Title: Factors Influencing The Development Of SME Clusters: Evidence From Pakistan Submission ID: 209 Authors: 1. Anjum Fayyaz* *corresponding author 2. Dr. Sarfraz Mian

University/Organization:

Email:

Lahore University of Management Sciences, Pakistan / Copenhagen Business School, Denmark

[email protected]

School of Business, State University of New York at Oswego, USA

[email protected]

3. Dr. Faheem Ul Islam Lahore University of Management Sciences, Pakistan

[email protected]

ABSTRACT1: This paper explores various factors that influence the development of small and medium enterprise (SME) clusters and how their performance can be measured by using a proposed set of variables derived in this study. The research employed data obtained from various “diagnostic studies” conducted on 20 clusters covered by the UNIDO‟s clusters2 and networks3 development initiative in Pakistan. We rate the success or failure of cluster development approach in Pakistan by categorizing the clusters into „HOT‟, „WARM‟ and „COLD.‟ This categorization was based on variables such as application of cluster development methodology, presence of trained Cluster Development Agents (CDA), creation of social capital, key characteristics of clusters, ownership of partner agencies, involvement of governmental agencies and the role of interest groups such as industry chambers or associations in these clusters. These variables are derived from the literature and assigned weights in accordance with their relative importance determined while applying the UNIDO4 cluster development methodology. The paper concludes with a strategy to convert „COLD‟ into „WARM‟ and „WARM‟ into „HOT‟ clusters. Keywords: SME Clusters, Networks, Cluster Development Agents (CDA), Diagnostic Study, Categorizing, Performance Outcomes

2

LITERATURE REVIEW & FRAMEWORK FOR ANALYSIS The importance of clustering and networking has been recognized in the literature over the last more than 100 years. The pioneering work by Marshall (1920) explained the advantages arising from the concentration of many small businesses of similar character in particular localities which describe the benefits associated with the agglomeration economies. It is now well known that clusters offered SMEs, at the very least, external economic advantages, including economies of scale and scope. Co-operation between agents within the clusters and networks, through the sharing of information, resources, knowledge and technical expertise, and other forms of joint actions reduced the transaction costs and further enhanced the competitiveness as well as accelerated learning and technical innovation. Nadvi (1995) also talked about the importance of clustering and networking for the small firms operating in the environments that were industrially and infra-structurally underdeveloped. He took the examples of shoe clusters in Brazil and Mexico, cotton knitwear cluster in Tirupur in India, engineering and electronics cluster in Banglore in India and electronics cluster in Seoul, Korea. The objective of reviewing these case studies was to learn the production organisation and strategies of co-operation that raise efficiency and upgrade skills and technologies. The role of institutions, both local associations as well as government support bodies was also explored in facilitating the organisational and technical learning strategy and furthering the growth of the industrial clusters and networks. Nadvi and Halder (2002) have corroborated with Marshall‟s views using the example of Sialkot surgical instruments cluster in Pakistan. Malmberg, A., Solvell, O. and Zander, I. (1996) discussed the need for a renaissance in the study of external economies. They explained the need to understand the agglomeration economies, regional science, and industrial and economic policy. They also suggested that the most decisive economic policy influences were often at the local level. Humphrey & Schmitz (1995) took the evidences from both developed and developing countries that clustering and networking helped small- and medium-sized enterprises (SMEs) to raise their competitiveness. They also highlighted that the role of public policy in this process was less clear. The European experience suggested that local and regional government could play an important role, particularly where it cooperated with private sector organizations. They also realized that the clustering was more important for the developing countries. Research on clustering in developing countries had been inspired by the competitiveness of industrial districts in advanced countries. The success of the Third Italy and other European regions in the 1970s and 1980s exemplified the power of collective efficiency focused on the need of demanding customers. These experiences contributed to a shift in the industrial policy debate: towards more concern with (1) the role of regional and local governments, and (2) joint public/private sector initiatives.

3

They also talked about the experiences from Indonesia and Brazil where the dynamic clusters and networks were buyer driven and that public support for upgrading of SMEs had more impact in buyer driven than in supplier driven chains. Helping SMEs to become more customer oriented was a key element of effective assistance. Intervention was most effective when it was based on the „Triple C‟ approach, which was explained as customer oriented, collective and cumulative. Bazan and Schmitz (1997) found that the trust and the related concept of social capital were present in a cluster resulting in the probability of networking and trust building. Thompson (1967) explained the role of a network level coordinator or administrative organisation to facilitate the interaction among the network member firms. Numerous examples, in both industrialized and developing countries, demonstrated that SME clusters had established themselves as important and dynamic players within the international market responding to global competition challenges by capitalizing on local opportunities and collective competitive advantage. UNIDO‟s work in India, Pakistan, Nicaragua, Senegal, Zimbabwe, etc. is a manifestation of this aspect (UNIDO‟s brochure on SME Clusters and Networks). Empirical work by Larson (1992), Doz (1996), Gulati and Gargiulo (1999) have focused on the evolution of cooperative network relationships. They found that the multilateral networks involve direct interactions among many member organizations that may never have interacted with one another before. They also involved interactions between members firms and an administrative entity that coordinated network activities. Ghani & Fayyaz (2005) discussed the benefits of networking and trust building in their case study on Lahore Woven Garment Consortium (LGC) in Pakistan. They also talked about the improvement in the bargaining power of network members while working under a consortium. They also explained the importance of collaborative efficiency to explore and develop a new market. They also observed positive transactional and transformational outcomes for the members of the consortium due to networking and trust building. Fayyaz, Mian & Khan (2009), while discussing the entrepreneurship in Pakistan also elaborated the importance and role of support institutions and interest groups while developing SMEs. They also highlighted that the problems in the current support system for SMEs in Pakistan was due to the overlapping mission statements of agencies working to promote SMEs and entrepreneurs, non-availability of right business development service providers (public and private) and the non-professional management of interest groups such as chambers and associations in the industrial clusters. These issues had hampered the growth and sustainability of SMEs in Pakistan. They also highlighted the role of government agencies, which had overlapping mission statements while intervening in different SME sectors and clusters. These interventions

4

were required to be reconciled through defining the core competence of the agencies involved in a project. This in turn required the capacity building of the institutional mechanisms working on industrial revival and growth in the country. If we look at the UNIDO approach to SME clusters and networks evelopment, UNIDO has developed a methodology of initiating successful interventions aimed at fostering cooperative relations within SME clusters, which has been applied in a large number of developing countries throughout the world during the past ten years. A particularly successful programme of this kind has been under implementation in India since the mid1990s. This also began with a limited number of pilot clusters and has now expanded dramatically to cover more than 100 clusters in various industrial branches throughout the country, with the number of counterpart institutions engaged in implementing the programme having increased from two in 1997 to twenty in 2005. As a result, UNIDO can provide the appropriate technical assistance needed to enhance local capabilities for the development of under-performing SME clusters, inter alia by generating a shared development vision for each specific cluster, organizing the dispersed knowledge and resources towards common objectives, and helping the implementation of jointly-identified actions. The main features of this strategy are an emphasis on:

 A participatory approach to vision building: Under-performing clusters are characterized by internal conflicts and a very low level of social capital. Defusing such conflicts and providing the right environment for the development of much-needed mutual trust requires a shift in the vision and expectations of all cluster stakeholders. Based on the findings of a participatory diagnostic study, the UNIDO approach aims at identifying a new agenda for the cluster that enables the cluster stakeholders to understand and appreciate their mutual interdependence and the challenges they jointly face. Building upon a rich interaction among the cluster stakeholders, the approach aims at the identification of a shared action plan, consisting of the initiatives that will relieve the structural limitations that hinder the development of the cluster. The action plans so produced seek to maximize the developmental contribution of local cluster actors, including the targeted SMEs and their respective associations, public and private SME support institutions, and other enterprises along the value chain. This participatory approach is illustrated in Figure 2.

5

Figure 2: Participatory Approach to SME Cluster Development:

Initial focus on selected pilot clusters UNIDO

Localand Central Government

Later expansion of project to other clusters

Financial Institutions

Universities

Internal mea sures to enhance collective efficiency of selected SME clusters

International

CSOs

National CSOs International Experts

Activities and support services • Entrepreneurship development • Management and vocational training • Hard-/software management • Business linkages/networks and integration tinto national and global value chains • Linkages with national and international Sources of product development and technology • Marketing support • IT support • Investment support

CSOs: Civil Society Organizations

 Capability-building actions based on shared targets: A key feature of this approach is that it aims at sensitizing and connecting cluster actors, as also indicated in Figure 2. This is not regarded as an end in itself, but rather as a necessary step towards the identification of a shared vision of the cluster and the implementation of coordinated capacity-building actions. Owing to the initial mistrust that characterizes under-performing clusters, the strategy initially concentrates on short-term initiatives for smaller groups of enterprises supported by local institutions. These initiatives provide the opportunity for a progressive exercise of planning actions and assuming responsibilities.

 Sustainability of the cluster development process: The ultimate goal of this approach is to promote the emergence of an autonomous cluster governance framework ensuring that the local development process continues even after the completion of the UNIDO project. This implies the provision of a variety of capabilitybuilding services that do not necessarily involve the creation of new institutions or business

6

development services (BDS), but rather the networking of existing ones and bridging the gap between demand and supply. The responsibility for implementing this strategy lies with a specialist, the cluster development agent (CDA), who has been specifically trained by UNIDO in the fields of enterprise/cluster diagnostics, conflict management and resolution, network building, project management and project evaluation. The cluster development agent is expected to reside in the cluster for the duration of the project in order to gain a full understanding of the challenges and opportunities faced by the cluster entrepreneurs. By operating on a fulltime basis within the cluster, he/she is expected to establish a relationship of trust with and between the cluster entrepreneurs. He/she is also expected to enjoy a high degree of autonomy in mobilizing the technical, financial and human resources required for the development of the cluster, and to support and guide the entrepreneurs to identify and realize new opportunities while helping them to develop their own entrepreneurial talents. Finally, the UNIDO approach provides for the establishment of an institutional mechanism for the continuous training of cluster development agents to support the expansion of the programme to cover an ever-increasing number of under-performing clusters. This training is conducted through specialized courses, documentation of best practice, study tours, exchange of experience and joint learning workshops, etc. The literature above talks about the importance of clustering and networking and the benefits associated with the agglomeration economies. It also explained the linkages of external economies with the cooperation between the agents within the clusters and networks, through sharing of information, resources, knowledge and technical expertise. This strategy helped to reduce the transaction costs and enhanced the competitiveness, learning and innovation. As the UNIDO Methodology of clusters and networks development revolved around these parameters so the first variable under-consideration to categorize the clusters was, the “Application of Cluster Development Methodology in the selected clusters.” On the issue of presence of full time change agent (Cluster Development Agent – CDA) in the cluster, literature talks about administrative mechanism required to coordinate the activities of multilateral networks. At some of the places role of a network level coordinator or administrative organization was emphasized to facilitate the interaction among the member firms. In a similar way, UNIDO methodology of cluster development specifically considered the presence and role of a coordinator, called a change agent or a cluster development agent (CDA) in the cluster as one of the key variable for success or failure of a cluster. The third variable was an outcome of understanding of trust and social capital in a cluster. While evaluating the success or failure of a cluster, it was important to consider the aspect of social dynamics to decipher clusters into hot, warm and cold.

7

Although initial theories on cluster development talked about the importance of agglomeration economies and benefits associated with the clustering and networking but the most important aspect which UNIDO considered even before starting the application of methodology was the characteristic of a cluster, taken as a fourth variable in this paper. This included the economic contribution of the cluster, number of people employed and target markets of the selected cluster. Defining the ownership of partner agency and role of the government was based on the willingness and the core competence of the partner organizations while applying UNIDO cluster development methodology on a cluster. Fifth and sixth variables considered it important if the partner agencies had independent reporting mechanism for cluster development within their organizations. Furthermore, recognition and support of policy to the cluster development programme in Pakistan was a breakthrough, which led to the execution of the programme by the implementing wings of relevant ministries. Finally, allocation of a separate budget for the project in addition to the nomination of UNIDO trained focal point for the programme by a partner agency was taken into account for the study. These variables had been taken from the literature as well as the 20 diagnostic studies conducted under the clusters and networks development programme in Pakistan. Final and the seventh variable was the role of associations and chambers, named as interest groups in the literature. This represented the presence and voice of the core cluster actors i.e. producers and exporters in a specific cluster. The other components of this variable were ownership and understanding of the members of these groups in addition to the alignment of agendas of the interest groups with the developmental and political needs of a cluster. This is how the theoretical framework based on seven variables was developed by taking a look at the literature and the diagnostic studies conducted by UNIDO in Pakistan.

8

METHODOLOGY: Based on the expert opinion, relative weights were assigned to the variables selected for the study1. Categorization of clusters according to the score under each variable is detailed in the following table: Explanation of Variables to Categorise a Cluster as „Hot‟, „Warm‟ Or „Cold‟ I. Application of Cluster Development Methodology – 25 Serial No.

Components

Weighted (Numbers)

1.

Placement of a full time UNIDO trained CDA

5

2.

Diagnostic study developed as per UNIDO methodology

5

3.

Allocation of budget for implementation of sub-activities / CFCs Handholding mechanism / capacity / willingness of the partner agency to implement Impact on growth / profitability of the cluster / network

5

4.

5.

1

5

5

Remarks

This is one of the most important factors and starting point of cluster development approach. Cluster Development Agent (CDA) or Social Mobilizer is hired/nominated and trained through a structured programme of UNIDO before placing the person in the field for actual implementation. Although diagnostic study (DS) is a part of CDA training but sometimes partner agencies use a different model of conducting the diagnostic studies without following the required pattern already defined during the training process. Changing the template or using a different template hampers the implementation process and DS becomes a mix of apples and oranges. Most of the times the partner agencies have a predefined budget for the activities in the clusters for example establishment of training centre or a common facility centre (CFC). It is an important aspect if the partner agency has already nominated some funds for intervention in the cluster. It is a part of capacity building and willingness of the partner agency. It can also be taken as the seriousness linked with the track record of partner agency and its future mandate to move forward in this direction.

It is action-based programme linked with the external economies of scale, ultimately leading to some tangible benefits not only at the cluster level but also at the network and individual company level. That is why this is important

Several cluster and networking experts associated with UNIDO were consulted to arrive at a consensus decision to rank order the variable.

9

to understand the impact of cluster methodology in terms of profitability of the cluster in general and network/individual companies in specific.

II. Presence of Cluster Development Agent – CDA (Change Agent) – 15 Serial No. 1.

Components

Presence of a Full Time Change Agent from the Private Sector

Weighted (Numbers) 10

5

if CDA is present on part-time basis or placed for a limited time period say (1-2 years) 2. 3.

Authority with the Change Agent Resource allocation for the Cluster Development Awareness Workshops

3 2

Remarks

It is important to note that the characteristics of a change agent (CDA) including attitude, skill and germs to contribute are extremely important to intervene in a cluster. Furthermore, the neutrality of the change is one of the key factors to be taken into account for value addition to the cluster. For that matter, it is advisable to hire change agents from the private sector and from within the cluster to have professional and neutral outlook. In most of the cases, when government departments nominate their officers as change agents, the probability of transfers and postings after few months or a year jeopardizes the whole process of change management. Once the change agents are in place, there is a need to provide them with some authority and resources to work on the joint activities, which are not done by the stakeholders themselves in most of the cases. These joint activities include awareness creation workshops and seminars, exposure visits and sharing of best practices.

III. Social Dynamics of the Cluster – 15 Serial No. 1.

Components

Formal Networks Developed in the Cluster

Weighted (Numbers) 5

Remarks

Whenever networking and trust building process is initiated by a CDA in a cluster, some special purpose vehicles such as networks, self help groups or consortiums are used to build trust and create social capital in the cluster. Some of these

10

2.

Informal Network/s in the Cluster

5

3.

Mutual Cooperation present in the Cluster at informal level – no clear cut networking

5

networks or consortiums pass through life cycle through collaborative efficiency and joint efforts in such as way that they ultimately feel to become a legal network through registration from the security and exchange commission of Pakistan. These networks are then called formal networks. Availability of these networks in the cluster is an indication of change in the cluster. During the process of cluster development, it is not necessary that all the networks formed should pass through a legal structure. They can form informal networks or consortiums to achieve specific joint objectives and leave that network and move to some other network for some other objective. Examples of the activities related to these kind of networking can be joint purchase of raw material, joint skill development or opening a new market jointly, etc. In most of the clusters, there are inherent mutual cooperation mechanisms and some level of trust along the supply chain. People deal on the basis of trust and cooperate in building each others capacity through this mechanism. For example buyers or buying houses build the capacity of suppliers to match the quality requirements. This type of cooperation is not explicit but can be identified through a trained CDA working in the cluster.

IV. Cluster Characteristics – 10 Serial No. 1. 2.

3.

4. 5.

Components GDP Contribution No. of Establishments => 100 No. of Employees in the Cluster => 1000 Presence in Local Market Presence in Export Market

Weighted (Numbers) 2 2

2

2 2

Remarks UNIDO has a predefined criterion of selection of a cluster which includes GDP contribution, number and distribution of small, medium & large enterprise, total number of employees in the cluster, forward linkages such as number of local or international buyers. For simplification sake, we have taken GDP contribution positive for those clusters which are present in both local and the export markets.

11

V. Ownership of Partner Agency – 10 Serial No.

Components

Weighted (Numbers)

1.

Establishment of Cluster Development Cell

3

2.

Allocation of Separate Budget for Cluster Development

3

3.

Application of Methodology in the True Sense (Cluster Development or Network Development

3

4.

Relevant Officers of the Partner Agency Passed through Orientation / Training of Trainers on the Concept

1

Remarks

This is linked with the sustainability of intervention. If partner agency decides to develop an independent mechanism to initiate and run the programme, it moves to the long term sustainability of the system of intervention in a specific cluster. If the partner agency has a mandate to establish a common facility centre (CFC), it also adds value to the cluster but if the CFC is established without doing pre-requisites required through the application of cluster development methodology, the potential of that CFC will not be utilized to the fullest extent. It would require separate budget lines for beforemath, establishment and the aftermath of the intervention. Generally, cluster development would require, capacity building of the enterprises, especially CEO/Directors initially and then the other cluster actors including support institutions, academic institutions and interest groups such as chamber and associations. To sum up, it is a combination of capacity building of enterprises, institutes and strengthening of associations. On the other hand, Network-based methodology requires developing networks in the clusters and alignment of other stakeholders to help the networks grow in the cluster locally and internationally. It comes under training of trainers programme, which is run with two objectives: 1. to create awareness amongst the partner agencies on the importance of cluster development / industrial change management 2. To build capacities to launch and implement their own cluster development programme

VI. Priority of Government – 15 Serial No. 1.

Components

Cluster a Part of Trade Policy or

Weighted (Numbers) 10

Remarks

The clusters which are either announced in the trade policy or part of provincial policy for intervention, they are

12

2.

Policy by Planning & Development Dept. Monitoring & Evaluation by the Government through Establishment of a Cluster Development Cell in the Ministry

considered to be the priority clusters of government for intervention

5

It depicts the ownership of the government for cluster intervention at policy level. Ministries take the ownership and develop a mechanism of monitoring and evaluation in collaboration with the cluster stakeholders.

VII. Role of Industry Chambers / Associations – 10 Serial No. 1.

2.

3.

Components

Weighted (Numbers)

Presence of Chamber / Association Strong Ownership of Chamber / Association

2

Cluster-Specific Developmental Agenda of the Chamber/Associa tion

5

3

Remarks

If the chamber or association is not present in the cluster, there is a need to create some platform to give voice to the cluster. If chamber or association in the cluster has powerful agenda to fight with the government to get benefits for the members. It also includes the track record of the platforms to learn about the cluster-specific successful implementation of projects in the past. If the interest groups i.e. chambers / associations in the cluster have developmental agendas to add value to the businesses of the entrepreneurs in the cluster instead of inclination towards political mobilization.

13

DATA COLLECTION AND ANALYSIS For primary data collection at each of the 20 clusters under study, interviews were conducted with an average of 40 cluster actors including producers / exporters, raw material providers, buying houses, industry association/ chamber representatives, support institutions and the cluster development agents. In a nutshell, the study is based on 20 diagnostic studies, which in turn are based on a total of 800 interviews with their stakeholders. Table 1 shown below has four clusters categorized as „HOT‟ based on the score which is more than or equal to 70 out of the total score of 100 against the seven variables and their weights. Table – 1: “HOT Clusters” Variables to Categorise a Cluster as “Hot”, “Warm” Or “Cold”

Clusters

1.Cutlery Cluster Wazirabad

2.Fan Cluster Gujrat 3.Sports Goods Cluster Sialkot

4.Woven Garments Cluster Lahore

Role of Chamber / Association

Score

(15)

(10)

Out of (100)

Categorise “HOT” (=>70), “WARM ” (50 – 69) or “COLD” (<50)

7 (item 1 missing)

10 (item 2 missing)

2 (items 2,3 missing)

81

HOT

10 All

6 (items 1,4 missing)

10 (item 2 missing)

2 (items 2,3 missing)

78

HOT

15 All

10 All

4 (items 1,2 missing)

10 (item 2 missing)

5 (item 3 missing)

71

HOT

15 All

8 (item 2 missing)

7 (item 1 missing)

10 (item 2 missing)

2 (items 2,3 missing)

82

HOT

Application of Cluster Development Methodology

Presence of Change Agent (CDA)

Social Dynamics of Cluster

(25)

(15)

(15)

25 All

12 (item 2 missing – with full time CDA)

20 (item 4 missing)

Cluster Characteristics

Ownership of Partner Agency

Priority of Government

(10)

(10)

15 All

10 All

15 All

15 All

20 (item 5 missing)

7 (item 2 missing – no full time CDA)

25 All

15 All

Total Score

Analysis of “HOT” Clusters: Four clusters in Table 1 considered as “HOT” with the following common features:

14

Full time change agents while working in these clusters properly applied cluster development approach after conducting the diagnostic studies as per training by UNIDO. For these clusters, separate budgets were allocated not only for the CDAs but also for the activities in these clusters such as awareness creation workshops/seminars and sharing of best practices with the cluster stakeholders. In this way, active members of the cluster took the ownership of the concept and realigned their direction as per requirement of the cluster. An informal process of networking and trust building started with the principle firms and the support institutions in the cluster. This is how the initiation of trust building and networking took place for collaborative efficiency and joint efforts. Once the full time CDA was present and he/she started applying the cluster development methodology, the natural outcome was the trust building and networking of the enterprises as they started taking initiatives such as joint purchase of raw material and machinery, joint skill development, inventory management and process improvement. This is how the trust was converted into social capital and tangible benefits. Once the cluster actors moved to a higher level of trust and confidence they started working jointly to explore the demand side such as market development and diversification. In the “HOT” clusters, the characteristics of the cluster included contribution to the economy, number and types of enterprises and total number of employees, their presence in the local and export markets. In this case almost all four clusters under this category fulfilled these criteria and considered as important clusters for the economy. Although these clusters were “HOT” but most of the partner agencies did not establish independent cluster development cell. Nevertheless, they had cluster development focal points, trained through UNIDO training system with some allocation of budget depending upon the agency‟s mandate. In most of the cases, the budget was allocated only for hiring CDAs and establishing cluster development centres for the visibility. On top of that these clusters were the priority clusters of the government and part of trade policy either at provincial level or federal level. Interest groups such as chambers and associations were vibrant in these clusters. Networks had taken strong positions in the chambers and associations due to their active participation in the meetings of chambers and associations and change of vision. In the worst case scenario, Cutlery Cluster Wazirabad or Fan Cluster Gujrat had more than eight networks in each of these clusters and their voice was strong enough either to influence the government or their trade bodies.

15

Table – 2: “WARM Clusters” Table 2 shown below has seven clusters categorized as „WARM‟ based on the score which is between 50 to 69 out of the total score of 100 against the seven variables and their weights. Table – 2 Clusters

5.Autopart Cluster Lahore 6.Electrical Appliance Cluster Gujranwala 7.Gems & Jewellery Cluster Karachi 8.Knitwear Cluster Lahore 9.Leather Cluster Karachi 10.Leather Cluster Sialkot 11.WoodWorking Cluster Chinniot

Total Score

Variables to Categorise a Cluster as “Hot”, “Warm” Or “Cold” Application of Cluster Development Methodology

Presence of Change Agent (CDA)

Social Dynamics of Cluster

(25) 10 (3,4,5 items missing)

(15) 10 (2,3 items missing)

(15) 10 (1 item missing)

15 (4,5 items missing)

5 (2,3 items missing)

5 (2,3 items missing)

10 (1,4,5 items missing)

7 (2 item missing)

5 (1,2 items missing)

10 (3,4,5 items missing) 10 (1,4,5 items missing) 15 (4,5 items missing)

10 (2,3 items missing) 7 (2 item missing) 5 (2,3 items missing)

10 (1 item missing) 5 (1,2 items missing) 10 (1 item missing)

15 (4,5 items missing)

5 (2,3 items missing)

10 (3 item missing)

Cluster Characteristics

Ownership of Partner Agency

Priority of Government

Role of Chamber / Association

Score Out of (100)

(10)

(10) 7 (3 item missing)

(15) 10 (2 item missing)

(10) 5 (3 item missing)

10 All

4 (1,3 items missing)

10 (2 item missing)

10 All

4 (1,3 items missing)

10 All

10 All 10 All 10 All 6 (1,5 items missing)

Categorise “HOT” (=>70), “WARM ” (50 – 69) or “COLD” (<50)

62

WARM

2 (2,3 items missing)

51

WARM

10 (2 item missing)

5 (3 item missing)

51

WARM

4 (1,2 items missing) 4 (1,3 items missing) 4 (1,2 items missing)

10 (2 item missing) 10 (2 item missing) 10 (2 item missing)

5 (3 item missing) 5 (3 item missing) 5 (3 item missing)

59

WARM

51

WARM

59

WARM

4 (1,3 items missing)

10 (2 item missing)

0

50

WARM

Analysis of “WARM” Clusters: Seven clusters considered as “WARM” with the following common features:

16

Part-time change agents, after passing through the training by UNIDO on cluster development concept were nominated either from the government departments or from the private sector. Professionals nominated from the government were either transferred to other departments or promoted and repositioned whereas people from the private sector were placed in the clusters for limited time period say 1-2 years. Half-hearted application of the methodology and allocation of budget mostly for the tangible projects lacked the soft infrastructure required to develop networks through the change agents in these clusters. As there was no full time presence of the change agent so the impact assessment of the application of methodology was difficult in these clusters. On social dynamics, there were no formal networks in these clusters again due to the absence or limited presence of the change agents. Some informal networks developed or some mutual cooperation existed in these clusters with potential to be converted into tangible benefits for the stakeholders. In the „WARM‟ clusters, the characteristics of the clusters depicted the high importance of these clusters for the economy. They had significant contribution to the economy due to at least 100 or more establishments in the cluster and their contribution to the employment generation. These clusters were present in both local and export markets. Partner agencies in these clusters did not establish independent cluster development cell. Nevertheless, their officers or change agents were trained through UNIDO training system with some allocation of budget depending upon the agency‟s mandate. In most of the cases, the budget was allocated only for common facility centres. These clusters were also the priority clusters of the government and had been part of the trade policy either at provincial or federal levels. Chambers / associations were relatively vibrant and played their role in political influence to get subsidies or tax rebates for their members from the government. Woodworking Chinniot was the only cluster in this category, where chamber or association was either not present or not significant to provide a platform to the core cluster actors. Some informal networks were available in these clusters which was one of the basis to consider them as vibrant. These clusters can be facilitated for higher level of social capital, networking and trust building to ultimately convert them into hot clusters with tangible benefits to the stakeholders.

17

Table – 3: “COLD Clusters” Table 3 shown below has 9 clusters categorized as „COLD‟ based on the score which is less than 50 out of the total score of 100 against the seven variables and their weights. Table – 3 Clusters

Variables to Categorise a Cluster as “Hot”, “Warm” Or “Cold” Application of Cluster Development Methodology (25)

12. Ajrak Cluster Matyari – Sindh 13.Balochi Chappal Cluster Quetta 14.Banana Cluster Nawab Shah 15.Ceramic Pottery Cluster Gujrat 16.Electrical Appliance Cluster Lahore 17.Electrical Fittings Cluster Sargodha 18.Red Chillie Cluster Kunri 19.Rice Cluster Larkana 20.Textile SparePart

Presence of Change Agent (CDA)

Social Dynamics of Cluster

Cluster Characteristics

Ownership of Partner Agency

(15)

(10) 2 (1,2,3,5 items missing) 4 (1,3,5 items missing)

(10)

5 (2,3 items missing)

5 (1,3 items missing)

15 (4,5 items missing)

5 (2,3 items missing)

5 (1,2 items missing)

5 (1,3,4,5 items missing)

5 (2,3 items missing)

5 (1,2 items missing)

6 (1,5 items missing)

10 (3,4,5 items missing)

5 (2,3 items missing)

5 (2,3 items missing)

6 (2,3 items missing)

10 (3,4,5 items missing)

5 (2,3 items missing)

5 (1,3 items missing)

8 (5 item missing)

10 (3,4,5 items missing)

5 (2,3 items missing)

5 (1,2 items missing)

10 All

5 (2,3 items missing)

5 (1,2 items missing)

10 All

5 (2,3 items missing)

5 (1,2 items missing)

10 All

5 (2,3 items

5 (1,2 items

6 (2,3 items

5 (1,3,4,5 items missing) 5 (1,3,4,5 items missing) 10 (3,4,5 items

Categorise “HOT” (=>70), “WARM ” (50 – 69) or “COLD” (<50)

Priority of Government

Role of Chamber / Association

(15)

(10)

7 (1 item missing)

10 (2 item missing)

2 (2,3 items missing)

41

COLD

7 (1 item missing)

10 (2 item missing)

0

46

COLD

10 (2 item missing)

0

32

COLD

10 (2 item missing)

5 (3 item missing)

42

COLD

10 (2 item missing)

2 (2,3 items missing)

41

COLD

10 (2 item missing)

2 (2,3 items missing)

43

COLD

10 (2 item missing)

5 (3 item missing)

44

COLD

10 (2 item missing)

2 (2,3 items missing)

38

COLD

10 (2 item

5 (3 item

42

COLD

(15)

10 (3,4,5 items missing)

Total Score Score

1 (1,2,3 items missing) 1 (1,2,3 items missing) 1 (1,2,3 items missing) 1 (1,2,3 items missing) 4 (1,2 items missing) 1 (1,2,3 items missing) 1 (1,2,3

Out of (100)

18

Cluster Faisalabad

missing)

missing)

missing)

missing)

items missing)

missing)

missing)

Analysis of “COLD” Clusters: Nine clusters considered as „COLD‟ with the following common features: In these clusters, development of diagnostic studies and placement of CDAs for a limited time period was a common feature. In some cases even the CDAs were not nominated, diagnostic study was conducted by the officers nominated by the partner agencies with no understanding of application of methodology and no budget allocation after diagnosing the problems in these clusters. No formal or informal networks developed in these clusters. There was only a flavour of mutual cooperation in these clusters (also called regional concentrations). In these clusters, under the heading of cluster characteristics, it was difficult to assess the contribution of these clusters to the economy, even the number and types of enterprises and employment contribution was not significant. These clusters were mostly targeting the local markets. Partner agencies had no cluster development cell. In some cases, focal points were nominated or CDAs were hired temporarily to be trained through UNIDO training on clusters and network development approach. There was either no allocation of budget or some resources allocated for the common facility centres. Interestingly, all these „COLD‟ clusters had also been the priority clusters of the government and were part of trade policy either at provincial or federal levels. Chambers / associations were not active or in some cases not even present to table the issues of these clusters. In some cases, chambers or associations were present but they were working only for the political agendas to get rebates or subsidies from the government or donor agencies. There was no developmental agenda of these interest groups for the members of the platform.

19

CONCLUSIONS AND RECOMMENDATIONS I. Though diagnostic studies have already been conducted in all the 20 clusters underconsideration, due to the non-availability of full time cluster development agents (CDAs), most of the clusters could not move from „COLD‟ to „WARM‟ and from „WARM‟ to „HOT‟ categories. Thompson (1967) highlighted the role of a coordinator or an administrative organisation where as Larson (1992), Doz (1996), Gulati and Gargiulo (1999) talked about the administrative entity to coordinate network activities. So the role of a full time cluster development agent in the clusters is important to make sure the success of these clusters. Allocation of budget in most of the cases was only for the establishment of common facility centers (CFCs), which was insufficient to apply the holistic cluster development approach. Separate budget and resources were required to mobilize the partner agencies and stakeholders through the development of soft infrastructure such as networks, consortiums and self-help groups. Full time change agents, taken from the private sector, were required to be present in the clusters ideally for 3 years in „HOT‟ clusters, 3-5 years in „WARM‟ clusters and 5-7 years in „COLD‟ clusters. Resource allocation to the change agents was required along with some spending authority to conduct small joint activities such as awareness creation workshops, networking exercises and institution building of the partner agencies. II. On one hand, awareness creation and capacity building of the partner agencies was required, but on the other hand, cluster development approach should have been taken as part of future agendas of the partner agencies. Impact assessment on growth and profitability was only possible after bridging the gap between the cluster actors through networking and trust building. Fayyaz, Mian & Khan (2009), have already elaborated the importance and role of support institutions as it is important to build the capacity of the partner agencies for successful application of methodology. III. Understanding of social dynamics of the cluster in terms of track record of joint activities and norms of interaction of a specific regional concentration (cluster) were important for the change agents and the partner agencies. Mobilization of cluster actors, drivers and active members of the society were highly important to develop networks, consortium and self-help groups through visioning exercises by the CDAs. Strong handholding of cluster actors was required by the CDAs to develop formal and informal networks in the cluster to ultimately convert them into tangible benefits through external economies of scale. Ghani & Fayyaz (2005) while explaining the benefits of networking and trust building also tabled the issues related to improvement in the bargaining power of network members through collaborative efficiency using the concept of external economies. They also observed positive transactional and transformational outcomes for the members of the consortium due to networking and trust building IV. Selection of cluster for intervention was based on pre-defined criteria of UNIDO in collaboration with the partner agencies. For example clusters selected in partnership with Trade Development Authority of Pakistan (TDAP) were either already in export or had

20

potential to export in future. Gross domestic product (GDP) contribution, number of establishment, employment level and target markets were some of the factors taken into consideration while selecting the clusters. Clusters selected in partnership with Small & Medium Enterprise Development Authority (SMEDA) had been selected with a focus on employment generation and poverty alleviation. In conclusion, the clusters were to be selected carefully to get short-term and long-term benefits accordingly. V. Ownership of partner agencies was inescapable as the interventions required in the selected clusters were for a limited time period (say 5-7 years). Once the intervention was complete in a cluster it was important to share the lessons learnt, customize and replicate to other similar situations for higher level of benefits and outreach of the programme. It was also important to have an independent directorate of cluster development in the partner agency for true application of methodology with separate allocation of budget for sub-projects. Capacity building of relevant officers of partner agencies was a pre-requisite of the exercise. VI. SME Cluster Development should have been a separate item to be discussed and approved in the trade policy with clear monitoring and evaluation system. It was also important to establish a directorate by the partner agencies for long term sustainability of the concept and the application. Humphrey & Schmitz (1995) highlighted the role of public policy as one of the important factors in cluster intervention where local and regional governments could play an important role to improve the competitiveness of the clusters. VII. Interest groups such as chambers and associations were extremely important pillars of cluster development approach as they are the representative bodies of trade and commerce in Pakistan. They could be used as strong vehicles for transfer of ownership of the concept at the field level. If these platforms were not available in the clusters, they were to be created and if they were available, their capacities were required to be enhanced to get right benefits from the right partner to add value to their members. Fayyaz, Mian & Khan (2009) highlighted the importance and role of support institutions and interest groups while developing SMEs. Non-professional management of interest groups had already hampered the growth and sustainability of SMEs in Pakistan. The above explanation illustrates the strategy to convert the „COLD‟ clusters into „WARM‟ and „WARM‟ clusters into „HOT‟. Although the variables selected for categorization could have been elaborated more or new variables could have been added to understand the categorization in more detail but data collection and field level understanding required more homework through the addition and analysis of tangible variables attached to the concept. Although the paper gives one level of understanding of strategy to improve the health of clusters but the framework is required to be tested in different situations to come up with an improved version of framework for literary and empirical contributions.

21

REFERENCES Bazan, L., & Schmitz, H. (1997). Social Capital and Export Growth: An Industrial community in Southern Brazil, IDS Working Paper No. 361, University of Sussex. Doz, Y. L., (1996). The evolution of cooperation in strategic alliances: initial conditions or learning process, Strategic Management Journal, Vol. 17, Pp. 55-83. Fayyaz, A., Khan, J.H. and Mian, S.A. (2008). The impact of formal SME networks in emerging economies: a study of formal manufacturing networks in Pakistan, Int. J. Technology Intelligence and Planning, Vol. 4, No. 1, pp.115–129. Fayyaz, A., Mian, S.A. and Khan, J.H. (2009). State of entrepreneurship and globalisation in Pakistan, Int. J. Business and Globalisation, Vol. 3, No. 3, pp.271–287. Ghani, J. and Fayyaz, A. (2005). Lahore Woven Garments Consortium (LGC), Asian Journal of Management Cases, Vol. 4, No. 1, pp.65–86, Lahore University of Management Sciences, Pakistan. Gualti, R., and M. Gargiulo, (1999). Where do international networks come from?, Administrative Science Quarterly, Vol. 40, Pp. 619-652. Hobohm, S. (2006). SME Cluster and Network Development (CND) in Pakistan, Project Document, UNIDO. Humphrey, J. and Schmitz, H. (1995). Principles for promoting clusters & networks of SMEs, Small and Medium Enterprises Branch UNIDO (October 1995). Larson, A., (1992). Network dyads in entrepreneurial settings; A study of the governance of exchange relationships, Administrative Science Quarterly, Vol. 37, Pp. 76-104. Malmberg, A., Solvell, O. and Zander, I. (1996). Spatial Clustering, Local Accumulation of Knowledge and Firm Competitiveness, Geografiska Annaler. Series B, Human Geography, Vol. 78, No. 2. (1996), pp. 85-97. Marshall, A. (1920). Principles of Economics, 8th ed., Macmillan, London. Nadvi, K. (1995). Industrial Clusters and Networks: Case Studies of SME Growth And Innovation, Small and Medium Industries Branch UNIDO Headquarters Vienna, Austria Nadvi, K. & Halder, G. (2002). Local clusters in global value chains: Exploring dynamic linkages between Germany and Pakistan, IDS Working Paper No. 152, University of Sussex.

22

SME Brochure (2001). Development of Clusters & Networks of SMEs, Private Sector Development Branch, UNIDO Vienna. Thompson, J. (1967). Organizations in action, New York: McGraw-Hill Book Company.

END NOTES 1 The views expressed herein are those of the authors and do not necessarily reflect the views of the United Nations Industrial Development Organization (UNIDO) 2 UNIDO defines clusters as sectoral and geographical concentrations of enterprises that produce and sell a range of related or complementary products and, thus, face common challenges and opportunities. These concentrations can give rise to external economies such as emergence of specialized suppliers of raw materials and components or growth of a pool of sector-specific skills and foster development of specialized services in technical, managerial and financial matters. 3 Networks are groups of firms that cooperate on a joint development project complementing each other and specializing in order to overcome common problems, achieve collective efficiency and penetrate markets beyond their individual reach. Networks formed by SMEs only are termed horizontal, to distinguish them from those where one or more large-scale enterprises are involved which are of the vertical type. 4 United Nations Industrial Development Organization