The Role of Higher Education in High Tech Industry Development: What Can International Experience Tell Us?

Sachi Hatakenaka Independent researcher (sachi@alum.mit.edu)

Introduction This paper examines how different roles have been played by different higher education institutions in the development of high tech industry, and how different institutional characteristics influenced the roles they played. The key objective of the paper is to propose a framework for understanding different types of institutions. More specifically, the paper addresses the following three questions:

1. what are the characteristics of higher education institutions which define the nature of their contributions to high tech industry?

2. what are the different roles they play in the development of high tech industry? 3. how do these institutional characteristics develop?

The ultimate objective of the paper is to draw lessons for the future for higher education systems, particularly in developing countries. The background is that there is an extraordinary level of policy attention being given to higher education today. The reason is clear; with the rise of knowledge economies and science-based high tech industries, higher education is considered key to economic development. At the same time, the world has observed that high tech industry can emerge in a diverse set of countries including those with limited industrial bases or low levels of economic development, such as Ireland, China or India. It is not surprising that innovation policies to create and support high tech industries are a priority in almost all OECD and many developing countries. Developing an appropriate higher education sector is becoming a critical component of such policies. There is much written on the role of American universities in high tech industries and more generally in their industrial development, including some that compare their roles against those in Europe or Japan. For other countries, there are studies which conclude that a significant role was or is being played by well trained engineers and scientists, and by implication higher education systems. However, few studies go beyond referring to higher education as one of multiple – albeit often citical – factors. The comparative higher education literature describes similarities and differences across national higher education systems, but there tends to be little emphasis on the different roles they play in economic development. Lester (2005) makes an important step in arguing that different types of industrial transformations require different roles to be played by universities based on a comparative study of several countries. There is also an emerging literature on university-industry relationships in developing countries (Chun and Kenny 2007, Kroll and Leifner 2008, Ma 2007, Wu 2007, Yusuf and Nabeshima 2007), with some papers beginning to discuss the role of universities in economic development in a more specific way (Mazzoleni and Nelson 2007, Leifner and Schiller 2008). Mazzonleni and Nelson provide an in-depth analysis of the role of public research in economic catch up, and argues for the increasing importance of research for

1

economic catch up particularly in application oriented sciences and engineering (2007). They also provide a helpful summary about the nature of American research universities and their economic roles. Leifner and Schiller contributes to the debate through introducing the notion of ‘academic capabilities’ to think about the roles universities play in technological capacity development (2008). This paper seeks to extend our understanding about the role of higher education in the development of high tech industry from a range of countries. Wherever possible, this has been complemented by country specific literature on higher education systems and their development. The main emphasis has been given to two ‘recent’ high tech industrial sectors: ICT (including hardware as well as software) and biotechnology. The point of this paper is not to provide a full contrast of the paths taken, nor to examine all the contributing factors to high tech industry development. That high tech industry took off for a variety of reasons is taken as a given, and the paper focuses on what possible specific roles higher education has played within that context. The paper examines differences in high tech industry capabilities among countries – but only to the extent that they are related to different roles played by higher education systems. In order to explore the variety of roles played by higher education at different stages of industrial development, four sets of countries were included in the paper:

(a) the US, which has held a leading position in both industries from their birth; (b) Japan and Finland which entered the global competition in high tech products based on

significant industrial capability that already existed; (c) a group of late-developers including Taiwan and Korea which nonetheless were

building industrial capabilities around the time when these industries were born; and (d) a group of countries which developed high tech industry from a much less developed

industrial base (Ireland, Israel, China and India). The paper attempts to draw on comparative historical data to the extent they exist, so that the comparisons can be made not only across countries but also across time. The Framework for Understanding Different Institutional Characteristics What are the key dimensions that differentiate higher education institutions meaningfully? In this section, a framework for differentiating institutions is proposed using three different dimensions that measure institutional characteristics: responsiveness, basic science orientation, and selectivity (Fig 1). The first dimension, responsiveness, distinguishes institutions in terms of their ability and inclination to respond to changing practical and industrial needs. The second dimension, basic science orientation, distinguishes institutions in terms of the level of commitment to fundamental science. The third dimension has to do with the level of selectivity with which students and staff are recruited into the institution, and distinguishes elite institutions with those that are open to the greater mass. [Fig 1]

2

These three dimensions emerged in the course of the review of the various country cases, but were developed also based on prior academic work. The first two dimensions, responsiveness and basic science orientation, owe much to the work of Stokes (1997), in which he proposed two similar dimensions to distinguish between four different types of research orientation. Using these two dimensions, four different types of institutions could be distinguished as shown in a two by two matrix in Fig 2. [Fig 2] The top right cell is occupied by ‘relevant research’ universities, which conduct fundamental research, but with an eye on their relevance to the society. They are committed to fundamental science, but are also concerned about being useful and strive to remain responsive to the socio-economic needs. Surprisingly, there are not many institutions that fit this category. One primary example is a group of American research universities, which had developed embracing the value of relevance (Geiger 2004b, Rosenberg and Nelson 1994, Mazzoleni and Nelson 2007). When the American universities became the envy of the world because of their contributions in high tech industry such as biotechnology and IT industry, the world initially looked to all American universities as if they were all the same. Over time, it became clear that even within the American system, there were different types of universities, and that only a group of research universities were playing proactive roles in industrial innovation generally and in the development of high tech industry in particular. Some Israeli universities also appear to belong to this cell, though perhaps with less emphasis on basic science and greater focus on applied research. The top left cell is occupied by ‘basic research’ universities, which are driven principally by the core values of fundamental science, with little interest or capacity to respond to external needs. This is where the classic ivory tower universities with well developed research capabilities belong; indeed the vast majority of research universities in the world belonged to this cell, at least until recently, when economic relevance became a global catch phrase. These institutions value cutting edge science, reward scientific discoveries, and encourage their academics to actively define the direction of scientific research. Scientific developments were to be independent of practical requirements, as the latter were seen to be ‘corruptive’ influence for science. Indeed, until the 1990s, this cell was what most higher education institutions aspired to belong to. Many Japanese national universities developed to fit into this category, in spite of their origins as practically relevant universities - as will be discussed later. The bottom right cell comprises ‘practically oriented’ institutions. The main characteristics of these institutions are that they aspire to meet the practical skills and knowledge needs of the economy. They offer courses relevant to employers and often conduct consulting and application oriented research with and for employers. However, they have little interest or capacity for furtherance of generic scientific knowledge or fundamental research. Examples are diverse; they range from the Grandes Ecoles in France (though they began to develop basic research capacity recently), designed to provide elite professional education, and many polytechnics and their equivalents in other countries. Both Ireland and Finland reformed their higher education sectors by introducing this type of institution. Historically, the Japanese national universities would have belonged to this cell at their founding as they were established

3

to provide key technological skills needed for national modernization. On the other hand, their commitment to bring ‘science’ into practice would have made them higher up in the quadrant in comparison with traditional trade schools which would have focussed on any practical skills rather than scientific ones. Many Chinese universities also appear to have developed towards the right hand cell to become much more pratically-oriented over the last 20 years. Some institutions fit this cell naturally by the virtue of their emphasis on application oriented subject mix; engineering institutions such as the Indian Institutes of Technology are good examples of those. The bottom left cell is occupied by ‘teaching-focussed’ institutions, which are neither practically oriented in their teaching nor research-oriented. Liberal arts colleges in the US are good examples as they are committed to teaching broad based curricula, with emphasis on generic skills rather than vocational or professional content. They define their identity in terms of their commitment to teaching and conduct hardly any research. There are other types of teaching institutions, for instance, those that are more academically oriented. A vast majority of universities in developing countries fall into this category by default, simply because they do not have resources to become research active. In many such cases, their teaching is defined broadly by academic disciplines. Some institutions occupy space straddling the categories. For instance, many Japanese, Korean, and Taiwanese universities are teaching institutions but with greater practical orientation than others owing to their large offerings in subjects such as engineering. The third dimension ‘selectivity’ differentiates further the nature of institutions in each of the four cells. The more selective the institutions are in terms of incoming students, the more elitist they are. As such, Grandes Ecoles are differentiated from the bulk of polytechnic type institutions, as they are some of the most highly selective institutions, while the bulk of polytechnics are not. Selectivity is often a system-level characteristic; some higher education systems are explicitly selective, characterized by highly competitive entrance examinations to allocate talents across institutions, as in Japan and Korea. Other higher education systems are much less so, and are based on the notion that all universities command equal status, as in Germany and many other European states. As will be discussed later, selectivity appears to be an important characteristic which strengthened the roles institutions could play. Different Roles Played by Different Types of Institutions This section describes the roles played by the different types of institutions in the development of high tech industry, drawing on the experiences of the countries listed above. No single type of institutions appears to be necessary for the development of high tech industry. Rather, different countries have seen different types of institutions playing different roles. The footprints of university roles are sometimes visible in the shape and nature of the resulting industry, but at other times industry developed incorporating university contributions in a way that became almost invisible.

4

Relevant Research Universities. It is well established that some American research universities played critical roles in the birth and development of biotechnology and computer industry. For instance, fundamental scientific discoveries in university laboratories and resulting patents (such as Cohen and Boyer’s discovery of the recombinant DNA technique) led to the development of biotechnology as a whole new industry with strong linkages with academic scientific research (Henderson, Orsenigo and Pisano 1999). The early evolution of the biotechnology industry was characterized by a number of new firms, many of which were academic spinoffs – usually with academics working in conjunction with professional managers, backed by venture capital. The early academic inventors were labelled as ‘star scientists’ as they were key players both in basic science and in their applications and had a special role mainly because of the significant amount of tacit knowledge associated with the original scientific discoveries (Zucker and Darby 1996, Zucker and others 1998, Zucker and others 2002). The norms in American universities that permitted academics to get deeply engaged in commercialization, which had developed early given their value for being useful, played a key role in this process (Henderson, Orsenigo and Pisano 1999, Eskowitz 2002). American universities also played key roles in the development of the computer hardware and software industry, though their visibility varied over time (Bresnahan and Malerba 1999, Mowery 1999, Rosenberg and Nelson 1994). For instance, universities were actively engaged in the early development of computers in the 1940s (and indeed during this period, European universities such as Manchester, Cambridge and Berlin were as active as American universities). They also played a critical role in the development of microcomputers as exemplified by spinoffs from MIT, Stanford and University of Texas (only Cambridge University in the UK played a similar role in microcomputers in Europe). There is evidence that industry values the less ‘visible’ contributions made by universities. One survey found that industry saw general research findings or instrumentation and techniques developed through research as more important university contributions than prototypes (Cohen and others 2002). The same survey found that publications, meetings and conferences, informal interactions and consulting were more important than patents, as mechanisms through which industry gained access to university knowledge. Further, industry reported that university research was helpful in solving problems, rather than in inspiring and triggering new research. These results have to do with contributions made by American universities more generally, which are likely include a whole range of institutions including basic research institutions. Nonetheless, it is important to note that what is valued is not so much ‘application’ capabilities in universities, but ‘relevant’ research capability. There is also some evidence that the development of the software industry was greatly facilitated by an early establishment of computer science as a new discipline in universities (Mowery 1999). Universities were thereby capable not only of contributing to key knowledge formation but also of organizing and delivering education programmes to supply updated skills. Indeed, the American universities’ ability to create and legitimate computer science as a new field was unparalleled by European or Japanese universities (Mowery 1999).

5

In Israel, universities played similar roles, by supplying highly skilled human resources and in being a source of leading technologies not only in ICT but also in biotechnology and agriculture related research (de Fontenay and Carmel 2004, Breznitz 2007). Soon after its formation, the Israeli government made significant public investments in application oriented scientific research in universities, public research institutions and in the military. Their universities became well established scientific centres early, and today has world’s highest publication record in volume per capita and in their citations (Table 1 and Table 2). This did not stop them from becoming application-orientated also, as university-industry relationships had long been well established (Bresnitz 2004), and as their unusually high patenting records show with three universities among the top six patent issuers in the country (Bresnitz 2004, Trajtenberg 2001). Israel became one of the first R&D sites abroad for US multinational companies with Motorola establishing a research unit in 1963 (de Fonteney and Carmel 2004). Universities’ contribution tends to be overshadowed by that of the Military in the stories told about Israel’s high tech industry, but they have contributed to the emergence of an entrepreneurial sector directly and indirectly, through spinoffs as well as science parks, some of which had been established as early as the 1960s (Roper and Grime 2005). It is no accident that the resulting high tech industry in Israel was highly R&D oriented, and had cutting edge technological content. This provides a sharp contrast to other emerging economies. While in Ireland or India, their software industry largely grew providing software services, Israel’s software industry developed on the basis of niche software products with global applications. Israel’s innovation capabilities as judged by its patenting performance, for instance, lags behind US or Japan, but is comparable to Finland or Taiwan (Trajtenberg 2001). Israel’s ICT industry has a marked concentration on upstream R&D, compared with the Finnish ICT industry which covers a wider value chain ranging from R&D, production to marketing and distribution, or the Irish case, which has hardly any R&D (Roper and Grime 2005). In summary, relevant research universities appear to contribute through:

• conducting basic research in fields relevant to the developments in industry, and disseminating such knowledge through a varieties of activities such as patenting and licensing, academic spinoffs, consulting and advising

• providing updated skills in areas relevant to industry, both in undergraduate education and in advanced research training for masters and PhDs

Basic research universities. Since basic research universities are characterized by an academic ethos of scientific autonomy, it would not be surprising to find that basic research universities play less proactive roles than relevant research universities. Nonetheless, they can play sufficiently powerful roles both in research and education to support the development of high tech industry with high technological content, as is demonstrated by the Japanese example. It is a group of their national universities that functioned as basic research universities. One caveat is that the Japanese universities’ roles depended highly on the proactive nature of their emerging home industry. Some analysts coined in the term, ‘receiver active paradigm’ of technology transfer to describe this phenomenon; it was the ‘technology receiving’ industry which actively defined

6

how and when technology transfer took place rather than the universities (Kodama and Suzuki 2007). There is no question that Japanese universities in general played a critical role in education by supplying large numbers of engineers and scientists at undergraduate and masters levels, which were essential for the birth and development of the electronics industry. A group of Japanese national universities were particularly important in producing high calibre scientists and engineers as they were highly selective and had large science and engineering faculties. However, companies had to develop significant capability to train them to obtain the needed skilled through extensive in-house training. A form of division of labour developed whereby universities provided selection and general academic training while companies did the necessary specialist training, even for R&D personnel. Japanese companies have also been highly active in using universities to acquire new scientific knowledge. One unusual mechanism that brought university academics and industry together was the industry practice of sending their own staff to universities, not for any specific education programmes, but more generally to learn, sometimes for a whole year. This was often motivated by their desire to learn about some fields of science, in which industry had limited expertise (Hicks 1993, Darby and Zucker 1996). Accordingly, in biotechnology, ‘star scientists’ actively worked with company employees in their university labs – with resulting higher levels of co-authorships than in the US (Darby and Zucker 1996). Other researchers also find that Japanese academics have a level of relationships with companies equivalent to the US or UK, at least as measured by co-authorships (Pechter and Kakinuma 1999), but that companies had to be proactive in drawing scientific and technological information from the universities (Kodama and Suzuki 2007). In the early development of biotechnology, universities played a negligible role in forming new companies Japan, in a striking contrast to the US and UK where academic spinoffs were highly visible and dominant (Henderson and others 1999). This was partly because of weaknesses of Japanese science at least at the outset; there were literally no researchers in Japan working in genetic engineering at the time of the scientific breakthroughs (Darby and Zucker 1996). But even when they did catch up in science, universities as organizations did not permit their civil servant ‘star scientists’ to work in the same way in new companies (Darby and Zucker 1996). The Japanese universities were generally slow to respond to emerging industrial skills needs, for instance in new areas such as software related engineering, semiconductors and biotechnology (Baba and others 1996, Mowery 1999, Kobayashi 2001, Darby and Zucker 1996). They were also slow in establishing relevant research training. Although masters programmes in engineering became commonplace by the 1980s, PhD programmes in science and engineering remained small and unattractive for industry (Clark 1995, Shimizu and Mori 2001 also see Table 3). PhDs from universities were too narrow and too academically oriented for industry. Thus, industry recruited bright potential researchers with less academic training and trained them in-house – using the uniquely Japanese system in which doctorates were awarded by universities to dissertations which were written mainly in the companies, without detailed supervision by academics (Shimizu and Mori 2001). The companies took the extreme route of developing in-house training for cultivating PhD level researchers.

7

In summary, basic research universities can contribute to high tech industry in developing scientific capacity and in providing scientific information, though in the case of Japan, this required much more proactive work on the part of industry. In education, such universities can also provide large numbers of scientists and engineers as needed by the industry, though their training may not be fine tuned in terms of specific industrial technological needs. In Japan, their role in education appeared to be facilitated significantly by the fact that the universities were highly selective; selectively in universities was used actively by industry for identifying high calibre talents. Practically oriented Institutions. Practically oriented institutions played a unique role in offering engineering and science skills which were more specialized and directly relevant to industrial needs. In Ireland and Finland, professional institutions were created as an alternative to conventional university education, which was seen to be unresponsive to industrial needs. The institutional responsiveness was particularly important when new disciplines such as computer science emerged, and appeared to give countries a competitive edge, as was clear in the case of Ireland. In Ireland, the government led expansion of technically trained manpower in the 1970s and 1980s played a key role in attracting multinational companies (MNCs) including in ICT. The starting point was the consensus that emerged early among policy makers that technical education was critically needed, but that universities were overly academic and that different institutions were needed to provide the critically needed technical manpower (White 2001). Thirteen Regional Technical Colleges and two National Institutes of Higher Education were established in the 1970s for this purpose, representing the bulk of expansion in the tertiary sector. They were established specifically to be responsive to economic needs, and are today known to have well-established practices, for instance, in assessing industrial needs or in obtaining industrial inputs in curricular content (Breznitz 2007). Finland also made a decisive move in creating a set of responsive institutions when they established 29 polytechnics in the 1990s. By 2000, nearly 60% of entrants to higher education were going to polytechnics. Though it is early to understand the full impact of polytechnics in Finland, their performance reviews have been positive, with indications that their expansion coincided with the timing of major industrial needs, particularly in telecommunications (OECD 2003). In one survey of high tech industry in Northern Finland, 47% of companies said that polytechnics were important, compared with 38% which thought a university with a known regional orientation to be important and 10-20% that found other universities in Finland important (Juahiainen 2006). Both Korea and Taiwan invested in application oriented research capacity in government research institutes which played significant roles in developing domestic technological capacity within their high tech industry. The fact that such capacity was built outside universities was perhaps not surprising given that it is faster and easier to concentrate resources in dedicated institutions. However, such decisions did have consequences, particularly in the case of Korea,

8

in delaying the development of research capacity in universities - except in one notable exception of KAIS as discussed below. Taiwan’s experience in high tech industry cannot be told without the story of a public research institution called Electronic Research Service Organization (ERSO), a branch of Industrial Technology Research Institute (ITRI), which was at the heart of Taiwanese experience of acquiring technological know-how in semiconductors. ERSO orchestrated the first major technology transfer agreement with RCA in the 1970s, and subsequently developed an integrated circuit (IC) fabrication technology to establish a spinoff company, which in turn played a critical role in the diffusion of technological know (Wade 1990, Saxenian 2004). Since the Taiwanese industry comprised many small to medium scale enterprises (unlike Japan or Korea), they did not have the capability to undertake significant R&D in-house, and public investment in applied research was essential for their technological upgrading (Wade 1990). The Korean experience of investing in application oriented research capacity in government research institutes is somewhat less visible, though important in helping to build corporate research capacity. One example is the Korea Institute of Science and Technology (KIST) established in 1966, with Korean scientists and engineers brought back from abroad, one of the early efforts to reverse the brain drain. KIST was designed to undertake contract research both with industry and government, and was particularly active in facilitating technology transfer from foreign firms to local firms (Mazzoleni and Nelson 2007). KIST is also an interesting – and exceptional - example as it later became amalgamated with the Korea Advanced Institute of Science (KAIS), a graduate level education institution established in the 1970s specifically to create high level skills of industrial relevant, to fill the important gap in Korean higher education (Mazzoleni and Nelson 2007). It is possible – though not common - for research capacity developed in government research institutes to contribute directly to the evolution of higher education institutions. Teaching-focussed Institutions. In the early stage of high tech industry development, companies often recruited generic science and engineering graduates who were not necessarily trained in a specialization relevant to the industry. Their provision was a role predominantly played by teaching institutions in Japan, Korea, Taiwan, China, and India, mainly because many institutions simply had not yet developed research capacity. In all cases, governments emphasized science and engineering. Beyond such a general statement, there is a significant divergence in terms of what happened. In Japan, the government aggressively promoted science and engineering undergraduate programmes in response to growing industrial demand in the late 50s through the early 1960s, not only in national universities, but also in private universities which were largely teaching oriented (Nakayama 1995). Most private universities in Japan were teaching focussed institutions with limited resources to undertake sciences or engineering or research. It was through special government subsidies for science and engineering, which provided up to half of undergraduate student costs and two thirds of associated investment costs, that they managed to establish science and engineering programmes (Nakayama 1995). This push no doubt helped

9

Japan obtain a large number of scientists and engineers so that they were 27-28% of all undergraduates in the late 1960s through the 1980s (Table 4). Korea and Taiwan made earlier moves in emphasizing science and engineering relative to their industrial development – as they both had met massive brain drains in early days. In Korea, More than a third of undergraduate enrolments were in science and engineering by the mid 1960s (Table 4), and the proportion went up further to over 40% in the subsequent period in spite of the massive expansion of the higher education sector that was taking place. Korea today has one of the highest tertiary enrolment ratios in the world of 91% (Table 5), with one of the highest proportions of science and engineering (Table 6). In Taiwan, the proportion of science and engineering enrolments was about a quarter through the early 1970s, and gradually increased from the mid 1970s through the 1990s to over 40% (Table 7). In contrast, even in 2005, India’s engineers comprised only 6% of total tertiary enrolments, and even when combined with scientists, the proportion is significantly lower than others at 20% (Table 6). There is no magic number in terms of the share of science and engineering. One factor whose influence appears to go well beyond that of ‘numbers’ is selectivity. Japan, Korea, Taiwan, China and India share a common characteristic in having highly selective institutions which recruited high calibre students into engineering nationally, and which have set quality standards in engineering schools more generally. Further, these institutions invariably had a significant share of their institutional profile in engineering, and in that sense, their origins were closer to ‘practically-oriented’ institutions than generic teaching-focussed institutions. At the time of the take off, graduates from these selective schools were not in high demand in the labour market; they represented cheap high calibre resources with training in science and technology. Different Roles Revisited In this section, six different roles played by higher education institutions are summarized, mainly to highlight the fact that the nature of these roles depends both on the institutional characteristics and the industrial contexts. Education Roles. Different types of institutions have played different educational roles toward high tech industry development. Generic teaching institutions and basic research universities contributed to the production of generic scientists and engineers, and left the companies to take the responsibility for developing specialized knowledge and skills. This role was particularly pronounced when there was government intervention in preferentially expanding science and engineering programmes. Professionally oriented institutions played a more proactive role in keeping up with changing industrial skills needs and in supplying specialist knowledge and skills as industry evolved. Relevant research universities were able to play the similar proactive role at a higher level – in their research training. Research Roles.

10

Similarly, different types of institutions played different research roles as well. Academics in relevant research universities played the most proactive role in conducting research and in working directly with its applications, helped by universities’ organizational capacity to promote and support consulting, patenting/licensing as well as spinoffs. Basic research oriented universities played a less active role in furthering industries’ commercialization goals, but helped companies acquire better understanding of new scientific fields, provided advice and conducted joint research, albeit more often on universities’ own terms. Other institutions such as government research institutions or professionally oriented universities helped industry to acquire existing technologies through application oriented research. Spinoffs. University ‘spinoffs’ appeared to play a visible role when there was a significant capacity/knowledge gap between universities and surrounding industry, but the nature of this capacity/knowledge gap could be very different, as demonstrated by the case of biotechnology in the US and ICT in China. In biotechnology in the US, the knowledge gap had to do with fundamental scientific discoveries with significant related tacit knowledge. In China, the nature of the knowledge gap in ICT appears to have been about old technology which was nonetheless new in China. The capacity gap may also have arisen because of the highly selective nature of universities; universities may have been better positioned to assemble a group of high calibre individuals to lead and work in new ventures. There were significant examples of academic spinoffs associated with universities with relevant research capabilities. Thus, in biotechnology in the US, academics from research universities got directly involved in spinoffs, and there were similar stories in different fields in Israel. In Ireland, where universities did not develop research capacity until late, owing to the lack of government funding (Sands 2005), academic spinoffs were generally not frequent. However, one computer science department developed an early research capacity with European funding, and became known for its academic spinoffs (Breznitz 2007). A very different experience was found in China, where universities had become active in creating enterprises since the late 1980s, even when they had little research capability. Though they are sometimes described as ‘spin-offs,’ they are significantly different from the normal practice in that they are owned and managed by universities (Eun and others 2006). Some of these companies have been spectacularly successful, and include three of the most successful PC companies, Lenovo, Founder and Tongfang, which were enterprises created by the Chinese Academy of Sciences, Beijing University and Tsinghua University respectively. Some 40 university enterprises are already listed in stock markets in China and Hong Kong (Eun and others 2006). Interestingly, the knowledge content in these companies did not derive from significant scientific research, rather, it was the mechanisms through which skilled personnel moved from universities to the commercial sector (Chen and Kenney 2007). These enterprises were one simple mechanism through which universities could contribute to industrial capabilities in an

11

environment of very limited industrial capability (Eun and others 2006). They are closer to Japanese university start ups in the early phase of industrial development (Odagiri and Goto 1996) where academics could behave as arbitragers of Western technology and were in a good position to create companies, given an under-developed Japanese industrial context. It is not clear how long this practice of university enterprise will continue in China. Both the government and many universities may be going through a re-thinking process, as many enterprises have not been successful and managerial responsibilities are increasingly demanding (Ma 2007, Kroll and Leifner 2008). It is likely that China’s university enterprise experience was a phenomenon dictated by the specific context of underdeveloped industry and a high concentration of talents in universities. Science Parks. Taiwan provides the most visible example of a highly successful science park, which played a critical role in the development of its semiconductor industry. However, universities were not the main feature of the Science Park. In 1980, the Taiwanese government established a science park in Hsinchu, close to two of Taiwan’s best technical universities, and relocated ERSO to the park (Saxenian 2004). There was a certain division of labour in that applied and relevant basic research were undertaken by ERSO, while the universities appeared largely to supply graduates as manpower to the park. In China, many high tech zones, industrial and science parks have been established close to universities by the central government, local authorities, as well as the universities. The nature of their dynamic evolution is only beginning to be documented, though the diversity in approach is already evident (Chen and Kenny 2007). The one established by the local authorities in Zhongguancun in Beijing, close to the Chinese Academy of Science as well as to Beijing and Tsinghua Universities, is considered one of the most successful in attracting MNC R&D facilities. Another well known example is the one established by Tsinghua University on its campus, which has many MNC R&D units as well as spinoffs from the university. An interesting variant is Shenzhen whose High-Tech Industrial Park has been used to attract established universities to create a virtual campus. Licensing. This is one function that seems to be linked strongly to relevant research capacity in universities. However, its effectiveness depends also on the level of development of intellectual property related institutions in the country. In the celebrated case of the US, the patenting practice evolved as one of many mechanisms for technology transfer from universities, predicated upon huge government funding in relevant scientific research (Mowery and others 2004, Rosenberg and Nelson 1994). However, it is increasingly clear that for most universities in the US, patenting/licensing is a net drain rather than a source of additional income (Thursby and Thursby 2007). The wider acceptance of WTO and other international regimes related to intellectual property will undoubtedly have consequences for the paths that future developing economies can take in

12

catching up. While it is important for developing countries and their universities to increase their understanding about intellectual property, indiscriminate opening of technology transfer offices in universities with a view to patenting their inventions may be unwarranted. Brain Drain/circulation. Brain drain turned into brain circulation appears to play an increasingly important role in the introduction of high tech industry in developing countries. This undoubtedly reflects the increasingly global nature of the labour markets in developed countries, many of which are actively recruiting professionals from developing countries in areas of key skills shortages such as ICT. Such a brain drain can turn into brain circulation, once appropriate domestic conditions are developed for industrial development, when ‘brains’ that are enriched with international experience can return to their home countries. Indeed, one feature of Taiwan’s high tech development which distinguishes it from those of Japan or Korea, is the visible role played by their diaspora of engineers, particularly in Silicon Valley, and how their ‘brain drain’ turned into ‘brain circulation’ linked directly to the birth of high tech industry (Saxenian 1999, 2002, 2004, 2006). Almost eighty percent of the Taiwanese diaspora professionals in Silicon Valley had gone to the US to study and remained there for work (Saxenian 2002). Early generations of the diaspora were instrumental in informing the Taiwanese government about Silicon Valley and its unusual dynamism, the reason why the Taiwanese had an early policy for building a cluster rather like Silicon Valley (Saxenian 2004). In 1980, they invested in a science park in Hsinchu, close to two of Taiwan’s best technical universities, and relocated ERSO (Saxenian 2004). The science park became an active recipient of the returning diaspora with 40% of companies there started by returnees. (Wade 1990, Saxenian 2004). It is well known that Taiwan suffered from a massive brain drain which started in the 1950s and lasted until the 1990s. Interestingly, this did not stop the government from organizing programmes of overseas study to complement Taiwan’s own education system. In 1975, there were 2300 students studying under government auspices; in 1986 7000 were in such a programme (Wade 1990). This was happening in the midst of the massive brain drain, which was estimated as 20% of the all engineering graduates in the late 1970s (Hou and Gee 1993). Instead of stopping to send people, the Taiwanese government simply stepped up efforts to keep in touch with the diaspora community (Wade 1990), the benefit of which was finally felt in the 1990s when the brains started to return. China and India followed Taiwanese experience in brain drain to the Silicon Valley. In the survey of Silicon Valley professionasl, over half of Indians and nearly 80% of mainlanbd Chinese had originally come to the US for their education (Saxenian 2002). China was able to benefit doubly from diasporas, as hsinchu-Silicon Valley-Shanghai increasingly became connected to broadly defined overseas Chinese engineering including those originally from Taiwan (Saxenian 2006). In the early 2000s, the Chinese government also stepped up effort to recruit back the overseas Chinese. For India, the dominance of Indian entrepreneurs not only in Silicon Valley but more broadly in US high tech companies is well documented (Saxenian 2002, Wadha and others 2007). This highly entrepreneurial diaspora played a key

13

‘reputational intermediary’ role in connecting Indian firms to client US firms, or in brining American multinationals to India (Kapur 2002, Kapur and McHale 2005, Saxenian 2006). Even in the brain drain, selectivity appears to have played a role. Having a competitive system of higher education entrance meant that some engineering colleges and universities had an elite status – and it was largely graduates from such institutions who became critical members of the diaspora and formed vibrant alumni networks (Saxenian 2004). Such networks were important for solving technical problems, as former teachers as well as former classmates were important sources of technical information. In Taiwan, China and India, the brain drain was often triggered by prospects of graduate education abroad. In all three cases, graduates from the most select institutions actively pursued graduate education opportunities abroad and succeeded in finding places most notably in American research universities. It is easy to imagine that these graduates would have benefited from having been at elite home institutions in their search for the best graduate schools in the US, with active peer interactions providing information about studying abroad, and with letters of recommendation from home institutions having credibility. It is almost as though graduate schools in the US became a natural extension of the domestic education system. It is also interesting to see different approaches taken by different governments. The Taiwanese were sending fellowships abroad even at the height of their brain drain. They apparently did so with the belief that they will gain in the end. A decade later, the Chinese government was doing something similar. Both valued the foreign education sufficiently to send them abroad in the first place, but both also worked hard to bring them back. It is almost as though for the Taiwanese and Chinese governments, foreign institutions, particularly in the US, were a critical component of their higher education system, providing the kind of education that their domestic institutions could not provide. India, which was another country suffering from a severe brain drain particularly of high calibre graduates, was somewhat hostile to the idea of expatriate Indians, and also slower in orchestrating the use of their expatriate brains. (Saxenian 1999, 2006). How did the universities develop over time? While all the four types of institutions (relevant research, basic research, practically-oriented and teaching-focussed) played some economic role in the development of high tech industry, ‘responsive’ institutions in the right hand cells of fig 2 appear to play much more proactive and direct roles in helping high tech industry to emerge and evolve. And yet, such types of institutions do not seem to emerge naturally. In all cases, government had played a critical role in founding them and in influencing their missions and orientations. In this section, the experiences of different countries are reviewed, to examine the manner in which different types of institutions developed or evolved so to understand underlying dynamics. Building Responsiveness.

14

US. It was no accident that a group of responsive institutions emerged in the US. A whole group of American universities, the ‘land grant colleges’, were founded in the late 19th century, with an explicit objective to serve the practical needs of the community. The federal government supported their establishment through special grants, and continued to support their growth through funding some of their service activities such as agricultural extension. This group was a wide mix ranging from generic teaching institutions such as Cornell or Purdue to industrial training oriented institutions such as Massachusetts Institute of Technology. That universities could and should be useful to the society became a belief shared by government, the public and universities broadly. The ‘responsiveness’ of American universities in teaching was well established in these early days, and it was their responsiveness that brought about new subjects such as electrical, chemical and aeronautical engineering (Rosenberg and Nelson 1994). Before the second world war, private foundations provided key support to some universities, helping them to undertake research that was relevant and useful to the society (Geiger 2004a). However, it was the spectacular success of using science to develop key military technology during the second world war that brought about a new level of commitment to ‘useful scientific research’ for the government, universities and the public alike. The post-war government research funding was massive, but also ‘mission-oriented,’ coming from key agencies with application interests such as defence, health and energy (Rosenberg and Nelson 1994, Geiger 2004b). The role played by the Defence Advanced Research Project Agency (DARPA) is funding application oriented basic research was legendary; so much so that the American National Academy of Science recommended a creation of ‘DARPA-like’ agency in energy to ensure the continued competitiveness of American science (National Academy of Science 2005). Ireland and Finland represent interesting cases in which their universities appeared to have moved towards right hand cells to become more responsive. In each case, the government invested in creating a large group of practically oriented institutions as described above, but also pushed its universities to become more economically relevant. In each country, there had been concerns that universities were detached from industry, and were unable to meet changing labour market needs (OECD 2003, White 2001). In each country, investing to create sizeable alternatives to universities exerted pressure on universities also to become more sensitive to economic needs. In Ireland, government funded the establishment of new technical programmes which were considered relevant to industry in universities as well as in the technical institutions. In Finland, significant public investments were subsequently made in key strategic areas of R&D, with much emphasis on collaboration between companies and universities (Srinivas and Viljamaa 2008, Dahlman and others 2007, OECD 2003). Today, Finland appears to have a more diverse university sector, with variable levels of university inclination to work with industry. Some universities with technological orientation or regional development missions are much more active in working with industry, with much larger shares of funding from Finnish Funding Agency for Technology and Innovation (TEKES), which provides competitive grants for application oriented research (Ministry of Education, Finland 2005).

15

A good example is Oulu, which was a region that transformed itself from being a sleepy backwater to becoming a vibrant high tech centre, its university played an important role (Donnelly and Hyry 2004). Established in the 1950s with a specific mission to help the region develop, the University Oulu has been both proactive in its choice of disciplinary coverage and responsive in its education. An electrical engineering department was established early, well before industrial needs emerged, but they also responded quickly to meet emerging skills shortages, for instance in management education (Donnelly and Hyry 2004). The university’s responsiveness also led to the development of relevant research centres, in which both basic and applied research were undertaken, often jointly with industry and with some academic spinoffs. China. In China, the government has been taking steps to reform the higher education system since the 1980s – and in the retrospect, the steps taken have been remarkably consistent in pushing key universities towards the American relevant research university model (Ma 2007). In the S&T reform starting in the 1980s, the role of science and technology in economic development was emphasized in strong terms. While public research institutes were to re-orient the content of their research to meet economic needs, universities were to develop research capacity relevant for the society for the first time. The official endorsement did not mean government funding was forthcoming for research. Although the Chinese National Science Foundation was established in 1986 to provide competitive grants for basic research projects in public research institutions as well as in universities, university budgets had been cut and were extremely tight. Universities had strong incentives to generate their own incomes through industrial contracting (Ma 2007). This was the context in which universities began to develop certain responsiveness to industry, through contract research, through consulting as well as by setting up their own enterprises. The government has also been supporting the emergence of elite research universities through a series of special programmes, starting from the key university and key laboratory programmes in the 1980s, to the more famous Project 211 and Project 985 in the 1990s which concentrated government funding in top 100 and top 9 universities respectively (Ma 2007). Together with the gradual development of competitive funding for research, these resulted in strong incentives for universities to become research-oriented – to compete globally to become world class universities (Ma 2007). It is no accident that one of the first global rankings of universities was designed by one Chinese university – they were developing such indicators to inform themselves of their positions in the world. Moving Away from Responsiveness. The paths taken by Japanese universities provide an interesting contrast to that of the US. The starting point for Japanese universities was similar to that of the US, in that the first university, the University of Tokyo, was developed in the late 19th Century as a critical instrument for Japan’s modernization, specifically to teach useful scientific knowledge. It was highly unusual at the time for universities to have practically oriented faculties including engineering, agriculture, medicine, and law along with science and literature (Bartholomew 1989). Other imperial universities were established along the Tokyo University model, and collectively, they played an active role in introducing key technologies and knowledge essential for Japan’s early

16

industrialization. University academics and graduates were critical in the birth of many of Japan’s technology oriented companies, such as Toyota, Toshiba, and some pharmaceutical companies (Odagiri and Goto 1996). Indeed, Japanese versions of university ‘start-ups’ were visible in many corporate histories. Large companies habitually sponsored university facilities, and academics actively served as technical consultants (Hashimoto 1999). Even the largest industrial corporations had limited research facilities or capabilities in these days, and university academics were undertaking application oriented research that were relevant to such companies. During the post-war period, however, universities began to become isolated from industrial development on an evolutionary path to basic research universities. In a striking contrast to the US, where the successful role of science in military technology had reinforced the public confidence, Japanese academics moved away from the military-industry complex (Hatakenaka 2004). When industry began to acquire foreign technology in the 1950s, universities played a minimal role in the process (Hashimoto 1999). Industries recruited their own researchers, often from universities, to establish corporate laboratories in the 1960s (Nakayama 1991), marking the beginning of the new era when corporate research became a dominant force in industrial innovation (Nakayama 1996). In the late 1960s student uprisings led to further criticisms about university-industry relationships, and a strong mood to limit such relationships prevailed (Hashimoto 1999, Hatakenaka 2004). Various rules and norms emerged under this culture of censorship; the bulk of scientists and engineers were civil servants in national universities, subject to rules which forbade them from consulting with industry, and universities did not have the autonomy to create their own rules. Although universities still worked with industry, their modes of engagement were distinctly different from those of American universities. Relationships were formed principally by industries doing the leg work of coming to the universities, extracting information, and making every effort to learn from them – in a ‘receiver-active’ technical transfer model (Kodama and Suzuki 2007). There was surprisingly little interaction between universities and companies in terms of the content of education (Dore and Sako 1989). Rather, companies expected universities to be no more than a selection mechanism, and developed inhouse capabilities and practices to train them in the needed specializations. One important consequence of such a system was that Japanese universities did not learn to respond to emerging industrial skills need, and this was increasingly an issue for industry as it developed into advanced technology fields, particularly when new industries such as computer software, semiconductors and biotechnology emerged in the 1980s (Baba and others 1996, Kobayashi 2001, Darby and Zucker 1996). In a comparative analysis of Stanford University and Tokyo Institute of Technology (TIT), curricular change was found to be infrequent and limited in scope in TIT compared with that in Stanford, where updating took place organically and almost constantly (Hoshino and others 2003, Harayama 2001). The universities’ lack of responsiveness was particularly pronounced in research training. Although masters programmes in engineering became commonplace by the 1980s, PhD programmes in science and engineering remained small, and highly academic. Even when they expanded PhD programmes in the 1990s, little effort was made to reform the curricular content, and 40% of surveyed industry reported skills mismatch as a reason why they did not recruit

17

PhDs (Shimizu and Mori 2001). This was in some ways an inevitable and consistent development given the corporate personnel practice of recruiting the best and brightest early and training them in-house. Indeed, a form of postgraduate education in sciences and engineering developed within industry rather than within universities (Clark 1995). Developing a Basic Research Orientation Developing research capacity of whatever nature is often itself a difficult step in the development of higher education systems. In Korea, universities played little visible roles in the transition that Korean electronics companies made during the 1980s to high tech fields such as computers and semiconductors, due to their inadequately developed research capacity resulting from government underinvestment (Kim 1993, Kim 2000). Korean universities were particularly under-resourced given their rapid and massive expansion in undergraduate enrolments in the 1970s and the 1980s (Kim 1993). The student staff ratio declined from 23 in 1966 to 36 in 1985 (Kim 1993), and nearly a half of national and public universities revenues came from tuition and fees by 1985 (KCUE 1990). Korea’s public spending is strikingly low in per student terms even today, compared with other countries (Table 8 and 9). It was not until the late 1990s, when government for the first time concentrated large amounts of resources in the best institutions for research and graduate education, that their research performance began to improve in a significant manner (Kim and Nam 2007). Resource scarcity was not all bad news for universities, however. Some of the quoted statistics indicate a relatively high level of engagement with industry. About a half of university research had been funded by industry in the early 1990s (Kim 2000), and one survey of industry found that nearly a half of surveyed companies had worked with universities (Sohn and Kenny 2007). The industry share of research funding declined in the late 1990s to about 15%, presumably reflecting increasing levels of government funding in research (see Table 10). However, it remains higher than the 6% in OECD countries, indicating that industry engagement continues to be important, even if university contribution is not readily acknowledged, and in spite of their poor performance record in proactive measures such as academic spinoffs (Sohn and Kenny 2007). Some would argue that research orientation was slow to develop in Korean universities, because of the historical division of labour in which public research institutions rather than universities were responsible for research (Lee 2000, Sohn and Kenny 2007). Similar difficulties are encountered in India, where there is a well established group of public research institutions whose status is higher than the universities (Jayaram 2007). The best researchers tend to be recruited into research institutions, leaving universities in a difficult position to develop research capabilities. The fact that these research institutions have no teaching responsibilities makes the situation worse in that the best researchers are not contributing to the training of the next generation researchers (Jayaram 2007). Nonetheless, there is no question that public research institutions can be spectacularly successful in the development of high tech industry, as the Taiwanese experience with ERSO shows. It is hard to imagine similar results could have been obtained in the short time period if a similar capacity had been established in a university. Still, the fact that relevant research

18

capacity was concentrated outside universities probably represents a lost opportunity for Taiwanese universities as well. Finland’s experience shows that it is possible to ‘link’ applied research capacity in public research institutes to universities in a productive way. When Finland made its bid to turn itself into a knowledge economy in the early 1990s, one key component of its policies was large public investment in collaborative R&D for universities, public research institutes and companies, orchestrated by the Finnish Funding Agency for Technology and Innovation, TEKES (Dahlman and others 2007). China provides an example in which the research role of universities was promoted early, despite the existence of public research institutes. In the S&T reform starting in the 1980s, the Chinese government took action both in re-orienting government research institutions to become economically relevant, and in recognizing the research role of universities for economic development for the first time. Concluding Remarks This paper has argued that higher education institutions can play a variety of roles in supporting high tech industry. The roles they play in turn can influence the shape and nature of high tech industry. The paper also argued that not all higher education institutions or systems can or need to play the same set of roles. Rather, the roles they play would be determined by the characteristics that the institutions have developed over the years in terms of their responsiveness, their basic science orientation and their selectivity. Goverments can play a critical role in influencing the orientations of institutions –through subtle or not so subtle means. The production of scientists and engineers of high standards and in sufficient numbers has been one common element across all countries. While some countries depended on the large production of generic scientists and engineers, others benefited from sizable investments in specialist training with an emphasis on practical skills. However, in those countries with only generic training in science and engineering, selectivity of institutions appeared to play a key additional role in enabling industry to recruit appropriate manpower. For a high tech industry needed to develop with high technological content, it was important to have sufficient research capacity in universities to provide advanced scientific skills in the labour force. While proactive industries could do wonders in extracting key scientific capacity and training from any universities with scientific research capacity, as the Japanese case showed, it is more helpful for industry if there was ‘responsiveness’ on the part of universities. It is perhaps no accident that in countries with proactive and relevant research universities such as the US and Israel, new industries or new strands of industries emerged and grew readily. Even from universities with limited research capacity, high tech industry did benefit – but often predicated upon cheap high calibre labour and production advantages rather than on technological content. Ireland benefited much from specialist training as provided by practically oriented institutions, while China or India emerged as high tech centres even

19

without such training capabilities. Perhaps in countries such as Ireland where labour costs cannot be as cheap as in China or India, it was important that their labour force provided the value added through additional skills. Overall, institutions in the right hand cells of Fig 2 appear to contribute more dynamically and directly than those in the left hand cells. Responsiveness is a key dimension for developing countries to encourage in their higher education institutions. Responsiveness does not appear to emerge naturally. Rather, all the cases examined had some specific interventions – usually from government. There are standard practices that can be used to introduce a more scientific orientation in higher education systems, such as peer review or competitive research funding processes, and developing countries have an option to make use of the international infrastructures such as overseas PhD training or international journals. In contrast, there is no simple way to encourage institutions to become responsive. The ‘demand’ to which institutions must be responsive is often local, intangible and ill-defined. Governments may require institutions to provide new educational programmes or to expand them, but simply following such a top-down requirement does not make them responsive. The Korean government’s central requirement to keep up the quota for engineering programmes may not have led institutions to become responsive. Indeed, it is possible that the more frequently a government interferes through regulations or even through financial incentives, the less institutions learn to be creative and strategic about what they do. In Ireland and Finland, new groups of institutions had to be established from scratch in order to build responsiveness into the system. Without such concerted efforts, universities appear to drift towards the lefthand side of Fig 2, as a result of peer pressures and academic norms, towards becoming basic research universities or academically oriented teaching institutions. To the extent that most countries do not have sufficient human or monetary resources to establish research oriented institutions at the outset (one exception being Israel), it appears essential for a country to have a strategy that promotes developments upwards and to the right in Fig 2 so that at least some institutions become relevant research institutions. Founding ethos does seem to matter, as shown by the Land Grant Colleges in the US or some institutions with technological and regional development missions in Finland. Equally, public statements and expectations that universities should be relevant can also help. The American government’s expectation that universities must be useful for agriculture was a starting point in their funding of agricultural extension activities in universities. The Chinese government’s explicit demands that universities play a key role in technological upgrading of the society no doubt provided a powerful starting point for them to work with industry; the fact that ‘funding’ for such activities had to be sought from individual enterprises rather than some central agency was probably helpful as universities had to learn to work with a variety of ‘stakeholders.’ On the whole, there appear to be more examples of institutions which developed responsiveness first, then followed with fundamental science orientation (one good example being Massachusetts Institute of Technology – see Etzkowitz 2002); it is probably easier to develop in that way rather than to attempt it the other way around.

20

Today, economic responsiveness of universities has become a global mantra. The last decade has seen an ever increasing number of government policies trying to push universities to develop such a dimension, leading to some changes in academic values. For instance, it would be unusual to meet an academic in any OECD country who would refute the need for a university to be economically relevant. Notwithstanding such a dynamic change taking place, most institutions have a long way to go to develop a full set of internal and external mechanisms to be responsive and proactive. As seen from the example of Japanese national universities, which are as collaborative with industry as are US institutions, they still substantively lag in proactive technology transfer metrics such as licensing and spinoffs. Indeed, in order for them to become truly ‘proactive,’ a whole new set of relationships are likely to be needed, not only with industry but also with as other players ranging from venture capital to proactive research funding bodies. It takes a full eco-system to help nurture responsive universities. One advantage that developing countries have over old economies may be the potential to develop new institutions that together can make up such a complete system. Fundamental research orientation is something that requires serious commitment and investment, usually from government. Investments are needed not only in human capital in terms of advanced research training, but also in infrastructure such as libraries and laboratories, and direct expenses associated with research. It is an orientation that cannot be developed overnight. However, it is also an orientation that is hard to change once introduced, as PhD trained academics usually gravitate towards research rather than teaching, and there is often a sense of elitism associated with being research-oriented – and the more ‘basic’ the more elite. There is an academic drift that tends to gravitate towards the left hand cells of Fig 2; active interventions are usually needed to counter this. ‘Selectivity’ is one dimension of higher education institutions which often emerges naturally as the demand for higher education expands and the competition for places increases. Such selectivity needs to be underpinned by robust selection systems that are fair but also provide good learning incentives for future students. This is not easy, but worth considerable attention, as selection systems can influence directly the reputational hierarchy of institutions as well as learning incentives of students for years to come. Some examinations inadvertently encourage students to become tactical rather than learning oriented. Equally, some scoring systems give spurious impression of accuracy. The fact that all three dimensions are important in enabling institutions to play key roles does not mean that all institutions must be the same. On the contrary, as the case sof the US or Finland show, there is a strong case for requisite variety – or diversification within a country. Indeed, it is reasonable for a national system of higher education to seek to have as much diversity as possible in terms of types of institutions. One historical legacy that the case countries share is some form of national championship of science and technology. The post war US was a special place in terms of policy makers endorsing the role of science for societal benefit; that commitment and belief led to the development of a whole host of federal funding institutions to support fundamental science for different reasons (Geiger 2004b). Similarly, without the highest commitment from Jawaharlal

21

Nehru, the first post-independence leader in India, the Indian Institutes of Technology would never have received the extraordinary technical assistance from the US, Soviet Union and Germany to develop the high standards that they did. Without the commitment of Deng Xiaoping, the Chinese government would not have committed as many resources to science and technology, which in turn helped attract the best and the brightest to the field. Such visible support of science and technology at the highest levels probably helped to foster commitment to science and technology within the society such that students more readily choose to study science and engineering. Such national championship is not easy to maintain over time. Science is expensive and it takes a long time before its impact can be felt. It is easy for commitment to science to give way to fiscal pressures, particularly given that it is never clear how much is needed. It is also not easy to cultivate a high level of interest within the society to induce students to choose science and engineering. Many OECD countries, such as the US, UK and Japan are increasingly concerned about the level of interest among their students in science. Even in India, there is an emerging concern that fewer students are taking a science track. One final point for reflection is the role of governance arrangements at both institutional and national levels. Quality assurance considerations often lead government bodies to become overly prescriptive about inputs as well as outputs. This is a significant dilemma in developing countries, where there is little tradition and culture about the quality of higher education. Attempts by governments to ‘control’ quality on the basis of inputs can lead to institutions becoming similar to each other and to being passive in defining what they are. To make the matters worse, some dimensions are easier to measure than others: publications and citations are often used to evaluate the quality of scientific research, but there are no established metrics that look at responsiveness. Metrics used to ‘control’ or ‘guide’ are often themselves skewed. Tight regulatory control by government can also be a serious impediment when institutions need to respond to the changing needs of the nation. And yet the desire to ‘respond’ to national needs both from labour markets and from future students, particularly in dynamically developing countries, often pushes governments to centralize key decisions about the size and directions of the higher education system. An institutional capacity to ‘respond’ to complex demands is unlikely to develop if institutions are not allowed to make strategic decisions. Managerial autonomy and academic autonomy to decide what to teach and research are critical elements for fostering responsiveness in institutions. At the same time, institutions are unlikely to be able to make good strategic decisions without governance arrangements which allow appropriate structures and processes to develop internally. The key for the future is to put in place governance arrangements at both institutional and system levels which meet the balance of such competing concerns and which guide evolving higher education systems effectively for the needs of the nation.

22

Fig 1: Three dimensions

responsiveness

selectivity

basic science orientation

23

24

Fig 2: Institutional characteristics: responsiveness and basic science orientation

yes Basic science orientation

no

Relevant Research

Institutions

Practical Institutions

Basic Research

Institutions

Teaching Institutions

Responsiveness to economic needs

yes no

Table

1: S

cienc

e and

engin

eerin

g artic

les by

coun

try: 1

995-

2005

Av

erag

e ann

ual c

hang

e (%

)

Reg

ion/

coun

try/e

cono

my

1995

2000

2005

Pulib

catio

n pe

r m

illion

pop

ulat

ion

1995

–20

0020

00–

2005

1995

–20

05Fi

nlan

d 4,

077

4,84

44,

811

914

3.5

–0.1

1.7

Irela

nd

1,21

81,

581

2,12

050

25.

46.

05.

7C

hina

9,

061

18,4

7941

,596

3115

.317

.616

.5In

dia

9,37

010

,276

14,6

0813

1.9

7.3

4.5

Isra

el

5,74

16,

290

6,30

992

61.

80.

10.

9Ja

pan

47,0

6857

,101

55,4

7143

43.

9–0

.61.

7S

outh

Kor

ea

3,80

39,

572

16,3

9634

120

.311

.415

.7Ta

iwan

4,

759

7,19

010

,841

474

8.6

8.6

8.6

Uni

ted

Sta

tes

193,

337

192,

743

205,

320

678

–0.1

1.3

0.6

fing

inst

ing

rie

s/

.

t i

NOTE

S: A

rticle

coun

ts fro

m se

t of jo

urna

ls co

vere

d by S

cienc

e Cita

tio In

dex (

SCI)

and S

ocial

Scie

nces

ita

tion I

ndex

(SSC

I). A

rticles

clas

sifed

by ye

ar of

ion an

d ass

igned

to

regio

n/cou

ntry/e

cono

my on

basis

of in

stitut

ional

addr

ess(e

s) lis

ted on

artic

le. A

rticles

onra

ction

al-co

unt b

asis,

i.e., f

or ar

ticles

with

colla

bora

tins

titutio

ns fr

om m

ultipl

e cou

ntec

onom

ies,

each

coun

try/ec

onom

y rec

eives

frac

tiona

l cre

dit on

basis

of pr

opor

tion o

f its p

artic

ipat

itutio

ns.

nC

i pu

blica

t

SOUR

CES:

Tho

mson

Scie

ntific

, SCI

and S

SCI, h

ttp://s

cienti

fic.th

omso

ncom

/prod

ucts/

caeg

ories

/citat

ion/;

pIQ, In

c.; an

d Nati

onal

Scien

ce F

ound

ation

, Divi

sion o

f Scie

nce R

esou

rces S

tatist

ics,

spec

ial ta

bulat

ions.a

s pub

lishe

d in S

cienc

e an

d En

ginee

ring

Indic

ator

s 200

8, p

opula

tion

data

from

the

Wor

ld Ba

nk d

atab

ase

acce

ssed

May

7,2

008,

Natio

nal S

tatis

tics,

Taiw

an,

25

Tabl

e 2:

Rel

ativ

e pr

omin

ence

of s

cien

tific

lite

ratu

re: 1

995

and

2003

(R

elat

ive

cita

tion

inde

x)

R

ank

Cou

ntry

19

9520

03

1

Sw

itzer

land

1.

189

1.15

2

2

Uni

ted

Sta

tes

1.01

31.

026

5 U

nite

d K

ingd

om

0.83

00.

864

8 Fi

nlan

d 0.

755

0.82

6

12

Ire

land

0.

662

0.76

4

14

Is

rael

0.

682

0.74

2

15

E

U-1

5 0.

681

0.73

7

19

H

ong

Kon

g 0.

438

0.67

2

22

Ja

pan

0.55

10.

575

24

Sin

gapo

re

0.42

20.

509

26

Chi

le

0.34

30.

478

28

Sou

th K

orea

0.

320

0.43

9

29

Th

aila

nd

na0.

432

30

Cze

ch R

epub

lic

na0.

417

31

Arg

entin

a 0.

304

0.41

1

32

S

love

nia

na0.

409

33

Sou

th A

frica

0.

325

0.40

4

34

Ta

iwan

0.

321

0.40

1

35

M

exic

o 0.

349

0.38

5

36

P

olan

d 0.

295

0.36

0

37

B

razi

l 0.

314

0.35

9

38

S

lova

kia

na0.

339

40

Bul

garia

0.

166

0.32

5

41

C

hina

0.

218

0.29

3

42

In

dia

0.18

20.

284

43

Cro

atia

0.

257

0.28

1

44

Tu

rkey

0.

214

0.27

8

45

S

audi

Ara

bia

0.22

00.

256

26

na =

not

app

licab

le

EU

= E

urop

ean

Uni

on

NO

TES

: Rel

ativ

e pr

omin

ence

of s

cien

tific

lite

ratu

re m

easu

red

on b

asis

of r

elat

ive

cita

tion

inde

x of

cou

ntry

/eco

nom

y, i.

e., c

ount

ry/e

cono

my’

s sh

are

of c

ited

liter

atur

e ad

just

ed fo

r its

sha

re o

f pub

lishe

d lit

erat

ure.

Cou

ntry

/eco

nom

y’s

cita

tion

of o

wn

liter

atur

e ex

clud

ed. I

ndex

of 1

.00

indi

cate

s co

untry

/eco

nom

y’s

shar

e of

cite

d lit

erat

ure

equa

ls it

s w

orld

sha

re o

f sci

entif

ic li

tera

ture

. Ind

ex >

1.00

or <

1.00

indi

cate

s co

untry

/eco

nom

y ci

ted

rela

tivel

y m

ore/

less

than

indi

cate

d by

its

shar

e of

sci

entif

ic li

tera

ture

. C

ount

ries/

econ

omie

s w

ith s

hare

of w

orld

pub

licat

ions

in c

ited

field

<0.

10%

dur

ing

perio

d ei

ther

exc

lude

d or

na.

Cita

tions

on

fract

iona

l-cou

nt b

asis

, i.e

., fo

r cite

d ar

ticle

s w

ith c

olla

bora

ting

inst

itutio

ns fr

om m

ultip

le c

ount

ries/

econ

omie

s, e

ach

coun

try/e

cono

my

rece

ives

frac

tiona

l cre

dit o

n ba

sis

of p

ropo

rtion

of i

ts p

artic

ipat

ing

inst

itutio

ns. E

U-

15 c

onsi

sts

of 1

5 E

U c

ount

ries

befo

re e

xpan

sion

of m

embe

r cou

ntrie

s on

May

1, 2

004.

Eas

t Asi

a-4

cons

ists

of C

hina

(inc

ludi

ng H

ong

Kon

g), S

inga

pore

, Sou

th K

orea

, an

d Ta

iwan

. S

OU

RC

ES

: Tho

mso

n IS

I, S

cien

ce C

itatio

n In

dex

and

Soc

ial S

cien

ces

Cita

tion

Inde

x, h

ttp://

ww

w.is

inet

.com

/pro

duct

s/ci

tatio

n/; i

pIQ

, Inc

.; an

d N

atio

nal S

cien

ce

Foun

datio

n, D

ivis

ion

of S

cien

ce R

esou

rces

Sta

tistic

s, s

peci

al ta

bula

tions

. S

cien

ce a

nd E

ngin

eerin

g In

dica

tors

200

6

Tabl

e 3:

Tot

al e

nrol

lmen

ts: a

dvan

ced

rese

arch

(PhD

) pro

gam

mes

19

70

1980

1990

2000

2003

2006

% o

f tot

al

(200

6)

Chi

na

54

,038

108,

737

...

...

Finl

and

19

,750

19,8

4622

,145

7.2

Indi

a

55,0

1965

,357

36,5

19-

Irela

nd

2,

904

3,81

65,

146

2.8

Isra

el

6,

647

7,94

49,

715

3.1

Japa

n 13

,243

18

,211

28,3

5459

,007

68,2

4575

,028

1.8

Rep

ublic

of K

orea

28,9

2434

,712

41,0

551.

3Ta

iwan

16

6 67

34,

437

13,8

2221

,658

29,8

392.

3U

nite

d S

tate

s

293,

002

306,

889

388,

685

2.2

Sou

rce:

UN

ES

CO

dat

abas

e ac

cess

ed M

ay 7

, 200

8, E

duca

tion

Sta

tistic

s, M

inis

try o

f E

duca

tion,

Sci

ence

and

Tec

hnol

ogy,

Jap

an; E

duca

tion

stat

istic

s, M

inis

try o

f Edu

catio

n, T

aiw

an

Tabl

e 4:

Und

ergr

adua

te s

cien

ce a

nd e

ngin

eerin

g en

rollm

ents

in J

apan

and

Kor

ea (i

nclu

ding

agr

icul

ture

exc

ludi

ng m

edic

ine)

1960

19

65

1970

19

75

1980

19

85

1990

19

95

2000

20

05

Japa

n 13

6,81

8238

,596

375,

5984

43,1

8045

1,90

4463

,336

524,

2016

11,3

5162

5,37

1590

,549

%

of t

otal

23

%27

%28

%27

%26

%27

%26

%26

%25

%24

%

Kor

ea

37

,099

59,2

6474

,410

166,

1373

37,6

2441

9,89

1523

,002

%

of t

otal

3534

3641

3640

44

S

ourc

e: M

EX

T M

onbu

kaga

kuto

keiy

oran

200

6, W

eidm

an a

nd P

ark

2000

27

Tabl

e 5:

Ter

tiary

gro

ss e

nrol

lmen

t rat

io

19

70

1

97

5

19

80

1

98

5

19

90

1

99

5

20

00

2

00

5

20

06

Chi

na

0 1

2 3

3 5

8..

22

F

inla

nd

13

28

32

3448

70

83

92

93 In

dia

5 5

5 6

6 7

1011

12

Irel

and

12

17

18

2231

40

49

58

59 Is

rael

18

23

29

33

36

41

5058

58

Jap

an

18

26

31

2831

42

47

55

57 K

orea

, Rep

. 7

9 15

34

39

52

7390

91

Taiw

an

3956

8284

Uni

ted

Sta

tes

47

55

56

6072

81

69

82

82S

ourc

e: U

nesc

o da

taba

se a

nd W

orld

Ban

k da

taba

se a

cces

sed

May

7, 2

008,

Edu

catio

n in

dica

tors

from

Min

istry

of

Edu

catio

n, T

aiw

an

*Tot

al n

umbe

r of s

tude

nts

in te

rtiar

y ed

ucat

ion

incl

udin

g gr

adua

te s

choo

ls a

s a

perc

enta

ge o

f the

pop

utat

ion

in th

e fiv

e ye

ar a

ge g

roup

follo

win

g on

from

sec

onda

ry s

choo

l lea

ving

age

Ta

ble

6: T

otal

terti

ary

enro

llmen

ts in

sci

ence

and

eng

inee

ring

20

05

% o

f tot

al e

nrol

lmen

ts

Sci

ence

/Eng

inee

ring

ratio

Sci

ence

Eng

inee

ringT

otal

S

cien

ceEn

gine

erin

gS&

Eng

Chi

na

1443

129

5536

1231

8352

821

8%30

%38

%0.

3Fi

nlan

d 35

468

8082

730

5996

12%

26%

38%

0.4

Indi

a **

+ 16

8950

4 69

6609

1177

7296

14%

6%20

%2.

4Ire

land

22

851

1923

318

6561

12%

10%

23%

1.2

Isra

el

2996

7 56

812

3109

3710

%18

%28

%0.

5Ja

pan

1187

04

6685

2640

3830

23%

17%

19%

0.2

Rep

ublic

of K

orea

26

4259

10

2284

532

2487

58%

32%

40%

0.3

Uni

ted

Sta

tes

1537

243

1154

9711

7272

044

9%7%

16%

1.3

Sou

rce:

UN

ES

CO

dat

abas

e ac

cess

ed M

ay 7

, 200

8, fo

r Chi

na, f

igur

es a

re fr

om 2

004

(Ji 2

006)

**

Indi

a's

data

on

2000

are

act

ually

from

yea

r 200

1

28

Tabl

e 7:

Tai

wan

's g

radu

ates

in s

cien

ce a

nd e

ngin

eerin

g (in

clud

ing

agric

ultu

re e

xclu

ding

m

edic

ine)

1960

1965

1970

1975

1980

1985

1990

1995

Num

ber

1655

2689

7987

1971

625

672

3796

550

490

8293

8 %

tota

l 27

%25

%24

%33

%36

%42

%44

%46

%S

ourc

e: L

in 2

004

Ta

ble

8: P

ublic

exp

endi

ture

per

pup

il as

a %

of G

DP

per

cap

ita

(Ter

tiary

leve

l)

1999

20

00

2004

20

05

C

hina

90

...

...

...

Japa

n 15

18

2119

R

epub

lic o

f Kor

ea

8 ...

9

...

Fi

nlan

d 41

39

3735

Ire

land

29

31

2425

Is

rael

33

32

26...

Uni

ted

Sta

tes

27

...

2323

In

dia

...

9194

58

Sou

rce:

UN

ES

CO

dat

abas

e ac

cess

ed M

ay 7

, 200

8

29

Tabl

e 9:

Ter

tiary

edu

catio

n ex

pend

iture

s as

% o

f GD

P

1999

20

00

2004

20

05

Chi

na

Priv

ate

...

...

...

...

Pu

blic

0.

4...

...

...

Tota

l

...

...

...

...

Japa

n P

rivat

e 0.

60.

60.

80.

9

Pub

lic

0.5

0.5

0.5

0.5

To

tal

1.

11.

11.

31.

4R

epub

lic o

f Kor

ea

Priv

ate

1.7

...

1.9

...

Pu

blic

0.

4...

0.

5...

Tota

l

2.2

...

2.3

...

Finl

and

Priv

ate

...

...

0.1

0.1

P

ublic

1.

71.

71.

71.

7

Tota

l

...

1.7

1.8

1.8

Irela

nd

Priv

ate

...

...

...

0.2

P

ublic

1.

0...

0.

90.

9

Tota

l

...

...

...

1.2

Isra

el

Priv

ate

...

0.9

1.0

...

P

ublic

1.

21.

11.

0...

Tota

l

...

2.0

2.0

...

Uni

ted

Sta

tes

Priv

ate

...

1.7

2.0

2.0

P

ublic

1.

10.

91.

11.

1

Tota

l

...

2.6

3.1

3.0

Indi

a P

rivat

e ...

-

0.2

...

Pu

blic

...

0.

91.

0...

Tota

l

...

0.9

...

...

Sou

rce:

UN

ES

CO

dat

abas

e as

acc

esse

d M

ay 7

, 200

8

30

Tabl

e 10:

Aca

dem

ic R&

D fin

ance

d by

indu

stry

, by s

elect

ed O

ECD

coun

tries

: 198

1–20

06

(Per

cent)

Year

Fr

ance

Ge

rman

y Ja

pan

South

Kor

ea

UK

U.S.

Al

l OEC

D Ch

ina

Russ

ia 19

81

1.3

1.8

1.0

NA

2.8

4.4

2.9

NA

NA

1986

2.0

5.8

1.7

NA

5.7

6.5

4.5

NA

NA

19

91

4.2

7.0

2.4

NA

7.8

6.8

6.0

NA

NA

1996

3.2

9.2

2.4

50

.5 6.7

7.1

6.9

NA

23

.7 20

01

3.1

12.2

2.3

14.3

6.2

6.5

6.4

NA

26.5

2002

2.9

11

.8 2.8

13

.9 5.8

5.8

6.2

NA

27

.2 20

03

2.7

12.6

2.9

13.6

5.6

5.3

6.0

35.9

27.9

2004

2.7

13

.2 2.8

15

.9 5.1

5.1

6.2

37

.1 32

.6 20

05

NA

NA

NA

15.2

NA

5.2

NA

36.7

29.3

2006

NA

NA

NA

NA

NA

5.3

NA

NA

NA

So

urce

: Scie

nce a

nd E

ngine

ering

Indic

ators

2008

31

32

References Agarwal, Pawan. 2007. Higher education and labour market in India. Paper presented at the World Bank ABCDE conference. Athreye, Suma S. 2005. “The Indian Software industry.” In Ashish Arora and Alfonso Gambardella eds., From underdogs to tigers: the rise and growth of the software industry in Brazil, China, India, Ireland, and Israel. Oxford: Oxford University Press Baba, Y., S. Takai, and Y. Mizuta. 1996. “The user driven evolution of the Japanese software industry: the case of customized software for mainframes.” In D. C. Mowery, ed. The international computer software industry: a comparative study of industry evolution and structure. New York: Oxford University Press Bartholomew, James R. 1989. The Formation of Science in Japan, New Haven and London: Yale University Press Basant, Rakesh and Chandra, Pankaj. 2007. “Role of educational and R&D institutions in city clusters: an exploratory study of Bangalore and Pune regions in India.” World Development 35(6): 1037-1055 Bresnahan, Timothy F, and Franco Malerba. 1999 “Industrial dynamics and the evolution of firms’ and nations’ competitive capabilities in the world computer industry.” In David C. Mowery and Richard R. Nelson, eds., Sources of industrial leadership: studies of seven industries. Cambridge: Cambridge University Press Breznitz, Dan. 2004. “The Israeli software industry.” In Ashish Arora and Alfonso Gambardella eds., From underdogs to tigers: the rise and growth of the software industry in Brazil, China, India, Ireland, and Israel. Oxford: Oxford University Press Breznitz, Dan. 2007. Innovation and the State: Policital Choice aqnd Strategies for Growth in Israel, Taiwan and Ireland. New Haven and London: Yale University Press. Chun, Kun and Martin Kenney. 2007. “Universities/research institutes and regional innovation systems: the cases of Beijing and Schenzhen.” World Development 35(6):1056-1074 Clark, Burton R. 1995. Places of inquiry: research and advanced education in modern universities. Berkeley, Los Angeles and London: University of California Press Cohen, W. M., R.R. Nelson and J.P. Walsh. 2002. Links and impact: the influence of public research on industrial R&D. Management Science 48 (1): 1-23 Dahlman, Carl, Jorma Routti, and Pekka Yla-Anttila. 2007. Finland as a knowledge economy: elements of success and lessons learned. Washington: World Bank Institute

33

Darby, Michael R. and Lynne G. Zucker. 1996. Star scientists, institutions and the entry of Japanese biotechnology enterprises. NBER Working Paper 5795. de Fontaney, Catherine and Erran Carmel. 2004. “Israel’s Silicon Wadi: The Forces behind Cluster Formation.” In Bresnahan, Timothy and Gambardella, Alfonso eds. Building high tech clusters: silicon valley and beyond, Cambridge: Cambridge University Press. Donnely, Tom and Martti Hyry. 2004. “Urban and Regional High Technologies: the Case of Oulu.” Loca Economy 19(2): 134-149 Dore, Ronald P. and Mari Sako. 1989. How the Japanese learn to work. London and New York: Routledge Eun, Jong-Hak, Keun Lee and Guisheng Wu. 2006. “Explaining University-run enterprises in China: A theoretical framework for university-industry relationship in developing countries and its application to China.” Research Policy 35: 1329-1346 Etzkowitz, Henry. 2002. MIT and the rise of entrepreneurial science. London and New York: Routledge Geiger, Roger. 2004a. To advance knowledge: the growth of American Research Universities 1900-1940. New Brunswick and London: Transaction Publishers Geiger, Roger. 2004b. Research and relevant knowledge: American Research Universities since World War II. New Brunswick and London: Transaction Publishers Harayama, Yuko. 2001. “Silicon Valley no sangyo hatten to Stanford daigaku no curriculum hensen.” (Silicon Valley’s industrial development and curricular change in Stanford University). In Aoki, Masahiko, Akihiko Sawa, and Michio Daito, eds. 2001. Daigaku kaikaku: kadai to souten. Tokyo: Toyokeizai shinposha, pp. 31-66 Hashimoto, Takehiko. 1999. “The hesitant relationship reconsidered: university-industry cooperation in postwar Japan.” In Lewis Branscomb, Fumio Kodama, and Richard Florida, eds. 1999. Industrializing knowledge: University-industry linkages in Japan and the US. Cambridge and London: MIT Press Hatakenaka, Sachi. 2004. University-industry partnerships in MIT, Cambridge and Tokyo: storytelling across boundaries. New York: Routledge Henderson, Rebecca, Luigi Orsenigo, and Gary P. Pisano. 1999. “The pharmaceutical industry and the revolution in molecular biology: interactions among scientific, institutional and organizational change. In Mowery, David and Nelson, Richard ed. Sources of industrial leadership: studies of seven industries. Cambridge 1999

34

Hicks, Diana. 1993. “University-industry research links in Japan.” Policy Sciences 26: 361-395 Hoshino, Yu, Yuko Harayama, and Shigeo Hatatani. 2003. “Daigaku no gaibu henka heno taio sei (Responsiveness of universities to external change)” in Harayama, Yuko ed. Sangaku Renkei. Tokyo: Toyokeizaishimposha Hou, Chi-Ming and San Gee. 1993. “National Systems Supporting Technical Advance in Industry: The Case of Taiwan.” In Richard R. Nelson, ed., National Innovation Systems: A comparative Analysis. New York and Oxford: Oxford University Press Jayaram, N. 2007. “Beyond retailing knowledge: prespects of research-oriented universities in India.” In Philip G. Altbach, and Jorge Balan, eds., World Class Worldwide: Transforming Research Universities in Asia and Latin America. Baltimore: the Johns Hopkins University Press Ji, Zhou. 2006. Higher education in China. Thompson Learning Juahiainen, Jussi S. 2006. “Multipolis: High-technology Network in Northern Finland.” European Planning Studies 14(10): 1407-1428 Kapur, Devesh. The Causes and Consequences of India’s IT Boom. 2002. India Review 1(2): 91-110 Kapur, Devesh and McHale, John. 2005. “Sojourns and software: international mobile human capital and high tech industy development in India, Ireland and Israel.” In Arora, Ashish and Gambardella, Alfonso ed. 2005, From underdogs to tigers: the rise and growth of the software industry in Brazil, China, India, Ireland, and Israel. Oxford: Oxford University Press Kim, Ki-Seok, and Sunghee Nam. 2007. “The Making of a World-Class University at the Periphery.” In Altbach, Philp G, and Jorge Balan eds., Transforming Resaerch Universities in Asia and Latin America: World Class Worldwide. Baltimore: The Johns Hopkins University Press Kim, Limsu. 1993. “National System of Industrial Innovation: Dynamics of Capability Building in Korea.” In Richard R. Nelson, ed., National Innovation Systems: A comparative Analysis. New York and Oxford: Oxford University Press Kim, Limsu. 2000. “Korea’s National Innovation System in Transition.” In Limsu Kim, and Richard R. Nelson, eds., Technology, learning and innovation: experiences of newly industrializing economies. Cambridge and New York: Cambridge University Press Kobayashi, Shinichi. 2001. “Daigaku kyoiku no shokugyoteki relevance to daigaku no soshiki sekkei.” (Relevance of university education to work and organizational design).

35

In Aoki, Masahiko, Akihiko Sawa, and Michio Daito, eds., Daigaku kaikaku: kadai to souten. Tokyo: Toyokeizai shinposha. Kodama, Fumio and Jun Suzuki. 2007. How Japanese companies have used scientific advances to restructure their businesses: the receiver-active national system of innovation. World Development, 35(6): 976-990 Korean Council for University Education (KCUE). 1990. Korean Higher Education: Its development, Aspects and Prospects. Kroll, Henning, and Ingo Leifner. 2008. “Spin-off enterprises as means of technology commercialisation in a transforming economy- Evidence from three universities in China.” Technovation 28: 293-313 Lee, Won-Young. 2000. “The role of science and tgechnology policy in Korea’s industrial development.” In Limsu Kim, and Richard R. Nelson, eds., Technology, learning and innovation: experiences of newly industrializing economies. Cambridge and New York: Cambridge University Press Leifner, Ingo and Daniel Schiller. 2008. “Academic capabilities in developing countries- a conceptual framework with empirical illustrations from Thailand.” Research Policy. 37: 276-293. Lester, Richard. 2005. Universities, Innovation and the competitiveness of local economies: a summary report from the Local Innovation Systems Project – Phase I. MIT IPC Working Paper IPC-050-010 Lin, Tin-Chun. 2004. “The role of higher education in economic development: an empirical study of Taiwan case.” Journal of Asian Economics 15: 355-371 Ma, Wanhua. 2007. “The Flagship University and China’s Economic Reform.” In Altbach, Philip G., and Balan, Jorge, World Class Worldwide: Transforming Research Universities in Asia and Latin America. Baltimore: The Johns Hopkins University Press Mazzoleni, Roberto and Richard R. Nelson 2007. “Public research institutions and economic catch up.” Research Policy 36: 1512-1528. Ministry of Education, Finland. 2005. OECD thematic review of tertiary education. Country Background report for Finland. Moore, Gordon and Kevin Davis. 2004. “Learning the Silicon Valley Way.” In Bresnahan, Timothy and Gambardella, Alfonso eds., Building high tech clusters: silicon valley and beyond. Cambridge: Cambridge University Press.

36

Mowery, David. 1999. “The computer software industry.” In Mowery, David and Nelson, Richard ed. Sources of industrial leadership: studies of seven industries. Cambridge: Cambridge University Press Mowery, David, Nelson, Richard R., Sampat, Bhaven N., Ziedonis, Arvids A. 2004. Ivory tower and industrial innovation: university-industry technology transfer before and after the Bayh-Dole Act. Stanford: Stanford University Press Nakayama, Shigeru. 1991. Science, technology and society in postwar Japan. Kegan Paul International Nakayama, Shigeru. 1995. Kagaku gijutsu no sengoshi. Tokyo: Iwanamishoten National Academy of Sciences. 2005. Rising above the Gathering Storm: Energizing and Employing America for Brighter Economic Future. Washington: National Academies Press National Science Foundation. 2007. Asia’s Rising Science and Technology Strength: Comparative Indicators for Asia, the European Union and the United States. Washington, D.C.: National Science Foundation Nelson, Richard ed. 1993. The national innovation system: a comparative analysis. New York, Oxford: Oxford University Press Odagiri, Hiroyuki and Akira Goto. 1996. Technology and industrial development in Japan: building capabilities by learning, innovation and public policy. Oxford: Oxford University Press Organization for Economic Cooperation and Development (OECD). 2003. Reviews of Naitonal Policies for Education: Polytechnic Education in Finland. Paris: OECD Pechter, Kenneth and Sumio Kakinuma. 1999. “Co-authorship linkages between university research and Japanese industry.” In Branscomb, Lewis, Kodama, Fumio and Florida, Richard (eds), Industrializing knowledge: University-industry linkages in Japan and the US. Cambridge and London: MIT Press, pp. 102-127 Roper, Stephen and Seamus Grimes. 2005. “Wireless valley, silicon wadi and digital island – Helsinki, Tel Aviv and Dublin and the ICT global production network.” Geoforum 36: 297-313 Rosenberg, Nathan and Richard Nelson. 1994. “American universities and technical advance in industry.” Research Policy (23): 323-248 Sands, Anita. 2005. “The Irish software industry.” In Ashish Arora and Alfonso Gambardella eds., From underdogs to tigers: the rise and growth of the software industry in Brazil, China, India, Ireland, and Israel. Oxford: Oxford University Press

37

Saxenian, Annalee. 1994. Regional Advantage: culture and competition in Silicon Valley and Route 128. Cambridge and London: Harvard University Press. Saxenian, Annalee. 1999. Silicon Valley’s New Immigrant Entrepreneurs. San Francisco : Public Policy Institute of California Saxenian, Annalee, Yasuyuki Motoyama, and Xiaohong Quan. 2002. Local and Global Networks of Immigrant Professionals in Silicon Valley. San Francisco: Public Policy Institute of California Saxenian, Annalee. 2004. “Taiwan’s Hsinchu Region: Imitator and Partner for Silicon Valley.” In Bresnahan, Timothy and Gambardella, Alfonso ed., Building high tech clusters: silicon valley and beyond. Cambridge: Cambridge University Press. Saxenian, Annalee. 2006. The new Argonauts: regional advantage in a global economy. Cambridge and London: Harvard University Press Shimizu, Kiyoyuki, and Kazuhiro Mori. 2001. Nihon no rikokei daigakuinkyoiku no bapponteki kaikaku (Fundamental reform for graduate schools in Japan) in Masahiko Aoki, Akihiro Sawa, and Michio Daito eds., Daigaku kaikaku: kadai to souten. Tokyo: Toyokeizai shinposha. Sohn, Dong-Won, and Martin Kenny. 2007. “Universities, clusters, innovation systems: the case of Seoul, Korea.” World Development 35(6): 991-1004 Stokes, Donald E. 1997. Pasteurs quadrant: basic science and technological innovation. Washington, D. C.: Brookings Institution Press Srinivas, Smita and Kimmo Viljamaa. 2008. ”Emergence of Economic Institutions: Analysing the Third Role of Universities in Turku, Finland.” Regional Studies 42(3); 323-341 Thursby, Jerry G. and Marie C. Thursby. 2007. “University licensing.” Oxford Review of Economic Policy 23(4): 620-639 Trajtenberg, Manuel. 2001. “Innovation in Israel 1968-1997: a comparative analysis using patent data.” Research Policy 30: 363-389 UNCTAD. 2005. World Investment Report: Transnational corporations and the internationalization of R&D. New York and Geneva: 2005 Wade, Robert. 1990. Governing the market: Economic theory and the role of government in East Asian industrialization. Princeton: Princeton University Press

38

Wadhwa, Vivek, AnnaLee Saxenian, Ben Rissing and Gary Gereffi. 2007. “America’s New Immigrant Entrepreneurs.” http://memp.pratt.duke.edu/downloads/americas_new_immigrant_entrepreneurs.pdf Walsh, Kathleen. 2003. Foreign high-tech R&D in China: risks, rewards and implications for U.S.-China relations. Washington: The Henry L. Stimson Center White, Tony. 2001. Investing in people: higher education in Ireland from 1960-2000. Dublin: Institute of Public Administration Wu, Weiping. 2007. “Cultivating research universities and industrial linkages in China: the case of Shanghai.” World Development 35(6): 1075-1093 Xiwei, Zhong, and Yang Xiangdong. 2007. “Science and technology policy reform and its impact on China’s national innovation system.” Technology in Society 29: 317-325 Yusuf, Shahid and Kaoru Nabeshima. 2007. How universities promote economic growth. Washington: The World Bank Zhou, Ji. 2005. Higher education in China. Cengage Learning Zucker, Lynn G. and Michael R. Darby. 1996. “Star scientists and institutional transformation: patterns of invention and innovation in the formation of the biotechnology industry.” Proceedings of the National Academy of Sciences, 93(23): 12709-12716 Zucker, Lynne G., Michael R. Darby, and Marilyn B. Brewer. 1998. “Intellectual human capital and the birth of US biotechnology enterprises.” The American Economic Review. 88(1): 290-306 Zucker, Lynne G., Michael R. Darby, and Jeff S. Armstrong. 2002. “Commercializing knowledge: university science, knowledge capture and firm performance in biotechnology.” Management Science 48(1): 138-153

39