A. Lotufo, Transferncia de Tecnologia, (Ed.Komedi, Campinas, 2009). Available at http://www.fortec-br.org/site/documentos/ on Nov 30, 2010.

3 Adam Smith, The Wealth of Nations, (1776).

4 J. Miller, Upset the natural equilibrium, inInnovation breakthrough thinking at 3M, DuPont, GE,Pfizer and Rubbermaid, eds. Rosabeth Kanter, J. Kao and F. Wiersema (Harper Business, New York,1997).

5 E. Mansfield, Contributions of new technology to the economy, in Technology, R&D and theEconomy, eds. Bruce Smith and Claude Barfield. p. 125 (The Brookings Institutions, Washington, DC,1996).

6 E. Mansfield, Contributions of new technology to the economy, in Technology, R&D and theEconomy, eds. Bruce Smith and Claude Barfield, p. 132 (The Brookings Institutions, Washington, DC,1996).

7 E. Mansfield, Contributions of new technology to the economy, in Technology, R&D and theEconomy, eds. Bruce Smith and Claude Barfield, p. 132 (The Brookings Institutions, Washington, DC,1996)

8 Photonics Spectra, pp. 2425, April 1999

9 RedOTRI, Informe de la Encuesta 2008. Available athttp://www.redotriuniversidades.net/index.php?option=com_docman&task=doc_download&gid=848&Itemid=33&mode=view, Nov 30, 2010.

10 M.M. Godinho, L.M. da Silva and R. Cartaxo, Anlise da actividade das Oficinas de Transfernciade Tecnologia e de Conhecimento (OTIC) e dos Gabinetes de Apoio Promoo da PropriedadeIndustrial (GAPI) de mbito acadmico, a report for OTIC UTL. Available athttp://www.otic.reitoria.utl.pt/file_download/307 on Nov 30, 2010.