Mapping innovation in the global photovoltaic industry: a bibliometric approach to cluster identification and analysis

The photovoltaic industry acts as a key force in the transition toward a sustainable energy production model. Only a niche market a few years ago, photovoltaic panels and installations are now becoming a mainstream electricity provider. As a highly dynamic, multi-technology and globally distributed industry it is a challenge to identify and quantify its appearance in regional clusters. This challenge is even greater for its critically important innovation activities, including reaching higher efficiency of the cells, and increasing (other) functional and design qualities, like flexibility and color, as well as developing cheaper production methods. Yet because of the industry?s important role in securing a sustainable energy supply and in contributing to the regional economies in which it is established, a deeper understanding of the its global presence and activities is of significant scientific importance and policy relevance. This research follows the approach of 'paper trail' of the industry's innovation process as revealed in patents and scientific publications. By using these documents as sources from which to extract indicators for innovative activity, inputs, outputs and collaboration networks, a detailed picture of the photovoltaic industry innovation and its constituent regional clusters is constructed. This allows not only for the identification and analysis of major regional changes and global shifts of the industry and variation in cluster types, but also enables the estimation of models to identify at least some of the critical factors in photovoltaic clusters' innovation growth pattern, in so far as they can be revealed by bibliometric indicators. In order to select the relevant documents, we first determine which technologies are involved in the photovoltaic industry, like concerning the materials of the cells and shape of the panels, and derive this from expert opinion. We use the USPTO patent database and the Scopus database to retrieve relevant data published between 2005 and 2014. This data is used to carry out a multiple regression model estimation. Providing an understanding of the global location changes and growth and underlying factors in photovoltaic cluster's innovation development is new. The analysis exemplifies the currently increasing scientific attention to the role of cities and regions in transitions of socio-technical systems towards higher levels of sustainability, while referring to local seedbed conditions, including knowledge spillovers, and to networking power of technology actors like multinational companies and universities. The results offer insights to policy makers who aim to avoid barriers to innovation arising from global shifts.