The deployment of new energy technologies and the need for local learning

The deployment of new energy technologies and the need for local learning

Energy Policy 101 (2017) 274–283 Contents lists available at ScienceDirect Energy Policy journal homepage: www.elsevier.com/locate/enpol The deploy...

328KB Sizes 0 Downloads 28 Views

Energy Policy 101 (2017) 274–283

Contents lists available at ScienceDirect

Energy Policy journal homepage: www.elsevier.com/locate/enpol

The deployment of new energy technologies and the need for local learning a,⁎

b

MARK

a

Lena Neij , Eva Heiskanen , Lars Strupeit a b

International Institute for Industrial Environmental Economics (IIIEE), Lund Univeristy, Box 196, 221 00 Lund, Sweden Consumer Society Research Centre, University of Helsinki, P.O. Box 24, FI-00014, Finland

A R T I C L E I N F O

A BS T RAC T

Keywords: Energy technology deployment Distributed energy technologies Technological learning Solar photovoltaic (PV) Local learning

The objective of this paper is to identify local aspects of technological learning in the deployment of solar photovoltaic (PV), a globally important form of distributed energy technology. We review literature in the economics of innovation and economic geography to identify the need for local learning when adopting new technologies and seek evidence on the local aspects of learning processes in the deployment of new (energy) technologies. The analysis focuses on the empirical evidence of learning processes in PV deployment. Our findings show that learning for PV deployment exhibits characteristics of local learning identified in the innovation literature (tacit knowledge, shared narratives, user relations and learning in interorganizational networks). In addition, we show that competencies in the deployment of PV rely on learning processes that are closely connected to the historically and geographically distinctive characteristics of the built environment. We also find evidence of the significance of proximity in (local) learning, as well as examples of knowledge being codified over time into national and global knowledge flows. We conclude with policy implications that acknowledge the importance of local learning for deployment.

1. Introduction The transition towards a low carbon society is a major challenge that will require advances in innovation research and our understanding of technological change and innovation policy. More specifically, the transformation of the energy system will require a good understanding of the nature of technological change related to renewable energy technologies, including both theoretical conceptualizations and empirical evidence. A special case in point is the process of deployment of renewable energy technologies, i.e., the measures needed to get a technology into use and make it work in local contexts. This can involve symbolic work of domestication and societal embedding of new technologies (Sørensen 2013), but also very practical work of technology selection, design, acquisition, commissioning, installation and use, as well as the requisite administrative procedures (such as land use planning and permitting). So far, the innovation literature has had limited focus on the deployment of new energy technologies and such processes have rarely been explicitly defined or examined in the literature (e.g. Mignon and Bergek, 2016). This research gap in the innovation literature on deployment of renewable energy technologies has opened up for research on learning (i.e., the development of technological capabilities) related to deployment of technologies such as wind turbines and solar photovoltaics (PV) (Langniß and Neij, 2004; Shum and Watanabe, 2007, 2008, 2009;



Dewald and Truffer, 2012; Strupeit and Neij, 2016; Strupeit, 2016). These studies have highlighted the localized character of deployment and learning processes, thus opening the question: to what extent does deployment of different renewable energy technologies rely on specifically local learning and policy support. In this paper we define local as a concept referring to a geographical and administrative area that is smaller than a nation state (i.e., a region, municipality or city), while acknowledging that the concept of local also suggests particular institutional, cultural and social commonalities and connections that may or may not coincide with geographical location (Maskell and Malmberg, 1999). Several researchers have highlighted the role of local governments in climate innovation (Hodson and Marvin, 2010; Bulkeley and Castán Broto, 2013) and particularly in the development of capabilities for renewable energy deployment (McCauley and Stephens, 2012; Mattes et al., 2014). Nevertheless, the deployment processes of new energy technologies, and their specifically local nature, remain largely underexplored and poorly conceptualized in the economics of innovation literature. The intention of this paper is to provide support for improvement in innovation and energy policy. Our objective is to advance knowledge on technological change and the local deployment of new energy technologies, with a special focus on the learning aspects of deployment. We seek evidence of the extent to which various learning processes in the deployment of new (energy) technologies are local. We argue that the

Corresponding author. E-mail addresses: [email protected] (L. Neij), Eva.Heiskanen@helsinki.fi (E. Heiskanen), [email protected] (L. Strupeit).

http://dx.doi.org/10.1016/j.enpol.2016.11.029 Received 31 March 2016; Received in revised form 25 September 2016; Accepted 17 November 2016 0301-4215/ © 2016 Published by Elsevier Ltd.

Energy Policy 101 (2017) 274–283

L. Neij et al.

refers to knowledge and embodied skill that is difficult to articulate, let alone codify (Polanyi, 1983; Polanyi, 1996/1997, p 136; Maskell and Malmberg, 1999). Tacit knowledge is sticky and thus may not travel easily beyond the context in which it was generated (Gertler, 2003). Because of this it is difficult to share, and it usually only moves with a small group of people sharing common traits or practices. Over time, pieces of such tacit knowledge become codified: in the case of technological learning, scientific and technological knowledge represents such codified knowledge. While formal, codified knowledge often is created by scientific and technological R & D, knowledge related to processes of doing certain tasks, using certain technologies and interaction among actors often remains tacit and highly localized, drawing on experience rather than codified and mobile knowledge (Lundvall and Johnson 1994; Jensen et al., 2007). While the modern learning economy exhibits a strong tendency to codify knowledge, Maskell and Malmberg (1999) argue that not all pieces of knowledge are equally codifiable. Moreover, they argue that some tacit knowledge is needed in order to use codified knowledge, since codified knowledge does not carry with it the quality judgments and embodied routines required to use knowledge and build up consistent capabilities to continue using and accumulating it. A build-up of internal competencies is necessary for capturing external (codified) knowledge, appreciating its value and making use of it (Dosi, 1988). Even formal professional knowledge involves a tacit component: Dosi and Nelson (2013) argue that industries at a given time exhibit “shared general design concepts” and “problem-solving heuristics” which include normative aspects like criteria for assessing performance, as well as characteristic ways of solving problems and directions in which solutions are sought. Hence, professional engineering knowledge, while much of it is distributed in codified form, also includes an element of tacit appreciation, orientation and propensity to use certain types of knowledge for certain types of problems. The importance of codified and tacit knowledge varies by type of innovation. Asheim et al. (2007) distinguish between two different knowledge bases: analytical and synthetic. Industries (actors) building on an analytical knowledge base draw on scientific knowledge (genetics, biotechnology, IT) and hence, links to universities are important. Knowledge application is in the form of new products, processes and radical innovations. Synthetic knowledge is more relevant for the deployment of technologies, where the innovation takes place mainly in response to the need to solve specific problems arising in the interaction with clients and suppliers. Compared to an analytical knowledge base, learning concerns know-how, craft and practical skill and is often oriented towards the efficiency and reliability of new solutions, or the practical utility and usability of products meeting the needs of the clients (i.e, issues that are particularly important in the deployment of new technologies). According to Asheim et al. (2007), localized learning is more important for industries (actors) drawing on a synthetic knowledge base, although an analytical knowledge base can be drawn from a regional supportive infrastructure (e.g. local universities). Jensen et al. (2007) argue that the most successful firms combine both synthetic and analytical knowledge in their innovation processes. A final point about the content of learning is its downside. Since learning and accumulation of tacit knowledge occurs continually, firms develop routines that are highly durable and path dependent and limit their capacity to respond to changes in the external environment (“competency trap”, see Levitt and March, 1988). Under changing conditions (e.g. rapid changes in factor costs) “unlearning” of existing routines and commitments may be necessary.

evidence and conceptualization of who learns in the deployment of new technologies, what is learned, and how space matters for this type of learning is still patchy. Mapping this territory from a systematic conceptual basis of local learning concepts is important for the identification of policy implications for different spatial levels. The analysis of the local aspects of learning for technology deployment builds on two integrative reviews. First, we review literature in the economics of innovation (Nelson and Winter, 1977; Anon, 1988, 1992) and economic geography (Asheim, 1996; Maskell and Malmberg, 1999) on spatial aspects of technological learning. Second, we apply the conceptualizations found in this literature to empirical evidence of learning processes in PV deployment (see Annex 1 for detailed review methods). Based on these reviews, we aim to offer a more systematic understanding of the following questions: What type of learning processes related to the deployment of new energy technologies can be identified? To what extent are they local? How can these learning processes be supported? We have chosen to analyse the deployment of PVs since it is a technology of major interest in the energy policy discourse on renewable energy technologies and most likely to rely on local learning for its deployment (Shum and Watanabe, 2008). PV modules are deployed and rendered into functioning energy production systems locally, whereas the development and production of PV modules is not a local activity. This may imply that local learning has a special role in the deployment of this technology, also given the emerging evidence on the geographically uneven adoption of PVs within countries (Müller and Rode, 2013; Graziano and Gillingham, 2015; Schaffer and Brun, 2015; Balta-Ozkan et al., 2015; Rode and Weber, 2016). Moreover, the deployment of PV modules relates to the downstream segment of the PV value chain, a segment that is likely to be at least in part localized and different from the upstream manufacturing industry, which deals with the development and production of new energy technologies. Our paper is outlined as follows: Section 2 presents a review of the technological learning literature and its main conceptualizations concerning what is learned, who learns and how space influences technological learning. Section 3 presents the results of our integrative review of the literature concerning PV deployment, and compares this with the technological learning literatures. Section 4 discusses our findings and their limitations, and Section 5 presents conclusions and implications for policy and further research. 2. Conceptualization of local learning for technological change The conceptual review on local learning for the introduction of new technologies presented in this section builds on literature in economics of innovation (e.g. Nelson and Winter, 1977; Lundvall 1988, 1992; Lundvall and Johnson 1994) and economic geography (e.g. Asheim, 1996; Maskell and Malmberg, 1999; Malmberg and Maskell, 2002). The point of departure is the learning economy (Lundvall and Johnsson, 1994), and our focus is on the local (sub-national) level of innovation rather than the national systems of innovation (Freeman, 1992). The review identifies concepts related to the introduction of new technologies and local learning processes. Although the paper has a focus on the deployment of new energy technologies, this review is broader and captures (local) learning processes related to technological change in development, manufacturing, and deployment under the following headings: (1) what is learned locally, (2) who learns and how and (3) how does spatial proximity support learning. 2.1. What is learned locally?

2.2. Who learns and how?

Knowledge is partly contained within a geographic (local) space due to the tacit nature of knowledge. Unlike mere information, knowledge requires social interaction, observation and personal communication for its transmission (Audretsch and Feldman 2004). Tacit knowledge

In the literature, learning has been considered on the level of individuals, organizations, collectives and – with some qualifications – 275

Energy Policy 101 (2017) 274–283

L. Neij et al.

on the level of entire “learning regions” (Asheim, 1996; Florida, 1995; Simmie, 1997) or “regional innovation systems” (Freeman 1992, Lundvall 1992, Nelson 1993). Individuals naturally learn through both formal education and training and via experience and social interaction with other individuals and groups. Organizational learning goes beyond the sum of learning of the organizations’ employees. Like individuals, organizations learn from experience via trial-and-error experimentation and via directed search (Levitt and March, 1988; Cohen and Sproull, 1996). Experience is interpreted, shared and passed on via processes of collective sensemaking of past success and failure, but also more formally in documents, accounts, files, standard operating procedures and rule books. Firms develop internal procedures and routines, which accumulate lessons learned that endure the turnover of personnel and the passage of time (Levitt and March, 1988). Moreover, organizational learning accumulates in the organizational structure of the firm (e.g. what types of functions and professions are represented and at which level of hierarchy), as well as in individual employees’ skills, qualifications and training and in the fixed capital owned by the firm, such as certain types of equipment and machinery (Maskell and Malmberg, 1999). This “organizational memory” directs the firm's search for solutions and also influences its capacity to absorb new knowledge from the external environment (Levitt and March, 1988). The notion of local learning implies that learning also occurs among and between organizations (firms). Cooperation and alliances with other organizations (firms) provide a diversified knowledge base and experiences that provide opportunities for learning. This perspective is highlighted in the concept of “learning by interacting” (Lundvall, 1988; Lundvall and Johnson 1994). Here, learning and feedback processes between suppliers and customers are particularly important (Lundvall, 1988). Since supplier and customer relations are usually non-exclusive, such interaction can lead to learning in supply networks where, for example, a supplier learns from its customer and transfers the results of that learning to another customer (Morgan, 2007). Maskell and Malmberg (1999) further argue that proximity plays an important role in such interfirm learning, since firms draw on a shared base of resources, institutions, social and cultural structures. Formal or informal networks are defined in the innovation literature as an important node for interactive learning (Lundvall, 1988) and an essential way to capture and absorb external knowledge in various forms. Networks can support local learning and link up with sources of codified knowledge at the national or global level. Such a model of local nodes pooling up with global distributed networks has been identified for several types of firms (Asheim, 2012). Local learning will also benefit from pooled resources and knowledge spillovers, which occur because it is difficult to completely own knowledge and because knowledge is not destroyed when used (Audretsch and Feldman 2004). Thus, knowledge flows between organizations (firms) via employees, via suppliers and customers, via educational and professional organizations, and via products and services. Moreover, knowledge requires significant upfront costs to produce but virtually none to reproduce, and the use of knowledge implies increasing returns, i.e. learning begets further learning by increasing organizations' capacity to absorb new information (Cimoli et al., 2006). Collective learning, a concept coined by the European group of researchers called GREMI, refers to learning occurring in “innovative milieux”, i.e., local business environments encompassing a specific production system and a specific culture (Camagni, 1991; Keeble et al., 1999; Lawson and Lorenz, 1999). Collective learning builds on a “substrate” (Camagni, 2002) created by supply networks and local labour markets. The relationships and shared language created by this interaction as well as more informal socialization allows firms in such environments to collectively “transcode” external knowledge into locally relevant decisions and business ideas by drawing on informal contacts, imitation and mutual assessment of rumours. These informal

networks facilitate informal ex ante coordination of private decisions, thus leading to “informal collective action”, i.e., mutually compatible responses to opportunities or problems (Camagni, 2002). While the original notion of collective learning applied to communities hosting innovative small enterprises, Keeble et al. (1999) have successfully applied this concept to the Cambridge region, and collective learning among local technology-based firms and the university. Learning regions (Florida, 1995; Asheim, 1996; Morgan, 2007) and regional innovation systems (Cooke et al., 1997; Asheim and Isaksen, 2002) are concepts that suggest an even more encompassing and systemic scope for local learning. These perspectives imply that in addition to firms, their suppliers and customers, “regions” can build up absorptive capacity and enhance their ability to innovative by combining local and externally sourced knowledge. Building on similar ideas as “innovative milieux”, the notion of the learning region draws on the concepts of interactive learning (Lundvall, 1988) and on the embeddedness of innovation in institutional routines and shared conventions (Morgan, 2007). The learning regions concept suggests that regional authorities can stimulate networked learning, knowledge spillovers and “buzz” through purposive regional development strategies that favour informal information exchange and the combination of diverse knowledge bases, which in turn facilitates the absorption of external knowledge. The similar concept of regional innovation system (Cooke et al., 1997; Asheim and Isaksen, 1997, 2002) stresses the importance of a regional knowledge hub or support centre, while highlighting the role of collective tacit knowledge alongside codified knowledge, and the role of place-specific, contextual knowledge in local innovation. In sum, the local and regional dimension of learning places emphasis on the human capital developed by interactions between firms and local colleges, informal contacts and casual information exchange, and “innovative surplus” or synergies emerging from shared cultural, psychological and political perspectives of those involved in the same specialization (Lundvall and Borrás 1997). This facilitates the use of tacit knowledge, which requires trust, which in turn requires stable and repeated interactions where organizations (firms) are able to monitor each other over time and sink investments in each other. 2.3. How does spatial proximity support learning? Lorenzen (2007) presents a convincing argument for why tacit knowledge is predominantly (though not uniquely) transferred via physical proximity. He attributes the characteristic spatial aspect of learning to social capital and social institutions, which are not exclusively, but predominantly local. Proximity facilitates communication by making the exchange of casual and non-critical information effortless and less costly – also beyond the dyadic relations of customer-supplier relations. Hence, proximity is particularly important for “weak ties” (i.e., non-dependency relations) and the development of social trust, which enables knowledge to credibly flow also via casual relations. Lorenzen (2007) thus argues that while social relations can span (and hence even tacit knowledge can flow) across spatial boundaries, social capital consists of dense and rich overlapping matrices of social interactions and social institutions. It is this overlap of both “strong” (e.g. supply-chain) and “weak” (e.g. professional) ties that depends on proximity and place and reduces the effort needed to transfer tacit knowledge. This is because the clustering of some social relations spills over to others (e.g., supplier-customer relations originate in casual encounters or project-based co-operation) and because social norms, created over repeated interactions, foster social trust, which enables the free flow of information by allowing the finite number of agents in a place-based community to monitor each other's behaviour. Moreover, the distribution of knowledge and learning is facilitated by face-to-face communication, buzz, and local networking. Face-toface interaction offers the capacity for interruption, feedback, and learning (Nohria and Eccles, 1992), which are important for the 276

Energy Policy 101 (2017) 274–283

L. Neij et al.

Heiskanen et al., 2014; Smith et al., 2014; Palm, 2016; Strupeit, 2016. The literature suggests that as PV markets mature, local tacit knowledge has tended to become codified over time. The initially tacit nature of installing PV modules in the early German PV market of the 1990 s (Dewald and Truffer, 2011, 2012) is describedby Strupeit (2016) as residing in the concrete know-how, craft and skills in conventional craft firms (installers) who, due to the absence of commercially available technologies, had to improvise and develop their own solutions for mounting systems and cabling. With time, however, some tacit knowledge becomes codified in artefacts: mounting technology and other deployment-related hardware became more standardized and commercialised (Shum and Watanabe, 2009). As tacit knowledge became embodied into technology, certain local skills related to installation became less important and the respective knowledge has transferred towards upstream manufacturing firms of deployment-related products such as mounting systems, inverters, and other ancillary technology. Similarly, the synthetic and initially local tacit knowledge obtained through the experience with installing PV system technologies in Germany gradually became codified into written or electronic format, such as handbooks, technical installation guidelines and software tools (Strupeit, 2016). This codification facilitated the dissemination of initially tacit knowledge across wider geographies and audience groups, which is evidenced in increasingly uniform spatial distribution patterns across Germany (Rode and Weber, 2016). Codification of deployment knowledge also enabled and facilitated its further diffusion in formal educational programmes (Schaffer and Brun, 2015), which gradually became uniform at the national level. These courses and vocational trainings covered a broad range of deployment-related topics, such as various aspects of installation, maintenance, safety, marketing and sales, finance and legal matters. Zhang et al. (2012) describe the formal education programmes for installers as important to legitimize the workmanship and quality of PV installations in an emerging market like Hong Kong. Jaegersberg and Ure (2011) show that access to this kind of formal training and qualification for SMEs is a key factor differentiating mature from emerging PV markets. While tacit knowledge for the installation of PV systems has shown to become codified, there is less evidence in the literature on local knowledge related to the integration of PV into the physical and historical infrastructure being codified (see Verhees et al., 2013; Karakaya et al., 2015). We even argue that the deployment knowledge related to idiosyncrasies of the physical and historical infrastructure could be characterized as distinctively local and will therefore require local learning. Moreover, the interpretation and application of the growing body of the codified and increasingly complex knowledge base requires complementary skills and tacit knowledge in the local geographical and institutional context. Such tacit knowledge seems, however, to be poorly developed in several countries and, actors, such as architects, urban planners and building developers, have only fragmented knowledge and lack education in planning of PV deployment (Kanters and Horvat, 2012; Palm, 2014; Baborska-Narozny et al., 2016). For example, knowledge and information of PV aesthetics has been found to be insufficient (Wall et al., 2012). In fact, interactions and collaboration between the PV sector and the construction sector have been described as weakly developed in several countries (Verhees et al., 2013; Palm, 2014; Strupeit, 2016). Moreover, critical actors like architects, building developers and urban planners do not have the knowledge and skills to utilise such tools as solar maps/atlases and planning software, which might allow them to better exploit the potentialofsolar energy in the context of the local physical and historical infrastructure (Wall et al., 2012). Likewise, the understanding of user requirements and the development of relationships with users seems to be local and less amenable to codification (Shum and Watanabe, 2008). Several scholars describe a need to understand the user profiles and preferences and to develop relationships with users (Ulsrud et al., 2011; Dewald and Truffer, 2011;

transmission of tacit knowledge (Bathelt et al., 2004; Stroper and Venables, 2004). Buzz refers to non-deliberate exchange of information and knowledge outside formal collaboration, and can also pertain to the transmission of codified knowledge (Asheim et al., 2007; Bathelt et al., 2004). Asheim et al. (2007) say that buzz can be transferred both locally, face-to-face, and globally, electronically. Local networking provides an arena for both buzz and face-to-face communication, knowledge flows, the sharing of experience at the local level and also a local collective memory (Autant-Bernard et al., 2013; Binz et al., 2014). Due to the significance of proximity in local learning, many authors see a particular role for local public policy, or national policy directed particularly toward the local level, in supporting local learning and innovation. In terms of policy implications, these lines of research emphasize the importance of fostering local knowledge networks and interactive learning, rather than mere top-down dissemination of scientific knowledge from centres to peripheries (Asheim and Isaksen 1997; Morgan, 2007). In addition to formal knowledge development, the creation and renewal of skilled labour and knowledge (Asheim et al., 2011) and the fostering of science-industry interactions, policy recommendations include fostering informal interaction and e.g. local platforms for the dissemination of knowledge (Autant-Bernard et al., 2013). Some authors also stress the role of public policy in fostering the necessary “unlearning” of existing routines in the face of new circumstances. This can be done by encouraging openness to external influences, supporting diversity and removing barriers to the entry of outsiders (Autant-Bernard et al., 2013). 3. Local learning in solar PV deployment: emerging evidence and initial observations The review presented in Section 2 brings to light a number of aspects on local learning that are relevant for advancing our knowledge on the deployment of new energy technologies. The literature focuses more on processes of development and production of new technologies than on their deployment – but such learning processes could be of equal importance for deployment. In this section we elaborate further on the nature of learning for deployment drawing on the literature on the deployment of PV technologies (Annex 1), in order to identify:

• • •

what needs to be learnt (locally) for the deployment of PV who needs to be engaged in the deployment of PV and how how spatial proximity supports learning for the deployment of PV

3.1. What is learned locally? The implication of the literature review in Section 2.1 is that the analysis of learning related to the deployment of renewable energy technologies should take into consideration aspects of tacit knowledge as well a synthetic knowledge base characterized by concrete knowhow, craft and practical skills applied to improve efficiency and reliability. Based on a review of the PV literature we elaborate on these findings and provide conceptual examples. The literature on PV stresses the need for deployment-specific (tacit) knowledge and skills that relate to the adoption and adaptation of solar PV modules, i.e. technology selection, acquisition, commissioning, design, installation and use, as well as the requisite administrative procedures (such as land use planning and permitting). In the literature we find research on a generic technology that is adapted into variable local geographical, technological, and institutional contexts: most notably, local climates and patterns of insolation, building orientation and shading from the environment, local demands for configuration of PV system components, local building properties and building materials and traditions, the demands of the electricity grid and local permitting procedures (Zhang et al., 2012; Verhees et al., 2013; Graziano and Gillingham, 2015; Karakaya et al., 2015; 277

Energy Policy 101 (2017) 274–283

L. Neij et al.

Verhees et al., 2013; Smith et al., 2014). This group includes PV system owners and operators, planners, architects and construction firms, wholesalers and installers of PV technology, consultants, utilities and electricity traders, industry associations, NGOs, authorities, financers, insurers, etc. Moreover, we notice that learning for PV deployment involves a range of professions that are not specialized in PV, such as architects, engineers and technicians (Zhang et al., 2012; Smith et al., 2014; Palm, 2014; Strupeit, 2016), urban planners (McCauley and Stephens, 2012; Smith et al., 2014; Palm, 2014; Heiskanen et al., 2015), building permitting authorities (Dong and Wiser, 2013; Barbose et al., 2015) – and importantly – users (Rai and Robinson, 2013; Noll et al., 2014; Verhees et al., 2013; Baborska-Narozny et al., 2016). Learning related to PV deployment has occurred across the entire PV value chain, encompassing producers, intermediates, installers, construction companies and utilities (Shum and Watanabe, 2008; van Mierlo, 2012; Verhees et al., 2013; Heiskanen et al., 2015; Strupeit, 2016). Strupeit and Neij (2016) further show how banks and institutional investors gained experience and knowledge in PV financing, which resulted in organizational learning and dedicated routines in issuing loans to prospective PV users (see also Podewils, 2009; McCauley and Stephens, 2012 and Palit, 2013 on learning by financial institutions). In Germany, increased experience and learning has also resulted in more active insurance companies publishing specific guidelines in an attempt to increase installation quality (VDE, 2011). Various networks and associations played a crucial role for the development of regulatory institutions, not only for financing and insurance but also for the development of quality management standards (Strupeit, 2016). In the USA, local learning has been critical for the provision of locally relevant financial offerings, such as third party ownership (TPO) models (Sigrin et al., 2015; Overholm, 2015; Rai et al., 2016). Learning has been identified also in the interorganizational networks set up for installing PV in the built environment, i.e., networks that facilitate learning-by-interacting (Lundvall, 1988). van Mierlo (2012) and Heiskanen et al. (2015) have observed learning occurring in the value chains and the supply networks for building-sited PV demonstrations. Beyond immediate supply chain relations, McCauley and Stephens (2012) discuss learning in public-private partnerships related to the deployment of PV in public buildings, whereas Noll et al. (2014), McCauley and Stephens (2012), Verhees et al. (2013), Smith et al. (2014) and Livi et al. (2014) discuss the learning occurring in promotional networks – in particular, concerning user and institutional expectations. While many of these networks have been set up by advocacy groups and NGOs, in several countries they have been found to be important sites of learning by professionals; according to Dewald and Truffer (2012), 80% of solar initiatives in Germany included professionals such as craftsmen, installers and planners. In addition, solar initiatives, local climate agencies and consumer protections agencies provided information at the local level and machinery rings served as intermediaries for farmers (Dewald and Truffer, 2011). Some examples can also be found of the kind of collective learning discussed by Camagni (2002), which is primarily based on knowledge spillovers or pooled resources. In its most embryonic form, this occurs through transfer of lessons from one project to another, for example via employees or consultants (Shum and Watanabe 2008; Heiskanen et al., 2015). Overholm (2015) presents a convincing case of knowledge spillovers from a pioneering PV system integrator to followers. Bollinger and Gillingham (2014) also present quantitative evidence suggesting knowledge spillovers among US solar installers. Knowledge can also be shared purposively via cross-firm interactions along the vertical and horizontal axes of the PV value chain. One form of horizontal interaction observed in Germany has been the cooperation of installer firms from different trades (e.g. electrotechnology, roofing, sanitary/climate/plumbing). Specifically, these firms have worked together with the installation of PV systems, thereby complementing their trade-specific tacit knowledge and skills, which

Sigrin et al., 2015; Karakaya et al., 2016; Rai et al., 2016). Dewald and Truffer (2012) highlight the importance in the early German market of the emergence of installers as trusted sources of information for consumers. Learning for PV deployment also includes the development of what Shum and Watanabe (2008) refer to as “next bench design”, i.e., the adaptation of professional paradigms and the development of shared routines, design concepts and problem heuristics among the diverse firms involved in PV installation. For example, van Mierlo (2012) and Zhang et al. (2012) describe how the companies involved in delivering a solar system needed to align practices and develop shared codes and practices for dividing roles and responsibilities in order to deliver functioning system solutions. This type of learning can at first also involve aspects of “unlearning” of previous locked-in capabilities (Maskell and Malmberg, 1999) and institutions. For example, Dewald and Truffer (2012) describe how the German craft sector was initially sceptical toward solar technology and only with time and growing market size increasingly engaged with PV installations. In terms of institutional lock-in, Verhees et al. (2013) show how solar demonstrations served to highlight mismatches between solar and building technology standards. Another dimension of learning identified in the institutional context is the development of shared meaning and narratives about PV (Livi et al., 2014). This process typically occurs at the national or subnational level. For example, Smith et al. (2014) describe how the evolution of solar capabilities in the UK involved the linking of PV to salient discourses about environment, jobs and greening the economy (see also Verhees et al., 2013). Similar observations were made for Germany, where decades of public discourse on energy and sustainability matters (Bruns et al., 2011) served as the breeding ground in which firms and individuals ascribed PV technology a variety of meanings, such as promising business and employment opportunities, a more secure and sustainable energy supply, and a means for selfdetermination by rejecting dependencies on incumbent monopoly players in the energy sector (Strupeit, 2016). Dewald and Truffer (2012) describe how advocacy organizations such as solar initiatives and industry associations played an important role in in Germany in channelling and translating this public discourses into more concrete narratives about the potential role and opportunities of PV, and Noll et al. (2014) describe similar roles of solar community organizations in the USA. Finally, local learning in the institutional context involves the creation of public acceptance concerning the visual impact of PV installations. For example, Heiskanen et al. (2015) describe how successive demonstration projects in Finland led to the development of a distinctive “modest” architectural style of PV installation, which in turn helped grounding and legitimizing the technology in the local context. 3.2. Who learns and how? The implication of the literature review presented in Section 2.2 is that the analysis of learning related to the deployment of renewable energy technologies should take into consideration learning by different actors and at different levels, i.e. by individuals, firms and organizations, but also include learning in interorganizational networks and more diffuse collective learning. The learning by individuals may rely on formal education and training, but more important may be the learning based on experience and social interaction among individuals and organizations. In the text below we illustrate who learns and how in PV deployment. The deployment of PV technologies primarily relates to the downstream segment of the PV value chain, and in contrast to the upstream manufacturing industry that deals with the development and production of PV technology, the downstream segment is characterized by a highly diverse set of actors (Miller and Hope, 2000; Zhang et al., 2012; 278

Energy Policy 101 (2017) 274–283

L. Neij et al.

lated. In this section, we elaborate on the local characteristics of learning in PV deployment and the distinction between local learning versus global knowledge flows. There are inherently local aspects in PV deployment, since the most appropriate practices depend on local geography (insolation patterns), local building culture (e.g. building geometry, roof shapes, mounting and wiring practices), local urban structure (e.g. shading and roof space), local institutions and practices, and actors present in the area. Since installation is a service delivered at the customers' premises, at least small installer companies are likely to operate predominantly within a particular region (Colatat, 2009; Jaegersberg and Ure, 2011; Fabrizio and Hawn, 2013). There are also administrative boundaries defining the mandate of local authorities in tasks such as urban planning and development, building regulations and PV permitting (Dewald and Truffer, 2012; Palm, 2014; Heiskanen et al., 2015). This can fragment the development of institutions and infrastructures for PV deployment and lead to localized pattern of learning, as observed by Dong and Wiser (2013) and Seel et al. (2014). However, these ‘naturally’ local aspects are not usually mentioned in the literature on local technological learning (Section 2). The role of face-to-face interaction is particularly highlighted in the learning observed between users and the network of companies involved in PV deployment. New technologies involve significant uncertainties, and for example Dewald and Truffer (2012) and Strupeit (2016) have shown how the involvement of trusted local craftsmen was critical for the early PV market formation and the development of user-producer relations and the related knowledge flows (see also Karakaya et al., 2015; Mattes et al., 2014). Particularly in early markets, word-of-mouth communication between existing and prospective users of PV technology has served as an important source for local learning and the formation of legitimacy for PV (McCauley and Stephens, 2012; Dewald and Truffer, 2011, 2012; Noll et al., 2014; Sigrin et al., 2015; Rai et al., 2016). Local associations, such as solar initiatives (Dewald and Truffer, 2012; Noll et al., 2014; Graziano and Gillingham 2015; Strupeit 2016) have clearly played a role in PV deployment in the most advanced markets. Similarly, more generally for renewable energy, McCauley and Stephens (2012) and Mattes et al. (2014) highlight the critical role played by local universities and colleges. In the case of the Saxony solar cluster, Jaegersberg and Ure (2011) briefly describe the type of local learning elaborated by Lundvall and Borrás (1997), where the local and regional dimension of learning builds on the human capital developed by interactions between firms and local universities and colleges, informal contacts and casual information exchange, and synergies emerging from shared cultural, psychological and political perspectives of those involved in the same specialization. However, there are few thick descriptions of the rich overlapping networks and the trust created by repeated interactions described by Lorenzen (2007), or of the informal networks leading to ex ante coordination of private decisions, i.e., mutually compatible responses to opportunities or problems described by Camagni (2002). Few solar PV markets are sufficiently developed or studied that predominantly local learning could yet be discerned (Jaegersberg and Ure, 2011). However, some observations suggest the relevance of such processes, in particular in the early development of new markets and between users and installer firms. As described in Section 3.1 the local rooting of installer firms in Germany and proximity to their clients, their knowledge of local markets and a trust relationship with the (prospective) PV system owners were key factors for their success in the market (Bathke, 2009); Strupeit, 2016). The local embeddedness of installer firms also eased interaction between these two actor groups during the operational phase, thereby providing important feedback and learning about the performance of the PV system to the firm (Dewald and Truffer, 2012). Ad-hoc interactions of different craft trades allowed the combination of knowledge from different knowledge bases and enabled mutual

resulted in more efficient and high-quality installation processes. Such cooperation has often been informal and based on previous contacts with other firms in the region (Dewald and Truffer, 2011, 2012; Karakaya et al., 2015; Mattes et al., 2014; Strupeit, 2016). Over time, this kind of cooperation and alignment across craft trades has increased in Germany, due to the need to increase quality, streamline work processes and to gain acceptance and legitimacy for a new technology (which initially is perceived of as risky). In mature markets, there seems to be a tendency for (local) informal information exchange to formalize into formal training and education – some of which might be locally organized and delivered (Shum and Watanabe 2008; Jaegersberg and Ure, 2011; Zhang et al., 2012; Palm, 2014). As an example of cross-firm interactions along the vertical axis of the PV value chain, producers and intermediates of PV system components in Germany established formal and informal networks with their downstream clients. Specifically, these firms built strategic alliances with installer firms in order to gain access to the enduser. As an important element of these alliances, producers and intermediates organized training courses and workshops for installer firms (Strupeit, 2016). At these events, both upstream and downstream actors obtain and exchange knowledge and experience with regards to products, planning and installation as well as new technical and regulatory requirements. At the user-end of the value chain, several authors also highlight the interactive nature of learning about and in interaction with users in the development of competencies for PV deployment (Nagamatsu et al., 2006; Dewald and Truffer, 2012; Noll et al., 2014; Smith et al., 2014; Baborska-Narozny et al., 2016). As mentioned in Section 3.1, “next bench design” can evolve into standardized practices of PV system design and engineering (Shum and Watanabe, 2008) as well as codes concerning the roles and responsibilities of contracting parties (Zhang et al., 2012). For example, Strupeit (2016) show how various schemes for standardization, quality management and safety of PV system components have evolved over time, both in Germany (VDE, 2011; Solarpraxis, 2012) and internationally (e.g., IEC, 2004, 2005, 2008). The creation of voluntary quality management schemes for PV planning and installation (BSW-Solar and ZVEH, 2013; RALsolar, 2013) as well as region-specific industry standards on wind load zones (DIN, 2005a) and snow load zones (DIN, 2005b) further advanced the formal knowledge about goodpractice installation routines. In Germany, these rules have been developed in a highly interactive process by committees with representatives and experts from industry associations, producers of PV system components, as well as testing and research institutes. In emerging PV markets, SMEs lack access to such networks, and have thus been excluded from the rule-setting process, resulting in regulation that is perceived of as onerous (Jaegersberg and Ure, 2011). Learning is also observed within the public administration and among grid operators. Dong and Wiser (2013) show how permitting times and costs can vary significantly across US states, suggesting that administrative learning has occurred locally in those states with more efficient permitting (Burkhardt et al., 2015); however, the particular routines, roles and operating procedures have not been elaborated. Heiskanen et al. (2015) observe that municipalities have gained an increasingly important role as facilitators of solar installations in Finland. Strupeit (2016) show the evolution in the routines for the grid connection process in Germany and well as the development of the institutional framework for PV deployment (see also Seel et al., 2014; Barbose et al., 2015). 3.3. How does spatial proximity support learning? The review in Section 2.3 suggests that spatial proximity has a special role in supporting learning through e.g. knowledge spillovers, pooled resources, an educated, local workforce and local educational and research institutions. However, the role of spatial proximity in supporting particularly the deployment of technologies is less articu279

Energy Policy 101 (2017) 274–283

L. Neij et al.

physics but also institutional and aesthetic aspects that are characteristically local (Wall et al., 2012). Second, the literature indicates the importance of local learning in the case of PV. The deployment of PV includes processes such as technology selection, acquisition, commissioning, design, installation and use, as well as the requisite administrative procedures such as land use planning and permitting – some of these processes are analysed in the literature, others not. Particularly in early markets, the development of these processes is likely to rely on knowledge that may be tacit by nature and on a knowledge base that is likely to be synthetic rather than analytical (Asheim et al., 2007). This in turn, indicates a need to support local learning in terms of capacity building. The literature indicates that demonstration programs have been supportive for local learning and local capacity building. Third, our study indicates the importance of local interaction and networking. The deployment of PV further engages several actors in the supply chain, other than those in the development and production of PVs. Many of these are (at least initially) not specialized in PV or energy, but include building professionals, installers, architects, urban planners and permitting authorities, and even financial service companies. Case studies from several countries highlight the importance of learning-by-interacting (Lundvall, 1988; Lundvall and Johnson 1994) in the multi-actor value chains involved in the deployment of PV in the built environment. The learning by individuals may rely on formal education and training, but more important may be the learning based on experience and social interaction among and between individuals and organizations. In all, this indicate a need for initiatives that provide opportunities for networking. We also found some emerging suggestions of predominantly local learning processes in the deployment of PV that resemble the discussions on “innovative milieux” and “learning regions”. Further research might investigate to what extent local deployment processes draw on the rich overlapping of networks based on strong and weak ties and the trust created by repeated interactions described by Lorenzen (2007), and of the informal networks leading to ex-ante coordination of private decisions, i.e., mutually compatible responses to opportunities or problems described by Camagni (2002). There are also some studies suggesting that pooled resources and interaction between local firms and local educational and research institutions play a role in the learning process, as suggested by Cooke et al. (1997). These can be seen, in particular, in qualitative studies pertaining to the German market (Dewald and Truffer, 2012; Strupeit, 2016), they are visible in studies comparing emerging and mature markets (Jaegersberg and Ure, 2011), and they might in part underlie the uneven geographic uptake of PV observed in several quantitative studies (Graziano and Gillingham, 2015; Schaffer and Brun, 2015; Balta-Ozkan et al., 2015; Rode and Weber, 2016). These observations suggest that local networks and local communication, in terms of buzz and face-to-face, develop the knowledge and skills required to adopt the new technology to its surroundings. Like in learning regions, local regional “networks” are likely to be necessary (at least initially) for the absorption of external knowledge. While further research is merited, the results indicate not only the need for networking but also a need for local networking to support the deployment of new energy technologies. In all, the implication of the literature review is that place does matter, even though the “ubiquitification” of knowledge may reduce the importance of some spatial aspects of learning, especially as markets mature. In some cases, learning based on experiences and experimentation at the local level trickles up to the national level, and good practices on the local level contribute to national legislation, standards, and other codified knowledge. In this way, knowledge may become codified and available in global or national knowledge flows. Still, the interpretation and application of this codified knowledge requires a constantly evolving set of complementary skills among those actors who implement the technology at the local level (cf. Dosi, 1988; Dosi and Nelson, 2013). In the case of PV, knowledge related to the physical

learning. It is also clear that interaction and a collective long-term process of awareness-raising and meaning ascription served to create the legitimacy of solar energy in Germany (Jacobsson and Lauber, 2006; Dewald and Truffer, 2011, 2012; Strupeit 2016). Since local solar initiatives and chambers of commerce played an important role in this process, it is likely that this aspect of learning was predominantly local (though knowledge could also flow between localities), and that it was initially to some extent sticky and reliant on dense and overlapping local social networks enabling trust and informal coordination, as evidenced by Dewald and Truffer's (2012) and Rode and Weber's (2016) analysis of uneven local market formation in Germany. Due to the significance of proximity in (local) learning we sought in the literature a discussion on local governmental policy implications, but with limited results. Policy analyses for the deployment of PV have primarily focused on schemes of financial support such as taxes, subsidies, feed-on tariffs and tradable permits (albeit capacity development is highlighted in studies on rural PV electrification policy in developing countries, see Miller and Hope, 2000; Ulsrud et al., 2011, 2015). Although economic instruments have created conditions and opportunities for local learning and the formation of legitimacy for new energy technologies, they were not targeted at specifically supporting local learning. Detailed analyses of the emergence of PV competencies in local and national contexts have highlighted the role of demonstration programmes (e.g. Brown and Hendry, 2009; Bossink, 2015; Heiskanen et al., 2015). They have also shown how early PV advocates have struggled to create comprehensive and coordinated networks (van Mierlo, 2012; Verhees et al., 2013; Smith et al., 2014). The lack of access to specialized competencies and business networks is also highlighted in international comparison of emerging to mature PV clusters (Jaegersberg and Ure, 2011). However, the policy literature has paid limited attention to the role of policy incentives that can foster knowledge and training in localized know-how, enhance craft and practical skills, and support interactive learning and the formation of local knowledge networks. Yet, increasing evidence of spatially uneven diffusion patterns suggests that local policies might be important (Noll et al., 2014; Li and Yi, 2014; Graziano and Gillingham, 2015; Schaffer and Brun, 2015). Shum and Watanabe (2009) emphasize a need for broader policy support covering the numerous actors in the PV value chain that are involved in and could benefit from local learning and Jaegersberg and Ure (2011) show that lack of such policy support and representation is a central problem for SMEs in emerging and struggling PV value chains. 4. Discussion The literature review on local learning in economic geography and economics of innovation brings to light a number of aspects of local learning that are relevant for advancing our knowledge on the deployment of new energy technologies. The review on solar PV deployment, on the other hand, shows that the literature on the deployment of PV is fragmented and that only a few studies explicitly focus on the processes of learning for solar PV deployment. Nevertheless, the reviews contribute important insights. First, the literature review on processes of learning for PV deployment shows aspects of deployment that are distinctively local and will require local learning that have not been emphasized in the geography of innovation literature before. Some of these pertain to the physical and historical infrastructure and to historical patterns of governance and planning of the built environment, where PV systems are predominantly deployed. This natural and historical geographical fragmentation of the physical and institutional environments in which technologies are deployed is an aspect that is not discussed in the technological learning literature that we reviewed, and appears to be characteristic to the deployment process. Further research could investigate this in more detail, addressing not only localized building 280

Energy Policy 101 (2017) 274–283

L. Neij et al.

(local) learning, but we also identify several examples of knowledge being codified over time and transferred to global and national flows of knowledge. Based on the significance of proximity in local learning we stress the importance of further analysing how innovation policy (and energy policy) could better support the local adaptation of new technologies to their surroundings and support local capacities to absorb external knowledge. The paper further lends support to tentative policy implications and provide guiding principles for local learning in PV deployment. First, the paper shows that local learning processes build up important knowledge on how to integrate new distributed technologies in the built environment. Support of such processes falls quite naturally to local governments, which have responsibility for spatial and urban planning. Second, in emerging markets, such local processes may also be seen as ways for more codified knowledge on the national level to be created ‘from the bottom up’. For this reason, local government programmes and processes may be worthy of national and pan-national innovation and energy policy support. Third, local learning occurs in interaction among several actors involved in the PV deployment value chain or supporting it. Explicit policy support for interaction, e.g., the development of networks, relationships and trust that support the flow of local and non-local knowledge, may be merited in order to support learning. Fourth, demonstration programmes have the potential to support local learning and build local networks – demonstrations should thus be included in the policy mix, particularly in the case of emerging PV markets.

and historical infrastructure as well as to historically determined patterns of governance and planning will most probably rely on local learning. In the literature we found that many of the (local) learning processes in Germany were developed and supported by local and national professional networks. The creation of such networks seems to rely on an (local) trust and shared meaning and narratives about PV (Dewald and Truffer, 2012; Strupeit, 2016; Mattes et al., 2014; Karakaya et al., 2015). These networks resulted in easily accessible information, educational activities, the advancement of technologies, institutional routines and standards. The literature clearly indicates that these networks have been essential for the successful deployment of PV. Based on the fact that we find little evidence in the literature on the formation of stable professional networks for the deployment of PV in other countries, we ask ourselves if the creation of local networks can benefit from governmental support, in terms of local programmes, local dialogue initiatives or demonstration projects. Such networks could interact in the formation of new (global and national) artefacts, routines and educational activities, but also support the absorption of external knowledge. Moreover, governments might support a local communication infrastructure and provide local education and training based on (and perhaps even improving on) local experiences and practice. Further research should establish at which point such networking is most helpful: whether it requires a certain degree of underlying social mobilization as well as a certain national framework creating reliable market signals (as has been the case in Germany). Our review of the literature is based on the existing research on PV deployment available at the present time. With a greater number and depth of country and local case studies focusing explicitly on learning processes, similarities and differences among learning patterns in deployment processes, and the temporal evolution of local learning processes might be better observed and analysed. With more and better empirical material, it might also be possible to consider extending existing economic research on local learning (Asheim, 1996; Asheim and Isaksen, 2002; Camagni, 1991; Cooke et al., 1997; Lorenzen, 2007; Maskell and Malmberg, 1999) to identify distinctive characteristics that relate to deployment. Our current paper merely hypothesizes that geographical fragmentation of the physical and institutional environments in which technologies are deployed, as well as the involvement of a wide range of non-specialized actors in the deployment process (implying an important role for tacit and synthetic knowledge) might be factors that makes local learning even more important for deployment processes than for the development and production of new technologies. Further research is needed to verify or falsify these hypotheses.

Acknowledgements The authors are very grateful to the Swedish Energy Agency and the Academy of Finland (grant 293405) for financial support. Appendix A. Supporting information Supplementary data associated with this article can be found in the online version at doi:10.1016/j.enpol.2016.11.029. References Asheim, B.T., 1996. Industrial districts as “learning regions”: a condition for prosperity? Eur. Plan. Stud. 4 (4), 379–400. Asheim, B.T., 2012. The changing role of learning regions in the globalizing knowledge economy: a theoretical re-examination. Reg. Stud. 46 (8), 993–1004. Asheim, B.T., Coenen, L., Vang, J., 2007. Face-to-face, buzz, and knowledge bases: sociospatial implications for learning, innovation, and innovation policy. Environ. Plan. C 25 (5), 655. Asheim, B.T., Isaksen, A., 1997. Location, agglomeration and innovation: towards regional innovation systems in Norway? Eur. Plan. Stud. 5 (3), 299–330. Asheim, B.T., Isaksen, A., 2002. Regional innovation systems: the integration of local ‘sticky’ and global ‘ubiquitous' knowledge. J. Technol. Transf. 27 (1), 77–86. Asheim, B.T., Lawton Smith, H., Oughton, C., 2011. Regional innovation systems: theory, empirics and policy. Reg. Stud. 45, 875–891. Audretsch, D.B., Feldman, M.P., 2004. Knowledge spillovers and the geography of innovation. Handb. Reg. Urban Econ. 4 (Cities Geogr.), 2063–3073. Autant-Bernard, C., Fadairo, M., Massard, N., 2013. Knowledge diffusion and innovation policies within the European regions: challenges based on recent empirical evidence. Res. Policy 42 (1), 196–210. Baborska-Narozny, M., Stevenson, F., Ziyada, F.J., 2016. User learning and emerging practices in relation to innovative technologies: a case study of domestic photovoltaic systems in the UK. Energy Res. Soc. Sci. 13, 24–37. Balta-Ozkan, N., Watson, T., Mocca, E., 2015. Spatially uneven development and low carbon transitions: insights from urban and regional planning. Energy Policy 85, 500–510. Barbose, G., Darghouth, N.R., Weaver, S., Feldman, D., Margolis, R., Wiser, R., 2015. Tracking US photovoltaic system prices 1998–2012: a rapidly changing market. Prog. Photovolt.: Res. Appl. 23 (6), 692–704. Bathelt, H., Malmberg, A., Maskell, P., 2004. Clusters and knowledge: local buzz, global pipelines and the process of knowledge creation. Prog. Hum. Geogr. 28 (1), 31–56. Bathke, R., 2009. Installateure profitieren vom direkten Kundenkontakt. Handelsblatt. Binz, C., Truffer, B., Coenen, L., 2014. Why space matters in technological innovation systems—Mapping global knowledge dynamics of membrane bioreactor technology. Res. Policy 43 (1), 138–155. Bollinger, B., Gillingham, K., 2014. Learning-by-Doing in Solar Photovoltaic Installations. Yale University Working Paper.

5. Conclusions and potential policy implications The reviews presented in this paper show that learning for PV deployment exhibits characteristics of local learning identified in the innovation literature: it has a strong tacit component, it is synthetic and adapts new technologies to existing user needs and contexts, and aligns practices and develops shared codes for dividing roles and responsibilities in value chains. It also involves a strong component of developing shared meanings and narratives about PV. Such knowledge and skills develop primarily in local networks of firms, users and other organizations such as solar initiatives. In addition, we show that competencies in the deployment of distributed energy technologies rely on learning processes that are closely connected to the historically and geographically distinctive characteristics of the built environment, and which add a further layer of density to the localized knowledge and skill emerging in the complex value chains and social networks involved in PV deployment. We also identified processes in which local learning becomes codified and standardized on a national level. The study shows that spatial, social and cultural proximity play a role in 281

Energy Policy 101 (2017) 274–283

L. Neij et al.

Environ. 15 (2), 175–185. Lawson, C., Lorenz, E., 1999. Collective learning, tacit knowledge and regional innovative capacity. Reg. Stud. 33 (4), 305–317. Levitt, B., March, J.G., 1988. Organizational learning. Annu. Rev. Sociol., 319–340. Li, H., Yi, H., 2014. Multilevel governance and deployment of solar PV panels in U.S. cities. Energy Policy 69, 19–27. Livi, C., Jeannerat, H., Crevoisier, O., 2014. From regional innovation to multi-local valuation milieus. In The Social Dynamics of Innovation. Networks, 23–41, (Taylor & Francis). Lorenzen, M., 2007. Social Capital and Localised Learning: Proximity and Place in Technological and Institutional Dynamics. Urban Stud. 44 (4), 799–817. Lundvall, B.-Å., 1988. Innovation as an interactive process - from User-Producer Interaction to the National System of Innovation. In: Dosi, G., Freeman, C., Nelson, R., Silverberg, G., Soete, L. (Eds.), Technical Change and Economic Theory. Pinter Publishers, London. Lundvall, B.-Å (Ed.), 1992. National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning. Pinter Publisher, London. Lundvall, B.Å., Borrás, S., 1997. The globalising learning economy: implications for innovation policy. Rep. DG XII, Comm. Eur. Union, 34–39. Lundvall, B.-Å., Johnson, B., 1994. The learning economy. J. Ind. Stud. 1, 23–42. Malmberg, A., Maskell, P., 2002. The elusive concept of localization economies: towards a knowledge-based theory of spatial clustering. Environ. Plan. A 34 (3), 429–450. Maskell, P., Malmberg, A., 1999. The competitiveness of firms and regions –”Ubiquitification” and the importance of localized learning. Eur. Urban Reg. Stud. 6, 9–25. Mattes, J., Huber, A., Koehrsen, J., 2014. Energy transitions in small-scale regions–What we can learn from a regional innovation systems perspective. Energy Policy 78, 255–264. McCauley, S.M., Stephens, J.C., 2012. Green energy clusters and socio-technical transitions: analysis of a sustainable energy cluster for regional economic development in Central Massachusetts, USA. Sustain. Sci. 7 (2), 213–225. Mignon, I., Bergek, A., 2016. System-and actor-level challenges for diffusion of renewable electricity technologies: an international comparison. J. Clean. Prod. 128, 105–115. Miller, D., Hope, C., 2000. Learning to lend for off-grid solar power: policy lessons from World Bank loans to India, Indonesia, and Sri Lanka. Energy Policy 28 (2), 87–105. Morgan, K., 2007. The learning region: institutions, innovation and regional renewal. Reg. Stud. 41, 147–159. Müller, S., Rode, J., 2013. The adoption of photovoltaic systems in Wiesbaden, Germany. Econ. Innov. New Technol. 22 (5), 519–535. Nagamatsu, A., Watanabe, C., Shum, K.L., 2006. Diffusion trajectory of self-propagating innovations interacting with institutions—incorporation of multi-factors learning function to model PV diffusion in Japan. Energy Policy 34 (4), 411–421. National Innovation Systems. A Comparative. In: Nelson, R. (Ed.), Analysis. Oxford University Press, New York/Oxford. Nelson, R.R., Winter, S.G., 1977. In search of a useful theory of innovation. In Innovation, Economic Change and Technology Policies, 215-245. Birkhäuser Basel. Nohria, N., Eccles, R., 1992. Face-to-face: making network organizations work. In: Nohria, N., Eccles, R. (Eds.), Networks and organizations: Structure, form, and action, (. Harvard Business School Press, 288–308. Noll, D., Dawes, C., Rai, V., 2014. Solar Community Organizations and active peer effects in the adoption of residential PV. Energy Policy 67, 330–343. Overholm, H., 2015. Collectively created opportunities in emerging ecosystems: the case of solar service ventures. Technovation 39, 14–25. Palit, D., 2013. Solar energy programs for rural electrification: experiences and lessons from South Asia. Energy Sustain. Dev. 17 (3), 270–279. Palm, A., 2014. An emerging innovation system for deployment of building-sited solar photovoltaics in Sweden. Environ. Innov. Soc. Transit. 15, 140–157. Palm, A., 2016. Local factors driving the diffusion of solar photovoltaics in Sweden: a case study of five municipalities in an early market. Energy Res. Soc. Sci. 14, 1–12. Podewils, C., 2009. Noch kl emmen die Kredit. Nicht.: für Den. Bau Von. Kleina. und für gute Gross. verleihen die Banke. nach wie vor Geld. Photon Feb, 92–97. Polanyi, M., 1983. The tacit dimension, Gloucester. Mass., Peter Smith, 1983. PolanyiM., 1996/1997. The TacitDimension. Routledge and Kegan, London. Rai, V., Reeves, D.C., Margolis, R., 2016. Overcoming barriers and uncertainties in the adoption of residential solar PV. Renew. Energy 89, 498–505. Rai, V., Robinson, S.A., 2013. Effective information channels for reducing costs of environmentally-friendly technologies: evidence from residential PV markets. Environ. Res. Lett. 8 (1), 014044. RALsolar, 2013. RAL Gütezeichen Solar [WWW Document]. RAL Gütegemeinschaft Solarenergieanlagen e.V. URL〈http://www.ralsolar.de〉 (accessed 11.15.13). Rode, J., Weber, A., 2016. Does localized imitation drive technology adoption? A case study on rooftop photovoltaic systems in Germany. J. Environ. Econ. Manag. 78, 38–48. Schaffer, A.J., Brun, S., 2015. Beyond the sun—Socioeconomic drivers of the adoption of small-scale photovoltaic installations in Germany. Energy Res. Soc. Sci. 10, 220–227. Seel, J., Barbose, G.L., Wiser, R.H., 2014. An analysis of residential PV system price differences between the United States and Germany. Energy Policy 69, 216–226. Shum, K.L., Watanabe, C., 2007. Towards an institutions-theoretic framework comparing solar photovoltaic diffusion patterns in Japan and the United States. Int. J. Innov. Manag. 11 (04), 565–592. Shum, K.L., Watanabe, C., 2008. Towards a local learning (innovation) model of solar photovoltaic deployment. Energy Policy 36 (2), 508–521. Shum, K.L., Watanabe, C., 2009. An innovation management approach for renewable energy deployment—the case of solar photovoltaic (PV) technology. Energy Policy 37

Bossink, B.A.G., 2015. Demonstration projects for diffusion of clean technological innovation: a review. Clean. Technol. Environ. Policy 17, 1409–1427. Brown, J., Hendry, C., 2009. Public demonstration projects and field trials: accelerating commercialisation of sustainable technology in solar photovoltaics. Energy Policy 37, 2560–2573. Bruns, E., Ohlhorst, D., Wenzel, B., Köppel, J., 2011. Renewable Energies in Germany's Electricity Market: a Biography of the InnovationProcess. Springer, Dordrecht, Netherlands. BSW-Solar and ZVEH, 2013. Photovoltaik Anlagenpass [www Document]. Photovoltaik Anlagenpass. URL〈http://www.photovoltaik-anlagenpass.de/〉 (accessed 11.15.13). Bulkeley, H., Castán Broto, V., 2013. Government by experiment? Global cities and the governing of climate change. Trans. Inst. Br. Geogr. 38 (3), 361–375. Burkhardt, J., Wiser, R., Darghouth, N., Dong, C.G., Huneycutt, J., 2015. Exploring the impact of permitting and local regulatory processes on residential solar prices in the United States. Energy Policy 78, 102–112. Camagni, R., 1991. Innovation networks: spatialPerspectives. John Wiley & Sons, Inc, London. Camagni, R., 2002. On the concept of territorial competitiveness: sound or misleading? Urban Stud. 39 (13), 2395–2411. Cimoli, M., Dosi, G., Nelson, R., Stiglitz, J.E., 2006. Institutions and policies shaping industrial development: An introductory note. Scuola Superiori Sant’Anna: LEM Working Paper Series 2006/02. (Online: 〈http://www.sssup.it/UploadDocs/5672_ 2006_02.pdf〉) Cohen, M.D., Sproull, P. (Eds.), 1996. Organisational learning. Sage, London. Colatat, P.C., 2009. Photovoltaic Systems, the experience curve, and learning by doing: who is learning and what are they doing?(Doctoral dissertation, Massachusetts Institute of Technology). Cooke, P., Uranga, M.G., Etxebarria, G., 1997. Regional innovation systems: institutional and organisational dimensions. Res. Policy 26 (4), 475–491. Dewald, U., Truffer, B., 2011. Market Formation in Technological Innovation Systems— Diffusion of Photovoltaic Applications in Germany. Ind. Innov. 18, 285–300. Dewald, U., Truffer, B., 2012. The local sources of market formation: explaining regional growth differentials in German photovoltaic markets. Eur. Plan. Stud. 20 (3), 397–420. DIN, 2005a. DIN 1055 Einwirkungen auf Tragwerke: Teil 5 Schnee- und Eislasten. DIN, 2005b. DIN 1055 Einwirkungen auf Tragwerke: Teil 4 Windlasten. Dong, C., Wiser, R., 2013. The impact of city-level permitting processes on residential photovoltaic installation prices and development times: an empirical analysis of solar systems in California cities. Energy Policy 63, 531–542. Dosi, G., 1988. Sources, procedures and microeconomic effects of innovation. J. Econ. Lit. 26, 1120–1171. Dosi, G., Nelson, R.R., 2013. Evol. Technol.: Assess. State–Art. Eurasia. Bus. Rev. 3 (1), 3–46. Fabrizio, K.R., Hawn, O., 2013. Enabling diffusion: how complementary inputs moderate the response to environmental policy. Res. Policy 42 (5), 1099–1111. Florida, R., 1995. Toward the learning region. Futures 27, 527–536. FreemanC., 1992. Technology Policy and Economic Performance: Lessons from Japan. Frances Pinter, London. Gertler, M.S., 2003. Tacit knowledge and economic geography of context, or The undefinable tacitness of being (there). J. Econ. Geogr. 3, 75–99. Graziano, M., Gillingham, K., 2015. Spatial patterns of solar photovoltaic system adoption: the influence of neighbors and the built environment. J. Econ. Geogr. 15, 815–839. Heiskanen, E., Jalas, M., Rinkinen, J., Tainio, P., 2014. The local community as “lowcarbon lab”: Promises and perils. Environ. Innov. Soc. Transit. 14, 149–164. Heiskanen, E., Nissilä, H., Lovio, R., 2015. Demonstration buildings as protected spaces for emerging sustainable solutions – the case of solar building integration in Finland. Forthcoming inJournal of Cleaner Production. Hodson, M., Marvin, S., 2010. Can cities shape socio-technical transitions and how would we know if they were? Res. Policy 39 (4), 477–485. IEC, 2004. International standard IEC 61730-1: photovoltaic (PV) module safety construction - Part 1: requirements for construction. Ed. 1. 0, 2004–2010. IEC, 2005. International Standard IEC 61215: Crystalline silicon terrestrial photovoltaic (PV) modules – Design qualification and type approval. Second edition 2005-04. IEC, 2008. International Standard IEC 61646: Thin-film terrestrial photovoltaic (PV) modules - Design qualification and type approval. Edition 2.0, 2008-05. Jacobsson, S., Lauber, V., 2006. The politics and policy of energy system transformation—explaining the German diffusion of renewable energy technology. Energy Policy 34 (3), 256–276. Jaegersberg, G., Ure, J., 2011. Barriers to knowledge sharing and stakeholder alignment in solar energy clusters: learning from other sectors and regions. J. Strateg. Inf. Syst. 20 (4), 343–354. Jensen, M.B., Johnson, B., Lorenz, E., Lundvall, B.Å., 2007. Forms of knowledge and modes of innovation. Res. Policy 36 (5), 680–693. Kanters, J.Horvat, M., 2012. Solar energy as a design parameter in urban environments. Proceedings from SHC 2012: International Conference on Solar Heating and Cooling for Buildings and Industry, San Francisco. Karakaya, E., Hidalgo, A., Nuur, C., 2015. Motivators for adoption of photovoltaic systems at grid parity: a case study from Southern Germany. Renew. Sustain. Energy Rev. 43, 1090–1098. Karakaya, E., Nuur, C., Hidalgo, A., 2016. Business model challenge: lessons from a local solar company. Renew. Energy 85, 1026–1035. Keeble, D., Lawson, C., Moore, B., Wilkinson, F., 1999. Collective learning processes, networking and ‘institutional thickness' in the Cambridge region. Reg. Stud. 33 (4), 319–332. Langniß, O., Neij, L., 2004. National and international learning with wind power. Energy

282

Energy Policy 101 (2017) 274–283

L. Neij et al.

accelerating access to electricity services through a socio-technical design in Kenya. Energy Res. Soc. Sci. 5, 34–44. Ulsrud, K., Winther, T., Palit, D., Rohracher, H., Sandgren, J., 2011. The solar transitions research on solar mini-grids in India: learning from local cases of innovative sociotechnical systems. Energy Sustain. Dev. 15 (3), 293–303. van Mierlo, B., 2012. Convergent and divergent learning in photovoltaic pilot projects and subsequent niche development. Sustain.: Sci., Pract. Policy 8 (2), 4–18. VDE, 2011. VDE Institute certifies photovoltaic modules using new test procedure [WWW Document]. URL 〈http://www.vde.com/en/Institute/News/Documents/1137_PI_Intersolar_PV_eng.pdf〉 (accessed 11.22.11). Verhees, B., Raven, R., Veraart, F., Smith, A., Kern, F., 2013. The development of solar PV in The Netherlands: a case of survival in unfriendly contexts. Renew. Sustain. Energy Rev. 19, 275–289. Wall, M., Probst, M.C.M., Roecker, C., Dubois, M.C., Horvat, M., Jørgensen, O.B., Kappel, K., 2012. Achieving solar energy in architecture-IEA SHC Task 41. Energy Procedia 30, 1250–1260. Zhang, X., Shen, L., Chan, S.Y., 2012. The diffusion of solar energy use in HK: What are the barriers? Energy Policy 41, 241, (24).

(9), 3535–3544. Sigrin, B., Pless, J., Drury, E., 2015. Diffusion into new markets: evolving customer segments in the solar photovoltaics market. Environ. Res. Lett. 10 (8), 084001. Simmie, J. (Ed.), 1997. Innovation, Networks and Learning Regions?. Jessica Kingsley, London. Smith, A., Kern, F., Raven, R., Verhees, B., 2014. Spaces for sustainable innovation: solar photovoltaic electricity in the UK. Technol. Forecast. Soc. Change 81, 115–130. Solarpraxis, 2012. PV+Test: Independent Photovoltaic Module Test [WWW Document]. URL 〈http://www.pvtest.de/index_en.html〉 (accessed 3.13.12). Sørensen, K., 2013. Beyond innovation. Towards an extended framework for analysing technology policy. Nord. J. Sci. Technol. 1, 12–23. Stroper, M., Venables, A.J., 2004. Buzz: face-to-face contact and the urban economy. J. Econ. Geogr. 4, 351–370. Strupeit L. An innovation system perspective on the drivers of cost reduction for emerging energy technologies: the case of photovoltaic deployment in Germany, IIIEE working paper, 2016 Strupeit, L., Neij, L., 2016. Cost dynamics in the deployment of photovoltaics. Renew. Sustain. Energy Rev., (in press). Ulsrud, K., Winther, T., Palit, D., Rohracher, H., 2015. Village-level solar power in Africa:

283