Science and Public Policy, 37(8), October 2010, pages 571–582                                                            D...
Knowledge demand in the South                                                                 processes. Policies for scie...
Knowledge demand in the South   If physical capital is accumulated without                       built from innovation sur...
Knowledge demand in the SouthPeople’s Republic of China, 2007); the enrollment in                            Many developi...
Knowledge demand in the SouthLall exemplifies the conscious building of these            …are uncertain in that their succ...
Knowledge demand in the SouthThe mismatch between demand and supply and, par-                      Fostering knowledge dem...
Knowledge demand in the South   People’s motivations to seek access to higher          On Latin American science, technolo...
Knowledge demand in the Southtertiary education. Instruments related to demand-        Table 2. Comparison between groups ...
Knowledge demand in the South   There is surely ample room for improvement for           cultural consequences. In particu...
Knowledge demand in the South   Another reason to consider that inclusive innova-          people. The term import substit...
Knowledge demand in the Southsuggest that weak knowledge market demand is a                            2009. Available at:...
Knowledge demand in the South   Alphonse 2007. Globalisation, Education and Development.             UNCTAD 2007. The Leas...
Copyright of Science & Public Policy (SPP) is the property of Beech Tree Publishing and its content may not becopied or em...
Upcoming SlideShare
Loading in …5

Weak Knowledge Demand in the South: Learning Divides and Innovation Policies


Published on

  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Weak Knowledge Demand in the South: Learning Divides and Innovation Policies

  1. 1. Science and Public Policy, 37(8), October 2010, pages 571–582 DOI: 10.3152/030234210X12767691861137; Weak knowledge demand in the South: learning divides and innovation policies Rodrigo Arocena and Judith Sutz Many developing countries confront serious problems in benefitting from the advancement of knowledge; a main difficulty being to expand the learning processes. Related policies have been jeopardized by weak market demand for knowledge. Both supply-side and demand-side science and technology and innovation policies have thus been below expectations. This paper argues that to reverse this long trend, current policies can profitably be complemented by a set of innovation policies conceived as social policies, which can simultaneously answer a strong social demand for knowledge and expand endogenous innovative capabilities.T ODAY’S RICH COUNTRIES ARE the knowledge-rich countries: access to knowl- edge and use of knowledge are both wide-spread, knowledge has been their ‘lever of riches’(Mokyr, 1992). That is even more important today: a technology transfer. Since then, several developing countries have improved some knowledge-related indicators, like higher education enrollment and the number of researchers. However, many of them have not, most of them being within the UN Conference‘knowledge-based and innovation-driven economy’ on Trade and Development (UNCTAD) classifica-(de la Motte and Paquet, 1996) has emerged in a set tion of ‘least developed countries’ (UNCTAD,of countries frequently termed ‘developed countries’ 2007). In general terms, despite important differ-or the ‘North’. In comparison, most other countries ences between the countries, the problem ofare more or less poor in terms of access to knowledge for development persists.knowledge and use of that knowledge. This can be Two intertwined explanations for this persistencetermed the problem of knowledge for development. deserve particular attention. The first relates to theThe causes of the problem have been explored from prospects for learning, which are of utmost im-different standpoints (Herrera, 1971; Cooper, 1973; portance if an economy wants to transformSteward, 1977; Landes, 1998; Rosenberg and knowledge into one of its fundamental resourcesFrischtack, 1985). During the second half of the 20th (Lundvall, 1992: 1). Such prospects are not brilliantcentury it was recognized as a policy problem by in developing countries. Poor countries are stuck inmost developing countries. Policy instruments were poverty because, amongst other reasons, they spe-put in place, ranging from measures to fortify the cialize in activities that are devoid of learning poten-scientific infrastructure to those aimed at stimulating tial (Reinert, 2007: xxviii). The second explanation relates to the demand for knowledge, a main driver of the process of learning: such demand is weak inJudith Sutz (corresponding author) is at the Comisión Sectorial developing countries (Cimoli et al., 2009; Rodrik,de Investigación Científica, Universidad de la República 2007). Analyzing these two issues in an interrelatedJackson 1303, Código Postal 11200, Montevideo, Uruguay; E- way can be an Ariadne’s thread leading to a bettermail:; Tel: (598-2) 408 33 09. Rodrigo understanding of the obstacles that deter a better useArocena is at the Rectorado de la Universidad de la República, of knowledge for developmental purposes.18 de julio 1824, Código Postal 11100. Montevideo, Uruguay; The analysis of this interrelatedness is based on aE-mail:; Tel: (598-2) 409 84 26. The authors wish to thank the anonymous referees and the ed- re-examination of the concept of learning and of theitor of this journal for their valuable comments on earlier sources of knowledge demand. The social processesversions of this paper. of advanced learning increasingly influence theScience and Public Policy October 2010 0302-3427/10/080571-12 US$12.00  Beech Tree Publishing 2010 571
  2. 2. Knowledge demand in the South processes. Policies for science, technology and in- Rodrigo Arocena holds a doctorate in mathematics and in development studies from Venezuela’s Central University. novation that do not take this situation into account He has been a full professor of mathematics and is now a are bound to increase such an imbalance. In the full professor of science and development at the University words of Reinert: of the Republic, Uruguay. He has developed several related graduate and postgraduate teaching programs in Uruguay and other Latin American countries. His broad field of re- …successful cases of economic development search is in development, science and technology. His cur- prove the importance of simultaneously provid- rent research interests include: the relationship between ing not only a flow of better educated people, technology and inequality in developing countries as well as the transformation of Latin American universities. Currently but also jobs where their skills are demanded. he is the rector of the University of the Republic. … Nations that only address the supply side of educated people end up educating for migra- Judith Sutz holds a doctorate in socio-economics of devel- opment from the Université of Paris 1; she is an electrical tion. (Reinert, 2007: 320–231, emphasis in the engineer and has a Master’s degree in developing planning original) from Venezuela’s Central University. She is a full professor of science, technology and society at the University of the Policies for science, technology and innovation are Republic, Uruguay and the academic coordinator of the uni- versity’s research council. Her research interests include: limited in their capacities to fortify market knowledge the specific conditions for innovation in developing demand, which depends primarily on the productive countries, and problems associated with the production and structure and other factors beyond their reach. But social use of knowledge in such countries. She is a member of the scientific board of the Global Network on Economics they can systematically combine efforts with other of Learning, Innovation and Competence-building Systems public policies in order to foster the social demand for (Globelics). knowledge. This demand is extremely diverse: in developing countries the demand for ‘inclusive inno- vations’ is particularly important. These are charac- terized as solutions to problems that primarily affectactual role of knowledge. Broadly speaking, these deprived populations and are truly innovative, that is,processes have two main aspects: they exhibit some degree of newness and originality. Mobilizing learning by solving through this demand Learning by studying at high level in a way that is one of the ways in which to cope with the problem makes it possible for a person to continue acquir- of knowledge for development. ing knowledge throughout their life. The rest of this paper is divided into five sections. Learning by participating in knowledge- The first section deals with the characterization of demanding activities that include problem the above-mentioned processes of learning, and with solving. the ‘learning divide’ among countries that result from differences in the opportunities to perform suchBriefly speaking, these are learning by studying and processes. The second section deals with the rolelearning by solving. Knowledge-rich countries of knowledge demand in fostering learning pro-are such because many people have opportunities cesses. The third section shows that learning byto learn both by studying and by solving. Such op- solving is particularly difficult to foster in develop-portunities are distinctively less frequent in other ing countries. The fourth section considers relatedcountries: thus a major ‘learning divide’ emerges. lessons stemming from the experience of Latin Learning by studying is mainly related to knowl- American policies for science, technology and inno-edge supply: learning by solving is mainly related to vation. It follows that policies should foster the so-knowledge demand. The weakness of market knowl- cial demand of knowledge, particularly payingedge demand in most developing countries makes it attention to ‘inclusive innovations’. The final sectionparticularly difficult to foster learning by solving, exemplifies this rise to persistently unbalanced learning Learning divides in the 21st century Without the type of knowledge acquisition thatLearning by studying is mainly related comes from formal processes of studying, neitherto knowledge supply: learning further knowledge production nor the application of knowledge is possible. Without the opportunity toby solving is mainly related to utilize the acquired knowledge, such knowledgeknowledge demand. Weaknesses in looks like a library that continues purchasing newmost developing countries make it books, but is seldom used. Making mistakes and be- ing able to overcome them is a key component ofparticularly difficult for them to foster any learning process: without the opportunity tolearning by solving solve problems this key component is absent from the learning process at nation level. As Lall put it almost 20 years ago:572 Science and Public Policy October 2010
  3. 3. Knowledge demand in the South If physical capital is accumulated without built from innovation surveys, but is not available the skills or technology needed to operate it ef- for the vast majority of developing countries. An- ficiently, NTC [nation technological capabili- other, more restricted, indicator is the percentage of ties] will not develop adequately; or if formal all researchers working in firms; again, it has serious skills are created but not combined with tech- problems with coverage. The ratio of gross expendi- nological effort, efficiency will not increase ture on R&D to gross domestic product (GDP) and dynamically. (Lall, 1992: 170) the part financed by firms are very popular indica- tors of the importance given to science, technologyLundvall and Soete (2002) also suggested a differen- and innovation in a given country. The first of thesetiation between two types of learning: is widely available. It can be taken as a proxy for opportunities for learning by solving, even if a rather In national education systems people learn spe- indirect one, because it makes sense to infer that the cific ways to learn. In labour markets they ex- more a country invests in R&D in proportion to its perience nation specific incentive systems and wealth, the more opportunities knowledgeable peo- norms that will have an impact on how and ple will have to work in knowledge-demanding ac- what they learn. (quoted in Robertson et al., tivities. In fact, as is elaborated later in this paper, 2007: 160) one of the reasons why most developing countries continue to have dismal figures for R&D/GDP is be-We refer to the process of acquiring knowledge and cause such knowledge-demanding activities areskills through the formal process of studying at ter- scarce.tiary level as ‘learning by studying’. The process of Table 1 shows these indicators for several coun-acquiring knowledge by systematically using ad- tries, some of which are known for having economiesvanced knowledge for solving problems is termed that are to a large extent innovation-driven. Others are‘learning by solving’. included for purposes of comparison. One good indicator of the opportunity for learning The figures in Table 1 are consistently high forby studying is given by the enrollment in higher ed- developed countries. In some cases, like Brazil anducation. Such information is available for most South Africa, the figure for R&D/GDP is not as lowcountries. A different indicator of the supply of as in most developing countries, reaching a thresh-knowledgeable people is the number of researchers old of 1%, but the rest of the indicators are much(full-time equivalents) in relation to the population. weaker. Argentina is relatively well placed in termsAs would be expected, these indicators are linked. of the enrollment in higher education and number ofCountries usually score high or low on both indica- researchers, but scores low in the proxies for learn-tors (see Table 1 for some selected countries). ing by solving. China exhibits figures which are ful- The opportunities for learning by solving are dif- ly comparable with those of developed countries inficult to approach by means of a single indicator. A some indicators, but it still lags behind in others.relatively good indicator would be the percentage of However, its rhythm of change is impressive:all human resources in science and technology between 1996 and 2006 its gross expenditure on(S&T) working in businesses; this indicator can be R&D multiplied by six (Ministry of S&T of theTable 1. Some indicators of capacities to produce and to use knowledge for diverse countriesCountries HE gross Researchers GERD/GDP GERD % funded Researchers working enrolment ratio (FTE) per million by business in business firms inhabitants enterprises (% of total researchers)USA 82 4.671 2.67 66.4 79.2Sweden 75 6.139 3.68 63.9 72.8South Korea 96 2.044 3.47 73.7 68.5China 22 1.071 1.49 70.4 68.3Japan 58 8.840 3.45 77.7 66.1Denmark 80 5.431 2.57 59.5 65.1Finland 94 7.382 3.47 68.2 56.8France 55 3.440 2.10 52.4 55.5UK 59 3.695 1.84 47.2 48.8Spain 68 2.784 1.28 47.1 43.5Brazil 30 629 1.02 47.9 37.4Mexico 26 460 0.50 46.5 31.1 (2003)India 13 137 0.80 19.8 16.8Argentina 62 980 0.51 29.3 15.5South Africa 15 382 0.96 43.9 …Colombia 33 151 0.18 27.0 0.3Source: UIS, available at <>, 2007 reference year or latest year available; for South Africa, Southern Africa Regional Universities Association.Science and Public Policy October 2010 573
  4. 4. Knowledge demand in the SouthPeople’s Republic of China, 2007); the enrollment in Many developing countries have experiencedhigher education increased from 7% to 22% between periods of economic growth stemming from differ-1999 and 2007 (UNESCO, 2010). ent combinations of the first three of these processes. We can put each country in a ‘map’ (see Figure 1) But backwardness has been substantially and per-by using enrollment in higher education and manently diminished only in the comparatively fewR&D/GDP, respectively, as proxies for learning by cases when the fourth process became relevant. Re-studying and learning by solving. markable examples have been provided during the Figure 1 schematizes a learning divide between 20th century by the Scandinavian countries and bydeveloped countries and several other countries East Asia where learning processes expanded(Arocena and Sutz, 2000). This notion recalls how quickly.difficult it is for many countries to cope with the en- Importing knowledge has been of paramount im-rollment revolution and to use knowledge for devel- portance in every successful story of development.opment, that is, to cross the learning divide. But the problem of knowledge for developmentNevertheless, it can be crossed, as is shown by the could not be solved solely by importing knowledge.differing positions of South Korea over time. This is connected with the following assertions due to Lall (1990: 22): Why is knowledge still a weak lever for  Without investment capabilities ‘there is little un- development in many countries? derstanding of the technology transferred, leading to subsequent high costs of operation and a lackMokyr (1992: 4–6) asserts that economic growth of upgrading’.stems from four distinct processes:  Process innovation capabilities are necessary, par- ticularly because ‘buying new technology may be investment, that is, increases in capital stock; quicker and less risky than innovation, but, to be trade and division of labor; successful and economical, also requires a high economies of scale or, more generally, size ef- degree of technical skills’. fects; and  ‘Capabilities to follow trends in product develop- increases in the stock of human knowledge, ment, (almost as difficult as independent innova- including technological and institutional changes, tion)’, are necessary because new technologies which are the source of the so-called Schumpet- are often not transferable to independent erian growth. firms. Israel Japan Sweden Finland South Korea (2007) South Korea (1997) Germany Austria Denmark USA Australia France Canada UK Norway Czech R. China South Korea (1985) Spain Italy Russian Fed. Brazil Tunisia South Africa Turkey Lithuania Pakistan India Malaysia Chile Mexico Argentina South Korea (1975) Costa Rica Uruguay Bolivia Colombia Vietnam Peru Ecuador Algeria  Figure 1. Illustration of the ‘learning divide’ Notes: Vertically, from bottom to top, R&D/GDP: <0.5, 0.5–1, 1–2, 2–3.5 Horizontally, from left to right, gross enrolment in tertiary education (%): <30, 30–50, 50–60, 60–75, >75 Source: As for Table 1, Arocena and Sutz (2000)574 Science and Public Policy October 2010
  5. 5. Knowledge demand in the SouthLall exemplifies the conscious building of these …are uncertain in that their success depends oncapabilities in South Korea, and shows that Sub- the ability and readiness of the actors operatingSaharan Africa is an example of the lack of such ef- in the system to absorb and maximise the po-forts. In Latin America importing industrially em- tential of these incentives. Developing demand-bodied knowledge was a failure if it was not side policy instruments is of equal relevancecombined with endogenously generated knowledge for European and national systems, and relate(Fajnzylber, 1984). Nowadays technical cooperation to policy actions designed to enhance the tech-with least developed countries also faces the diffi- nological competencies of the system byculties of building long-term capacities and the like- emphasising the role of the user.lihood of limiting or inhibiting local alternatives(Bell, 2007: 10). A case study in an economically depressed part of Endogenous generation of knowledge is a neces- France shows:sary complement to the success of knowledge im-ports, particularly when knowledge in the making is …that it is not sufficient to legislate a policyimported. A general feature of the changes in the for making scientific and technologicalproduction and use of knowledge is that both are in- knowledge available in order to create a rela-creasingly intertwined. Advanced learning, scientific tionship of synergy between the developmentresearch, technological innovation and economically of universities and economic development. Inefficient diffusion of innovations become more en- the case in point, the problem is the lack of de-tangled. Thus, the last process becomes increasingly mand on the part of the main economic actorsdifficult if it is not closely connected with the other (industrial firms). (Laperche, 2002: 168)three. The need to cope with such issues ‘systemically’ A report on the state of S&T in Latin America indi-has also been reinforced by their context-dependent cates that the most important factor explaining theaspects. Health and life sciences in Latin America shortage of linkages between knowledgeable peopleowe, in no small measure, their comparatively dis- and institutions and production is still the weaknesstinguished history to that phenomenon. Some of the of the knowledge demand from enterprises (RICYT,main health problems in the continent were quite 2008: 22).specific. They could not be solved by knowledge Of late, demand has been a rather forgotten issue:transfer alone: systemic building of research and in-novation capabilities was needed. Today, generally Imagine trying to cut a piece of paper with justspeaking, problems and disciplines with strong con- one blade of a pair of scissors. It’s near impos-text-dependent aspects are expanding their influence sible. Yet that is what we try to do with innova-in knowledge-related activities. Health, environ- tion policy. (...) Innovations are the product ofment, nutrition, and sanitation are all examples of the creative interaction of supply and demand.that. Related problems often have solutions that are However, in focusing on how to increase theaffordable in the North in terms of research, devel- supply of innovative businesses, policymakersopment and production costs as well as in terms of have lost sight of the importance of demand.the purchasing power of the expected customers, but (Georghiou, 2007: 1–2)that are not affordable in poorer contexts. When onlyunaffordable solutions exist, the problem is still That is particularly worrying for developingopen and requires specific research (Srinivas and countries. We conjecture that one of the main causesSutz, 2008). of the learning divide is the weak knowledge de- Summing up, the new centrality of learning and mand addressed to endogenous sources ofinnovation processes gives new weight to an old knowledge by the productive structure of developingidea: to make knowledge a lever of development, countries. In Latin America this has been and still is‘massive’ endogenous capabilities and opportunities a dominant trait.for creatively using advanced knowledge are needed.But that is quite difficult when demand for The Latin American and Caribbean productionknowledge is weak, in quantity and/or in quality. pattern on the one hand, induces private sector The role of such demand has been frequently and enterprises to express a meager demandstressed. Porter (1990: 86) stated that home demand for knowledge, and on the other hand, leadsconditions: domestic agents to mostly seek outward ori- ented linkages, privileging foreign companies …shapes the rate and character of improvement and research laboratories that already have and innovation by a nation’s firms sound reputation and worldwide recognized experience in effective and efficient S&T ef-Porter argued further that the quality of home de- forts. Thus a mismatch ensues between de-mand is more important than its quantity. Lundvall mand-side needs and supply-side offerings,and Borrás (1997: 123) stressed that the effects of hampering policies’ impact. (Cimoli et al.,supply-side policy instruments: 2009: 12)Science and Public Policy October 2010 575
  6. 6. Knowledge demand in the SouthThe mismatch between demand and supply and, par- Fostering knowledge demand needs specific andticularly, the problem of the knowledge outward ori- context-dependent policies. It cannot be assumedented linkages, was already well understood by the that the same policies will be efficient where suchLatin American scholars of S&T 40 years ago demand is rather strong and also where it is weak.(Sabato and Botana, 1968). It leads to a sort of glob- Concerning the developing world, what is assert-alization of the sociology of science concept, the ed is the weakness of market demand for knowledge.Matthew effect famously proposed by Merton According to Rodrik, the explanation is that entre-(1968): those knowledge and innovation producers preneurs think that, on average, innovative activitiesthat receive the knowledge demand from developing offer low profits. Usual policies for fosteringcountries get stronger in terms of their ability to sat- knowledge demand focus on firms. So the problemisfy such demand; conversely, those that even hav- requires a more diversified set of policies. Furthering the knowledge skills to solve problems are not support for this assertion will be given in the sectioncalled to solve them in their own countries, will of this paper devoted to a brief review of thechange profession, migrate or exploit their skills achievements and limitations of Latin Americanwell below their possibilities. knowledge policies. Now we turn to another reason Consequently, as Rodrik (2007: 101) has for focusing on knowledge demand: the most diffi-commented: cult aspects of learning processes are directly de- pendent on the level of such demand. …innovation in the developing world is con- strained not on the supply side but in the de- mand side. That is, it is not the lack of trained Two aspects of fostering learning scientists and engineers, absence of R&D labs, processes: a difficult and a more or inadequate protection of intellectual proprie- difficult task ty that restricts the innovations that are needed to restructure low-income economies. Innova- Learning by studying is less difficult to achieve than tion is undercut instead by lack of demand from learning by solving, the latter being directly related its potential users in the real economy – the en- to the level of knowledge demand. The former pro- trepreneurs. And the demand for innovation is cess is not at all easy. Although enrollment in higher low in turn because entrepreneurs perceive new education is steadily growing globally, in most de- activities to be of low profitability. veloping countries this growth is far from enough to redress the gap. Nevertheless, even if results areKnowledge supply per se does not create knowledge slow to achieve, and moreover some processes, likedemand. This has been recently recognized in India, emigration, can jeopardize them, the move towardswhere around one-third of all graduates are in scien- increasing the opportunities to study at tertiary leveltific fields but 66% of those holding advanced de- is distinguishable (see Figure 2). Understanding whygrees and unemployed are also in those fields this is so helps us to better grasp the difficulties of(Bagla, 2005). fostering learning by solving. Figure 2. Tertiary gross enrollment ratio by geographical region, 2000 and 2007 Note: data includes all post-secondary students (ISCED4, 5 and 6) Taken from Altbach et al. (2009: v)576 Science and Public Policy October 2010
  7. 7. Knowledge demand in the South People’s motivations to seek access to higher On Latin American science, technology andeducation can be relatively independent of market innovation policiessituations. Even if increasing pressures are exertedupon universities to transform them into more Supply-side policies in Latin America started asmarket-dependent organizations, they still have a classical science policies: in some cases they becamerelative autonomy from purely market considera- more elaborate S&T policies. Resources and posi-tions. Universities are seen as enabling organiza- tion in the government agenda rose and fell in dif-tions par excellence in the knowledge economy: ferent periods. On average they have hardly everexpanding them is usually considered a legitimate been high, but the balance differs greatly fromaim of public policy. The advancement towards a country to country. Nevertheless, they have contrib-higher rate of enrollment in higher education in de- uted to the building of a fairly significant researchveloping countries has an increasingly positive citi- structure. During the period 1996–2006 the numberzenry valuation. Thus the move towards learning of full-time equivalent researchers in Latin Americathrough studying, including at higher levels, has a and the Caribbean grew 85%, reaching 234,661. Thegrowing level of political support. So, it can be ex- weight of the region in terms of researchers at globalpected that this leg of the learning opportunities, level also grew significantly during that period, fromnotwithstanding all the difficulties involved, will on 2.9% to 3.8% (RICYT, 2008: 16, 17).average, become steadily stronger (see Altbach et The outcomes of supply-side policies in terms ofal., 2009). research capacities have been highly influenced by We now turn to the other leg of the learning op- three factors. One is the already stressed weak mar-portunities. Learning by solving usually involves ket demand for endogenous knowledge generation.some sort of human group, where learning is fos- Lack of continuity and weakness of public policiestered by the cross-fertilization of what different peo- have also been relevant, Brazil being a partial excep-ple in the group know. Even if each person learns in tion. We see different combinations of periods ofthis way, the most productive result is that the group almost complete neglect with periods of mild atten-as such learns. tion to these issues and with some short periods of fairly strong official push. Those two ‘external’ fac- It is important to distinguish between the con- tors help to explain the relevance of a third factor: cepts of ‘learning community’ and ‘collective the internal dynamics of organized research. The of learning individuals’. Learning communities Latin American research structure has been shaped demonstrate as such learning behavior, i.e. to a large extent by claims stemming from some as a whole they have the ability to construc- large public universities and by the different and tively interact with change at a level that trans- changing levels of self-organizing capabilities, of in- cends the simple sum of the various ternational connections and of the national lobbying individuals who constitute the community. power of networks of researchers, mainly working in (Visser, 1999: 7) those universities. OECD innovation policies have provided the ra-The strength of ‘learning communities’, stemming tionale for the Latin American wave of innovationfrom systematic processes of applying knowledge to policies. Some of its mechanisms were frequentlyproblem solving, is strongly dependent on the level adopted, like tax reductions for innovation-relatedof knowledge demand. investments and competitive funds to promote asso- Any group where learning by solving occurs re- ciative innovation efforts. Others were scarcelyquires searching how to solve problems. Searching adopted, for instance those directed towards the in-is always costly and time consuming, be it by: ternal transformation of the absorptive capacities of firms, like support of small- and medium-sized en- …looking up the answer in a source known to terprises (SMEs) for hiring their first employee with contain the answer, or an extended search for a problem solution that may not exist. (Nelson and Winter, 1982: 64)The propensity to search depends on the perceived In 1996–2006 the number of full-timerewards of finding solutions to unsolved or not satis- equivalent researchers in Latinfactorily solved problems. Such rewards need to besufficiently attractive to overcome inertia and the re- America and the Caribbean grewluctance to incur the different costs that searching 85%, reaching 234,661. During thatinvolves. Again, the opportunities for searching, a period the region’s share ofkey component of learning by solving, are stronglydependent on the level of knowledge demand. Even researchers at global level grew fromif fostering learning by studying is not an easy task 2.9% to 3.8%in developing countries, learning by solving is evenmore difficult.Science and Public Policy October 2010 577
  8. 8. Knowledge demand in the Southtertiary education. Instruments related to demand- Table 2. Comparison between groups of innovative firms in Uruguay and in Europeside innovation policies were hardly ever adopted,like the most pro-active mechanisms of technologi-cal public procurement. Innovation policies have not Percentage of Percentage ofachieved great success. Innovative firms continue to innovative firms innovative firms based on that excel in allbe a small proportion of all firms. Their main inno- endogenous aspects ofvation efforts are still importing machinery and competences that innovative processequipments. The capacities built through the supply- have characteristic that have indicated in row characteristicside S&T policies are still under-utilized by firms, (Uruguay, 2006) indicated in rowwhich continue to constitute a marginal labor market (statistics forfor researchers and, in the case of SMEs, even for Europe in 2004)university graduates. Innovative 89.8 88.4 The above assertions can be illustrated by the technologically incase of Uruguay. In 1985 the country recovered product anddemocratic rule, after 12 years of dictatorship. S&T processessupply-side policies have been implemented since Performing internal 64.1 70.9then, backed by international organizations. A pre- R&Dviously dismal situation has been moderately im- Receiving public 0.5 28.5 support for innovationproved in terms of the number and training ofresearchers, research infrastructure and salaries, Source: based on Bianchi et al. (2009)grants, mobility and research quality. Those poli-cies, mainly supported by public money, have notbeen accompanied by a really intensive demand for usefulness of the available knowledge supply. Inno-local research and knowledge capabilities from in- vation policies of the type implemented so far industry or services. Policy instruments aimed at fos- Uruguay have not advanced much towards this kindtering innovation at firm level were implemented, of outputs. For example, 80% of small firms do notbut it was tacitly assumed that a main reason why have any employee with a background in S&T ter-firms do not rely on knowledge and are not too in- tiary education, a trend first measured in 1986novative is that they lack sufficient money to do (Argenti et al., 1989) and confirmed in the last in-so. Consequently different types of financial in- dustrial innovation survey of 2006 (Bianchi et al.,struments were developed, from tax alleviation to 2009).co-financing for innovation projects. These instru- To date, innovation policies that could foster thements were not extensively used. Even between productive use of endogenous knowledge capabili-Uruguayan innovative firms the reliance on the ties in Uruguay and other Latin American countriesavailable public support was quite low when com- have had limited success. Such policies are frequent-pared with such reliance in Europe. ly defined as a sort of ‘cut and paste’ from the poli- To establish a meaningful comparison we take cies of developed countries, so local specificitiestwo groups of firms, in Uruguay and in Europe, are not sufficiently taken into account. Informationcharacterized by being the most innovative within about the innovativeness of firms and related issuesthe universe of all industrial firms. The definition of is frequently gathered through surveys aimed moresuch firms is, for Uruguay: ‘Innovative based on the at establishing international comparisons than an-endogenous competences of the firm’, representing swering relevant questions in a given context.7.1% of all industrial firms (Bianchi and Gras, Competitive funds for innovation would be de-2006). For Europe the definition was: ‘companies manded by many firms only if the prevailing high(that) shine in all aspects of the innovative process innovative passivity is redressed to some extent byand support this activity through the requesting of some ‘localized’ instruments that are more pro-advice from experts as well as hiring new graduates active and even customized. If this does not hap-from universities at home and abroad’ (European pen, only those few firms that already know thatCommission, 2004), representing 20% of the sam- local knowledge is important will apply for funds.ple. Table 2 shows that the two groups of firms are Demonstration effects are needed to build strongerquite similar regarding innovation, while it also bridges between the relatively weak knowledgeshows how they differ in relation to their utilization demand from firms and national knowledge supply.of public support for innovation. Policies could identify some important and para- On the other hand, indicators of achievement for digmatic problems affecting groups of firms andan innovation policy can include an increase in the support specific programs to search solutions. Theseproportion of firms that employ knowledgeable peo- ‘tailored demand-side’ innovation policies areple as well as a widening of their relations with scarcely seen in manufacturing. The situation is dif-knowledge producers. Given that no business would ferent in agriculture, where the information aboutmake investments in this direction unless it consid- production problems in need of knowledge is per-ers it profitable, such outputs can be reasonably con- manently updated through different types of institu-sidered to be a shift towards recognition of the tional arrangements.578 Science and Public Policy October 2010
  9. 9. Knowledge demand in the South There is surely ample room for improvement for cultural consequences. In particular, it will foster adifferent kinds of demand-side policies directed to better match between national problem-solvingfirms’ innovative behavior. It is worth bearing in capabilities and demands for knowledge and innova-mind, however, that such improved policies will tion in society at large.probably face a high wall of a structural nature, If these assertions are likely to be correct, one cannamely the type of productive activities in which envisage breaking the vicious circle and starting anational firms are engaged and the main sources of virtuous one. Increasing demand can be a self-profits obtained by the entrepreneurs that own and sustained process if, when demand increases, it isgovern such firms. met by a satisfactory supply, that fosters further de- The question has two parts: first, if the knowledge mand and also enhances supply. That outline of ademand of businesses is difficult to foster due to strategy for mobilizing knowledge demand towardsstructural reasons; and second, without the strong national capabilities is based on inclusive innova-fostering of knowledge demand the learning divides tions. Thus the potential role of such innovationswill not be overcome, then what sectors of social life needs to be assessed.would be adequate to stem knowledge demand That potential stems, first of all, from the factthrough new types of innovation policies well that some problems faced by deprived people haveadapted to promote processes of human and sustain- not gained insights from knowledge because theyable development? have not been addressed by modern S&T. In some cases this is due to commercial reasons, but in other cases such problems are ‘below the radar’ (Clark, et Policies to foster social demand al., 2009) of present mainstream research and inno- of knowledge vation. Helping R&D agendas as well as business- es’ innovation agendas to pick up some of theseImproving the innovative capabilities of firms is problems and to give them full attention may have aessential for development. In many developing positive impact on the achievements of social poli-countries structural obstacles position the perceived cies. Some of these problems are of tantalizingrisk threshold for starting innovative efforts at too complexity, like many in the realm of health, wherehigh a level. This is particularly true for those inno- the non-existence of solutions may be related to in-vative efforts related to a more intensive use of sufficient knowledge. In other cases the lack of so-advanced knowledge. Consequently, knowledge lutions can be traced back to an unposed question ordemand stemming from business continues to be a problem which was not highlighted. A Uruguayanweak, especially the part of such demand directed to example of a solution found once the problem itlocal supply. Thus supply-side S&T policies lack solved received attention is the potable water auton-legitimacy, as is seen in the endless re-negotiation of omous unit, a transportable engine able to cleanthe amounts of the public investment on R&D. Such mud and other physical contaminants from water. Itinvestment is difficult to legitimize if society at large is small and light enough to be transported in an or-does not make an extensive use of endogenous dinary truck. It proved particularly useful for serv-knowledge-based capabilities. If those capabilities ing communities after different types of climaticare not expanded then development is jeopardized, disturbance, or to provide drinkable water to small,which perpetuates a vicious circle. isolated communities, where the distance to urban We posit that one way to break such a vicious cir- centers makes the expansion of water services toocle is to widen the approach and the scope of inno- expensive. The problem was highlighted when Uru-vation policies, while also conceiving them as social guayan soldiers serving under the UN in Africapolicies. Innovation policies as social policies are were frequently moving from place to place, nonethose that point at detecting and fostering knowledge of which had drinkable water available. The innova-demand that is both socially very justified, and is tion was developed by engineers from the armyaddressed to endogenous providers of knowledge. and the public enterprise devoted to water supplyInclusive innovations are the expected outputs of and sanitation. The device is now used in severalsuch policies. They are required to improve the well- countries.being of the most deprived part of the population. This is simply an anecdote, it illustrates that inThe needs that social policies should fulfill and that some cases posing a problem can lead to research,research and innovation could try to solve are huge. development and innovation efforts that achieve aIn Latin America at least, after the disastrous social successful solution. A systematic effort to highlighteffect of the economic policies guided by the ‘Wash- social problems and to make them the focus of aington consensus’ recommendations, social policies specific branch of innovation policies could turn an-have been widely recognized as a legitimate and im- ecdotes like this and many others into trends. If thisportant sphere of public intervention. If innovation is achieved, more solutions could be found for thepolicies are able to show that they are strategic for type of problems we are discussing. Of course, hav-the implementation of social policies, they will re- ing solutions does not mean that they will be usedceive stronger support. If innovation policies as so- effectively: the issue is that solutions must existcial policies are successful, that will have ample before they can be used effectively.Science and Public Policy October 2010 579
  10. 10. Knowledge demand in the South Another reason to consider that inclusive innova- people. The term import substitution is valid becausetions are vital for the well-being of the most de- these innovations solve problems that could eventu-prived part of the population is the high cost in the ally be solved through imports. The term competi-international market of existing solutions to some tive is valid because these innovations provide bettertypes of problem. In many cases, the problem of af- solutions than the imported ones, due to their cost,fordability is the result of political issues – i.e. intel- adaptability to local conditions or both factors.lectual property rights (IPR) protection and World Several examples of this kind of inclusiveTrade Organization regulations: several vaccines are and competitive import substitution from Uruguayexamples of that. However, in some cases, lack of can be given. We shall briefly mention one in theaffordability may be related to genuinely high costs realm of health, which is actually used in severalof production. Turning things upside down and de- public hospitals in Uruguary. The problem posed inveloping a radically different way of looking into the technical terms was that the halogen lamps to treatproblem, which can lead to a radically different severe jaundice through photo-therapy used in themethod of production, can be a good alternative to a intensive care divisions of the maternity wards ofheavy subsidy of the existing solutions. public hospitals had a high rate of failure due to the In the case of the Haemophilus influenzae type short life of the light bulbs. Replacement bulbs wereb (Hib), responsible for meningitis and different very expensive, thus several lamps were not operat-kind of sepsis in small children, full-immunization ing. This photo-therapy involves directing a veryschemes were in place in most industrialized precise beam of blue light onto the baby in order tocountries by 1991, following the innovation that de- allow the expulsion of the bilirubin molecule, whoselivered a workable vaccine. The Global Alliance for concentration causes the jaundice. The problemVaccines and Immunization started support for ac- posed in social terms is that premature babies have aquiring Hib vaccines in 17 low-income countries in higher probability of being affected by severe jaun-2005. dice. Such babies are frequently from the most de- prived sector of the population and are treated in Middle-income and developing countries have public hospitals. At the international level, the exist- been hesitant to introduce the vaccine because ing solution was lamps made from light emitting di- of its relatively high price. (Akumu et al., odes (LEDs), semiconductor devices that have a 2007:1) very long life. These lamps were expensive, mainly because LEDs are low intensity devices, so many of So far (...) immunization against this disease has them are required to reach the necessary level of in- reached only a fraction of the children living tensity. The new solution was based on the design of in low-income countries. (WHO, 2006: 451) an optical device to be incorporated into the lamp that concentrates the light, obtaining the same inten-One road towards inclusive innovation strategy is to sity with a much smaller number of LEDs. The in-try to develop alternative solutions to the existing novation was due to a young electrical engineer andones, combining the needed technical characteristics atomic physicist at the Universidad de la República.with affordable costs. In the case of the vaccine The new lamp (commercially named BiliLed) wasagainst Hib this strategy was followed in Cuba, in produced by a national electronic firm. It was in-cooperation with a Canadian university. The alterna- vented because a physician was enraged by the prob-tive consisted of developing a synthetic vaccine, that lems with the existing lamps, and he commented onis, a purely chemical vaccine, whose production the problem to a researcher who was able to relatecosts would be considerably lower than those of the the problem to experimental work in his academiccurrent, biological vaccine. It took more than a dec- laboratory (Geido et al., 2007). Again, this is an an-ade to achieve success as the R&D part of the pro- ecdote, told in an extremely stylized way, withoutcess was tantalizing complex (Vérez-Bencomo et entering into the many difficulties that paved theal., 2004). But, at the end of 2004 one million doses way to the final device. The point is that anecdotesof this new vaccine, whose effectiveness is fully are useful to foresee possible trends: we posit thatrecognized by the WHO, were inoculated into those the trend that this anecdote suggests is not onlyCuban infants who were aged less than two months. necessary but also possible. In the last paragraphs we referred, first, toinnovations oriented towards ‘not-yet-existing-solutions’, and then to innovations geared towards Conclusionwhat can be termed ‘inclusive and competitive im-port substitution’. The latter refers to problem- General comparisons (enrollment in higher educa-solving innovative efforts that follow to some extent tion, percentage of researchers in the working popu-a different heuristic path than that used to achieve lation etc.) show that the knowledge supply isthe existing solutions. The term innovation is valid quantitatively weaker in developing countries thanbecause we are talking about different solutions. The in the North. Nevertheless, indicators like theterm inclusive is valid because it is attempted to amount of investment in knowledge production andprovide solutions to problems which affect deprived the number of knowledgeable people hired by firms,580 Science and Public Policy October 2010
  11. 11. Knowledge demand in the Southsuggest that weak knowledge market demand is a 2009. Available at: < acsip/index.php/acsip09/atlc09/paper/view/160/207>, last ac-more structural problem. Such demand is frequently cessed 2 June 2010.exogenously oriented, widening the mismatch be- Cimoli, M, C J Ferraz and A Primi 2009. Science, technology andtween capabilities already acquired and opportuni- innovation policies in global open economies: reflections from Latin America and the Caribbean, GCG Georgetown Universi-ties to use them. Historical experience and empirical ty, 3(1), 32–59. Available at < suggest that fostering knowledge demand revistas/articulo_118_1238082739847.pdf>, last accessed 2is both more difficult and more specific than foster- June 2010. Clark, N, J Chataway, R Hanlin, D Kale, R Kaplinsky et al. knowledge supply, so it is located at the heart of Below the radar: What does innovation in the Asian driverthe development problem. economies have to offer other low income economies. London, Innovation policies conceived partly as social pol- UK: Economic and Social Research Council. Available at <>, last ac-icies can help to face that problem. Such policies can cessed 2 June a set of incentives, for academia, for busi- Cooper, Ch 1973. Science Technology and Development: Thenesses and for social policies, to make the involve- Political Economy of Technical Advance in Underdeveloped Countries. London, UK: Frank Cass.ment in inclusive innovation attractive to each of the de la Motte, J and G Paquet eds. 1996. Evolutionary Economicsactors. Social policies, conceived partly as innova- and the New International Political Economy. London, UK:tion policies, can systematically build a set of rele- Pinter. European Commission 2004. Flash Eurobarometer No. 164, In-vant problems in need of knowledge-based nobarometer. Available at < Thus they can act as an effective demand- flash/fl164_en.pdf>, last accessed 2 June of science, technology and innovation capabilities Fajnzylber, F 1984. La industrialización trunca de América Latina. Buenos Aires, Argentina: Centro Editor de América Latina.from among others, the endogenous providers of Geido, D, H Failache and F Simini 2007. BiliLED low cost neona-knowledge and innovation. The combination of the- tal phototherapy, from prototype to industry. Journal of Phys-se two types of policies can strengthen knowledge ics: Conference Series 90 012024. Georghiou, L 2007. Demanding innovation. Lead markets, publicsupply and knowledge demand and, even more im- procurement and innovation, NESTA National Endowmentportantly, can help to connect the former and the lat- for Science, Technology and the Arts, Provocation 02. Availa-ter. In this sense, these types of innovation policies ble at < Conference/2007%20Annual%20Conference/Position%20and of social policies can be useful for overcoming Papers/Luke_georghiou_provocation.pdf>, last accessed 2the mismatch between local knowledge-based prob- June 2010.lem-solving capabilities and societal problems which Herrera, A 1971. Ciencia y Política en América Latina. México DF: Siglo XXI Editores.need such capabilities in order to be solved. Lall, S 1990. Building Industrial Competitiveness in Developing Countries, Vol. 1. Paris, France: Development Centre Studies, OECD. Lall, S 1992. Technological capabilities and industrialization.References World Development, 20(2), 165–186. Landes, D 1998. The Wealth and Poverty of Nations: Why SomeAltbach, P, L Reisberg and L Rumbley 2009. Trends in global are so Rich and Some so Poor. New York: W W Norton. higher education: tracking an academic revolution. Report for Laperche, B 2002. The four key factors for commercialising re- the UNESCO 2009 World Conference on Higher Education. search: the case of a young university in a region in crisis. Available at < OECD Higher Education Management and Policy, 14(3), 149– 2844977e.pdf>, last accessed 2 June 2010. 173. Available at <, A O, M English, J A G Scott and U Griffiths 2007. Eco- 37444681.pdf#page=146>, last accessed 2 June 2010. nomic evaluation of delivering Haemophilus influenzae type b Lundvall, B A 1992. Introduction. In National Systems of Innova- vaccine in routine immunization services in Kenya. Bulletin of tion, B A Lundvall ed., pp 1–16. London, UK: Pinter. the World Health Organization, 85(7), 511–516. Available at: Lundvall, B A and L Soete 2002. GLOBELICS: GLOBal Network < for Economics of Learning, Innovation and Competence- html>, last accessed 2 June 2010. building Systems. Aalborg, Denmark: Department of BusinessArgenti, G, C Filgueira and J Sutz 1989. Ciencia y tecnología. Un Studies, University of Aalborg. Available at <http://www. diagnóstico de oportunidades. Montevideo, Uruguay: Edicio- nes de la Banda Oriental. =16&print=>, last accessed 2 June 2010.Arocena, R and J Sutz 2000. Interactive learning spaces and de- Lundvall, B A and S Borrás 1997. The globalising learning econ- velopment problems in Latin America. DRUID Danish Re- omy: Implications for innovation policy. Report to the Commis- search Unit for Industrial Dynamics Working Paper 13/2000. sion of the European Union, TSER Program. Brussels: Available at < European Commission. ds2000-86.pdf>, last accessed 2 June 2010. Merton, R 1968. The Matthew effect in science. Science,Bagla, P 2005. Indians embrace science, but can’t always prac- 159(3810), 56–63. tice it. Science, 309(5744), 2142. Ministry of Science and Technology of the People’s Republic ofBell, M 2007. Technological learning and the development of pro- China 2007. China S&T Statistics Data Book. Available at duction and innovative capacities in the industry and infra- < structure sectors of the least developed countries: what roles 109573867344872.pdf>, last accessed 2 June 2010. for ODA? UNCTAD The Least Developed Countries Report Mokyr, J 1992. The Lever of Riches, Technological Creativity and 2007 Background Paper. Available at < Economic Progress, Oxford, UK: Oxford University Press. sections/ldc_dir/docs/ldcr2007_Bell_en.pdf>, last accessed 2 Nelson, R and S Winter 1982. An Evolutionary Theory of Eco- June 2010. nomic Change. Cambridge, MA: Harvard University Press.Bianchi, C and N Gras 2006. Innovative behavior and economic Porter, M 1990. The Competitive Advantage of Nations. New performance in the Uruguayan manufacturing industry 2001– York: Free Press. 2003. Paper presented at International ProACT Conference, Reinert, E 2007. How Rich Countries got Rich and Why Poor Innovation Pressure, held 15–17 March 2006, Tampere, Countries Stay Poor. London, UK: Constable. Finland. RICYT 2008. El estado de la ciencia 2008. Buenos Aires, Argen-Bianchi, C, N Gras and J Sutz 2009. Cooperation for innovation tina: Centro Redes. Available at < as knowledge exchanges: a view from innovation surveys. Pa- interior/interior.asp?Nivel1=6&Nivel2=5&IdDifusion=25>, last per presented at the Atlanta Conference on Science, Technol- accessed 2 June 2010. ogy and Innovation Policy, held Atlanta, GA, 2–3 October Robertson, S, M Novelli, R Dale, L Tikly, H Dacha and NScience and Public Policy October 2010 581
  12. 12. Knowledge demand in the South Alphonse 2007. Globalisation, Education and Development. UNCTAD 2007. The Least Developed Countries Report, 2007. Ideas, Actors and Dynmics, Report Educational Papers No. Knowledge, technological learning and innovation for devel- 68. Bristol, UK: Department of International Development, Uni- opment. Available at < versity of Bristol. Available at < ldc2007_en.pdf>, last accessed 2 June 2010. people/academicStaff/edslr/publications/10slr>, last accessed UNESCO 2010. UNESCO Institute for Statistics 2010. Available 2 June 2010. at <, D 2007. One Economics, Many Recipes: Globalization, aspx?ReportId=167>, last accessed 23 September 2010. Institutions, and Economic Growth. Princeton, NJ: Princeton Vérez-Bencomo, V, V Fernández-Santana, Eugenio Hardy, Maria University Press. E Toledo, Maria C Rodríguez et al. 2004. A synthetic conju-Rosenberg, N and C Frischtak 1985. International Technology gate polysaccharide vaccine against Haemophilus influenzae Transfer: Concepts, Measures, and Comparisons. New York: Type b, Science, 305(5683), 522–525. Praeger. Visser, J 1999. Overcoming the underdevelopment of learning: ASabato, J and N Botana 1968. La ciencia y la tecnología en el transdisciplinary view. Paper presented at the Annual Meeting desarrollo futuro de América Latina. Revista de la Integración, of the American Education Research Association, held Mon- 1(3), 15–36. treal, Canada, 19–23 April 1999. Available at <http://www.Srinivas, S and J Sutz 2008. Developing countries and innovation:>, last ac- searching for a new analytical approach. Technology in Socie- cessed 2 June 2010. ty, 30(2), 129–140. WHO 2006. Weekly epidemiological record, No. 47, 445–452.Stewart, F 1977. Technology and Underdevelopment. London, Available at < UK: Macmillan. index.html>, last accessed 2 June 2010.582 Science and Public Policy October 2010
  13. 13. Copyright of Science & Public Policy (SPP) is the property of Beech Tree Publishing and its content may not becopied or emailed to multiple sites or posted to a listserv without the copyright holders express writtenpermission. However, users may print, download, or email articles for individual use.