LEARNING, INNOVATION AND GROWTH: A COMPARATIVE POLICY APPROACH TO EAST ASIA AND LATIN AMERICA

Transcription

1 Paper published in in Science and Public Policy, vol. 24, no. 6. LEARNING, INNOVATION AND GROWTH: A COMPARATIVE POLICY APPROACH TO EAST ASIA AND LATIN AMERICA Guillermo Cardoza Belfer Center for Science and International Affairs (CSIA), Research Fellow J.F. Kennedy School of Government, Harvard University LASPAU Fellow Executive Director, Latin American Academy of Sciences Cambridge, 1997.

2 2 Learning, Innovation and Growth: A Comparative Policy Approach to East Asia and Latin America Guillermo Cardoza 1 tel. (617) Guillermo_Cardoza@harvard.edu Abstract How have East Asian economies managed to grow so rapidly, while reducing the inequality of income distribution and increasing international competitiveness? Although it is still too early to know the consequences of the ongoing financial crisis in East Asia, it is important to highlight the extraordinary accomplishments achieved by these countries between 1960 and the mid-1990s and to stress the rich experiences from which developing countries can continue to draw lessons. This study concentrates attention on the development of more comprehensive approach helping to explain the systemic nature of industrial dynamics and the interdependence of economic variables involved in development strategies. The spectacular momentum of East Asian growth achieved during this period, and now threatened by the financial crisis, is the result of the establishment of a virtuous circle of learning, innovation and economic growth. 1 The author is Research Fellow at the Center for Science and International Affairs (CSIA), John F. Kennedy School of Government, Harvard University, Fellow at LASPAU: Academic and Professional Programs for the Americas Professor of International Economics and Technological Development at Universidad Central de Venezuela, and Executive Director of the Latin American Academy of Sciences (ACAL). This article is based on research funded by the Consejo Nacional de Investigaciones Científicas y Tecnológicas (CONICIT) de Venezuela, the Fundación Simón Bolívar para la ACAL, the Alberto Vollmer Foundation and the CSIA. The author wants to thank Lewis B. Branscomb and Dwight H. Perkins for detailed comments on a first draft and colleagues at CSIA and ACAL with whom he has discussed the themes raised in this article. He is also grateful to LASPAU for the administrative support.

3 3 Outward-looking growth and equality: the East Asian development path Just after World War II, Taiwan and Korea employed the usual array of tariffs, quotas, and multiple exchange rates to promote import substitution. However, Taiwan and Korea became aware of the limitations of applying substitution policies in relatively small markets. At the beginning of the 1960 s, they reacted to the slowing of economic development by undertaking several restructuring measures such as the unification and devaluation of currencies and the elimination of import restrictions (Kuznets, 1988). These actions were conceived of as a preliminary step to lay the groundwork for an exportled economic growth model. In the framework of this outward-looking industrialization strategy, East Asian firms benefited from protection and subsidies while the various industries matured, mastered their respective technologies and gained a foothold in the markets. This is an important difference from the import substitution policies applied in Latin American countries where firms grew in protected markets and were managed under the rent seeking model contrary to innovative behavior. Indeed, the preferential treatment offered by the governments of these countries to their infant industries lasted too long, precluding their exposure to the stimulating effect of international competitiveness. The results of the industrialization strategies applied in each region are very different. Certainly, between 1985 and 1994 East Asian countries registered an average annual growth rate of GDP per capita of 6.9% while Latin American countries exhibited an average rate of only 0.6% for the same period. As can be observed in Figure 1, from 1950 to 1990 the GDP per capita for the newly industrialized economies of Taiwan, Korea, Singapore and Hong Kong increased 9, 8, 7 and 6.6 times respectively. Comparatively, the Latin American countries showed much more modest increases. Even Brazil and Mexico, the countries showing the greatest growth in GDP per capita, only increased 3 and 2.6 times during the same period.

4 4 Fig.1- Latin America (6) and East Asia (4): Growth of GDP per capita in constant dollars ,000 12,000 9,000 6,000 3,000 - Brazil Argentina Colombia Chile Mexico Venezuela Korea Hong Kong Singapore Taiwan (*) (*) Data for 1951 Source: Summers and Heston, The Penn World Table (Mark 5), What is even more impressive about the economic performance of East Asian countries is that high rates of economic growth have been accompanied by more equal income distribution 2. As shown in Table 1, income inequality, measured by the ratio of income shares of the richest 20 percent and the poorest 20 percent of the population (World Bank, 1993a), is considerably lower for East Asian economies, indicating a more equitable distribution of income across households than in Latin America 3. By contrast, since the beginning of the 1980's Latin American countries have been experiencing continuous economic and social deterioration which is reflected by low (or negative) average annual growth rates of GDP and by large segments of their populations living in poverty (Table 1). 2 As explained by Perkins (1986) the benefits of growth in East Asia were widely distributed. And this wide distribution in turn contributed to political stability that made further growth possible 3 As suggested by Birsdall et al. (1995) policies adopted by governments in East Asia which we label share growth may be an important part of the explanation for the remarkable association of rapid growth and low inequality.

5 5 TABLE 1: LATIN AMERICA (6) AND EAST ASIA (7): COMPARISON OF POPULATION, GDP, GDP PER CAPITA, INCOME INEQUALITY AND POVERTY Population GDP GDP per capita Income People in poverty (%) in millions Aver. Annual Growth Rate (%) Inequality (*) Urban Rural Argentina Brazil Colombia Chile Mexico Venezuela Korea Hong Kong Singapore Taiwan Indonesia Malaysia Thailand (*) As proposed by the World Bank (1993a) income inequality is measured by the ratio of the income shares of the richest 20 percent and the poorest 20 percent of the population. Sources: World Bank, World Development Report 1992, 1993b and 1996; UNDP, Human Development Report, Another important fact to mention about the economic performance of the East Asian countries is the accelerated expansion of their industrial activities. As shown in Figure 2, for the periods and the average annual growth rates of industry were several times higher in East Asia than those registered by Latin American economies for the same periods. It should also be noted that the severe economic crisis of the 1980 s, characterized by growing fiscal deficits, high external debts and inflationary processes, considerably affected the growth of industry in Latin America.

6 6 Fig. 2- East Asia (4) and Latin America (5): Average annual growth rate of Industry and Brazil Colombia Chile Mexico Venezuela Korea Singapore Malaysia Thailand Political and economic stability have also been put forward as one of the explanations of sustained growth in East Asia. As noted by Perkins (1986): rapid economic growth in the region coincided with governments that were stable for long periods of time and which gave a high priority to creating a climate conducive to investment in growth. Indeed, until the late 1990 s a judicious management of budget deficits, external debts and exchange rates allowed East Asian countries to elude inflationary means of financing (Page, 1994). By contrast, a lack of fiscal discipline and the recurrence of money creation to cover public deficits led countries like Argentina and Brazil to suffer hyperinflationary episodes. Additionally, as observed in Table 2, the enormous weight of the external debt and the uncertainty caused by high inflation resulted in drastic reductions of investments in Latin America during the 1980 s. Although countries of both groups exhibit similar levels of participation in terms of their industrial output as a percentage of GDP, they present different shares of manufactured goods in their total exports. Table 3 demonstrates that, in spite of the international

7 7 economic recession of the 1980 s, East Asian economies maintained relatively high average annual growth rates of exports and significantly increased the shares of manufacture goods in total exports. In fact, manufactures represented the lion s share of merchandise exports for Korea, Hong Kong, Malaysia and Thailand in 1993, whereas primary commodities still represent more than half of Latin American exports (Argentina, 68%; Colombia, 60%; Chile 80%, and Venezuela, 86%) 4. It is worth noting that the share of technology-intensive products, like machinery, electronic products and transport equipment, has grown very quickly in East Asian economies. This comparison reveals the rapid shift from labor intensive to technology intensive economies in East Asia. TABLE 2: LATIN AMERICA (6) AND EAST ASIA (6): COMPARISON OF GOVERNMENT DEFICIT/SURPLUS, INFLATION, DEBT AND INVESTMENTS Central Goverm't Inflation Debt Debt Service Gross Domestic Overall Deficit/ Average Annual As percent of As percent of Investment - Av. Surplus (% GNP) Rate (%) Exports Exports Annual Growth Rate (%) Argentina Brazil Colombia Chile Mexico Venezuela Korea Hong Kong Singapore Indonesia Malaysia Thailand Sources: World Bank, World Development Report 1992, 1993b and All Latin American countries experienced drastic reductions of the average annual growth rates of imports during the 1980 s (Table 3). The scarcity of foreign exchange to import capital goods and intermediary products has affected the competitiveness and the export capacity of Latin America.

8 8 As a matter of fact, export-driven strategies are frequently cited in the literature among the most important factors explaining the success stories of East Asia (Kuznets, 1988; World Bank, 1993a; Page, 1994) 5. Export activities allowed East Asian economies to achieve industrial specialization and to adopt a better pattern of resource allocation. In addition, export firms could also make better use of available production capacities and benefit from economies of scale. Certainly, the challenge faced by export firms for meeting the high international standards of quality and design was an additional incentive to introduce best practices and offer ongoing training to personnel in charge of mastering the new technologies. Furthermore, continuous modernization of the export sector increased the demand to the governments for better services infrastructure and to the educational and S&T systems for highly qualified workers and scientists. Exports, investments and employment: the mechanics of growth in East Asia. Both Young (1993) and Krugman (1994) support the argument that the extraordinary economic performance of East Asian economies can be primarily explained by the accumulation of physical capital and a highly qualified labor force, while other authors emphasized the positive relationship of investments, low levels of government consumption and economic growth (Kuznets, 1988; Barro and Lee, 1994). Rodrik (1995) goes even further, affirming that investment is the main factor explaining the economic growth of Korea and Taiwan. Indeed, even if the average annual growth rates of gross domestic investment in East Asian economies decreased during the 1980 s, they remained relatively high compared to 5 Another factor considered responsible for the rapid economic growth of East Asian economies put forward by Barro and Lee (1993) in their cross-country analysis is the impact of initial conditions, represented by initial values of GDP per capita, educational attainment and life expectancy. In their study of the determinants of economic growth they found that these initial conditions were the most important factor contributing to growth in Korea and Taiwan during the period. Similarly, Rodrik (1995) argues that, in spite of their low income levels, in 1960 Taiwan and Korea had favorable initial conditions, mainly represented by a better educated labor force and a more equal distribution of income and wealth. He states that these factors made both countries ready for economic take-off, in the sense that the latent return to capital accumulation was high.

9 9 those of Latin America (Table 2). With the exception of Chile, all Latin American countries exhibited a drastic TABLE 3: LATIN AMERICA (6) AND EAST ASIA (6): COMPARISON OF EXPORTS, MANUFACTURE, EXPORTS AND IMPORTS Exports Imports Average Annual Total Manufactures Average Annual Growth Rate (%) (millions) % Growth Rate (%) Argentina ,020 15, Brazil ,100 43, Colombia ,920 8, Chile ,710 11, Mexico ,600 61, Venezuela ,900 15, Korea ,500 96, Hong Kong , , Singapore ,400 96, Indonesia ,900 40, Malaysia ,000 58, Thailand ,510 54, Sources: World Bank, World Development Report 1992, 1993b and reduction in investments during this decade. It is important to mention that the sustained increase of investments in East Asian countries was possible because of the growing manufactured export revenues (Table 3) and the high rates of domestic savings (Tables 4).

10 TABLE 4: LATIN AMERICA (6) AND EAST ASIA (6): COMPARISON OF INVESTMENTS, SAVINGS, EMPLOYMENT, EARNINGS PER EMPLOYEE AND, PRIVATE CONSUMPTION Gross Domestic Gross Domestic Growth of Earnings per Private GDP Investments Savings Employment in Employee Consumption per capita (% GDP) (% GDP) Manufacture(%) Growth rate (%) Average Annual Growth Rate (%) Argentina Brazil Colombia Chile Mexico Venezuela Korea Hong Kong Singapore Indonesia Malaysia Thailand Sources: World Bank, World Development Report 1992, 1993b and

11 11 Another factor contributing to the economic growth of East Asia cited by Young (1993) and Krugman (1994) is the rapid expansion of employment in manufacturing due mainly to the adoption of an outward development strategy that allowed for the rapid increase of labor demand and the incorporation of new entrants in the economy. Certainly, resources from loans, domestic savings and export revenues were used to finance productive activities that generated new jobs. The rapid growth of paid employment in manufacturing that registered in Korea (35%), Malaysia (141%) and Singapore (45%) between 1986 and 1995 contrast with the decrease in manufacturing employment experienced by Argentina, Brazil, Mexico and Venezuela (Table 4). In the words of Young (1993) the crucial characteristic of the newly industrialized countries of East Asia is not that they have unusually rapid growth in manufactures, but, rather, that they have successfully expanded investment and employment in manufacture, hence the title: industrializing. However, it should be noted also that high rates of investments and continuous technological progress permitted East Asian economies to create more highly skilled and better paid jobs and, as a consequence, to expand private consumption. Indeed, sustained investments, both in physical and human capital, allowed these countries to expand manufacturing employment and increase earnings per employee during the period. As implied by data presented in Table 4, the growing rates of earnings had an important positive impact on domestic savings and internal consumption in East Asian economies. By contrast, most Latin American countries exhibited modest or negative growth rates of earnings per employee 6 and sharp reductions in their consumption levels during the 1980 s compared with previous periods 7. 6 Between 1980 and 1990 Korea, Hong Kong, Singapore and Taiwan registered considerable increases in average wages (230%, 117%, 159% and 280% respectively), whereas Latin American countries exhibited more modest increases or even negative as it was the cases of Colombia (-4.5%), Mexico (-8%) and Venezuela (-55%). 7 Although Brazil exhibited relatively high rates of earnings per employee since 1970, it is important to note that private consumption fell considerably during the hyperinflationary period of the 1980 s.

12 12 The expansion of demand allowed for better use of economies of scale permitting cost reductions and low rates of inflation. In the words of Birdsall et al. (1995) strong domestic demand gave East Asia s early manufacturers a competitive advantage in international markets, by giving them the opportunity to test market labor intensive goods and achieve economies of scale. Furthermore, rapid growth of exports and increases in domestic consumption and investments contributed to the expansion of the aggregate demand and stimulated employment and economic growth in the East Asian region. 50 Fig. 3: Latin America (4) and East Asia (2): Growth in Investment, Employment, Earnings per employee, Consumption and GDP per capita GDI (Av.annual growth rate / ) Employment (Growth rate / ) Earnings per employee (Growth rate / ) Private Consumption (Av.annual growth rate / ) GDP per capita (Av.annual growth rate / ) Argentina Brazil Mexico Venezuela Korea, Rep. Singapore

13 13 Summarizing, the data presented in Fig. 3 suggests that East Asian economies went through a complex process of creation of a self-sustained pattern of economic growth based on the establishment of synergetic relationships among productive investments, allowed by continuous growth in savings and exports, employment generation and increases in consumption. However, even if very valuable from an analytical perspective, this presumed self-sustained pattern of economic growth doesn t yet provide a satisfactory answer to the question of how East Asian economies have managed to grow so rapidly until the mid-1990s, while reducing the inequality of income distribution and increasing their productivity and competitiveness in international markets. Knowledge accumulation: a key factor for economic performance in East Asia Public and private investment in education and in R&D contributed to the accumulation of knowledge required for the export of skill-intensive manufactures. In fact, as discussed previously, East Asian success stories share two common factors: 1- sustained demand for highly skilled human resources brought about by the export-oriented growth strategies and, 2- provision of a well-trained work force by high quality education systems and continuous on-the-job training programs. Certainly, a broad knowledge base has become the cornerstone for building modern societies. Sustained investments in education and training are needed for the construction of the so-called knowledge-intensive economies. As knowledge becomes the main source of economic growth, expansion of human capital should be at the core of any development strategy. Several cross-country analyses have shown that education attainment has a positive impact on growth (Barro and Lee, 1994; Easterly and Levine, 1996) and contributes to reducing inequality (Birdsall, 1995). Similarly, other authors have argued that the availability of a highly skilled labor force allows for increasing global productivity and enhancing the profitability of investments in the East Asian economies (Perkins, 1986;

14 14 Hou and San Gee, 1993; Page, 1994; Rodrik, 1995). As Grossman and Helpman (1991a) stated, the process of knowledge accumulation generates endogenously the productivity gains that sustain growth in the long run. As shown in Table 5, countries of Latin America and East Asia exhibited similar levels of public expenditures on education measured as percentage of GNP for the years 1980 and Nevertheless, as can be observed in Fig. 4, for the period the average annual growth rate of public expenditures in education in Indonesia, Korea and Thailand was much higher than in Latin American countries. Fig. 4- Latin America (6) and East Asia (3): average annual growth rate of public expenditure on education / Argentina Brazil Colombia Chile Mexico Venezuela Korea Indonesia Thailand Both sets of countries also registered increases in their enrollment rates in all educational levels. However, rapid educational expansion in Latin America was not accompanied by a corresponding improvement in quality. As observed in Table 5, in 1992 the public expenditures in primary and secondary education per eligible student -measured in purchasing power parity (PPP)- was considerably higher in Korea than in the Latin American countries. This indicator, although imperfect, gives a rough idea of the quality of education and makes evident the Korean emphasis in providing broad based, quality

15 15 education. By contrast, Korea s public expenditures per student on higher education are only a fraction of that of Latin American countries.

16 TABLE 5: EDUCATION IN LATIN AMERICA (6) AND EAST ASIA (6): COMPARISON OF PUBLIC EXPENDITURE, ADULT ILLITERACY, ENROLMENT RATES, EXPENDITURE PER STUDENT AND THIRD LEVEL STUDENTS. Public Expenditure Adult Net enrolment ratios Expenditure Third level Students in Tertiary on Education illiteracy Primary Secondary Tertiary per student in PPP* students per Engineering per students as % of GNP (%) (%) (%) (%) primary secondary tertiary 100,000 inhab. Million inhab. abroad (%)** Argentina Brazil Colombia Chile Mexico Venezuela Korea Hong Kong Singapore Indonesia Malaysia Thailand * Current expenditure per pupil estimated in Power Purchase Parity (International dollars) ** Tertiary students abroad as a percentage of those at home. Sources: World Bank, World Development Report 1992, 1993b and 1996; UNESCO, World Education Report, 1995; UNDP, Human Development Report, 1996; UNESCO, Statistical Yearbook, 1993,

17 17 Besides, as shown in Figure 5, compared with East Asia, countries like Colombia, Brazil and Venezuela show a relatively small percentage of students reaching the fifth grade. High rates of repetition and drop-outs reflect not only the social and economic problems but the low quality and the lack of efficiency exhibited by Latin American educational systems. Fig. 5- Latin America (5) and East Asia (3): percentage of 1991 cohort reaching grade Brazil Colombia Chile Mexico Venezuela Korea Hong Kong Singapore In 1995, adult literacy rates in Korea, Hong Kong and Singapore reached 98%, 92% and 91% respectively. With the exception of Brazil and Mexico, countries that still exhibited double digit adult illiteracy rates in 1995, Latin American nations have also managed to reduce illiteracy (Table 5). A similar trend can be observed in the net enrollment ratios in primary school in Asian countries. In 1992, Korea, Singapore and Taiwan exhibited net enrollment rates of 100% while an important percentage of children don t attend primary school in countries like Brazil (10%), Colombia (17%), Chile (14%) and Venezuela (12%).

18 18 It is also important to mention the notable effort made by countries like Korea in secondary education reflected in a net enrollment rate of 85% 8. In the same Table it can be observed that the enrollment ratios of students at the tertiary level (technical institutes and universities) registered an important growth between 1980 and 1992 in Korea (from 15% to 48%) and in Hong Kong (from 10% to 21%), whereas enrollment rates in Brazil remained unmodified, around 12%. As shown in Table 5, in 1994 the number of students in engineering per million inhabitants was considerably higher in Korea than in other countries considered in this study. These data demonstrate the emphasis that disciplines such as engineering, directly related with industrial production, receives in Korea and gives an indication on the outcome of education policies oriented to enhance competitiveness in manufacturing activities. Nevertheless, it is worth mentioning the remarkable efforts of Argentina, Chile, Colombia and Venezuela to increase the number of students at the third level. Actually, with the exception of Brazil, all Latin American countries registered a substantial growth in the number of students in higher education per 100,000 inhabitants during the period (Table 5). However, as discussed in previous studies (Cardoza and Villegas, 1996; Cardoza, 1997), the quantitative expansion of third level education in Latin America has not been accompanied by comparable increments in quality and relevance of content mainly because of the low qualification of faculty and their lack of participation in research activities. As indicated by UNESCO-CRESALC (1996) between 1950 and 1992, the number of students enrolled in higher education increased by a factor of 30 (from to 8 millions) and the number of teachers by a factor of 40 (from to one million). Nevertheless, according to UNESCO only 20 % of teaching personnel of institutions of higher education have graduate training and only 10% of them conduct teaching-research activities on a regular basis. 8 Asian governments have been highly committed to providing broad based, quality education by allocating a high share of their public expenditure on education to basic education. For instance, in 1993 Korea dedicated 81% of is current public expenditure to basic education (including pre-primary, primary an secondary school).

19 19 Another interesting common characteristic of East Asian economies is the strong emphasis on programs of overseas training of human resources. As observed in Table 5, the tertiary students abroad, measured as the percentage of those at home, was very small in Latin American countries as compared with Hong Kong (50%), Malaysia (28%) and Singapore (25%) for the period It is also important to mention that as a result of the adoption of repatriation policies, based on the offering of international competitive salaries and excellent working conditions, countries like Korea, Taiwan and Singapore are recovering many of their students and professionals sent abroad for training. On the contrary, Latin America is suffering from a growing loss of students and highly trained scientists and professionals through migration to developed countries due to the lack of appropriate incentive systems and the deterioration of social, political and economic conditions 9 (Alvarez, 1992; Cardoza and Villegas, 1997). Vocational schools and private sector training programs have also played a decisive role in providing the technical training that firms in Taiwan and Korea required of workers in the process of technological upgrading (Deyo, F. 1987). Certainly, as technologies advance they gain in complexity and demand skilled technicians and professionals able to make optimum use of them. The current economic reforms in Latin America demand more awareness and responsiveness from educational and research institutions, particularly universities and public laboratories, in order to identify and satisfy the specific needs of industries in terms of personnel training and R&D 10. As a matter of fact, the availability of a broad base of knowledge, represented by highly trained human resources, strong R&D 9 Alvarez, (1993) estimated that from 1961 to 1990 approximately 500,000 professionals and highly qualified workers emigrated from Latin American countries to the United States. 10 In the coming years increasing demand for scientist and engineers in charge of the technological upgrading required by industry should be addressed by higher education institutions in those countries committed to economic restructuring in Latin America. In order to be prepared to adequately meet these demands, universities must undergo profound academic and administrative reforms. These complex structural changes have already been undertaken in several universities in the region and preliminary results allow us to be optimistic about the results of the reform processes.

20 20 capacities and efficient information systems, has become a critical determinant in the current processes of openness and liberalization of Latin American economies. Research and Development (R&D), productivity and international trade in East Asia In addition to education and training of human resources, R&D activities are essential for building knowledge-based economies. Academic and industrial R&D are key elements for the creation and accumulation of knowledge required to strengthen innovation capabilities needed for maintaining increasing rates of productivity and continuously enhance the competitiveness of firms interested in having access to international markets. As argued by Grossman and Helpman (1991b) resources devoted to R&D contribute over time to productivity in the production of final goods, as well as to the stock of scientific and engineering knowledge. Educational institutions and public and private R&D laboratories, as well as engineering and design centers are called upon to play a fundamental role in the generation of a minimum level of endogenous R&D required for the effective diffusion and absorption of technologies. In the words of Branscomb (1995) the ability to capture scientific and technological assets in world markets requires a command of sufficiently unique technologies domestically to be able to make mutually beneficial exchanges The figures for R&D personnel, expenditures in R&D as percentage of GDP, and number of scientific publications and patents per million inhabitants for selected countries of Latin America and Asia are presented in Table 6. Data on R&D personnel per million inhabitants show that the adoption of policies actively supporting training allowed Asian countries to increase their R&D personnel more significantly. Indeed it can be observed that in 1994, Korea, Singapore and Taiwan exhibited numbers several times higher than Chile and Argentina, the Latin American countries showing the highest figures.

21 21 With the exception of Mexico and Venezuela, which exhibited decreases in R&D expenditure during the period, all other Latin American countries showed increases in this indicator. However, none has reached the level of 1% of GDP (Table 6). By contrast, East Asian countries registered impressive increases approaching the levels of industrialized countries. The R&D expenditures of Taiwan, South Korea, Singapore doubled, tripled and quadrupled respectively during the same period. TABLE 6: LATIN AMERICA (6) AND EAST ASIA (7): COMPARISON OF EDUCATION AND R&D EXPENDITURES, R&D PERSONEL, PUBLICATIONS, PATENTS, AND PRODUCTIVITY R&D R&D Publications Patents * Growth rates Value Personnel per Expenditure per million per million of GDP added per million inhab. As % of GDP inhabitants inhabitants per worker worker % Argentina (93) (92) Brazil 298 (83) 235 (93) Colombia (82) Chile (93) (92) Mexico 194 (84) (80) 0.32 (93) Venezuela 279 (83) 208 (92) Korea (92) Hong Kong Singapore (91) Taiwan (90) (91) Indonesia (88) (91) Malaysia (92) (91) Thailand (91) 0.02 (86) 0.16 (91) * Number of applications for patents per million inhabitants Sources: World Bank, World Development Report 1992, 1993b and 1996; UNESCO, World Education Report, 1995; Human Development Report, 1996; SCI, ; United Nations, Statistical Yearbook several issues. In the early stages of industrialization, East Asian countries relied mainly on foreign sources of technology and imitative learning strategies to support their technological, managerial and organizational upgrading. However, as these economies approach the technological frontier, they have remained committed to developing sound R&D capabilities providing them with the necessary indigenous learning and innovation capabilities. This explains the rapid increase in R&D expenditure registered in Asian

22 22 economies between 1981 and As mentioned in a previous work (Cardoza and Villegas, 1996) these important differences between East Asia and Latin American regions have their roots in the different roles assigned by these two groups of countries to S&T in their development processes. Most important, the increases in R&D expenditures in the Asian countries are almost entirely due to the large contribution of the private sector. Figure 6 shows the distribution of R&D activities performed by universities, industries, governmental and private nonprofit research organizations for selected countries of Asia and Latin America. The majority of R&D in Latin American countries is concentrated in universities, while in Asian countries the greater part of research is performed by industry. It is striking to note the scarce participation of industry in R&D in Latin America when compared to Asian countries. At the beginning of the 1990s, the bulk of R&D activities, mostly carried out in universities and governmental research, was funded by Latin American governments (Villegas, R. and Cardoza, G., 1995) whereas in Asian countries the greater part of research was supported by industry (Fig. 7). Fig. 6- Latin America (4) and East Asia (2): R&D by performer (around 1990) University Governmental organization Non-Profit organization Industry 100% 80% 60% 40% 20% 0% Argentina Chile Mexico Venezuela Korea Singapore

23 23 Fig. 7- R&D in selected Latin American and Asian countries by source of funds (around 1990) Goverment Private 100% 80% 60% 40% 20% 0% Argentina Chile Mexico Korea Singapore A similar analysis can be made when comparing scientific production by field for both groups of countries. As can be observed in Figure 8, the greatest percentage of scientific papers published by Latin American countries in 1993 were in basic sciences and medical sciences (Argentina, 67%; Brazil, 90%; Chile, 95%; Mexico, 89%, Venezuela, 94%), whereas in the Asian countries there was a more balanced distribution between basic and applied sciences with an important proportion of publications in engineering and new technologies (South Korea, 36%; Singapore, 26%; Taiwan, 42%). This taxonomy by field reveals the orientation towards applied research in East Asian countries, as well as the close links established between academic research and industry (Cardoza, 1997). Moreover, when comparing the number of patents per million inhabitants, Korea, Hong Kong and Singapore exhibited significantly higher figures than Latin American countries indicating the clear orientation of R&D systems in these East Asian countries towards applied activities (Table 6). The significant proportion of research in engineering and new technologies is aimed at supporting high-tech industries in their processes of continuous innovation and indicates the extent to which economic priorities are taken into account in the S&T research agenda of Asian countries.

24 24 Fig. 8 - Latin America (6) and East Asia (3): Distribution of scientific publications by field Basic Sciences Medical Sciences Engineering & New Technologies 100% 80% 60% 40% 20% 0% Argentina Brazil Colombia Chile Mexico Venezuela Korea Singapore Taiwan These data suggest that Latin American countries are lacking a more balanced basic and applied research agenda 11. However, balancing the research agenda doesn t mean that basic sciences should receive less support 12. Rather, applied research should be developed to meet the demands posed by economic restructuring and the adoption of a new pattern of insertion in the international trade based on the export of high value added products. Latin American countries could also find inspiring examples in the East Asian region useful for the design and implementation of policies to promote the collaboration of universities and public R&D institutions with the productive sector, and to encourage 11 It is necessary to recognize that in spite of the unfavorable conditions that countries like Argentina, Brazil, Chile, Colombia, Mexico and Venezuela have confronted in recent years, they have made important efforts to create national scientific communities able to meet the international standards of scientific production. However, Latin American countries in general not only exhibit lower investment in R&D but a lower output in applied research as measured by publications and patents in engineering and new technologies. 12 Mansfield in his studies on the impact of R&D on firm s productivity has noted the importance of long-term R&D: innovative output seemed to directly related to the percentage of its R&D expenditures devoted to basic research.

25 25 industries to create in-house R&D facilities. Governments are called upon to play a decisive role in the establishment of cooperative R&D-based projects that bring together private business and the public R&D laboratories in universities and research centers for targeted technologies and strategic industrial sectors. This is an urgent task to enhance innovation capabilities and strengthen the competitive advantages of Latin American economies 13. Summarizing, the continuous increase of public and private investments in high quality education and R&D in a stable and receptive environment, allowed East Asian economies to gain and maintain growth momentum during the last 30 years. As data suggest, investments in education and R&D not only have positive effects on knowledge accumulation (scientific publications) and industrial innovation (patents) but also seem to have a positive impact on employee productivity 14. In fact, as shown in Table 6, Hong Kong, Korea, Singapore and Taiwan, countries with high rates of investments in education and R&D, registered average growth rates of GDP per worker of 5.3%, 5,7%, 4.1% and 3.6% between 1980 and 1988 whereas growth rates in Latin American countries were stagnant or negative during the same period. Moreover, Table 7 shows how these increases in productivity permitted Asian countries to significantly expand their manufacture exports to the OECD member countries between 1970 and TABLE 7: LATIN AMERICA (6) AND EAST ASIA (6): COMPARISON OF OCDE IMPORTS OF MANUFATURED GOODS AND COMPOSITION OF 1993 IMPORTS OF MANUFACTURES (%) 13 As concluded by Nelson (1993) in his comparative analysis of national innovation systems, most countries having strong innovative and competitive capabilities in fields like fine chemicals, pharmaceutics, agriculture and electronics entertain solid and continuous relationships with R&D laboratories in universities and public research centers. For instance, Korean electronics industries have benefited from ongoing assistance from public R&D centers, through cooperative research projects and consulting services, to keep up to date technologically and enhance their problem solving capability. 14 As stated by Freeman (1982) "it would not be unreasonable to regard education, research and experimental development as the basic factors in the process of growth, relegating capital investment to the role of an intermediate factor."

26 26 Value of imports Composition of 1993 of manufactures by imports of manufactures origin in millions US$ textiles chemical Machinery Transport Other clothing Electronics equipment Argentina Brazil Colombia Chile Mexico Venezuela Korea Hong Kong Singapore Indonesia Malaysia Thailand Sources: World Bank, World Development Report In conclusion, the availability of highly skilled human resources and sound learning and innovation capabilities allowed East Asian economies the continuous growth of productivity required for the successful implementation of an export-oriented development strategy. As a matter of fact, the capacity to generate and adapt new knowledge made it possible for export industries in East Asia to rapidly introduce the technical progress necessary to compete in the international markets. Export activities, furthermore, allowed these economies to specialize and provided the financial resources permitting them to bring together, in a self-reinforcing relationship, the factors contributing to economic and social development. In fact, the rapid expansion of export activities created a sustained demand for highly skilled workers, allowing for increases in wages and consequently, for higher levels of domestic consumption and savings. The next section will develop a conceptual framework to address the question of how exports, investments and technology progress were combined, in the framework of an outward looking strategy, with other factors like savings, employment, education,

27 27 research and development (R&D), management techniques and institutional arrangements, to create synergetic relationships conducive to a better economic output. The virtuous circle of learning, innovation and growth Cross-economy analyses indicate that rapid factor accumulation, both of physical and human capital and labor growth, can explain most of the output growth of East Asian economies (Young, 1993; Barro and Lee, 1994; Krugman, 1994; Birdsall et al., 1995; Rodrik, 1995; Easterly and Levine, 1996). These studies have also made insightful contributions to our understanding of the role played by post-war land reforms 15, industrial policies and investments in health, housing and basic education to the successful adoption of the outward-oriented industrialization strategies in the Asian region. However, these attempts to establish the proximate sources and determinants of growth in East Asian economies have overemphasized the role of single factors or focused on linear causalities that conceal the multidimensional and systemic nature of economic and social development processes. As a matter of fact, in the absence of favorable conditions not a single factor on its own could satisfactorily account for the rapid economic growth of Asian economies. Although most specialists agree on the importance of factors like capital deepening, level of education of the work force, technology upgrading and macroeconomic and political stability in Asian development, very little attention has been given to explain the complex connections established between the multiple variables involved, at the micro and macroeconomic levels, in the development process. An adequate explanation should take into account the right combination of factors and policies. Such an inclusive view of the 15 In the words of Perkins (1986): Japan, South Korea, and Taiwan all experienced a major transfer of land from landlords to peasant tillers in the late 1940s and the early 1950s. Furthermore, the transfer of land titles, particularly in Korea and Japan, involved a substantial element of confiscation to the landlords. There was, as a result, significant redistribution of wealth to former tenant farmers.

28 28 determinants of growth would prevent unnecessary biases that create misleading analysis and ultimately render policy recommendations worthless. Toward this purpose, the present study centers attention on the development of a more comprehensive conceptual framework helping to explain the linkages established among the elements entering into the processes of learning, innovation and economic growth. This integrative approach would allow for a better understanding of the connections between technological, managerial, organizational and institutional upgrading and the overall economic and social development processes. The next section will support the argument that the spectacular momentum and sustainability of East Asian growth are the result of the establishment and strengthening of a virtuous circle of learning, innovation and economic growth. Learning, innovation and growth: anatomy of a miracle Nowadays productivity and the international competitiveness depend heavily on the rapid accumulation of knowledge and the effective diffusion of technologies and best practices. Strengthening a firm s capacity to identify, incorporate and adapt technologies has become a central task for countries committed to opening and liberalizing their economies. Indeed, the successful adoption of an export-oriented development model in East Asian countries has been made possible not only by strong investments in physical and human capital but also by the adoption of incentive policies oriented to the continual improvement of learning and innovation capabilities. Industrialization strategies implemented by Japan, South Korea, Taiwan, Singapore, Hong Kong and more recently Malaysia, Indonesia and China have been based in large part on the rapid diffusion and assimilation of foreign and domestic technologies, continuous innovation in management and organizational techniques and the adoption of more favorable institutional frameworks for the development of economic activities. Indeed, the successful implementation of outward-looking strategies by Asian economies has been possible because of the

29 29 establishment of a complementary and self-reinforcing pattern of knowledge accumulation, industrial innovation and productivity growth. Diagram 1 proposes a possible configuration of the virtuous circle of learning, innovation and economic growth. This policy approach tries to put into perspective the frequently overlooked relationships among the macro and micro-economic factors involved in the learning processes and international trade. To facilitate the understanding of the dynamics, this schema offers a rapid, though admittedly simplified story that highlights only the principal connections among the major determinants of economic growth. As described in the previous section, the right side of Diagram 1 emphasizes the existing linkages between the ability to assimilate new entrants into the labor market, the effect that higher salaries have on private consumption and the impact that domestic savings have on productive investments. The left side of Diagram 1 brings together the main factors contributing to the ongoing absorption of knowledge required to create the learning and innovation capabilities needed for achieving a better economic performance. As implied in Diagram 1, the resulting expansion in export of higher value-added products enables the closing of the virtuous circle by providing the financial resources required to accelerate the rate of innovation and generate new, better-paying jobs. Thus, this integrative approach supports the notion that economic performance depends on the establishment of selfreinforcing relationships among education, R&D and learning, the principal activities leading to the expansion of the knowledge base, innovation processes, productivity growth, and international trade.