Innovative Cluster Organizations in Tanzania
A Minor Field Study evaluating cluster performance and actor collaborations within the clusters included in ISCP-Tz
IDA STADENBERG
Master of Science Thesis Stockholm, Sweden 2016
Innovative Cluster Organizations in Tanzania
A Minor Field Study evaluating cluster performance and actor collaborations within the clusters included in ISCP-Tz
Ida Stadenberg
Master of Science Thesis INDEK 2016:32 KTH Industrial Engineering and Management
SE-100 44 STOCKHOLM
This study has been carried out within the framework of the Minor Field Studies Scholarship Programme, MFS, which is funded by the Swedish International Development Cooperation Agency, Sida.
The MFS Scholarship Programme offers Swedish university students an opportunity to carry out two months’ field work, usually the student’s final degree project, in a country in Africa, Asia or Latin America. The results of the work are presented in an MFS report which is also the student’s Bachelor or Master of Science Thesis. Minor Field Studies are primarily conducted within subject areas of importance from a development perspective and in a country where Swedish international cooperation is ongoing.
The main purpose of the MFS Programme is to enhance Swedish university students’ knowledge and understanding of these countries and their problems and opportunities. MFS should provide the student with initial experience of conditions in such a country. The overall goals are to widen the Swedish human resources cadre for engagement in international development cooperation as well as to promote scientific exchange between unversities, research institutes and similar authorities as well as NGOs in developing countries and in Sweden.
The International Relations Office at KTH the Royal Institute of Technology, Stockholm, Sweden, administers the MFS Programme within engineering and applied natural sciences.
Erika Svensson Programme Officer
MFS Programme, KTH International Relations Office
KTH, SE-100 44 Stockholm. Phone: +46 8 790 6561. Fax: +46 8 790 8192. E-mail: erika2@kth.se www.kth.se/student/utlandsstudier/examensarbete/mfs
Master of Science Thesis INDEK 2016:32
Innovative Cluster Organizations in Tanzania
A Minor Field Study evaluating cluster performance and actor collaborations within the clusters included in ISCP-Tz
Ida Stadenberg
Approved
2016-07-01
Examiner
Anders Broström
Supervisor
Kristina Nyström
ABSTRACT
Cluster Organizations, as a means of promoting competition and innovation in industrial clusters, have become increasingly popular over the world. Cluster organizations aim to increase growth and competitiveness of clusters within a region, and have become a central part of economic policy-making across the world. Recently, the concept has been used to a larger extent as a tool for economic development and poverty alleviation. This thesis seeks to examine the cluster organizations that are part of the Sida funded program Innovation Systems and Cluster development in Tanzania (ISCP-Tz), by evaluating performance, goals and development of the program based on cluster facilitators perceptions, and assess linkages and actor collaborations between clustered actors. The data in this thesis is collected through a telephone-administered questionnaire, as well as interviews and visits to cluster sites. The results show a positive impact on cluster firms performance as assessed by cluster facilitators, but show that actor collaborations in many cases are inadequate and need to be improved.
Key-words
Economics, Development Economics, Economic development, Cluster, Cluster organization, Cluster Initiative, Agglomeration economics, Innovation, Tanzania, ISCP-Tz
ACKNOWLEDGEMENTS
I would like to thank my supervisor, Associate professor Kristina Nyström, for her valuable guidance and support during the thesis process. Your suggestions, comments and encouragement have been very appreciated.
I also want to express my deepest gratitude to COSTECH for giving me the opportunity to write this thesis, for introducing me to key persons and for valuable input during my time in Tanzania.
I would like to express a special thank you to Dr. Dugushilu Mafunda, Furaha Kabuje, Dan Nerén and Julieth Kweka, for your precious time spent in helping me during the process.
Additionally, I would like to thank the respondents of the survey for taking your time to provide valuable and useful information.
I would also like to thank Professor Ramon Wyss for introducing me to COSTECH and the Innovation Systems and Cluster development Program in Tanzania, and for providing me with guidance and valuable feedback. I would also like to thank Göran Lindqvist, Director of Research at Stockholm School of Economics, for excellently providing me with inspiration and knowledge during the early stage of the thesis process.
A special thank you to the Swedish International Development Cooperation Agency, Sida, for giving me the opportunity to carry out this study in Tanzania within the Minor Field Studies Scholarship Programme, MFS.
Lastly, I want to express my deepest gratitude to my sister Elin for your endless support in life.
TABLE OF CONTENTS
1. INTRODUCTION ... 1
1.1 Background ... 1
1.1.1 Clusters ... 1
1.1.2 Sida’s support of cluster organizations in Tanzania ... 1
1.1.3 Tanzania National Innovation System ... 2
1.2 Previous research and my contribution ... 3
1.3 Purpose ... 3
1.4 Limitations and sustainability implications ... 4
1.5 Outline of the thesis ... 4
2. THEORETICAL FRAMEWORK ... 5
2.1 Cluster theory ... 5
2.1.1 Benefits of agglomeration ... 5
2.1.2 The role of competition ... 6
2.2 The cluster life cycle ... 7
2.3 Cluster organizations – the gap model ... 7
3. LITERATURE REVIEW ... 10
4. CLUSTER ORGANIZATIONS IN ISCP-TZ ... 12
5. METHODOLOGY ... 14
5.1 Research design ... 14
5.1.1 Data collection ... 14
5.1.2 Questionnaire ... 14
5.1.3 Translation of questionnaire ... 15
5.1.4 Interviews and cluster visits ... 15
5.2 Limitations and validity of the study ... 16
5.2.1 Validity and reliability ... 16
5.2.2 Challenges with cluster policy evaluations ... 16
6. EMPIRICAL ANALYSIS ... 18
6.1 Results ... 18
6.2 Analysis ... 26
6.2.1 The gap between firms ... 26
6.2.2 The government gap ... 27
6.2.3 The capital gap ... 29
6.2.4 The academia gap ... 29
6.2.5 The gap between clusters ... 30
6.2.6 Economic impact ... 30
7. CONCLUSIONS AND SUGGESTIONS FOR FUTURE RESEARCH ... 32
8. REFERENCES ... 34
APPENDIX 1. TANZANIA ... 37
APPENDIX 2. QUESTIONNAIRE ENGLISH ... 38
APPENDIX 3. QUESTIONNAIRE SWAHILI ... 45
1. INTRODUCTION
1.1 Background 1.1.1 Clusters
Cluster organizations (COs) as a means of promoting competition and innovation in industrial clusters, have become increasingly popular all over the world. Recently, it is also used to a larger extent as a tool for economic development and poverty alleviation (UNIDO, 2010). COs are organized attempts to increase growth and competitiveness of clusters within a region, and have now become a central part of economic policy-making across the world. Although COs are used and adopted in most parts of the world today, they started out in the developed world, and have increased in popularity in developing countries after year 2000. International organizations like the UN and the World Bank are using clusters as an economic development tool, which has resulted in many donor-initiated COs (Ketels, Lindqvist and Sölvell 2006).
This thesis uses Porter’s definition of clusters as “a geographically proximate group of interconnected companies and associated institutions in a particular field, linked by commonalities and complementarities” (Porter 2008: 215). A Cluster initiative (organization) is defined as “organized efforts to increase growth and competitiveness of clusters within a region, involving cluster firms, government and/or the research community.” (Ketels, Lindqvist and Sölvell, 2003: 9). The terms cluster initiatives and cluster organizations will be used interchangeably throughout the thesis.
1.1.2 Sida’s support of cluster organizations in Tanzania
Tanzania is a low-income country in Sub-Saharan Africa, GDP per capita was $955.11 2014 according to The World Bank (2015a). Despite the low GDP, the overall macroeconomic performance in the country is strong and Tanzania has enjoyed stable economic growth rates recent years. There are many challenges; a large part of the population lives below the poverty line and inequality increases as GDP increases, leaving many people behind (The World Bank, 2015b)2.
Sweden and Tanzania have a long history of development cooperation, over 50 years, with the aim to reduce Tanzania’s aid dependency. The support goes through the Swedish International Development Cooperation Agency (Sida) and has covered many areas over the years. According to the recent strategy adopted in 2013, the current focus areas for the cooperation with Tanzania are: jobs and development of energy and agricultural markets, improved education and increased entrepreneurship, strengthened democratic accountability and transparency, increased awareness of human rights (Sida, 2015).
Sweden has, through Sida, supported cluster organizations in Tanzania since 2005 through a program called Innovation Systems and Cluster development Program Tanzania (ISCP-Tz).
ISCP-Tz is part of Innovation Systems and Cluster development Program East Africa (ISCP-
1As a comparison, GDP per capita in Sweden 2014 was $58,898.9
2 See Appendix 1 for general data about Tanzania
EA), which includes Tanzania, Uganda and Mozambique. The idea of the program came up when representatives from the participating countries attended the 6th Global Conference on Innovative Clusters in Gothenburg, Sweden, 2003, a conference organized by VINNOVA (the Swedish Innovation Agency) and TCI (the Competitiveness Institute). The support went through the University of Dar es Salaam (UDSM) in the beginning, but was changed to the Tanzania Commission for Science and Technology (COSTECH)3 in 2011, thereby moving the coordination to a national level. The project started with the support of eight clusters in 2006, and has now grown to 67 clusters in 2015. The innovative cluster organizations aims to contribute to regional and national economic growth, and to strengthen the business environment and make a positive socio-economic impact. The overall aim with the cluster-based initiative is to strengthen the linkages between the cluster actors through collaborative activities, to enhance knowledge transfer and thereby improve innovation, value addition and competitiveness. This in turn will contribute to poverty alleviation, strengthening of local agricultural products and natural resources, preservation of the environment and improved gender equality. The COs combine university-industry-government relationships in a triple helix model4; in that way the clusters will lead to increased productivity, higher quality of products and services and generate employment opportunities (Rath et al, 2012; Rydhagen and Trojer, 2014).
1.1.3 Tanzania National Innovation System
Tanzania formulated a long term National Development Vision 2025 in the year 2000. The vision aims at transforming Tanzania to a middle-income country by 2025, with a focus on five attributes: good governance, high quality livelihoods, peace, stability and unity, a well educated and learning society and a competitive economy capable of producing sustainable growth and shared benefits (United Republic of Tanzania, 2000). To be able to reach the goals, Tanzania is working according to five-year development plans (FYDP). The first one came into effect 2011 and focuses on strengthening the country’s infrastructure; roads, port, energy, and information and communication technology. The second one from 2016 highlights the importance of developing the industrial sector, and the third five-year development plan from 2021 will focus on making manufacturing and service sectors more competitive. The three five-year plans build upon each other, and the success of the former is crucial for the implementation of the latter.
Overall, the plans aim to increase the productivity of the agricultural sector and transform the structure of the economy from a mainly agrarian to a mixed economy. Before the five-year plans came into force, Tanzania was working according to National Strategy for Growth and Reduction of Poverty, known in Tanzania as MKUKUTA (Mkakati wa Kukuza Uchumi na Kupunguza Umaskini Tanzania), and MKUZA (Mkakati wa Kukuza Uchumi Zanzibar) in
3 COSTECH is a parastatal organization with the responsibility to coordinate and promote research and technology development within Tanzania. The organization is an advisory organ to the Government concerning all matters related to science and technology, for example policy formulation, allocation of resources and to facilitate national, regional and international cooperation in scientific research and technology development.
4 A triple helix model is a model that enhances the importance of collaboration between universities, governments and industries. The model stresses the role of universities in innovation, and emphasizes that innovation and economic development comes from interaction between the three elements.
Zanzibar. MKUKUTA aimed at reaching sustained high economic growth and alleviate poverty, but after realizing that poverty remained high despite the country’s high rates of economic growth, the FYDPs were implemented (Economic and Social Research Foundation, 2015).
A review of the National Innovation System (NSI) by the Ministry of Communication, Science and Technology (2012) concludes that one of the weaknesses of the NSI in Tanzania is the lack of partnership and collaborations between academia, industry and the government. Improved collaborations between R&D institutions, the industry and the government are important for establishing IPR systems, enhancing innovations and commercialization of indigenous technologies. The same applies to R&D activities, which are stated to be supply- rather than demand driven. This means that the connection between research institutions and the private sector needs to be improved so that immediate needs are satisfied and urgent innovation opportunities utilized. The research results need to be converted to services, products and processes to be able to enhance the business environment and contribute to socio-economic development. Further, the report recommends a system for protecting and commercializing local resources and knowledge, combined with a national IPR framework (Ministry of Communication, Science and Technology, 2012).
1.2 Previous research and my contribution
Most of the research and literature concerning clusters and cluster organizations are made in a developed country context, and are therefore not always applicable in a developing country.
Most COs in developed countries are initiated through industries or governments, whereas in developing countries many COs are donor-initiated. Donor-initiated COs usually take place where government support for clusters and competitiveness is low, and where the level of trust is lower; hence they usually operate in a different environment. Donor-initiated COs tend to focus more on “basic” industries that are not yet well developed, and company- or government initiated COs in developed countries are more focused on enhancing innovative capacity in already developed high-tech clusters. Additionally, previous research show that COs in developing countries are less likely to have quantified targets and goals (Ketels, Lindqvist and Sölvell, 2006).
This thesis contributes to the research of cluster initiatives in developing regions by focusing on the clusters included in the Innovation Systems and Cluster development Program in Tanzania.
The thesis shows how the cluster initiatives work and collaborate, and provides an assessment of the general performance of clusters in Tanzania.
1.3 Purpose
The main purpose with the thesis is to assess the economic impact of the cluster organizations included in the ISCP-Tz and evaluate bridge building between the clustered actors, thereby contribute to the research of clusters in developing regions. Further, my research questions are:
• What are cluster organizations perceptions of cluster performance, goals and development?
Theoretical framework Literature
review CO's in ISCP-Tz Methodology Empirical
analysis Conclusions
• How has the linkages and actor collaborations within the clusters evolved since the implementation of the ISCP-Tz?
1.4 Limitations and sustainability implications
Difficulties with data collection of economic indicators of clustered firms limit the thesis in terms of the programs’ economic impact on firms. A majority of the firms within the clusters are informal and do not keep track on their financial performance. Therefore, data on performance is based on cluster facilitators’ assessments. Further, the cluster facilitators’ assessments serve as an indication of the economic performance, but it needs to be stressed that they are based on their subjective view.
The thesis aims to examine collaborations and linkages among the clusters in general, on an aggregate level, and do not provide deep insights into each cluster. Since the focus is on Tanzania, the results cannot be generalized to a larger population than the clusters included in the sample.
Sustainability aspects in this thesis concern economic- and social sustainable development. The aims with the Innovation Systems and Cluster development Program in Tanzania is to make a positive socio-economic impact that contributes to poverty alleviation, through strengthening the business environment. Hence the economic and social development goes hand in hand in this matter. The thesis will contribute to economic and social sustainable development through evaluating the impact of the cluster program.
1.5 Outline of the thesis
Part two provides a theoretical framework with cluster theory as a foundation, followed by a theory of cluster life cycles and the role of cluster organizations. Part three focuses on previous studies conducted on clusters and cluster initiatives, part four describes the cluster program in Tanzania, and part five describes the methods and research design used for data collection. Part six presents the results from data collection followed by an analysis, and finally part seven provides a conclusion and suggestions for future research.
2. THEORETICAL FRAMEWORK
2.1 Cluster theory
2.1.1 Benefits of agglomeration
“A cluster is a geographically proximate group of interconnected companies and associated institutions in a particular field, linked by commonalities and complementarities” (Porter 2008, 215). Cluster theory states that firms that are located in clusters benefit from agglomeration economies, such as; scale economies, external benefits of production, spillover effects and access to specialized labor. The reason for this is their proximity to each other, both geographically and by activities. Benefits also stem from collaborations between the business community, government, and supporting institutions like research organizations and financial supporting institutions, which creates value through joint interests (The World Bank, 2009).
Cluster theory stems from theories of agglomeration and localization economics. One of the first contributors to agglomeration economics was Alfred Marshall (1890), who analyzed the reason why firms locate in close proximity to each other. Marshall stated that the main reason for firms in related industries to cluster within the same area is the physical conditions available there.
This means that it did not have as much to with agglomeration externalities caused by firms, but rather exogenous factors like natural resources that draw firms to the same place (McCann and Folta, 2008). Further, Marshall highlighted three benefits of agglomeration, related to reduced transport costs of firms when it comes to people, ideas and goods. The first one, about people, relates to labor market pooling. Clustered firms can benefit from a large pool of specialized labor, facilitate matching between labor and firms and maximize productivity. The second, about ideas, concerns knowledge transfer between clustered firms, such as workers exchanging tacit knowledge. The third, goods, is related to the proximity to supporting industries, like complementary products or downstream suppliers, which reduces transaction costs (Ellison, Glaeser and Kerr, 2010).
Even though the benefits of agglomeration dates back to Marshall in the 1890s, the larger strand of the literature about agglomeration economies and evaluations of clusters received increased attention with Michael Porter in the 1990s. According to Porter (2008), extensive literature about clusters and location economics were written in the first fifty years of the 1900s, then moving out of the mainstream economics for some time, to return again in the 1990s. One of the reasons for this might be that past localization theories, and arguments promoting agglomeration were developed within a different industrial landscape, and that globalization changed the pillars that these theories were based upon. Cluster theory today is adjusted to globalization and the dynamic economic landscape in which the firms operate (Porter, 2008).
Clusters appear in all kinds of industries and economies, both within basic and high-tech industries. They also vary in size; both the number of firms and the size of existing firms. The essence of cluster benefits lies in the linkages between its members, both within the vertical and horizontal chain and when it comes to supporting functions. Supporting firms can include infrastructure providers, training institutions, firms that produce complementary products, education, research, technical support etc. (Porter, 2008). According to Porter (2008), the
location of firms affects their competitive advantages, which in turn affects productivity and productivity growth. The description above stating that the essence of a cluster lies in its linkages between its members fits well when it comes to competition as “the productivity of a location rests not on the industries in which its firms compete, but on how they compete” (Porter, 2008:
225). A high technological firm in a location that lacks high technological infrastructure, transportation, well-educated workers etc. will hence not generate the productivity and innovations it otherwise could.
Being located in a cluster can increase access to specialized inputs and reduce costs. When a firm outsource to another firm located in the same cluster, local outsourcing, costs such as transportation costs, transaction costs etc. can be reduced. It also reduces the risk of moral hazard and other opportunistic behavior, since the contracting firms are members of the same cluster.
Additionally, local outsourcing facilitates support services like repairs, installations and trainings. Just like increased access to inputs is valuable by firms within the cluster, so is increased access to complementarities by customers and workers. Potential customers visiting a cluster are able to visit the intended firm, but also firms offering complementarities located nearby. Firms can also take advantage of joint marketing. A large cluster could also influence government’s investments in public goods such as infrastructure, education or fairs (Porter, 2008).
2.1.2 The role of competition
In Porter’s work The Competitive Advantage of Nations (1998) he uses four interrelated forces to demonstrate the effect of location on competition, a model that is commonly referred to as Porter’s diamond. All four parts are important when describing the context in which a firm operate, but the part focusing on clusters is mainly the part of the model named related supplier or support industries, and even more importantly through the linkages between all four parts.
FIGURE 1. PORTERS DIAMOND
Source: Porter, M. (2008) On Competition. Harvard Business Press, p. 227.
Factor (input) conditions have an indirect effect on the productivity of firms. Factor conditions include infrastructure, human resources, information, the legal system, universities and research institutes, and their degree of efficiency matter for the firms within the cluster.
Firm strategy, structure and rivalry refer to the context in which the clustered firms operate.
Low rivalry generates low productivity, according to Porter (2008), since competition forces firms to innovate and be productive. The local rivalry is constituted of the rules and norms within the cluster, but the macroeconomic conditions, such as political stability, and the microeconomic policies, like the tax system, intellectual property rules and labor market policies, within a country also influences the level of competition within a cluster.
Demand conditions have direct impact on the productivity of firms, and the firms’ ability to compete by differentiating.
Related and supporting industries, and the linkages between them, are important features of a cluster and its benefits.
2.2 The cluster life cycle
The benefits of agglomeration seem to change with time as clusters change and decline, a process referred to as the life cycle of a cluster (Ketels and Memedovic 2008; Menzel and Fornahl 2009). The model by Menzel and Fornahl (2009) suggests that the development of a cluster through a cluster life cycle depends on the level of heterogeneity among the actors of a cluster, and thereby also by its size. Large clusters that consist of many firms has a larger potential to diversify in terms of technology and size, while small clusters need to be specialized for the actors to take advantage of each other. The size and heterogeneity of a cluster will evolve as it moves through the four different stages of the cluster life cycle. In the emerging stage, the cluster consists of a few firms with growing potential, the level of innovation is increasing and so does heterogeneity as more firms enter. In the growth stage, the cluster becomes more specialized and focused. According to the model, the level of heterogeneity reaches the highest peak between the emergence and growth stages. The next step, sustainment, shows a matured stage where the level of heterogeneity is decreasing and the cluster enters a sustainable path. For the cluster to be sustainable, it needs to maintain the level of renewal and heterogeneity to keep the innovation level and avoid decline. The renewal can occur by adapting new technology to the cluster or using existing knowledge within the cluster to adapt to new environments.
The benefits of maintaining heterogeneity within a cluster are connected to the linkages between the actors within it. The knowledge and experiences by the different actors have to accessible, which is made by strengthening actor collaborations (Menzel and Fornahl, 2009).
2.3 Cluster organizations – the gap model
Although the mechanisms behind cluster theory and improved relations between its members are clearly beneficial to firms, network failures seem to hinder clustered members to take full advantage of the external economies of scale. This is where cluster organizations come in. COs are used to strengthen the clusters and increase the competitiveness of the firms within them.
Figure 2 shows a cluster and the main actors that form a cluster; the government, research institutions, capital providers, education institutions and the most central actor – firms (Ketels, Lindqvist and Sölvell, 2013).
FIGURE 2. ACTORS WITHIN A CLUSTER
Source: Ketels C., Lindqvist G., Sölvell Ö. (2013) The Cluster Initiative Greenbook 2.0, Ivory Towers Publishers Stockholm, p. 37
If actors within the cluster would collaborate perfectly, they would be able to enjoy benefits of agglomeration and contribute to innovation and economic growth. However, there are barriers between the actors that prevent collaboration and thereby innovation, barriers that Ketels, Lindqvist and Sölvell (2013) refer to as gaps. These are illustrated in the gap model in figure 3.
According to Ketels, Lindqvist and Sölvell (2003) COs are “ (…) collaborative actions by groups of companies, research and educational institutions, government agencies and others, to improve the competitiveness of a specific cluster” (Ketels, Lindqvist and Sölvell, 2003).
FIGURE 3. THE GAP MODEL
Source: Ketels C., Lindqvist G., Sölvell Ö. (2013) The Cluster Initiative Greenbook 2.0, Ivory Towers Publishers Stockholm, p. 38
Ketels and Memedovic (2008) emphasize three approaches of cluster organizations. The first one focuses on creating a platform for interaction between the actors, so as to enable collaborations and knowledge transfer. The second highlights the importance of collaboration between the private and public sector, since it is important that the government is aware of what kind of investments and policies that affect the firms’ success. The third highlights the importance of
Research institutions
Education institutions Capital providers
Government
Firms
The gap between firms The research gap
The education gap The government
gap
The capital gap The global
market gap
The gap between clusters
collaborations and joint actions between the clustered firms, and the activities they perform together to reach their goals (Ketels and Memedovic, 2008). Joint actions can take place vertically or horizontally; where vertical cooperation refers to joint actions among firms at different stages along the supply chain, and horizontal cooperation means joint actions among competitors (McCormick, 1998).
3. LITERATURE REVIEW
Ketels, Lindqvist and Sölvell (2003, 2006, 2013) have made contributions to the cluster literature through their evaluations of cluster organizations across the world. The Cluster Initiative Greenbook (Ketels, Lindqvist and Sölvell, 2003) and the Cluster Initiative Greenbook 2.0 (Ketels, Lindqvist and Sölvell, 2013), conduct a Global Cluster Initiative Survey (GCIS) to perform a comprehensive evaluation of drivers and success factors of clusters across industries and countries. They evaluate objectives, performance, how the clusters operate and cluster policies. Their research includes survey responses from 233 and 356 cluster managers in 2003 and 2013, respectively. The largest part of their sample constitutes OECD countries. For example, only two African countries are included in sample of the GCIS performed 2003 and 2013, South Africa in the first and Tanzania in the latter.
A larger diversity of countries are included in the report Cluster Initiatives in Developing and Transition Economies (Ketels, Lindqvist and Sölvell, 2006), where the GCIS is performed on 450 cluster organizations across the world. The report compares cluster organizations in developed, transition and developing country contexts, so as to contribute to a better understanding of their differences and similarities. They find that the objectives, activities and performances of the clusters differ according to the context. For example, they find that COs in developing countries to a larger extent are donor-initiated and focused on agriculture and basic industries, while the COs in advanced economies typically are government-initiated and characterized by high technology. The survey conducted in their report denotes the importance of contextualization (Ketels, Lindqvist and Sölvell, 2006).
Klofsten (2009; Klofsten et al, 2015) derives general success factors of clusters, based on the Business Platform Model (Klofsten, 1992), and case studies of five Swedish clusters. His findings are based on interviews with the facilitator of each cluster, which result in five general factors that contribute to success of a cluster: Idea, Activities, Critical mass, Commitment and driving forces and Organization. The study does not evaluate the clusters but gives examples of ways to operationalize the success factors into interview questions for data collection (Klofsten, 2009).
Previous literature brings up a number of obstacles that can hinder the clustered actors to benefit from agglomeration economies, obstacles more common in developing regions (UNIDO, 2010;
McCormick, 1998). One reason is that small-scale firms are more prone to prioritize short term interests, meaning that they might reject long term benefits because of high short term costs or investments (UNIDO, 2010). In addition, transaction costs are usually high, especially in regions with low levels of trust. Low levels of trust hinder collaborations and innovation since it makes firms reluctant of sharing information. Developing regions generally have weak institutions, which implies that it is difficult to enforce sanctions in case of opportunistic behavior (UNIDO, 2010; McCormick, 1998). The weak institutions combined with low levels of trust hinder the development of partnerships in the long run. However, if trust levels would be higher, there is a possibility for it to balance the negative effects of weak institutions, by establishing a reputation mechanism that work as a legal contract (UNIDO, 2010). A common socio-cultural identity within the cluster can provide a basis for trust and hinder opportunistic behavior (Schmitz,
1995). Further, the paper by UNIDO (2010) states that relationships between local governments and the private sector are weak and that financial institutions often are designed for large-scale firms and hence is unable to deliver services to small-scale enterprises (UNIDO, 2010).
Sonobe, Akoten and Otsuka (2011) study the relationship between years of schooling on enterprise size and growth, in an informal metalworking cluster in Nairobi. Their findings show that highly educated entrepreneurs are more likely to innovate to increase profitability of the firm. An interesting aspect about their data is that they use the variable number of workers as enterprise size and growth, and do not take revenue, value-added or similar measures into account. The paper states that there seem to be difficult gathering data on such measures in enterprise surveys in Sub-Saharan Africa in general, especially in the informal sector.
Porter’s definition5 of clusters has been criticized for being broad and unclear, and one of the reasons is that he uses the term ‘geographical proximity’ without defining a clusters boundaries.
This has made previous literature question the importance of geographical proximity, since the definition does not include any limits on how large a cluster can be. The difficulties in delimiting the boundaries of a cluster taken together with the fact that outcomes are hard to measure, makes evaluation of clusters difficult (Swords, 2013). This is discussed in an article by Martin and Sunley (2003), which states that the reason for the popularity of clusters lies in the incompleteness of its definition. The authors mean that the fact that the cluster concept is rather vague has saved it from being tested and evaluated, like models and theories usually are. The concept of clusters seems to be universally accepted even though there is a lack of empirical evidence for its benefits. Further, they argue that in cases where clustered firms have shown economic growth, one should be careful to interpret it as causality, as there are many possible influences that contribute to a firms’ success than its location relative to other firms (Martin and Sunley, 2003). Aziz and Norhashim (2008) argue that cluster analysis lacks a holistic framework that includes all actors within the cluster. The frameworks used in analysis usually concentrate on the development of firms, while ignoring other clustered actors and their progress (Aziz and Norhashim, 2008). The involvement of many different actors in a cluster aggravate efforts to set objectives, since expectations and goals probably differ between firms, local politicians, coordinators etc. Fromhold-Eisebith and Eisebith (2008) suggest that cluster evaluation should divide its focus in objective and subjective goals, where the former evaluates indicators commonly associated with clusters that allow for comparison, while the latter evaluates subjective outcomes for the different actors. However, they too, emphasize the problems of comparisons of cluster developments, and highlight regional influences as the main obstacle.
5 “A cluster is a geographically proximate group of interconnected companies and associated institutions in a particular field, linked by commonalities and complementarities” (Porter 2008, 215)
4. CLUSTER ORGANIZATIONS IN ISCP-TZ
The cluster organizations included in ISCP-Tz operate in diverse industries, mainly focused on basic industries like agriculture, food and basic manufacturing, while a few of them are capital- intensive. This is in line with previous literature (Ketels, Lindqvist and Sölvell, 2006) that emphasizes that developing countries, where donor-initiated cluster organizations dominate, to a larger extent than advanced and transition economies are focused on basic industries. In contrast, advanced economies typically focus on high tech industries. The clusters are geographically distributed over the whole country, but a majority of them are situated in the regions; Dar es Salaam, Morogoro or Zanzibar. The cluster organizations in Tanzania are presented according to industry group in Table 1.
TABLE 1. INDUSTRY GROUPS
Industry group Cluster initiative Region
Agriculture, food, basic manufacturing
Vegetable Seed CI Oil Seed CI
Arusha Babati
Beekeeping CI Bukombe
Nutraceuticals CI Dar es Salaam
Eastern Regions Mushroom CI Dar es Salaam
Handloom CI Dar es Salaam
Textile CI Dar es Salaam
Furniture CI Dar es Salaam
Tomato CI Iringa
Cassava Processing CI Kibaha
Soap CI Kigoma
Cassava CI Kigoma
Sisal CI Kishapu
Small Scale Sisal farming CI Korogwe
Rice Processing CI Magugu
Bee Keeping CI Manyoni
Food Processors CI Morogoro
Rice Processors CI Morogoro
Furniture CI Morogoro
Meat CI Morogoro
Poultry Keeping CI Morogoro
Textile CI Morogoro
Sunflower CI Mpwapwa
Fish Farming CI Mwanga
Soap CI Mwanza
Livestock Keepers CI Nkasi
Mushroom CI Ruvuma
Sunflower CI Singida
Fish Farming CI Ugweno
Sea Weed CI Zanzibar
Fruit, Vegetables and Spice Processing Zanzibar Unguja
Poultry CI Zanzibar Unguja
Soap CI Zanzibar Unguja
Fruit, Vegetables and Spice Processing Zanzibar Pemba Fish Farming and Processing CI Zanzibar Pemba
Honey CI Zanzibar Pemba
Capital intensive manufacturing Metal works CI Small scale mining CI
Dar es Salaam Kilindi Engineering and Metal works CI Morogoro
Milling CI Morogoro
Oil Processing CI Morogoro
Engineering CI Shinyanga
"High tech", advanced services Bio Fuels CI ICT CI
Dar es Salaam Dar es Salaam
Tourism Cultural Heritage Tourism CI Bagamoyo
Mwenge Wood Carving CI Tourism CI
Cultural Heritage Tourism CI Cultural Heritage CI
Dar es Salaam Morogoro Tanga
Zanzibar Unguja
Other Educational Services CI Dar es Salaam
The clusters included in ISCP-Tz are natural agglomerations, so called organic clusters that existed prior to the program. The program has then built up and strengthened cluster organizations so as to enhance innovation and competitiveness of the cluster. ISCP-Tz has supported the cluster organizations through capacity building to cluster facilitators, business support, technology development, support with linking the clusters to markets and financial institutions, and also through providing entrepreneurship- and business skills. A cluster facilitator is the person who is responsible for the development of the cluster organization. The facilitators are appointed by COSTECH, and earlier by UDSM, and are persons with various backgrounds and functions. So as to prepare them for the role as a facilitator, they have been provided trainings, like leadership training, entrepreneurship and management training. The cluster facilitators work voluntarily and are hence not paid. The business support has focused on formalizing firms through product certifications, registration of business name and adapting to certain standards. This has been done through connecting the cluster organizations to Tanzania Bureau of Standards (TBS) and Tanzania Food and Drugs Authority (TFDA) etc. The cluster organizations have been exposed to technology development centers and research institutions to facilitate technology development. When it comes to finance, the cluster organizations have been provided with seed funds, to support the establishment of the cluster and activities to get started.
Additionally, the program has established links with banks to facilitate access to finance for clustered firms (COSTECH, 2016).
TABLE 2. EXAMPLES OF CLUSTER INITIATIVES Zanzibar Seaweed
The Seaweed cluster consists of Seaweed farmers, exporters, buyers, academic actors represented by the University of Dar es Salaam (UDSM) and Institute of Marine Sciences and government actors like the Department of Fisheries and the Department of Agriculture. The cluster exports dry seaweed to international markets, and has also engaged in value-creation by the introduction of new products such as seaweed soaps, seaweed oil, seaweed juice, jam and cookies. These are not exported but sold on the local market.
Morogoro Rice Processing Cluster
The Rice Processing cluster consists of rice farmers, rice processors and rice traders, as well as representatives from government through Morogoro Municipal Agricultural office and the District Trade office, and academia through Cholima research center. The cluster focus on increasing rice-farming productivity and improve the quality standards as well as market linkages, and has also expressed a need to improve financial linkages so as to increase access to capital.
Eastern Regions Mushroom Cluster
The Eastern Regions Mushroom cluster consists of mushroom farmers, mushroom processors and spawn makers, and has been collaborating with the College of Engineering and the Faculty of Science at the University of Dar es Salaam (UDSM), as well as with the local governments.
The efforts have been directed towards sharing of experiences among farmers and trainings of mushroom cultivation and increasing the quality of mushroom spawns.
Source: COSTECH, Knowledge products from sti clusters
5. METHODOLOGY
5.1 Research design
The method in this thesis is a combination of quantitative and qualitative elements, mainly focusing on the quantitative aspect. A number of previous studies evaluating cluster organizations emphasize the importance of combining quantitative and qualitative measures (Klofsten, Bienkowska, Laur and Sölvell, 2015; Klofsten, 2009; Diez, 2001). The quantitative part of the thesis collects data on clustered firms performance and actor collaborations within the cluster organizations, through a telephone administered questionnaire. The qualitative aspect aims to contribute to a deeper understanding of the cluster organizations; used both before the construction of the questionnaire as a means of adjusting the questions to fit the Tanzanian context, and during the data collection as a complement to the questionnaire in form of interviews.
5.1.1 Data collection
The data in this thesis is collected using an interviewer-administered questionnaire. When using a telephone questionnaire, the response rate is usually higher than for an internet-mediated questionnaire (Saunders, Lewis and Thornhill 2009), and this is particularly true for Tanzania where Internet access is low and not all respondents have access to email. The method of using an interviewer-administered questionnaire was chosen because it makes it possible to reach a large sample of respondents, despite the large distance between the clusters.
The respondents to the questionnaire are the cluster facilitators of the cluster organizations included in the ISCP-Tz. A cluster facilitator is the person who has the overall responsibility for the cluster organization, and the one that has received training through the program. Hence the facilitator is the person who has knowledge about the development of the cluster, and the collaborations that take place between the actors within it. The names and contact details to cluster facilitators were provided by COSTECH.
According to COSTECH (2016) there are 67 cluster organizations included in the program.
However, after going through the number of clusters with COSTECH, the remaining number is 51. The reason for the reduced number is that some of the clusters included in the 67 clusters were identified but were not included in the program, and hence did not receive support. Another reason is that previously sponsored clusters left the program. Additionally, the cluster organizations initiated after year 2013 were excluded, since the cluster organizations need some time for the collaborations and connections to grow, and hence time before evaluation. Out of the 51, COSTECH provided names and contact details to 45 cluster facilitators. This means that the final number of respondents that were contacted and asked to participate in the study by responding to the questionnaire was 45.
5.1.2 Questionnaire
The questionnaire aims to collect data on the existence of actor collaborations so as to be able to analyze cluster gaps and bridge building in accordance with the Gap Model. It also includes a
part where the cluster facilitators are asked to provide financial performance data of the clustered firms (See the questionnaire in Appendix 1).
The questions designed for the questionnaire are a mix of questions developed for this thesis, and questions used in previous literature. The questionnaire consists of nine closed-ended questions and one open-ended question in the end where the respondents are given the opportunity to share their view. In four of the nine closed-ended questions, there is a possibility for the respondents to add a choice that fits their situation better than the ones outlined in the questions. Questions 5, 6, 8 and 9 in the questionnaire are adopted from Ketels, Lindqvist and Sölvell (2006), and slightly adapted. The adoption of questions from Ketels, Lindqvist and Sölvell (2006) will enable a comparison of the results to a large sample of developing regions included in their research. The questions constructed for this thesis are developed with inspiration from the cluster literature, question 2, 3 and 4 are related to the Gap Model described in section 2.
The questionnaire consists of rating questions in matrices, one ranking question and one open question. The rating questions are designed with a likert scale, and are a way to collect opinion data (Saunders, Lewis and Thornhill 2009). In this case, assessment on the cluster development, and ratings on the frequency of different activities. The ranking question collects data on the relative importance of the different alternatives, in this case the goals of the cluster organization (Saunders, Lewis and Thornhill 2009). Finally, the open question allows for additional input that the respondents want to share. The questions were pre-coded prior to data collection to facilitate analysis, with numbers representing each alternative.
5.1.3 Translation of questionnaire
The first version of the questionnaire was constructed in English, and then translated to Swahili.
The translation was made by a translator provided by COSTECH, and proofread and corrected by a cluster expert. When translating a questionnaire, it is important to bear in mind the different meanings of words; the lexical meaning, idiomatic meaning and experiential meaning.
Translating the questions due to their lexical meaning, which is the direct translation of words, might give a result that is far from what the question was intended to ask. To make sure the questionnaire did not include direct translations or words unfamiliar to the context, a cluster expert at COSTECH proofread, and corrected, the translated questionnaire.
5.1.4 Interviews and cluster visits
Since most of the cluster theories are developed in a developed country context, interviews with cluster facilitators and visits to cluster sites were included, so as to have their input on the theories and material. The interviews were semi-structured, so as to enable new thoughts and ideas, but at the same time stay within the subject of clusters. Interviews with four cluster facilitators were conducted, and they were chosen according to proximity and industry group.
Due to time- and financial constraints, visits and interviews had to be conducted in close proximity to the COSTECH office in Dar es Salaam. The interviewees were spread out according to industry group, with two clusters representing the group Agriculture, food, basic manufacturing, one in Dar es Salaam and one on Zanzibar, and one cluster within Capital intensive manufacturing, and lastly one within Tourism, both in Dar es Salaam. Visits were also
made to the Seaweed cluster initiative and Soap cluster initiative at Zanzibar, Mwenge Wood Carving cluster initiative, Textile cluster initiative, Handloom cluster initiative and Eastern Mushroom cluster initiative in Dar es Salaam.
5.2 Limitations and validity of the study 5.2.1 Validity and reliability
The validity of a questionnaire refers to the ability of the questionnaire to measure what is intended. Saunders, Lewis and Thornhill (2009) explain different kinds of validity of a questionnaire, two of them will be discussed here; content validity and criterion-related validity.
The first one refers to the extent to which the questions in the questionnaire cover appropriate information. To ensure a high content validity and hence include questions adjusted to the cluster facilitators, I have studied previous literature concerning cluster organizations and discussed the proposed questions in the questionnaire with key persons at COSTECH. It was of importance to discuss the questionnaire with experts at COSTECH since most of the theories concerning cluster organizations are made in the contexts of developed countries. By receiving their input, the questionnaire was adjusted to make sure the questions are relevant for the cluster facilitators in Tanzania. Additionally, a pilot test including two respondents was undertaken, in order to make sure the questions are suitable and understandable. After the pilot test, two changes were made in the questionnaire. The first one was made in question 2A and 2B where the respondent noted there was no option for selecting every month, hence that was added. The second change was made regarding the translation of the English word cluster, also in question 2A and 2B, where the first version used the Swahili translation nguzo, which was changed to kongano.
Criterion-related validity of the questionnaire refers to the extent to which the questions measure what is intended; in this case performance and actor collaborations within the cluster organizations included in ISCP-Tz.
Reliability refers to whether the questionnaire will result in consistent findings when used with different samples and in different time periods. The response rate in this thesis is not 100 %, which means a different sample could have generated different results. Different time periods could also generate different results, since economic performance and collaborations evolve and develop over time (Saunders, Lewis and Thornhill, 2009).
5.2.2 Challenges with cluster policy evaluations
According to Schmiedeberg (2010), there are four main challenges with cluster policy evaluations; the organization of evaluation, how to define performance, the time lag between policy intervention and impact, and data availability. The first one refers to a principal-agent problem since many evaluators are assigned by the policy makers and hence are likely to satisfy the policy maker instead of conducting an independent evaluation. Even though this thesis is written independently of any organization, the collaboration with COSTECH, for example in terms of being introduced to cluster facilitators, could influence the answers of the respondents.
The second challenge about how to define performance lies in defining which outcome the
evaluation should focus on. An evaluation can focus on the cluster itself, the collaborations between the actors and the growth of the cluster, or it could focus on firm performance or macroeconomic factors evaluating regional performance. Hence it is important to define the boundaries of the cluster and which indicators to measure. It is also important to bear in mind that firm- and regional performance is influenced by external factors as well, and there is also a time lag between the intervention and the impacts, which makes it difficult to evaluate possible causalities. The fourth and last challenge brought up by Schmiedeberg (2010) concerns data availability. Data availability is a challenge that has been prominent in this thesis, since a majority of the firms do not keep track on their financial performance.
6. EMPIRICAL ANALYSIS
6.1 Results
This section provides the results from the questionnaire.
Most of the cluster initiatives included in the survey were initiated during 2012-2013, 18 of 32.
Six of them were initiated 2006-2008 and four 2009-2011, and initiation year is missing for the four cluster organizations left. 22 % of the respondents are women, and 78 % are men.
Figure 4 shows that being part of a cluster organization has led to an increase in sales, wages, innovation (as measured by new products and improved products), turnover and number of employed, based on an assessment by the cluster facilitators. 90 % of the respondents indicate that being part of a cluster organization has led to an increase in improved products and turnover, and the results are almost as high for sales, 87 %. Some of the respondents indicate that there is no difference by being part of a cluster organization, and a few report a decrease. It is worth noting that 10 %, which is approximately 3 cluster organizations, have experienced a decrease in sales since they became part of a cluster organization.
Figure 5 shows the contact frequency between the clustered firms and the actors within the cluster. 60 % of the firms within the cluster are in contact every quarter or less, which means a very low contact frequency. Only 40 % of the firms are in contact with other firms within the same cluster monthly or weekly, and when it comes to government institutions the results show 27 %. However, a majority of the firms are never, or less frequently than every year, in contact with academia, financial institutions, other cluster organization and international markets6.
6A low contact frequency with international markets is explained by the fact that most firms do not operate on the international market
10% 3% 7%
0% 3% 3%
3%
23%
37%
10% 6%
26%
87%
74%
57%
90% 90%
71%
0%
25%
50%
75%
100%
Sales Wages New products
Improved products
Turnover Number of employees
Percentage of respondents
FIGURE 4. CHANGE IN INDICATORS
Question 1: In your view, has being part of a cluster organization led to a change in the following indicators for the firms within the cluster?
Decrease No difference Increase
83%
59%
58%
53%
47%
29%
17%
28%
32%
31%
47%
35%
0%
13%
10%
16%
6%
35%
0% 25% 50% 75% 100%
International markets Financial Institutions Academia Other cluster organizations Regional markets Governmental institutions
FIGURE 6: CONTACT FREQUENCY CO's
Question 2b: How often is the cluster organization in contact with the following actors
Never or less frequently than every year
Every year or every quarter Every month or every week 70%
65%
58%
55%
37%
30%
23%
27%
23%
29%
28%
57%
43%
37%
3%
13%
13%
17%
7%
27%
40%
0% 25% 50% 75% 100%
International markets Other cluster organizations Financial Institutions Academia Regional markets Government institutions Firms in the same CO
FIGURE 5: CONTACT FREQUENCY FIRMS
Question 2a: How often are the firms within the cluster organization in contact with the following actors?
Never or less frequently than every year
Every year or every quarter Every month or every week
The results concerning contact frequency in figure 6 between the cluster organizations, managed by the facilitator, and the actors, show a similar pattern. The actor with the highest contact frequency is governmental institutions, where 35 % of them are in contact monthly or weekly.
Figure 7 shows the priorities given to the different collaborations. It shows that 52 % of the respondents give high priority to firm-to-firm collaboration within the cluster organization, 35 % give high priority to collaboration between firms and government, 29 % give high priority to collaboration between clusters, 27 % give high priority to collaboration with other markets, 23 % give high priority to collaborations between firms and financial institutions, and lastly 17 % give high priority to firm-academia collaborations. 20 % have not engaged in any firm-academia collaboration, and the corresponding number for firms and financial institutions is 16 %.
0%
10%
20%
30%
40%
50%
60%
Firm to firm collaboration within the cluster
organization
Firms and governmental
institutions
Collaboration with other clusters in similar
or other sectors
Collaboration with other markets (local,
regional or international)
Firms and financial institutions
Firms and academia
FIGURE 7: PRIORITIZED COLLABORATIONS
Question 3: Please rank the following collaborations according to priority:
Have not engaged in any collaboration
Low priority Medium priority High priority
0%
10%
20%
30%
40%
Share of respondents
FIGURE 8. CONTACT FREQUENCY WITH PRIORITIZED ACTORS
High priority and are in contact quarterly, monthly or weekly High priority and are in contact yearly, less frequently than once a year or never
Figure 8 combines the respondents who assigned high priority to the different actors, with the level of contact frequency for the same actors. The green bars show the share of respondents who assigned both high priority to the collaboration, and chose quarterly, monthly or weekly contact frequency with the same actor. This is to see if the respondents who assigned high priority to the different collaborations have a higher contact frequency with those actors. The results show that prioritized collaborations have a higher contact frequency for firm-to-firm and firm-to- government, and to a small degree when it comes to firm-to-academia. For the other actors, contact frequency is low although the respondents assigned them high priority.
Figure 9 shows the average reply of five cluster organizations in the regions Morogoro, Zanzibar (including both Unguja and Pemba) and Dar es Salaam. The comparison between these three particular regions was made because they have the highest concentration of clusters. The figure shows that contact frequency with the clustered actors are on average higher in the clusters located in Morogoro than in the other regions, with all actors except regional and international markets, where Zanzibar dominates the previous, and Dar es Salaam the latter.
