Technopolis Development: An Assessment of the Hsinchu Experience

PLEASE SCROLL DOWN FOR ARTICLE

On: 26 December 2008

Access details: Access Details: [subscription number 788856048] Publisher Routledge

Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

International Planning Studies

Publication details, including instructions for authors and subscription information:

http://www.informaworld.com/smpp/title~content=t713426761

Technopolis development: An assessment of the Hsinchu experience

Chien-Yuan Lin a

a Institute of Building and Planning, National Taiwan University, Taipei, Taiwan

Online Publication Date: 01 June 1997

To cite this Article Lin, Chien-Yuan(1997)'Technopolis development: An assessment of the Hsinchu experience',International Planning Studies,2:2,257 — 272

To link to this Article: DOI: 10.1080/13563479708721682

URL: http://dx.doi.org/10.1080/13563479708721682

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should be independently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with or arising out of the use of this material.

Technopolis Development: An Assessment of the

Hsinchu Experience

CHIEN-YUAN LIN

Institute of Building and Planning, National Taiwan University, 1 Sec. 4, Roosvelt Rd., Taipei, Taiwan.

ABSTRACT The Hsinchu Science-Based Industrial Park (HSIP) in Taiwan is a government-led

investment project located 50 miles south of Taipei established nearly 15 years ago. Its continuous growth in terms of revenue and investment has been recognized as a successful model of technology park development. Although regional development was not considered during plan-ning, its economic achievement has indirectly impacted on the technopolis development in the long term. To accommodate the housing demands, the Hsinchu Science City Development Plan was prepared in 1991 to promote the technopolis development. However, it is regarded as a failed project. This paper will analyse the success factors of the park development and the failure factors of the Science City project. In addition, based on the Hsinchu experience, the institutional paradox of government-led technopolis development is discussed.

Introduction

Production systems in the global economy have been revolutionized by techno-logical changes. Cities which grew by traditional manufacturing industries are giving way to so-called technopolises where the economy is driven by high technologies. Technopolis development differentiates between spontaneous growth and planned occurrences. The former is common in the USA., such as Silicon Valley in California and Research Triangle in North Carolina, where technology-oriented regions have developed as a result of the uncoordinated actions of private business who happen to locate together due to the presence of commonly valued attributes, especially research universities, good climate and, perhaps, small-scale government investments (Luger, 1996). The latter is typical in Asian nations, where strategic investments, especially by central government, are intended to create technology-oriented cities to serve as engines of growth for their respective regions. Compared with the planned developments of Tsukuba Science City in Japan and Taedok Science Town in Korea, the develop-ment of Hsinchu Science City in Taiwan demonstrates some unique experiences in that the government-led projects in the production-oriented technology park (Hsinchu Science-based Industrial Park, HSIP) have successfully stimulated national and local economic development but failed in promoting the physical development needed in the context of a technopolis.

Taiwan is an island of 36 000 km2, with a total population of 21.35 million in

1995. As one of the newly industrialized countries (NICs), Taiwan has experi-enced rapid economic growth in the past three decades. In 1965, the per capita

1356-3475/97/020257-16 © 1997 Carfax Publishing Ltd

GNP was a mere US$217, but it reached US$12 439 in 1995. However, criticism has been levelled that urban and regional planning/management in Taiwan has dismally failed to keep pace with its economic achievements. The experience of the Hsinchu Science City project has evidently demonstrated that the manage-ment of physical developmanage-ment is much more difficult and complicated than the promotion of economic growth in the case of technopolis development. This paper is intended to shed some light on the critical factors that have influenced the success of the technology park and the failure of the development planning of Hsinchu as a technopolis. The assessment of Hsinchu's experience should provide some valuable information to planners in countries where governments are involved in planning and development of technology parks and technopo-lises.

Global Economy and the Technopolis

The increasing globalization of economic activities has been the most significant development in the world economy during the past few decades. 'Globalization' of an economic activity is qualitatively different from 'internationalization'. It is a more advanced and complex form of internationalization, implying a degree of functional integration between internationally dispersed economic activities (Dicken, 1992). International labour division associated with globalization is not only enabled by information technology and advanced international transporta-tion, but was also enhanced by the extraordinary events which occurred in 1989, including socialism's collapse and the end of the Cold War. With the opening of the Berlin Wall in November 1989, an extraordinary sequence of events began unfolding to, ultimately, transform global economic geography. Communist governments were ousted throughout Eastern Europe, and later subsequently in the former Soviet Union (Berry el ah, 1993).

Cities are the regional economic centres exercising urban functions to serve their peripheral areas. They are complex places that offer a variety of supportive services and benefits that have developed into a sophisticated and dynamic balance to respond efficiently to their distinctive opportunities and needs. Restated, the value must result from the 'functions' that are encouraged in the city rather than from the 'shell' that is provided by the physical infrastructure. Influenced by the growing borderless economy, the nature and structure of these economic activity centres and regions are correspondingly transformed. Cities interact each other and are directly or indirectly inter-linked as a network of global cities. In addition to the conventional global network of 'national capital cities', cities linked in different functional networks are designated according to their major economic functions, e.g. 'airport city' (Conway, 1992); 'seaport city'; and 'technopolis' or 'technopole' (Hodgson, 1992; Castells & Hall, 1994). Among them, 'technopolis' can be viewed as a city of technology in which activities pivot. In addition, technopolises are places where the arts of science and the application of scientific knowledge are the driving forces of a region's economic activity.

In recognizing the increasing importance of high-tech industries in terms of economic development and regional development, an increasing number of developed and less developed countries are encouraging the development of high-tech industries directly or indirectly. Developing a technology park or science park is frequently viewed as a prominent instrument to establish an

innovative milieu or core area of regional development (Goldstein & Luger, 1993). Such a trend has recently flourished in Asian countries (Luger & Lee 1994). The Hsinchu Science-based Industrial Park (HSIP) in Taiwan is a govern-ment-led project with nearly 15 years of experience. It is contended by Castells & Hall (1994) that at least 15 to 25 years are necessary for a technology park development to reach full impact. Assessing the Hsinchu experience in terms of technopolis development should provide valuable information for pertinent individuals and agencies.

The geographical area of a technopolis is composed roughly of two parts: the technology park(s) that serves as the engine of economic development, and the surrounding area where the regional development is influenced by the technol-ogy park. To review the technopolis development of Hsinchu, this paper discusses the development experience of HSIP and the planning of Hsinchu Science City Development. The critical success factors of HSIP and failure factors of the Science City project are analysed as well. Finally, in addition to the discussion of the future development of the Hsinchu technopolis, an institutional paradox of government-led technopolis development is discussed.

Development of Hsinchu Science-based Industrial Park

The Hsinchu Science-based Industrial Park (HSIP) is situated between Hsinchu City and Hsinchu County, 15 minutes by car from the centre of the city. Travel time to Taipei by freeway is about an hour and a half, to the CKS International Airport about 40 minutes, and to the major harbours of Taichung (south of the park) and Keelung (to the north) approximately two hours in each case (Figure 1).

While most technology parks are purely private sector real estate investments in western countries, Hsinchu is a purely government-led project planned and established by the National Science Council at the central government level, which started operating in December 1980. HSIP has attempted to attract investment by high-tech companies and spur the growth of high-tech industry in Taiwan. Over the past 15 years, the Taiwan government has allocated a total of US$483 million for the HSIP. These funds have been used primarily to develop and maintain the park's physical and non-physical infrastructure, and for educational institutions within the park area. Currently, HSIP includes 424 ha of developed land; another 156 hectares of land is currently under develop-ment.

The number of companies in the HSIP grew from seven in 1980 to 180 in 1995. HSIP companies are classified into six categories: integrated circuits (IC), computers and peripherals, telecommunications, optoelectronics, precision machinery and materials, and biotechnology. HSIP firms' combined sales reached US$11.3 billion in 1995. Of the companies inside the park, 36 are foreign-owned and 144 are locally owned. Aggregated investment reached US$5.6 billion by the end of 1995 (Tables 1 and 2). Local sources accounted for 88% of HSIP investment capital, while foreign sources accounted for 12% (HSIPA, 1996a).

The significant contribution of HSIP's production has been widely recognized and confirmed in the literature (Wang, 1991; Luger & Lee, 1994; Lin & Yang, 1994). More specifically, HSIP plays an influential role in Taiwan's high-tech industrial development. HSIP houses nearly all the IC manufacturing companies

CKS International

Legend:

Figure 1. Geographic location of Hsinchu Science-based industrial park.

in Taiwan. IC products are the key components that the island's high-tech industries have heavily relied on, and are Taiwan's largest import items since the late 1980s. With 47 firms in operation, HSIP's computers and peripherals industry in 1995 recorded total sales of US$4, 493 million. According to statistics, the total value of information products from Taiwan in 1995 was US$21.3 billion, US$19.7 billion of which was accounted for by hardware products. Taiwan is the world's third largest supplier of information products, surpassed only by Japan and the USA. In many product categories, an extremely high proportion of Taiwan's output originates from the HSIP (Table 3). Among the leading

Table 1. Summary of HSIP operations Year 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 No. of companies 26 37 44 50 59 77 94 105 121 137 140 150 165 180 193 Registered capital (US$ million) 38 59 110 144 177 546 769 1345 2149 2377 2901 3072 4403 7088 11447 Corporate sales (US$ million) NA NA 238 264 450 866 1737 2124 2443 2903 3406 4810 6706 10 940 5834 No. of employees NA 3583 6454 6670 8275 12 201 16 445 19 071 22 356 23 297 25148 28 416 33 538 42 257 50 792

Source: Hsinchu Science-Based Industrial Park Administration (June 1996) Statistics Quarterly.

Table 2. HSIP industries in 1995

Industry Integrated circuits Computers and peripherals Telecom. Opto-electronics Precision machinery and materials Bio-technology Total No. of (No.) 56 41 30 26 18 9 180 firms (%) 31 23 17 14 10 5 100% No. of employees (No.) 22 496 11148 4071 3270 1041 231 42 257 (%) 53 26 10 8 2 1 100% Corporate sales ($)* 5591 4593 643 379 94 8 11308 (%) 49 41 6 3 0.8 0.2 100

Note: *Unit = US$ millions).

Source: Hsinchu Science-Based Industrial Park Administration (1996a).

companies, Acer Incorporated established itself as the sixth largest supplier of personal computers in the world in 1995 (HSIPA, 1996a).

Critical Success Factors of HSIP

Location factors are frequently treated as the determinants of technology park development and discussed by geographers and urban planners. Technology, market, labour and capital are the four basic elements required for high-tech industries to survive. Technology parks with location factors deemed appropri-ate to capture and maintain these four elements usually enjoy a higher likelihood of success. Location advantages such as proximity to a metropolitan

Table 3. Market share of HSIP's information products Product type Monitors Desktop computers Image scanners Portable computers LAN adapters Taiwan's world ranking 1 3 1 1 1 World market share (%) 70 10 64 27 38 HSIP companies' domestic market share (%) 56 52 50 42 39 Source: Hsinchu Science-Based Industrial Park Administration (1996a)

area, easy connection to good air transportation, affiliation with a university and R&D institutions, and high-quality living environment are considered prerequi-sites for technology park development in many related reports. However, they are not sufficient conditions. As Castells & Hall (1994) pointed out, the California culture for Silicon Valley development cannot be copied easily in other places. Luger & Goldstein (1991) concluded in an extensive study that park policies and procedures are also critical factors influencing the success of park development. Many technology parks have failed, despite appropriate location factors.

Given that HSIP has successfully achieved its pre-defined operation goals, identifying critical success factors of HSIP is a worthwhile task undertaken herein to compare with other success models. In political slogans, it is very easy to attribute success to the excellence of the government's lead and support. However, if the right actions are not taken, government-led projects do not automatically deserve success. The question is, what are the right actions to be undertaken by the government? As Porter (1990) contended, if industries clus-tered in a certain area are to be successful in a global market, they must be supported by certain factors with a competitive advantage. Important factors, although incomplete, that may have been forming the competitive advantages of HSIP are identified and discussed below.

Development Strategy

The development strategy is important to a park's survival and success. Econ-omic theory suggests that a country in the early econEcon-omic development stage should encourage inward investment via transnational companies and then develop sufficient technological capacity so as to improve and rejuvenate exist-ing industries. In contrast, the developed nations should emphasize basic scientific research and its technological applications, particularly in respect of key technologies (Castells & Hall, 1994; Luger & Lee, 1994). The production-oriented strategy of HSIP's early development and targeting at the information industry has paved the right path for successful operations.

Location Advantage

Although location factors are not sufficient conditions, they influence a park's competitive advantage. HSIP enjoys almost every success factor identified previ-ously (Luger & Goldstein, 1991). For instance, within a radius of two miles, HSIP

can easily gain access to two universities with sound reputations in engineering (National Chiao-Tung University and National Tsing-Hua University); an ad-vanced research unit (Industrial Technology Research Institute, ITRI); and interchanges of the two major freeways. In addition, CKS international airport and Taipei (the biggest metropolitan area in Taiwan) can be reached by car within 45 minutes and 75 minutes, respectively. Since HSIP is a production-oriented technology park, firms have heavily relied on the CKS International Airport for transporting imported components and exported products. The proximity to Taipei metropolitan is not only beneficial to HSIP for so-called proximity to innovative milieu and convenient recruitment of skilled labour, but also critical to the production links made with existing electronic firms in Taipei. It is often overlooked, that most early established firms in HSIP are subsidiary investments and spin-offs of electronic firms in Taipei, Acer Incorporated being one of them (Hsieh, 1994).

ITRI is important to HSIP in terms of providing directors and managers of R&D and technical fields, utilizing R&D facilities and offering technical consult-ing. Regarding the connections with two universities near HSIP, Kung (1995) confirmed that two universities are important to the training and recruitment of technology labour for HSIP; however, their contributions to the technology acquisition of HSIP are trivial. As is generally accepted, the prohibition of private business involvement in faculties of a public university is the major obstacle to university spin-off.

Physical Environment

Recruiting high-tech professionals is the most direct method of acquiring tech-nology. However, they are not numerous and most live in a high-quality living environment in well-developed nations. Undoubtedly, recruiting high-tech pro-fessionals is not an easy task for a new technology park. It is not just a matter of how much to pay; rather, it is a matter of how to provide an environment that is attractive to families of high-tech professionals and high-tech companies, since most high-tech professionals are relatively young. Most are married and their children attend school. Meanwhile, high-tech companies prefer to be located in a high-quality environment for the corporation's image.

To provide an attractive environment, the new town approach of HSIP's physical planning is a positive contribution. Not only was the landscape care-fully planned, but mixed land use was planned to provide industrial, residential and recreational areas. The industrial areas, in addition to standard factory buildings and sites for custom-designed factories, include green areas set aside to give the park a unique verdant environment. The residential areas include not only housing, but sports facilities. Most importantly, the National Experimental High School in HSIP is the most important resource valued by employees' families, particularly those having returned from overseas. The school is the only 'national' school in Taiwan that provides an educational service from kinder-garten to senior high school with bilingual programmes.

Property Development

In addition to a high-quality physical environment, the property development strategy has contributed to HSIP's success in other ways: the Tease only' policy

and the 'expansion' approach. Since HSIP is a government-led project, lands were acquired and developed by HSIP and then leased to high-tech firms at a relatively cheap rate. However, promising tenants were selected according to their high-tech attributes and development potential. Furthermore, HSIP may exempt collection of rental for no more than five years for land leased to a high-tech industry whose technology is deemed of great value to industrial development. The 'lease only' policy not only reduces the capital burden, but also helps high-tech firms to start-up. They have a higher likelihood of succeed-ing because their growth potential has been recognized in the review process. While other technology parks have limited growth space, the continuous 'expan-sion' of HSIP has provided opportunities for high-tech firms to grow and new investments to enter. Consequently, energy for HSIP can be continuously enhanced.

Administrative Services

In addition to the physical environment, efficient and high-quality administrat-ive services are viewed as critical factors to high-tech firms. Service agencies in the HSIP include the Science Park Administration (SIPA), a customs office, two banks, a post office and a telecommunications station. SIPA operates under the auspices of the National Science Council at the central government level. SIPA's main responsibilities are to provide HSIP companies with comprehensive ser-vices and to maintain the park. SIPA's work includes planning, investment services, labour management, business services, building control, public works, land administration, computer network services, warehousing, sanitation, emer-gency services and security. Notably, some of the services are supposed to be provided or supervised by local governments or other governmental agencies. To improve efficiency and quality, they are transferred to SIPA by lawmaking. Owing to such a unique arrangement, the so-called 'one-stop' service was enabled. The customs service provided directly by the Ministry of Financial Affairs is particularly important to firms in HSIP owing to their heavy reliance on transportation speed to compete in the global market.

Taxation Incentives

Duty free and tax exemption statuses are exercised to encourage high-tech production and R&D in HSIP. Machinery and equipment imported for use by HSIP enterprises are exempted from import duties and dues, commodity tax and business tax. Within two years from the date on which it begins to market its product or to render services, high-tech firms in HSIP are eligible to select any fiscal year in the four-year period from such date for exemption from profit-seeking enterprise tax for a consecutive period of five years. Furthermore, if additional investment is made, a profit-seeking enterprise income tax due on additional income may be exempted for a consecutive four years. Taxation incentives not only enhance HSIP's desirability to high-tech investment, but also help the capital accumulation of high-tech firms in HSIP.

Social Culture

Local culture is a subtle but important factor in developing a technopole, as

emphasized by Castells & Hall (1994). It is believed that the Chinese social linkage between Silicon Valley and Hsinchu has helped the start-up of high-tech firms (particularly IC, computer and peripherals manufacturers). Statistics show that returning expatriates have played a vital role in HSIP's growth. By the end of 1995, the number of returned expatriates in the HSIP was 2080, and the number of park companies founded by members of this group was 79 (HSIPA, 1996a). With the knowledge and ideas they have brought back, returned expatri-ates have done much to raise the levels of technology and R&D in the HSIP. In addition, Table 4 depicts the nationality distribution of foreigners currently living in Hsinchu city during the year 1995. It provides us another clue to show the close social link between the USA and Hsinchu.

HSIP's Impact on Technopolis Development: An Assessment

Although regional development was not a primary motivation for establishing HSIP in the beginning, its continuous economic prosperity has eventually trickled impacts down upon technopolis development, although slowly. Wang (1991) conducted the first comprehensive evaluation of HSIP's social and econ-omic impacts on technopolis development (including administration areas of Hsinchu City and County) based on official statistics from 1980 to 1989. He concluded that the regional economic effects owing to HSIP's additional job opportunities were not significant despite outstanding production performances of firms in HSIP. At that time, Hsinchu still had a population loss problem although it was slowing down. Castells & Hall (1994) comprehensively reviewed the HSIP experience in their worldwide investigation of technopoles. Other than a summary of previous findings regarding the appropriateness of HSIP's loca-tion, Castells & Hall (1994) investigated the role of government in particular, concluding that "Hsinchu provides material evidence of the impact of the developmental State on the new shores of the world economy" (p. 109). Obviously, the spatial impacts of local development takes a longer time than economic performance to be evident.

In addition to the aforementioned comprehensive reviews, many other inves-tigations of HSIP's impacts have been undertaken looking at different aspects. Based on previous findings and updated supplementary data, in this paper the regional impacts are summarized in the dimensions of regional economic contribution, population and technology transfer respectively.

Regional Economic Contribution

The average production value of each job in HSIP had reached US$120 000 per year in 1989, as targeted to be achieved in 1991 according to its operation strategy (Wang, 1991). According to HSIPA's recent estimation, HSIP has di-rectly and indidi-rectly contributed to regional economic development in job creation and local business income (HSIPA, 1996b). Some 43 372 jobs were provided by the park as of the end of 1995. Among which 20 633 persons (47.6%) are residents of Hsinchu City; and 12 533 (28.9%) are residents of Hsinchu County. For every 10 persons employed in the manufacturing sector, three are working in HSIP. In addition, 6000 jobs were created by the 200 satellite firms

Year 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 USA 103 155 159 185 192 283 300 381 485 672 401 409 Table 4. Philippine 7 7 7 3 9 10 14 19 20 29 18 152 Nationality of Korea 5 2 4 7 7 10 9 9 9 14 18 18 foreign residents France Canada 8 1 1 1 1 4 5 5 8 10 11 9 4 7 3 10 15 18 19 25 27 31 15 19 Thai 4 5 6 10 7 6 12 20 20 22 36 715 in Hsinchu Japan 19 36 28 26 22 51 59 82 101 121 44 71 Others 152 165 172 163 170 169 183 254 299 353 279 333 Total 302 378 380 405 423 551 601 795 969 1252 822 1726

Source: Hsinchu City (1996) Annual Statistics of Hsinchu City in 1995.

in Hsinchu area. Meanwhile, the labour structure in Hsinchu area was also enhanced by the creation of HSIP's jobs. In HSIP, the proportion of employees with at least a junior or technical college education exceeded 50%; the national average is 7.5%. HSIP workers' average age was 30 years; with females compris-ing 51.5% of the population.

In addition, as estimated, about US$1 billion was spent in Hsinchu region by the workers and firms located in HSIP during 1995, the 17 000 immigrated workers in 1995 have had a tremendous impact on the local housing market. The park also contributed to the city and county US$2.61 million and US$1.81 million, respectively, in terms of estate tax in 1995.

Population

For a long time, Hsinchu area (including Hsinchu City and County) suffered from population loss. For instance, the population in Hsinchu area was 648 145 persons in 1981, the year HSIP began operation, when there was a net social increase rate (immigration—emigration) of — 1%. Since then, an increasing population has immigrated into the area. The population impact of HSIP was concluded to be insignificant by Wang (1991) when Hsinchu still had a net loss of 1972 persons in social migration in 1988. At the end of 1995, 42 257 jobs opportunities were created in HSIP, and the total population in Hsinchu area was 340 255, while the social increase rate reached 0.2%. Restated, Hsinchu area enjoys more immigrants owing to HSIP's development.

Comparing the local impacts of foreign direct investment in export processing zones (EPZ) and HSIP, Lin & Yang (1995) indicated that while EPZ created many jobs that mainly attracted young female singles who stayed in a single dormi-tory, HSIP created more jobs with higher income that attracted married couples with a higher education level. In terms of their population impact, this difference implies that HSIP created more housing and public service demand than EPZ. In fact, the average productivity of each employee in HSIP was US$17.3 million per year in 1993, whereas the national average was US$8.1 million. Higher consumption capability can be imagined. In fact, Wang (1993) investigated the consumption characteristics of HSIP employees, finding strong links with Taipei

Table 5. R&D expenditures and number of HSIP patents in 1994 Integrated circuits, computers and peripherals Telecommunications Optoelectronics Precision machinery Biotechnology Total R&D expenditures (US$ million) 169.95 74.11 34.88 17.70 5.63 2.89 5.16 R&D expenditure sales (%) 5.8 2.7 7.3 15.4 8.1 6.7 4.8 Number of foreign patents (A/B)» 541/165 78/43 5/5 27/13 0/0 1/0 652/226 Number of domestic patents (A/B)* 648/68 84/26 2 / 2 8/5 0/0 0/0 742/101

Note: *A = number of patents applied for; B = number of patents granted.

Source: HSIPA (June 1996) Statistics Quarterly o/Hsinchu Science-Based Industrial Park.

City despite the fact that the local commercial environment in Hsinchu city had improved significantly.

Technology Transfer

Kung (1995), in the most recent assessment, examined the technology linkage of high-tech firms in HSIP through questionnaires and interviews. Kung found that 'self R&D' is clearly the most important way of enhancing technology strength for the 41 responding firms surveyed. Table 5 lists the R&D expenditures and number of patents of firms in HSIP in 1994. This table reveals that IC, computers and peripherals industries are the key areas of R&D investment. While the R&D expenditures sales ratio was 5.4% in HSIP, the national average was only 1%. However, Kung also concluded that except for the deep linkage between ITRI and high-tech firms in HSIP, not much technology transfer within the local area had occurred.

Based on the theory of endogenous technology growth and R&D expendi-tures, Luger & Lee (1994) investigated the experience of HSIP development, concluding that Taiwan possesses the essential ingredients to implement a strategy of endogenous knowledge-based growth. Moreover, HSIP's develop-ment model should be modified if an additional science park is to be developed.

Planning the Technopolis: Hsinchu Science City

Integrating with local development was not seriously considered in the early phase of HSIP development. However, owing to the continuous and rapid growth of employment in HSIP for a decade, a shortage of housing with a high-quality living environment became a critical issue in the late 1980s. A movement for residential community development by the employees in HSIP and ITRI was sparked and promoted by the island-wide real estate price escalation in Taiwan. To accommodate the demand of housing development, SIPA proposed a comprehensive plan for the development of Hsinchu Science City to coordinate and integrate regional development (covering the area of Hsinchu City and Hsinchu County to a total of 1531.69 km2) with HSIP's

development for a term of 25 years. The short-term target year is 1996 and long-term target year 2005. Major development concepts of Hsinchu Science City are described and discussed below (HSIPA, 1993).

Technology Development

To support technology industries located in HSIP and neighbouring areas, a metropolitan business centre and technology interchange centre to enhance the milieu of innovation and high-tech industrial development are to be developed. In addition, based on a regional development policy, a comprehensive R&D centre and renewal projects for existing industrial areas are to be promoted to upgrade existing industries in the Hsinchu area.

Living Environment

Planning and facilitating the development of residential communities for the technology employees are primary concerns. In addition to low-density com-munity development in Shiangshan and Baoshan hillside areas, comcom-munity renewal projects near HSIP are also to be promoted. These developments are aimed at diversifying the high-quality housing to be provided at an appropriate price in Hsinchu Science City.

It is planned that commercial centre and business buildings are to be devel-oped by using existing industrial lands. Improving commercial and business facilities should be helpful to developing Hsinchu into an international city. In addition, as the Hsinchu area is rich in nature and cultural resources, it is expected to provide historic areas and parks for the residents in Science City.

Infrastructure Construction

In addition to construction projects that are already included in national con-struction projects such as the Northern Second Freeway, Western-Seashore Expressway and High Speed Rail projects, two important local construction projects are planned: a mass rapid transit system for Hsinchu area to improve the transportation service level and a teleport project to enhance the information infrastructure.

International Technopolis

Hopefully, the Science City will be internationally attractive to high-tech indi-viduals in terms of a living and research environment. In other words, Hsinchu Science City aimes to become an international technopolis by improving the physical environment. In addition, implementation programmes are also pro-posed to promote international exchange of academic and R&D activities.

Failure Factors of Hsinchu Science City Project

The target year for short-term development of Hsinchu Science City project was 1996. As a performance assessment shows that none of the construction projects newly added to the development plan has been initiated to date, the govern-ment-led Hsinchu Science City project is obviously a failure. Critical failure factors can be summarized as below.

Unwillingness of Local Governments

Hsinchu Science City covers the administration areas of Hsinchu City and Hsinchu County. Although the Science City planning is directly initiated and prepared by HSIPA at the central government level, land-use control and infrastructure construction are under the official authority and are the responsi-bility of local governments. Not only is coordination difficult because the City Mayor and County Governor belong to two different political parties, but also because both local governments are strapped for financial resources; neither of the local governments is willing to promote regional development projects.

Difficult]/ of Land-use Conversion

Zoning is the main instrument of land-use control in Taiwan. Because of the four-tier administration structure, the approval process in applying for land-use conversion is not only time consuming, but also inconsistent among different review agencies. Furthermore, since most land in Hsinchu is in hillside areas, the Order of Building Prohibition in Water Resource Preservation Areas in 1990 stopped many development applications and infrastructure projects.

Insufficient Legal Basis for Private Construction of Infrastructure

In the development plan, most construction projects were planned for invest-ment by the private sector. However, the only legal basis for public construction of infrastructure with incentives was not made available until 1995. Restated, private provision of infrastructure was totally impossible before, that time.

Organizational Powerlessness of Task Group

Recognizing the importance of coordination among different government sec-tors, the Task Group for the Promotion of Hsinchu Science City Development was formed in 1991. Members of the task group come from various related agencies in both local and central governments. Although the task group was expected to coordinate and promote the implementation of development projects, it has proved to be of limited effect. The task group has no power in the decision processes of resource allocation and land development approval. Six meetings of the task group have been held since establishment. At the end of 1996, the last meeting had been held a year earlier, in December 1995.

Future Development of Hsinchu Technopolis

Technology is the fuel that empowers economic development; advanced technol-ogy is the mark of a nation's strength. During the past 15 years, HSIP has played a prominent role in reshaping Taiwan's industrial base and bringing about the development of high-value-added products. The HSIP will continue to grow. According to HSIP's 10-year operation plan, the goals to be achieved in 2003 are sales revenues of US$50 billion, employee numbers reaching 80 000 and number of companies reaching 300.

In addition to the existing developed area of 424 ha, an expansion project of 156 ha is under way. The mixed land-use development of HSIP with a scale of

580 ha within an independent administration boundary allows HSIP to operate like a new town. According to the competitive advantage model (Porter, 1990), the personal computer manufacturing industry in Taiwan has established some leading advantages for global competition, and firms in HSIP should be able to maintain the prosperity of the information industry for some time to come. Based on the theory of growth pole and growth centre and by the nature of a technopolis where technology-induced economic activity provides the pivot for the urban development, spill-over effects on regional development resulting from HSIP's economic prosperity will become increasingly evident.

A system providing a full range of network services was installed in the HSIP in 1991. This system is widely used, particularly for Internet and World Wide Web access. In July 1995, with the government's National Information Infrastruc-ture development programme under way, SIPA completed the installation of a park-wide asynchronous transfer mode (ATM) broadband network integrating automated customs clearance and other information systems. This effort should make HSIP more capable of accommodating the needs of an information economy.

Looking at the infrastructure required for networking, many ongoing projects are favourable for developing the Hsinchu technopolis in terms of urban development and global connections. Privatization of telecommunication ser-vices took effect in July 1996. Information flow will be stimulated and favourable to long-distance interactions. The newly completed Northern Second Freeway has relieved the traffic congestion between Taipei and Hsinchu significantly. In the future, a new high-speed rail station will be situated near HSIP, which will encourage Hsinchu's interaction with other local cities. The expansion project of CKS international airport into a regional air hub will definitely be helpful to international passenger transportation and cargo transportation between Hsinchu and elsewhere in the world.

Difficulties encountered in coordinating various governmental agencies at the central and local levels will continue to persist in future technopolis develop-ment. Similar to the nature of metropolitan development, an inter-governmental coordination mechanism is a critical factor influencing the efficiency and quality of public service provision. This issue might become even more serious owing to the continuing decentralization movement in Taiwan's bureaucracy.

Conclusion

Building a technopolis is not a policy that is likely to prove successful in the short term, as emphasized by Castells & Hall (1994). It might take 15 to 25 years for the full impacts to become evident, since the spatial development is organic in character, and networks and linkages are established over time. Judging from the 15 years' experience of HSIP development, spontaneous technopolis devel-opment is evident.

The economic success of HSIP (the core of the technopolis) has evidently stimulated the regional development of surrounding areas in the long term despite the fact that regional development was not planned, even not con-sidered, in the earlier stages. While technology is treated as the cornerstone to make a technology park come alive, the market provides the energy for the technology park to survive in the long term. The prosperity of the PC manufac-turing industry has played a driving role in the continuous growth of HSIP.

HSIP has long been criticized as 'not scientific enough' because it has more production firms than R&D labs. However, such production-oriented develop-ment is really the key for HSIP to maintain its economic vigour. As was argued by Luger & Lee (1994), HSIP's early success in production has made it more attractive and capable of stimulating indigenous R&D.

Although many technology parks were planned to be developed, not all projects produced favourable results. The performance of a government-led technology park project is a function of the support level and the efficiency of public administration. Since the HSIP project is a purely government-led invest-ment project, the governinvest-ment's role should be of particular importance to HSIP's development and deserves discussion to conclude this paper.

HSIP's success has benefited from governmental support in many facets, as discussed earlier. Among those support items, however, the institutional set-up is critical in terms of administrative efficiency. The bureaucratic hierarchy in Taiwan is vertically a four-tier system (State, Province, City/County, Town/ Hsiang), while the lower tiers are notorious for their poor efficiency, and difficulties of horizontal coordination among different government agencies are often complained of. Through lawmaking (Statute for the Establishment and Administration of a Science-Based Industrial Park), HSIPA is authorized to take over tasks of local governments in land use and building management, indus-try/business administration and public works, and to supervise the branch offices of related public services including customs office, high school, banks, post office and telecommunications station. With the centralization of authority to HSIPA, high environmental quality and efficient public service can be maintained for the residents and firms in the park. Furthermore, although it seems ironic for the National Science Council to sponsor the operation of a production-oriented park (HSIP), whereas the Ministry of Economic Affairs runs the research unit (ITRI), in fact, the role exchange has made collaboration and interchange much easier, particularly in the aspects of R&D and spin-offs.

Interestingly, while such an institutional set-up contributes so much to the success of HSIP, it becomes an obstacle to the implementation of Hsinchu Science City Development Plan as contended herein. How can such an institu-tional paradox be solved? This a question suggested for inquiry in future studies.

References

Berry, B.J.L., Conkling, E.C. & Ray, D.M. (1993) The Global Economy (Englewood Cliffs, NJ: Prentice-Hall International Editions).

Castells, M. & Hall, P. (1994) Technopoles of the World (London and New York: Routledge). Conway, M. (1993) Airport Cities 21 (Atlanta, GA: Conway Data).

Dicken P. (1992) Global Shift: The Internationalization of Economic Activity (New York and London: Guilford Press).

Goldstein, H.A. & Luger, M.I. (1993) Theory and practice in high-tech economic development, in: R. Bingham & R. Mier (Eds) Theories of Local Economic Development, pp. 147-174 (Newbury Park, CA: Sage Publications).

Hodgson, B. (1992) Technopolis: challenges and issues: a tale of (at least) three cities, in: D.V. Gibson, G. Kozmetsky & R.W. Smilor (Eds) The Technopolis Phenomenon (Lanham, MD: Rowman & Littlefield).

Hsieh, H.J. (1994) Characteristics of the Spatial Distribution of Information Industries in Taiwan, master's thesis, National Taiwan University Taipei (in Chinese).

Hsinchu Science-based Industrial Park Administration (1993) Development Plan of Hsinchu Science City (Hsinchu, in Chinese).

Hsinchu Science-based Industrial Park Administration (1996a) Annual Report of Hsinchu Science-based

Industrial Park in 1995 (Hsinchu).

Hsinchu Science-based Industrial Park Administration (1996b) Economic contribution of Hsinchu

Science-based Industrial Park to local development, unpublished report (in Chinese).

Kung, S.F. (1995) The role of science parks in the development of high technology industries—with special

reference to Taiwan, PhD dissertation, University of Cambridge, UK.

Lin, C.Y. & Yang, S.P. (1994) Technology park and local economic development, in: Proceedings

of International Symposium on City Planning, p p . 491-198 (Fukuoka: Japanese City Planning

Association).

Lin, C.Y. & Yang, S.P. (1995) Foreign direct investment and local direct investment in Taiwan, in: Proceedings of International Symposium on City Planning, (Kwanju: Korean Urban Planning Association).

Luger, M.I. (1996) Spontaneous technopolises and regional restructuring: the case of Research Triangle, NC, in: Proceedings of The International Symposium on the Technopolis, Its Vision and Future, p p . 5-30 (Taejon, Korea: World Technopolis Organizing Committee).

Luger, M.I. & Goldstein, H.A. (1991) Technology in the Garden (Chapel Hill, NC: University of North Carolina Press).

Luger, M.I. & Lee, Y.J. (1994) The role of science parks in East Asia economic development, in:

Proceedings of the 1994 International Symposium on Industrial Park Development Policy and Planning,

p p . VI-1-19 (Taipei: National Taiwan University).

Porter, M.E. (1990) The Competitive Advantage of Nations (New York: Free Press).

Wang, H.K. (1993) An Analysis of the Living Periphery of HSIP's Employees (Taipei: Institute of Building and Planning, National Taiwan University, in Chinese).

Wang, K. (1991) An Evaluation of the Attraction Effects of High Tech Industries—Case Study of HSIP (Chungli, Taiwan: National Chung-Young University, in Chinese).