This thesis investigates whether and how public policies can help build industrial clusters. The research applies a case study method based on 60 interviews to the emerging information communication technology (ICT) clusters in Scotland and Hong Kong. The analysis uses Molina’s sociotechnical constituency (STC) framework and its associated ‘diamond of alignment’, which help focus on two interrelated dimensions: 1) the complex technical and social aspects of the design, implementation and evolution of the Scottish and Hong Kong ICT clusters and 2) the difficulties of developing a cluster in the context of major diversity of organisational interests and patterns of interaction. This research revealed that the cluster building effort in the two regions has been fraught with difficulties due to misalignment between the perceptions and pursuits of policy makers and the interests of industry members. This thesis concludes that cluster building is an evolutionary process of sociotechnical alignment which can be facilitated by feedback and learning. It also suggests that for successful cluster building to take place, policy makers should focus on stimulating the processes of cluster formation, including the building up of the technological capabilities of the industrial actors, while facilitating the integration of the major actors’ interests and demands with the policy programmes. Cluster building involves the development of new ways of thinking as well as the practice of networking; it necessitates the coherent effort of collective learning and a long-term commitment to change the existing technological system. A long-term adaptive policy programme should be pursued to focus on effectively aligning the interests and pursuits of the different actors in the cluster at various stages.