Action strategies for strengthening industrial clusters in southern Taiwan

Technology in Society 28 (2006) 533–552

Action strategies for strengthening industrial clusters

in southern Taiwan

$

Ting-Lin Lee



Department of Asia Pacific Industrial and Business Management, National University of Kaohsiung, 700 Kaohsiung University Road, Kaohsiung 811, Taiwan

Abstract

For the past decade, the Taiwanese government has applied the policy of ‘‘North heavy, South light’’ to put more emphasis on development in northern Taiwan instead of the south. This has resulted in uneven development between the northern and southern regions, especially when introducing, developing, and supporting high technology and resource allocation in education. Industry clusters have become the centerpiece of economic development policy in many parts of the world. In this study, it is assumed that there are two categories of less-advantaged regions in Taiwan: (1) older industrialized regions dominated by labor-intensive/capital-intensive industries, and (2) industrial regions that have merged with potential high-technology small firms but still lack infrastructure. This study discusses the following points: (a) How do industrial clusters work in action? (b) What barriers do less-advantaged industrial clusters face? (c) What action strategies promote less-advantaged industrial clusters?

r2006 Elsevier Ltd. All rights reserved.

Keywords: Innovation system; Industrial cluster; Social infrastructure; Taiwan

1. Introduction

Traditional agriculture and heavy chemical engineering industries have long been the main industries in southern Taiwan owing to the government’s long-held policy of ‘‘North heavy, South light.’’ As a result, the south lacks development in the fields of social science, www.elsevier.com/locate/techsoc

0160-791X/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.techsoc.2006.09.007

$

An earlier vision of this paper was presented at the Portland International Conference on Management of Engineering and Technology (PICMET), Seoul, Korea, July 31–August 4, 2004.

Corresponding author. Tel.: +886 7 5919 246; fax: +886 7 5919 428. E-mail address:linda_lee@nuk.edu.tw.

technology industries, and innovation—even education levels are lower (see Table 1). Furthermore, with professional workers moving toward northern Taiwan, and the unbalanced distribution of educational resources and development, it is not possible to construct innovative industrial clusters in the south. If the situation continues in this classic feedback loop, it will result in even greater concentration of advanced industries in northern Taiwan.

This research focuses on industrial clusters in two regions. The first is in southern Taiwan, an area that is labor-concentrated and oriented toward basic, mature industries in a region that has already lost its competitiveness to other developing countries. The second cluster is based on high technology, although newly developed industrial clusters have not yet achieved economies of scale or cluster effects and are just at the primary stage, with incomplete infrastructures, few marketing effects, and low company rental rates in comparison to other mature clusters (e.g. Hsinchu Industrial Science Park).

The purposes of this research are to observe the current development of industry clusters in southern Taiwan in a systematized and benchmarking analysis approach, to identify and investigate the barriers that less-advantaged industrial clusters face, and to develop possible action strategies for social and value networks that could build innovative industrial clusters. The paper is structured as follows. Section 2 describes the nature of industry clusters in the southern Taiwan. Section 3 defines ‘‘cluster’’ and its implications for social infrastructure. Section 4 provides an analysis of this research, including methodology, procedures, and scope of the study. Section 5 briefly describes the status of and barriers to industrial clusters in southern Taiwan. Section 6 highlights action strategies for the less-advantaged industrial clusters in southern Taiwan, and the final Section 7 offers policy implications and conclusions.

2. Overview of industrial clusters in Taiwan 2.1. Industrial parks

Taiwan has over 80 industrial parks spread throughout the island on all coasts, covering 11,000 ha of land, of which 1686 ha are located in southern Taiwan. Industrial parks have been the cornerstone of Taiwan’s industrial development since their creation in the 1970s, and are a foundation of the island’s economic success. Under the guidance of the Industrial Development Bureau (IDB), Taiwan’s industrial parks were originally designed to manage land development while simultaneously considering rising industrial growth and

Table 1

Workforce comparison of Taipei and Kaohsiung metropolitan areas City/county Number of Ph.D.s and

Master’s degrees

Residents age 15+ Ph.D. and Master degree/ Residents age 15+ (%)

Taipei City 87,654 2,145,892 3.32

Taipei County 36,491 2,900,785 1.26

Kaohsiung City 18,607 1,216,475 1.53

Kaohsiung County 7495 995,139 0.75

Source: Taipei City Statistical Overview 2002; Taipei County Statistical Overview 2002; Kaohsiung City Statistical Overview 2002; Kaohsiung County Statistical Overview 2002.

environmental concerns. Today, the concept of industrial zones has expanded to include science parks and parks that emphasize manufacturing support for high-tech industries. Development goals for industrial parks include:



provide adequate space for companies wishing to build factories,



generate the economic benefits of industry clustering,



prevent disturbances to industry and the peripheral environment,



promote local economic development,



create an environment that emphasizes high added-value production,



foster the development of strategic industries.

2.2. Economic Processing Zones

In addition to industrial parks, there are also Economic Processing Zones (EPZs). These were initiated in the late 1950s and 1960s to set up export processing zones in Taiwan. Their four goals were: ‘‘Attract Foreign Industrial Investment,’’ ‘‘Explore External Trade,’’ ‘‘Increase Job Opportunities’’ and ‘‘Introduce Modern Technology.’’ At the time, the government was facing financial difficulty, there was a shortage of foreign reserves, population was increasing, and unemployment rates were high, so there was an urgent need for the government to develop industries for export. At the same time, manufacturing industries in other industrialized countries were beginning to move overseas. A high-efficiency, free, special zone was needed to facilitate export processing. The government established the first export processing zone in Taiwan, the Kaohsiung Export Processing Zone (KEPZ), in 1966. With a unified authority and highly efficient system predicated on simplified procedures, EPZs have played a vital role in Taiwan’s economic development and made outstanding contributions to the accumulation of foreign exchange and the introduction of new technologies.

2.3. Science-based parks

The objectives of science-based industrial parks are (a) to establish a base from which to develop high-tech industries, and (b) to cater exclusively to the needs of high-tech development, utilizing resources from industry, government, and academia to create an innovative environment that smoothly integrates R&D and manufacturing, and promotes the upgrade of Taiwanese industries.

A major science park located in northern Taiwan is Hsinchu Science-Based Park (HSP), covering 1175 ha of land. It was the island’s first science park and is the home of Taiwan Semiconductor, the world’s largest made-to-order IC manufacturer.

Located between the southern cities of Tainan and Kaohsiung, and covering more than 1608 ha of land, Southern Taiwan Science Park (STSP) is a conglomeration of Tainan Science Park and Luchu Science Park. It serves as home to a growing number of IT and electronics companies and also accommodates the biotechnology and agricultural industries (seeTables 2 and 3).

3. Definition of a cluster

According to Porter[1]a cluster is ‘‘a geographically proximate group of interconnected

complementarities.’’ More typically, a cluster is defined as a concentration of companies and industries in a geographic region, which are interconnected by the markets they serve and the products they produce as well as the suppliers, trade associations and educational

institutions. Another definition by Enright [2] points out that an industry cluster is ‘‘a

group of business enterprises and non-business organizations for whom membership within the group is an important element of each member firm’s individual competitive-ness. Binding the cluster together are buyer–supplier relationships, or common

technologies, common buyers or distribution channels, or common labour pools’’ [3].

Rosenfeld[4]indicates that an industry cluster is ‘‘a geographically bounded concentration

of similar, related or complementary businesses, with active channels for business transactions, communications and dialog that share specialized infrastructure, labor markets and services, and that are faced with common opportunities and threats.’’ A broad

Table 2

A comparison of Hshinchu Science-based Park (HSP) and Southern Taiwan Science Park (STSP)

HSP STSP Number of companies 384 157 Number of employees 113,011 32,793 Education Total 7336(6%) 166(0.5%) Others 30,015(27%) 10,631(32.4%)

Senior High School 25,571(23%) 10,378(31.6%)

Junior College 27,329(24%) 7211(22%)

Bachelor’s degree 21,465(19%) 4159(12.7%)

Master’s degree and above 1295(1%) 248(0.8%)

Area (ha) 1175 1608

R&D expenditures/Sales 7.8% N/A

Note: STSP includes Tainan Park (638 ha+extension 400 ha) and Kaohsiung Park (570 ha). Source: HSP and STSP websites, 2004 statistics

Table 3

Company categories in SHP and STSP

HSP STSP

Number of firms Sales statistics (NT$ Hundred million)

Number of firms Sales statistics (NT$ Hundred million)

Integrated circuits 164 7427.38 22 831.54

Computers and peripherals 58 1382.45 3 9.04

Telecommunications 52 605.30 15 8.72

Opto-electronics 61 1312.63 47 1685.76

Precision machinery 21 92.47 41 46.00

Biotechnology 28 25.39 29 11.55

Total 384 10,859.22 157 2592.61

cluster captures important linkages and spillovers of technology between industries, which are fundamental to competition, productivity, and innovation.

Most definitions of industry clusters refer to the geographic scope of the cluster, and the importance of proximity. While geographic scope is important in defining clusters, every cluster has different geographic requirements, so there is no uniform definition of the

appropriate geographic scope of a cluster.1Most authors argue for a broader definition of

clusters that are defined by both horizontal and vertical relationships, and include both direct and indirect linkages.

The geographic scope of a cluster ranges from a single city or region to a country or even a network of regions in neighboring countries. Thus, clusters vary in size, depth, and level of aggregation across geographic localization, among industries, and from cluster to cluster. Furthermore, the content of a cluster changes when new requests are made as a result of government policy.

The main economic goals of a cluster are to enhance economic performance, lessen transportation costs, decrease production costs, and achieve external economies of scale.

Doeringer and Terkla[14]believe in ‘‘geographical concentrations of industries that gain

performance advantages through co-location.’’ Performance advantages can be achieved through cluster corporations that create extensive learning opportunities, increase productivities, reach economies of scale and scope, share risks and R&D costs, allow greater flexibilities, and shorten time-to-market for new products or new technology processes.

4. Implications for social infrastructure

Recently, the role of social infrastructure in defining industry clusters has become

pervasive in the literature. Rosenfeld[4]points out that information flow is critical in an

effective industry cluster and, in order to facilitate information exchange, social infrastructure is required. He points out that while the characteristics of a cluster may be present, it is not necessarily an effective cluster. An effective cluster must also include social interaction, trust, and a shared vision in order to create its dynamic nature. It is the dynamic interactions and functional relationships between firms and industries that

characterize a cluster[14]. Furthermore, Hart and Simmie[15]indicate that agglomeration

economies could be divided into three sub-sets: Types A, B, and C.2 These types are

contained within the Agglomeration Economies/Local Production Network Paradigm Set, i.e. the three subsets share common elements of the main set. In Type C, social networks are established between individuals within firms and between individuals in different firms. Although speed is important, the Type C clusters actively promote innovation rather than 1_{Several theoretical concepts have been established to describe the characteristics of such regional systems,}

for example: industrial districts, industrial clusters, innovative milieux, regional innovation systems, etc. (see

[5–13]).

2_{Cohesive clusters (Type A) are groups of firms that initially locate together to reduce costs. It is assumed that}

transport costs are a function of weight and distance. The concern is to keep the costs of movement associated with material assembly and subsequent distribution to the market to a minimum. New Industrial Districts (Type B) tend to be knowledge-based, often with a high proportion of companies in high-tech sectors. They rely extensively on R&D for the creation of new products. Innovative Milieu (Type C) is based on the work of researchers who emphasize the importance of social capital to promote innovation.

simply responding rapidly to it; they work together to promote common medium- and long-term innovative goals.

Generally speaking, the cluster concept focuses on the linkages and interdependencies among actors in the value chain to produce products, services, and innovations. Clusters differ from other forms of cooperation and networks in that the actors involved in a cluster are linked in a value network. The cluster concept goes beyond ‘‘simple’’ value chains in which forms, operating in the same end-product market and belonging to the same industry group, cooperate in aspects such as R&D, demonstration programs, collective marketing, or purchasing policy. Clusters are often cross-sectoral (vertical and/or lateral) networks made up of dissimilar and complementary firms specializing around a specific link or knowledge base in the value network.

5. Methodology and structure

The following discussion draws onFig. 1and itemizes choices made within the range of

options listed there. The figure shows the research scope and study flow. 5.1. Choosing dimensions of measurement

The primary method for exploring the questions posed in the Abstract was a series of lateral/parallel comparative case studies in different fields of science-based or technology-based industries in several locations. The principal mode of data collection was in-depth interviews with firms, university researchers, administrators, and policy-makers. Drawing primarily on these interviews, and augmented by benchmarking analysis using local and regional statistical data as appropriate, we traced the scientific and industrial development of the field in each location, then analyzed and compared the circumstances. Based on the contents of each cluster in this research, three dimensions of measurement—Linkage,

Geography, and Time—[16]were taken into account and adopted in order to choose the

appropriate analytical clusters (seeFig. 1).

In the geographic dimension, industrial clusters were chosen by the co-location of firms in specific regions. The industrial districts observed here are all localized in a specific area (southern Taiwan). For the linkage dimension, firms within the chosen cluster were members of the same extended value-network and most firms were manufacturing industries (a labor-based cluster), while for the time dimension only declining or potential (future) clusters were considered. After a comprehensive survey, two types of clusters located in the south of Taiwan were selected for further analysis based on the features of being ‘‘localized value-networks,’’ and in the categories of being ‘‘declining’’ or ‘‘potential.’’ They were: (a) the Southern Taiwan Science-based Park (STSP), which includes Tainan Science Park and Luchu Science Park, and (b) 19 industrial parks located in southern Taiwan plus three EPZs.

5.2. The choice of the Cluster Level of Analysis

The first research question is explored using a combination of case study and interview approaches. The cluster level of analysis is grounded on the OECD concept identified in the present case as the ‘‘meso level,’’ which involves inter- and intra-industry linkages in

different stages of the production chain of similar end products[17]. The analysis is further carried out via benchmarking.

5.3. The choice of Cluster Content Analysis

Following Porter’s[18]classification, it was found that most of these clusters in southern Taiwan are ‘‘vertical-integrated’’ clusters, i.e. buyer–seller relationships exist among the firms within the cluster.

NIS RIS Cluster Linkage _Geography Time Innovation System + Interacting learning + Policy Instruments

General Cluster Analysis Focused Cluster Analysis

Methodology

Qualitative (Interview) Quantitative (LQ)

Suggested Guiding Strategies In - d e p th

In te rv ie w

Knowledge flow + Incentive & competitive mechanism The Second Technology Corridor in South of Taiwan Localized Non-localized Analytical dimension of industry cluster Literature review of innovation system Research Questions Value-chain cluster Labor-based cluster Innovation cluster

1. Southern Taiwan Science-based Park 2. Southern Technology Industrial Park (includes 22

Industrial Parks)

Existing Declining Emerging Potential

Cluster level of analysis

Macro Level Meso Level Micro Level Content Analysis Vertical Cluster Horizontal Cluster

5.4. The choice of Methodology

The second question involved qualitative analysis using in-depth interviews and case comparisons between Science-based Parks and Industrial Parks, comparing such things as infrastructure, authority, public facilities, education, tax incentives, and source of

development funds (refer to Table 4). This study evaluates the barriers facing

less-advantaged industrial clusters.

Table 4

Science-based Park versus Industrial Park Factors for

Comparison

Science-based Park Industrial Park

Authority National Science Council Ministry of Economic Affairs Development focus Centered on National Foresight project Focus on industrial technology

development project Public facilities and

infrastructure

 Water and power supply

 Environment protection and industrial safety

 Traffic and telecommunications information facilities

 Entertainment, humanity, culture facilities

 Water and power supply

 Environment protection and industrial safety

 Traffic and information facilities

Mode of exploitation (Development Fund)

Most funds come from government budget According to the Statue for Upgrading Industries,’’ the Industrial Development Bureau of the Ministry of Economic Affairs selects a development and management organization to raise funds by itself; very little funds from government Tax incentives  No taxes are levied on park enterprises

for imported machinery for self use, raw materials, fuel, materials and semi-finished goods. No paperwork required for no-tax treatment, guarantees, bookkeeping or tax withholding.

 Park enterprises that export products or labor services are subject to zero enterprise tax.

 Park enterprises may act in accordance with regulations that form part of the Statute for Upgrading Industries, and apply for a five-year tax exemption or a reduction in personal investment tax.

 Reduced taxes for transshipment businesses where operators pay tax on 10% of revenue

 Tax incentives as stipulated in the Statute for Upgrading Industries

 Goods for export from EPZs are exempt from Commodity Tax

 Customs duty exemption on machinery and equipment, raw materials, fuel, materials and semi-finished products that are imported for private use, and finished products transshipped by trade/warehouse operators.

 Zero business tax on exported goods, exported labor, purchased goods

 Deed tax exemption on newly built plants in EPZs and buildings purchased from the EPZs

 Housing tax at EPZs half of normal rate (1.5% vs. 3%)

5.5. The choice of Policy Instrument

The action strategies for promoting dynamic networks that support less-advantaged industrial clusters are proposed by means of a ‘‘general cluster analysis’’ instead of a ‘‘focused cluster analysis’’[19].

6. Characteristics of southern regional clusters

Revised from the model of innovation system [20], the characteristics of southern

regional clusters, including five sub-systems and an exogenous macro environment, are

shown in Fig. 2. These characteristics [21] can be used to compare clusters with other

similar clusters in southern Taiwan or with clusters in northern Taiwan by identifying weaknesses and selecting appropriate actions.

6.1. Human resources sub-system 6.1.1. Workforce skills and availability

Under the government policy of ‘‘North heavy, South light,’’ southern Taiwan has for a long time been unable to obtain adequate resources for development. Evidence indicates that people with higher skills, or with possibilities for training, or with further education, leave the region seeking jobs in the bigger cities, such as Taipei and Hsinchu. This phenomenon is especially obvious for employees in Science-based Parks. The labor force

Human Resources (HR) Manpower (University; Public, Private, or Quasi-government Research Institute) Education & Training

Innovation Commercialization (I) (Product &Process Innovation ) Process Innovation Product Innovation New Innovation Capacity

Financial (Capital Market) (F) Government R & D Funding Private R & D Budget

Venture Capital

Foreign Depository receipts (ADR,GDR, ECB)

Science and Technology (ST)

Transfer Pattern Venture, Alliance, Licensing,Turn key etc..)

RD capacity (Education , Experience, Technology process capacity, Patents, Publications , Annual R&D expenditure per person )

Product Market (M) Rival or Partner Forward & Backward chain Market Share

Knowledge & Skill

Embedded Knowledge

Equipment and New Plant Investment

New Products Labor

Cost

International Environment

Government Macro Policy of Economic, Legal & Financial

Mental Skill

Internalization Manpower Prediction & Recruitment

Marketing Investment Essential Managers

Technological Skill Demand

New S & T Demand VC Demand Market Incentives (spin-off, joint Technology Transferring Investment Profits Sales profits

Fig. 2. Innovation system and its interactions. Source: Revised from Lee and von Tunzelmann[20]. Note: Dotted line means market pull, true line means technology push.

that remains in southern Taiwan has more technical skills than research skills (18.4% of the total are administrators, 37.3% are technicians, 44.3% are non-technical labor). These characteristics affect the regional labor market in disadvantaged clusters, suggesting that there are problems in renewing the skills base of the region. For many industries in southern Taiwan, it is difficult to interest younger people with R&D skills to work in the industries in this region.

6.1.2. Education and training

In a recent study by Lee [22], it was pointed out that there is always a shortage of

advanced or senior researchers in southern industries, no matter whether in the high-tech or traditional industries. Every year the Science Parks offer professional training services, such as lectures, workshops, and seminars; HSP offered training for over 9000 people/time, while STSP offered training for more than 800 persons/time in 2004. It is obvious that, regardless of the quality or quantity, the workforce is insufficient in southern Taiwan. Therefore, determining how to incubate a local embedded workforce and to build a labor pool for industries in southern Taiwan is the key issue. The lack of talent, funding, and infrastructure results in a shortage of skilled workforce, an insufficient R&D capacity, and a low level of education and training.

6.2. Science and technology sub-system 6.2.1. R&D capability

Taiwan’s SMEs account for 98% of all firms, most of them less inclined, due to limited capital, to develop a new product or new process technology when technological know-how or applications can be bought in the marketplace. Besides, most SMEs in the Industrial Park are in the manufacturing industry and are, by nature, subcontracted producers. Thus they do not incubate their own R&D capacity, instead licensing in technology or forming strategic alliances. Furthermore, there are few lead firms within the cluster, so there are no cluster products or technologies. Although SMEs at STSP are almost all high-tech firms, the scale (of researchers, capital, size, finance, etc.) is still limited.

6.2.2. Innovation and imitation

As described above, because only a few lead firms are located in the southern cluster, it is difficult to conceive and develop new and state-of-the-art technologies and products, much less bring them to market. R&D in all of Taiwan is relatively insufficient, regardless of conditions or scale, and this circumstance is even worse in southern Taiwan. Innovations seldom come from an R&D department but instead through turnkey projects, licensing, or alliances. More specifically, new technologies are transferred from a leading foreign company and then imitated or reverse engineered.

6.3. Market sub-system

6.3.1. Presence of market leaders and innovators

The most successful clusters have lead firms that are part of global networks and markets, and they employ people who are active in international professional associations and build personal networks. However, there are few acknowledged market leaders and

magnet firms in the southern Taiwan cluster. Despite the presence of high-tech industries, such as optoelectronics and semiconductors, in STSP, most optoelectronic firms are start-ups, whereas semiconductor firms are mostly branches of SHP. Thus it is difficult to access international associations and markets, and to keep pace with international trends. Nevertheless, there are a few headquarter operations setting up in the southern clusters (see

Table 5). Among the 22 Industrial Parks in southern Taiwan, this situation is much worse

than in STSP.

6.3.2. Proximity to suppliers

The firms within the southern industrial cluster appear to constitute a kind of supplier–user pattern of relations. No matter where the cluster is, firms in the geographic area automatically form supplier–user relationships. For example, the optoelectronic and semiconductor industries are concentrated in the STSP and Tainan Technology Industrial Park; the IC, electronic and information industries center are in the Nantze EPZ; the petrochemical industry is concentrated in the Lin-Yuan Industrial Park; the vehicle components industry is gathered in Pintung Industrial Park. Each is close to sources of primary and secondary supplies, materials, and services that minimize transaction costs and maximize interactions.

6.4. Financial sub-system

The industrial structure in southern Taiwan clusters is dominated by small, sometimes micro, companies. There are about 3243 firms in the southern industrial parks, while there are 3951 firms in the northern ones. Capital investment in the southern industrial parks is about NT$1670 million, while investment northern parks is about NT$1708 million (see

Table 6). It is almost axiomatic that the R&D budgets of SMEs in southern Industrial Park

are relatively limited. 6.4.1. Capital availability

To support the government policy of national industrial development, some banks allow investing and provide needed finance to high-tech firms. They supply commercial loans at very low interest rates to erect plant and purchase machinery. This service is common in each science-based or technology-based cluster, and it encourages the development of science and technology as well as reduces the burden on firms in the cluster. However, venture capital companies are generally located in just two or three of the large metropolitan cities. In 2003, there were 241 venture capital funds, but only four funds were

established in southern Taiwan [23].

Table 5

Geographical distribution of Operation Headquarters established in Taiwan

Northern Taiwan 93

Central Taiwan 25

Southern Taiwan 14

Taipei City 62

Kaohsiung City 3

Venture capital is also not especially patient and therefore tends to limit itself to companies or industries that are entering the takeoff stage, not those in the mature stage. In May 2002 the Executive Yuan unveiled a new program called ‘‘Two Trillion and Twin Star Industries Development Plan.’’ The ‘‘two trillion’’ refers to the semiconductor and flat-panel display industries—industries that are relatively mature so there are few investment opportunities for venture funds. Besides, clusters in southern Taiwan are mostly in traditional manufacturing, which is not of interest to venture capital funds. Although optoelectronic industry centers (such as the flat-panel display industry) are in STSP, the industry is too mature to interest the venture capital industry. All this means that firms in the emerging or older industry clusters fail to obtain working or start-up capital.

6.5. Government policy

The development of government policy has its own stages. Each stage has different questions that need to be solved, different policy tools that must be used, different target groups to be faced, different duties to be implemented, different policy contexts to be separately dealt with, and perhaps different policy needs.

The establishment of STSP imitated the successful experience of HSP and it is successfully resolving the problems of growth that were limited in HSP. Faced with the varied conditions of space and time, however, the government might have different policy tools to improve STSP’s growth. It seems clear that government should assist traditional manufacturing industries in southern industrial parks, which are facing the need to globalize, upgrade their technologies, and transform themselves to remain competitive in the future.

6.5.1. Specialized services

‘‘Specialized services’’ here means public-sector services, such as technology extension services, technology centers, export assistance, small business centers, and private-sector services provided by designers, engineering consultants, accountants and lawyers who have special knowledge of the cluster. Most of the clusters in Taiwan offer convenient services,

Table 6

The competitiveness of Northern and Southern Industrial Parks

Northern Industrial Park Southern Industrial Park (include 3 EPZs)

Number of Industrial Parks 15 22

Area in Total (ha) 1793.5 1686

Number of firms 3951 3243

Number of employees 189,968 165,515

Capital (million NT$) 1,708,034.91 1,669,831.59 Education (%)

Junior college and higher 25.03 23.90

Senior High School 37.25 35.17

Junior High School and under 37.72 40.93

Source: 2004 Management Announcement from the Technology Industrial Park, Industrial Development Bureau, Ministry of Economic Affairs (IDBMOEA) website:http://idbpark.moeaidb.gov.tw

such as bureaus for tariff duties to handle import–export affairs, logistic centers for applying to Customs, or one-stop services that offer employment information. In addition, the authorities also provide preferential loans, tax relief, and long-term financing, among others. The only drawback is that these technology extension services do not achieve economies of scale and thus result in high administration costs.

6.5.2. Networks and alliances

An ‘‘edging out’’ effect prevails among the southern clusters. They compete for firms, resources, government grants and subsidies, national major construction projects such as international airports, harbors, high-speed railways, and national research institutes. A kind of selfish regionalism embedded in each cluster results in resource dispersion or overlapping. Therefore, some cluster members work together cooperatively in joint ventures, production, marketing, training, or problem solving.

6.6. Macro environment

In an illustration of the ‘‘global village’’ effect, trends and changes in the international environment affect changes in other nations, i.e. a slight movement in one part of the international environment may affect the historical context, political networking, industrial structure, social relationships, or government technology policy. Given the possibilities of this trend, every country tries to strengthen and develop its industries, strengthen their comparative advantages, and encourage emerging industries to develop innovative capacity. Moreover, trends toward regional integration are gradually developing, as seen in the North American Free Trade Agreement (NAFTA), the Free Trade Area of the Americas (FTAA), the European Union (EU), and the Association of South East Asian Nations (ASEAN).

As a result of these international trends, global competition becomes more intense, product life cycles may be shortened, the protection of intellectual property rights is more carefully monitored, marketing distribution channels are broadened, and the management of supply chains becomes more important than ever. When these phenomena are followed with the development of information technology and the Internet, the effects will be greatly increased.

6.7. Social capital

Social capital transports information and accelerates imitation inside a cluster. Unlike in HSP, there have rarely been conspicuous alumni relationships among cluster members in southern Taiwan. There is also a geographic barrier for emerging clusters that are typically located in a rural region and do not have access to good transportation (basic infrastructure is under construction in STSP at the moment). Therefore, the social networks appear as weak links regardless of the scale and degree of activity among local businesses and civic associations in the region, the frequency of interaction, and the informal networks of personal business and related contacts.

6.7.1. External connections

Clusters that focus exclusively on internal linkages cut themselves off from sources of new knowledge and technology. Only a few large firms in each cluster in southern Taiwan

have joint ventures (albeit fragmentary), contracts, alliances with firms in other regions/ countries, or elements of contact and communication with experts in other regions. The knowledge of international benchmarking practices is limited. Without access to such practices, clusters are limited to learning only within their borders and have difficulty achieving competitive positions.

6.7.2. Shared vision and leadership

Most firms in the region are small, which means they face unique problems related to lack of specialist capacity, bounded vision (e.g. a lack of awareness of innovations due to low resources and knowledge bases, and limited expertise) and often strong locational dependency. These characteristics affect their approach to innovation and quite possibly their attitudes toward external sources of technological support. Although most leaders in these clusters take responsibility for collective competitiveness, they appear to be satisfied with current achievements instead of looking outward for additional competitive possibilities.

6.8. Summary

To summarize, the barriers associated with the five sub-systems in southern regional clusters are as follows.



High-tech and R&D professional talent is inadequate; the clusters lack people with

basic skills, industry knowledge and industriousness, and consistent work habits.



Educational achievement is relatively low.



Upgrading the pace of traditional industry is slow; emerging industries are still being

incubated.



Small firm-based clusters are less aware of the new production methods and struggle to

cope with a highly concentrated market.



Few leading firms that are part of global networks and markets.



Venture capital unwilling to invest in emerging or mature industry clusters.



Government is unable to loosen its rigid policy of cross-strait navigation, which further

affects international trade.



The foreign labor policy remains inflexible.



Many rural regional clusters, which increases travel time; a workforce with limited

potential, which indirectly cuts-off learning and affects the ability to develop new ideas.



Inadequate infrastructure (water, electricity, transportation, plant). Older industry

clusters have limited land and outdated infrastructure; emerging industry clusters have inconvenient transportation and poorer quality of life.



The effects of ‘‘edging-out’’ in which each industrial cluster competes for firms,

resources, government grants and subsidies, national major construction projects, national research institutes to invest in its own cluster and the expense of cooperation with other clusters to mutual benefit.



Weak linkages between the scale and degree of activity among local business and civic

associations in the region; less frequent interactions, informal networks of personal business-related contacts.

7. Action strategies for less-advantaged industrial clusters

Cluster strategies are driven by employers’ needs, not by the economic interests of

particular individuals or places [24]. Many suggestions for less-advantaged industrial

clusters begin with ways to raise skill levels and make the right connections, including learning from benchmarking clusters, and readjusting or reconfiguring successful factor strategies to match the cluster development efforts.

To contribute to these issues, a number of actions are suggested, to promote dynamic network innovation policies in southern Taiwan.

7.1. Quality of employment

It is a common occurrence in the industrial clusters in southern Taiwan that they are unable to recruit a workforce that contains the quality and quantity of employees needed. Mature or emerging clusters face shortages of skilled labor for upgrading or transition. Therefore, it is not surprising that the workforce is often the most pressing problem facing a cluster and its first choice for collective action.

The first action strategy is to strengthen the educational support system and its connection to the southern regions. What a high-technology industry needs is an educational support system of continually upgraded human quality. Expanding the numbers of higher education institutes, such as departments for natural sciences and applied sciences and technology, and programs that offer Masters and Ph.D. degrees, is a first step to developing a high-technology industry. The aim should be to integrate the universities in the southern region in series–parallel connections to provide the teaching staff, courses, and skills that industries need. New students who fail to qualify for a career program can strengthen their professional skills through pre-training or pre-employment. With these connections, poorly skilled or less-experienced employees can still enroll in courses in the universities that allow them to earn qualifications or take part in a cooperative project to enhance their opportunities for pre-training. Pre-employment is a kind of part-time job, which gives students opportunities to experience a practical career life during summer vacation.

7.2. Establish incentive mechanisms

In general, the authority in each regional industrial cluster receives support by actively participating in the policy of introducing a new ‘‘Industrial Upgrading Statute.’’ On the taxation side, the ‘‘Statute for Encouragement of Investment’’ stipulates that firms undertaking R&D expenditures can enjoy a discount on their land tax and other levies. On the financial side, the authority provides marginal loans for new equipment and other aspects of technology transfer, educational training, and promoting the development of the industrial structure. However, the policy structure is still limited for applicants. We suggest broadening the scope and number of incentives in the future. Besides, the Taiwan-style employee profit-sharing and stock-ownership system (TSE-PSSO) is another incentive for encouraging employees.

7.3. Facilitate business networks and skills alliances

Networks within clusters provide firms with external economies. Business networks provide members with economies of scale by sharing costs, experiences, and risks.

Networks have proved to be effective in aiding less-advantaged regions and small companies, and have proved useful for spurring training among SMEs that have been reluctant or unable to invest in upgrading the skills of their incumbent workforce.

Soft networks (i.e. associations) are of two types: trade associations for the firms themselves, and entrepreneurial manager associations for individuals. Regional learning connections can be built via formal trade associations to raise the capacity for innovation. With regard to tacit knowledge or know-how, it can be obtained through informal personal relationships, entrepreneurial managers’ associations, or in a similar way. Soft networks promote joint production, marketing, technology cooperation, or product development networks among their members, thereby generating sales and creating jobs. 7.4. Pursue niches

Finding an effective leverage for a regional cluster is crucial. Moreover, realizing twice the result with half the effort is even more important for a less-advantaged cluster. By using SWOT analysis, a less-advantaged cluster can find its own niche(s), and then work to combine its regional assets (perhaps expertise or local resources) to create a competitive edge and to drive industry upgrading.

The same goes for firms. They should identify and exploit their core competencies and flexibilities in the wider structure of the international division of labor.

Finally, individuals can find their own niche in the clusters, perhaps something they are clever at and can do that for the cluster—thus building a personal technology/skills portfolio of developed efforts, qualified credentials, and work experiences that relate to that niche.

7.5. Cultivate innovation capacity

In the policy of creating a supportive environment for raising ‘innovation capacity’, the main policy focuses fall on improving R&D capacity, cultivating technological human resources, strengthening cooperation with international companies, building the infra-structure to promote rapid diffusion of technologies, and stimulating both supply of and demand for new knowledge. Creativity comes from the cultivation of different forms of art, such as culture, art, music, painting, fine art, etc. Except for professional skills, firms should provide and support non-R&D related innovation. Similarly, to strengthen the individual’s artistic appreciation and the capacity for self-discipline are crucial as well. 7.6. Build a ‘‘Center for Technology Development’’ in southern region

A ‘‘Center for Technology Development’’ is similar to a technology R&D center. Based on a mode of international division of labor, the advanced country is devoted to radical innovation; Taiwan focuses on the refinement/redesign of this radical innovation through technology transfer; finally Taiwan transfers this advanced technology to countries with lower operating costs.

As far as benefiting regional integration, the Center should include the R&D center integrated with a technology center and a technology exploiting company. The integrated R&D center would be drawn from government, universities, and corporations, the

technology centers or technology exploiting companies would be sponsored by private enterprise.

The main tasks of such a Center would be to: (a) promote the agglomeration of technology exploiting companies; (b) form technology policy; (c) coordinate cooperative technology innovation between universities and industry, and spread research results, emerging technologies, technology standards, and product standardization; (d) assist with research and development of manufacturing techniques, management efficiency, market information, and automation; and (e) subsidize applied research and high-risk projects. In such a structure, the local government in each regional cluster would integrate all the R&D resources, establish a mechanism for cooperation, and build a technology platform for R&D. Such a platform would provide information and consulting on technology, or perhaps even encourage strategic alliances among industries/firms.

7.7. Create a trial market for innovative products and processes

A trial market is an experimental market that may lie offshore or a special area for new products and technologies that are unable to build sales performance and market rewards in the initial stage. Through this trial market, new products and technologies can be tested for their degree of product acceptability and market recognition, then go back for further re-design of the prototype or re-trial the technology, as well as developing marketing strategies. If the new product or technology sells well in the offshore area, then it could be expanded to other areas and markets.

7.8. Combine regional assets with new technologies to encourage venture capital investment Combining regional assets with emerging technologies to create and innovate new products and technologies or new firms and industries might very well stimulate venture capital investment in the regional cluster. The spin-off and start-up firms or emerging industries in which VC would invest could be agricultural biotechnology, ocean fishery technology, nano-chemical industry, and so forth. Therefore, raising risk tolerance, especially for angel investors, should be taken into consideration.

Authorities in southern Taiwan should ensure that access is open, diversified, and available, while also providing needed funding and giving support to extension-type services. These would include supporting delivery services (technology, business assistance, capital, and training), developing tech-savvy law and accounting firms, setting up venture capital funds, preferential loans, tax relief, long-term financing, etc. Once the spin-off/ startup firms emerge, authorities can also give angel investors an excellent opportunity to invest in sunrise industries.

7.9. Loosen related policy and regulations

Although government relaxed some of the limitations on mainland Chinese professionals or foreign technology professionals coming to Taiwan for high-tech research, there are still many limitations. According to current law on foreign labor, the central government controls most of the decisionmaking, regardless of wages or number of employees. Power for deciding foreign labor policy should be handed instead to the local authority.

For academic degrees such as Masters and Doctorates, science and engineering students who are on compulsory active duty in the Army could be allowed to directly enter industry as a substitute for service in the Army. However, the quota distributed to industries is limited and concentrates largely on high-tech industry. Instead it should also be opened to other industries, especially traditional manufacturing and service industries.

7.10. Loosen relations between the two sides of the Strait

Most firms and industries agree that the government should be open in its cross-straits economic and trade policies instead of habitually forbidding Taiwan industry, for political reasons, to step westward to China. Many Taiwanese manufacturers in labor-intensive industries have set up factories on the mainland because of the availability of cheap labor and low overhead costs there. Clusters in southern Taiwan should be allowed to expand their boundary in order to take advantage of more opportunities in mainland China. In other words, clusters based on an international division of labor should be able to go into other geographic areas and exploit markets there and reduce overhead costs wherever possible.

The same applies to firms in the less-advantaged clusters. Firms and SMEs should try to connect with other clusters/micro-clusters in mainland China to determine if there are possibilities for cooperative enterprises. Similarly for individuals, they should be allowed to extend their informal or exclusive connections in order to find other potential opportunities to enrich their life, experience, and knowledge.

7.11. Co-opetition replacing a zero-sum effect

To make a comprehensive survey, most industrial clusters are given similar preferential measures in terms of tax incentives, low rents on land and factory, infrastructure, facilities, and so forth. However, the ‘‘edging out’’ effect described earlier exists among some clusters, especially in the southern industrial cluster. To avoid this occurrence, authorities in southern Taiwan should organize a planning mode based on the cluster division of labor and strive for more markets and opportunities. Each cluster has its own development focus in terms of its advantages or strengths. They would draw on each other’s strengths to offset their own or others’ weaknesses, thereby generating cooperation and competition, instead of a zero-sum effect. Once the cake is bigger (by expanding the market), a few micro-enterprises or start-up firms can be spun-off or carved out. Under this strategy of ‘‘co-opetition,’’ whether among clusters, firms, or individuals, they can share norms and intelligence among firms and eventually become grounded in a competitive regional economy. Such an economy is in fact built on the bases of social interactions, linkages, and markets.

8. Policy implications and conclusions

As a whole, the southern clusters are less advantaged in terms of workforce, education, infrastructure, finance, knowledge, interactions, and personal networks. However, the authorities in southern Taiwan are beginning to recognize the need to raise their entire competence in order to establish an integral innovation environment based on their knowledge and capacity.

As for the clusters, they have the potential to change and upgrade because they are learning to recognize what their niches are, where to implement them to maximize benefits, and how to find leverage by means of previous successful or failed experiences. Gradually, they are learning to cope with the difficulties and global trends. I believe they will figure out how best to use the cluster framework to bring more people together and place them into successful regional innovation systems in order to implement strategies that accomplish their goals.

A subsidiary aim of this paper is to explore policy implications.

1. First, it is important for national and (more specifically) regional governments to be familiar with the particular needs of firms in the areas for which they are responsible. Regional policies should take into account the evidence concerning firms’ innovation needs, and the inadequacies or perceived unimportance of the existing technological infrastructure in the region. Public support must be directed to those aspects of the innovation process in which firms are actually involved.

2. The public sector should promote collaboration between firms and existing regional institutions such as colleges and universities. Central government, together with public sectors in the regions, may aid the transformation by means of a wide range of initiatives related to the development of knowledge-based structures and the acceleration of learning-by-doing. 3. By creating and developing physical infrastructures, such as telecommunications and

transport systems which firms rate as highly important, the public sector can reduce some of the drawbacks of being located in a less-advantaged region, and enhance the advantages of the location. Thus, integration of regional infrastructures with national infrastructures must be an important part of a policy that stimulates sustainable economic growth in the region.

4. The education system must develop a flexible system that makes it possible to combine vocational and higher education to attract new students. More specifically, policies should address the current problems associated with attracting and retaining skilled labor. The ‘‘brain-drain’’ effect should be countered, especially in the rural region clusters.

5. Firms in these regions encounter higher barriers to gaining access to information, technology, and knowledge that are relevant for production and innovation. These barriers make it difficult for the firms to participate fully in technological development in the relevant markets because the functionalities of various networks and channels are divided. Local authorities should aggregate relevant information and publish it for all firms in the cluster, thus ensuring fair and open access that is widely available to all. Acknowledgements

The author would like to thank Nick von Tunzelmann (SPRU, Sussex University) for providing helpful advice, as well as the National Science Council of Taiwan for funding this project (NSC 92-2416-H-390-006).

References

[1] Porter M. Clusters and the new economics of competition. Harvard Bus Rev 1998.

[2] Bergman EM, Feser EJ. Industrial and regional clusters: concepts and comparative applications. West Virginia University. 1999. Available from: /http://www.rri.wvu.edu/WebBook/Bergman-Feser/contents. htmS.

[3] Whalley J, Hertog PD. Clusters, innovation and RTOs: a synthesis of the findings from the RISE cluster studies. One project of RISE, vol. 11. Glasgow and Utrecht, 2000.

[4] Rosenfeld SA. Bringing business clusters into the mainstream of economic development. Eur Plann Stud 1997;5(1):3–23.

[5] Becattini G. The Marshallian industrial district as a socio-economic notion. In: Pyke F, Becattini G, Sengenberger W, editors. Industrial districts and inter-firm co-operation in Italy. Geneva: International Institute for Labour Studies; 1990. p. 37–51.

[6] Maillat D, Lecoq B. New technologies and transformation of regional structures in Europe: the role of the milieu. Entrepreneurship Regional Dev 1992;4:1–20.

[7] Pyke F, Sengenberger W. Industrial districts and local economic regeneration. Geneva: International Institute for Labour Studies; 1992.

[8] Scott A. The role of large producers in industrial districts: a case study of high technology systems houses in southern California. Regional Stud 1992(26):265–75.

[9] Camano RP. The concept of innovative milieu and its relevance for public policies in European lagging regions. Regional Sci 1995(74):317–40.

[10] van Dijk MP. Flexible specialization, the new competition and industrial districts. Small Bus Econ 1995(7):15–27.

[11] Markesan A. Sticky places in slippery space: a typology of industrial districts’. Econ Geography 1996(72):293–313.

[12] Lawson C. Territorial clustering and high-technology innovation: from industrial districts to innovative milieux. Working Paper No. 54. ESRC Center for Business Research, University of Cambridge, 1997. [13] Rabellotti R. External economies and cooperation in industrial districts. Houndmills: Macmillan; 1997. [14] Doeringer PB, Terkla DG. Business strategy and cross-industry clusters. Econ Dev Q 1995(9):225–37. [15] Hart DA, Simmie J. Innovation, competition and the structure of local production networks. Local

Economy 1997: 235–46.

[16] Feser EJ. Introduction to regional industry cluster analysis. North Carolina: Chapel Hill; 2001. [17] OECD. Boosting innovation: the cluster approach. Paris, 1999.

[18] Porter M. The competitive advantage of nations. New York: Free Press; 1990.

[19] Bergman EM, Feser EJ. Industry clusters: a methodology and framework for economic co-operation and development. In: Roelandt P, denHertog TJA, editors. Boosting innovation: the cluster approach. Paris: OECD; 1999.

[20] Lee TL, von Tunzelmann N. A dynamic analytic approach to national innovation systems: the IC industry in Taiwan. Res Policy 2005;34(4):425–40.

[21] Rosenfeld SA. Creating smart systems: a guide to cluster strategies in less favoured regions. A report to the European Union-Regional Innovation Strategies, 2002. /http://www.rtsinc.org/publications.htmlS.

[22] Lee TL. The industrial analysis for fitting Kaohsiung future development. Project report. Kaohsiung, Taiwan, 2003.

[23] TVCA website./http://www.tvca.org.tw/S

[24] Rosenfeld SA. Just clusters: economic development strategies that reach more people and places—a synthesis of experiences. Regional Technologies Strategies, 2002.

Ting-Lin (Linda) Lee is currently an Assistant Professor in the Department of Asia-Pacific Industrial and Business Management, National University of Kaohsiung, Taiwan. She received her Ph.D. degree in Science and Technology Policy at PREST, University of Manchester, UK, and an MBA at Sun Yat-Sen University, Taiwan. She worked for more than 12 years in Taiwan central and local government, rising to section chief. Her current research focuses on science and technology policy, innovation and R&D management, industrial policy, and the industry practical application of system dynamics. A recent paper was published in Research Policy.