848 KN OW LEDGE ECOSYSTEM FORM ATION : AN IN STITUTION AL AN D ORGAN ISATION AL PERSPECTIVE Argyro Alm panopoulou KNOWLEDGE ECOSYSTEM FORMATION: AN INSTITUTIONAL AND ORGANISATIONAL PERSPECTIVE Argyro Almpanopoulou ACTA UNIVERSITATIS LAPPEENRANTAENSIS 848 Argyro Almpanopoulou KNOWLEDGE ECOSYSTEM FORMATION: AN INSTITUTIONAL AND ORGANISATIONAL PERSPECTIVE Acta Universitatis Lappeenrantaensis 848 Dissertation for the degree of Doctor of Science (Economics and Business Administration) to be presented with due permission for public examination and criticism in the Auditorium of the Student Union House at Lappeenranta- Lahti University of Technology LUT, Lappeenranta, Finland on the 27th of April, 2019, at noon. Supervisors Professor Paavo Ritala LUT School of Business and Management Lappeenranta-Lahti University of Technology LUT Finland PhD Kati Järvi Talent Vectia OY and Hanken School of Economics Finland Reviewers Associate Professor Llewellyn D.W. Thomas Ramon Llull University Spain Associate Professor Leena Aarikka-Stenroos Tampere University Finland Opponent Associate Professor Llewellyn D.W. Thomas Ramon Llull University Spain ISBN 978-952-335-354-1 ISBN 978-952-335-355-8 (PDF) ISSN-L 1456-4491 ISSN 1456-4491 Lappeenranta-Lahti University of Technology LUT LUT University Press 2019 Abstract Argyro Almpanopoulou Knowledge Ecosystem Formation: An Institutional and Organisational Perspective Lappeenranta 2019 108 pages Acta Universitatis Lappeenrantaensis 848 Diss. Lappeenranta-Lahti University of Technology LUT ISBN 978-952-335-354-1, ISBN 978-952-335-355-8 (PDF), ISSN-L 1456-4491, ISSN 1456-4491 The main purpose of this thesis is to provide a better understanding on knowledge ecosystem formation, that is, the factors and processes that result in the emergence of knowledge ecosystems and their particular forms of organisation. I mainly focus on the particularities of the institutional environment and its related institutions as well as the intermediate processes of identifying joint interests and developing a sense of a shared purpose. In this endeavour, I adopted institutional and organisational lenses and took a holistic, multi-level approach. Drawing on two qualitative studies, this thesis shows how the institutional environment can constrain, enable, and orient knowledge ecosystem formation and how the nature of the joint exploratory processes has implications for the organisation of knowledge ecosystems. First, this thesis contributes to the nascent knowledge ecosystem literature by providing a comprehensive understanding of the early stages of knowledge ecosystem formation and the influence of the socio-economic context during the pre-formation phase as well as of the organisation of ecosystems in such contexts, where hierarchical organisational forms are not feasible. My findings also contribute to the literature on non-traditional forms of organisation by advancing our understanding on the organising particularities of knowledge ecosystems as a type of meta-organisation. In addition, this thesis contributes to the institutional literature and recent calls to understand field transformation and related constraining forces by identifying a variety of field-sustaining mechanisms mutually reinforcing each other. For policymakers, this thesis provides insights on how public policy can support knowledge ecosystem formation. For practitioners, the findings provide suggestions on how to navigate the complexities and ambiguities of the early phases of knowledge ecosystem formation. Keywords: formation, institutional environment, knowledge ecosystem, meta- organisation, organisation Acknowledgements Thank you, Llewellyn DW Thomas, for your exceptionally constructive comments that helped me finalize the dissertation and for being my opponent. Thank you, Leena Aarikka-Stenroos, for your extremely valuable comments that helped me finalize the dissertation. Thank you, Kati Järvi and Paavo Ritala, for being my mentors. Thank you, my co-authors, for your input and collaboration on the publications. Thank you, Kirsimarja Blomqvist, for believing in me and pushing me out of my comfort zone. I will always remember the support you have given me. Thank you, DDI team, Martti Mäntylä, Kimmo Karhu and Suvi Lavinto, for all the insightful discussions and great collaboration. Thank you, my friends and colleagues at LUT University, for the peer-support and learning you provided all these years. Thank you, Business Finland, Academy of Finland, and Foundation for Economic Education, for providing the financial support. Thank you, Henna Järvi, for being my sydänystävä, for being there for me when I needed it the most. Thank you (Σας ευχαριστώ), mom and dad (Μαμά και μπαμπά), for your unconditional love and support, for believing in me and allowing me to follow the paths that I chose. (για την άνευ όρων αγάπη και υποστήριξη σας, για το οτι πιστέψατε σε μένα και μου επιτρέψατε να ακολουθήσω τα μονοπάτια που επέλεξα.) Thank you, Niko, for all the little things you do to make my life easier, for your patience, understanding and love. (για όλα τα μικρά πράγματα που κάνεις για να κάνεις τη ζωή μου πιο εύκολη, για την υπομονή σου, την κατανόησή σου και την αγάπη σου.) Thank you, Eleni, for being my sister, my ally and my best friend. I will always remember and cherish our life together. I wish you were here now. I miss you. I love you. Thank you, My son Georgios, for keeping me grounded. You are the love of my life. Argyro Almpanopoulou April 2019 Lappeenranta, Finland To Eleni My little sister that left too soon. Your smile, your sensitive heart, your strength, courage and fight for life will be ingrained in me forever. I miss you. I love you. Contents Abstract 3 Acknowledgements 5 Contents 9 List of publications 11 1 Introduction 13 1.1 Overview of the research gaps ................................................................ 15 1.2 Recognising research opportunities & constructing research questions . 17 1.2.1 Research opportunities ................................................................ 17 1.2.2 Research questions ...................................................................... 20 1.2.3 Positioning the research .............................................................. 22 1.3 Glossary ................................................................................................... 23 1.4 Organisation of the dissertation ............................................................... 25 2 Conceptual and theoretical background 27 2.1 The ecosystem concept ............................................................................ 27 2.1.1 Ecosystem concept in management ............................................ 28 2.1.2 Ecosystem-as-structure vs ecosystem-as-affiliation ................... 29 2.1.3 Ecosystem (both views) and other related concepts ................... 32 2.1.4 Knowledge ecosystems ............................................................... 35 2.2 Knowledge ecosystem formation: Lenses & sensitising concepts .......... 36 2.2.1 The institutional lens: An institutional approach to ecosystems . 38 2.2.2 The organisational lens: Ecosystem as a form of organisation ... 40 3 Methodology 45 3.1 Theoretical paradigm and rationale for the study .................................... 45 3.2 Research design ....................................................................................... 46 3.2.1 Research strategy ........................................................................ 46 3.2.2 Progressive focusing of my inquiry ............................................ 47 3.2.3 Methodological choices for individual studies ........................... 53 3.3 Trustworthiness criteria for my inquiry ................................................... 57 4 Publications and synopsis of the findings 61 4.1 Publication I – In defense of ‘eco’ in innovation ecosystems ................. 61 4.2 Publication II – Organisation of knowledge ecosystems: Prefigurative and partial forms ........................................................................................................ 62 4.3 Publication III – Innovation ecosystem emergence barriers: Institutional perspective ............................................................................................... 64 4.4 Publication IV – Emergence of energy services ecosystems: Scenario method as a policy enabler ....................................................................................... 65 5 Conclusions 69 5.1 Answering my research questions ........................................................... 69 5.2 Theoretical implications .......................................................................... 78 5.2.1 Knowledge ecosystem literature ................................................. 78 5.2.2 Literature on non-traditional forms of organisation .................... 80 5.2.3 Institutional literature .................................................................. 80 5.3 Policy and practice implications .............................................................. 81 5.4 Limitations and future research ............................................................... 82 References 85 Appendix A 105 Publications 11 List of publications This dissertation is based on the following papers. The rights have been granted by the publishers to include the papers in this dissertation. I. Ritala, P. and Almpanopoulou, A., 2017. In defense of ‘eco’ in innovation ecosystem. Technovation, 60–61(February), pp.39–42. II. Järvi, K., Almpanopoulou, A. and Ritala, P., 2018. Organization of knowledge ecosystems: Prefigurative and partial forms. Research Policy, 47(8), pp.1523– 1537. III. Almpanopoulou, A., Ritala, P. and Blomqvist, K., 2019. Innovation ecosystem emergence barriers: Institutional perspective. In: Proceedings of the 52nd Hawaii International Conference on System Sciences, pp.6357–6366. IV. Almpanopoulou, A., Bergman, J-P., Ahonen, T., Blomqvist, K., Ritala, P., Honkapuro, S. and Ahola, J., 2017. Emergence of energy services ecosystems: Scenario method as a policy enabler. Journal of Innovation Management, 5(1), pp.58–77. Authors’ contribution Paper I: Both authors contributed equally in writing and revising the paper. Paper II: K. J. and I jointly designed the study as well as collected and analysed the data. P.R. provided critical input during those stages. All authors equally wrote the first draft and revised and rewrote the manuscript. Paper III: I am the principal author and investigator. I wrote most of the first draft of the paper. I was mainly responsible for designing the study as well as collecting and analysing the data. P.R and K.B. provided input primarily in the theory and contributions chapters. All authors jointly revised and rewrote the manuscript. Paper IV: I am the principal author and investigator. I wrote most of the first draft of the paper. J-P. B. provided input in the scenario method chapter. T.A., S.H., and J.A. provided input in two of the case illustrations, and K.B., P.R. provided critical input. I was responsible for revising and rewriting the manuscript. 13 1 Introduction Ecosystems – why should we care? Why is it important for modern organisations to build or belong to an ecosystem or even several ecosystems? What is so distinctive about ecosystems, and why should managers, policymakers, and academics understand their peculiarities? In modern society, where everything is highly interconnected and where problems are universal, highly complex, and urgent, solutions cannot be found in silos, by single organisations, industries, or disciplines. The complexities of this era call for expertise and capabilities that no single organisation holds or is capable of developing on its own. Thus, single organisations often struggle to keep up or are simply not prepared to respond to the uncertainties and challenges of today’s settings (Furr and Shipirov, 2018), which renders them interdependent with others and their environment. Ecosystems allow a diverse set of actors (individual and organisational) to come together in the search for solutions, creating knowledge, and defining and commercialising new offerings (e.g. Järvi, Almpanopoulou and Ritala, 2018; Dattée, Alexy and Autio, 2018). James Moore (1993) introduced the ecosystem concept in the business and management field. Specifically, Moore (1993, 1996) adopted the biological metaphor of ‘ecosystem’ to describe how organisations and individuals interact and evolve in systems, operating similarly to those we can witness in nature. The key insights – which were developed by other authors later on – were built on the systemic nature of ecosystems, including the principle of a shared environment, co-evolution, interdependence, and ecosystem leadership (e.g. Moore, 1993; Iansiti and Levien, 2004; Autio and Thomas, 2014; Aarikka-Stenroos and Ritala, 2017). However, while the business ecosystem literature has been cumulating for decades, there is very little consensus on the usefulness and definition of the term itself. In fact, the ecosystem term has heavily dispersed to more specific viewpoints besides generic reference to business, resulting in a plurality of definitions1 and prefixes, such as innovation ecosystems (e.g. Adner, 2006; Adner and Kapoor, 2010), platform ecosystems (e.g. Ceccagnoli, et al., 2012), service ecosystems (e.g. Akaka, Vargo and Lusch, 2013), knowledge ecosystems (Clarysse, et al., 2014), and, recently, entrepreneurial ecosystems (e.g. Autio, et al., 2018). In this thesis, I particularly focus on knowledge ecosystems. Knowledge ecosystems have a narrow scope on early knowledge creation and search. In this type of ecosystem, multiple actors join forces to create new knowledge in a pre-competitive setting (van der Borgh, Cloodt and Romme, 2012; Clarysse, et al., 2014; Valkokari, 2015). Therefore, activities such as exploitation and commercialisation are not the focal point (Valkokari, 2015). Prior literature has mostly described knowledge ecosystems as geographically co- located hotspots steered by either universities (Clarysse, et al., 2014) or firms (van der Borgh, Cloodt and Romme, 2012) that focus on a collaborative knowledge search (Valkokari, 2015). Knowledge ecosystems might form around particular technological or societal issues (Dougherty and Dunne, 2011) or form to address a number of scientific problems, which can progressively lead to knowledge exploitation and actor-specific 1 See Table 3 for definitions of the mentioned constructs. 1 Introduction 14 appropriation (Franzoni and Sauermann, 2014; Perkmann and Schildt, 2015). Building on these conceptualisations, in Järvi, Almpanopoulou, and Ritala (2018, p.1524) (publication II), we view knowledge ecosystems as ‘organisations comprising diverse actors bound together by a joint search for valuable knowledge while having independent agency also beyond the knowledge ecosystem’. In this conceptualisation, the role of co- location is not highlighted due to the capabilities of technology to facilitate collaborative knowledge work across geographical distances (cf. Still, et al., 2014). Knowledge ecosystems might be the key to resolve large-scale scientific and societal problems as well as promote growth and social welfare as successful collaboration in such collectivities might lead to unprecedented scientific breakthroughs. The ATLAS project at CERN (European Organisation for Nuclear Research) is an example of an extraordinarily large-scale and complex knowledge ecosystem involving approximately 3000 experts from 181 independent institutions globally, aiming to develop a new particle detector to discover the Higgs boson (see Tuertscher, Garud and Kumaraswamy, 2014; Atlas.cern, 2019). At smaller scales, policymakers in many countries also increasingly recognise the potential of knowledge ecosystems for regional and national growth as well as for boosting competitiveness and accelerating innovation (Ritala and Gustafsson, 2018). For example, in Australia and Finland, publicly funded cooperative research centres were established with the goal to pursue and implement collaborative research programmes between industry and academic institutions (see Halme, et al., 2014; Sinnewe, Charles and Keast, 2016). Nevertheless, sustaining long-term collaboration among actors in such initiatives is not an easy task. In fact, problems related to e.g. resourcing, agenda setting, monitoring contributions, and evaluating performance often arise and can impede the collaborative efforts (see e.g. Sinnewe, Charles and Keast, 2016). Considering the impact knowledge ecosystems such as the examples above might have for science, innovation, and society in general, it is worthwhile to understand more about the processes and factors that can support forming and sustaining such configurations of multiple actors with diverse backgrounds and expertise. This thesis provides a better understanding on knowledge ecosystem formation, that is, the factors and processes that result in the emergence of knowledge ecosystems and their particular forms of organisation (see also Ebers, 1997). I particularly focus on the particularities of the institutional environment and its related institutions as well as the intermediate processes of identifying joint interests and developing a sense of a shared purpose. Specifically, my findings demonstrate how the institutional factors influence knowledge ecosystem formation and how the nature of these intermediate processes shape the organisation of knowledge ecosystems. At a practical level, understanding those factors and processes allows policymakers and practitioners to recognise the issues that can hamper the formation of ecosystems and be prepared for what to expect in their innovation endeavours. In addition, as described at the beginning of this section, the complexities and uncertainties of the current fast-moving era call for key actors to join forces for collective strategic action, and although self-organising is an inherent characteristic of ecosystems (Peltoniemi, 2006), my study shows that policy interventions might be essential for facilitating these processes (see also Clarysse, et al., 2014). Finally, 15 my findings help in recognising the organising requirements and challenges of reaching (and maintaining) a joint understanding and shared vision among ecosystem actors and inform practitioners about the necessary management capabilities to navigate through these challenges. In the next section, I provide an overview of the research gaps that this thesis aims to help fill. 1.1 Overview of the research gaps Research gap 1 – Concept The meaning and applicability of the ecosystem concept has been a cause of debate, especially among academics (see e.g. Oh, et al., 2016; Tsujimoto, et al., 2017), and the multiplicity of prefixes applied to the concept in the business and management field has created more confusion and even doubt over the value of the concept in explaining any inter-organisational phenomena. This lack of consensus related to the meaning, scope, boundaries, and theoretical foundations of the ecosystem concept have led authors to claim that ecosystems are a ‘…faulty analogy to natural ecosystems, and [are] therefore a poor basis for the needed multi-disciplinary research and policies addressing emerging concepts of innovation’ (Oh, et al., 2016, p.1). The knowledge ecosystem is one of the most recent additions to the ecosystem concepts repertoire, and the sceptics might well argue why we need another ecosystem concept and how it compares to other ecosystem concepts. Recent literature has described the meaning of the concept and stated that knowledge ecosystems are different to other types of ecosystems mainly in terms of their focus on early knowledge generation and actors’ connectivity in geographical proximity (e.g. Clarysse, et al., 2014; Valkokari, 2015). Nevertheless, there is still need to unpack the terminology and distil its most important aspects in explaining interdependent knowledge creation and search. In this thesis, I argue for the value of considering knowledge ecosystems separately and attempt to refine our understanding on how the different types of ecosystems relate to one another, particularly regarding the innovation process. Understanding all these aspects contributes to the ongoing scholarly debate on the value and usefulness of the concept and responds to calls for more reflection regarding its conceptual and theoretical underpinnings (see Ritala and Gustafsson, 2018). Research gap 2 – Context and dominant approach Popular and extensively studied examples of ecosystems are those around focal actors, technologies, or platforms, such as Google (Iyer and Davenport, 2008; Rao, 2016) and Cisco (Li, 2009). The business world is full of examples like the above (see also SAP, Intel, IBM, Wikipedia), where focal actors leverage their positions as market leaders and their technological advantage to create ecosystems of diverse actors and accelerate the development of innovation (Ritala and Almpanopoulou, 2017; Baiyere, 2018). In fact, the current scholarly interest in ecosystems has dominantly focused on such top-down organising by focal actors in stable and already established ecosystems in the information and communication technology (ICT) industry, often organised around technological 1 Introduction 16 platforms.2 The very nature of the ICT industry, which is systemic and characterised by intense technological interdependences (Ethiraj and Puranam, 2004; Barnett, 1990), explains the popularity of the ecosystem concept in examining ICT-based collectivities. However, the emphasis on these ecosystems has left relatively unexplored ecosystems in traditional industries, such as energy, for instance. Organisations in such industries might face an utterly different set of challenges that ask for diverse solutions or approaches. For example, evidence from the aerospace industry (see Ritala, et al., 2013) shows that aerospace companies, when compared to ICT firms, have struggled more in developing an ecosystem mindset and thus in managing and facilitating collaboration in ecosystems consisting of multiple partners. An important aspect of my thesis is to address the lack of understanding on the organisation of ecosystems in such contexts, other than the typical ICT collectivities, where organising is not defined by a powerful focal actor or technological platform. Such contexts (see for example the case of the ATLAS project) are usually characterised by increased complexities and uncertainties that make hierarchical organisational forms unfeasible since no single actor possesses the required knowledge to dictate decisions or set the direction (see also Fjeldstad et al. 2012; Tuertscher, Garud and Kumaraswamy, 2014). Consequently, the organising requirements and challenges of such ecosystems are essentially different from the extensively studied examples of actor-centric or platform-based ecosystems; and to reap the benefits of a successful collaboration among multiple diverse actors in these types of ecosystems, it is important to deepen our understanding on how they can be organised. Research gap 3 – Phenomenon of ecosystem formation The focus of attention in the literature taking a top-down, focal actor perspective has primarily been on ecosystem creation, namely on the actions and strategies through which a focal actor constructs its ecosystem. This stream of literature sees focal actors as the ones defining the shared goal in that the ‘system-level goal corresponds to the goals of its architects’ (Gulati, Puranam and Tushman, 2012, p.573). Therefore, creating an ecosystem is ‘a form of endogenous strategic action’ (Autio and Thomas, 2018, p.1), meaning that focal actors can (indirectly) manipulate or shape the perceptions of potential ecosystem stakeholders concerning the value, for example, of their technological platform. Using their network positions, market power, and resources, focal actors can shape or orchestrate the relationships and interdependencies between ecosystem players by managing engagement and connection (Paquin and Howard-Grenville, 2013) as well as membership (Nambisan and Sawhney, 2011) and configuring distinguished roles for ecosystem actors (Williamson and De Meyer, 2012). Further, the ecosystem literature has described ecosystem creation as a systemic process that focal firms must try to control by 2 A platform refers to a coordinating artefact that focal actors utilise, or the technologies, systems, etc. that other ecosystem actors can utilise, to improve their own performance (Cusumano and Gawer, 2002; Iansiti and Levien, 2004; Li, 2009; Tiwana, Konysnski and Bush, 2010; Autio and Thomas, 2014). For example, a platform can refer to an operating system, a cloud service, or a hardware platform. 17 becoming competent in keeping pace with ecosystem dynamics, specifically by road mapping and pre-empting while simultaneously enacting resonance (Dattée, Alexy and Autio, 2018). Alternatively, and particularly in the case of complex knowledge ecosystems, a focal actor might not be identifiable at the early stages of ecosystem formation, and in fact, the birth of an ecosystem relies on a collaborative exploration process and the strategic agency of multiple actors rather than the strategic agency of a focal firm (see also Attour and Barbaroux, 2016). However, we still do not fully understand how these ecosystems are formed and what the institutional factors and specific processes are that influence their formation (see for recent calls Autio and Thomas, 2014; Overholm, 2015; Barile, et al., 2016; Autio, et al., 2018; Möller and Halinen, 2017; Seidel and Greve, 2017). The lack of understanding on these factors and processes that shape ecosystem formation represents an important gap given that the choices and actions of actors looking for opportunities to form or join ecosystems are increasingly influenced by the institutional environment and its institutions, rules, and culture in which they are embedded (see Schreyögg and Sydow, 2011). 1.2 Recognising research opportunities & constructing research questions The process of constructing my (final) research questions did not follow the linear path of (1) identifying research gaps, (2) developing research questions to answer those gaps, and finally (3) designing the research to answer these questions. This thesis is a product of ‘progressive focusing’ (a process I describe in detail in section 3), meaning that I gradually refined my focus and research questions based on unanticipated findings while in the field and a complex iteration of data and theory (see Sinkovics and Alfoldi, 2012). Therefore, a combination of ongoing engagement with the ecosystem literature and my fieldwork led to the recognition of the research opportunities described below and the formulation of my final research questions. 1.2.1 Research opportunities Research opportunity 1 – Taking a holistic and multi-level approach During my ongoing immersion in the ecosystem literature, I came to realise (as also described above in research gaps 2 and 3) that much of the scholarly work has been focal actor-centric since the more you move away from the single actor perspective, the more complex it gets methodologically. The formation of an ecosystem involves multiple dynamic interactions and interdependencies as well as co-evolution among many individual actors and including the wider context (Peltoniemi, 2006; Seidel and Greve, 2017). Considering the complexity and unpredictability of this process (Smith and Stacey, 1997), it becomes questionable whether a single powerful actor can plan, control, and/or design their ecosystem, and consequently investigations taking a single actor, top-down perspective might become problematic to comprehensively understand and embrace the 1 Introduction 18 complex and multifaceted nature of the phenomenon. Alternatively, unlike biological ecosystems, social or organisational ecosystems encompass learning, purposeful action, and efforts by individual actors to influence their ecosystems (Garnsey and Leong, 2008; Garnsey, Lorenzoni and Ferriani, 2008; Valkokari, 2015; Barile, et al., 2016). Thus, there is a need for focus shifts across multiple levels – from individuals, to the organisational level, to the ecosystem level, to the field, and to the wider context (Barile, et al., 2016). By taking into account multiple stakeholder views as well as adopting a multi-level perspective, I attempt to contribute to the knowledge ecosystem literature by providing a comprehensive understanding of the factors and processes that shape the formation of knowledge ecosystems and their particular forms of organisation. Research opportunity 2 – Adopting a temporal lens Along with adopting multiple levels of analysis and empirically considering multiple actors’ perspectives, this thesis adopts a temporal lens, which aims to deepen even further our understanding of knowledge ecosystem formation. This opportunity arose while in the field. I soon recognised that what happens during the time of potential ecosystem emergence (i.e. pre-formation phase) and the specific socio-economic context in which organisational actors are embedded during this time can influence the emergence or non- emergence of an ecosystem. Organisational actors during the pre-formation period face much uncertainty, which can leave certain opportunities unrealised (Kirsch, Moeen and Wadhwani, 2013; Gustafsson, et al., 2016). Previous research on industry emergence has recognised the importance of this period and has primarily documented the existence of the pre-formation stage in cases of ‘successful’ emergence (Agarwal and Bayus, 2002; Golder, Shacham and Mitra, 2009; Kirsch, et al., 2014; Gustafsson, et al., 2016). However, there is still much to learn about the temporal context of complex phenomena, such as ecosystem formation, particularly about how different eras or stages affect these (Kunisch, et al., 2017). Thus, a deeper understanding of the pre-formation phases and the factors that influence ecosystem formation is essential for informing actors that look to direct strategic action and commit resources (Kirsch, et al., 2014). In this thesis, I address this lack of understanding on how the socio-economic context during the pre-formation phase can influence ecosystem formation. Research opportunity 3 – Seeing knowledge ecosystem formation through institutional and organisational lenses During the data collection, the analysis, and the emerging findings, I was exposed to additional research opportunities – first, the role of the institutional environment in enabling and constraining ecosystem formation and, second, that ecosystems as contemporary forms of organising with no strict boundaries or hierarchies are to a degree organisable (see also King, Felin and Whetten, 2010; Puranam, Alexy and Reitzig, 2014; Dobusch and Schoeneborn, 2015). However, to make sense of my emergent findings, I needed new lenses that could provide some preliminary direction and help me interpret my initial findings. Therefore, I turned to institutional theory (e.g. Scott, 2014; DiMaggio 19 and Powell, 1983; Sudabby, Bitektine and Haack, 2017) and the literature on non- traditional forms of organisation (e.g. Ahrne and Brunsson, 2005, 2011), respectively. The institutional lens: An institutional approach to ecosystems The ecosystem literature has recently applied institutional lenses and highlighted their appropriateness in studying the dynamics and boundaries of ecosystems as well as the early phases of ecosystem creation (Autio and Thomas, 2014, 2018). In particular, the organisational/institutional fields and legitimacy provide useful lenses for understanding or interpreting such phenomena. For instance, Autio and Thomas (2018) suggest that ecosystem creation entails the establishment of institutional legitimacy among future ecosystem stakeholders. The institutional literature has extensively acknowledged that the emergence of new technologies and innovations comprises compound institutional dynamics that, apart from a focus on technological changes, require attention to user practices, regulation, infrastructures, and symbolic and cultural issues (e.g. Geels, 2002, 2004; Markard, Wirth and Truffer, 2016). The value and prospects of a focus on organisational and institutional fields has also been emphasised in the wider organisation literature (see e.g. Davis and Marquis, 2005). The notion of field connotes the existence of a community of organisations that partakes in a common meaning system and whose participants interact more frequently and fatefully with one another than with actors outside the field (DiMaggio and Powell, 1983; Scott, 2014; Zietsma, et al., 2017). It has been argued that there is a need for more empirical investigations on the field level, mainly due to the fact that considerable economic change and socio-technical transitions are not confined within boundaries of single organisations or industries (Davis and Marquis, 2005; Geels and Schot, 2007). In this thesis, institutional lenses are adopted to examine the role and impact of the wider institutional context on knowledge ecosystem formation. Institutional theory and concepts such as legitimacy and institutional field provide a suitable lens for making sense of phenomena regarding how social systems are formed and assessing the institutional forces that can restrain, construct, and empower these systems (Suchman, 1995; Scott, 2014; Berthod, 2016; Zietsma, et al., 2017). For knowledge ecosystem formation, examining the complete range of institutional phenomena at the field level is particularly important, given the co-evolving and interdependent nature of ecosystems and their wider environments (e.g. Aarikka-Stenroos and Ritala, 2017; Möller and Halinen, 2017). The organisational lens: Ecosystem as a form of organisation Organisations have evolved significantly since the first and highly influential organisation theories were postulated. In fact, through the various types of inter-organisational cooperative behaviour, several forms of organisation have emerged. Ecosystems constitute a form of organisation that can incorporate organising elements such as membership, rules, monitoring, etc. (cf. Gulati, Puranam and Tushman, 2012) while still allowing the actors to be loosely coupled and to self-organise. According to Moore (2013), ecosystems can be viewed as a form of organisation ‘that shows promise in 1 Introduction 20 achieving shared purposes, sharing value among many contributors, and in bringing the benefits of technology to a range of people, cultures and problems far beyond what earlier systems have achieved’ (p.3). However, when shifting the focus from single organisations to ecosystem-like structures, it gets harder to define what is or is not an organisation (Wilhoit and Kisselburgh, 2015, p.573). While ecosystems might be considered less organisable compared to single organisations, they meet the conditions of patterning and structure that create at least a viable setting for organisation (see Schreyögg and Sydow, 2010). Nonetheless, much is still unclear about the organising processes and elements in ecosystems; therefore, there is still a theoretical need to deepen our understanding of ecosystems as forms of organisation (see for recent calls Moore, 2013; Gawer, 2014; Jacobides, Cennamo and Gawer, 2018). In this thesis, I address this theoretical need and examine the organisation of ecosystems by integrating the previously distinct literature on non-traditional forms of organisation. Specifically, I have used concepts such as meta-organisation, partial organisation, and organisationality to sensitise my analysis as these concepts provide a suitable lens when trying to make sense of fluid, non-hierarchical forms of organisation, such as ecosystems (see Ahrne and Brunsson, 2011; Gulati, Puranam and Tushman, 2012; Dobusch and Schoeneborn, 2015). 1.2.2 Research questions Considering the identified research gaps and research opportunities, the main purpose of this study is to theorise on the institutional factors and intermediate processes that influence knowledge ecosystem formation. By adopting institutional and organisational lenses, and taking a holistic, multi-level approach to embrace the complex nature of the core phenomenon, I ask the following main research question: How are knowledge ecosystems formed? Subsequently, the main research question is divided in four sub-questions. Their connection to the recognised research gaps and opportunities as well as the publications addressing them is presented in Table 1. Table 1. Research questions, gaps, and publications Research question Research gaps and opportunities Objectives Supportive literature Publication Why is it important to consider knowledge ecosystems, and how do they relate Gap 1 (Concept) To argue on the value of considering interdependent knowledge creation and search separately and explain the intricacies of knowledge (Autio and Thomas, 2014; Valkokari, 2015: Clarysse, et al., 2014; Oh, et al., 2016; Tsujimoto, et al., Publications I and II 21 to other types of ecosystems? ecosystems compared to other ecosystem types. 2017; Ritala and Gustafsson, 2018). How are knowledge ecosystems organised? Gap 2 (Context and dominant approach), 3 (Phenomenon) Opportunity 1 (holistic and multi-level approach), 3 (theoretical lenses) To provide a better understanding on knowledge ecosystems’ particular forms of organisation and how the nature of intermediate processes influences them. (Valkokari, 2015; Overholm, 2015; Barile, et al., 2016; Gawer, 2014; Jacobides, Cennamo and Gawer, 2018) Publication II How do institutional factors shape knowledge ecosystem formation? Gap 2 (Context and dominant approach), 3 (Phenomenon) Opportunity 1 (holistic and multi-level approach), 2 (temporal lens), 3 (theoretical lenses) To provide a better understanding on how factors in the broader institutional environment during pre-formation phase shape knowledge ecosystem formation. (Overholm, 2015; Barile, et al., 2016; Seidel and Greve, 2017; Autio, et al., 2018) Publication III How can public policy drive knowledge ecosystem formation? Gap 2 (Context and dominant approach), 3 (Phenomenon) Opportunity 1 (holistic and multi-level approach), 2 (temporal lens), 3 (theoretical lenses) To suggest mechanisms by which the institutional environment can facilitate the formation of knowledge ecosystems and particularly the intermediate processes of identifying common interests and defining shared purpose. (Overholm, 2015; Barile, et al., 2016; Seidel and Greve, 2017; Autio, et al., 2018) Publication IV The first sub-question aims to discuss the value and usefulness of the knowledge ecosystem concept in explaining interdependent knowledge creation and search as well as its relationship to other ecosystem concepts. Publications I and II address this question. The second sub-question takes an organisational perspective on ecosystems and is discussed in publication II. The third sub-question examines the impact of the wider institutional environment on knowledge ecosystem formation. Publication III address this 1 Introduction 22 question by examining the constraining influence that the institutional environment might have on the early pre-formation phases, when an ecosystem has yet to emerge. Finally, the fourth sub-question aims to highlight the enabling impact that public policy can have on knowledge ecosystem formation and the intermediate processes of identifying common interests and defining common goals and is discussed in publication IV. 1.2.3 Positioning the research This thesis is positioned in the intersections of three main literature streams (see Figure 1). It primarily contributes to the knowledge ecosystem literature, from where I adopted my main concepts. Then, as already discussed in section 1.2.1, the institutional literature and literature on non-traditional forms of organisation have provided the necessary lenses and sensitising concepts for interpreting my core phenomenon of knowledge ecosystem formation. The knowledge ecosystem literature is branch of the broader ecosystem literature that draws from the ecosystem-as-affiliation perspective. According to this perspective, ecosystems are communities of interconnected actors defined by their affiliations, usually to a focal actor or platform (see e.g. Moore 1993, 1996; Iansity and Levien, 2004; Autio and Thomas, 2014). However, as was explicated in the previous sections, in certain contexts ecosystem actors might not be affiliated to a focal actor or platform but rather to the broader community. This stream of literature emphasises cross-industry boundaries, interdependence, and symbiosis among ecosystem actors (Iansiti and Levien, 2014) and best describes community-level phenomena or other macro-level interactions (Adner, 2017), which makes it a relevant perspective for the purposes of this thesis. In section 2.1.2, I provide a more elaborate discussion on this stream of literature and its relation to knowledge ecosystems. To conclude, I would like to clarify positioning the formation concept under the ecosystem literature stream. In this thesis, I utilise formation as it is defined in the broader inter-organisational relationships literature (see section 1.3) and because I consider the ecosystem literature a branch of this broader literature stream, I have positioned the concept in the ecosystem literature. Nevertheless, formation has been a central focus in the institutional literature and particularly research on organisational fields (see e.g. Lawrence and Suddaby, 2006; Greenwood and Suddaby, 2006; Zietsma and Lawrence, 2010), as we will also see in section 2.2.1. In section 1.2.1 I have already discussed how I use the institutional literature and literature on non-traditional forms of organisation in my research as well as why they are relevant for making sense of knowledge ecosystem formation. In section 2.2, I further elaborate on the value of these literature streams and the particular concepts I utilise to sensitise my analysis. 23 Figure 1. Positioning of the research The following section briefly outlines a glossary of the main concepts discussed in this thesis. 1.3 Glossary My aim in this section is to provide the reader with a glossary of key concepts that will serve as a quick aid on the working definitions used in this thesis. Therefore, its purpose is not to thoroughly explain the key constructs. For a more comprehensive review, I would kindly refer the reader to section 2. I present the terms per literature stream in alphabetical order in Table 2. Table 2. Definitions of key concepts Literature stream Concept Definition Knowledge ecosystems literature Co-evolution ‘…multiple dynamics that interact with one another over time’ (Aarikka-Stenroos and Ritala, 2017, p.25) Emergence The process of complex systems rising out of a set of interactions; though, the connection between the actions and motives of individual actors and the systemic 1 Introduction 24 outcome is uncertain (Smith and Stacey, 1997; Peltoniemi, 2006) Formation The factors and processes that influence the emergence of knowledge ecosystems and their particular forms of organisation (see e.g. Ebers, 1997) Knowledge ecosystem ‘A knowledge ecosystem consists of users and producers of knowledge, organized around joint knowledge search’ (own definition in Järvi, Almpanopoulou and Ritala, 2018, p.1533) System ‘A specific set of components (actors, organizations, entities) that are interdependent, but independent of other systems (e.g. von Bertalanffy, 1956)’ (Ritala and Almpanopoulou, 2017, p.39) Institutional literature Field ‘a community of organizations that interact together “frequently and fatefully” (Scott, 1995, pp.207-8) in a “recognized area of institutional life” (DiMaggio and Powell, 1983, p.148)’ (Zietsma, et al., 2017, p.391) Institutions ‘Institutions comprise regulative, normative, and cultural-cognitive elements that, together with associated activities and resources, provide stability and meaning to social life’ (Scott, 2014, p.56) Legitimacy ‘…a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions’ (Suchman, 1995, p.574 in Scott, 2014) Literature of non- traditional forms of organisation Meta-organisation Form of organisation that comprises legally autonomous individuals or organisations bound by a system-level goal (Gulati, Puranam and Tushman, 2012) Organisation Any ‘multiagent system with identifiable boundaries and a system-level goal toward which the constituent actor’s efforts are expected to contribute’ (Puranam, Alexy and Reitzig, 2014, p.163) Organisationality The convergence of organisational identity and actorhood in which social collectives achieve the minimum criteria of an organisation, i.e. the interconnectedness of decision making, collective entity, and related identity (Ahrne and Brunsson, 2011; Dobusch and Schoeneborn, 2015) Partial organisation Form of organisation including only one or few elements of organising – i.e. membership, hierarchy, 25 monitoring, rules, and sanctions (Ahrne and Brunsson, 2011) 1.4 Organisation of the dissertation The dissertation consists of two parts. Part I provides the overview of the study, and Part II presents the individual publications. I organise Part I into five sections. Section 1 (Introduction) informs the reader about the research background, research gaps, recognised research opportunities, and related research questions. In addition, I provide a glossary of key terms. Section 2 (Conceptual and theoretical background) reviews the ecosystem concept in prior literature as well as its intricacies and the theoretical lenses adapted in this study. Section 3 (Methodology) introduces the theoretical paradigm and rationale of the study, the research design, and the methodological choices for the individual studies. Section 4 (Publications and synopsis of the findings) summarises the key findings of the individual publications presented in Part II. Finally, section 5 (Conclusions) provides answers to the research questions, the study’s theoretical and practical contributions, and suggestions for future research. 27 2 Conceptual and theoretical background This section opens by introducing the ecosystem concept. First, I briefly describe its origin from ecology as well as how and why management scholars adopted it to conceptualise various inter-organisational relationships and a range of interactions in business and innovation. Then, I concentrate particularly on knowledge ecosystems – which are the focus of this thesis. Finally, in sub-section 2.2 I describe the theoretical lenses utilised to understand the core phenomenon of my thesis, namely the formation of knowledge ecosystems. 2.1 The ecosystem concept The ecosystem is a central concept in the field of ecology. Etymologically the word ecosystem derives from the Greek οίκος, meaning ‘home’, and σύστημα, meaning ‘system’. Ecologists in the late 1800s to early 1900s introduced several terms (such as biocoenosis, microcosm, biosystem, etc.) to describe the complex interdependence of biotic and abiotic components (Golley, 1993). However, Sir Arthur Tansley (1935) first articulated the basic definition of the ecosystem concept in 1935. He defined the ecosystem as a biotic community or assemblage and its related physical environment in a particular place (Pickett and Cadenasso, 2002). With his definition and etymology of the term, Tansley (1935) wanted to emphasise the complex interaction between living organisms (humans included) and non-living components (Evans, 1956; Winterhalder, 1984). Before we delve into the application of the concept in the social sciences and particularly management, there are a few key features of the original concept that are important to note. First, the ecosystem concept is scale-independent (Allen and Hoekstra, 1992; Willis, 1997) as ecosystems might be of any size given the existence of living organisms, physical environment, and interactions within them (Pickett and Cadenasso, 2002). For example, ecosystems might vary from a patch of soil with plants and micro- organisms or an anthill to the entire planet Earth (Willis, 1997; Pickett and Cadenasso, 2002). Nevertheless, ecosystems have explicit spatial boundaries that need to be (desirably) specified (Willis, 1997; Pickett and Cadenasso, 2002). Another important characteristic of the ecosystem concept is that it is not restricted by narrow assumptions of equilibrium or of complex or stable systems (Pickett and Cadenasso, 2002), which makes ‘ecosystem’ a complex but also versatile concept. Since its inception, the ecosystem concept has been widely used in ecology as well as in the fields of anthropology and other social sciences that later adopted the concept (Winterhalder, 1984). The growing popularity and attractiveness of the ecological concept among anthropologists was not only because of its ability to holistically study humans in their natural and physical environment (Moran, 1990) but also because of the potential they saw in the similarities between natural and social systems in regard to organisation and processes (Winterhalder, 1984). Early applications of the ecosystem perspective in economics dates back to Rothschild (1990), who used the ecosystem concept as analogy to describe the economy. According to Rothschild (1990, p.xi), ‘a 2 Conceptual and theoretical background 28 capitalist economy can best be comprehended as a living ecosystem. Key phenomena observed in nature – competition, specialization, co-operation, exploitation, learning, growth, and several other – are also central at business life’. In Rothschild’s (1990) analogy companies serve as living organisms and industries as species. Three years later Moore introduced the ecosystem concept to the management field. In the following section, I describe in more detail the application of the biological concept in the field of business and management and how it differs from other related concepts in the field. 2.1.1 Ecosystem concept in management Management scholars frequently embrace biological concepts to describe organisational phenomena, such as the emergence and evolution of organisations and human systems, since these concepts capture well the complexities that shape and sustain such systems (Mars and Bronstein, 2017). Moore, in his 1993 seminal article (p.76), argued that a company should not be viewed as a member of a particular industry but as part of a business ecosystem where firms co-evolve capabilities and work cooperatively and competitively to support new products, satisfy customer needs, and eventually incorporate the next round of innovations. In these inter-organisational arrangements phenomena such as specialisation, exploitation, learning, and growth are also central (Iansiti and Levien, 2004; Basole, 2009). Since Moore’s initial introduction of the concept, the business and management literature has broadly used the ecosystem term to describe the connections and interdependency between actors with some common or complementary characteristics that encourage or enable the exchange of resources and creation of value (e.g. Williamson and DeMeyer, 2012). The ecosystem term, however, has heavily dispersed to other, more specific viewpoints besides generic reference to business, resulting in a plurality of definitions and prefixes, such as innovation ecosystems, platform or technology ecosystems (e.g. Ceccagnoli, et al., 2012; Wareham, et al., 2014), service ecosystems (e.g. Akaka, Vargo and Lusch, 2013), and, recently, entrepreneurial ecosystems (e.g. Autio, et al., 2018). Table 3 summarises the most dominant ecosystem concepts and their basic definitions. Table 3. Plurality of definitions and prefixes Prefix + ecosystem Basic definition Business ecosystem ‘An economic community supported by a foundation of interacting organizations and individuals – the organisms of the business world. This economic community produces goods and services of value to customers, who are themselves members of the ecosystem. The member organizations also include suppliers, lead producers, competitors, and other stakeholders. Over time, they co- evolve their capabilities and roles, and tend to align themselves with the directions set by one or more central companies. Those companies holding leadership roles may 29 change over time, but the function of ecosystem leader is valued by the community because it enables members to move toward shared visions to align their investments and to find mutually supportive roles’ (Moore, 1996, p.26) Entrepreneurial ecosystem ‘…a distinct type of cluster that specializes in harnessing technological affordances (Gibson, 1977) created by digital technologies and infrastructures…and combines them with spatial (i.e. proximity-related) affordances to support a distinctive cluster dynamic that is expressed through the creation and scale-up of new ventures’ (Autio, et al., 2018, p.74) Innovation ecosystem ‘network of interconnected organizations, organized around a focal firm or a platform, and incorporating both production and use side participants, and focusing on the development of new value through innovation’ (Autio and Thomas, 2014, p.3). Knowledge ecosystem ‘Geographical hotspots … where local universities and public research organizations play a central role in advancing technological innovation within the system’ (Clarysse, et al., 2014, p.1164) Platform ecosystem A network comprising a platform ownerand all providers of complements that increase the value of a platform (Gawer and Cusumano, 2008; Ceccagnoli, et al., 2012) Service ecosystem ‘…relatively self-contained, self-adjusting systems of resource-integrating actors connected by shared institutional logics and mutual value creation through service exchange’ (Vargo and Akaka, 2012, p.207) 2.1.2 Ecosystem-as-structure vs ecosystem-as-affiliation Despite the plurality of prefixes and definitions, Adner (2017) recognised that there are actually two main conceptualisations of ecosystems, namely ecosystem-as-affiliation and ecosystem-as-structure (see Table 4). Ecosystem-as-affiliation refers to the view of ecosystems as communities of interconnected actors defined by their affiliations to a focal actor or platform, whereas the ecosystem-as-structure view sees ecosystems as ‘configurations of activity defined by a value proposition’ (Adner, 2017, p.40). Adner (2017) argues that the two approaches are compatible, meaning that one approach does not exclude the other even though they are conceptually distinctive. The perspective of 2 Conceptual and theoretical background 30 ecosystems as networks of affiliated actors (see e.g. Moore, 1996; Iansiti and Levien, 2004) puts particular emphasis on issues regarding the accessibility of an ecosystem (i.e. the degree of openness) to potential actors as well as macro-level interactions (Adner, 2017). Alternatively, in the ecosystems-as-structure approach (see e.g. Adner, 2006; Adner and Kapoor 2010, 2016), the focus is on a specific value proposition and the actors and activities needed for that proposition to materialise. Table 4. Ecosystem-as-structure vs ecosystem-as-affiliation Ecosystem-as-affiliation Ecosystem-as-structure Definition A network of interconnected organisations that focuses on a shared purpose and is usually organised around a focal actor or a platform. The alignment structure of the multilateral set of partners that need to interact in order for a focal value proposition to materialise. Starting point Community, actors Value proposition Emphasis on Cross-industry boundaries, the rise of interdependence, and the potential for symbiosis, macro- level processes, and interactions Activities to materialise a value proposition Supporting literature e.g. Moore (1993, 1996), Iansiti and Levien (2004), Autio and Thomas (2014), Rong and Shi (2014) e.g. Adner (2006, 2012, 2017); Adner and Kapoor (2010, 2016) Ecosystem-as-structure view The ecosystem-as-structure view places the value proposition at the core of the ecosystem. Therefore, ecosystems according to this view (see Adner, 2006, 2012, 2017; Adner and Kapoor, 2010, 2016) are configurations of aligned activities and actors that are necessary for a value proposition to materialise. Membership in ecosystems-as- structure is not open-ended but clearly defined based on whether or not members’ participation is necessary for the value proposition to happen. The emphasis is usually on how an interdependent set of actors interacts in order to generate and commercialise innovations that create value for the end customer (i.e. materialise a value proposition) and how to adequately coordinate these efforts in order for the materialisation to be successful (e.g. Adner and Kapoor, 2010; Adner, 2012; Kapoor and Lee, 2013). The 31 degree to which participating actors can align their activities and flows will be decisive to the success of the shared value creation endeavour (Adner, 2017; Jacobides, Cennamo and Gawer, 2018) Recently, Jacobides, Cennamo, and Gawer (2018), with the aim to further enrich the scholarly understanding of ecosystems and contribute to the discussion initiated by Adner (2017), proposed a pragmatic and narrower perspective on ecosystems, which they defined as ‘a set of actors with varying degrees of multilateral, nongeneric complementarities that are not fully hierarchically controlled’ (p.10). With their conceptualisation, Jacobides, Cennamo, and Gawer (2018) intended to highlight modularity, different types of complementarity, and the design of distinct roles governed by shared rules as key elements of ecosystems. In other words, they see the role of modular structures and complementarity in facilitating ecosystem emergence by allowing multiple interdependent but distinct actors to align without being hierarchically managed. To summarise, the ecosystem-as-structure perspective starts by defining a value proposition; the value proposition is then the one defining the boundaries of the ecosystem, i.e. the set of actors that need to interact for the proposition to occur (see Adner, 2017). It is worth mentioning here that ecosystem-as-structure bears close resemblance to the supply or value chain constructs with the difference that the ecosystem concept enables taking into account the possible shifts in actors’ positions or activities flow (i.e. dynamics). Ecosystem-as-affiliation view On the other side, the ecosystem-as-affiliation view starts with a community of actors often affiliated to a focal actor or a platform. This perspective of the ecosystem concept is the one closer to the biological concept, considering its emphasis on community-level interactions and processes. For example, the ecosystem-as-affiliation perspective puts particular emphasis on the symbiotic relationships among ecosystem actors and their co- evolution potential (Moore, 1996; Iansiti and Levien, 2004; Li, 2009) as well as how the ecosystem as a whole can adapt and evolve in order to survive unexpected changes in the broader environment (Moore, 1993; Iansiti and Levien, 2004; Basole, 2009). In addition, this literature stream has paid particular attention to issues related to the access and openness of an ecosystem (e.g. Rohrbeck, Hölzle and Gemünden, 2009) and the role of the hub or keystone actor for providing the necessary stability to the ecosystem (Iansiti and Levien, 2004; Dhanaraj and Parkhe, 2006; Pellinen, et al., 2012). A recent turn in the ecosystem-as-affiliation view has been the growing scholarly interest (both conceptual and empirical) in understanding early-stage ecosystems and how these collectivities come into being in the first place. Complementing the increasing research on the orchestration and governance of ecosystems while upholding a focal-actor perspective, Dattée, Alexy, and Autio (2018) show how leading actors can ‘manipulate’ ecosystem creation when uncertainty is high and there is no pre-defined value proposition. Specifically, the authors show how the creation of an ecosystem becomes a 2 Conceptual and theoretical background 32 ‘collective discovery’ orchestrated or controlled by the leading actor, who delays committing too many resources until a desirable shared vision is crystallised. Additionally, Autio and Thomas (2018) also identified mechanisms that leading actors can use to control ecosystem creation, which include manipulating potential stakeholder cognitive and normative legitimacy as well as perceptions of technological and economic instrumentality. In addition to the recent contributions on understanding the creation or emergence of ecosystems, there have been significant advancements related to the conceptual or theoretical underpinnings of the ecosystem construct. In particular, Autio and Thomas (2014, 2018) introduced institutional theory and concepts such as organisational field and legitimacy into the ecosystem analysis. Particularly, Autio and Thomas (2014) argue that ecosystems can be viewed as organisational fields since they encompass key components of organisational fields, namely institutional logics, a variety of actors (e.g. individuals, organisations, associations, etc.), and governance structures (see also Scott, 2014). Complementing this conceptualisation, in Ritala and Almpanopoulou (2017) (publication I), we highlight the component of temporal and spatial boundaries within the definition of ecosystems. Space and time are also key when defining boundaries for biological ecosystems (Willis, 1997; Post, et al., 2007). The boundary issue is important for understanding any particular organisation or social system (Gulati, Puranam and Tushman, 2012), and in ecosystems as open social systems (see Anderson, 1999; Scott, 2014; Scott and Davis, 2016), boundaries must be determined empirically (cf. Scott, 2014). Apart from geographical scope and temporal scale, other useful indicators for determining ecosystem boundaries are permeability and types of flow (Valkokari, 2015). This thesis adopts the ecosystem-as-affiliation perspective, considering that my starting point is not a pre-defined value proposition but a community of actors. In addition, since this thesis investigates at its core macro-level phenomena (i.e. ecosystem formation), I view the ecosystem-as-affiliation perspective as the most suitable approach. However, as also discussed in the previous section, I do not consider affiliation to a focal actor or platform but instead to the broader community (see discussion about research gap 3 in section 1.2). Before I discuss the main concept of this thesis (i.e. knowledge ecosystems), it is worthwhile to briefly inform the reader about other concepts that have been used to describe various inter-organisational phenomena and how they compare to the ecosystem concept (both structure and affiliation views are considered here). 2.1.3 Ecosystem (both views) and other related concepts Regardless of the conceptual advancement in recent years, the debate regarding the value and explanatory utility of the ecosystem concept for organisational phenomena is still ongoing. Some authors discard the usefulness of the concept (e.g. Oh, et al., 2016), and others argue the importance of an ecosystem perspective for understanding how modern organisations function in increasingly complex networks and systems (Adner, 2017; Mars and Bronstein, 2017). However, despite the growing significance of the concept, an ecosystem perspective ‘may not be necessary – and is never sufficient’ (Adner, 2017, 33 p.53) for understanding and explaining inter-organisational relationships and interactions in business and innovation. In fact, there are numerous – and at times much more established – concepts to describe phenomena such as collaborative innovation and knowledge creation. The following table, adapted from Adner (2017, p.54), however expanded to include the concept of clusters and systems of innovation as well as missing elements related to the ecosystem-as-affiliation view, aims to inform the reader about some of the main alternative concepts in the management field, including their focus and how they compare with the ecosystem concept. However, a broader review and comprehensive comparison of the mentioned constructs is outside the scope of this thesis. Table 5. Comparing the ecosystem concept with other related concepts; adapted from Adner (2017) Concept Authors Core issue/key features Comparison to ecosystem concept/missing element Platforms e.g. Gawer and Cusumano (2002); Parker, Van Alstyne and Choudary (2016) Access, incentives, and control with focus on technology Platforms are concerned with the governance of interfaces; ecosystems are concerned with the rise or structure of interdependence Multi-sided markets e.g. Hagiu and Wright (2015) Access, incentives, and control with focus on transactions Multilateral arrangements that do not have a broker role; contestability roles among partners. Indirect links Networks and alliances e.g. Gulati (1999); Powell, Koput and Smith-Doerr (1996) Observed patterns of connectivity Explicit design and alignment strategies; flow of value, who hands off to whom (vs collaborating on invention; building on knowledge) Business models e.g. Osterwalder, Pigneur and Tucci (2005); Zott, Amit and Massa (2011) Plan for value creation and capture for focal firm Indirect links among partners; boundary logic; necessary consistency of models across all partner firms 2 Conceptual and theoretical background 34 Supply chain and value chain e.g. Porter (1985); Simchi- Levi (2005) Make vs buy decisions; bargaining; partner reliability Partners off the critical path; multilateral dynamics; alignment strategies; not hierarchically controlled (Jacobides, Cennamo and Gawer, 2018) Industry structure e.g. Bain (1959) Nature, sources, and management of rivalry Specifics of value creation; innovation and changes to the industry value proposition Industry architecture e.g. Jacobides (2005) Vertical division of labour New dependencies that arise and lie outside the traditional value chain Value net 1/ Value net 2 e.g. Brandenburger and Nalebuff (1996) / e.g. Christensen and Rosenbloom (1995) Competing with complementors / lock into cost structure of supply chain Structure: how the arrangement of actors affects value creation and value capture / dynamics: the emergence and evolution of the network; value appropriation and use (Autio and Thomas, 2014) Systems of technology e.g. Hughes (1993) Social and technological dimension of systems; bottlenecks Absence of explicit structure Open innovation e.g. Chesbrough (2003); von Hippel (2006) Sources of innovation; governance and quality of communities and participants Integration of multiple actors; multilateral dynamics Cluster e.g. Porter (1998); Tallman, et al. (2004) Mechanisms supporting firm- and cluster-level competitive advantage (Autio, et al., 2018); regionality/locality; concept of industry central Dynamic interactions of numerous interdependent actors; inclusion of downstream side actors (Autio and Thomas, 2014) Systems of innovation e.g. Cooke (2001); The ability of national and regional economies to generate innovation; Emphasis on market mechanisms; evolutionary nature of 35 Lundvall, et al. (2002) emphasis on non- market institutions (Autio, et al., 2018; Papaioannou, Wield and Chataway, 2009) interconnections among various actors, innovative activities, and their environment (Papaioannou, Wield and Chataway, 2009) The above table helps us realise that there is no concept perfectly explaining the complexities of modern inter-organisational relationships. Each of the above concepts, ecosystems included, carries its own assumptions, key features, and focal points. As I described in the beginning of this section, the ecosystem concept is a highly versatile, scalable concept with a great ability to capture interdependences among multiple actors and the role of the wider institutional environment as well as phenomena such as co- evolution, collaborative innovation, and knowledge creation (Basole, 2009; Autio and Thomas, 2014; Aarikka-Stenroos and Ritala, 2017). For these reasons, I considered the ecosystem concept to be a suitable conceptual lens for studying collaborative knowledge creation, where multiple actors depend on each other and need to co-develop the required capabilities to achieve any knowledge-creation goals. In the following section, I concentrate on knowledge ecosystems, which are the focus of my thesis, and elaborate on what phenomena they describe the best and how management scholars have been utilising them. 2.1.4 Knowledge ecosystems In order to understand knowledge ecosystems, it is important to distinguish them from the broader innovation ecosystem concept. The innovation ecosystem concept is increasingly utilised to understand this rising interdependence and the systemic nature of innovation (e.g. Rohrbeck, Hölzle and Gemünden, 2009). Innovation ecosystems are seen as ‘network[s] of interconnected organizations, organized around a focal firm or a platform, and incorporating both production and use side participants, and focusing on the development of new value through innovation’ (Autio and Thomas, 2014, p.3). In addition, innovation ecosystems encompass a broader scope of exploration and exploitation, or the process of invention-to-commercialisation (e.g. Dattée, Alexy and Autio, 2018; Valkokari, 2015). The knowledge ecosystem is a much narrower concept, with a scope on early knowledge creation and search. In this type of ecosystem, multiple actors join forces to create new knowledge in a pre-competitive setting (van der Borgh, Cloodt and Romme, 2012; Clarysse, et al., 2014; Valkokari, 2015). Therefore, activities such as exploitation and commercialisation are not the focal point (Valkokari, 2015). As it is one of the most recent additions to the ecosystem concepts repertoire, there is much conceptual and empirical work to be done to clarify its most important features as well as its conceptual and theoretical underpinnings. Prior literature has mostly described knowledge ecosystems as geographically co-located hotspots steered by either 2 Conceptual and theoretical background 36 universities, research institutes (Leten, et al., 2013; Clarysse, et al., 2014), or firms (van der Borgh, Cloodt and Romme, 2012) that focus on collaborative knowledge search (Valkokari, 2015). Knowledge ecosystems have been viewed as forming around particular technological or societal issues (Dougherty and Dunne, 2011), addressing a number of scientific problems, which can progressively lead to knowledge exploitation and actor-specific appropriation (Franzoni and Sauermann, 2014; Perkmann and Schildt, 2015). Building on these conceptualisations, in Järvi, Almpanopoulou, and Ritala, (2018, p.1524) (publication II), we view knowledge ecosystems as ‘organizations comprising diverse actors bound together by a joint search for valuable knowledge while having independent agency also beyond the knowledge ecosystem’. In this conceptualisation, the role of co-location is not highlighted due to the capabilities of technology to facilitate collaborative knowledge work across geographical distances (cf. Still, et al., 2014). As mentioned in the previous paragraph, the main purpose of knowledge ecosystems is to create knowledge collaboratively. To do so, autonomous actors must participate in a joint search for solutions to highly complex problems that no single actor alone can solve. This involves a change of ‘the situation from one in which appropriators act independently to one in which they adopt coordinated strategies to obtain higher joint benefits’ (Ostrom, 1990, p.39). However, before the search for solutions, ecosystem actors have to engage in a search for problems, which broadly refers to domains within which a solution search can start (e.g. Nickerson and Zenger, 2004; Felin and Zenger, 2014; Lopez-Vega, Tell, and Vanhaverbeke, 2016). Finding the problem is important because how complex the problem is has implications for how actors should organise for the solution search (Nickerson and Zenger, 2004; Felin and Zenger, 2014). Knowledge work across multiple organisational boundaries requires coordination mechanisms that can enable navigating the complexities that such arrangements involve. The literature has suggested a variety of such mechanisms or practices, including boundary objects (Carlile, 2002, 2004; Bechky, 2003), representations of work (Barley, 2015), boundary organisations (O’Mahonny and Bechky, 2008; Perkmann and Schildt, 2015), the open sharing of contributions (Franzoni and Sauermann, 2014), and the selective revealing of knowledge (Alexy, George and Salter, 2013). There is still a great deal to learn about how knowledge ecosystems are formed and how they can be organised to accomplish highly uncertain and complex goals. This thesis is an attempt to enrich our understanding of these phenomena, and to do so I have adopted institutional and organisational lenses. 2.2 Knowledge ecosystem formation: Lenses & sensitising concepts The wider ecosystem literature (particularly in the ecosystem-as-affiliation perspective) has primarily focused on how ecosystems are or can be deliberately created by powerful focal actors (e.g. Autio and Thomas, 2018; Dattée, Alexy and Autio, 2018). The emergence of ecosystems (i.e. how complex systems arise out of a set of interactions) as well as the phases prior to the birth of an ecosystem, i.e. its formation, is a much less understood phenomenon. Ecosystem scholars have mainly proposed ecosystem lifecycle 37 models to describe the emergence and evolution of ecosystems. For example, Moore (1993) has proposed four stages of the ecosystem lifecycle – birth, expansion, leadership, and renewal – the first three of which relate to the emergence of an ecosystem. Recently, Thomas (2013) empirically found three phases of emergence – initiation, momentum, and control – and showed that the emergence sequences of diverse ecosystems were significantly different. Specifically, he found that the initiation phase is similar across ecosystems but that the following momentum and control phases exhibit increasing variation as the ecosystem grows and takes on idiosyncratic characteristics. The literature on industry emergence might also offer a useful insight on the emergence of ecosystems. For example, Gustafsson, et al. (2016) identified three phases of industry emergence. The first stage involves a disruption to the existing industrial order, triggering a second, co- evolutionary stage, which includes four sub-processes related to developments in technology, markets, activity networks, and industry identity. The convergence of these sub-processes leads to the third, growth stage and the rise of a new industry. However, when it comes to formation (which is the primary focus of this doctoral dissertation), much of our knowledge comes from the broader literature on inter-organisational relationships and institutional literature, particularly research on field formation. In the inter-organisational network literature, formation refers to the factors and processes that result in the emergence of inter-organisational networks and their particular forms of organisation (Ebers, 1997). The inter-organisational literature has analysed the formation of inter-organisational networks at the actor level, the relational level, and the institutional level (Ebers, 1997). At the actor level, the focus has been mainly on the motives of individual actors to create or enter networking relationships or on the particular strategies for alliance or network formation (see e.g. Ozcan and Eisenhardt, 2009; Hallen and Eisenhardt, 2012). At the relational and institutional levels, the primary focus has been on the conditions that either facilitate or hinder different forms of inter-organisational collaboration (for relational conditions see e.g. Gulati, 1995, 1999; and for institutional conditions e.g. Powell, 1990; Powell et al. 1996). Recently, Valente and Oliver (2018) proposed a model of meta-organisation formation. They particularly found that meta-organisation formation is based on a specific set of actor-level enabling conditions, such as experiential embeddedness, complex systems framing, and receptiveness to coordination innovation. In their model, meta-organisation formation involves a four-step process: translation, norm changing, role redefinition, and equitable collaboration. The identified sub-processes allowed the range of distinct parties and their accompanying insights to be brought together so that the resultant meta-organisation was isomorphic with the complexity that it had to address in its broader environment (Valente and Oliver, 2018). Finally, the institutional literature has primarily considered enabling conditions for field formation (e.g. Battilana, 2006; Greenwood and Suddaby, 2006) and proposed mechanisms of field formation, such as ‘institutional work’ (e.g. Battilana, Leca and Boxenbaum, 2009; Lawrence and Suddaby, 2006) and ‘field-configuring events’ (e.g. Lampel and Meyer, 2008; Meyer, Gaba and Caldwell, 2005 in Thomas, 2013). For example, related to the enabling conditions for field formation, Greenwood and Suddaby 2 Conceptual and theoretical background 38 (2006) focus on institutional entrepreneurship and consider when embedded influential actors are empowered and motivated to act as institutional entrepreneurs. Lounsbury and Crumley (2007) provide a more complex model of field formation that offers a more comprehensive understanding of the processes that lead to the emergence of a new practice field, such as theorisation, contestation, and boundary definition. In this thesis, I adopted the above-discussed conceptualisation of formation (see Ebers, 1997) and particularly focus on the influence of the institutional environment and its related institutions on knowledge ecosystem formation as well as the implication of the nature of intermediate processes on an ecosystem’s form of organisation. In the following sections, I first discuss what an institutional lens can bring to our understanding of ecosystem formation and then merge insights from the ecosystem literature with the literature on fluid forms of organisation. 2.2.1 The institutional lens: An institutional approach to ecosystems Before I discuss how an institutional lens helps us interpret the formation of ecosystems, it is useful to understand what institutional theory is and to define some of its core concepts, particularly focusing on those sensitising my empirical analysis. Institutional theory places institutions at the heart of organisations and generally social life. Institutions encompass regulative (i.e. rules, laws, sanctions), normative (i.e. values and norms, what is appropriate, how things should be done), and cultural-cognitive (i.e. common beliefs, shared logics) features, which, in connection with activities and resources, bring permanence and meaning to social systems (Scott, 2014, p.56). Institutions are thus complex, enduring social structures, comprising symbolic elements (rules, norms, and beliefs), social activities, associated behaviours, and material resources (Giddens, 1984; Scott, 2014). Scott (2014) highlights the centrality of human behaviour and action in transforming or maintaining institutions, which nonetheless arise in human interaction (see also Geertz, 1973; Hallett and Ventresca, 2006; Barile, et al., 2016). Conforming to established institutions enables social systems to achieve legitimacy, eliminate uncertainty, and improve lucidity in their actions and activities (Berthod, 2016). Organisations, as social systems, do not operate in isolation, and in order to survive and flourish in their environment they require social acceptability and credibility, namely legitimacy (Scott, et al., 2000; Scott, 2014). Legitimacy refers to ‘a generalized perception or assumption that the actions of an entity are desirable, proper, or appropriate within some socially constructed system of norms, values, beliefs, and definitions’ (Suchmam 1995, p.574) and can be also characterised as regulative, normative, and cognitive (see Scott, 2014; Markard, Wirth and Truffer, 2016; Suddaby, Bitektine and Haack, 2017). First, regulative legitimacy relates to the degree to which a social system (i.e. organisation) aligns with prevailing processes for rule-setting, monitoring, and sanctioning. Second, normative legitimacy refers to the ‘degree of congruence or fit between the actions, characteristics, and form of the organization and the beliefs and cultural values of the broader social environment within which it exists’ (Suddaby, 39 Bitektine and Haack, 2017, p.454). Finally, cognitive legitimacy refers to the extent to which the organisation’s taken-for-granted expectations are synced to those of the wider environment in which it is embedded (see e.g. Aldrich and Fiol, 1994). In short, legitimacy is ‘a fundamental condition of social existence’, embodying perceived congruence with rules, normative values, or alignment with cultural-cognitive frameworks (Scott, 2014, p.72). Legitimacy serves as ‘an anchor-point of a vastly expanded theoretical apparatus addressing the normative and cognitive forces that constrain, construct, and empower organizational actors’ (Suchman, 1995, p.571). A distinctive attribute of institutional theory and concepts such as legitimacy is that they provide a useful lens for making sense of questions regarding why and how organisations operate the way they do and how they are formed or designed (Berthod, 2016). Recognising these features of institutional theory, the ecosystem literature has recently applied institutional lenses and highlighted its relevance for studying the dynamics and boundaries of ecosystems as well as the early phases of ecosystem creation (Autio and Thomas, 2014, 2018). Establishing institutional legitimacy across potential ecosystem actors is viewed as a key aspect of ecosystem creation and evolution (Aldrich and Fiol, 1994; Sine, David and Mitsuhashi, 2007; Autio and Thomas, 2018). Ecosystem creation is greatly determined by the focal actor’s ability to facilitate or manipulate the development of a shared institutional logic, which in turn will influence the coherence, permanence, and predictability of cooperative ecosystem building endeavours (Autio and Thomas, 2018). Another important concept in institutional theory is the organisational field. Organisational fields were originally defined as ‘those organizations that, in the aggregate, constitute a recognized area of institutional life: key suppliers, resource and product consumers, regulatory agencies, and other organizations that produce similar services or products’ (DiMaggio and Powell, 1983, p.148). The concept is increasingly utilised in studying a range of contexts from delineated systems to the less structured territories (Scott, 2014). Institutional fields have been more broadly seen to form around issues, opinions, politics, norms, debates, a set of products or services, or other organisational arrangements (Hoffman, 1997; Zietsma, et al., 2017). Organisational fields allow one to clearly conceptualise and empirically assess the relevant environment for a given organisation by putting emphasis on the analysis of particular relational linkages and patterns of actions (Scott, 2014; Zietsma, et al., 2017). In addition to the usefulness of the concept in understanding the nature of the environment for a given organisation, it is in itself a useful level of analysis for examining social systems and processes (Scott, 2014). For ecosystem formation, examining the complete range of institutional phenomena at the field-level is particularly important, given the co-evolving and interdependent nature of ecosystems and their wider environments (e.g. Aarikka-Stenroos and Ritala, 2017; Möller and Halinen, 2017). The environment, characterised by its degree of maturity and institutionalisation as well as a set of other dimensions, such as complexity, solidity, and munificence, influences the formation of any social system (e.g. Dill, 1958; Scott, 2014; Möller and Halinen, 2 Conceptual and theoretical background 40 2017). Moore (1993) highlighted the role of the wider environment as a trigger for the renewal (or death) of these ecosystems. Particularly, he argued that issues such as the rise of new ecosystems, sudden changes in environmental conditions (e.g. new regulations or customer buying behaviours), and changes in macro-economic conditions can threaten the existence and vitality of mature ecosystems. The institutional literature contends that the broader environment and accompanying institutions can constrain, that is, restrict, the set of possibilities for action; enable, meaning open up, possibilities for action (see e.g. Giddens, 1984; Scott, 2014; Dijk, Wells and Kemp, 2016; Markard, Wirth and Truffer, 2016); or, recently, orient (i.e. make more likely to settle on some possibilities out of the enabled) action (Cardinale, 2018). However, there is still very little understanding about how these influences of the wider environment manifest in the early phases of ecosystem formation, when potential ecosystem actors look to direct strategic action and commit resources (Kirsch, et al., 2014), and there is a need for initiative taking and collective action to create a shared vision (Pellinen, et al., 2012; Autio and Thomas, 2018). What is happening during the time of potential ecosystem emergence (i.e. pre-formation phase) and the specific socio-economic context in which the organisational actors are embedded during this time can influence the emergence or non-emergence of an ecosystem. Organisational actors during the pre-formation period face a great deal of uncertainty, which can leave certain opportunities unrealised (Kirsch, et al., 2014; Gustafsson, et al., 2016). Previous research on industry emergence has recognised the importance of this period and has primarily documented the existence of the pre-formation stage in cases of ‘successful’ emergence (Agarwal and Bayus, 2002; Golder, Shacham and Mitra, 2009; Kirsch, et al., 2014; Gustafsson, et al., 2016). Overall, for all the reasons explicated in this section, I see the institutional lens, and particularly the concepts of legitimacy and organisational field, as highly useful for understanding how the institutional environment and the involved actors can influence the early phases of ecosystem formation (i.e. pre-formation). 2.2.2 The organisational lens: Ecosystem as a form of organisation Organisations have significantly evolved since the first and highly influential organisation theories (e.g. March and Simon, 1958) were proposed. With firms increasingly engaging in various types of inter-organisation relationships, there has become a greater need for forms of organisation other than that of a single firm. Ecosystems can be viewed as such distinct forms of organisation, showing ‘promise in achieving shared purposes, sharing value among many contributors, and in bringing the benefits of technology to a range of people, cultures and problems far beyond what earlier systems have achieved’ (Moore, 2013, p.3) and can comprise elements of organisation such as membership, rules, monitoring, etc. (Ahrne and Brunsson, 2011; Gulati, Puranam and Tushman, 2012). In the following paragraphs, I will elaborate on the particularities of ecosystems as forms of organisation and how meta-organisation, partial organisation, and organisationality help to make sense of those. Before we examine the intricacies of organising in ecosystems and the sensitising concepts of this thesis, it is useful to have closer look at how previous literature has defined the concept of organisation and what features it encompasses. 41 March and Simon (1993) defined organisations as ‘systems of coordinated action among individuals and groups whose preferences, information, interests or knowledge differ’ (p.2). For them, ‘organization theories describe the delicate conversion of conflict into cooperation, the mobilization of resources and the coordination of effort that facilitate the joint survival of an organization and its members’ (p.2). Drawing on the commonalities of previous conceptualisations of the organisation, Puranam, Alexy, and Reitzig (2014) suggested that an organisation refers to a ‘multiagent system with identifiable boundaries and system-level goals (purpose) toward which the constituent agent’s efforts are expected to make a contribution’ (p.163). The authors also suggested organising to be a particular kind of problem-solving process (see also Weick, 1969; Newell and Simon, 1972) focusing on four general problems of task division, task allocation, reward provision, and information provision. These conceptualisations show that the concept of organisation can incorporate descriptions of organisations beyond the boundaries of the individual firm and other traditional forms. Nevertheless, as Wilhoit and Kisselburgh (2015) acknowledge, when shifting the focus from single organisations to ecosystem-like structures, it gets harder to define what is or is not an organisation (p.573). Due to the multiplicity of actors, and complex interdependences among them, ecosystems might be viewed as less organisable compared to, for example, single firms or other entities; nevertheless, they do meet the minimum conditions for an organisation, namely patterning and structure (see Schreyögg and Sydow, 2010). Ecosystems do not follow a bureaucratic basis of authority, meaning they are not governed by strictly centralised hierarchical structures (cf. Gulati, Puranam and Tushman, 2012). Regardless of the power of a focal actor, ecosystem members are still independent, self-governing entities preserving control over their resources and decision making over their membership status, contribution, and agreement with collective goals (Berkowitz and Bor, 2017). The continuity and survival of ecosystems rest on the active involvement of multiple actors. However, the degree to which an ecosystem’s boundaries are open or closed can vary (e.g. Gulati, Puranam and Tushman, 2012), from entirely open (Franzoni and Sauermann, 2014) to more confined partnerships focusing, for instance, on the specific needs of a single actor (van der Borgh, Cloodt and Romme, 2012). In addition, who is part of an ecosystem might also change over time as their priorities or plans evolve and are not necessarily in line with the collective direction (Barry and Rerup, 2006; Berkowitz and Bor, 2017). Further, not all actors are equally or simultaneously active all the time (Davis and Eisenhardt, 2011; Davis, 2016). The instabilities related to participation (which are intensified in the early stages of ecosystems) can blur the boundaries between the ecosystem and its environment and create a ‘governance void’ characterised by unclear roles, rules, and other governing structures (Garud, Jain Tuertscher, 2008; Järvenpää and Välikangas, 2016). Finally, the autonomy of ecosystem actors can make the creation of and adherence to law-like or unilaterally set rules problematic; consequently, ecosystems often depend on decentralised decision-making processes and voluntary self-regulation (e.g. to collectively agreed standards, rules, and norms) to sustain or boost interaction and collective action (Berkowitz and Bor, 2017). However, the degree of openness in decision making or rule setting can also vary significantly (consider, for example, Apple vs Google vs Linux ecosystems). 2 Conceptual and theoretical background 42 A growing stream of literature on organising beyond traditional organisations has suggested several concepts that can help in making sense of fluid, non-hierarchical forms of organisation, such as ecosystems. First is the concept of the meta-organisation, introduced by Ahrne and Brunsson (2005), to facilitate a better understanding on the organisation of inter-organisational collective action. Meta-organisations consist of legally autonomous actors bound by an overarching, system-level goal (Gulati, Puranam, and Tushman, 2012). Ecosystems can be interpreted as meta-organisations in which independent, self-governed actors converge toward a system-level goal of, for example, creating knowledge, innovation, or value. Meta-organisations are described as ‘decided social orders’ to underscore that organising elements essential for the recurrence of social interaction are an outcome of collective decision making (Ahrne, et al., 2016; Berkowitz and Bor, 2017). In such meta-organisations, actors’ motives and perceptions might not be aligned and are not bound by any formal authority or the employment contracts of a traditional organisation (Ahrne and Brunson, 2005; Gulati, Puranam and Tushman, 2012). A second and related concept for non-traditional forms of organisation is that of partial organisation. Partial organisation, compared to complete, formal organisation, comprises only one or a few of the organising elements (i.e. membership, rules, monitoring, hierarchy, and sanctions) (Ahrne and Brunsson, 2011). Meta-organisations can be partially organised as they might not be able to incorporate all of these elements (Ahrne, Brunsson and Seidl, 2016; Berkowitz and Dumez, 2016); as Berkowitz and Bor (2017) explain, meta-organisations often have weak central authority and low sanctioning power. The concept of partial organisation provides a helpful lens for understanding ecosystem coordination in the absence of some organising elements, such as hierarchical structures (cf. Ahrne and Brunsson, 2011), and how these types of collectivities can function flexibly while maintaining individual actor autonomy (see Felin and Zenger, 2014). To complement Ahrne and Brunsson’s (2005, 2011) emphasis on the interconnectedness of decision making as a minimum criterion for an organisation, Dobusch and Schoeneborn (2015) introduced the concept of organisationality. The term ‘organisationality’, the authors argued, together with interconnected decision making, organisational identity, and actorhood, represents the minimum criteria of an organisation (Ahrne and Brunsson, 2011; Dobusch and Schoeneborn, 2015). Dobusch and Schoeneborn’s (2015) conceptualisation brings more detail into what is and what is not an organisation by proposing ‘a more gradual differentiation’ between organisations and non-organisations (Dobusch and Schoeneborn, 2015, p.1006). Finally, organisationality helps us to comprehend how fluid social collectives, such as ecosystems, accomplish collective actorhood,3 which also entails the formation of a collective identity (Drepper, 2005; Schreyögg and Sydow, 2010; King, Felin and Whetten, 2010; Dobusch and Schoeneborn, 2015). To conclude, the concepts detailed above provide a valuable lens for making sense of the organisation of ecosystems, particularly focusing on knowledge search and creation. 3 Actorhood refers to the notion that a collective is perceived or recognised as an actor that both de jure and de facto is capable of acting (King, Felin and Whetten, 2010; Dobusch and Schoeneborn, 2015) 43 Figure 2 summarises the discussion above and demonstrates how I utilise the institutional and organisational lenses. In the following section, I will present a synopsis of the findings of the individual publications, in which the concepts presented in this section served as sensitising constructs. Figure 2. Theoretical lenses for studying knowledge ecosystem formation 45 3 Methodology This chapter details the methodology of the present research. I begin by introducing the theoretical paradigm and rationale of the study. After elaborating my research design, I introduce the methodological choices for the individual studies. Finally, I conclude with a discussion about the trustworthiness of my study. 3.1 Theoretical paradigm and rationale for the study Paradigms are ‘overarching philosophical systems’ (Lincoln, 2005, p.230) that encompass the researcher’s epistemological, ontological, and methodological assumptions (Guba and Lincoln, 1994; Denzin and Lincoln, 2005). There are multiple theoretical paradigms – positivism, post-positivism, critical theory, interpretivism, etc. – that inform organisational research, and each carries diverse and often competing ideological assumptions. My aim in this chapter is not to go through and compare all the available paradigms competing for acceptance in guiding organisational inquiry. Instead, I will focus on the philosophical premises that guide my study to ensure that the etic view (i.e. researcher’s viewpoint [Patton, 2002]) that frames my inquiry is clear to the reader; for a comprehensive comparison of paradigms, I encourage readers to refer to Guba and Lincoln (1994). The underlying theoretical paradigm of my study is constructivism. The ontological foundation of constructivism, namely what is the nature of reality, lies on relativism (Lincoln and Guba, 2013). Specifically, constructivism starts with the assumption that social reality is multiple, processual, and constructed: ‘Realities exist in the form of multiple mental constructions, socially and experientially based, local and specific, dependent for their form and content on the persons who hold them’ (Guba, 1990, p.27). This view of the nature of reality entails a subjectivist epistemological assumption, which means that knowledge is created in interaction between the researcher and participants and is person- and context-specific (Guba and Lincoln, 1994; Lincoln and Guba, 2013; Charmaz, 2014). In line with Charmaz (2014) and Lincoln and Guba (2013), I view subjectivity as inseparable from social life in that people’s experiences and/or understanding of any physical reality or social interaction determine their actions and responses to it. Constructivism has been previously proposed as a viable alternative approach to the study of managerial and organisational phenomena, such as strategic management and business networks (see e.g. Mir and Watson, 2001; Peters et al, 2013). In network research, a constructivism perspective posits the possibility of creating credible interpretations of network processes and interactions by combining several stakeholder views and engaging them in retrospective and/or prospective sense-making of their lived experiences (Weick, 1995; Halinen, Törnroos and Elo, 2013). Holding a constructivism worldview thus, meaning seeing the world as socially constructed, allows me to understand in context and from multiple perspectives (see also Willis, 2007) ecosystem formation. By definition, 3 Methodology 46 ecosystems involve multiple actors, and these actors comprise people (e.g. individuals, groups, organisations); thus, there is always some intentional action or conscious choice involved (see also Peltoniemi 2006; Valkokari, 2015). With human behaviour being fairly undetermined, multidirectional, and often contradictory (Bouchikhi, 1998) – and individual stakeholder decisions and actions possibly causing counter-responses from others, which then multiply in manifold interdependencies across the ecosystem – the need for a perspective that allows the understanding of such complex interactions is imperative. In fact, constructivism builds on such complexity and postulates that social systems (such as in my case knowledge ecosystems) are both an emerging outcome of and a context for the actions and interactions among multiple stakeholders, thus allowing the formulation of questions that seek to understand how these systems form and what shapes certain outcomes (Bouchikhi, 1998; Peters, et al., 2013). 3.2 Research design 3.2.1 Research strategy The answers one gives to the ontological and epistemological questions have implications for the answer one gives to the methodological question: How does one go about acquiring knowledge? (Lincoln and Guba, 2013). Given the premises of relativism and subjectivism, the methodology suitable to constructivism ‘must be one that delves into the minds and meaning-making, sense-making activities of the several knowers involved’ (Lincoln and Guba, 2013, p.40). Thus, given the complexity of a phenomenon such as the formation of ecosystems and the multiple actors involved, I chose a qualitative inquiry as it enables the investigation of complex and subjective experiences in context (Burrell and Morgan, 1979; Lincoln and Guba, 1985), provides flexibility, and allows for the emergence of unexpected findings (Sinkovics and Alfoldi, 2012). Qualitative inquiry is also particularly appropriate for novel and relatively understudied phenomena (Bansal, Smith and Vaara, 2018) where there is little prior work, as in my case concerning the formation of knowledge ecosystems. Often, ecosystem scholars take for granted the existence of ecosystems and thus neglect the earlier phases of ecosystems (Valkokari, 2015). The ecosystem literature (see e.g. Isckia, 2009; Li, 2009) has primarily focused on how focal actors build or operate in an ecosystem, usually taking a single- actor, top-down perspective. Empirical research taking the whole ecosystem as a level of analysis is quite rare, possibly due to the methodological difficulties arising from the innate complexities of ecosystems (Järvi and Kortelainen, 2017). However, there has been some scholarly work (see e.g. Clarysse, et al., 2014; van der Borgh, Cloodt and Romme, 2012; Still, et al., 2014) taking the whole ecosystem as a unit of analysis to study, for example, network structure, ecosystem lifecycle, or collaborative value creation. Similarly, in my study, I also take a more holistic approach and try to understand the phenomenon of ecosystem formation through multiple actors’ perspectives, and multiple levels of analysis, in relatively unexplored empirical contexts, moving away from the ICT sector (see section 3.2.3 for more elaboration on the empirical settings). 47 Qualitative strategies are increasingly employed when studying networks or, more generally, social systems as they allow researchers to understand complex processes in context and enable building (or refining) theory that is grounded in the experiences of those living with and creating the phenomena (Shah and Corley, 2006). In network research, qualitative methods have been seen as particularly useful for studying, for example, network development, processes, and how actors interact with others in their environment (Jack, 2010) as well as when trying to understand the context within which networks are formed or operate and interpret their influence on specific processes (Jack, 2010; Borgini, Caru and Corva, 2010; Järvensivu and Törnroos, 2010). Similarly, qualitative inquiry is suitable for my study given that my goal is to theorise on the complex processes of formation in ecosystems and to provide deep understanding on the influence of the wider institutional environment on these processes – ‘After all, good theory building requires the rich knowledge that only qualitative methods can really provide’ (Jack, 2010, p.132) 3.2.2 Progressive focusing of my inquiry My thesis originated from two broader research projects, one on ecosystems’ evolution dynamics in the Finnish context and the other on the digital disruption of the energy sector in Finland and the possible emerging ecosystems – study 1 originated from the former project and study 2 from the latter project. However, none of the wider projects focused particularly on the formation and organisation of knowledge ecosystems. Rather, this focus emerged during data collection and analysis and was interpreted as a relevant aspect of the empirical phenomenon and thus a promising theme for theorising. Namely, my research design developed iteratively and progressively (see Figure 3) after I realised the complexities of the research contexts and gradually refined the focus to reflect the emerging issues (see Parlett and Hamilton, 1972; Sinkovics and Alfoldi, 2012). I will discuss the progressive focusing of the overall dissertation in the following paragraphs. The research process for the individual studies is elaborated in section 3.2.3. 49 Figure 3. Progressive focusing for my study; adapted from Sinkovics and Alfoldi (2012) 3 Methodology 50 Phase 1: Choosing my topic and designing my inquiry I was introduced to ecosystem concept while still an undergraduate student, and I was fascinated by how multiple different organisations (even competing organisations) could come together and how such complex collectives could be governed or coordinated. My first task as a doctoral student was to choose a topic and conduct a thorough literature review to build the theoretical and conceptual foundations needed for my research. As I was reviewing the ecosystem literature, I came to realise that there is actually very little research about the early phases of ecosystems. My second supervisor was working on project 1 at that time and already doing some fieldwork in the context of the Strategic Centers for Science, Technology, and Innovation (SHOKs), focusing on four mature ecosystems and how they are coordinated. While in field, she made a similar observation that there is indeed very little understanding on how ecosystems emerge. Due to this synergy, I joined the project and formulated my initial research question, which was about how ecosystems and their coordination mechanisms emerge. Given the scarcity of the extant research on this topic, I designed my study as a qualitative inquiry based on a constructivist worldview because of the constructivist perspective’s focus on the meaning and deeper understanding of phenomena rather than seeking to test the truth (Willis, 2007). At this stage, a multiple case study design was envisioned, and, in addition to the four mature ecosystems, three emerging ecosystems were purposefully sampled (Patton, 1990) to obtain a deeper understanding of what is actually happening in the early phases. Then, I built my interview guide (see Appendix A) and contacted my first interviewees. The first informants were chosen based on archival material. The informants had to be in a position to talk about the earlier phases of the ecosystems and their coordination. At the end of each interview, I asked the interviewees to suggest additional informants who had the knowledge to answer my questions (i.e. snowballing technique). In fact, informants are not chosen to be representative in any statistical sense; they are selected because they are knowledgeable about the researched topic and capable and willing to talk about it (Kumar, Stern and Anderson, 1993). Phase 2: Collecting data and starting the analysis In phase 2, I conducted 13 interviews4 with key informants holding diverse roles and hierarchical levels in the ecosystems under preparation. As data collection progressed, in addition to the broader questions asked, I started focusing and refining the interview guide on emerging issues. After I conducted these interviews, I began analysing the data in collaboration with my second supervisor.5 I utilised a constructivist variant of grounded theory analysis methods (Charmaz, 2014) to ensure philosophical and methodological congruence. This approach provided systematic, though versatile, procedures for 4 Interviews with key respondents of the mature ecosystems were conducted before these interviews as part of the broader project’s original purpose; however, the in-depth analysis started with the interviews from the emerging ecosystems. Thus, in Figure 3, the latter interviews are displayed in phase 2 and the former in phase 3. 5 For a detailed description of the collaborative analytical procedures, please refer to section 3.2.3, ‘Methodological choices for individual studies’. 51 analysing the qualitative data and the resulting construct theory (Strauss and Corbin, 1998; Charmaz, 2014), making it an appropriate analytical strategy for explaining the dynamic phenomenon under examination. In the constructivist variant of grounded theory, the researcher constructs the codes, meaning that coding is a result of the researcher’s actions, meanings, and interpretations (Charmaz, 2014). Constructivist grounded theory recognises that researchers hold preconceptions, previous ideas, and skills yet encourages the researcher to keep an open (not empty) mind while coding and to try to be aware of how his/her previous conjectures influence his/her understanding of the data (Charmaz, 2014; Dey, 1999). The analytical process of grounded theory starts with initial coding, where the aim is to remain open and construct codes that fit the data the best. This stage is followed by focused coding, where the researcher is focusing his/her analysis on the initial codes that make the most analytic sense. Focused coding then guides the formulation of core categories (for elaboration on the coding processes, please refer to Charmaz, 2014 or Saldana, 2009). While still analysing the interview data from the ecosystem cases in the Finnish SHOKs, I received an invitation to build a project proposal tailored to the topic of my doctoral dissertation – the emergence of new ecosystems – in a different empirical context that would provide me with a business-oriented ecosystem case in one of the most strategic industries in Finland, the energy sector. However, as we had no prior experience in this specific sector, we decided to conduct expert interviews in order to understand the current state of the energy sector and identify some interesting emerging ecosystem cases. After acquiring some background knowledge of the Finnish energy sector, eight leading experts in the field were identified and contacted for interviews. Phases 3 and 4: Refocusing the study, adopting new theoretical lenses The ongoing analysis of the data from both studies revealed new exciting paths for my research. First, during the analysis of the data from the Finnish SHOKS ecosystems (both mature and emerging), it became clear that the respondents were not only talking about the coordination of the ecosystems but that they were actually describing the broader organisation. At this stage, the second sub-question (How are knowledge ecosystems organised?) was formulated. Then, I turned to the literature on contemporary forms of organising (e.g. Kellogg, Orlikowski and Yates, 2006; Puranam, Alexy and Reitzig, 2014) and the concepts of meta-organisation (e.g. Gulati, Puranam and Tushman, 2012), partial organisation (e.g. Ahrne and Brunsson, 2011), and organisationality (Dobusch and Schoeneborn, 2015) to deepen my theoretical understanding on organising beyond traditional/formal organisations and to further sensitise my analysis. Sensitising concepts spark the researcher’s perception about a topic (van den Hoonard, 1997), provide lenses for interpreting or understanding experience (Charmaz, 2003), and offer initial but tentative ideas to pursue. Sensitising concepts may guide or lay the foundation for analysis but do not provide prescriptions to the inquiry (Bowen, 2006; Charmaz, 2014). 3 Methodology 52 Similarly, while analysing6 the expert interviews for project 2, it soon became evident that the institutional environment (e.g. regulations, existing norms, and culture) played a central role in this sector, especially in constraining the formation of new ecosystems. This emerging issue led to institutional theory (Scott, 2014) and the concepts of the institutional field (Zietsma, et al., 2017) and legitimacy (Scott, 2014). Then, the interview guide was updated to incorporate more questions about the role of institutions in the formation of ecosystems in the energy sector, and actors holding diverse roles in the sector were purposefully sampled to achieve maximum variation. A total of 18 interviews were conducted during this phase. I then analysed7 these interviews focusing particularly on the instances where the respondents were describing their views or experiences regarding the role of the institutional environment in the renewal of the energy sector. Parallel to the empirical work and the interviews with the experts, the immersion of the project team in nationwide projects focusing on the renewal of the energy sector led to the observation that the institutional environment can also enable or facilitate the formation of ecosystems, which later resulted in publication IV. At this stage, my third and fourth sub-questions were formulated (How do institutional factors shape knowledge ecosystem formation, and how can public policy drive knowledge ecosystem formation?). During my empirical work with both projects, I also continued advancing my conceptual and theoretical foundation regarding ecosystems and posed my first sub-question (Why is it important to consider knowledge ecosystems, and how do they relate to other types of ecosystems?), which resulted in publication I. This conceptual work influences my understanding of the knowledge ecosystem concept and its value in explaining interdependent knowledge creation and search and thus the way I view and interpret my data. Phase 5: Theorising, writing up the dissertation Making sense of and combining the empirical findings of two separate (even though related) studies is a challenging task. Grounded on empirical evidence from study 1, I demonstrated that in the formation phase the ecosystem actors engage in a joint search and theorise on how the nature of the joint search has implications for the organisation of the ecosystem. Then, based on study 2, I found a strong and interdependent set of institutional barriers at the field level that mutually reinforce each other, namely regulation and policymaking ambiguity, incumbent actor inertia, cognitive constraints on opportunity recognition, and institutional complexity. Then, in publication IV, I explicated how policy intervention can function as a facilitator for initial social interaction among potential ecosystem actors by creating forums for collective action or initiative taking. To conclude, I interpreted my findings in study 2 and the conjectures in publication IV as manifestations of constraining, enabling, and orienting the effects of the institutional environment in pre-formation phases, which can initiate or hinder the 6 Constructivist grounded theory analysis methods were also utilised for these interviews. 7 Constructivist grounded theory analysis methods were also utilised for these interviews. 53 formation of knowledge ecosystems. The outcome of my theorising process is illustrated in Figure 4 and is discussed in more detail in the Conclusions section of the thesis. 3.2.3 Methodological choices for individual studies In this section, I will describe the research process for the individual studies. Publication II resulted from study 1 and publication III from study 2. Publication I is conceptual, and publication IV uses case illustrations, thus neither uses the empirical studies.8 Research design study 1 Empirical setting The empirical context9 of study 1 is centred on dynamic and industry-driven research programmes pursued and implemented by SHOKs, established in 2006 by the Finnish government. The main aim of these centres was to boost the competitiveness and innovativeness of strategic Finnish industries (Halme, et al., 2014) as well as to ‘develop and apply new methods for cooperation, co-creation, and interaction’ (Tekes, 2016). The research programmes aim at developing new knowledge in a pre-competitive setting through collaborative, jointly defined research work. The joint efforts of the participants take the form of the recombination of existing knowledge and the creation of new knowledge and inventions, which can later be applied in new products, technologies, and services. The research programmes are built and operated by various companies, research institutions, and universities of different sizes and from diverse industries (i.e. information technology, health and wellbeing, energy, and construction). Monetary resources for the collective are provided by the funding agencies and companies and are allocated to the universities and research institutions according to agreed-upon programme practices. The context was purposefully chosen for the purposes of the broader project as even though the involved stakeholders were diverse and their goals heterogeneous, collaboration had to be coordinated if the collective was to accomplish its goal. Research strategy During the collection and analysis of the data from project 1, a theoretical interest in organising knowledge creation and knowledge work in knowledge ecosystems emerged, which we considered a promising theme for theory building. The study was originally designed as a multiple case study (e.g. Eisenhardt, 1989) of seven cases in which each research programme would represent a case and its own unit of analysis. While performing the within-case analysis (i.e. detailed write-ups for each case [Eisenhardt, 1989]) and cross-case analysis (i.e. searching for patterns, similarities, and differences across cases [Eisenhardt, 1989]), we realised that even though the programmes were very 8 To highlight the collective effort in these studies, I utilise the plural form of the first-person personal pronoun in my descriptions. 9 Please refer to publication II for a detailed description of the empirical setting. 3 Methodology 54 similar in terms of structure (e.g. stratification, roles, and work packages), they bore differences in the knowledge work and approaches to participation and coordination depending on the phase of the research programme, either being under preparation or ongoing. For instance, in programmes under preparation, participants had yet to identify the common goal and field of research and innovation, whereas in ongoing programmes, the work was conducted mainly within an already identified field. Therefore, what is being coordinated and how and who is participating depend on whether the research programme is in preparation or ongoing. Based on this notion, we redesigned the study as a collective case study (Stake, 1995) of two cases, with seven embedded units of analysis. We considered our cases to be instrumental (Stake, 1995) in that they enabled us to develop theory on the organisation of knowledge creation in knowledge ecosystems. Data collection and analysis The data collection process included semi-structured interviews and the gathering of relevant archival material. The use of different data collection methods in our study and the following triangulation of data increase the trustworthiness of our research findings. We interviewed a total of 27 (4 + 23) key respondents. We utilised archival material to identify the key respondents to interview first and used the snowballing technique to identify further respondents who had central roles in the activities and coordination of the research programmes. Specifically, to ensure our respondents’ suitability, our sample also reflects different hierarchical levels in the research programmes as we interviewed focus area directors, academic coordinators, programme directors, programme coordinators, work package leaders, and participants. The interviews were conducted during the fall of 2014 and spring of 2015. The first four interviews were conducted to gain a richer contextual understanding of the centres and research programmes. The following 23 interviews comprised our main data. Our respondents represent different actors in the research programmes and possess a variety of roles representing different hierarchical levels in the research programmes. Specifically, out of the 27 interviewees, 17 represent companies (11 non-profit and 6 for-profit), four represent public research organisations, and six represent universities. Our interview guide consisted of three main sections. First, background questions regarding the respondent and his/her organisation were asked. The second section of the interview guide focused on the preparation of the research programme, and the respondents were asked to describe what has happened in the programme, who its actors are, and what type of activities are performed in the programme preparation. Thirdly, questions about the current state of the programme were asked, focusing on the goals of the research programme as well as on how activities are coordinated in the programme. For the interviews concerning ongoing research programmes, we mainly focused on the third section of our interview guide; however, some questions related to the past were also asked. Fourteen of the interviews were conducted in Finnish and 13 in English, and they lasted between 35 and 85 minutes. They were tape-recorded and transcribed verbatim, allowing for a systematic analysis of the raw data. 55 We started our analysis with initial coding utilising NVivo software. This phase involved coding sentences or segments of the data depending on their richness. We tried to remain open to what our material suggested, and, when applicable, we utilised in-vivo coding to stay as close to the raw data as possible. The initial stage of data analysis for the larger research project sparked our interest in the organisation of knowledge ecosystems and the literature on contemporary forms of organising (e.g. Kellogg, Orlikowski and Yates, 2006; Puranam, Alexy and Reitzig, 2014). From that literature, we identified the concepts of meta-organisation (e.g. Gulati, Puranam and Tushman, 2012), partial organisation (e.g. Ahrne and Brunsson, 2011), and organisationality (Dobusch and Schoeneborn, 2015) as our sensitising concepts. These concepts suggested ‘directions along which to look’ (Blumer, 1954, p.7). While coding, we took an interrogative approach to the data, which means that we took them apart and examined how they were constituted (see Charmaz, 2014). When applicable, we coded with gerunds, enabling the detection of processes, actions, and organising elements. The initial codes varied in length from a couple of words to full sentences. In the second step of our analysis, we went through our initial codes and started to compare them. We then selected those that were most significant or appeared more frequently or that made the most analytic sense. We started to sort them and organise them into focused codes (see Charmaz, 2014). Along with the empirically grounded initial and focused codes, theory codes were also introduced (see also Kreiner, 2016). At the beginning of this phase, we went through the initial codes separately and kept individual memos. Through discussions and multiple iterations with the data and our initial codes, we then agreed on a core set of codes that had the most theoretical reach and centrality for our nascent analysis. Apart from our individual memos, we kept detailed memos of our discussions to ensure that all the valuable and creative ideas about our analysis, and possible relationships between our focused codes, were written down (see Charmaz, 2014). We started our analysis with the data from the research programmes under preparation. With the first author of publication II, we coded the interview transcripts independently, and we met frequently to discuss the coding and resolve issues in order to arrive at a common set of codes. This process ensured that the coders interpreted the data in a similar fashion and without missing any relevant information. Next, we analysed the data from the ongoing research programmes. The first author independently open-coded the interview transcripts, which we then discussed in a joint coding meeting. As we, with the first author, were responsible for coding, the third author in publication II participated in the analysis by providing a ‘new set of eyes’ (Stake, 2000) to challenge the empirical knowledge of the other two. The combination of open and focused coding allowed us to analyse what the cases were all about (within-case analysis). Then, we returned to our focused coding to develop core categories, a process performed in parallel with the cross- case analysis. Research design study 2 3 Methodology 56 Empirical setting The energy sector is currently transforming from centralised power generation and one- way power transmission to decentralised generation and two-way power flow, requiring advanced energy storages; solutions for energy efficiency; demand response; interconnections with other energy networks, such as, gas, heat etc.; and new services based on digitalisation and ubiquitous communications (Bose, 2010; Pinomaa, Ahola and Kosonen, 2011; Lawrence and Woods, 2015). However, this is anything but easy as the energy sector is highly regulated and capital intensive. The sector comprises a variety of stakeholders that each contribute to the development of the market environment, from for-profit organisations (energy producers, retailers, distributors, service providers, etc.) to non-profit organisations such as ministries, non-governmental organisations, universities, and research institutes specialised in the energy field. In Finland, there are few large producers that sell electricity to large-scale energy consumers and electricity retailers and also operate in retail markets. In addition, there is a large number of smaller producers. The majority of the retailers (around 70 currently in Finland) are local and regional electricity companies. The complexities of the new digital era and the inadequacies of past business models and practices to respond to the changing environment call for capabilities that no one actor alone possesses or is capable of developing on its own, which makes actors in the energy sector highly interdependent. Research strategy During our initial interviews to gain a rich contextual understanding of the Finnish energy sector, we realised that there was a pattern in our interviewees’ responses. They were discussing all the issues that actually constrain the sector from moving forward and new ecosystems from emerging or old ecosystems from transforming. This was an unanticipated finding and sparked our curiosity to learn more about the barriers in the wider institutional environment that inhibit ecosystem formation. Therefore, we refocused our study on this issue and turned to institutional theory (e.g. Scott, 2014) and specifically the regulative, normative, and cultural-cognitive institutional elements as sensitising lenses for theorising. In addition, we focused our study on the level of the organisational field (DiMaggio and Powell, 1983; Wooten and Hoffman, 2008; Zietsma, et al., 2017). We considered an in-depth qualitative inquiry to be the most appropriate strategy as it was consistent with the interpretive nature of the study, having the aim to understand, to ‘reconstruct’ multiple stakeholders’ perspectives and meanings (Burrell and Morgan, 1979; Weinstein, 1999; Willis, 2007; Peters, et al., 2013) regarding the constraints on ecosystem formation in the energy sector. Data collection and analysis We collected our study data through semi-structured interviews. Specifically, we conducted 26 interviews with key informants representing different stakeholders in the energy sector. We aimed for variation (see Patton, 1990) in roles and organisations represented to ensure that our sample comprised as many perspectives as possible and 57 thus to ensure the trustworthiness and authenticity of our findings (Lincoln and Guba, 1985). We utilised archival material to identify our first interviewees and then a snowballing technique by asking at the end of each interview for suggestions regarding further interviewees. Our data collection consisted of two phases. During the first phase, we interviewed eight experts in the energy sector. Our aim was to gain an overall understanding of the specific field and its current state in terms of digitalisation and to identify any interesting emerging ecosystems. As mentioned above, during this phase, we observed that there were certain obstacles restricting the emergence of new ecosystems, which subsequently led us to institutional theory. Accordingly, we conceptualised the energy sector as an organisational field and formulated our final research questions (What are the barriers that inhibit ecosystem emergence, and how are these barriers sustained?). In the second phase, we updated our interview guide to incorporate questions about the role of institutions and regulation and the roles and activities of the various actors in this regard. We then interviewed 18 additional key informants representing different actors of the organisational field. The interviewees (in both phases) were made up of a wide variety of experts, comprising six leading energy-sector-related academics, two research institute representatives, five policymakers, 10 company representatives, two industry association representatives, and one representative of a non-governmental organisation. In the first phase of the data analysis, I used NVivo software to independently start the analysis with initial coding. In this phase I coded sentences or segments of the data depending on their richness. I tried to remain open to what the material suggested and, whenever applicable, utilised in-vivo coding. The initial codes varied in length from a couple of words to full sentences. In the second phase of our analysis, my co-authors and I actively iterated the initial codes through discussions in multiple meetings. At the beginning of this phase, my co-authors went through the initial codes separately and provided comments, questioning the analytical decisions and helping to raise the level of abstraction. We then selected the codes that were most significant, that appeared most frequently, or that made the most analytic sense and started to sort them and organise them into focused codes (see Charmaz, 2014). In this stage, in addition to grounded codes, theory codes were also applied. Our approach to the data analysis followed an abductive approach (cf. Kreiner, 2016), comprising an iterative cycle of inductive patterns with a reflection back and forth with theory, which made it possible to draw broad patterns from the data. Additionally, the different analytical roles in the research team provided a more wide-ranging and diverse set of perspectives, which further increased the trustworthiness of our findings. 3.3 Trustworthiness criteria for my inquiry The ontological, epistemological, and methodological choices one makes also have implications regarding how the quality or goodness of one’s inquiry should be judged as well as what the most appropriate criteria are (Lincoln and Guba, 2013). The quality criteria for an inquiry cast in the constructivist paradigm, that is, with relativist ontological foundations and subjectivist epistemological premises, cannot be objectivist or foundational. In fact, Lincoln and Guba (1985) and Lincoln (1995) proposed 3 Methodology 58 trustworthiness criteria as the most appropriate for such a paradigmatic framework. Trustworthiness refers to whether the research outcomes are the result of a systematic process and can be trusted (Lincoln and Guba, 1985). Lincoln (1995) divided the trustworthiness criteria in two main categories, parallel methodologic criteria, which include credibility, transferability, dependability, and confirmability, and authenticity/ethical criteria, which include fairness, ontological authenticity, educative authenticity, catalytic authenticity, and tactical authenticity (refer to Table 6 for definitions). Parallel methodologic criteria correspond to the rigor criteria of objectivist research, namely validity, reliability, and generalisability, whereas authenticity/ethical criteria are innate to the constructivist paradigm and do not correspond to any conventional criteria. I utilise both sets of criteria to judge the quality of my overall research approach. I have briefly discussed in the methods section how we ensured the trustworthiness of our findings and interpretations in the individual studies; however, Table 6 provides a more comprehensive summary of the strategies employed to ensure the trustworthiness of my inquiry. Table 6. Trustworthiness criteria for my inquiry Trustworthiness criteria10 Method of addressing Parallel methodologic criteria Credibility (paralleling internal validity, establishing confidence in the findings and interpretations of a research study)  Triangulation of sources and researchers: multiple coders, multidisciplinary research team (project 2), different analytic roles in the research team, multiple interviewees holding different roles and representing diverse organisations  Peer debriefing: presenting emergent findings in seminars and conferences, papers undergoing double-blind peer review process and receiving extensive feedback and comments  Extended engagement and observation in the field: 2 research projects since 2014, a total of 57 interviews with key informants in the field, multiple other interactions with leading experts from multiple organisations Transferability (paralleling external validity, applicability of the findings and interpretations)  Thick descriptions of the empirical setting allow for better comprehension of findings and strengthens interpretations (see Geertz, 1973) 10 The reference sources for Table 6 are Lincoln (1995), Guba and Lincoln (1994), and Lincoln and Guba (2013). 59 Dependability (paralleling reliability, findings and interpretations an outcome of a consistent process)  Research process described in detail  Interview guides available  Path from empirical data to interpretations thoroughly explained  Reflexive memos (study 2) Confirmability (paralleling objectivity, findings and interpretations are a result of a dependable process of inquiry)  Triangulation of sources  Direct interview quotations used  Within and cross-case analysis of findings (study 2) Ethical/ Authenticity criteria: Fairness (views of multiple stakeholders have been accessed and taken into consideration)  Triangulation of sources  Peer debriefing  Prolonged engagement and persistent observation Ontological authenticity (stakeholders learning something new, enlargement of their personal constructions)  Dialectical conversations with interviewees  Openness of purpose Educative authenticity (enhanced understanding of constructions of others)  Publications resulting from the research studies  Presentations of results in seminars Catalytic authenticity (findings stimulating action)  Accessibility of final reports (i.e. academic publications) to all stakeholders Tactical authenticity (empowerment of action)  Negotiated data to be collected and their reporting  Maintenance of anonymity and confidentiality  Inclusion of representatives from multiple stakeholder groups 61 4 Publications and synopsis of the findings This section of the dissertation summarises the findings of the individual publications that comprise the second part of this thesis. I discuss each publication in terms of objectives, main findings, and contribution and provide a synopsis of each publication in Table 7. Synopsis of the individual publications and their main findings. The reader should refer to the original publications for a more detailed discussion. In three of the individual publications the innovation ecosystem label is utilised, however this label was applied before the notion of ‘knowledge ecosystem’ was more clearly understood. 4.1 Publication I – In defense of ‘eco’ in innovation ecosystems Main objective The ecosystem literature has been cumulating for decades, however there is very little consensus on the definition and usefulness of the term itself. This lack of consensus related to the meaning, scope, boundaries, and theoretical foundations of the ecosystem concept have led authors to claim that the ecosystem construct is a ‘…faulty analogy to natural ecosystems, and is therefore a poor basis for the needed multi-disciplinary research and policies addressing emerging concepts of innovation’ (Oh, et al., 2016, p.1). This paper is a reflection to this recent critique and an attempt to counter-argue why the ecosystem concept can be helpful for making sense of collaborative knowledge creation and innovation. Considering the increased scholarly interest in and popularity of the ecosystem concept among practitioners, it is important to assess more closely its possible merits as well as its conceptual, methodological, and theoretical underpinnings (cf. Ritala and Gustafsson, 2018) before any attempt to discard the concept from the scholarly discourse. Main findings This paper starts by unpacking the etymology of the ecosystem term and particularly highlights the role of the ‘eco’ prefix in the term (i.e. the ecological part) and what it brings to the management field. The ecological part of the ecosystem term refers to the interdependency and co-evolution among the various actors involved in an ecosystem. Moore had explicated the importance of these aspects in his seminal 1993 introduction of the concept in business and management. However, the second part of the term, namely ‘system’, should by no means be downgraded as it refers to the specific components making up the ecosystem and how these are connected while remaining independent entities. Nevertheless, it is highly important to specify clear boundaries when attempting to study any system or organisation considering the scalability of the concept, similar to studying biological ecosystems (e.g. Willis, 1997). This paper also proposes ways to move forward conceptually and methodologically and increase rigor. For example, methodologies such as simulation modelling can uncover the complex links among constructs or the effects of interactions among various organisational and strategic 4 Publications and synopsis of the findings 62 processes over time (Repenning, 2002; Zott, 2003; Davis, Eisenhardt and Bingham, 2007). Furthermore, qualitative process research can provide deep understanding on the ‘hows’ and ‘whys’ of these processes (Langley, 1999). Contribution This paper is an attempt to stimulate further discussion and development related to the foundations and value of the ecosystem concept in business and management. Therefore, it contributes to the ongoing scholarly debate on the value and usefulness of the concept and responds to calls for more reflection regarding its conceptual and theoretical underpinnings (see Ritala and Gustafsson, 2018). One key contribution of the paper concerns the proposed conceptualisation of the term, complementing previous conceptualisations (e.g. Autio and Thomas, 2014) and adding the components of temporal and spatial boundaries into the definition of ecosystems. Specifically, we define ecosystems as ‘systems that focus on innovation activities (goal/purpose), involve the logic of actor interdependence within a particular context (spatial dimension) and address the inherent co-evolution of actors (temporal dimension)’ (p.41). Regarding this thesis, the conceptual work in publication I has influenced my understanding of the ecosystem concept and its value in explaining collaborative knowledge creation and innovation and therefore how I view and interpret my data. 4.2 Publication II – Organisation of knowledge ecosystems: Prefigurative and partial forms Main objective Ecosystems can be viewed as a form of organisation ‘that shows promise in achieving shared purposes, sharing value among many contributors, and in bringing the benefits of technology to a range of people, cultures and problems far beyond what earlier systems have achieved’ (Moore, 2013, p.3). However, when shifting the focus from single organisations to ecosystem-like structures, it gets harder to define what is or is not an organisation (Wilhoit and Kisselburgh, 2015, p.573). While ecosystems might be considered less organisable compared to single organisations, they meet the conditions of patterning and structure that create a viable setting for organisation (see Schreyögg and Sydow, 2010). Nonetheless, much is still unclear about the organising processes and elements in ecosystems; therefore, there is still a theoretical need to deepen our understanding of ecosystems as forms of organising (Gawer, 2014). To address this issue, the following question is posed: How are knowledge ecosystems organised? – specifically focusing on: What is the nature of the knowledge search in knowledge ecosystems? Who participates in the knowledge search? How are knowledge search and knowledge-creation activities coordinated? Main findings 63 This paper shows how knowledge ecosystems are organised around collaborative knowledge creation and search, and, based on this notion, a refined, empirically grounded definition of the knowledge ecosystem is proposed: A knowledge ecosystem consists of users and producers of knowledge, organised around joint knowledge search (p.11). In addition, an important finding of this study is that knowledge ecosystems differ in terms of the nature and target of the joint search. Knowledge ecosystems are then conceptually differentiated into those searching for a knowledge domain and those searching within an identified knowledge domain. Specifically, it is proposed that the search for a knowledge domain involves probing and formulating a common goal, and the search within a knowledge domain involves selectively revealing and reinforcing the common goal. Finally, it is demonstrated that the nature of joint search has implications for how knowledge ecosystems are organised, and particularly two forms of organising are identified: prefigurative and partial. Prefigurative ecosystems involve informal coordination and affiliation-based participation, whereas partial ecosystems involve formal regulation and monitoring and membership-based participation. Contribution This paper enhances our understanding of how knowledge ecosystems search for and create new knowledge and how they are organised for these tasks and further proposes a refined, empirically grounded definition of knowledge ecosystems. Theoretically, this paper contributes to three main streams of literature: literature on knowledge search, literature on ecosystems, and literature on non-traditional forms of organising. Regarding the knowledge search stream, this paper complements previous findings on various governance modes for knowledge search in innovation contexts varying in degree of openness (Felin and Zenger, 2014; Lopez-Vega, Tell and Vanhaverbeke, 2016) by explicating knowledge ecosystems as specific forms of organising for problem identification, search, and solving. It is also suggested that processes of problem identification and problem solving may occur iteratively and, in many cases, that these processes may be even parallel or intertwined (see also von Hippel and von Krogh, 2015). Regarding the ecosystem literature, this paper departs from a focus on a single focal actor’s coordination efforts (Rohrbeck, Hölzle and Gemünden, 2009; Leten, et al., 2013; Ritala, et al., 2013) or the structural aspects of ecosystems (Still, et al., 2014; Clarysse, et al., 2014) and takes an organisation perspective to study the interdependence between the actors and collective efforts toward knowledge integration. Finally, regarding the literature on non-traditional forms of organising, this study adds more detail to Ahrne and Brunsson’s (2011) dichotomy of partial and complete organisations and also demonstrates the importance of the concept of organisationality (Dobusch and Schoeneborn, 2015) in a knowledge-creation context, especially in regard to understanding how collective actorhood is achieved. Related to this thesis, this paper answers the second sub-question – How are knowledge ecosystems organised? – and provides a better understanding on ecosystems as a form of organisation as well as their particular organising requirements and challenges. 4 Publications and synopsis of the findings 64 4.3 Publication III – Innovation ecosystem emergence barriers: Institutional perspective Main objective Moore (1993) highlighted the role of the wider environment as a cause for the renewal (or death) of mature ecosystems. Specifically, he viewed issues such as the rise of new ecosystems and sudden changes in environmental or macro-economic conditions as threatening to the survival and continuation of mature ecosystems. The institutional environment has been viewed as constraining, enabling, or orienting action (Giddens, 1984; Scott, 2014; Suddaby, Bitektine and Haack, 2017; Markard, Wirth and Truffer, 2016; Cardinale, 2018). However, there is still very little understanding about how these influences manifest in the pre-formation phases of ecosystems, when potential ecosystem actors look to direct strategic action and commit resources (Kirsch, et al., 2014), and there is need for initiative-taking and collective action to create a shared vision (Pellinen, et al., 2012; Autio and Thomas, 2018). This paper particularly focuses on the constraining function of the institutional environment and examines the dynamic counterforces for ecosystem formation. Specifically, the following question is posed: What are the barriers that inhibit ecosystem emergence, and how are these barriers sustained? Main findings Grounded on empirical evidence, this paper suggests four main barriers to innovation ecosystem emergence. First, incumbent actor inertia – sustained by actors’ adherence to the past business logic and operating principles and the concentration of influence within static and closed networks as well as the dominance of the legacy industry – was found to be a prominent barrier to new ecosystems forming and existing ecosystems transforming. Second, regulation and policymaking ambiguities, sustained by the slowness of the policymaking, short-sighted political vision, and geopolitical and economic risks, create uncertainty regarding the future direction of the energy policy and thus act as a major inhibitor for new investments and broader ecosystem initiatives. Third, cognitive constraints on opportunity recognition, sustained by perceived uncertainty over market opportunities, the dispersion of necessary capabilities and resources, and a lack of policy-driven incentives for innovation, present a significant barrier to both new players attempting to form new ecosystems and established actors making sense of socio- technical change and related opportunities. Finally, institutional complexity, sustained by complex regulatory processes and system-level renewal requirements, is seen as an important barrier to ecosystem formation. The findings suggest that policymaking ambiguities, the inertia of incumbent actors, and cognitive constraints on opportunity recognition mutually reinforce each other, and institutional complexity functions as an overarching barrier that further sustains the inflexibility of incumbent actors and policymakers and creates additional cognitive constraints on opportunity recognition. Contribution 65 This study shows that innovation ecosystem formation involves a multifaceted set of issues, which increases the importance of critically examining the wider institutional environment and particular fields (e.g. the energy sector) when attempting to understand how the grassroots of ecosystem formation is constrained. This paper identifies a variety of field-sustaining mechanisms mutually reinforcing each other, which broadly correspond to regulative, normative, and cognitive legitimacy (cf. Scott, 2014; Markard, Wirth and Truffer, 2016; Suddaby, Bitektine and Haack, 2017). These findings contribute to recent calls for more research to understand field transformation and related constraining forces (cf. Zietsma, et al., 2017). In addition, by having a better understanding of the constrains the wider environment can pose, the actors involved in ecosystem formation are better equipped to act proactively and search for possible solutions to potentially overcome some of these barriers. This paper contributes to the third sub-question of this dissertation (i.e. How do institutional factors shape knowledge ecosystem formation?) as it illuminates how the institutional environment can constrain the formation of ecosystems. 4.4 Publication IV – Emergence of energy services ecosystems: Scenario method as a policy enabler Main objective The early phases of ecosystem emergence present an important phenomenon about which there is much to learn, including questions such as how highly autonomous actors begin to organise themselves around ecosystems with shared visions and purposes. Self- organising has been seen as a key characteristic of ecosystems (Peltoniemi, 2006); however, policy interventions can be helpful for ecosystems being built around new technologies and innovations (Clarysse, et al., 2014). To provide a better understanding of how ecosystem emergence can be facilitated, this paper asks how public policy initiatives enable the emergence of ecosystems. The paper draws on several case illustrations from the energy sector in Finland. The energy sector is a highly regulated and capital-intensive sector, which makes the ‘natural’ emergence of new energy services ecosystems rather challenging; therefore, it is argued that the institutional environment can act as an enabler for the emergence of new ecosystems. Main findings The main argument in this paper is that policy-driven initiatives and associated mechanisms can enable or facilitate the formation of an innovation ecosystem. The case examples exemplify how potential knowledge and resources are mobilised for new ecosystem emergence, how the relevant stakeholders can create shared understandings of the future, and what kinds of triggering mechanisms can encourage passive actors to actively engage, take risks, and commit. Mechanisms such as scenario-building workshops can foster active dialogue and interactions among the numerous actors comprising the energy sector. In practice, relevant actors are invited to workshops and participative, facilitated discussions, through which they become aware of and share their 4 Publications and synopsis of the findings 66 views on the nature and impact of future developments in the energy sector. This activates them to take dynamic roles, work together, and build the necessary synergies in the planning, execution, and assessment of specific actions to respond to or influence these developments. These types of events and facilitated discussions enable social interactions among different actors in, for example, sharing information and knowledge, building joint roadmaps, and generally co-creating effective joint responses (e.g. improving current networked processes or building new business models) to the uncertainties of energy technology development and other economic, social, regulative, and political factors. In short, all this work facilitated by policy-driven initiatives can enable actors to engage in a search for joint goals and visions and to establish shared logics, which we consider a precondition for the formation of ecosystems. Contribution This study contributes to the understanding of ecosystem formation from a public policy intervention perspective and argues for the important role it can play in the early pre- formation phases. The birth of an ecosystem has often been viewed as process designed and controlled by a strong focal actor (see e.g. Moore, 1993; Isckia, 2009; Rohrbeck, Hölzle and Gemünden, 2009); however, in the absence of such an actor, other mechanisms (such as those explicated above) become useful. Since ecosystems are complex systems with open boundaries and constant in and out flows (Cilliers, 2001), the interdependencies and co-evolutions between public policy and private sector actors become a relevant issue to investigate. This paper initiates that discussion and aims to motivate future studies on the influence of policy intervention on the formation of innovation ecosystems. In regard to this thesis, this paper contributes to the fourth sub- question of this dissertation (i.e. How can public policy drive knowledge ecosystem formation?) as it illuminates how the institutional environment (in the form of policy- driven initiatives or interventions) can enable or facilitate the formation of ecosystems. 67 Table 7. Synopsis of the individual publications and their main findings Publication I Publication II Publication III Publication IV Title In defense of ‘eco’ in innovation ecosystems Organisation of knowledge ecosystems: Prefigurative and partial forms Innovation ecosystem emergence barriers: Institutional perspective Emergence of energy services ecosystems: Scenario method as a policy enabler Objective To reflect on the ecosystem concept and its usefulness. To understand how knowledge ecosystems are organised. To identify the institutional barriers that inhibit the emergence of ecosystems and understand how they are sustained. To propose how public policy initiatives can enable the emergence of ecosystems. Main findings and contribution This paper proposes that the ecosystem concept combines prominent features from natural ecology to inform the design of system-level activities, requiring conceptual and empirical rigor. The paper proposes that the nature of knowledge search has implications for the form of organisation of the ecosystem. Particularly, the nature of knowledge search differentiates between two types: searching for a knowledge domain and searching within a knowledge domain – and respectively two forms of organisation: prefigurative and partial form. The paper also provides empirical insights into how such organising takes place. The paper identifies four ecosystem emergence barriers and related field- sustaining mechanisms and explains how these barriers mutually reinforce each other and how the institutional complexity of the energy field functions as an overarching barrier. The paper suggests that policy-driven initiatives and associated mechanisms, such as the scenario method, function as an enabler for focusing the attention of relevant actors and identifying the triggering events that guide their activities toward a shared future. Contribution to the dissertation entity Provides some conceptual basis and help to argue on the value of the ecosystem concept for making sense of interdependent Provides a better understanding on how the nature of the intermediate process of joint search shapes the organisation of Provide a better understanding on how factors in the broader institutional environment during the pre- formation phase Suggests mechanisms by which the institutional environment can facilitate the formation of knowledge 4 Publications and synopsis of the findings 68 knowledge creation and search separately. knowledge ecosystems. shape knowledge ecosystem formation. ecosystems and particularly the intermediate processes of identifying common interests and defining shared purpose. 69 5 Conclusions The main purpose of this thesis is to theorise on knowledge ecosystem formation and specifically on the institutional factors and intermediate processes that shape the knowledge ecosystem’s emergence and forms of organisation. In my endeavour to provide a comprehensive understanding on and embrace the complexities of the core phenomenon, I adopted institutional and organisational lenses and took a holistic, multi- level approach. In this section, I first answer my research questions and propose the framework depicted in Figure 4 to address my main research question on how knowledge ecosystems are formed. Then, I will discuss the overall contribution of this thesis to theory and practice, and finally I will briefly explicate the main limitations of this thesis and propose future research avenues related to these limitations as well as the core phenomenon. 5.1 Answering my research questions How are knowledge ecosystems formed? As depicted in Figure 4, ecosystems are organised to perform a joint search, and the nature of this search shapes their form of organisation. However, the formation of an ecosystem can be influenced by the constraining, enabling, and orienting effects of the wider institutional environment in phases before the joint search is initiated (i.e. pre-formation phase). The following four sub-questions help me to elaborate my findings regarding knowledge ecosystem formation, and they cover the appropriateness and importance of the knowledge ecosystem concept for studying interdependent knowledge creation and search, the organisation of knowledge ecosystems, and the influence of the institutional environment and public policy on knowledge ecosystem formation. Why is it important to consider knowledge ecosystems, and how do they relate to other types of ecosystems? The core phenomenon under investigation here is the formation of collectives of multiple actors for collaborative knowledge creation and search. In my thesis, I conceptualise these collectives as knowledge ecosystems, and one question that often comes up is ‘why?’ – ‘why is the knowledge ecosystem concept a suitable lens for your study; isn’t the ecosystem term just a buzzword or a faulty analogy?’ Therefore, the first sub-question aims to firstly challenge my choice of conceptual lens and also to contribute to the ongoing scholarly debate on the value and usefulness of the concept for studying collaborative knowledge creation and search. The ecosystem concept has a history of over 80 years, with only 26 years in business and management. During these years, the concept has received much criticism from scholars from various disciplines (ecology included, e.g. O’Neill, 2001). In the social sciences, it has been criticised, for example, for its ambiguity related to boundary definitions and its neglect of human conscious and intentional decision making (see Winterhalder, 1984); and in the business and 5 Conclusions 70 management realm, it has been characterised as a flawed analogy to natural ecosystems, with no clear definition and practical usefulness (see Oh, et al., 2016). Despite the ambiguities around the definition and usage of the ecosystem concept in business and management, I argue that the ecosystem construct is a suitable conceptual lens for studying interdependent knowledge creation and search, where multiple actors are dependent on each other and needed to co-develop the required capabilities to achieve any knowledge-creation goals. I do agree that the concept carries many assumptions from its original field that might not be directly applicable to social systems. Nevertheless, it has a unique ability to capture complex interdependences among multiple actors, multilateral dynamics, and the role of the wider institutional environment as well as phenomena such as co-evolution, collaborative innovation, and knowledge creation (Basole, 2009; Autio and Thomas, 2014; Aarikka-Stenroos and Ritala, 2017; Mars and Bronstein, 2017). As a highly versatile, scale-independent concept, it can be used to analyse a variety of social systems and diverse forms of interdependent collective action. However, the issue about the boundary definition is an important one, as is also recognised in ecology: ‘The robustness of the concept is evident from its ability to be extended across scales. However, it is important that the scale should be specified’ (Willis, 1997, p.270, emphasis added). For the analysis of any particular social system (Gulati, Puranam and Tushman, 2012), knowledge ecosystems included, boundaries must be determined empirically (cf. Scott, 2014). The literature has proposed a number of suitable indicators for defining ecosystem boundaries, including geographical scope, temporal scale, permeability, and types of flow (Scott, 2014; Valkokari, 2015; Ritala and Almpanopoulou, 2017). The problem regarding the neglect of human conscious and intentional decision making (see Winterhalder, 1984) attributed to the original concept stems from the fact that humans (or homo sapiens) are not considered a component of the ecosystem but rather an outer interference. However, this issue is under debate in ecology, with scholars arguing for humans being a keystone species that can manipulate processes and biotic structures (see e.g. O’Neill, 2001). Knowledge ecosystems, by definition, involve multiple actors, and these actors comprise people (e.g. individuals, groups, organisations), making humans a key component of these types of ecosystems; thus, there is always some intentional action or conscious choice involved (see also Peltoniemi, 2006; Valkokari, 2015). Nevertheless, to improve the concept’s applicability to business and management, we might have to relax some its original axioms and possibly develop some new ones (Oh, et al., 2016; Ritala and Almpanopoulou, 2017) while still embracing useful features such as those explicated in the previous paragraphs. As already explicitly discussed in earlier sections of this thesis, knowledge ecosystems are about various actors getting together to collectively create and deliver some research output in form of new knowledge, and they are usually location-specific (see Clarysse, et al., 2014). On the other side, innovation ecosystems encompass a broader scope of exploration and exploitation as they focus on how actors get together to deliver a joint technological or innovation goal (see e.g. Adner and Kapoor, 2010). Then, business 71 ecosystems have been established around the direct business benefits of ecosystem actors, or, put differently, the baseline of this type of ecosystem is resource exploitation for value creation and capture (Valkokari, 2015). Therefore, knowledge, innovation, and business ecosystems occur in different phases of the invention-to-commercialisation process. Finally, the entrepreneurial ecosystem refers to ‘…a distinct type of cluster that specializes in harnessing technological affordances (Gibson, 1977) created by digital technologies and infrastructures…and combines them with spatial (i.e., proximity- related) affordances to support a distinctive cluster dynamic that is expressed through the creation and scale-up of new ventures’ (Autio, et al., 2018, p.74). The entrepreneurial ecosystem is thus rather different from the other ecosystem types as it is purposefully centred around entrepreneurial activity (see also Scaringella and Radziwon, 2018). How are knowledge ecosystems organised? During ecosystem formation, actors participate in a joint search for goals and visions to establish shared logics or meanings. In this thesis (and related publication II), it is theorised that the nature of this joint search shapes the ecosystem’s form of organisation. Particularly, two types of joint search and two types of corresponding forms of organisation are identified. Knowledge ecosystems can search either for a domain or within an already identified domain wherein a solution search can start (cf. Nickerson and Zenger, 2004), thus the joint search denotes the actual contend of knowledge creation. Then, depending on the nature of the joint search the ecosystem actors engage in, the degree of organisability, organising requirements, and challenges differ. Ecosystems searching for a domain are organised in a prefigurative form, whereas ecosystems searching within an already identified domain are organised in a partial form. In the following paragraphs I will elaborate on the key organising features of both ecosystem forms. One important aspect of any organisation relates to the relationship and inputs of individual actors to the wider organisational entity, and thus participation relates to how individual actors are connected and contribute to this entity (see Puranam, Alexy and Reitzig, 2014). In prefigurative ecosystems, actors’ participation is affiliation-based and self-resourced, meaning that actors are loosely connected to the ecosystems and that their contribution is based on voluntary action (see also Berkowitz and Bor, 2017; Cropper and Bor, 2018). Another important aspect of organisations is how to coordinate individual actors’ contributions as well as the collaborative efforts of ecosystem participants (see e.g. March and Simon, 1993; Puranam, Alexy and Reitzig, 2014). Without the presence of organising elements, coordination in prefigurative ecosystems is informal. Nevertheless, participants foresee the introduction of some elements of organising, such as for membership and monitoring. The role of a clear common goal and direction is important as it enables the ecosystem participants to connect, collaborate, and shape the focus of the collective (see also Moore, 1996; Iansiti and Levien, 2004). In prefigurative ecosystems, actors’ contributions and collaborative work are coordinated to formulate a common goal and identify a joint domain. However, a main organising challenge is the ambiguity related to decisions being in flux and the diverse expectations among 5 Conclusions 72 participants, often waiting for each other to take initiative and set the direction (see e.g. Pellinen, et al., 2012; Dattée, Alexy and Autio, 2018). Ecosystems that are partially organised have a distinct character and exercise their collective actorhood. In these ecosystems, the joint search occurs within an already identified domain, with actors engaging in solution-oriented knowledge-creation work. Actors’ participation is membership-based. Namely, ecosystem members have the right to use collective resources, are monitored for their contributions, and are bound to the ecosystem by their membership. Thus, organising elements such as rules and structures for monitoring are in place and used for the coordination of the joint knowledge-creation work. Specifically, coordination relies on formal monitoring and regulation based on predetermined criteria and plans. Despite the higher degree of organisability and organising elements being present, these ecosystems can face a diverse set of organising challenges related to individual actors’ distinctive plans and goals, which can interfere with the collective efforts. To avoid being caught up in working in silos, it is crucial for the ecosystem to find an appropriate balance between individual goals and the common goal to which ecosystem members can feel connected and through which they can achieve their own goals and plans (see also e.g. Nambisan and Baron, 2013). How do institutional factors shape knowledge ecosystem formation? What is happening during the pre-formation phase and the specific socio-economic context in which actors are embedded during this time can influence the emergence or non-emergence of an ecosystem and can leave certain opportunities unrealised (Kirsch, et al., 2014; Gustafsson, et al., 2016). Grounded on empirical evidence, I find a strong and interdependent set of institutional barriers at the field level that mutually reinforce each other, namely regulation and policymaking ambiguity, incumbent actor inertia, cognitive constraints on opportunity recognition, and institutional complexity constraining ecosystem formation. Institutional complexity functions as an overarching barrier that further sustains the rigidity of incumbent actors and policymakers and creates additional cognitive constraints on opportunity recognition. In the following paragraphs, I will elaborate on the dynamic influences between the identified field-level barriers. Incumbent actors’ reluctance to change due to their path-dependent pasts (i.e. ‘incumbent actor inertia’) fortifies the cognitive constraints on opportunity recognition. Specifically, taken-for-granted assumptions; expectations; the legitimacy of the existing business logics (Markard, Wirth and Truffer, 2016; Suddaby, Bitektine and Haack, 2017); and the alleviating effect of established, commonly accepted norms (Scott, 2014) obstruct them from stepping out of their comfort zone and recognising opportunities for new business endeavours. Reversely, the identified cognitive constraints can feed into incumbent companies’ incapability to overcome organisational inertia (see Gilbert, 2005). For example, companies do not see the opportunity to develop offerings for customers to support, for example, energy efficiency or demand response due to the perceived market 73 uncertainties related to low electricity prices and lack of demand. Therefore, as they do not feel there is enough incentive and business value, they are not motivated to change, and thus inertia is intensified; in other words, companies do not respond to the transforming environment but instead linger with hitherto learned practices (see also Gilbert, 2005; Dutton and Jackson, 1987; Staw, Sandelands and Dutton, 1981). The uncertainty associated with the policy environment creates additional constraints on opportunity recognition. For instance, the discontinuity between governments inhibits energy companies in regard to determining which investments are sensible and viable. Furthermore, the lack of clear intentions and decisions regarding concrete actions from the policy side makes indistinct where the best opportunities lie and how actors should make use of common investments or what the nature of the services should be. Cognitive constrains for opportunity recognition can also further intensify policy ambiguities. For example, a basic assumption in the energy sector is that energy production must occur on a large scale, leading to policy putting emphasis on very large, centralised energy production – therefore, slowing down policymaking that could promote, for example, distributed, renewable energy generation, around which new ecosystems could or should form. One interesting finding of this thesis is that the complexity of the broader environment can function as an overarching barrier that sustains the previously identified regulative, normative, and cognitive constrains. With regard to incumbent actor inertia and the cognitive constraints on opportunity recognition, many of the actors in the energy sector lack an overall vision or understanding of what is required for a system-level transition. Incumbent actors become unable to overcome their organisational inertia and drive change in the existing field. For smaller actors with less bargaining power, the field’s complexity makes it challenging to recognise and tap into opportunities. Further, institutional complexity can fortify the ambiguities in policymaking since policymakers have to contemplate multiple objectives and often conflicting issues in their decision making. Nevertheless, it is important to note that institutional complexity might also function as a trigger for the development of new initiatives to begin with (see e.g. Smets and Jarzabkowski, 2013; Siltaloppi, Koskela-Huotari and Vargo, 2016). Thus, there is certainly no ‘black or white’ regarding the influences of the wider environment and related institutions. In relation to this remark, in the next section, I will discuss how the institutional environment can enable and orient ecosystem formation. How can public policy drive knowledge ecosystem formation? In this thesis, I argue that the institutional environment can have an enabling and orienting effect on the formation of ecosystems. Particularly, policy intervention, in the form of publicly funded strategic research initiatives focusing on topical issues in the field, can facilitate the initial social interaction among potential ecosystem actors by creating forums for collective action and initiative taking, which in turn can enable and orient ecosystem formation. 5 Conclusions 74 The enabling effects of the institutional environment have been widely discussed in the institutional literature, mainly referring to how the institutional environment makes actions possible or makes actors aware of possibilities for actions that were imperceptible before (Seo and Creed, 2002). However, enabling entails not only that the wider environment creates possibilities for actions but also that it actively encourages and guides actors’ actions to certain possibilities rather than others (Selznick, 1957; Thornton, Ocasio and Lounsbury, 2012; Scott, 2014). The latter effect of guiding was recently understood as orienting and conceptually separated from enabling (see Cardinale, 2018) to highlight that while settling on specific possibilities actors are still influenced by the institutional environment, and thus this process is not only based on a reflective evaluation of the enabled possibilities of action. In other words, ‘action cannot be reduced to choice among alternatives posited as such but also reflects ways of being and acting that are relatively enduring’ (Cardinale, 2018, p.142). In this thesis, I view these effects (enabling and orienting) of the institutional environment as conceptually distinct yet interwoven and parallel. Policy-driven research initiatives (such as the illustrated cases in publication IV) play an essential role in facilitating social interaction and the creation and exchange of knowledge among multiple actors (see also Autio, et al., 2008). Events such as scenario-building workshops or other types of participative, facilitated discussions foster active dialogue and social interaction among potential ecosystem actors, through which they become aware of and share their views on the nature and impact of future developments, build joint roadmaps, and generally co-create effective joint responses. This activates them to take on dynamic roles, work together, and build the necessary synergies in the planning, execution, and assessment of specific actions to respond to or influence these developments. Social interactions allow actors to layer their various meanings in ways that might result in unforeseen outcomes, functioning therefore as a key mechanism and source of shared meanings (Leibel, Hallett and Bechky, 2018). In short, the facilitated social interaction by policy-driven initiatives can enable actors to engage in a search for joint goals and visions and to establish shared logics, which I consider a precondition for the formation of ecosystem(s). Finally, regarding the orienting effect of the institutional environment, policy-driven initiatives have specific objectives and focus areas. Thus, social interaction facilitated under those initiatives not merely opens up all the possible opportunities for the participating actors but also guides or orients their interaction toward a narrower set of possibilities, depending on the aims and focus of the initiative. To give an example from the illustrated case in publication IV, the Neo-Carbon Energy project (funded by the former Finnish Funding Agency for Innovation) focused on the requirements for and opportunities of an emission-free, renewable energy system; therefore, the social interaction in, for example, scenario-building workshops was oriented toward this vision, leaving out traditional energy systems or energy sources. Orienting, as a distinct effect of the institutional environment, is a recent advancement in institutional theory, so there is still a great deal to be understood. Nevertheless, it provides much more insight into the 75 constraining and enabling dichotomy and much more clarity as to what enabling really means (see Cardinale, 2018). 77 Figure 4. Knowledge ecosystem formationTheoretical implications Conclusions 78 5.2 Theoretical implications This thesis contributes to three main literature streams: the knowledge ecosystem literature, the literature on non-traditional forms of organisation, and the institutional literature. 5.2.1 Knowledge ecosystem literature Conceptual discussion Considering the increased scholarly interest in the concept, this thesis attempted to distil the most important aspects of the ecosystem concept in explaining phenomena such as collaborative knowledge creation and search. In this endeavour, it contributes to recent efforts to explicate the foundations and usefulness of the concept (e.g. Clarysse, et al., 2014; Valkokari, 2015) as well as responds to calls for more reflection regarding its conceptual, methodological, and theoretical underpinnings (see Ritala and Gustafsson, 2018). By unpacking the terminology and the key features of the original concept in ecology, I have argued why the ecosystem concept is useful in business and management and particularly in the study of collaborative, interdependent knowledge creation and search. Furthermore, in this thesis, I have contributed to the conceptual development of the knowledge ecosystem concept. Particularly, in publication II, knowledge ecosystems are described as ‘…users and producers of knowledge, organized around joint knowledge search’ (Järvi, Almpanopoulou and Ritala, 2018, p.1533). In publication I, we highlighted the importance of considering temporal and spatial boundaries as core components of the ecosystem concept (Ritala and Almpanopoulou, 2017, p.41). Context and dominant approach Furthermore, this thesis has contributed to the understanding on the organisation of ecosystems in such contexts, where organising is not defined or enabled by a technological platform or a powerful focal actor. Much emphasis in the ecosystem literature has been placed on top-down organising by focal actors in stable and already established ecosystems in the ICT industry, often organised around technological platforms (see e.g. Iyer and Davenport, 2008; Li, 2009; Rao, 2016; Dattée, Alexy and Autio, 2018). This has left relatively unexplored ecosystems in other more traditional industries, or ecosystems that are not organised around technological platforms. While platforms need ecosystems to surround them in order to be successful, ecosystems do not necessarily have platforms at their core (Autio and Thomas, 2014). In this thesis, I have studied ecosystem formation in the Finnish energy sector as well as in the context of collaborative knowledge creation in pre-competitive, pre-commercialisation settings, particularly cross-industry (ICT, health, energy, and construction) research programmes in Finland. The energy sector is certainly influenced by digitalisation, however the industry is less modular, highly regulated, and capital intensive compared to ICT; and as this thesis shows, actors face a different set of challenges and complexities that have an impact on the formation of ecosystems. Conclusions 79 In addition, the focus of this thesis on knowledge ecosystems in the context of the Finnish SHOKs, characterised by interdependence among actors in knowledge integration, complements prior literature’s emphasis on interdependence among ecosystem actors from a technological perspective (see e.g. Wareham, Fox and Cano Giner, 2014) or from the customer perspective (see e.g. Williamson and De Meyer, 2012). Interdependences related to the integration of knowledge work across knowledge and organisational boundaries have been argued to restrain (Ethiraj and Posen, 2013) but also to boost new knowledge creation (Bruns, 2013; Perry, Hunter and Currall, 2016). This thesis and related publication II have shown how knowledge ecosystems organise for knowledge integration tasks; how, why, and to what extent ecosystem actors participate in knowledge creation; and how this participation is regulated as well as what rights and responsibilities are included. Phenomenon of ecosystem formation: An institutional and organisational perspective The existence of an ecosystem is usually taken for granted, and scholarly work has not paid enough attention to the earlier stages of ecosystem formation (including pre- formation) (Kirsch, et al., 2014; Valkokari, 2015; Overholm, 2015; Barile, et al., 2016; Autio and Thomas, 2018; Dattée, Alexy and Autio, 2018). This thesis enhances the understanding of the early stages of ecosystem formation and the influence of the socio- economic context during the pre-formation phase. Thus, it responds to recent calls for more comprehensive understanding on (1) the influence of the wider institutional environment on ecosystem formation (see Autio and Thomas, 2014; Barile, et al., 2016; Autio, et al., 2017; Möller and Halinen, 2017; Seidel and Greve, 2017) and contributes to the (2) theoretical need to understand the organising processes and elements in ecosystems. By taking an institutional perspective, I have shown (as discussed in section 5.1) how the institutional environment can influence ecosystem formation. Particularly, I have shown how the constraining, enabling, and orienting effects of the wider environment and related institutions manifest in ecosystem formation. The establishment of institutional legitimacy across potential ecosystem actors is viewed as a key aspect of ecosystem creation and evolution (Aldrich and Fiol, 1994; Sine, et al., 2007; Autio and Thomas, 2018). While acknowledging and contributing to the discussion on the importance of facilitating the development of shared meanings and logics across potential ecosystem actors, this thesis shows that established legitimacy in a given field (e.g. energy sector) – that is, perceived congruence with its rules and normative values or alignment with its cultural-cognitive frameworks (Scott, 2014) – can also function as a barrier to the emergence of new ecosystems. In other words, the social acceptability and credibility resulting in the perceived solidity, coherency, and predictability of established ecosystems can simultaneously hinder the emergence of new ecosystems. By integrating the previously distinct literature streams on non-traditional forms of organisation (e.g. Ahrne and Brunsson, 2011; Gulati, Puranam and Tushman, 2012; Dobusch and Schoeneborn, 2015) in the ecosystem literature, this thesis provides a Conclusions 80 comprehensive understanding on ecosystems as forms of organisation. In comparison to traditional organisations with clearly defined boundaries and hierarchical structures (Kellogg, Orlikowski and Yates, 2006), ecosystem organisation can be described as fluid (cf. Dobusch and Schoeneborn, 2015) and uncompleted or continuously evolving (Barry and Rerup, 2006). This thesis has shown how ecosystems can be coordinated in the absence of hierarchical structures (cf. Ahrne and Brunsson, 2011) and technological platforms and how despite their fluid organisation they can function efficiently while preserving the independence of individual actors (see also Felin and Zenger, 2014). Furthermore, this thesis supports earlier findings that view ecosystems forming as an outcome of a collaborative exploration process (in my case, a joint search) and the strategic agency of multiple actors (see Attour and Barbaroux, 2016). 5.2.2 Literature on non-traditional forms of organisation Even though there is sufficient understanding about how non-traditional forms of organisation, such as meta-organisations, differ from conventional organisations (i.e. the latter feature formal authority, hierarchical structures, employment contracts, etc.), there is much less understanding about different types of meta-organisations and how and why they vary. In this thesis and related publication II, I argued that knowledge ecosystems can be understood as a particular type of meta-organisation that incorporate mechanisms enabling knowledge creation and sharing (e.g. Snow, et al., 2011) and allow for collective action while maintaining the actors’ autonomy (Sydow, et al., 2012; Berkowitz and Bor, 2017; Cropper and Bor, 2018). Therefore, this thesis contributes to recent calls (see Berkowitz and Bor, 2017) for bridging the literature streams. Particularly, it advances our understanding on the organising particularities of knowledge ecosystems as a type of meta-organisation while further refining the division between partial and complete organisations (Ahrne and Brunsson, 2011) with the conceptualisation of the prefigurative form of organisation. This thesis also touches upon the concept of actorhood; therefore, it contributes to recent discussions on various degrees of organisationality and the existence of organisation without actorhood (see e.g. Dobusch and Schoeneborn, 2015; Grothe-Hammer, 2018). Prefigurative ecosystems lack a distinctive character and are not yet externally perceived as autonomous actors (e.g. by funding organisations). Nevertheless, despite the lack of externally recognised actorhood, some degree of organisationality is achieved because of the interconnected decision-making processes during the identification of a common goal as well as the presence of informal coordinating mechanisms and the expectation of formal organising elements. 5.2.3 Institutional literature This thesis contributes to the institutional literature particularly focusing on the institutional dynamics behind the emergence of new technologies and innovations (Geels, 2002; Geels and Schot, 2007; Dijk, Wells and Kemp, 2016; Markard, Wirth and Truffer, 2016). This literature has highlighted that, related to the emergence of new technology, it Conclusions 81 is not enough to focus on technological changes only but that one must also pay attention to issues such as user practices, regulation, and infrastructures as well as symbolic and cultural issues at the field level (e.g. Geels, 2002, 2004; Geels and Schot, 2007; Markard, Wirth and Truffer, 2016). In this thesis, I have identified a variety of field-sustaining mechanisms mutually reinforcing each other, which broadly correspond to regulative, normative, and cognitive legitimacy (cf. Scott, 2014; Markard, Wirth and Truffer, 2016; Suddaby, Bitektine and Haack, 2017). These findings contribute to recent calls for more research to understand field transformation and related constraining forces (cf. Zietsma, et al., 2017). Finally, my thesis contributes to the recent development in the institutional literature to suggest orienting as a distinct effect of the institutional environment (see Cardinale, 2018). Even though, there is still much to be understood. My thesis presents a specific expression of the orienting influence of the institutional environment through narrowly focused policy-driven initiatives. 5.3 Policy and practice implications This thesis also has significant contributions for policy and practice. Forming and sustaining configurations of many often powerful and highly interconnected actors are not easy tasks. Policymakers and practitioners have to be able to recognise the institutions (regulative, normative, cognitive) that can hamper innovation and the emergence of ecosystems supporting it. Awareness of the institutional barriers identified in this thesis can prepare practitioners for what to expect in their knowledge creation and innovation endeavours. Innovation and renewal in the current fast-moving era are only possible when key actors join forces for collective strategic action. Although self-organising is an inherent characteristic of ecosystems (Peltoniemi, 2006), my study shows that policy interventions might be essential for facilitating these processes (see also Clarysse, et al., 2014). Policy-driven research initiatives can help to overcome ambiguities related to initiative taking and the uncertainties regarding who should be setting the direction and planning concrete actions. The mechanisms described in this thesis facilitate interactions among potential ecosystem actors, which can further enable the development of shared meanings and logics that can initiate the formation of a more concrete ecosystem. Single organisations often struggle to keep up or are simply not prepared to respond to the uncertainties and challenges of today’s settings (Furr and Shipilov, 2018), which makes them interdependent with others and their environment. Knowledge ecosystems can be viewed as the means for regional and national growth as well as for boosting competitiveness and accelerating innovation (Ritala and Gustafsson, 2018) for solving large-scale and highly complex societal problems. Considering the importance of ecosystems in addressing the complexities of the current era, a significant amount of national resources is usually distributed to implement such initiatives; therefore, it becomes imperative to have a clear understanding on how to successfully execute them, such as what type of funding instruments are appropriate and can facilitate their development. For example, ecosystems searching for their domain and joint goal could benefit from financing instruments that aim to support actors’ participation in goal-setting or research idea development and cooperation-building initiatives. Business Finland Conclusions 82 (www.businessfinland.fi) has recently launched such a funding instrument, targeted at research organisations for formulating common innovation goals and creating new collaborations with companies. Considering the very early stage of this development in Finnish policy, it is not yet visible what its actual impact on ecosystem formation will be; however, it is a significant advancement in recognising the importance of resourcing early cooperation building initiatives. Alternatively, the role of R&D funding instruments is highly important for ecosystems searching within an already identified domain; as it can accelerate the development of innovations and solutions, policymakers with open dialogue and in a cooperative manner with ecosystem actors should determine which domains or goals are strategic and nationally worthy to justify public funding. For practitioners, company managers, and university representatives it is important to participate in multiple knowledge ecosystems in order to maximise the opportunity of finding valuable knowledge and accomplishing ambitious knowledge-creation goals. Considering the uncertainty in early phases knowledge ecosystems participating in multiple initiatives ensures that its knowledge-creation goals will be met. My findings also suggest that participating actors have to be active in the early phases when the collective is searching for the common knowledge domain and the overarching common goal. Being active will allow single actors to significantly influence the direction of ecosystem-level goals and make sure that they can accomplish their own goals. Finally, the findings of this thesis help in recognising the necessary organising requirements and related challenges for reaching (and maintaining) this joint understanding and shared vision among ecosystem actors and inform practitioners about the necessary management capabilities to navigate through these challenges. Prior literature (e.g. Clarysse, et al., 2014; Sinnewe, Charles and Keast, 2016) has highlighted the difficulties in maintaining collaboration among diverse actors (e.g. between industry and academic institutions) once public funding ends or when the ecosystem has to move from the pre-competitive, pre- commercial stage to the commercialisation stage. Thus, management capabilities related to, for example, resourcing and governance mechanisms enabling joint knowledge creation activities to be coordinated cost-effectively are essential for coherent, enduring collaboration within the ecosystem. 5.4 Limitations and future research This thesis carries limitations related to context-specificity and generalisability that are typically attributed to qualitative research. Nevertheless, the lack of generalisability in a statistical sense does not lower the analytical value of the findings of this thesis as it enhances our understanding of the studied phenomena and helps to extend existing theoretical contributions (see Yin, 2003). It is also important to mention here that statistical generalisability is not (usually) the primary goal for qualitative research; and particularly in the case of this thesis, given its constructivist premises, its inherent subjectivity and context-specificity are considered strengths rather than limitations (see also Weick, 1995; Willis 2007; Halinen, Törnroos and Elo, 2013; Bouchikhi, 1998; Peters, et al., 2013). Nonetheless, as there is still a great deal to learn about the formation of knowledge ecosystems and how the institutional environment can influence those Conclusions 83 processes, future studies could focus on contexts that are less investment-intensive or less heavily state-regulated industries compared to the energy sector studied in this thesis. Comparative analyses of different empirical settings as well as types of ecosystems could further advance our understanding on how ecosystems and their organisation are formed and how diverse features, requirements, and characteristics affect these processes. One limitation of this thesis is the lack of empirical data related to the enabling and orienting effects of the institutional environment on ecosystem formation. Nevertheless, I hope that this thesis will spark scholarly interest to empirically study these influences of the wider environment. Furthermore, an interesting future research avenue would be to study whether, when, and why certain features of the institutional environment or certain institutions can have multiple concurrent influences (i.e. constraining, enabling, and orienting) and how these manifest. For example, an interesting finding in this thesis and related publication III was that institutional complexity acted as a major barrier to ecosystem formation. However, previous literature has suggested that institutional complexity operates as a trigger for the development of new initiatives to begin with (see e.g. Smets and Jarzabkowski, 2013; Siltaloppi, Koskela-Huotari and Vargo, 2016). This dual role of complexity (and any other possible such instances) presents a fascinating topic for future research endeavours. In addition, there is still a great deal to be learned about the pre-formation stage, also particularly in regard to non-emergence, as previous research has mainly recognised the importance of this period and has primarily documented the existence of the pre-formation stage in cases of ‘successful’ emergence (Agarwal and Bayus, 2002; Golder, Shacham and Mitra, 2009; Kirsch, et al., 2014; Gustafsson, et al., 2016). Understanding this particular period especially in cases of non- emergence can help in avoiding similar drawbacks for future ecosystem formation endeavours or help relevant actors to be better prepared or equipped to respond to those as they emerge. Finally, there is still much to understand about ecosystems as forms of organisation. Future research could adopt a longitudinal process perspective to study how ecosystems and their organisation evolve and how particular organising elements change over time. An interesting future direction would also be to study the transition from organisation without actorhood (first degree of organisationality) to one with actorhood (second degree of organisationality) or one with an established collective identity (third degree of organisationality). To conclude, a topic for future investigation is also whether and how ecosystems, characterised by loosely coupled connections, can develop and sustain complex and enduring organisation without actorhood (see also Grothe-Hammer, 2018). 85 References Aarikka-Stenroos, L. and Ritala, P., 2017. Network management in the era of ecosystems: Systematic review and management framework. Industrial Marketing Management, 67, pp.23-36. Adner R., 2006. Match your innovation strategy to your innovation ecosystem. Harvard Business Review, 84(4), pp.98-107. Adner, R., 2012. The wide lens: A new strategy for innovation. London: Penguin. Adner, R., 2017. Ecosystem as structure: An actionable construct for strategy. Journal of Management, 43(1), pp.39-58. Adner, R. and Kapoor, R., 2010. Value creation in innovation ecosystems: How the structure of technological interdependence affects firm performance in new technology generations. Strategic Management Journal, 31, pp.306-33. Adner, R. and Kapoor, R., 2016. Innovation ecosystems and the pace of substitution: Re‐ examining technology S‐curves. Strategic Management Journal, 37, pp.625-48. Agarwal, R. and Bayus, B. L., 2002. The market evolution and sales takeoff of product innovations. Management Science, 48, pp.1024-41. Ahrne, G. and Brunsson, N., 2005. Organizations and meta-organizations. Scandinavian Journal of Management, 21, pp.429-49. Ahrne, G. and Brunsson, N., 2011. Organization outside organizations: The significance of partial organisation. Organisation, 18, pp.83-104 Ahrne, G., Brunsson, N. and Seidl, D., 2016. Resurrecting organization by going beyond organisations. European Management Journal, 34(2), pp.93-101. Akaka, M. A., Vargo, S. L. and Lusch, R. F., 2013. The complexity of context: A service ecosystems. Journal of Marketing Research, 21(4), pp.1-20. Aldrich, H. E. and Fiol, C. M., 1994. Fools rush in? The institutional context of industry creation. Academy of Management Review, 19, pp.645-70. Alexy, O., George, G. and Salter, A. J., 2013. Cui bono? The selective revealing of knowledge and its implications for innovative activity. Academy of Management Review, 38, pp.270-91. References 86 Allen, T. F. H. and Hoekstra, T. W., 1992. Towards a unified ecology. New York: Columbia University Press. Anderson, P., 1999. Complexity theory and organization science. Organ. Sci., 10(3), pp.216-32. Atlas.cern, 2019. The collaboration. Available at http://atlas.cern/discover/collaboration [accessed November 2018]. Attour, A. and Barbaroux, P., 2016. Architectural knowledge and the birth of a platform ecosystem: A case study. Journal of Innovation Economics & Management, 1(19), pp.11- 30. Autio, E., Kanninen, S. and Gustafsson, R., 2008. First-and second-order additionality and learning outcomes in collaborative R&D programs. Research Policy, 37(1), pp.59- 76. Autio, E., Nambisan, S., Thomas, L. D. and Wright, M., 2018. Digital affordances, spatial affordances, and the genesis of entrepreneurial ecosystems. Strategic Entrepreneurship Journal, 12(1), pp.72-95. Autio, E. and Thomas, L. D. W., 2014. Innovation ecosystems: Implications for innovation management. In: M. Dodgson, D. M. Gann, and N. Phillips, eds. Oxford handbook of innovation management. Oxford: Oxford University Press. pp.204-228. Autio, E. and Thomas, L.D.W., 2018. Tilting the Playing Field: Towards an Endogenous Strategic Action Theory of Ecosystem Creation. In: S. Nambisan, eds. Open innovation, innovation ecosystems, and entrepreneurship: multidisciplinary perspectives. New Jersey: World Scientific Publishing. pp. 111-40. Bain, J. S., 1959. Industrial organization: A treatise. New York, NY: John Wiley. Baiyere, A., 2018. Fostering innovation ecosystems: Note on the ISPIM innovation forum. Technovation, 69, p.1. Bansal, P., Smith, W. K. and Vaara E., 2018. New ways of seeing through qualitative research. Academy of Management Journal, 61(4), pp.1189-95. Barile, S., Lusch R., Reynoso, J. and Saviano M., 2016. Systems, networks and ecosystems in service research. Journal of Service Management, 27(4), pp.652-74. Barley, W. C., 2015. Anticipatory work: How the need to represent knowledge across boundaries shapes work practices within them. Organization Science, 26, pp.1612-28. References 87 Barnett, W. P., 1990. The organizational ecology of a technological system. Administrative Science Quarterly, 35, pp.31-60. Barry, D. and Rerup, C., 2006. Going mobile: Aesthetic design considerations from Calder and the constructivists. Organization Science, 17, pp.262-76. Basole, R. C., 2009. Visualization of interfirm relations in a converging mobile ecosystem. Journal of Information Technology, 24(2), pp.144-59. Battilana, J. 2006. Agency and Institutions: The enabling role of individuals’ social position. Organization Studies, 18, pp.93-117. Battilana, J., Leca, B. and Boxenbaum, E. 2009. How actors change institutions: Towards a theory of institutional entrepreneurship. Academy of Management Annals, 3(1), pp.65-107. Bechky, B. A., 2003. Sharing meaning across occupational communities: The transformation of understanding on a production floor. Organization Science, 14, pp.312- 30. Berkowitz, H. and Bor, S., 2018. Why meta-organizations matter: A response to Lawton et al. and Spillman. Journal of Management Inquiry, 27(2), pp.204-11. Berkowitz, H. and Dumez, H., 2016. The concept of meta-organization: Issues for management studies. European Management Review, 13(2), pp.149-56. Berthod, O., 2016. Institutional theory of organizations. In: A. Farazmand, ed. Global encyclopedia of public administration, public policy, and governance. doi:10.1007/978- 3-319-31816-5_63-1. Blumer, H., 1954. What is wrong with social theory? American Sociological Review, 18, pp.3-10. Borghini, S., Caru, A. and Cova, B., 2010. Representing BtoB reality in case study research: Challenges and new opportunities. Industrial Marketing Management, 39(1), pp.XX. Bose, A., 2010. Smart transmission grid applications and their supporting infrastructure. Trans. Smart Grid, 1(1), pp.11-19. Bouchikhi, H., 1998. Living with and building on complexity: A constructivist perspective on organizations. Organization, 2, pp.217-32. Bowen, G. A., 2006. Grounded theory and sensitising concepts. International Journal of Qualitative Methods, 5(3), pp.12-23. References 88 Brandenburger, A. M. and Nalebuff, B. J., 1996. Co-opetition. New York, NY: Crown Business. Bruns, H. C., 2013. Working alone together: Coordination in collaboration across domains of expertise. Academy of Management Journal, 56, pp.62-83. Burrell, G. and Morgan, G., 1979. Sociological paradigms and organisational analysis. London: Heinemann. Cardinale, I., 2018. Beyond constraining and enabling: Toward new microfoundations for institutional theory. Academy of Management Review, 43(1), pp.132-55. Carlile, P. R., 2002. A pragmatic view of knowledge and boundaries: Boundary objects in new product development. Organization Science, 13, pp.442-55. Carlile, P. R., 2004. Transferring, translating, and transforming: An integrative framework for managing knowledge across boundaries. Organization Science, 15, pp.555-68. Ceccagnoli, M., Forman, C., Huang, P. and Wu, D.J., 2012. Cocreation of value in a platform ecosystem: The case of enterprise software. MIS Quarterly: Management Information Systems, 36(1), pp.263-90. Charmaz, K., 2003. Grounded theory. In: J. A. Smith, ed., Qualitative psychology: A practical guide to research methods. London: Sage. pp.81-110. Charmaz, K., 2014. Constructing grounded theory. London, UK: SAGE Publications Ltd. Chesbrough, H. W., 2006. Open innovation: The new imperative for creating and profiting from technology. Boston, MA: Harvard Business Press. Christensen, C. M. and Rosenbloom, R. S., 1995. Explaining the attacker’s advantage: Technological paradigms, organizational dynamics, and the value network. Research Policy, 24, pp.233-57. Cilliers, P., 2005. The architecture of complexity (introduction). Émerg.: Complex. Organ., 7(3-4), pp.138-54. Clarysse, B., Wright, M., Bruneel, J. and Mahajan, A., 2014. Creating value in ecosystems: Crossing the chasm between knowledge and business ecosystems. Res. Policy, 43(7), pp.1164–76. Cooke, P., 2001. Regional innovation systems, clusters, and the knowledge economy. Industrial and Corporate Change, 10(4), pp.945-74. References 89 Cropper, S. and Bor, S., 2018. (Un)bounding the meta-organization: Co-evolution and compositional dynamics of a health partnership. Administrative Sciences, 8(3), pp.1-19. Cusumano, M. A. and Gawer, A., 2002. The elements of platform leadership. MIT Sloan Manag. Rev. 43(3), p.51. Dattée, B., Alexy, O. and Autio, E., 2018. Maneuvering in poor visibility: How firms play the ecosystem game when uncertainty is high. Academy of Management Journal. doi:10.5465/amj.2015.0869. Davis, G. F. and Marquis, C., 2005. Prospects for organization theory in the early twenty- first century: Institutional fields and mechanisms. Organization Science, 16(4), pp.332- 43. Davis, J. P., 2016. The group dynamics of interorganizational relationships: Collaborating with multiple partners in innovation ecosystems. Administrative Science Quarterly, [e- journal] doi:10.1177/0001839216649350. Davis, J. P. and Eisenhardt, K. M., 2011. Rotating leadership and collaborative innovation: Recombination processes in symbiotic relationships. Administrative Science Quarterly, 56, pp.159-201. Davis, J. P., Eisenhardt, K. M. and Bingham, C. B., 2007. Developing theory through simulation methods. Acad. Manag. Rev., 32, pp.480-99. Denzin, N. K. and Lincoln, Y. S., 2005. Introduction: The discipline and practice of qualitative research. In: N. K. Denzin and Y. S. Lincoln, eds. The Sage handbook of qualitative research. 3rd ed. Thousand Oaks, CA: Sage. pp.1-32. Dey, I., 1999. Grounding grounded theory. San Diego, CA: Academic Press. Dhanaraj, C. and Parkhe, A. 2006. Orchestrating innovation networks. Academy of Management Review, 31(3), pp. 659–669. Dijk, M., Wells, P. and Kemp, R., 2016. Will the momentum of the electric car last? Testing an hypothesis on disruptive innovation. Technological Forecasting and Social Change, 105(April), pp.77-88. Dill, W. R., 1958. Environment as an influence on managerial autonomy. Administrative Science Quarterly, 2, pp.409-43. DiMaggio, P. J. and Powell, W. W., 1983. The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields. American Sociological Review, 48, pp.147-60. References 90 Dobusch, L. and Schoeneborn, D., 2015. Fluidity, identity, and organizationality: The communicative constitution of Anonymous. Journal of Management Studies, 52(8), pp.1005-35. Dougherty, D. and Dunne, D. D., 2011. Organizing ecologies of complex innovation. Organization Science, 22, pp.1214-33. Drepper, T., 2005. Organization and society. In: D. Seidl and K. H. Becker, eds. Niklas Luhmann and organization studies. Copenhagen: Copenhagen Business School Press. pp.171-90. Dutton, J. and Jackson, S., 1987. Categorizing strategic issues: Links to organizational action. Academy of Management Review, 12, pp.76-90. Ebers, M. (1997). Explaining inter-organizational network formation. In: M. Ebers, ed. The formation of inter organizational networks. Oxford: Oxford University Press. pp.3- 40). Eisenhardt, K. M., 1989. Building theories from case study research. Academy of Management Review, 14, pp.532-50. Ethiraj, S. and Puranam, P., 2004. The distribution of R&D effort in systemic industries: Implications for competitive advantage. Advances in Strategic Management, 21, pp.225- 53. Ethiraj, S. K. and Posen, H. E., 2013. Do product architectures affect innovation productivity in complex product ecosystems? Advances in Strategic Management, 30, pp.127-66. Evans, F. C., 1956. Ecosystem as the basic unit in ecology. Science , 123(3208), pp.1127- 28. Felin, T. and Zenger, T. R., 2014. Closed or open innovation? Problem solving and the governance choice. Research Policy, 43, pp.914-25. Fjeldstad, Ø. D., Snow, C. C., Miles, R. E. and Lettl, C. 2012. The architecture of collaboration. Strategic Management Journal, 33, pp. 734-750. Franzoni, C. and Sauermann, H., 2014. Crowd science: The organization of scientific research in open collaborative projects. Research Policy, 43, pp.1-20. Furr, N. and Shipilov, A., 2018. Building the right ecosystem for innovation. MIT Sloan Management Review, Summer, pp.59-64. References 91 Garnsey, E. and Leong, Y. Y., 2008. Combining resource-based and evolutionary theory to explain the genesis of bio-networks. Industry and Innovation, 15(6), pp.669-86. Garnsey, E., Lorenzoni, G. and Ferriani, S., 2008. Speciation through entrepreneurial spin-off: The Acorn-ARM story. Research Policy, 37(2), pp.210-24. Garud, R., Gehman, J. and Giuliani, A. P., 2014. Contextualizing entrepreneurial innovation: A narrative perspective. Research Policy, 43, pp.1177-88. Garud, R., Jain, S. and Tuertscher, P., 2008. Incomplete by design and designing for incompleteness. Organization Studies, 29, pp.351-71. Gawer, A., 2014. Bridging different perspectives on technological platforms: Toward and integrative framework. Research Policy, 43, pp.1239-49. Gawer, A. and Cusumano, M. A., 2002. Platform leadership: How Intel, Microsoft, and Cisco drive industry innovation. Boston, MA: Harvard Business Press. Gawer, A. and Cusumano, M. A., 2008. How companies become platform leaders. MIT Sloan Management Review, 49(2), pp.28-35. Geels, F. W., 2002. Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study. Research Policy, 31, pp.1257-74. Geels, F. W., 2004. From sectoral systems of innovation to socio-technical systems: Insights about dynamics and change from sociology and institutional theory. Research Policy, 33, pp.897-920. Geels, F. W. and Schot, J., 2007. Typology of sociotechnical transition pathways. Research Policy, 36, pp.399-417. Geertz, C., 1973. The interpretation of cultures. New York: Basic Books. Gibson, J. J., 1977. The theory of affordances. In: R. E. Shaw and J. Bransford, eds. Perceiving, acting and knowing. Hillsdale, NJ: Lawrence Erlbaum Associates. pp.67-82. Giddens, A., 1984. The constitution of society: Outline of the theory of structuration. Cambridge: Polity Press. Gilbert, C. G., 2005. Unbundling the structure of inertia: Resource versus routine rigidity. Academy of Management Journal, 48(5), pp.741-63. Golder, P. N., Shacham, R. and Mitra, D., 2009. Innovations’ origins: When, by whom, and how are radical innovations developed? Marketing Science, 28(1), pp.166-79. References 92 Golley, F. B., 1993. A history of the ecosystem concept in ecology: More than the sum of the parts. New Haven: Yale University Press. Greenwood, R. and Suddaby, R. 2006. Institutional entrepreneurship in mature fields: The big five accounting firms. Academy of Management Journal, 49(1), pp.27-48. Grothe-Hammer, M., 2018. Organization without actorhood: Exploring a neglected phenomenon. European Management Journal. Available at: https://doi.org/10.1016/j.emj.2018.07.009 [accessed November 2018]. Guba, E. G., 1990. The alternative paradigm dialog. In: E. G. Guba, ed. The paradigm dialog. Newbury Park, CA: Sage. pp.17-27. Guba, E. G. and Lincoln, Y. S., 1994. Competing paradigms in qualitative research. In: N.K. Denzin and Y.S. Lincoln, eds. Handbook of qualitative research. Thousand Oaks, CA: Sage. pp.105-17. Gulati, R. 1995. Social structure and alliance formation: A longitudinal analysis. Administrative Science Quarterly, 40(4), pp.619-52. Gulati, R., 1999. Network location and learning: The influence of network resources and firm capabilities on alliance formation. Strategic Management Journal, 20, pp.397-420. Gulati, R., Puranam, P. and Tushman, M., 2012. Meta-organization design: Rethinking design in interorganizational and community contexts. Strategic Management Journal, 33, pp.571-86. Gustafsson, R., Jääskeläinen, M., Maula, M. and Uotila, J., 2016. Emergence of industries: A review and future directions. International Journal of Management Reviews, 18, pp.28-50. Hagiu, A. and Wright, J., 2015. Multi-sided platforms. International Journal of Industrial Organization, 43, pp.162-74. Halinen, A., Törnroos, J. A. and Elo, M. 2013. Network process analysis: An event-based approach to study business network dynamics. Industrial Marketing Management, 42(8), pp.1213-22. Hallen, B. L. and Eisenhardt, K. M. 2012. Catalyzing strategies and efficient tie formation: How entrepreneurial firms obtain investment ties. Academy of Management Journal, 55(1), pp. 35-70. References 93 Hallett, T. and Ventresca, M. J., 2006. Inhabited institutions: Social interactions and organizational forms in Gouldner’s Patterns of Industrial Bureaucracy. Theor Soc, 35, pp.213-36. Halme, K., Lindy, I., Piirainen, K. A., Salminen, V. and White, J. eds., 2014. Finland as a knowledge economy 2.0: Lessons on policies and governance. World Bank Publications. Hoffman, A., 1997. From heresy to dogma: An institutional history of corporate environmentalism. San Francisco, CA: The New Lexington Press. Hughes, T. P., 1993. Networks of power: Electrification in Western society, 1880-1930. Baltimore, MD: Johns Hopkins University Press. Iansiti, M. and Levien, R., 2004. Strategy as ecology. Harvard Business Review, March, pp.69-78 Isckia, T., 2009. Amazon’s evolving ecosystem: A cyber-bookstore and application service provider. Canadian Journal of Administrative Sciences, 26(4), 332-43. Iyer, B. and Davenport, T. H., 2008. Reverse engineering Google’s innovation machine. Harvard Business Review, April, pp.59-68. Jack, S. L. 2010. Approaches to studying networks: Implications and outcomes. Journal of Business Venturing, 25(1), pp.120-37. Jacobides, M. G., 2005. Industry change through vertical disintegration: How and why markets emerged in mortgage banking. Academy of Management Journal, 48, pp.465-98. Jacobides, M. G., Cennamo, C. and Gawer, A., 2018. Towards a theory of ecosystems. Strategic Management Journal. 39, pp.2255-76. Järvenpää, S. and Välikangas, L., 2014. Opportunity creation in innovation networks. California Management Review, 57, pp.67-87. Järvenpää, S. and Välikangas, L., 2016. From governance void to interactive governing behaviors in new research networks. Academy of Management Discoveries, 2, pp.226-46. Järvensivu, T. and Törnroos, J. A., 2010. Developing case study research with moderate constructionism and abduction: Conceptualization and a practical illustration. Industrial Marketing Management, 39(1), pp.XX. Järvi, K., Almpanopoulou, A. and Ritala, P., 2018. Organization of knowledge ecosystems: Prefigurative and partial forms. Research Policy, 47(8), pp.1523-37 References 94 Järvi, K. and Kortelainen S. (2017) Taking stock on empirical research on business ecosystems: A literature review. International Journal of Business and Systems Research, 11(3), pp.215-28. Kapoor, R. and Lee, J. M. 2013. Coordinating and competing in ecosystems: How organizational forms shape new technology investments. Strategic Management Journal, 34(3), pp. 274–296. Kellogg, K. C., Orlikowski, W. J. and Yates, J., 2006. Life in the trading zone: Structuring coordination across boundaries in post-bureaucratic organizations. Organization Science, 17, pp.22-44. King, B. G., Felin, T. and Whetten, D. A., 2010. Finding the organization in organizational theory: A meta-theory of the organization as a social actor. Organization Science, 21, pp.290-305. Kirsch, D., Moeen, M., Wadhwani, R. D., 2014. Historicism and industry emergence: Industry knowledge from pre-emergence to stylized fact. In: M. Bucheli and R. D. Wadhwani, eds. Organizations in time. Oxford University Press: New York. pp.217-40. Kreiner, G. E., 2016. Tabula geminus - A ‘both/and’ approach to coding and theorising. In: K. D. Elsbach and R. M. Kramer, eds. Handbook of qualitative organizational research. New York, NY: Routledge. Kumar, N., Stern, L. W. and Anderson, J. C., 1993. Conducting interorganizational research using key informants. Academy of Management Journal, 36(6), pp.1633–51. Kunisch, S., Bartunek, J., Mueller, J. and Huy, Q., 2017. Time in strategic change research. Academy of Management Annals, 11(2), pp.1005-64. Lampel, J. and Meyer, A. D. 2008. Field-configuring events as structuring mechanisms: How conferences, ceremonies. and trade shows constitute new technologies, industries and markets. Journal of Management Studies, 45(6), pp.1025-35. Langley, A., 1999. Strategies for theorizing from process data. Acad. Manag. Rev., 24(4), pp.691-710. Lawrence, M. and Woods, E., 2015. The energy cloud – Emerging opportunities on the decentralized grid. Navigant Research. Lawrence, T. B. and Suddaby, R. 2006. Institutions and institutional work. In S. R. Clegg, C. Hardy, T. B. Lawrence and W. Nord, eds. Handbook of organization studies, 2nd ed. London, UK: Sage. pp. 215-54. References 95 Leibel, E., Hallett, T. and Bechky, B., 2018. Meaning at the source: The dynamics of field formation in institutional research. Academy of Management Annals, 12(1), pp.154-77. Leten, B., Vanhaverbeke, W., Roijakkers, N., Clerix, A. and Van Helleputte, J., 2013. IP models to orchestrate innovation ecosystems. California Management Review, 55(4), pp.51-64. Li, Y. R., 2009. The technological roadmap of Cisco's business ecosystem. Technovation, 29(5), pp.379-86. Lincoln, Y. S., 1995. Emerging criteria for quality in qualitative and interpretive research. Qualitative Inquiry, 1(3), pp.275-89. Lincoln, Y. S. 2005. Content, lived experience, and qualitative research. In R. A. Swanson and E. F. Holton, III, eds. Research in organizations: Foundations and methods of inquiry. San Francisco: Berrett-Koehler. pp.221-32. Lincoln, Y. S. and Guba, E. G., 1985. Naturalistic inquiry. Thousand Oaks, CA: Sage. Lincoln, Y. S and Guba, E. G ., 2013. The constructivist credo. Taylor and Francis. Kindle edition. Lopez-Vega, H., Tell, F. and Vanhaverbeke, W. 2016. Where and how to search? Search paths in open innovation. Research Policy, 45, pp.125-36. Lounsbury, M. and Crumley, E. T. 2007. New practice creation: An institutional perspective on innovation. Organization Studies, 38(7), pp.993-1012. Lundvall, B.-Å., Johnson, B., Andersen, E. S. and Dalum, B., 2002. National systems of production, innovation and competence building. Research Policy, 31(2), pp.213-31. March, J. G. and Simon, H. A. 1993. Organizations revisited. Industrial and corporate change, 2, pp.299-316. Markard, J., Wirth, S. and Truffer, B., 2016. Institutional dynamics and technology legitimacy–A framework and a case study on biogas technology. Research Policy, 45, pp.330-44. Mars, M. M. and Bronstein, J. L., 2017. The promise of the organizational ecosystem metaphor: An argument for biological rigor. Journal of Management Inquiry, 27(4), pp.382-91. Meyer, A. D., Gaba, V. and Colwell, K. A. 2005. Organising far from equilibrium: Nonlinear change in organizational fields. Organization Science, 16(5), pp.456-73. References 96 Mir, R. and Watson A., 2001. Critical realism and constructivism in strategy research: Towards a synthesis. Strategic Management Journal, 22(12) pp.1169-73. Moore, J. F., 1993. Predators and prey: A new ecology of competition. Harvard Business Review, 71(3), pp.75-83. Moore, J. F., 1996. The death of competition: Leadership and strategy in the age of business ecosystems. New York: Harper Business. Moore, J. F., 2013. Shared purpose: A thousand business ecosystems, a worldwide connected community, and the future. Available at: https://www.arm.com/files/pdf/Shared_Purpose.pdf [accessed November 2018]. Moran, E. F., 1990. The ecosystem concept in anthropology. Boulder, Colorado: Westview Press. Möller, K. and Halinen, A., 2017. Managing business and innovation networks – From strategic nets to business fields and ecosystems. Industrial Marketing Management, 67, pp.5-22. Nambisan, S., 2018. Architecture vs ecosystem perspectives: Reflections on digital innovation. Information and Organization, 28, pp.104-6. Nambisan, S. and Baron, R. A., 2013. Entrepreneurship in innovation ecosystems: Entrepreneurs’ self-regulatory processes and their implications for new venture success. Entrepreneurship: Theory and Practice, 37, pp.1071-97. Nambisan, S. and Sawhney, M., 2011. Orchestration processes in network‐centric innovation: Evidence from the field. Academy of Management Perspectives, August, pp.40‐57. Newell, A. and Simon, H. A., 1972. Human problem solving. Englewood Cliffs, NJ: Prentice Hall. Nickerson, J. A. and Zenger, T. R., 2004. A knowledge-based theory of the firm – The problem-solving perspective. Organization Science, 15, pp.617-32. Oh, D. S., Phillips, F., Park, S. and Lee, E., 2016. Innovation ecosystems: A critical examination. Technovation, 54, pp.1-6. O’Mahony, S. and Bechky. B. A., 2008. Boundary organizations: Enabling collaboration among unexpected allies. Admin. Sci. Quart, 53(3), pp.422-59. References 97 O’Neill, R. V., 2001. Is it time to bury the ecosystem concept? (With full military honors, of course!). Ecology, 82(12), pp.3275-84. Osterwalder, A., Pigneur, Y. and Tucci, C. L., 2005. Clarifying business models: Origins, present, and future of the concept. Communications of the Association for Information Systems, 16, pp.1-25. Ostrom, E., 1990. Governing the commons: The evolution of institutions for collective action. Cambridge, UK: Cambridge University Press. Overholm, H., 2015. Collectively created opportunities in emerging ecosystems: The case of solar service ventures. Technovation, 39-40, pp.14-25. Ozcan, P. and Eisenhardt, K. M. 2009. Origin of alliance portfolios: Entrepreneurs, network strategies, and firm performance. Academy of Management Journal, 52(2), pp.246-79. Papaioannou, T., Wield, D. and Chataway, J., 2009. Knowledge ecologies and ecosystems? An empirically grounded reflection on recent developments in innovation systems theory. Environment and Planning C: Politics and Space, 27(2), pp.319-39. Paquin, R. L. and Howard-Grenville, J. A., 2013. Blind dates and arranged marriages: Longitudinal processes of network orchestration. Organization Studies, 34(11), pp.1623- 53. Parker, G. G., Van Alstyne, M. W. and Choudary, S. P., 2016. Platform revolution. New York, NY: Norton and Co. Parlett, M. and Hamilton, D. 1972. Evaluation as illumination: A new approach to the study of innovatory programs. Occasional paper, Centre for Research in the Educational Sciences, University of Edinburgh. Edinburgh. Patton, M., 1990. Qualitative evaluation and research methods. Beverly Hills, CA: Sage Publications. Patton, M., 2002. Qualitative research & evaluation methods. 3rd ed. Thousand Oaks, CA: Sage Publications. Pellinen, A., Ritala, P., Järvi, K. and Sainio, L. M., 2012. Taking initiative in market creation–A business ecosystem actor perspective. International Journal of Business Environment, 5, pp.140-58. Peltoniemi, M., 2006. Preliminary theoretical framework for the study of business ecosystems. Émerg.: Complex. Organ, 8(1), pp.10-19. References 98 Perkmann, M. and Schildt, H., 2015. Open data partnerships between firms and universities: The role of boundary organizations. Research Policy, 44, pp.1133-43. Perry, S. J., Hunter, E. M. and Currall, S. C. 2016. Managing the innovators: Organizational and professional commitment among scientists and engineers. Research Policy, 45, pp.1247-62. Peters, L. D., Pressey, A. D. Vanharanta, M. and Johnston, W. J., 2013. Constructivism and critical realism as alternative approaches to the study of business networks: Convergences and divergences in theory and in research practice. Industrial Marketing Management, 42(3) pp.336-46. Pickett, S. T. A. and Cadenasso, M. L., 2002. The ecosystem as a multidimensional concept: Meaning, model and metaphor. Ecosystems, 5, pp.1-10. Pinomaa, A., Ahola, J. and Kosonen, A., 2011. PLC concept for LVDC distribution systems. IEEE Commun. Mag., 49(12), pp.55-63. Porter, M., 1998. Clusters and the new economics of competition. Harvard Business Review, 76, pp.77-90. Porter, M. E., 1985. Competitive advantage: Creating and sustaining superior performance. New York, NY: Free Press. Post, D., Doyle, M., Sabo, J. and Finlay, J., 2007. The problem of boundaries in defining ecosystems: A potential landmine for uniting geomorphology and ecology. Geomorphology, 89(102), pp.111-26. Powell, W. W. 1990. Neither market nor hierarchy: Network forms of organization. Research in Organizational Behavior. 12, pp.295-336. Powell, W. W., Koput, K. W. and Smith-Doerr, L., 1996. Interorganizational collaboration and the locus of innovation: Networks of learning in biotechnology. Administrative Science Quarterly, 41, pp.116-45. Puranam, P., Alexy, O. and Reitzig, M., 2014. What's ‘new’ about new forms of organizing?. Academy of Management Review, 39, pp.162-80. Rao, L., 2016. Inside Google’s next ecosystem: Virtual reality. Fortune, 173(7). Available at: http://fortune.com/daydream-google-virtual-reality/ [accessed November 2018] Repenning, N. P., 2002. A simulation-based approach to understanding the dynamics of innovation implementation. Org. Sci., 13, pp.109-27. References 99 Ritala, P., Agouridas, V., Assimakopoulos, D. and Gies, O., 2013. Value creation and capture mechanisms in innovation ecosystems: A comparative case study. International Journal of Technology Management, 63, pp.244-67. Ritala, P. and Almpanopoulou, A. 2017. In defense of ‘eco‘ in innovation ecosystem. Technovation, 60, pp.39-42. Ritala, P. and Gustafsson, R., 2018. Q&A. Innovation and entrepreneurial ecosystem research: Where are we now and how do we move forward? Technology Innovation Management Review, July, 8(7), pp.52-7. Rohrbeck, R., Hölzle, K. and Gemünden, H. G., 2009. Opening up for competitive advantage – How Deutsche Telekom creates an open innovation ecosystem. R&D Management, 39, pp.420-30. Rong, K. and Shi, Y. 2014. Business ecosystems: Constructs, configurations, and the nurturing process. London: Palgrave Macmillan. Rothschild, M., 1990. Bionomics: Economy as ecosystem. New York: Henry Holt and Company. Saldana, J., 2009. The coding manual for qualitative researchers. London: Sage. Scaringella, L. and Radziwon, A. 2018. Innovation, entrepreneurial, knowledge, and business ecosystems: Old wine in new bottles? Technological Forecasting & Social Change, 138, pp.59-87. Schreyögg, G. and Sydow, J., 2010. Crossroads – Organising for fluidity? Dilemmas of new organizational forms. Organization Science, 21, pp.1251-62. Schreyögg G. and Sydow, J. 2011. Organizational path dependence: A process view. Organization Studies, 32(3), pp.321-35. Scott, W. R. 1995. Institutions and organizations. Newbury Park, CA: Sage. Scott, W. R., 2014. Institutions and organizations. 4th ed. Thousand Oaks, CA: Sage. Scott, W. R. and Davis, G. F., 2016. Organizations and organizing: Rational, natural and open systems perspectives. Routledge: New York. Scott, W. R., Rueff, M., Mendel, P. J. and Garonna, G. A., 2000. Institutional change and healthcare organizations. Chicago: University of Chicago Press. Seidel, M. D. L. and Greve, H. R., 2017. Emergence: How novelty, growth, and formation References 100 shape organizations and their ecosystems. Research in the Sociology of Organizations, 50, pp.1-27. Selznick, P., 1957. Leadership in administration: A sociological interpretation. Evanston, IL: Row, Peterson. Seo, M.-G. and Creed, W. E. D., 2002. Institutional contradictions, praxis, and institutional change: A dialectical perspective. Academy of Management Review, 27, pp.222-47. Shah, S. K. and Corley, K. G., 2006. Building better theory by bridging the quantitative – Qualitative divide. Journal of Management Studies, 148(8), pp.1821-35. Siltaloppi, J., Koskela-Huotari, K. and Vargo, S. L., 2016. Institutional complexity as a driver for innovation in service ecosystems. Service Science, 8(3), pp.333-43. Simchi-Levi, D., 2005. Designing and managing the supply chain. New York, NY: McGraw-Hill College. Sine, W. D., David, R. J. and Mitsuhashi, H., 2007. From plan to plant: Effects of certification on operational start-up in the emergent independent power sector. Organization Science, 18, pp.578-94. Sinkovics, R. R. and Alfoldi, E. A., 2012. Progressive focusing and trustworthiness in qualitative research – The enabling role of computer-assisted qualitative data analysis software (CAQDAS). Management International Review, 52, pp.817-45. Sinnewe, E., Charles, M. B. and Keast, R., 2016. Australia's Cooperative Research Centre Program: A transaction cost theory perspective. Research Policy, 45, pp.195-204. Smets, M. and Jarzabkowski, P., 2013. Reconstructing institutional complexity in practice: A relational model of institutional work and complexity. Human Relations, 66 (10), pp.1279-1309. Smith, M. Y. and Stacey, R., 1997. Governance and cooperative networks: An adaptive systems perspective. Technological Forecasting and Social Change, 54(1), pp.79-94. Snow, C. C., Fjeldstad, Ø. D., Lettl, C. and Miles, R. E., 2011. Organising continuous product development and commercialization: The collaborative community of firms model. Journal of Product Innovation Management, 28, pp.3-16. Stake, R. E., 1995. The art of case study research. Thousand Oak, CA: Sage Publications, Inc. References 101 Stake, R. E., 2000. Case studies. In: N. K. Denzin and Y. S. Lincoln, eds. Handbook of qualitative research. Thousand Oaks, CA: Sage Publications, Inc. pp.435-54. Staw, B. M., Sandelands, L. and Dutton, J., 1981. Threat rigidity effects in organizational behavior. Administrative Science Quarterly, 26, pp.501-24. Still, K., Huhtamäki, J., Russell, M. G. and Rubens, N., 2014. Insights for orchestrating innovation ecosystems: The case of EIT ICT Labs and data-driven network visualisations. International Journal of Technology Management, 66, pp.243-65. Strauss, A. and Corbin, J., 1998. Basics of qualitative research: Techniques and procedures for developing grounded theory. Thousand Oaks, CA: Sage Publications, Inc. Suchman, M. C., 1995. Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20, pp.571-610. Suddaby, R., Bitektine, A. and Haack, P., 2017. Legitimacy. Academy of management annals, 11, pp.451-78. Sydow, J., Windeler, A., Schubert, C. and Möllering, G., 2012. Organizing R&D consortia for path creation and extension: The case of semiconductor manufacturing technologies. Organization Studies, 33, pp.907-36. Tallman, S., Jenkins, M., Henry, N. and Pinch, S., 2004. Knowledge, clusters, and competitive advantage. Academy of Management Review, 29(2), pp.258-71. Tansley, A. G., 1935. The use and abuse of vegetational concepts and terms. Ecology, 16, pp.284-307. Tekes, 2016. Strategic Centres for Science, Technology and Innovation (SHOK). Available at: http://www.tekes.fi/en/programmes-and-services/strategic-centres/ [accessed November 2016] Thomas, L. D. W., 2013. Ecosystem emergence: An investigation of the emergence processes of six digital service ecosystems. Doctoral dissertation, Imperial College Business School. Thornton, P. H., Ocasio, W. and Lounsbury, M., 2012. The institutional logics perspective: A new approach to culture, structure, and process. Oxford: Oxford University Press. Tiwana, A., Konysnski, B. and Bush, A. A., 2010. Platform evolution: Coevolution of platform architecture, governance, and environmental dynamics. Information Systems Research, 21(4), pp.675-87. References 102 Tsujimoto, M., Kajikawa, Y., Tomita, J. and Matsumoto, Y., 2017. A review of the ecosystem concept – Towards coherent ecosystem design. Technological Forecasting and Social Change, doi:10.1016/j.techfore.2017.06.032. Tuertscher, P., Garud, R. and Kumaraswamy, A., 2014. Justification and interlaced knowledge at ATLAS, CERN. Organizational Science, 25(6), pp.1579-1608. Valente, M. and Oliver, C., 2018. Meta-organization formation and sustainability in Sub- Saharan Africa. Organizational Science, 29(4), pp.678-701. Valkokari, K., 2015. Business, innovation, and knowledge ecosystems: How they differ and how to survive and thrive within them. Technology Innovation Management Review, 5(8), pp.17-24. van den Borgh, M., Cloodt, M. and Romme, A. G. L., 2012. Value creation by knowledge‐based ecosystems: Evidence from a field study. R&D Management, 42, pp.150-69. van den Hoonaard, W. C., 1997. Working with sensitizing concepts: Analytical field research. Thousand Oaks, CA: Sage. Vargo, S. L. and Akaka, M. A., 2012. Value cocreation and service systems (re) formation: A service ecosystems view. Service Science, 4(3) pp.207-17. Vargo, S. L., Wieland, H. and Akaka, M. A., 2015. Innovation through institutionalization: A service ecosystems perspective. Industrial Marketing Management, 44, pp.63-72. von Bertalanffy, L., 1956. General systems theory. Gen. Syst. 1, pp.1-10. von Hippel, E., 2006. Democratizing innovation. Cambridge, MA: MIT Press. von Hippel, E. and Von Krogh, G., 2015. CROSSROADS – Identifying viable ‘need– solution pairs’: Problem solving without problem formulation. Organization Science, 27, pp.207-21. Wareham, J., Fox, P. B. and Cano Giner, J. L., 2014. Technology ecosystem governance. Organization Science, 25, pp.1195-1215. Weick, K. E., 1969. The psychology of organizing. Reading, MA: Addison-Wesley. Weick, K. E., 1995. Sensemaking in organizations. Thousand Oaks, CA: Sage Publications. References 103 Weinstein, M., 1999. Railbirds, scouts, and independent booksellers: Towards a constructivist Interpretation of new business formation. Irish Marketing Review, 12(1), pp.3-17. Wilhoit, E. D. and Kisselburgh, L. G., 2015. Collective action without organization: The material constitution of bike commuters as collective. Organization Studies, 36, pp.573- 92. Williamson, P. J. and De Meyer, A., 2012. Ecosystem advantage. California Management Review, 55, pp.24-44. Willis, A. J., 1997. Forum. The ecosystem: An evolving concept viewed historically. Funct. Ecol., 11(2), pp.268-71. Willis, J., 2007.Foundations of qualitative research. London: Sage. Winterhalder, B., 1984. Reconsidering the ecosystem concept. Rev. Anthropol. 11, pp.301-13. Wooten, M. and Hoffman, A. J., 2008. Organizational fields: Past, present and future. In: R. Greenwood, C. Oliver, K. Sahlin, and R. Suddaby, eds. The Sage handbook of organisational institutionalism. London: Sage. Yin, R. 2003. Case study research: Design and methods. Thousand Oaks: Sage. Zietsma, C. and Lawrence, T. B., 2010. Institutional work in the transformation of an organizational field: The interplay of boundary work and practice work. Administrative Science Quarterly, 55(2), pp.189-221. Zietsma, C., Groenewegen, P., Logue, D. M. and Hinings, C. B., 2017. Field or fields? Building the scaffolding for cumulation of research on institutional fields. Academy of Management Annals, 11, pp.391-450. Zott, C., 2003. Dynamic capabilities and the emergence of intraindustry differential firm performance: Insights from a simulation study. Strateg. Manag. J., 24, pp.97-125. Zott, C., Amit, R. and Massa, L., 2011. The business model: Recent developments and future research. Journal of Management, 37, pp.1019-42. Appendix A Interview guide: Study 1  The research programme goes by the name of an ecosystem. Do you see it as a suitable label for it, or would you prefer to call in by another name? Is there any difference between the labels? How would you define ‘ecosystem’? Background information:  Could you tell me a little about yourself and your position in your organisation?  What is your present position in this company/organisation?  What are your major responsibilities in the company/organisation?  Let’s talk now a bit more specifically about XX research programmes. Could you elaborate a bit on XX’s activities and role, particularly regarding these programmes?  Choice of theme? Participants? Proposal writing? Application to funding agency?  What is your organisation’s role after the programme gets the funding and the research begins?  How do you see that the actors of these programmes are being connected to each other? What are the relations between the various actors?  Let’s talk about the research programme; tell me little about your/your organisation’s role and responsibilities in this research programme. How would you compare this programme with the previous ones? Examples of responsibilities… Research programme (past):  Could you tell me little about the birth of this programme?  Who initiated it and how?  Who were the key individuals and stakeholders? How did they help to initiate the programme in practice?  Why and how did your organisation engage in the programme?  What was the main motivation behind the programme?  How would you describe the risks and costs involved?  What were your role and responsibilities in the creation of programme?  What kind of support did you need and by whom?  What kind of challenges have you faced?  How were the particular companies and other participants selected?  How did you manage the actual work flow?  Has the research begun? If yes, when?  Has the research programme changed in any way since then? If yes, how?  Any change in the participants? What about the programme itself? Research programme (present):  Please tell me what this research programme is like (objectives, participants, rules, etc.).  How are the objectives communicated among the different partners?  How do you see the goals of the different partners?  What different partners are needed? And why?  Are there any agreed upon formal or informal rules that the partners have to obey?  How would you describe the decision-making process in the research programme?  How do you see yourself in this process?  How do you see your/your organisation’s expectations in relation to the objectives of the research programme?  What are your expectations regarding the development of this research programme? Thinking back over your remarks…  Is there anything else of importance you’d like to add?  Is there anything that we didn’t talk about that appears relevant?  Which partners to interview next? Interview guide: Study 2 Background information  Could you tell me briefly about yourself?  What is your present position in this company/organisation?  What are your major responsibilities in the company/organisation?  How long have you been involved in or following the development of the energy sector? General situation in energy sector renewal (digitalisation, disruption, and services)  How do you see the current situation in the Finnish energy sector renewal?  Where do you see opportunities for innovation, new services, and business development in general?  Where do you see the risks or hindrances?  What would you consider the most disruptive technologies for the energy sector? (By disruption we mean issues that are able to transform the logic of how the energy sector works. Disruption can make existing business unprofitable and create completely new business models.)  What is the role of digitalisation (i.e. digital technologies, ICT)? What new services have become available due to it or will be created in the future? Institutions and regulation  How do you see the role of institutions for energy sector renewal?  What opportunities or hindrances do you see? Please give examples  What type of institutional or regulatory changes do you think that energy sector renewal requires? Role and motivation  How is your organisation involved in the energy sector renewal? Could you give examples of some specific actions?  Could you elaborate on what motivates you and your organisation to participate in the energy sector renewal and new services development?  What kind of benefits do you see for your organisation?  How would you describe the risks and costs involved? What does it mean for you and your organisation? Energy sector key actors  Who are in your opinion the forerunners in the energy sector?  Who would you consider as key/the most important actors in the energy sector currently (and in the future)?  How do you see that they each contribute to the energy sector renewal (digitalisation and new services development)?  To whom of these actors is your organisation most connected? How do you see the relations of your organisation with the different actors? To conclude:  What are your expectations regarding the development/future of the energy sector in the next 5-10 years? In particular, what is the role of digitalisation?  What kind of role will Finland have in this development?  Thinking back over our discussion, is there anything else of importance you’d like to add or anything that we didn’t talk about that appears relevant?  Finally, who are the other visionary individuals that we should interview next? Publication I Ritala, P. and Almpanopoulou, A. In defense of ‘eco’ in innovation ecosystems Reprinted with permission from Technovation 60–61(February), pp.39–42, 2017 © 2017, Elsevier Contents lists available at ScienceDirect Technovation journal homepage: www.elsevier.com/locate/technovation In defense of ‘eco’ in innovation ecosystem☆ A B S T R A C T Innovation ecosystem is an increasingly popular but all too often ambiguously utilized concept across academia, policy and business. In their recent well-argued critique of the concept, Oh et al. (2016) called it a “flawed analogy” that is potentially dangerous for its lack of rigor. In this letter, we reflect on this critique and examine pathways to resolve some of the issues pointed out. We suggest that, at its best, the ecosystem analogy combines salient features from natural ecology to inform the design of system-level innovation management activities. This requires a great deal of conceptual and empirical rigor, and we outline a number of ideas for future research in this regard. 1. Introduction: What should we do with an analogy that became too popular? Business and management research boasts a high variety of concepts, reflecting the process of on-going change in the broader society, business, and technologies. From time to time, new “buzz” concepts emerge, attracting a lot of researcher and practitioner attention, only to fade away as new trends steal the spotlight. Some of these concepts last longer, creating independent and impactful fields of inquiry; either way, only time can determine their longevity. Innovation ecosystem is one of such concepts. Reflecting the ever-increasing connectivity of innovation activities, it joins the long list of other terms describing the networked and systemic nature of innovation. References to innovation ecosystems have appeared in policy and business discussions, and academics have followed suit with a surge of case studies, conceptualizations, and other approaches that seek to understand and explain the phenomenon. The problem is that there is no consensus on the term's definition, scope, boundaries, or theoretical roots. In a recent article in Technovation, Oh et al. (2016) elaborated a well-grounded critique of the concept of innovation ecosystem, suggesting that it is a flawed analogy and does not necessarily add much value to the existing innovation systems literature. We agree wholeheartedly that the concept is used loosely and inconsistently, resulting in its ambiguous input to scholarly discourse. However, given the rapid expansion of interest in the concept, it seems worthwhile to pursue greater academic rigor and concept clarity in its use. Thus, this letter is a reflection to the critique posed by Oh and colleagues, aiming to assess the potential merits of innovation ecosystem as a self-standing concept, and to examine ways to move forward. 2. Unpacking the terminology Based on Durst and Poutanen's (2013) review, Oh et al. (2016, p. 2) argue that the innovation ecosystem literature pays insufficient attention to the “dialog with multiple constituents”. Drawing from this, we argue that the term ecosystem demands attention to both of its parts: eco and system. Coupled with a focus on innovation activities, this serves to define a point of departure for a rigorous investigation of innovation ecosystems. The prefix eco in innovation ecosystems implies a specifically ecological aspect. As delineated by Moore (1993) in introducing the ecosystem concept to management studies, these ecological aspects relate to the interdependency among different actors, and to the co-evolution that binds them together over time. The term co-evolution refers to a two-way interaction between two entities that may induce change in some direction (Peltoniemi, 2006). In innovation ecosystems, which comprise numerous actors in different layers, actor's decisions may cause counter-responses from other actors. This behavior is multiplied in complex interdependencies across the ecosystem. Thus, it is essential to understand and take account of the link between micro and macro behaviors of ecosystem actors, as well as the cooperative and competitive interactions among them (Peltoniemi, 2006; Overholm, 2015), which affect the balance and dynamics of the ecosystem (Valkokari, 2015). From a systems science perspective, the term system refers to a specific set of components (actors, organizations, entities) that are interdependent, but independent of other systems (e.g., von Bertalanffy, 1956). In fact, innovation ecosystem could be fundamentally portrayed as a specific application of a complex adaptive system (see e.g. Anderson, 1999; Cilliers, 2005). Such underlying systems foundation is also recognized in the ecosystem literature (e.g., Peltoniemi, 2006). In understanding any system or organization, the boundary issue is particularly important (Gulati et al., 2012). For biological ecosystem boundaries, both space and time are seen to play a crucial role (Post et al., 2007). This analogously applies to innovation ecosystems as well – even if they could be considered as open social systems (see Anderson, 1999; Scott and Davis, 2016), at least some semi-stable boundaries could be drawn. Typically, this involves identification of the focal firm, such as “Lego's ecosystem” (Hienerth http://dx.doi.org/10.1016/j.technovation.2017.01.004 Received 3 September 2016Received in revised form 17 January 2017Accepted 18 January 2017 ☆ Oh, D. S., Phillips, F., Park, S., & Lee, E. (2016). Innovation ecosystems: A critical examination. Technovation, 54, 1–6 Technovation 60–61 (2017) 39–42 Available online 01 February 2017 0166-4972/ © 2017 Elsevier Ltd. All rights reserved. MARK et al., 2014), or discussion of the innovation or technology around which the system is formed (e.g., Battistella et al., 2013). Ecosystem boundaries could also be traced via geographical scope (local vs. regional or national vs. global); temporal scale (past to future or static snapshot vs. dynamic interaction); permeability (open vs. closed); or types of flow (knowledge, value, material) (Valkokari, 2015). Finally, there is the pre-fix innovation, which can be defined as creation of new knowledge and inventions, and the successful commercial adoption of those to the markets (Crossan and Apaydin, 2010). Innovation ecosystems involve both of these aspects. In fact, Oh et al. (2016) also recognize that the tension between “research economy” and “commercial economy” might be behind the recent rise of the innovation ecosystem discussion. This is a good point, and also recently articulated by Clarysse et al. (2014), who found that the knowledge ecosystem (i.e., the research economy) and the business ecosystem (i.e., the commercial economy), are partially separate but intertwined within the broader context of innovation activities. Inventions, ideas, and discoveries can be pursued by anyone, but the notion of successful commercialization hints strongly at the involvement of private sector actors. Unsurprisingly, then, innovation ecosystem has been adopted to describe profit-driven systems of innovation around focal companies, technologies and platforms (Li, 2009; Adner and Kapoor, 2010; Ritala et al., 2013; Overholm, 2015). However, there has also been a long standing tendency for public policy to support innovation initiatives in the name of economic development and societal progress. As Oh et al. (2016) note, this means that the concept of innovation ecosystem has begun to infiltrate spaces more traditionally described by such concepts as innovation system, triple-helix, or cluster. This has led to ambiguous usage and application of the concept. Thus, we call for a more mindful use of the terms eco and systems, as well as knowledge creation and market adoption aspects of innovation activities. This requires both rigorous scholarly work and careful usage of the concept. 3. Toward greater rigor in innovation ecosystem research One critique of the concept is that the deliberately designed ecosystems in business and innovation do not actually resemble natural ecosystems (Oh et al., 2016). The content and scope of the ecosystem concept is also debated in natural ecology see O’Neill (2001), increasing the likelihood of unproductive cross-disciplinary borrowing unless the conceptual underpinnings are clearly understood. On the other hand, technology and management literature involves also other powerful ecological analogies, shaping our understanding of the underlying phenomena. These include evolutionary economics (Nelson and Winter, 1982) and its subsequent iterations (e.g., Teece et al., 1997), with broadly adopted concepts of evolution and adaptation. Other evolutionary concepts such as technological speciation (Adner and Levinthal, 2002) and exaptation (Andriani and Cattani, 2016) also abound. Another ecological analogy widely used in management and innovation studies is “ambidexterity.” Originally referring to the ability to use right and left hands equally well (Oxford Dictionaries, 2016), the concept is used in innovation and management studies to refer to capability for simultaneous exploration and exploitation (O’Reilly and Tushman, 2013). Biologically, two hands look very much alike; in an organizational context, exploration and exploitation certainly don’t. This example shows that borrowing from biology need not always fully replicate the original term to be useful for scholarly purposes in another domain. Nevertheless, even with plausible analogy, its application may be less satisfactory. Oh et al. (2016) find a lack of consistency in the use of the concept of innovation ecosystems to describe firm-led ecosystems, digital platforms, regional innovation ecosystems, university-led ecosystems, and so on. While it is true that the concept of ecosystem is in many cases used very loosely, a birds-eye view suggests that these approaches share several common features. First, innovation is a goal or focus of the ecosystem in all cases; it is the actors, contexts, and boundaries that change. Second, they typically entail one or several focal entities that are central to the ecosystem and help to define its boundaries —for instance, a particular firm (Li, 2009; Hienerth et al., 2014), technology (Overholm, 2015), or digital platform (Cusumano and Gawer, 2002). So, although they are open social systems, it seems that innovation ecosystems are deliberately designed and evolve around key set of entities, at least at a particular point in time. If conceptual clarity poses a challenge for innovation ecosystem scholarship, empirical inquiry may present an even larger obstacle, and it is no surprise that much of the research consists of case studies. However, even if qualitative case inquiries are likely to be appropriate for the study of such complex entities, the demands for embracing the concept in full are arguably near to insurmountable. In fact, Oh et al. (2016) argue that sound measurement of progress of co-evolution in innovation ecosystem is beyond current scientific capabilities. Given the multiple interdependencies between organizations, technologies, individuals and institutions, we agree that this seems a fair assessment. However, the same problem plagues studies on any multi-actor networks. Thus, even granted the impossibility of a perfect research design, we believe it is worthwhile to engage in various forms of academic inquiry over the important real-life phenomena. For instance, with the help of simulation modeling and related theories (e.g., systems theory, control theory), innovation ecosystem studies can evolve from being merely descriptive to become more predictive. Simulation studies (e.g., system dynamics, agent-based modeling) can also more successfully comprehend the complexity and dynamicity of innovation ecosystems. In general, simulation is useful for theory development, as it can expose the complex connections among constructs or the results of interactions among numerous organizational and strategic processes, especially as they unfold over time (Repenning, 2002; Zott, 2003; Davis et al., 2007). In addition, we also see the value of qualitative process research for the study of dynamic phenomena like innovation ecosystems as it can provide rich understanding on the hows and whys of these processes (Langley, 1999). 4. The way forward: Some solutions and open questions In this short letter, we have addressed the burgeoning interest in the topic of innovation ecosystems and the related conceptual ambiguity challenges. We do not believe that our arguments are definitive or one-size-fits-all. Hoverer, we hope to spark active discussion and stimulate future work with improved conceptual and empirical rigor. To conclude, we summarize our arguments by responding to the four innovation ecosystem research challenges identified by Oh, p 5) et al. (2016). 4.1. Whether and how innovation ecosystems differ from national and regional innovation systems Oh et al. (2016) argue convincingly that the established innovation systems literature already include the necessary ingredients for discussing national and regional innovation systems. They also list potentially differentiating features of recent innovation ecosystem studies. Among these, the “market-driven ecosystem movement” - may represent one key to unlock this conceptual ambiguity. In his early scholarly introduction of the business ecosystem concept, Moore (1993) was already pointing to the importance of an ecosystem perspective on innovation. In addition, other P. Ritala, A. Almpanopoulou Technovation 60–61 (2017) 39–42 40 much cited sources of innovation ecosystems (e.g., Adner and Kapoor, 2010), treated the concept very much as a market-driven phenomenon and did not consider policy issues in the same way as the innovation systems literature. This discussion brings us back to considering the role of public initiatives and of private business at large. At best, market-driven innovation ecosystems do what the profit-seeking corporations do best in any case: innovate, compete, create customer value, and subsequent economic progress. At the same time, publicly funded regional innovation systems and triple-helix initiatives should not be overlooked. Whatever the terms used to describe these phenomena, we believe that all different types are needed to facilitate innovation activities regionally, nationally, and globally. While only time will determine the traction of these concepts in different applications, perhaps the best solution is to scope innovation ecosystems in more market-driven initiatives while using other more established concepts to discuss public policy. There is certainly shades of grey between the contexts, which calls for even more care in choosing the proper concepts. 4.2. Measurement of innovation ecosystem performance In the innovation management literature, there are several more or less established ways of measuring performance, including objective and subjective ways to measure outputs and processes. Some of these approaches could be viewed from ecosystem actor perspective and aggregated to system or sub-system levels. In reality, however, performance measurement of any collective multi-actor phenomenon is a difficult task; tensions and contradictions between actor- and system-specific goals are likely to emerge, raising questions about whose performance (and what) should be measured. Measurement could also remain completely external to the system, assessing the relative competitiveness of the ecosystem against competing innovations and technologies. In any case, as in any other study, performance metrics should be linked to the level of analysis and the research question. 4.3. Similarities and differences between natural and innovation ecosystems While the similarities of natural and innovation ecosystems may prove useful for extending scholarship (as in the case of ambidexterity), any differences may prove dysfunctional unless properly acknowledged. As Oh, et al. (2016, p. 2) observe, “An innovation ecosystem is not an evolved entity. Rather, it is designed.” However, as previously noted by Nelson and Winter (1982), all organizations and technologies evolve over time, and the closer you look, the more evolution you find. That being so, the unique features of purposeful design and evolutionary nature may make the innovation ecosystem concept viable for examining real-world phenomena in both of these important respects. For that reason, it is important for ecosystem scholars to understand which parts of the ecosystem are (and can be) engineered, and which parts are self-organized or co-evolved. Oh et al. (2016) highlight the difference in scope of natural and innovation ecosystems—that is, while natural ecosystems are local, innovation ecosystems can be global. However, in the biology literature the ecosystem scope is not that strictly defined either, as “an ecosystem may range from anthill to the entire biosphere of the globe” (Willis, 1997, p. 269). In fact, Willis argues that the concept's robustness is evident in its capacity for extension across scales, as well as its accommodation of both holistic and reductionist approaches. These benefits could equally apply to management and innovation studies. At the same time, this scalability of the concept makes its application problematic, as any networked innovation activity could be labeled an “ecosystem.” In this regard, we call for critical thinking, focusing on the added value of the concept and all of its components. To enhance the concept's applicability, we agree with Oh et al. (2016, p. 5) that “innovation ecosystem theorists may relax some axioms of ecology (and perhaps introduce a small number of additional ones) in order to fit the needs of artificial ‘ecosystems.” The development of such axioms is of fundamental importance to scholarship in the field. We believe that by utilizing some of the useful features of ecological thinking (e.g., co-evolution) and systems thinking (particularly complex adaptive systems), innovation ecosystem studies can embrace their research objects more holistically, as well as more realistically. 4.4. Distinguishing the levels at which the term is used As we iterated earlier, innovation ecosystem studies often seem to focus on dominant entities that determine their boundaries, such as hub firms, technologies, and platforms (see also Autio and Thomas, 2014). However, while the focal point of boundary definition may be a particular entity or platform, the level of the analysis can refer to the system itself. The challenges of such system-level analysis are many, and could be partially addressed by the use of different methodologies and making clear choices over level and unit of analysis. In this regard, we suggest that the term innovation ecosystem should ideally be used in respect of systems that focus on innovation activities (goal/purpose), involve the logic of actor interdependence within a particular context (spatial dimension) and address the inherent co-evolution of actors (temporal dimension). References Adner, R., Levinthal, D.A., 2002. The emergence of emerging technologies. Calif. Manag. Rev. 45 (1), 50–66. Adner, R., Kapoor, R., 2010. Value creation in innovation ecosystems: how the structure of technological interdependence affects firm performance in new technology generations. Strateg. Manag. J. 31 (3), 306–333. Anderson, P., 1999. Complexity theory and organization science. Organ. Sci. 10 (3), 216–232. Andriani, P., Cattani, G., 2016. Exaptation as source of creativity, innovation, and diversity: introduction to the special section. Ind. Corp. Change 25 (1), 115–131. Autio, E., Thomas, L., 2014. Innovation ecosystems. Oxf. Handb. Innov. Manag., 204–288. Battistella, C., Colucci, K., De Toni, A.F., Nonino, F., 2013. Methodology of business ecosystems network analysis: a case study in telecom Italia future centre. Technol. Forecast. Soc. Change 80, 1194–1210. Cilliers, P., 2005. The architecture of complexity (introduction). Émerg.: Complex. Organ. 7 (3–4), 138–154. Clarysse, B., Wright, M., Bruneel, J., Mahajan, A., 2014. Creating value in ecosystems: crossing the chasm between knowledge and business ecosystems. Res. Policy 43 (7), 1164–1176. Crossan, M.M., Apaydin, M., 2010. A multi‐dimensional framework of organizational innovation: a systematic review of the literature. J. Manag. Stud. 47 (6), 1154–1191. Cusumano, M.A., Gawer, A., 2002. The elements of platform leadership. MIT Sloan Manag. Rev. 43 (3), 51. Davis, J.P., Eisenhardt, K.M., Bingham, C.B., 2007. Developing theory through simulation methods. Acad. Manag. Rev. 32, 480–499. Durst, S., Poutanen, P., 2013. Success factors of innovation ecosystems – Initial insights from a literature review. In R. Smeds & O. Irrmann (eds.) CO-CREATE 2013: The Boundary- Crossing Conference on Co- Design in Innovation,pp. 27–38. Aalto University Publication series SCIENCE+TECHNOLOGY 15/2013. Gulati, R., Puranam, P., Tushman, M.L., 2012. Meta-organization design: rethinking design in interorganizational and community contexts. Strateg. Manag. J. 33 (6), 571–586. Hienerth, C., Lettl, C., Keinz, P., 2014. Synergies among producer firms,firms, lead users,users, and user communities:communities the: the case of the LEGO producer-user ecosystem. J. Prod. Innov. Manag. 31 (4), 848–866. P. Ritala, A. Almpanopoulou Technovation 60–61 (2017) 39–42 41 Langley, A., 1999. Strategies for theorizing from process data. Acad. Manag. Rev. 24 (4), 691–710. Li, Y.R., 2009. The technological roadmap of Cisco's business ecosystem. Technovation 29 (5), 379–386. Moore, J.F., 1993. Predators and prey: a new ecology of competition. Harv. Bus. Rev. 71 (3), 75–83. Nelson, R., Winter, S., 1982. An Evolutionary Theory of Economic Change. Harvard University Press, Cambridge, MA. Oh, D.S., Phillips, F., Park, S., Lee, E., 2016. Innovation ecosystems: a critical examination. Technovation 54, 1–6. O'Neill, R. V., 2001. Is it time to bury the ecosystem concept? (With full military honors, of course!) Ecology, 82(12), pp. 3275-3284 O'Reilly, C.A., Tushman, M.L., 2013. (Past, present, and future)Organizational Ambidexterity 27. The Academy of Management Perspectives, 324–338. Overholm, H., 2015. Collectively created opportunities in emerging ecosystems: the case of solar service ventures. Technovation 39, 14–25. Oxford Dictionaries, 2016. Definition of ambidextrous in English. Retrieved from: 〈https://www.oxforddictionaries.com/definition/english/ambidextrous? Q=Ambidexterity#ambidextrous__5/〉 Peltoniemi, M., 2006. Preliminary theoretical framework for the study of business ecosystems. Émerg.: Complex. Organ. 8 (1), 10–19. Post, D., Doyle, M., Sabo, J., Finlay, J., 2007. The problem of boundaries in defining ecosystems: a potential landmine for uniting geomorphology and ecology. Geomorphology 89 (102), 111–126. Repenning, N.P., 2002. A simulation-based approach to understanding the dynamics of innovation implementation. Organ. Sci. 13, 109–127. Ritala, P., Agouridas, V., Assimakopoulos, D., Gies, O., 2013. Value creation and capture mechanisms in innovation ecosystems: a comparative case study. Int. J. Technol. Manag. 63 (3- 4), 244–267. Scott, W.R., Davis, G.F., 2016. Organizations and Organizing: Rational, Natural and Open Systems Perspectives. Routledge, New York. Teece, D.J., Pisano, G., Shuen, A., 1997. Dynamic capabilities and strategic management. Strateg. Manag. J. 18 (7), 509–533. Valkokari, K., 2015. Business, innovation, and knowledge ecosystems: How they differ and how to survive and thrive within them. Technol. Innov. Manag. Rev. 5, 17–24. von Bertalanffy, L., 1956. General systems theory. Gen. Syst. 1, 1–10. Willis, A.J., 1997. Forum. The ecosystem: an evolving concept viewed historically. Funct. Ecol. 11 (2), 268–271. Zott, C., 2003. Dynamic capabilities and the emergence of intraindustry differential firm performance: insights from a simulation study. Strateg. Manag. J. 24, 97–125. Paavo Ritala⁎, Argyro Almpanopoulou Professor, School of Business and Management, Lappeenranta University of Technology, P.O. Box 20, FI-53851 Lappeenranta, Finland Doctoral student, School of Business and Management, Lappeenranta University of Technology, P.O. Box 20, FI-53851 Lappeenranta, Finland E-mail address: ritala@lut.fi ⁎ Corresponding author. P. Ritala, A. Almpanopoulou Technovation 60–61 (2017) 39–42 42 Publication II Järvi, K., Almpanopoulou, A., and Ritala, P. Organization of knowledge ecosystems: Prefigurative and partial forms Reprinted with permission from Research Policy 47(8), pp.1523–1537, 2018 © 2018, Järvi, K., Almpanopoulou, A., and Ritala, P. Acknowledgement: Right for first publication: Research Policy licensed under a Creative Commons Attribution License Research Policy 47 (2018) 1523–1537 Available online 26 May 2018 0048-7333/ © 2018 Lappeenranta University of Technology. Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/BY/4.0/). Research Policy 47 (2018) 1523–1537 1524 Research Policy 47 (2018) 1523–1537 1525 Research Policy 47 (2018) 1523–1537 1526 Research Policy 47 (2018) 1523–1537 1527 Research Policy 47 (2018) 1523–1537 1528 Research Policy 47 (2018) 1523–1537 1529 Research Policy 47 (2018) 1523–1537 1530 Research Policy 47 (2018) 1523–1537 1531 Research Policy 47 (2018) 1523–1537 1532 Research Policy 47 (2018) 1523–1537 1533 Research Policy 47 (2018) 1523–1537 1534 Research Policy 47 (2018) 1523–1537 1535 Research Policy 47 (2018) 1523–1537 1536 Research Policy 47 (2018) 1523–1537 1537 Publication III Almpanopoulou, A., Ritala, P., and Blomqvist, K. Innovation ecosystem emergence barriers: Institutional perspective Reprinted with permission from Proceedings of the 52nd Hawaii International Conference on System Sciences pp. 6357-6366, 2019 © 2018, Almpanopoulou, A., Ritala, P., and Blomqvist, K. Acknowledgement: Right for first publication: Proceedings of the 52nd Hawaii International Conference on System Sciences licensed under a Creative Commons Attribution License Innovation Ecosystem Emergence Barriers: Institutional Perspective Argyro Almpanopoulou Lappeenranta University of Technology argyro.almpanopoulou@lut.fi Paavo Ritala Lappeenranta University of Technology paavo.ritala@lut.fi Kirsimarja Blomqvist Lappeenranta University of Technology kirsimarja.blomqvist@lut.fi Abstract Innovation ecosystems are built around new technologies, ideas, and innovations and their supporting actors and structures. However, the emergence of ecosystems is constrained by a host of institutional, system-level barriers in the existing organizational field that inhibit the legitimacy, resourcing, and growth of new initiatives. Through an empirical study in the Finnish energy sector, we find a strong and interdependent set of regulative, normative, and cultural–cognitive barriers that restrict the emergence of innovation ecosystems with new technologies. In particular, we identify a set of barriers and related field-sustaining mechanisms. The findings offer important implications for the theory and practice of innovation ecosystem emergence and related system-level barriers. 1. Introduction Innovation ecosystems enable actors, technologies, and institutions to come together to create and commercialize new products and services (e.g., [33], [31], [12]). As open social systems (e.g., [7]), they enable dynamic inflows and outflows of resources and provide a shared institutional logic for the emergence of different types of innovation [46]. However, creating new ecosystems is not easy. As new innovation ecosystems emerge1, they often disrupt existing social, technological, and organizational fields and regimes (e.g., [20], [5]). Thus, as new ecosystems pursue new trajectories and paths, effectively replacing some old ones [3], they often face both deliberate and ‘passive’ resistance from different types of actors and institutions. 1 Emergence describes how complex systems arise out of a set of interactions; however, the connection between the actions of individual actors and the systemic outcome is uncertain [40]. In addition to new ecosystem emergence, we also consider the transformation or renewal of a mature ecosystem to be a form of emergence, since it involves profound restructuring and other organization-level changes [26] that can lead to unpredictable and even surprising outcomes at the ecosystem level. The literature on innovation and strategy has provided considerable evidence of entry barriers, a topic that has been discussed since the classic Porterian analysis of industry forces [30]. However, the entry barrier literature has typically focused on the barriers for single actors, rather than on the dynamic counterforces that prevent the emergence of whole ecosystems. More recently, scholars have begun to examine how the entry of new technologies and related actors and institutions occurs in various system-level settings (e.g., [10], [5], [13], [24]). Further, the literature on organizational and institutional fields has examined pathways to field change, including the entrance of new technologies and actors [49]. However, a comprehensive understanding of barriers and constraining mechanisms is largely absent in the innovation ecosystem literature, which has focused mostly on how existing businesses build and manage their ecosystems (e.g., [33], [47], [31]). This literature recognizes that the rise of new ecosystems, sudden changes in environmental conditions (e.g., new regulations or customer buying behaviors), and changes in macroeconomic conditions can threaten mature ecosystems [26]. In other words, the research on innovation ecosystems sees the wider environment as a trigger for the renewal (or death) of these ecosystems. Yet, this literature has paid less attention to how the broader environment and accompanying institutions might create barriers to innovation ecosystem emergence and the pre-formation phases of innovation ecosystems. In these phases, actors are still looking for opportunities to develop new innovations for the field, and ecosystem emergence typically requires collective action, a jointly adopted vision, and actors taking a lead on various issues [6, 29]. By contrast, the absence of these conditions acts as a barrier to ecosystem emergence. The outcome of this phase is unpredictable; however, actors’ choices and actions are increasingly influenced by and embedded in institutions, and the rules and culture that serve as institutional building blocks [35]. Existing innovation ecosystem research has not provided overarching evidence of the barriers that inhibit the processes (e.g., collective action, initiative taking) that lead to ecosystem emergence. 1SPDFFEJOHT PG UIF OE )BXBJJ *OUFSOBUJPOBM $POGFSFODF PO 4ZTUFN 4DJFODFT ]  63* I॒QTIEMIBOEMFOFU *4#/  $$ #:/$/%  1BHF  To address these research gaps and better understand the systemic nature of ecosystem emergence barriers, we follow [20], who suggest that socio-technical transitions can be examined through analyses of organizational and institutional fields [14] [49]. The institutional literature has long recognized that the emergence of new technologies and innovations involves complex institutional dynamics that require not only technological changes, but also a focus on user practices, regulation, infrastructures, and symbolic and cultural issues (e.g., [18], [19], [25]). These, in turn, might create a strong level of institutionalization, which is sustained through an interplay with both issue fields, in which powerful actors push back against radical developments, and market exchange fields, in which transactions are dominated by incumbent actors (cf. [49]). Institutional lenses have recently been applied to the literature on innovation ecosystems. [43] suggest that an institutional approach—and, particularly, an organizational fields approach—is useful for studying the dynamics and boundaries of ecosystems. Furthermore, [6] suggest that creating new ecosystems requires the establishment of institutional legitimacy among relevant stakeholders. In the current paper, institutional lenses are adopted to examine the diversity and strength of the institutions new ecosystem initiatives face and to understand the heretofore understudied dynamic counterforces of innovation ecosystem emergence. Based on this foundation, we propose the following research question: What are the barriers that inhibit ecosystem emergence, and how are these barriers sustained? To answer our research question, we conduct a qualitative inquiry with rich empirical evidence from the organizational field of the Finnish energy sector. Specifically, to examine ecosystem emergence, we select the “digitalization” of the energy sector as our empirical context. Utilizing an empirical study, we identify four ecosystem emergence barriers and related field-sustaining mechanisms. We find that the Finnish energy sector includes a strong and interdependent set of regulative, normative, and cultural–cognitive barriers that restrict ecosystem emergence. In particular, we build a model that explains how regulation and policymaking ambiguity, incumbent actor inertia, and cognitive constraints for opportunity recognition mutually reinforce one another. Our model also explains how the institutional complexity of the energy field functions as an overarching barrier. The results contribute to the understanding of the pre- emergence phase of ecosystems and related institutional barriers. The paper proceeds as follows. First, we discuss the conceptual background of innovation ecosystems and their emergence, followed by a broader discussion of institutional barriers. Second, we describe our methodological choices and then draw conclusions regarding ecosystem emergence barriers. Finally, we discuss the implications for theory and practice and suggest future research directions. 2. Conceptual background 2.1. Innovation ecosystem emergence The concept of innovation ecosystems is widely debated (see, e.g., [34], [27], [1], [44]); however, consensus is forming around some key features. In particular, we follow the recent conceptualization of [32, p. 41], who define innovation ecosystems as “systems that focus on innovation activities (goal/purpose), involve the logic of actor interdependence within a particular context (spatial dimension) and address the inherent co-evolution of actors (temporal dimension).” Innovation ecosystem actors typically include private firms that develop new technologies, universities and research institutions, and complementary firms that provide necessary technological components, inputs, and market access [3]. The existing research on innovation ecosystems has examined how such ecosystems are purposefully built by leading firms [33], [2], [31], as well as how they emerge around broader socio-technical regimes and industry-crossing economic developments [13], [34]. In this study, we focus particularly on the latter context: a broader system-level interdependence of actors that engage in innovation activities. Ecosystems are built around interdependencies of actors, technologies, and institutions [1]; therefore, the emergence of an ecosystem is a complex feat in itself. However, analyzing actors’ pursuits with respect to new ecosystem creation is not sufficient; we must also understand the existing and incumbent actors, technologies, and institutions that provide the field- level context within which (potential) emergence occurs. While ecosystems span several industry boundaries [47], [26], their emergence shares some features with industry emergence. In examining industry emergence, [24] differentiate among three phases. In the first stage, a disruption to the existing industrial order triggers a second, co-evolutionary stage, which includes four sub-processes related to developments in technology, markets, activity networks, and industry identity. The convergence of these sub-processes leads to the third, growth stage and the birth of a new industry. Similarly, during their emergence phase, innovation ecosystems 1BHF  first begin to disrupt existing actors, technologies, and institutions (e.g., [3]), while simultaneously confronting resistance from incumbents. Second, as an ecosystem develops, a co-evolution of new and existing actors, technologies, and institutions occurs (see [3]). Finally, innovation ecosystems enable the commercialization of ideas, inventions, and technologies, creating new businesses, industries, and ecosystems and, thus, integrating the systems’ explorative features through exploitation [45]. In the current study, we are particularly interested in the early pre-emergence phase, during which actors pushing for new initiatives are still struggling to find ways to foster ecosystem creation. To understand this phase, we turn our attention to the institutional barriers formed at the level of organizational fields (i.e., the context in which ecosystem emergence occurs). 2.2 Institutional barriers in organizational fields In general, institutions act as constraints and facilitators for innovation and technological development [13], [25]. Institutions are broadly seen to affect all organizational actions and interactions within a particular field (e.g., the energy sector) and to include regulative, normative, and cultural–cognitive aspects (see, e.g., [19], [25]). In the current study, we view the field level as the context for analyzing institutional barriers to ecosystem emergence. Organizational fields are defined as “those organizations that, in the aggregate, constitute a recognized area of institutional life: key suppliers, resource and product consumers, regulatory agencies, and other organizations that produce similar services or products” [14 p. 148]. The field level has been the key frame for analyses in institutional theory, as it explains the relevant contexts for institutional phenomena [49]. Recently, institutional theory has also begun to look at institutional fields more broadly, including fields formed around opinions, politics, norms, debates, and organizational arrangements (cf. [49]). For ecosystem emergence, examining the full variety of institutional phenomena at the field level is particularly important, given the co-evolving and interdependent nature of ecosystems and their business environments (e.g., [1]). When analyzing institutional barriers for innovation ecosystem emergence, examining legitimacy is particularly important. As suggested by [6], new ecosystem creation involves building legitimacy across various stakeholders, involving regulative issues, technological aspects, and cognitive and symbolic meanings. This broadly follows the tradition of institutional theory, in which legitimacy is divided into regulative, normative, and cognitive (see [36], [25], [41]). Regulative legitimacy refers to the degree to which an organization (or ecosystem) aligns with existing processes for rule-setting, monitoring, and sanctioning. Normative legitimacy is defined as “a degree of congruence or fit between the actions, characteristics, and form of the organization and the beliefs and cultural values of the broader social environment within which it exists” [41, p. 454]. Finally, cognitive legitimacy refers to a high degree of alignment between an organization’s “taken-for- granted” expectations and its environment (see, e.g., [4]). For ecosystem emergence, all three types of legitimacy are required; in other words, all three types of legitimacy might appear as constraining forces in the organizational field. Therefore, analyzing the institutional forces and related legitimacy is important for understanding not only the context of ecosystem emergence, but also the barriers that might prevent emergence from happening in the first place. 3. Methods The following sections discuss our methodological choices. After elaborating the research strategy and how it evolved, we describe the empirical setting of the study. Finally, we discuss our approach to the data collection and analysis. 3.1. Research design and empirical setting This study originated from a broader research project on the emergence of innovation ecosystems that initially did not focus on institutional barriers. Rather, this focus emerged during the data collection and analysis, and we interpreted it as a prominent feature of the empirical phenomenon and, thus, a promising theme for theorizing. Therefore, we progressively focused [28], [38] (see Figure 1) our study on the emerging issue of ecosystem emergence barriers and formulated our final research question as follows: What are the barriers that inhibit innovation ecosystem emergence, and how are these barriers sustained? We then turned to institutional theory (e.g., [14], [36]), which we utilized to sensitize our theorizing. Specifically, we focused our study on the level of the organizational field [14], [48], [49], initially choosing to examine the regulative, normative, and cultural– cognitive institutional elements [36]. To answer our research questions, we relied on an in-depth qualitative inquiry, which we consider to be consistent with our 1BHF  research goals and the exploratory nature of the study (see [15]). The empirical context of this study focuses on the energy sector in Finland, which comprises a variety of actors ranging from major players to several middle- sized firms and an increasing number of innovative start-ups. The energy sector is an interesting empirical context because it is one of the industries least disrupted by digitalization both in Finland and globally (compared, for instance, to the media and telecommunications sectors). Digitalization and related business models represent a new socio-technical regime [20] in the energy sector, which is currently emerging and will eventually replace some older business models and practices. As a highly regulated and capital-intensive sector, we believe that the energy sector is well suited to the study of the institutional barriers to ecosystem emergence. For the purposes of the current study, the energy sector represents the level of analysis of an organizational field in which we examine these barriers and related field-sustaining processes. Incumbent and established actors in the energy sector are interdependent because they must co- develop the capabilities to respond to this new era. Therefore, we perceive the energy sector as a feasible empirical context for studying the barriers to ecosystem emergence. 3.2. Data collection Our data collection process comprised semi- structured interviews. We interviewed 26 key informants representing different organizations to ensure a variety of viewpoints and, thus, to increase the validity of the findings [23]. We first utilized archival material to identify the key respondents to interview and then used a snowballing technique to identify further respondents. We collected our data in two phases (see Figure 1). First, we interviewed eight experts in the energy sector to gain a general understanding of the specific field and its current state in terms of digitalization. During this phase, we observed that there were certain barriers that obstructed the emergence of new ecosystems. This led us to turn to institutional theory. Accordingly, we conceptualized the energy sector as an organizational field and formulated our final research question. In the second phase, we updated our interview guide to include questions about the roles of institutions and regulations, as well as the roles and activities of various actors. We then conducted 18 additional interviews with key informants representing different actors within the organizational field. The interviewees represented a wide variety of experts, including six leading energy sector academics, two research institute representatives, five policy makers, ten company representatives, two industry association representatives, and one representative of a non-governmental organization. 3.3. Data analysis Our analytical procedure was guided by the principles of grounded theory (GT). The GT approach provides tools with great potential for analyzing processes [8], making it a relevant and well-aligned analytical strategy for explaining the dynamic phenomenon under examination. In the first phase of the data analysis, the first author independently started the analysis with initial coding utilizing NVivo. This phase involved coding sentences or segments of the data depending on their richness. We tried to remain open to what our material suggested and used in-vivo coding when applicable. The initial codes varied in length from a couple of words to full sentences. In the second phase of our analysis, the second and third authors were actively involved through discussions and multiple iterations of the initial codes. At the beginning of this phase, the second and third authors examined the initial codes separately and provided comments, questioning the analytical decisions and helping to raise the level of abstraction. We then selected the codes that were the most significant, appeared most frequently, or made the most analytic sense and started to sort and organize them into focused codes (see [8]). Our approach to the data analysis followed an iterative cycle of inductive patterns involving reflection back and forth with theory (cf. [22]) that made it possible to draw broad patterns from the data. Additionally, the different analytical roles allowed for researcher triangulation, yielding a more comprehensive and Figure 1. Progressive focusing of our study adapted by [38] 1BHF  heterogeneous set of perspectives, which we synthesized during the analysis process. The findings presented in the following section are an outcome of this analysis process. 4. Findings Our empirical analysis revealed four main barriers to innovation ecosystem emergence: incumbent actor inertia, regulation and policymaking ambiguities, cognitive constraints for opportunity recognition, and institutional complexity. In the following sections, we elaborate on each of the four barriers, the mechanisms that sustain them, and how they mutually reinforce each other. Figure 2 depicts our overall findings regarding these institutional counterforces to innovation ecosystem emergence. 4.1 Incumbent actor inertia Though we identified some innovative and flexible players pushing for energy sector digitalization, we found incumbent actors’ inertia to be a prominent barrier to new ecosystem emergence and the evolution of existing ecosystems. Overall, our informants described the organizational field as static and conservative. Incumbent actors were perceived as hesitant to drive change; their adherence to the past business logic and operating principles sustained the existing field instead of allowing for renewal. The following quotation vividly captures this issue: “They want large power plants, and historically, they've learned that this is the right thing to do, and maybe it has been the right thing to do, and it's okay, but now times are changing. But if you are within this group and within this bubble, it's very difficult to completely revise your thinking and think of the roadmap that: How do we go from this point A to the new immaterial digital world?” This trend includes energy companies, which fear cannibalizing existing investments/business and are reluctant to test new business models. Related to this issue is the lengthy life cycle of investments in the energy sector. For example, power plant investments have a life cycle of 35 to 40 years, and companies expect to keep the plant running for that time. The long economic lifetimes of these kinds of investment create a kind of natural inertia among incumbent actors. In addition, the concentration of influence within static and closed networks sustains the bargaining power and legitimacy of these actors. The strong in- group socialization within a relatively small and homogeneous group of influential actors and individuals leads to the formation of “bubbles” of consensus thinking, which result in high normative institutional barriers. In fact, the small number of influential individuals and their tight interconnections when making key decisions concerning, for example, energy policy and other decisions that affect the field, leads to a lack of outsider perspectives and can result in new digital business model initiatives being left on the sidelines: “Energy policy will then, basically, be done behind closed doors, and [those] who are inside the closed doors, they will determine very much the contents of the energy policy.” These large incumbents have developed a situation similar to a monopoly, creating a culture and mindset that lacks innovation and customer focus. Legacy industry dominance relates to the strong presence and influence of current incumbent industries, such as the forestry (or “bio-economy”) industry and the nuclear industry, among other dominant sectors. Our respondents viewed the increased availability of bio-based energy in Finland, as well as the access to key resources and financial support that historically lies within these sectors, as potentially harmful for the transformation of the energy sector. The power of these sectors stems from their role as major consumers and providers of energy, which has encouraged the Finnish government to “safeguard” them. As new and nimble players willing to innovate in energy efficiency and distributed energy production emerge, they find it hard to gain equal access to resources and infrastructure. 4.2 Regulation and policymaking ambiguities The ambiguity in regulation and policymaking is a major inhibitor for new investments and broader ecosystem initiatives, as it creates uncertainty regarding the future direction of the energy policy. Though most of our respondents perceived the government’s policy targets (e.g., de-carbonizing the energy system by 2050) as rather progressive, they argued that how these targets will be met is still unclear. We found that the slowness of the policymaking environment reinforces these Figure 2. Institutional counterforces to innovation ecosystem emergence. 1BHF  ambiguities related to the policy vision and the actual action plan. Our respondents perceived that energy policy always comes a few years behind international development and fails to recognize and promote advanced policies that could grasp the swift technological changes taking place. The slowness of policymaking is also related to the shortsighted political vision. The transition of political regimes (i.e., the parliament) every four years creates discontinuity in the policy environment and keeps some governments from pushing radical changes and making concrete action plans for the future. This uncertainty hampers any new ecosystem emergence via digitalization or otherwise. Finally, geopolitical and economic risks can intensify ambiguities in the policymaking environment. The dependency on other countries for energy is not considered a good pre-condition for progressive policymaking, as any attempt to gain energy independence could mean that big energy producers lose their influence and power, and it is uncertain what their reaction would be. Therefore, geopolitical and economic risks can slow progress in the policymaking environment, which can have a negative impact on ecosystem emergence due to uncertainties regarding the future. 4.3 Cognitive constraints for opportunity recognition An important barrier to ecosystem emergence concerns cognitive constraints to opportunity recognition. These constraints involve both new actors trying to form new ecosystems and established actors attempting to make sense of socio-technical change and related opportunities. First, our respondents perceived a great deal of uncertainty over market opportunities, which makes it difficult to identify viable business ventures in the energy sector. In particular, the lack of demand for new services and products and the low electricity prices discourage development and reduce the economic viability of investments. In addition, the dispersion of necessary capabilities and resources creates further constraints for collectively recognizing and exploiting opportunities. Finnish companies are very small, and they do not have the necessary resources to develop new services and products for final consumers. There are many electricity vendors and distribution companies scattered throughout Finland, resulting in a lack of interconnection among these spatially and thematically dispersed players, making the collective creation of opportunities rather challenging. Furthermore, regulation that could support the development of and/or give incentives for new business ventures is lacking. “The policy or legislation that could maybe, sometimes, catalyze this kind of services is nonexistent. I am not very positive that the Finnish energy policy would be that innovative in the future, or that it could [support the] creation of new services.” The overall lack of policy-driven incentives for innovation intensifies the constraints for opportunity creation and discovery. The following quote vividly describes this situation: “I’ve been in the industry for long enough to understand that, wherever I put my head, some way or the other, policy will crop up behind the corner. When you look at installing new renewable capacity or managing flexibility… when you have house A and house B wanting to talk to each other, policy forbids it. It is not possible. And here, you would like to see energy resources being shared so that the energy would never leave the neighborhood. It would stay in the neighborhood, and you would not need the huge cable to the neighborhood because the neighborhood could maybe have storage within that community. Now, we’re putting in the big cable so that we can produce the energy in a centralized plant, in a volume-efficient way, burning some sort of fuel, at some sort of location.” Finally, the tendency of policy decisions to lag behind technological developments inhibits the implementability of new services and, thus, obstructs the innovation and commercialization processes of services providers: “…like, for instance, if we would say to a network, ‘We can make sure that you can [get] five years’ more life out of your substation with this flexibility management service,’ the network will go, ‘well, that is all fine, but, in the model, I am only reimbursed for ten years of using the station. Every year after that, it takes out my benefits from my balance sheet because of the regulatory model. If I don’t buy a new base station, I’m going to lose money.’” This is a representative example of policy hindering development and creating major risks for energy companies considering adapting to services. 4.4 Institutional complexity Institutional complexity was perceived as a key hindrance for the renewal of the energy sector. The energy field is rather complicated, with multiple objectives and logics that can be partially conflicting and may lack easy solutions. “When researchers make calculations and models in Excel, they keep adding rows on how the Finnish electricity system should look in 2050, so start from scratch… and yet, there is a long history of existing infrastructure, so you cannot assume us to dismantle the existing infrastructure and start a new one from scratch... So, a mere academic calculation on how the energy system could be 1BHF  transformed with unlimited resources, it’s a bit too theoretical, and it’s not applicable in practice.” Thus, the transition to a distributed energy system in which energy is provided and consumed using smart, digital solutions is an extremely difficult, system-level challenge with system-level renewal requirements. For example, there are objectives for climate policy both at the EU level and nationally. The primary means to achieve these goals is through reduced emissions, investments in renewable energy, and energy efficiency. However, these have become separate objectives, which complicates things even further. For example, managing emissions alone does not necessarily provide sufficient incentive to change the energy system. The complex regulatory environment can slow decision-making and, thus, development. The energy sector is one of the most central industries in Finnish society, which increases the role of authorities. Compared to other industries, authorities highly influence the business environment in the energy sector. The multiple authorities and ministries involved in decision-making also complicates the regulatory environment, creating unnecessary bureaucracy and decelerating investments. 4.5 Synthesis: Institutional counterforces to innovation ecosystem emergence Our analysis suggests the mutual reinforcement of regulation, policymaking ambiguities, the inertia of incumbent actors, and cognitive constraints for opportunity recognition inhibit innovation ecosystem emergence. In addition, institutional complexity functions as an overarching barrier that further sustains the rigidity of incumbent actors and policymakers and creates additional cognitive constraints for opportunity recognition. As described in Error! Reference source not found., incumbent actors’ unwillingness to change due to their path-dependent histories (i.e., “incumbent actor inertia”) feeds into and reinforces the cognitive constraints for opportunity recognition. Specifically, taken-for-granted assumptions, the legitimacy of current business logics [25], [41], and the stabilizing influence of shared norms [36] make it difficult for incumbent actors to identify opportunities for new business. “We have such a strong tradition in that area, so if you build a pulp mill and you can sell the pulp for the global markets, then you are in a position where this energy comes from almost nothing invested. So, it’s not expensive at all in Finland.” On the other hand, cognitive constraints for opportunity recognition can also reinforce incumbent firms’ inability to overcome organizational inertia. For instance, due to market uncertainties related to the low electricity prices and a lack of demand, energy companies do not see the opportunity to develop products and services for customers to support, for example, energy efficiency or demand response. Thus, as there is little motivation to change, the inertia of energy companies increases; they do not respond to the changing environment, but instead continue with previously learned practices (see also [21], [16], [42]). The ambiguities in the policy environment create further cognitive constraints for opportunity recognition. First, the discontinuity between political regimes makes it difficult for energy companies to know which investments are wise and sustainable. In addition, the absence of explicit intentions and decisions concerning concrete measures, as perceived by many of our respondents, does not send the necessary policy signals about where the best opportunities lie. Finally, the traditional view in the energy sector that energy production must occur on a large scale has led to policies focusing on very large centralized energy production. These policies have failed to promote distributed, renewable energy generation, which is where new energy sector ecosystems are emerging. However, renewables have gradually gained a very large market share and do not require large centralized units. Therefore, as policy ambiguities based on the current policy environment increase, it is unclear how actors in the energy sector should make use of common investments in networks or what the nature of the services should be. Prior literature has highlighted the central role of the wider environment in fostering the acceptance of innovation and supporting and sustaining changes once they occur [36]. However, according to our findings, the complexity of this wider environment can create an additional systemic barrier that actually sustains the full range of cognitive, normative, and regulative constraining forces. First, with respect to incumbent actor inertia and the cognitive constraints on opportunity recognition, many of the actors in the energy sector lack an overall vision or understanding of what is required for a system-level transition. Therefore, as “this kind of big picture, it is missing,” incumbent actors become unable to overcome their organizational inertia and drive change in the existing field. For smaller actors with less bargaining power, the field’s complexity makes it challenging to recognize and tap into opportunities because “it takes quite a lot of more complex business models and networks.” Second, the complexity of the overall energy field also reinforces the ambiguities in policymaking. In particular, as described in the previous sections, policymakers must consider multiple 1BHF  objectives and sometimes conflicting factors in their decision-making. Hence, there is no “quick fix where you put out some easy solution,” especially when “as regulators, we will still have to be equal and transparent and not discriminate rules in the future for these guys who might not be able to participate in the market.” 5. Discussion and conclusions Our study has examined ecosystem emergence barriers in the energy sector. As summarized in Figure 1, we found four system-level barriers: incumbent actor inertia, regulation and policymaking ambiguities, cognitive constraints for opportunity recognition, and institutional complexity. These interlinked barriers sustain the existing status quo and prevent emerging and established actors from creating new ecosystems in the field. A better understanding of these barriers would enable focal actors driving new ecosystem initiatives to identify potential system-level hindrances and find solutions that potentially overcome some of these barriers. The study also informs policymakers on the difficulties in generating new ecosystems in an established and institutionalized field. The study contributes to several literature streams, which we discuss below. Practical and policy implications, as well as limitations and future research directions, are then discussed. 5.1 Theoretical contributions First, our study contributes to the little-researched avenue of ecosystem emergence and the pre- emergence phase of ecosystems. As discussed by [45], scholars tend to take the existence of an ecosystem for granted, meaning that not enough attention has been paid to the earlier stages of an ecosystem (i.e., its emergence). While some studies have discussed how individual actors can deliberately build innovation ecosystems (e.g., [33], [47], [31]), our study shows that the “natural” emergence of broader innovation ecosystems involves a much more multifaceted set of issues. Thus, it is important to critically examine the broader institutional environment and particular organizational fields (e.g., the energy sector) when analyzing how the grassroots emergence of innovation ecosystems is constrained. Our empirical findings identify several institutional counterforces that together act as system-level barriers. Such understanding is crucial for ecosystem scholars, as it provides a more holistic outlook than the classic entry barriers literature. Furthermore, while our findings mostly relate to the emergence barriers of innovation ecosystems, they might also support a deeper understanding of the barriers to broader ecosystem renewal and transformation, though this is a question for further research to elaborate. Second, our study contributes to the research on institutional barriers to new technology (e.g., [18] [19], [20], [13], [25]), focusing particularly on the field level [14], [48]. Our study shows that organizational fields involve a variety of field-sustaining mechanisms that mutually reinforce one another over time. Interestingly, our findings on “regulation and policymaking ambiguity,” “incumbent actor inertia,” and “cognitive constraints for opportunity recognition” are closely linked to the respective dimensions of regulative, normative, and cognitive legitimacy that are used in institutional theory to explain the emergence of new technology. Therefore, different dimensions of legitimacy seem to be an important pre-condition for ecosystem emergence. However, our findings complement these existing perspectives in the institutional literature by providing a more overarching view of the dynamic and interlinked barriers to ecosystem emergence (see Figure 1). These findings contribute to the calls for more research to understand field changes and related restraining forces (cf. [49]). Thus, our study informs the stream of research on how field change is inhibited by incumbents and “institutional elites” with the power to manage and constrain change. 5.2 Practical and policy implications Our results provide practical insights for actors seeking to understand field-level change and “system- level” innovations. Our study first reveals the systemic interdependence and consequent inertia hampering the adoption of digitalization and the exploration of business models. The heavy and very long investment cycles of traditional energy sources make interdependent key actors hesitant to pursue any disruptive forces. Compared to other traditional industries, it is more likely that the new digital business models would come from outside rather than inside the traditional industry. Yet, due to the energy industry’s systemic nature, and without the support of incumbent players, the emergence of innovation ecosystems is difficult, if not impossible. However, the incumbent players lack the capabilities and the mindset required to build customer-driven digital services. This is a vicious circle that is further hampered by the lack of active policies and regulation supporting industry renewal. With the advent of new technologies, we expect similar challenges to apply across different industries. Indeed, most major innovations require changing 1BHF  and/or challenging existing institutions (regulative, normative, cognitive); therefore, we expect that our results will also apply to other contexts. With respect to the energy sector, it seems that the movement toward digitalization and related renewal is only possible if influential individuals and key stakeholders can join forces for collective strategic action. A joint understanding and shared vision of the energy sector transition is needed, as is the ability to influence the policymaking, regulation, and infrastructures required for research, development, and piloting [17]. However, this transition also requires new players. Innovation rarely comes from industry incumbents [9], and this seems to be the prevailing situation in the Finnish energy sector. 5.3 Limitations and future research directions This study has context-specific and generalizability-related limitations inherent to any exploratory qualitative study, including sector and country specificity. Our findings concerning barriers might be most applicable to other investment-intensive industries with strong state regulation. However, we expect that the results provide a useful overview of the institutional complexities and field-sustaining mechanisms that inhibit ecosystem emergence. Based on the findings, and acknowledging the limitations, our study provides several avenues for future research. Researchers could focus on different types of processes through which emergence barriers are dissolved, such as institutional work, cooperative interactions between incumbents and entrants [2], and relevant market mechanisms [13]. For instance, it would be interesting to study how individual actors (private or public) can help address and resolve ecosystem emergence barriers by reducing uncertainty, generating collective vision, and creating various types of incentives to join a new ecosystem [see also 6]. In this regard, research integrating social movements, organization theory [15] and institutional entrepreneurship could be potentially useful lenses for examining the emergence of new ecosystems [11]. Finally, it would be interesting to examine institutional complexity (a major barrier to emergence) in more depth. For instance, some institutional scholars have suggested that institutional complexity is also a facilitator of new initiatives [39], [37]. This double role of complexity as both a barrier and an enabler is a fascinating direction for future research in innovation ecosystem emergence. 6. References [1] L. Aarikka-Stenroos and P. Ritala, “Network management in the era of ecosystems: Systematic review and management framework,” Industrial Marketing Management, vol. 67, pp. 23-36, 2017. [2] R. Adner and R. Kapoor, “Value creation in innovation ecosystems: How the structure of technological interdependence affects firm performance in new technology generations,” Strategic Management Journal, vol. 31, pp. 306-33, 2010. [3] R. Adner and R. Kapoor, “Innovation ecosystems and the pace of substitution: Re‐examining technology S‐curves,” Strategic Management Journal, vol. 37, pp. 625-48, 2016. [4] H. E. Aldrich and C. M. Fiol, “Fools rush in? The institutional context of industry creation,” Academy of Management Review, vol. 19, no. 4, pp. 645-70, 1994. [5] S. S. Ansari, R. Garud, and A. Kumaraswamy, “The disruptor's dilemma: TiVo and the US television ecosystem,” Strategic Management Journal, vol. 37, pp. 1829-53, 2016. [6] E. Autio and L. Thomas, “Tilting the playing field: Towards an endogenous strategic action theory of ecosystem creation,” in Open Innovation, Innovation Ecosystems, and Entrepreneurship: Multidisciplinary Perspectives, S. Nambisan, Ed. New Jersey: World Scientific Publishing, 2016. [7] E. Autio and L. Thomas, “Innovation ecosystems: Implications for innovation management,” in The Oxford Handbook of Innovation Management, M. Dodgson, N. Philips, & D. M. Gann, Eds. Oxford: Oxford University Press, 2016, pp. 204-28. [8] K. Charmaz, Constructing Grounded Theory. London: SAGE Publications Ltd., 2014. [9] C. Christensen, The Innovator’s Dilemma. Cambridge, MA: Harvard Business School Press, 1997. [10] B. Clarysse, M. Wright, J. Bruneel, and A. Mahajan, “Creating value in ecosystems: Crossing the chasm between knowledge and business ecosystems,” Research Policy, vol. 43, pp. 1164-76, 2014. [11] M. T. Dacin, J. Goldstein, and W. R. Scott, “Institutional theory and institutional change: Introduction to the special forum,” Academy of Management Journal, vol. 45, pp. 45- 57, 2002. [12] B. Dattée, O. Alexy, and E. Autio, “Maneuvering in poor visibility: How firms play the ecosystem game when uncertainty is high,” Academy of Management Journal, vol. 61, no. 2, pp. 466-98, 2018. [13] M. Dijk, P. Wells, and R. Kemp, “Will the momentum of the electric car last? Testing an hypothesis on disruptive innovation,” Technological Forecasting and Social Change, vol. 105, pp. 77-88, April 2016. [14] P. J. DiMaggio and W. W. Powell, “The iron cage revisited: Institutional isomorphism and collective rationality in organizational fields,” American Sociological Review, vol. 48, pp. 147-60, 1983. [15] A. Dubois and L. E. Gadde, “Systematic combining: An abductive approach to case research,” Journal of Business Research, vol. 55, no. 7, 2002, pp. 553-60. [16] J. Dutton and S. Jackson, “Categorizing strategic issues: Links to organizational action,” Academy of Management Review, vol. 12, pp. 76-90, 1987. 1BHF  [17] N. Fliegstein and D. McAdam, “Toward a general theory of strategic action fields,” Sociological Theory, vol. 29, pp. 1-26, 2011. [18] F. W. Geels, “Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study,” Research Policy, vol. 31, pp. 1257-74, 2002. [19] F. W. Geels, “From sectoral systems of innovation to socio-technical systems: Insights about dynamics and change from sociology and institutional theory,” Research Policy, vol. 33, pp. 897-920, 2004. [20] F. W. Geels and J. Schot, “Typology of sociotechnical transition pathways,” Research Policy, vol. 36, pp. 399-417, 2007. [21] C. G. Gilbert, “Unbundling the structure of inertia: resource versus routine rigidity,” Academy of Management Journal, vol. 48, no. 5, pp. 741-63, 2005. [22] B. G. Glaser and A. L Strauss, Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine, 1967. [23] B. R. Golden, “Research notes. The past is the past—Or is it? The use of retrospective accounts as indicators of past strategy,” Academy of Management Journal, vol. 35, no. 4, pp. 848-60, 1992. [24] R. Gustafsson, M. Jääskeläinen, M. Maula, and J. Uotila, “Emergence of industries: A review and future directions,” International Journal of Management Reviews, vol. 18, pp. 28-50, 2016. [25] J. Markard, S. Wirth, and B. Truffer, “Institutional dynamics and technology legitimacy—A framework and a case study on biogas technology,” Research Policy, vol. 45, pp. 330-44, 2016. [26] J. F. Moore, “Predators and prey: A new ecology of competition,” Harvard Business Review, vol. 71, no. 3, pp. 75-86, 1993. [27] D. S. Oh, F. Phillips, S. Park, and E. Lee, “Innovation ecosystems: A critical examination,” Technovation, vol. 54, pp. 1-6, 2016. [28] M. Parlett and D. Hamilton, “Evaluation as illumination: A new approach to the study of innovatory programs,” Edinburgh: Centre for Research in the Educational Sciences, University of Edinburgh, 1972. [29] A. Pellinen, P. Ritala, K. Järvi, and L. M. Sainio, “Taking initiative in market creation—A business ecosystem actor perspective,” International Journal of Business Environment, vol. 5, no. 2, pp. 140-58, 2012. [30] M. E. Porter, Competitive Advantage: Creating and Sustaining Superior Performance, New York: Free Press, 1985. [31] P. Ritala, V. Agouridas, D. Assimakopoulos, and O. Gies, “Value creation and capture mechanisms in innovation ecosystems: A comparative case study,” International Journal of Technology Management, vol. 63, pp. 244-67, 2013. [32] P. Ritala and A. Almpanopoulou, “In defense of 'eco' in innovation ecosystem,” Technovation, vol. 60, pp. 39-42, 2017. [33] R. Rohrbeck, K. Hölzle, and G. H. Gemünden, “Opening up for competitive advantage—How Deutsche Telekom creates an open innovation ecosystem,” R&D Management, vol. 39, pp. 420-30, 2009. [34] C. G. Sandström, “The non-disruptive emergence of an ecosystem for 3D printing—Insights from the hearing aid industry's transition 1989–2008,” Technological Forecasting and Social Change, vol. 102, pp. 160-8, 2016. [35] G. Schreyögg and J. Sydow, “Organizational path dependence: A process view,” Organization Studies, vol. 32, no. 3, pp. 321-35, 2011. [36] W. R. Scott, Institutions and organizations, 4th ed. Thousand Oaks, CA: Sage, 2014. [37] J. Siltaloppi, K. Koskela-Huotari, and S. L. Vargo, “Institutional complexity as a driver for innovation in service ecosystems,” Service Science, vol. 8, no. 3, pp. 333-43, 2016. [38] R. R. Sinkovics and E. A. Alfoldi, “Progressive focusing and trustworthiness in qualitative research—The enabling role of computer-assisted qualitative data analysis software (CAQDAS),” Management International Review, vol. 52, pp. 817-45, 2012. [39] M. Smets and P. Jarzabkowski, “Reconstructing institutional complexity in practice: A relational model of institutional work and complexity,” Human Relations, vol. 66, no. 10, pp. 1279-1309, 2013. [40] M. Y. Smith and R. Stacey, “Governance and cooperative networks: An adaptive systems perspective,” Technological Forecasting and Social Change, vol. 54, no. 1, pp. 79-94, 1997. [41] R. Suddaby, A. Bitektine, and P. Haack, “Legitimacy,” Academy of Management Annals, vol. 11, pp. 451-78, 2017. [42] B. M. Staw, L. Sandelands, and J. Dutton, “Threat rigidity effects in organizational behavior,” Administrative Science Quarterly, vol. 26, pp. 501-24, 1981. [43] L. D. Thomas and E. Autio, “The fifth facet: The ecosystem as an organizational field,” Academy of Management Proceedings, p. 10306, 2014. [44] M. Tsujimoto, Y. Kajikawa, J. Tomita, and Y. Matsumoto, “A review of the ecosystem concept—Towards coherent ecosystem design,” Technological Forecasting and Social Change, 2017. [Online]. Available doi: 10.1016/j.techfore.2017.06.032 [45] K. Valkokari, “Business, innovation, and knowledge ecosystems: How they differ and how to survive and thrive within them,” Technology Innovation Management Review, vol. 5, pp. 17-24, 2015. [46] S. L. Vargo, H. Wieland, and M. A. Akaka, “Innovation through institutionalization: A service ecosystems perspective,” Industrial Marketing Management, vol. 44, 63- 72, 2015. [47] P. J. Williamson and A. De Meyer, “Ecosystem advantage,” California Management Review, vol. 55, pp. 24- 46, 2012. [48] M. Wooten and A. J. Hoffman, “Organizational fields: Past, present and future,” in The Sage Handbook of Organisational Institutionalism, R. Greenwood, C. Oliver, K. Sahlin, and R. Suddaby, Eds. London: Sage, 2008, p. 130- 147. [49] C. Zietsma, P. Groenewegen, D. M. Logue, and C. B. Hinings, “Field or fields? Building the scaffolding for cumulation of research on institutional fields,” Academy of Management Annals, vol. 11, pp. 391-450, 2017. 1BHF  Publication IV Almpanopoulou, A., Bergman, J-P., Ahonen, T., Blomqvist, K., Ritala, P., Honkapuro, S., and Ahola, J. Emergence of energy services ecosystems: Scenario method as a policy enabler Journal of Innovation Management 5(1), pp.58–77, 2017 © 2017, Almpanopoulou, A., Bergman, J-P., Ahonen, T., Blomqvist, K., Ritala, P., Honkapuro, S., and Ahola, J. Acknowledgement: Right for first publication: Journal of Innovation Management licensed under a Creative Commons Attribution License Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 HANDLE: http://hdl.handle.net/10216/103566 SM: Dec/2015 AM: Feb/2017 ISSN 2183-0606 http://www.open-jim.org http://creativecommons.org/licenses/by/3.0 58 Emergence of Energy Services Ecosystems: Scenario Method as a Policy Enabler Argyro Almpanopoulou1*, Jukka-Pekka Bergman1, Tero Ahonen2, Kirsimarja Blomqvist, Paavo Ritala1, Samuli Honkapuro2, Jero Ahola2 1School of Business and Management, Lappeenranta University of Technology, Finland; *School of Business and Management, Lappeenranta University of Technology; P.O Box 20, Lappeenranta, Finland 2School of Energy Systems, Lappeenranta University of Technology, Finland argyro.almpanopoulou@lut.fi, Jukka-Pekka.Bergman@lut.fi, Tero.Ahonen@lut.fi, Kirsimarja.Blomqvist@lut.fi, Paavo.Ritala@lut.fi, Samuli.Honkapuro@lut.fi, Jero.Ahola@lut.fi Abstract. The very nature of the energy sector, as a highly regulated and capital- intensive sector, as well as the challenges imposed by the global transition to renewable energy, have made the emergence of innovation ecosystems, which are necessary for the development and commercialization of new solutions, rather challenging. We examine the emergence of energy services ecosystems from a policy perspective, suggesting the scenario method as an enabler for focusing the attention of relevant actors and identifying triggering events that guide their activities toward a shared future. We illustrate our arguments using three case examples from Finnish public policy. Our study contributes to the nascent literature of ecosystem emergence and public innovation policy in the field of energy services. Keywords: Ecosystem, innovation, emergence, energy services, scenario method, research policy. 1 Introduction The literature on business and innovation ecosystems has been accumulating for over two decades, beginning with the seminal contribution by Moore (1993). The ecosystem concept has since attracted significant attention, especially within the body of practitioner and managerial literature, which has largely focused on how ecosystems can be managed around focal actors, technologies, or platforms (e.g. Iansiti and Levien, 2004; Iyer and Davenport, 2008; Rohrbeck et al., 2009; Williamson and De Meyer, 2012). Furthermore, research on innovation ecosystems has concentrated on how actors organize into systems around new developments, technologies, and ideas (e.g. Autio and Thomas, 2013; Ritala et al., 2013). One important question that is still rather untapped relates to how innovation and business ecosystems emerge: that is, how actors begin to organize themselves around interdependent ecosystems with shared goals, visions, and purposes. While self-organizing is a key attribute of business ecosystems (Peltoniemi, 2006), policy interventions are often helpful Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 59 when ecosystems are being built around new technologies and innovations (Clarysse et al., 2014). To better understand how ecosystem emergence can be facilitated, in this paper, we focus on how public policy initiatives enable the emergence of ecosystems around energy sector innovations. Existing literature has begun to study, for example, the role of public funding and knowledge in enabling ecosystem emergence (Clarysse et al., 2014). In the fields of renewable energy and energy services, facilitating the emergence of new business ecosystems is an especially relevant public policy context. While the literature on energy policy has identified the importance of public policy initiatives (e.g. Lewis and Wiser, 2007; Lund, 2007), there is still not sufficient evidence of the particular mechanisms that enable participants to focus their attention and cognition toward mutually shared goals and future development paths. In this paper, we suggest that scenario methods can function as public policy intervention mechanisms for enabling and facilitating the emergence of a new energy service ecosystem. We frame our arguments within a hierarchy of systems, including both the broader national innovation system and the business and innovation ecosystems that emerge with (and without) the influence of the national innovation system. For example, the national innovation system consists of universities, research centers, large and small firms, and various legal and regulatory institutions. By the term energy services ecosystem, we refer to an innovation ecosystem consisting of both private and public actors interacting in various innovation- and business-related activities. In this sense, we build upon a recent conceptualization of innovation ecosystems as “clusters (physical or virtual) of innovation activities around specific themes (e.g., biotechnology, electronics, pharmaceutical and software)” (Ritala et al., 2013, p.248). Our paper uses several case illustrations from Finland to understand ecosystem emergence in the energy services sector. The energy sector in Finland (and worldwide) is a highly regulated and capital-intensive sector, which makes the “natural” emergence of new energy services ecosystems rather challenging. Thus, we argue that, especially in this context, the Finnish innovation system can play an important role as an enabler for the emergence of new energy services ecosystems. We specifically concentrate on policy interventions and related scenario work as mechanisms that facilitate the emergence of new ecosystems, including, in our case, the energy services ecosystem. We argue that the scenario method and related processes focus the attention of various ecosystem actors, while also supporting the triggering events that guide future development. To support our argumentation, we examine three cases of different research programs financed by TEKES (the national agency for innovation development) and Academy of Finland innovation system strategic initiatives. Recent literature has focused on the transformation from loosely coupled research and development collaborations to more determined business and innovation ecosystems (Möller and Rajala, 2007; Aarikka-Stenroos and Sandberg, 2012; Clarysse et al., 2014). Another stream of literature has examined how ecosystems are built and how they emerge Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 60 in the first place (e.g. Moore, 1993; Ritala et al., 2013). Our study contributes to these streams of literature from a public policy intervention perspective, as we suggest that the scenario method and related processes can play an important role in the emergence of new innovation ecosystems. With this paper, we aspire to initiate discussion and inspire future studies on the impact of policy intervention on the emergence of innovation ecosystems: a phenomenon that is little studied. We argue that the potential of the innovation ecosystem may not be fully realized without such mechanisms as the scenario process. Using our case examples, we illustrate how potential knowledge and resources are mobilized for new ecosystem emergence, how the relevant stakeholders can create shared understandings of the future, and what kinds of triggering mechanisms can encourage passive actors to actively engage, take risks, and commit. Our paper is organized as follows: We begin with a brief discussion of the emergence of ecosystems, followed by a brief description of the role of scenario methods for focusing and triggering this emergence. Next, we present three illustrative public policy cases in the field of energy services, focusing in particular on the attitudes, cognitions, decisions, and actions of relevant actors participating in the scenario method and related processes. 2 Understanding the emergence of ecosystems The innovation ecosystem, as a concept, has been used to describe the increasing emphasis on the interdependency and co-evolution of individual actors (Autio and Thomas, 2013), such as suppliers, customers, governments, and universities. A seminal contribution to the literature of ecosystems in the business and innovation context was made by James Moore (1993), who adopted the biological metaphor of the “ecosystem” to describe how organizations and individuals interact and evolve in systems that operate very similarly to those that we can observe in nature. The key insights, which were later developed by other authors, were built on the systemic nature of ecosystems, including the principles of shared environment, co-evolution, interdependence, and ecosystem leadership (e.g. Moore, 1993; Iansiti and Levien, 2004). Recently, the scope of the term “ecosystem” has expanded significantly to include platform ecosystems (e.g. Thomas et al., 2014), technology ecosystems (e.g. Wareham et al., 2014), and service ecosystems (e.g. Akaka et al., 2013). The birth and evolution of ecosystems has been one of key topics ever since the seminal contribution by Moore (1993), who established the concept of the ecosystem life cycle, which consists of steps of birth, expansion, leadership, self-renewal, and decline/death. However, the main focus of ecosystem literature has been on explaining or solving issues faced by the focal actor or the ecosystem leader (e.g. Iansiti and Levien, 2004). Specifically, prior literature has widely studied how focal actors operate in ecosystems and how they create and organize them by imposing rules for other actors. Empirical investigations of large, incumbent companies and their already established ecosystems have represented the main approach in much of the extant ecosystem research (e.g. Iyer and Davenport, 2008; Isckia, 2009; Rohrbeck et al., 2009). Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 61 Prior literature either implicitly or explicitly grants significant power to the focal actor in designing the innovation ecosystem, neglecting the roles and influence of other, non-focal (e.g. entrepreneurial) actors within the ecosystems they inhabit (e.g., Ozcan and Eisenhardt, 2009; Hallen and Eisenhardt, 2009). As stated earlier empirically, the innovation ecosystem literature has largely studied innovation ecosystems organized around a technological platform (e.g., Gawer and Cusumano, 2014; Wareham et al., 2014) or a single focal actor (e.g., Leten et al., 2013), assuming that this focal actor can direct the future of the ecosystem as a whole. However, this approach is rather myopic, since the key to the emergence of innovation ecosystems is the connection between micro and macro behaviors and the cooperative and competitive interactions among individual actors (Smith and Stacey, 1997; Peltoniemi, 2006). Namely, emergence refers to the phenomenon through which individual actors’ motives and actions lead to unpredictable population-level behavior (Peltoniemi, 2006). In other words, emergence occurs as a result of dynamic interactions and coevolutions among individual actors that lead to unanticipated outcomes, such as the rise of larger entities (e.g. innovation ecosystems that exhibit properties possessed by none of the systems’ actors) (Holland, 1997; Midgley, 2008). Simply put, the whole is larger than the sum of its parts. Further, when the link between action and long-term outcome is lost in the interactions between the actors and the system, it is impossible for an external actor or powerful member of the system to control or design the system’s behavior. Instead, the behavior emerges (as described by Smith and Stacey, 1997, p.83). In innovation ecosystems, unlike in biological ecosystems, selection forces are not unknown to those experiencing them; instead, they involve learning and deliberate efforts by purposive actors to influence their environment (Garnsey and Leong, 2008; Garnsey et al., 2008). Therefore, Garnsey and Leong (2008) argue that actors can deliberately transform their environments, including the very selection forces that act upon them. This indicates the scope for proactive decision making and motivated action (cf. Penrose, 1995, p. 3). In fact, we argue that investors and policy makers, as members of the wider innovation ecosystem, are in a position to influence the emergence and methods of operation of the forces of selection (see Garnsey and Leong, 2008; Clarysse et al., 2014). For example, through well-informed financial and networking support, these individuals are able to enable the emergence of the innovation ecosystems necessary to support the commercialization of emerging technologies (Garnsey and Leong, 2008). However, as Clarysse et al. (2014) show, policy makers’ support for research programs seeking knowledge creation does not automatically trigger the emergence of innovation ecosystems, since the value creation processes of innovation ecosystems are significantly different, implying that policies to support innovation ecosystems must be specifically tailored. The energy services ecosystem can be viewed as a complex system (see Cilliers, 2001) which is subject to constant inflows and outflows and which evolves over time. The system consists of actors, activities, and processes that are interdependent. The ecosystem evolves through changes in the actors themselves, as well as collective, system-level co-evolutions stemming from internal and external influences. During the process of emergence, the relevant actors appear and begin to conduct activities that are (at least partially) Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 62 interdependent from those of other actors. The actors also begin to coordinate their activities, with each taking a different role in the ecosystem (Moore, 1993; Iansiti and Levien, 2004). In order for innovation ecosystems (as social structures) to be sustained, there must be interactions among actors that are sufficiently recurrent and personal to create shared understandings, legitimations, and relations of acknowledged interdependence (Giddens, 1984). We view the role of the knowledge and shared cognition of ecosystem actors as an important precondition for emergence. We argue that one key benefit of the emergence of innovation ecosystems is the production and combination of knowledge necessary for innovation, which is dispersed among different, previously unconnected actors. Thus, an innovation ecosystem can be viewed as an integrating mechanism that allows for both knowledge exploration and knowledge exploitation (Valkokari, 2015) and that enables its actors to jointly address complex problems (Leten et al., 2013). Furthermore, we claim that innovation system-level policy tools and mechanisms can make such knowledge visible and provide opportunities for the actors who are potentially forming an ecosystem to create a shared vision and agenda. In particular, we focus on the scenario method as an intentional process that can focus the attention of ecosystem actors, enable the necessary social interaction, and facilitate a shared cognition over triggering events that guide actors towards a shared and plausible future. In the following section, we discuss the role of the scenario method as an enabler of ecosystem emergence. 3 Scenario method as an enabler of ecosystem emergence Scenarios are means to affect future development. The fundamental idea behind the scenario approach is to provide a structured way to create a dynamic and ongoing social interaction among individuals and to expand people’s thinking (Wack, 1985a; Wack, 1985b; Schoemaker, 1995; Schwartz, 1996). Scenarios express the vision and aims of a certain group of stakeholders. They help organizations and individuals develop and broaden the strategic thinking on possible future realities and facilitate an understanding of the fundamental drivers of business, market, and technological trends and changes (Masini and Vasquez, 2003; Wack, 1985b). Scenarios describe the complexity of phenomena that cannot be formally modelled (Schoemaker, 1997). Scenario processes make it possible to assess the competitive landscape in a new light, revealing alternative future development paths (Godet, 2000; Schoemaker, 1997). In the process of strategy-making, the scenario method has been used to create a holistic understanding of complex environments to focus actors’ operations towards a desired future (Schoemaker, 1993; Schoemaker, 1995). The use of scenarios reflects an organization’s proactive orientation (Godet, 2000), enhancing its organizational flexibility to respond to environmental uncertainty and future actions. The scenario method can provide a structured approach for dynamic and ongoing interactions among organizations to create intentional strategic conversations and dialectic processes (Schwartz, 1996). Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 63 The scenario process is established when there is a need for influencing the development of an organization or wider business environment (Wack, 1985a; Wack, 1985b), such as, in our case, energy service ecosystems. Since, from a certain stakeholder point of view, scenarios are intentional, they seek to produce new knowledge and focus the performance of participating organizations. This scenario process can be set by a single organization or political decision makers to influence public and business organizations’ decision making (Schwartz, 1996). Scenario networks vary from intra-organizational working groups to inter-organizational virtual networks, where individuals have access to a wider knowledge base, connections become more interactive, and more holistic interpretations are formed. For the purposes of this study, we view the scenario method as an enabler of ecosystem emergence, which takes place through 1) focusing the attention of ecosystem actors towards a certain direction, 2) enabling social interactions, and 3) making visible the triggering events that have a strong effect on the perceived futures of ecosystem actors. Scenario methods enable such focusing processes and the subsequent discovery of triggering events, which, together, help to facilitate ecosystem emergence when relevant actors are involved and influenced by the scenario work. First, scenarios are means to focus and communicate strategic intent with the organization and the wider stakeholder network. Second, as a structured process, it has been seen as an effective management tool facilitating social interaction in a networked context to explore the environment in order to understand complexity or recognize alternative paths to the most desired goal (de Jouvenel, 2000; Roubelat, 2000; Bergman et al., 2006). Therefore, the scenario process serves as a catalyst for channeling organizational resources towards new opportunities and goals. In other words, the scenario process works as a facilitated and structured context by enabling a group of individuals to serve as intermediaries (or interfaces) in interactions between the internal and external environments and by amalgamating them into a network to work on the same task under a shared vision (van der Heijden, 2002). Third, scenario processes can trigger the involved actors’ activities, thus leading them to address and develop resources towards shared goals. When there is a goal of affecting the development of industry or society, the shared vision is developed among the most influential stakeholders and disseminated to the wider stakeholder network to trigger the desired actions. Scenarios are descriptions of the most desired development paths toward these commonly accepted goals. They may provide new business opportunities or even trigger large-scale industry-level renwal. Fig. 1 summarizes the role of scenario methods as policy tools in enabling ecosystem emergence by focusing actors’ attention, discovering important triggering events that guide these actions, and identifying plausible scenarios that can be shared among ecosystem actors. Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 64 Fig. 1. Process for the scenario work. 4 Emergence of energy service ecosystems research in Finland Emerging ecosystems in the field of sustainable energy production, energy efficiency, and new services represent one of the feasible areas for scenario use methods. Since the energy sector is currently a subject of political and financial interest in Finland, the topics mentioned above are prominently visible in research programs funded by the Academy of Finland and TEKES (the Finnish Funding Agency for Innovation). These are the two most important state-owned financiers for research and innovation in the Finnish innovation system, and their objective is to create renewal and growth. These programs are introduced here to clarify the background of the three illustrative cases studied in following sections. In order to foster industrial renewal, political decision makers have recently enforced structural and financial changes within the Finnish innovation system. One of these policy making instruments has been the launching of strategic research initiatives for political decision making and (radical) industrial renewal. As a result of these changes, a new financing body, the Strategic Research Council (SRC) at the Academy of Finland, was established to provide funding for long-term and program-based research aimed at finding solutions to the major challenges facing Finnish society. The most important objectives of Focusing processes •Focus ecosystem actors and their cognition towards converging future developments • Focus is pre- determinated or takes shape during scenario work Triggering events •Events (technological and social) that are found during focusing processes •Have a strong expected effect in the future from the point of view of ecosystem actors • Events can be used to affect to enhance plausible scenarios Plausible scenarios •Descriptions of future developments paths including triggering events •Reveal plausible paths and triggering events that enable the ecosystem emergence Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 65 selected SRC programs are to provide support for evidence-based policy; to develop solutions for the regeneration of Finnish society; and to propose ideas for the future of business, industry, and working life. In 2015, the SRC launched programs related to the energy transition in Finnish society (SET) and the disruption of digital technologies in industry, including in the energy sector (DDI) have started. In addition, the Academy- sponsored project “Change in Business Ecosystems for Local Renewable Energy and Energy Efficiency—Better Energy Services for Consumers (USE)” applies the idea of business ecosystems to a context that extends actor networks from businesses to consumers and public actors. TEKES strategic research openings are projects seeking to achieve breakthroughs, create new high-level competences, and develop significant new areas of growth in Finland, all in pursuit of a larger goal of fostering the renewal of the Finnish economy. TEKES points out that these projects must have high levels of novelty, including truly new perspectives or unique combinations of topics, and that they need to have the potential to create significant and lasting change in Finnish economy. Furthermore, the visions of these projects must be simultaneously feasible, concrete, and challenging, since the projects will create competences that can be used to achieve goals that may initially seem impossible. The Neo- Carbon Energy project is one of the TEKES strategic research openings. Its objective is to establish a perspective on the needs, business opportunities, and societal implications of an emission-free energy system; to study the connections between the electricity grid and large-scale seasonal energy storage; and to explore its integration with other energy sectors (Landowski, 2014). One of the recognized key factors in the emergence of energy ecosystems is digitalization, which supports the transformation of an energy system from a centralized system to a more distributed one. This energy transformation is especially visible in Germany, the leading EU country in terms of its use of photovoltaic solar energy systems due to its Energiewende policy (Pegels and Lütkenhorst, 2014; International Energy Agency, 2015). These two factors will provide opportunities for new service development, industry renewal, and, thus, new businesses, which are being studied, with the help of scenario methods, in Academy (SET and DDI) and TEKES-funded projects (Neo-Carbon). These business models can change dramatically as the role of the customer transforms from that of a “consumer” to that of a “prosumer” (Pagani and Aiello, 2010). For example, in the consumer energy sector, digitalization is now visible through the use of Automatic Meter Readings (AMRs), which allow the remote monitoring of customer energy consumption with one-hour resolution; the use of Nord Pool spot price-based tariffs; and the development of services related to these options. AMRs can be considered physical components of smart grids, providing means for the automated control of active resources, including distributed generation, energy storage, and demand response (DR), which refers to flexibility in energy consumption (Koivisto et al., 2015). A promising service-based example of demand response applications is that of electric heating systems, which may alter their operation according to a given price or frequency signal to allow a required DR to be fulfilled without harm to the end user (There, 2015). AMR also provides technical Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 66 infrastructure for other third-party energy services, thus motivating the efficient use of energy and the active management of electric power quality (Logenthiran et al., 2012). - 5 Case projects and the role of the scenario method Scenario methods are currently applied in three different research projects within the energy sector (see Fig. 2). Of these, TEKES-funded Neo-Carbon Energy first launched with publicly available scenarios in 2014. Academy of Finland-funded projects Smart Energy Transition (SET) and Digital Disruption of Industry (DDI) followed in October 2015. This section introduces each project and its scenario work. Fig. 2. Timeline of the three projects. 5.1 Neo-Carbon Energy: Scenarios through Futures Cliniques The Neo-Carbon energy project seeks to design the operation principles and key components of a renewable energy system based solely on wind, solar, and sustainable hydro and biomass. Since the main challenge in solar and wind power is the intermittency of their generation, the key focus lies in seasonal storage solutions and solutions enabling the bridging of the electric power system with other energy systems, such as gas networks, transportation fuels, heat networks, industrial chemicals, etc. The main proposed solution for the energy storage problem is the power-to-gas (P2G) process, through which synthetic natural gas, SNG (i.e. methane), is produced from CO2 and H2 during times of excess solar and wind production. The natural gas infrastructure provides nearly infinite storage capacity for chemical energy, and the P2G solution can integrate the different energy systems (heat, power, and transportation). The aim of the Neo-Carbon Energy project scenarios is to recognize possible radically different energy futures with novel technology solutions and to identify what kinds of businesses these solutions can create. One key question involves how to present the Neo- Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 67 Carbon Energy system as attractive to citizens. In Neo-Carbon, there are four future scenarios for the year 2050 (illustrated in Fig. 3), all of which are transformative. In all scenarios, the world has undergone a third industrial revolution (see Rifkin, 2011), which includes revolutions in both energy production and communication technologies. In each scenario, energy is produced according to the Neo- Carbon energy model; however, the implementation of this solution, as well as people’s lifestyles, values, cultures, and business concepts, vary from scenario to scenario (Heinonen et al., 2015). Fig. 3. Four transformative scenarios in the Neo-Carbon Energy project (Heinonen et al., 2015) Tentative scenarios have been tested in the Futures Clinique (a participatory and exploratory future workshop, which is designed to anticipate especially radical futures and surprising effects [i.e. black swans]; see Heinonen and Ruotsalainen, 2013), during which participants (e.g. project members, government, business, and third sector representatives) work around a variety of scenario sketches. One of the challenges in employing such transformative scenarios, which involve varying socio-cultural aspects, seems to be that they might be overly abstract for primarily technology-oriented experts. However, since these experts were involved in the scenario processing in the Futures Clinique, they were, at least to some extent, committed to the ideas presented. Nevertheless, there is still work to be done to strengthen the links between these future scenarios and technical and economic-oriented research work. The Neo-Carbon energy project provides benefits for Finnish industry by introducing a novel energy system to leading industrial partners, educating decision makers, supporting Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 68 corporate-level strategy development, and identifying concrete business cases. Finally, the project designs and builds prototypes of the selected key technical devices, which the system requires in order to work. During the project, key companies can identify their roles within the energy system value chain and decide how they will subsequently invest in the subject. Ultimately, the project lays the foundations for a novel energy system and enables Finland to lead the transition toward this type of energy system, thus turning it into a business opportunity. The project’s research work is carried out by a multidisciplinary research team from the VTT Technical Research Centre of Finland, Lappeenranta University of Technology and the University of Turku Finland Futures Research Centre. The advisory board comprises industrial partners and provides internal pitching for the project by quarter-annually reviewing the outcomes of the project and directing its work. 5.2 Smart Energy Transition (SET) Disruptive technologies have been defined as advances that will transform life, business, and, ultimately, the global economy (Manyika et al., 2013). Renewable energy production and storage technologies are potentially disruptive technologies because they change not only the way we produce energy, but also the way we use energy, do business with energy, and innovate. Therefore, smart energy solutions can cascade into new business ecosystems, leading to radical shifts in the roles of producers, service providers, and consumers. The Smart Energy Transition project was launched in October 2015, and analyzes the ongoing global transition and its impacts on Finnish society, including, in particular, the potential benefits for cleantech, digitalization, and the bioeconomy. The SET consortium consists of seven Finnish universities and research institutes and four other organizations involved in researching and actively facilitating a sustainable smart energy transition in Finland. The work of these actors is divided into six work packages (see Fig. 4), whose progress is advised and accelerated by three expert panels and a transition arena for the demonstration of obtained results. Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 69 Fig. 4. Description of work packages in the SET project (SET, 2015). Scenario work is needed in the SET project to clarify the possibilities revealed by disruptions in the energy sector. Compared to the more generic scenario work done in the Neo-Carbon Energy project program, the scenario work in the SET project begins by providing an overall perspective on available solutions to produce, store, and consume renewables-based electrical energy through literature reviews and workshops. Hence, the focus is first on technical aspects. However, once these aspects have been explored and first alternative scenarios are constructed, policies, society, etc. will be considered. Based on the literature review and existing Neo-Carbon scenarios, two alternative scenarios for the year 2030 will be drafted. These will be publicly introduced in workshops and modified according to the results of the Delphi query, which is used to provide input about triggering events related to these scenarios. 5.3 Digital Disruption of Industry (DDI) The focus of the Digital Disruption of Industry project is on the economic and social implications of this disruption. The DDI project studies how the underlying fabric of current industries—how they operate, how they organize themselves, how they reason about their business and partners, and how they strategize—will be contested when novel digital (institutional) infrastructures with their own rules, norms, and mindsets begin to take form. The project focuses on several sectors of industry from an institutional perspective, which facilitates an evaluation of changes both in the national context and from a broader comparative perspective. The DDI project will yield a comprehensive study of the impact of digitalization, not only to industry itself, but also to its ecosystem partners, its stakeholders, and, more widely, its relevant societal institutions, such as business practices and models, regulation, management, and governance. Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 70 The DDI consortium consists of five Finnish universities and research institutes: Aalto, VTT Technical Research Centre of Finland, Lappeenranta University of Technology, ETLA and the University of Turku. This group collaborates with several organizations (large industry, small- and medium-sized enterprises (SMEs), startups and innovators, RTDI actors, government bodies, employees, customers, and consumers) in the targeted industrial and ICT ecosystems. These different actors collaborate in the planning, execution, and assessment of specific interventions related to concrete cases of digital disruption, the challenges involved, and the impacts on stakeholders. Further, within the consortium, there is close interaction with regard to information sharing, interactions with other stakeholders, roadmaps and scenarios, joint publications, events, and action plans for managing the disruption. The DDI is divided into five research work areas, which simultaneously tackle the two overall objectives of the project: the research objective of synthesizing an increasingly expressive scientific understanding of digital societal disruption, as seen through the lens of industry, and the policy objectives of creating an effective policy response to the institutional challenges raised by this disruption and of charting a route for Finnish companies and society through this change. The scenario process in DDI serves as a tool for active dialogue and interactions among policy, research, industry, and citizens, and this shared awareness creates the foundation for the research project. The scenario method will be used throughout the project to continuously analyze the context of digital disruption and industry transformation. Meta-scenarios will be used to identify the main driving forces of the operative environment, as well as the triggers beyond the shared cognitive frames that inform changes in future development paths. Further, the created meta-scenarios will provide normative descriptions of the uncertainties related to technology development, economic and social factors, and regulative and political actions for the next 15 years. 6 Focusing processes and triggering events in the case projects As described in the earlier focusing processes, social interactions and triggering events represent essential elements and outputs of scenario work. Focusing processes include events in which the different parties and their ways of thinking can be directed towards possible future paths. These enable the various actors to engage in vivid discussions and challenge possible future scenarios. Through these interactions, actors create a common understanding and a shared vision of the future. A trigger can represent an issue or event that is expected to "trigger" a chain of events or a future path to the future. Triggering events can either inspire or occur during scenario work, but in both cases, these events attract the interest of various parties to engage in the scenario work in order to prepare for the future. In the case of the Neo-Carbon Energy project, specific scenario work has already been performed, and in the SET and the DDI, the scenario work is ongoing. The SET and Neo- Carbon Energy projects are highly interlinked, since the SET builds upon the initial results of the Neo-Carbon project’s scenario work (see also Fig. 2 and the discussion above). These Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 71 two projects are more technologically oriented, with DDI taking a broader and more business-oriented perspective by focusing on multiple interconnected industries, including energy. Table 1 provides examples of focusing and triggering events for the scenario work in the Neo-Carbon Energy project, as well as for the planned and/or initial scenario work in the SET and DDI projects. The data resulting from the scenario work and the use of Delphi queries (see e.g. Glenn and Gordon, 2009) will reveal more detailed triggering events, such as abrupt changes in the energy production system. Table 1. Illustrative examples of focusing processes and triggering events in case projects Case project Role of scenario methods Focusing processes utilized/to be utilized Triggering events identified Neo-Carbon To provide insights into how the future RES-based energy world might be realized in four radically different transformative ways. Futures Clinique; different foresight methods (e.g. Futures Window, identification and impact analyses of weak signals and black swans, scenario narratives, etc.). Increasing peer-to-peer approaches, prosumerism, ecological awareness, the boom of startups with open-source principles, the increasing dominance of technological giants, and ubiquitous ICT. Smart Energy Transition consortium (SET) To provide an understanding of the rate and direction of energy transformation towards the selected scenarios Workshops with project partners; Delphi study with expert panels (tech, users, policy) for determining possible triggering events in the assumed scenarios. First workshop results: new startups and export companies, scarcity of resources, ecological disasters Digital Disruption of Industry (DDI) To enable active dialogue and interactions among the different actors of the wider energy ecosystem. Workshops with expert panel discussions; scenario work through the workshops; utilization of SET project Delphi study results applicable to this project. First ideas based on expert discussions: the shift in the Internet of Things from hype to reality as a techno- economic-social disruption that is expected to significantly influence the relative competitiveness of firms and nations. The scenario methods in the case projects serve as tools for fostering active dialogue and interactions among the various actors involved in the energy sector (e.g. policy makers, research institutions, companies, entrepreneurs, and even citizens). First, through the scenario work in the case projects, the different stakeholders can jointly recognize the driving forces and alternative future paths of the energy sector. In practice, the interested and relevant parties are invited in workshops and participative, facilitated discussions, through which they become aware and share their views of the nature and impact of future Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 72 developments in the energy sector, which then mobilize them to take an active role, work together, and build the necessary synergies in the planning, execution, and assessment of specific actions to respond or influence these developments. The scenario methods enable social interactions among different actors in, for example, sharing information and knowledge, building joint roadmaps, and generally co-creating effective joint responses (e.g. improving current networked processes or building new business models) to the uncertainties of energy technology development, as well as other economic, social, regulative, and political factors. The strategic research programs set by policy makers provide the incentive for and expectation that various stakeholders set up scenario processes that will enable them to learn about one another and the potential futures of the energy sector through social interaction. In this way, the scenario process can become a focal mechanism for the emergence and birth of interdependent ecosystem(s) with shared goals, visions, and purposes. 7 Conclusions This paper has discussed the emergence of new ecosystems in the area of energy services from a public policy perspective. We have developed a view of scenario methods as mechanisms that help to focus the attention of potential and current actors, as well as to create visibility for triggering events that are leading future developments. In so doing, our paper answers recent calls to better understand how public policy can help the creation of business and innovation ecosystems (Clarysse et al., 2014), as well as the birth and emergence of ecosystems in general (Ritala et al., 2013). The results suggest a range of research, policy, and practical implications, which are discussed in the following. 7.1 Research implications Our study contributes to the research on ecosystem emergence in general and provides implications for policy research in the field of energy services. First, as we discussed in the beginning of the study, ecosystem emergence is among the most important, but least studied phenomena within business and innovation ecosystem literature. Our papers suggest that facilitating the emergence of ecosystems might be necessary when self-organizing is not progressing sufficiently quickly. Ecosystem coordination is often managed by a strong focal actor (see e.g. Moore, 1993; Iyer and Davenport, 2008; Isckia, 2009; Rohrbeck et al., 2009); however, in the absence of such an actor, other mechanisms become useful in enabling the emergence and growth of ecosystems. This also highlights the essential linkages between the literatures of business and innovation ecosystems and public policy research streams, such as the research on triple-helix and national innovation systems (e.g. Martin and Johnston, 1999; Etzkowitz and Leydesdorff, 2000). Since business and innovation ecosystems are complex systems with open boundaries and constant inflows and outflows (Cilliers, 2001), the interdependencies and co-evolutions between public policy and private sector actors is an issue that is relevant Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 73 for practically any study related to emerging technologies and innovations. Second, it has been suggested that research programs, as policy making instruments, play an important role in the creation and exchange of knowledge among participating actors (Autio et al., 2008). In this study, we have argued that scenario workshops provide the time and space for ecosystem actors to share explicit and tacit knowledge. The role of these workshops is further accentuated in situations in which businesses do not yet see concrete business opportunities and when capturing the benefits of these opportunities requires the learning and development of competences among various ecosystem actors. Especially in the highly regulated energy industry, individual actors may not have sufficient incentive to take risks, take on ecosystem leadership roles, or invest in the building of ecosystems for new energy services (Iansiti and Levien, 2004). Thus, we argue that scenario work is a usable approach to study the future of emerging energy service systems. 7.2 Policy and practical implications In the Finnish new energy services context, there is no clear focal actor, single technology, or technology platform. This is also the case for many emerging technologies, which tend to face the “chicken and egg” problem. To overcome the chasms among initiative-taking, followership, and concrete actions, research programs and scenario work can be seen as especially helpful. Strategic research programs implemented by policy makers can be seen as (knowledge) platforms for connecting various ecosystem actors, since they build interdependence and require some level of coordination. Here, the recurrent interactions among knowledgeable and resourceful actors enabled by scenario work and related process can trigger the emergence of a more concrete ecosystem, which will begin to self-organize towards plausible future scenarios. In our research, we not only illustrate the use of the scenario method as a focusing and triggering mechanism for a single strategic research program, but also show the importance of the knowledge sharing mechanisms that link different strategic research programs. Each strategic research program has a specific focus, which may not be sufficient, on its own, to turn individual research and development collaborations into a concrete ecosystem. Rather, the knowledge sharing mechanisms function as linking mechanisms that connect complementary research programs to a larger knowledge ecosystem (Clarysse et al., 2014). Yet, without the scenario method and process as a focusing and triggering mechanism, the system could suffer from inertia and fail to realize its potential. Our results highlight the potential benefits of scenario work in this regard. 7.3 Limitations and future research directions This paper has limitations, especially regarding its generalizability. The case evidence presented in this paper should be treated as illustrative, since its purpose is to showcase the potential usage of scenario methods in the energy services sector context, rather than to prove cause-and-effect relationships. For instance, the SET and DDI research projects are still nascent; thus, their outcomes should be seen as plans for the actual scenario work to be Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 74 carried out. Overall, we have revealed the very first results of the scenario work. Future studies may build on the ideas presented on this paper in several ways. First, there is a need for studies to understand how scenario methods can facilitate ecosystem emergence. Our conceptualization of focusing processes and triggering events could serve as a foundation for conducting more data-rich case studies or broader quantitative studies. Second, more context-aware studies are needed to understand how energy sector actors, in particular, organize within ecosystems. Since many of the major global challenges are related to renewable energy, we need to know more about how business and innovation ecosystems are built around these challenges and what public policy can do to facilitate this development. 8 References Aarikka-Stenroos, L., & Sandberg, B. (2012). From new product development to commercialization through networks. Journal of Business Research, 65(2), 198-206. Akaka, M.A., Vargo, S.L., & Lusch, R.F. (2013). The complexity of context: A service ecosystems approach for international marketing. Journal of International Marketing, 21(4), 1-20. Autio, E., Kanninen, S., & Gustafsson, R. (2008). First-and second-order additionality and learning outcomes in collaborative R&D programs. Research Policy, 37(1), 59-76. Autio, E., & Thomas, L.D. (2013). Innovation ecosystems: Implications for innovation management. In M. Dodgson, D.M. Gann, N. Phillips (Ed.), The Oxford Handbook of Innovation Management, 204-288, Oxford: Oxford University Press. Bergman, J-P., Viljainen, S., Kässi, T., Partanen, J., & Laaksonen, P. (2006). Managing the exploration of new operational and strategic activities using the scenario method— Assessing future capabilities in the field of electricity distribution industry. International Journal of Production Economics, 104, 46-61. Cilliers, P. (2001). Boundaries, hierarchies and networks in complex systems. International Journal of Innovation Management, 5(2), 135-47. Clarysse, B., Wright, M., Bruneel, J., & Mahajan, A. (2014). Creating value in ecosystems: Crossing the chasm between knowledge and business ecosystems. Research Policy, 43(7), 1164-1176. de Jouvenel, H. (2000). A brief methodological guide to scenario building. Technological Forecasting and Social Change, 65, 37-48. Etzkowitz, H., & Leydesdorff, L. (2000). The dynamics of innovation: from National Systems and “Mode 2” to a Triple Helix of university–industry–government relations. Research Policy, 29(2), 109-123. Garnsey, E., & Leong, Y.Y. (2008). Combining resource-based and evolutionary theory to explain the genesis of bio-networks. Industry and Innovation, 15(6), 669-686. Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 75 Garnsey, E., Lorenzoni, G., & Ferriani, S. (2008). Speciation through entrepreneurial spin- off: The Acorn-ARM story. Research Policy, 37(2), 210-224. Gawer, A., & Cusumano, M. A. (2014). Industry platforms and ecosystem innovation. Journal of Product Innovation Management, 31, 417-433. Giddens, A. (1984). The Constitution of Society. Cambridge: Polity Press. Glenn, J.C., & Gordon, T.J. (2009). Futures Research Methodology Version 3.0. The Millennium Project. Godet, M. (2000). The art of scenarios and strategic planning: Tools and pitfalls. Technological Forecasting and Social Change, 65, 3-22. Hallen, B. L., & Eisenhardt, K. M. (2012). Catalyzing strategies and efficient tie formation: How entrepreneurial firms obtain investment ties. Academy of Management Journal, 55, 35-70. Heinonen, S., Karjalainen, J., & Ruotsalainen, J. (2015). Towards the Third Industrial Revolution— Neo-carbon energy futures clinique I. Finland Futures Research Centre. FFRC eBook 6/2015. Accessed 30th November 2015. https://www.utu.fi/fi/yksikot/ffrc/julkaisut/e-tutu/Documents/FFRC-eBook-6- 2015.pdf. Heinonen, S. & Ruotsalainen, J. (2013). Futures clinique—Method for promoting futures learning and provoking radical futures. European Journal of Futures Research, 1(7), 1-11. Holland, J.H. (1997). Emergence. Philosophica, 59(1), 11–40. Iansiti, M., & Levien, R. (2004). Strategy as ecology. Harvard Business Review, 82(3), 68- 81. Isckia, T. (2009). Amazon’s evolving ecosystem: A cyber-bookstore and application service provider. Canadian Journal of Administrative Sciences, 26(4), 332-343. Iyer, B. & Davenport, T. H. (2008). Reverse engineering: Google's innovation machine. Harvard Business Review, April, 59–68. Koivisto, V., Järventausta, P., & Mäkinen, S. J. (2015). Role of public regulators in demand response business ecosystem—Case New York State electric power market. In the Proceedings of 12th International Conference on the European Energy Market (EEM). pp. 1-5. IEEE. Landowski, E. (2014). Strategic research openings 2/2014. Accessed 30th November 2015. http://www.TEKES.fi/en/whats-going-on/application-schedules/strategic-research- openings-22014/ Leten, B., Vanhaverbeke, W., Roijakkers, N., Clerix, A., & Van Helleputte, J. (2013). IP models to orchestrate innovation ecosystems: IMEC, a public research institute in nano-electronics. California Management Review, 55, 51-64. Lewis, J.I. & Wiser, R.H. (2007). Fostering a renewable energy technology industry: An international comparison of wind industry policy support mechanisms. Energy Policy, Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 76 35(3), 1844-1857. Logenthiran, T., Srinivasan, D. & Shun, T.Z. (2012). Demand side management in smart grid using heuristic optimization. IEEE Transactions on Smart Grid, 3(3), pp. 1244- 1252. Lund, P.D. (2007). Effects of energy policies on industry expansion in renewable energy. Renewable Energy, 34(1), 53-64. Manyika, J., Chui, M., Bughin, J., Dobbs, R., Bisson, P. & Marrs, A. (2013). Disruptive technologies: Advances that will transform life, business, and the global economy. Accessed 30th November 2015. http://www.mckinsey.com/insights/business_technology/disruptive_technologies Martin, B. R., & Johnston, R. (1999). Technology foresight for wiring up the national innovation system: experiences in Britain, Australia, and New Zealand. Technological Forecasting and Social Change, 60(1), 37-54. Masini, E., & Vasquez, J. (2003). Scenarios as seen from a human and social perspective. Technological Forecasting and Social Change, 65, 49-66. Midgley, G. (2008). Systems thinking, complexity and the philosophy of science. Emergence: Complexity and Organization, 10(4), 55-73. Moller, K. & Rajala, A. (2007). Rise of strategic nets – New modes of value creation. Industrial Marketing Management, 36(7), 895-908. Moore, J.F. (1993). Predators and prey: A new ecology of competition. Harvard Business Review, 71(3), 75-83. Ozcan, P. & Eisenhardt, K.M. (2009). Origin of alliance portfolios: Entrepreneurs, network strategies, and firm performance. Academy of Management Journal, 52, 246-279. Pagani, G.A., & Aiello, M. (2010). Towards a service-oriented energy market: Current state and trend. Service-Oriented Computing: ICSOC 2010 International W orkshop, Revised Selected Papers, 203-209. Pegels, A., & Lütkenhorst, W. (2014). Is Germany’s energy transition a case of successful green industrial policy? Contrasting wind and solar PV. Energy Policy, 74, 522-534. Peltoniemi, M. (2006). Preliminary theoretical framework for the study of business ecosystems. Emergence: Complexity and Organization, 8(1), 10-19. Penrose, E. (1995). The Theory of the Growth of the Firm. Oxford: Oxford University Press. Rifkin, J. (2011). The third industrial revolution: How lateral power is transforming energy, the economy, and the world. London: Palgrave Macmillan. Ritala, P., Agouridas, V., Assimakopoulos, D., & Gies, O. (2013). Value creation and capture mechanisms in innovation ecosystems: A comparative case study. International Journal of Technology Management, 63(3-4), 244-267. Rohrbeck, R., Hölzle, K., & Gemünden, H. G. (2009). Opening up for competitive Journal of Innovation Management Almpanopoulou et al. JIM 5, 1 (2017) 58-77 http://www.open-jim.org 77 advantage–How Deutsche Telekom creates an open innovation ecosystem. R&D Management, 39(4), 420-430. Roubelat, F. (2000). Scenario planning as a network process. Technological Forecasting and Social Change, 65, 99-112. Schoemaker, P.J.H. (1993). Multiple scenario development: Its conceptual and behavioral foundation. Strategic Management Journal, 14, 193-213. Schoemaker, P. J. H. (1995), Scenario planning: A tool for strategic thinking. Sloan Management Review, W inter, 25-40. Schoemaker, P. J. H. (1997). Disciplined imagination – from scenarios to strategic options. International Studies of Management & Organization, 27(2), 43-70. Schwartz, P. (1996). The art of the long view— Planning for the future in an uncertain world. New York, USA: Doubleday Dell Publishing Inc. SET (2015). Smart energy transition project description. Accessed 30th November 2015. http://www.smartenergytransition.fi/?page_id=177andlang=en. Smith, M. Y., & Stacey, R. (1997). Governance and cooperative networks: An adaptive systems perspective. Technological Forecasting and Social Change, 54(1), 79-94. There (2015). There corporation solutions. Accessed 30th November 2015. http://www.therecorporation.com/solutions/. Thomas, L.D.W., Autio, E., & Gann, D.M. (2014). Architectural leverage: Putting platforms in context. Academy of Management Perspectives, 28(2), 198-219. Valkokari, K. (2015). Business, innovation, and knowledge ecosystems: How they differ and how to survive and thrive within them. Technology Innovation Management Review, 5, 17-24. van der Heijden, K., Bradfield, R., George, B., Cairns, G., & Wright, G. (2002). The sixth sense— Accelerating organizational learning with scenarios. Chichester, UK: John Wiley and Sons, Ltd. Wack, P. (1985a). Scenarios uncharted waters ahead. Harvard Business Review, Sept-Oct, 73-89. Wack, P. (1985b). Scenarios: Shooting the rapids. Harvard Business Review, Nov-Dec, 139-150. Wareham, J., Fox, P. B., & Cano Giner, J. L. (2014). Technology ecosystem governance. Organization Science, 25, 1195-1215. Williamson, P.J. & De Meyer, A. (2012). Ecosystem advantage: How to successfully harness the power of partners. California Management Review, 55(1), 24-46. ACTA UNIVERSITATIS LAPPEENRANTAENSIS 810. KÄYHKÖ, JORMA. Erityisen tuen toimintaprosessien nykytila ja kehittäminen suomalaisessa oppisopimuskoulutuksessa. 2018. Diss. 811. HAJIKHANI, ARASH. Understanding and leveraging the social network services in innovation ecosystems. 2018. Diss. 812. SKRIKO, TUOMAS. Dependence of manufacturing parameters on the performance quality of welded joints made of direct quenched ultra-high-strength steel. 2018. Diss. 813. KARTTUNEN, ELINA. Management of technological resource dependencies in interorganizational networks. 2018. Diss. 814. CHILD, MICHAEL. Transition towards long-term sustainability of the Finnish energy system. 2018. Diss. 815. NUTAKOR, CHARLES. An experimental and theoretical investigation of power losses in planetary gearboxes. 2018. Diss. 816. KONSTI-LAAKSO, SUVI. Co-creation, brokering and innovation networks: A model for innovating with users. 2018. Diss. 817. HURSKAINEN, VESA-VILLE. Dynamic analysis of flexible multibody systems using finite elements based on the absolute nodal coordinate formulation. 2018. Diss. 818. VASILYEV, FEDOR. Model-based design and optimisation of hydrometallurgical liquid- liquid extraction processes. 2018. Diss. 819. DEMESA, ABAYNEH. Towards sustainable production of value-added chemicals and materials from lignocellulosic biomass: carboxylic acids and cellulose nanocrystals. 2018. Diss. 820. SIKANEN, EERIK. Dynamic analysis of rotating systems including contact and thermal- induced effects. 2018. Diss. 821. LIND, LOTTA. Identifying working capital models in value chains: Towards a generic framework. 2018. Diss. 822. IMMONEN, KIRSI. Ligno-cellulose fibre poly(lactic acid) interfaces in biocomposites. 2018. Diss. 823. YLÄ-KUJALA, ANTTI. Inter-organizational mediums: current state and underlying potential. 2018. Diss. 824. ZAFARI, SAHAR. Segmentation of partially overlapping convex objects in silhouette images. 2018. Diss. 825. MÄLKKI, HELENA. Identifying needs and ways to integrate sustainability into energy degree programmes. 2018. Diss. 826. JUNTUNEN, RAIMO. LCL filter designs for paraller-connected grid inverters. 2018. Diss. 827. RANAEI, SAMIRA. Quantitative approaches for detecting emerging technologies. 2018. Diss. 828. METSO, LASSE. Information-based industrial maintenance - an ecosystem perspective. 2018. Diss. 829. SAREN, ANDREY. Twin boundary dynamics in magnetic shape memory alloy Ni-Mn- Ga five-layered modulated martensite. 2018. Diss. 830. BELONOGOVA, NADEZDA. Active residential customer in a flexible energy system - a methodology to determine the customer behaviour in a multi-objective environment. 2018. Diss. 831. KALLIOLA, SIMO. Modified chitosan nanoparticles at liquid-liquid interface for applications in oil-spill treatment. 2018. Diss. 832. GEYDT, PAVEL. Atomic Force Microscopy of electrical, mechanical and piezo properties of nanowires. 2018. Diss. 833. KARELL, VILLE. Essays on stock market anomalies. 2018. Diss. 834. KURONEN, TONI. Moving object analysis and trajectory processing with applications in human-computer interaction and chemical processes. 2018. Diss. 835. UNT, ANNA. Fiber laser and hybrid welding of T-joint in structural steels. 2018. Diss. 836. KHAKUREL, JAYDEN. Enhancing the adoption of quantified self-tracking wearable devices. 2018. Diss. 837. SOININEN, HANNE. Improving the environmental safety of ash from bioenergy production plants. 2018. Diss. 838. GOLMAEI, SEYEDMOHAMMAD. Novel treatment methods for green liquor dregs and enhancing circular economy in kraft pulp mills. 2018. Diss. 839. GERAMI TEHRANI, MOHAMMAD. Mechanical design guidelines of an electric vehicle powertrain. 2019. Diss. 840. MUSIIENKO, DENYS. Ni-Mn-Ga magnetic shape memory alloy for precise high-speed actuation in micro-magneto-mechanical systems. 2019. Diss. 841. BELIAEVA, TATIANA. Complementarity and contextualization of firm-level strategic orientations. 2019. Diss. 842. EFIMOV-SOINI, NIKOLAI. Ideation stage in computer-aided design. 2019. Diss. 843. BUZUKU, SHQIPE. Enhancement of decision-making in complex organizations: A systems engineering approach. 2019. Diss. 844. SHCHERBACHEVA, ANNA. Agent-based modelling for epidemiological applications. 2019. Diss. 845. YLIJOKI, OSSI. Big data - towards data-driven business. 2019. Diss. 846. KOISTINEN, KATARIINA. Actors in sustainability transitions. 2019. Diss. 847. GRADOV, DMITRY. Experimentally validated numerical modelling of reacting multiphase flows in stirred tank reactors. 2019. Diss. 848 KN OW LEDGE ECOSYSTEM FORM ATION : AN IN STITUTION AL AN D ORGAN ISATION AL PERSPECTIVE Argyro Alm panopoulou ISBN 978-952-335-354-1 ISBN 978-952-335-355-8 (PDF) ISSN-L 1456-4491 ISSN 1456-4491 Lappeenranta 2019