Regional Institutional Arenas for Social Innovation: A Mixed Methods Research

Abstract Social innovation is conceptualised as having two intimately related pillars: institutional innovation and locally embedded innovation, in the sense of social economy. Two main research questions were addressed: how political, institutional innovation is fostered and how does it influence social economy? A mixed methods research was conducted in the Mühlviertel NUTS3 region. Despite a framework of enhanced autonomy and institutional innovation for the main stakeholders, both macro and micro analysis illustrate a lack of intermediate space to: a) link the innovative agenda to high-state political agendas, and b) link institutional innovation to social economy.


Introduction
Whether we describe institutions as the roots of political science (Peters 2012), or more modestly, as a political dimension of renewed attention (Lecours 2005), policy studies can hardly give convincing interpretations without developing an institutional account (Hall and Taylor 1996). Initially, the article defines social innovation at the local level as resting on the two pillars of: (a) institutional innovation and (b) locally embedded innovation, in the sense of social economy (Moulaert and Nussbaumer 2005). The present article tries to investigate the main research question of how social innovation is fostered at the regional level. The article follows a two-stage analysis: the macro and micro dynamics of political, institutional innovation are explored. Accordingly, the relationship between institutional innovation and social economy is researched.
The present article makes a twofold contribution. On the one hand, it tries to illuminate the political, institutional struggles and processes for the development of institutional innovation and social economy. Parallel to that, it contributes to one of the major challenges of new institutionalism, namely the development of a self-contained apparatus to interpret institutional change. New institutionalist theories are susceptible to critique, due to the fact that they tend to refer to external shocks in order to explain institutional change (Przeworski 1991). This renders new institutionalism a theoretical lens for better interpreting institutional continuity and reproduction, rather than change and transformation (Peters 2012). By applying the sociological institutionalism literature on the research field, we try hence to synthesise a solid and selfcontained methodological tool to interpret institutional change and innovation.
In the first part of this article, a theoretical outline is presented. The concept of social innovation is clearly defined and an analytical framework is developed. The second part deals with the research design and the methodology of the case study. The NUTS3 Austrian region of M€ uhlviertel has gone through a long-term, successful, regional development process. Decentralised regions with a strong government activism are considered fruitful fields for social economy to flourish (Stephan, Uhlaner, and Stride 2015). Nonetheless, there have been few research projects that discuss the role of formal institutions in social economy (Dacin, Dacin, and Matear 2010;Estrin, Mickiewicz, and Stephan 2013), and ever fewer empirical contributions to this discussion. That seems as a major gap of the applied research, in which the present endeavour will try to make empirical contributions. In the third part, by developing a political narrative of the region, the institutional corpus and arenas are outlined. In the last part of our article, we are focussing on the micro-dynamics of institutional change, by facilitating an in-depth research of institutional agents. Eventually, for concepts that remain strongly related to the particular social contexts (Bacq and Janssen 2011), drawing generalisations from the case study would seem superficial. Considering the limitations of the present research thus, merely to shift the analytical attention to specific institutional processes and potentialities seems as a major precursor for future applied research.

Engaging to social innovation: a theoretical framework
Social innovation seems to be at the heart of the social economy literature (Defourny and Nyssens 2013). The two terms appear closely related, since structural features of social enterprises render them vehicles of social innovation (Gardin 2006;Brandsen and Pestoff 2009). For a discourse that might still be at an infancy stage (Dees and Battle Anderson 2006;Dorado 2006), with terms and notions that appear rather subjective (Somerville and McElwee 2011), the epistemological assumptions that guide the empirical research should be clearly identified. Thus, the article follows the identification of the concept of social innovation as was theorised by Moulaert and Nussbaumer: Social innovation at the local level rests on two pillars, institutional innovation (innovation in social relations, innovations in governance including empowerment dynamics) and innovation in the sense of the social economy-i.e. satisfaction of various needs in local communities … Yet both pillars are intimately related (Moulaert andNussbaumer 2005, 2071).
The relation of the two pillars seems empirically grounded, since the institutionalisation of social enterprises across Europe has contributed to the development of new public schemes and legal frameworks, which in their turn became channels for social innovation (Defourny and Nyssens 2013). This process of institutionalisation is illuminated by the 'Institutional Entrepreneurship' theoretical strand, which was initially introduced by DiMaggio (1988). The term institutional entrepreneurship refers to the practices of agents 'who have an interest in particular institutional arrangements and who leverage resources to create new institutions or to transform existing ones' (Maguire, Hardy, and Lawrence 2004, 657). Thus, social enterprises that become embedded to the institutional framework, trigger dynamics for institutional change and innovation. The present article focuses on the reversed causality of the relationship, namely how political, institutional innovation can trigger locally embedded innovation, in the sense of social economy.
Two main lines of argumentation describe the broader interaction between political institutions and social economy organisations. First, there is the institutional void perspective, which assumes that social economy motivation increases when we have less active governance, abundant social problems and a greater demand for social enterprises (Dacin, Dacin, and Matear 2010;Estrin, Mickiewicz, and Stephan 2013). On the contrary, the institutional support perspective implies that countries with more active governments will support and thus enhance social economy (Korosec and Berman 2006;Zahra and Wright 2011). Quantitative researches suggest that the institutional support perspective better describes the reality, since it is the active governments that help foster the generation of social enterprises (Stephan, Uhlaner, and Stride 2015), by 'crowding in' rather than 'crowding out' further financial support (Andreoni, Payne and Smith 2014;Heutel 2014). This 'crowding in' also concerns the institutional innovation capital, since local innovation policies seem to work only inside embedded national social-security systems (Moulaert and Nussbaumer 2005). Nonetheless, a precondition for that appears to be a minimum of shared political agendas between the various state levels and institutions. Otherwise, the institutional stakeholders often block each other's actions. Overall, the quality of collaboration between local and national authorities influences the capacity to face the basic needs (Moulaert and Nussbaumer 2005) and hence the dynamics of social innovation. This process is conceived as an unbreakable chain of innovation: Innovation in local governance can only work in concordance with innovations in national and global governance and social security (Moulaert 2000, 61).
Thus, the implementation of social innovation at the local level, appears to be strongly related to the involvement of support networks, including public institutions (Moulaert and Nussbaumer 2005). Potential seeds of institutional innovation hence depend on the wider opportunity structure (Tarrow 1994) in which they find themselves. Nonetheless, as urban social movement literature shows, local initiatives have difficulties in accumulating the sufficient power to dispute the dominant governance practices (Fainstein and Hirst 1995;Mayer 2000;Pickvance 2003). Thus, they depend heavily on their ability to struggle in the wider institutional arena, to challenge broader policies, practices and discourses (Gonz alez and Healey 2005). Innovation opportunities thus usually emerge as cracks to power conflicts and institutional collisions (Gonz alez and Healey 2005). Hajer (1995) following a neo-Gramscian approach, emphasises the need of diffusion of the new institutional logic to the broader institutional landscape to become hegemonic and dominant. These cracks thus need to expand and diffuse to the institutional space, in order for the institutional relations to work collectively towards the desirable ends.
The present article thus conceptualises the political framework not as a homogenous unity but as a nexus of inter-institutional interactions, with multiple boundaries and agendas (Jessop 1990). In other words, the political corpus is approached as an institutional arena that whereas often inflexible, remains open to institutional opportunities that could expand the space for change and innovative ventures. The first task of the article, is to outline the macro, institutional framework, in order to identify the dynamics and struggles between the different levels of authority and the potential cracks for institutional innovation. Applying institutional theory to social economy, can provide us with new insights for both social economy and institutional theory (Dacin, Dacin, and Matear 2010;Mair and Marti 2006).
Thus, in order to proceed, the article draws methodological tools from the 'New Institutionalism' literature, and more specifically from the 'Sociological Institutionalism' strand. The 'New Institutionalism' framework emerged during the 1980s as a counterreformation of the old institutionalism, adopting the assumptions of the latter, but further enriching them with research and theoretical tools that had informed both behaviouralism and rational choice theory (Peters 2012). James March and Johan Olsen who are considered the initial proponents of new institutionalism (March and Olsen 1989), shifted the analytical attention to the integration of institutions. 'Sociological Institutionalism', i.e. one of the strands of new institutionalism, arose out of organisation theory (Powell and DiMaggio 1991) and essentially enacted a conjugation of the normative with the cognitive perspectives, by seising the intellectual dividends that a partial eclecticism can bring (Peters 2012). Institutions under that scope are defined as entailing 'regulative', 'normative', and 'cultural-cognitive' pillars (Scott 2014). Thus, norms, values, culture and ideas all play a significant role in the conceptualisation of institutions (Dobbin 2006).
Nonetheless, despite different tendencies, most institutionalists have difficulties in explaining institutional change and innovation and repeatedly come back to the idea that external shocks are at the root of changes (Przeworski 1991). The inability to interpret change and transformation constitutes one of the main critical arguments against new institutionalism (Peters 2012). By placing so much emphasis on convergence, and isomorphic dynamics (DiMaggio and Powell 1983), autonomous decisionmaking has been undermined and removed from the picture (Dowding 1994). Thus, one of new institutionalism's main challenge is to develop an integrated and autarkic framework for explaining the change of institutions. The present article aspires to contribute to this endeavour.
An institutional account has thus to be self-contained, but still not explicitly endogenous as it might be initially assumed. The most important reason for that is that global dynamics have significantly blurred the boundaries of the environments (Harty 2005) by rendering the institutional setting multi-layered and highly complicated. Even the distinction between endogenous and exogenous seems to have little value in such a context. The example of the European Union is the most indicative, since a European institution is difficult to be categorised as endogenous or exogenous, in relation to a sub-state nation. Thus, the term of endogeneity that might be used is a rather moderate one, by taking into consideration the blurred boundaries of the distinction.
To trace the potentialities of endogenous institutional change, the article focuses on the macro and micro-dynamics of the institutions. One of the central assumptions of sociological institutionalism is that of the institutional isomorphism (Powell and DiMaggio 1983), namely a process by which an actor gradually incorporates existing institutional arrangements, structures and practices (Peters 2012). A progressive loss of the inner characteristics is a result of those pressures. The theory of isomorphism to some extent goes back to Weber and his theory that there will be a convergence towards a rational format of bureaucracy (Peters 2012). Thus, isomorphic pressures act as a counter effect and curb innovative initiatives (Defourny and Nyssens 2013). The quest for institutional innovation essentially becomes a quest against the institutional, isomorphic counter-dynamics. In other words, what is at stake is the willingness and ability of the stakeholders to generate institutional innovation, despite the isomorphic pressures.
This endeavour dwells on the issues of identity and place, as well as their interconnection (Healey 2002). The institutional capacities for innovation are engaged to a dialectic relationship between material practices and the meanings that the agents attach to their environment (Gonz alez and Healey 2005), i.e. what Richardson and Jensen (2003) have identified as cultural sociology of space. In other words, institutions are embedded in the political economy of a concrete time and space (Gonz alez Ceballos 2003). Their proximity and their level of embeddedness to their environment, thus influences their capacity to internalise environmental adjustments and by that their susceptibility to change and innovation. Swyngedouw (2005) argues that innovative political conditions emerge in a context of a widening and diverse gap between state and civil society. Advocates of this theory argue that change will be demanded when somehow, existing institutions become discredited or delegitimized (Blyth 2001;Cortell and Peterson 1999). This argument comes from the liberal democratic literature, which suggests that in order to be functional, governing institutions should be considered legitimate by a substantial majority of the population (Catt and Murphy 2003). For purposes of securing legitimacy thus, opportunities of change emerge (Dimaggio and Powell 1983), in cases leading highly embedded institutions to adopt innovation and change, in order to increase this legitimacy (Carvalho et al. 2017).
Environmental factors also enter the equation by shaping consciousness, identity and discourse of the members of the institutions. Planning theory connects specific geographical micro-practices to structural characteristics (Fainstein 2001;Gualini 2001;Healey 2004). More specifically, strong geographical and historical identities enact a sense of caring for the region and a strong willingness to promote the development and the benefits of the region (Gonz alez and Healey 2005). Additionally, shared solid geographical and historical identities among the members of the team, enhance cooperation and a common sense of belonging. Both elements foster self-esteem and social approval and hence an expanded space for experimentation and innovation (Gonz alez and Healey 2005).
The interaction with the societal context though is not exhausted in the institutional embeddedness but expands to the institutional relations and arenas of governance. An investigation of potential for transformation thus needs to examine actors and networks, stakeholders, arenas, discourses and frames of meaning, practices and routines (Gonz alez and Healey 2005), namely what the present article conceptualises as institutional mobility. Institutional mobility is identified as containing two strands: the external mobility of the institution as well as the internal mobility. The external mobility is investigated through the network of interactions, arenas, support and collision schemes of the broader institutional framework (Moulaert and Nussbaumer 2005). A high level of mobility and interaction between civil society and institutions might export policy debates (Montpetit 2005) and render the institutions open to a chain of small adaptations that in the end might amount to a substantial change to the original institution (Thelen 2003).
The internal mobility of the institutions dwells on the learning processes for the members of the institutions, which are considered principal means of adaptation to the newborn significations (Levinthal and March 1993). Institutions seem to adjust to their environment through processes of learning. Thus, changes of the environment constitute a set of opportunities for the institution, as well as a threat to its established pattern of behaviour (Peters 2012). This set of opportunities comes through a process of gradual legitimisation of the new practices (Tolbert and Zucker 1996). Even more directly, research suggests that education is positively related to social economy (Estrin, Mickiewicz, and Stephan 2013).
Most sociological institutionalists consider institutions as the embodiment of symbols, scripts and routine (Harty 2005), which guide actor's appropriate place and course of action. What renders change a very difficult task is exactly this internalisation of norms and thus the crystallized roles that every time members of institutions feel obliged to play. Due to the difficulty of addressing all the alternatives, to find the optimal one, actors stop analysing once they are satisfied with the knowledge that a solution serves their interest (Lindblom 1959). Thus, the range of solutions is significantly limited due to a logic of appropriateness and the tendency to limit the spectrum of alternatives. These solutions are usually the mainstream, routinised responses to the problems, belonging to the 'garbage can' of decision-making (Cohen, March and Olsen 1972). This tendency enacts a framework of homogeneity among the members of the institution, in terms of how they make sense of their collective worlds and engage cognitively and bodily in their day-to-day routines (Hajer 1995;Healey et al. 2003). This homogeneity is what the present article conceptualises as institutional cohesion.
When it comes to institutional innovation, the level of institutional cohesion, is one of the most substantial but also controversial issues. The ambiguity emerges due to two contradictory tendencies that cohesion raises: in the first aspect, a high sense of cohesion enhances the logic of appropriateness and further limits the choices of the members (Clemens and Cook 1999), whereas on the contrary, cohesion raises social approval between the members. The later constitutes a universal preference of human beings and is expressed as status, esteem, respect and honour (Smith 2010). Thus, social approval and trust between the members can trigger deviation from the routine, in terms of experimentation and innovation since a malperformance is more easily forgiven, when the person who commits it is being trusted (Offe 1999). The level of trust, hence directly influences the level of experimentation and innovation (Offe 1999).
Finally, the research has to engage to the micro-practices of the members of the institutions and their receptiveness to innovation. Sociological institutionalism seems more capable of interpreting those dynamics, due to the cognitive expansions that it introduced to the normativism of new institutionalism. More specifically, advocates of sociological institutionalism espouse a principle of bounded rationality for the members of the institutions (Powell and DiMaggio 1991). By that, they don't assume fully rational, fully informed agents, but rather rational agents that proceed to subjective, cost-benefit evaluations. At the micro-level, thus the research has to focus on resources and the ability of actors to deploy these, in order to implement their agendas for change (Peters 2012). One major aspect of this process is for actors to realise that the consequences of alternatives are promoting their own self-interest.

Research design and methodology
The M€ uhlviertel is an Austrian NUTS3 region, which belongs to the state of Upper Austria. The Upper Austria state is divided in four parts, one of them being the M€ uhlviertel. The others are Hausruckviertel, Traunviertel and Innviertel. M€ uhlviertel is a prosperous statistical region that is constantly growing in economic terms. The GDP adjusted by current market prices constantly rises from 2007 at 3.896 to 5.279 million euros at 2015, i.e. a growth of 74% during an eight years period. At the same time the population remains moderately stable, 204 thousand people at 2007, 206 thousand people at 2015. M€ uhlviertel is a primarily rural region with a percentage around 4,5% share of the Gross Value added by the primary sector, in relation to the Austrian average, which is significantly lower at 1,4%. The good economic situation of the region is also reflected to the unemployment levels, since in the three out of four M€ uhlviertel districts, unemployment is significantly low (Rohrbach 3%, Freistadt 4,2%, Perg 4,1%) with only Urfhar Umgebung (7,9%) converging to the Austrian average of 8,5% 1 . Overall, thus we are talking about a prosperous rural region. What field observations and interviews also reveal is that M€ uhlviertel has gone through a steady regional development via decentralised bottom-up processes. An implication of this success is the fact that the region of M€ uhlviertel contains six LEADER sub-regions: Perg Strudengau, M€ uhlvierter Alm, M€ uhlvierter Kernland, Sterngartl-Gusental, Donau-B€ ohmerwald, Urfahr-Umgebung 2 . A successful regional development, reflected to a strong LEADER network makes the existence of six LEADER sub-regions possible. The number and the small size of the LEADER regions indicate the embeddedness to the local level. It also indicates a strong bottom-up perspective and a high level of decentralisation, with subnational states responsible for 34% of public expenditure, ranking Austria the 14th most decentralised OECD country (OECD 2016).
Field observations during a two month internship revealed the need for an integrated mixed method plan. There were two fundamental reasons for that: a. By researching the dynamics of change we are not interested only on the 'what, where, when and who' but also for the 'why and how' of the processes. In other words, merely to trace potential causalities between variables wound not be enough, especially when investigating such multifaceted phenomena. What we are interested in thus, is to explain how and why these factors are related to one another. b. The assessments of the members cannot be adequately described only through a qualitative evaluation. A qualitative assessment of the extroversion of the institution for example, can range from a positive to a negative statement. On the contrary, having a numerical value could give a more concrete description of the assessment. c. On the other hand, for those parameters that we explore we cannot have strictly numerical data. For the quantitative dimension thus, we essentially elaborated quantitated qualitative data that tried to give numerical values to assessments. Thus, the conjugation of a quantitative and a qualitative dimension has been a structural feature of the research, from the very beginning.
The research of the present article followed a sequential exploratory design as summarised by Creswell (2014), where the qualitative data is collected first, followed by the collection and analysis of quantitative data. The purpose of this design is to initially develop a theoretical background upon which a research instrument (a survey in our research) is then applied, to describe specific variables. In the first phase, the research presupposed a triangulation of qualitative data. The institutional corpus was systematically compiled with resource to three sources: field observations, documents, and fifteen semi-structured interviews that were conducted in the research region with political experts that engage to social innovation and social economy from different perspectives.
The list of interviewees was generated, by following a combination of 'key informant sampling' with 'snowball sampling'. The key informant sampling was facilitated in order to conduct in-depth interviews of a selected (non-random) group of social innovation experts (Lavrakas 2008). Nonetheless, since the expertise on social innovation is not a robust, clearly defined quality, the subjects of the research guided the research by identifying their mutual relationships (Lavrakas 2008) among the population of the experts, following a snowball process. Eventually, the experience of the subjects were incorporated to the sampling process and were triangulated with field observations and documents. Is the list of interviewees exhaustive of the 'social innovation expertise' quality on the region? Most probably not, and it would be utopic to request something like that; still it seems exhaustive of the types of approaches and perspectives of the political experts of M€ uhlviertel to the social innovation framework.
Eventually, six LAGs managers, the manager and a member of the regional development office of the NUTS3 M€ uhlviertel, two politicians that are now elected members of the Upper State parliament, the representative of ELARD and president of the LEADER forum, an ex-mayor and pioneer of regional development, the manager of an organisation-facilitator of social innovation and two social entrepreneurs. All the interviews were anonymized for reasons of political sensibility. The questions of the interviews could be categorised as belonging to three main categories: regional development challenges, where questions about the main problems of the region were asked, as well as projects that were facilitated from the interviewee's institution to address those problems. Secondly, the interaction of the institution/organisation of the interviewee with other political institutions was explored. By asking them about specific connections to specific institutions a network of connections was outlined. Finally, by investigating discourse familiarity, perceptions and practices, innovative interventions and the interaction with the social economy sector were researched. The method that was used to facilitate the data was Grounded Theory. An integrated political narrative was developed in order to trace the institutional interactions and collisions, as well as to focus our analytical attention to the main institutional stakeholders that facilitate social innovation.
Once those agents were identified, a survey was developed drawing from the sociological institutionalism literature, to explore the micro-dynamics and the factors that can generate institutional change and innovation. More specifically, the research has been conducted through a questionnaire to the members of the institutions/facilitators under investigation. There were thirty, closed-ended linear scale questions from one to ten. The scale was chosen following previous surveys rom the European Network for Rural Development (ENRD), targeting the same subjects, for purposes of comparability. The questions investigated attitudes, practices, perceptions and assessments of the members of the institutions in reference to three main categories: institutional mobility and embeddedness, institutional cohesion, and receptiveness to social innovation and social economy 3 . Twenty-eight questionnaires were distributed, fourteen of which were answered, maintaining a response rate of fifty percent. The method that was used to elaborate the data was Descriptive Statistics. Eventually, institutional change was described via a joint-display and an integrated political narrative that was developed (Creswell 2014).
A macro-analysis: outlining the institutional arena Field observations of the M€ uhlviertel region, as well as qualitative data highlight the role of the Local Action Groups (LAGs) in the regional development process and more particularly in the fields of social innovation and social economy. Contrary to many European states, the managing authorities of Austrian LAGs are public organisations. While working with social enterprises of the field, talking to locals and policy makers of all levels, the terms social innovation and social economy came usually with reference to the LAGs and the LEADER processes. Indicatively, managers of LAGs have often been found involved to social enterprises (Chatzichristos 2018k). Relating social economy to the regional development policies, Austrian experts evaluate the European Social Fund (ESF) and the LEADER measures as the only European initiatives that have 'moderately positive impact' to the social economy sector (EESC 2016). The importance of the LEADER policies for the Austrian social economy seems to confirm the field observations. LAGs appear as those development agencies that facilitate social economy on the Austrian local level.
The analysis of the institutional arenas thus commences with how regional development agents of LAGs perceive of other political institutions and whether and how mismatches and institutional struggles develop. It might be reasonably objected, that by directly engaging to the role of the LAGs and considering them as the atomic unit of the research, the analysis ignores lower, municipal and communal initiatives. The response to that comes from the literature on social movements, which stresses the difficulties of small-scale local initiatives to accumulate sufficient power in order to manipulate the environment in which they find themselves incorporated (Fainstein and Hirst 1995;Mayer 2000;Pickvance 2003). Thus, such initiatives need to relate to national-scale alliances or to attach themselves to broader social movements and formal government processes (Fainstein and Hirst 1995). Local initiatives do need to accumulate power from a solid, concretely defined but still low-level political institution, which in the case of M€ uhlviertel region seems to be the LAG. For the M€ uhlviertel context thus, LAGs act as those embedded mediators that unify and facilitate the processes of a socially innovative regional development for municipalities and communities. Qualitative data display constant and productive relationships between LAGs and municipalities, with LAGs often being generated bottom-up, as local initiatives (Chatzichristos 2018k).
A further reason of the critical engagement to the LAG network is a high controversy in interviewing politicians, implied from a political identity that seems to heavily influence their responses and their general stance towards the researched topics. This became quite apparent during interviews with politicians (Chatzichristos 2018a;Chatzichristos 2018c). Interviewees thus should be as least biased as possible and this parameter might exclude politicians a priori. At least when the biases themselves are not subjected to research. The members of LAGs appear as those least prejudiced agents between European, National and local levels as well as intermediaries between the civil society and the politicians. Adequately distanced from the political debates and firmly grounded to the needs of the civil society, the M€ uhlviertel LAGs appear as the smallest unifying institution that facilitate social innovation and social economy.
Regarding thus LAGs as the primary institutional agents, the institutional struggles with the higher political levels seems overall of a low intensity. Drawing from field observations and qualitative data, LAGs seem to enjoy a satisfactory cooperation with the higher political levels, by constantly highlighting their autonomy and their entrenched field of action. They often have a 'high respect for the work of the politicians' of the higher political levels (Chatzichristos 2018b) and a 'fairly good cooperation with the Upper Austria state', due to a prestigious status of the sub-region (Chatzichristos 2018k) or even the support that they receive (Chatzichristos 2018f). Even more importantly, they often have common perceptions that vertically penetrate the political spectre, from the Federal state to Upper Austria and to the LAGs, specifically in 'the connection of tourism with agriculture' 4 (Chatzichristos 2018k; Chatzichristos 2018f). Usually, the connection with the Federal state remains only on the level of perceptions, since the autonomy of the municipalities is a principle of the Austrian constitution (Chatzichristos 2018a). Thus, references to struggles with the national institutions usually abstain from the interviews.
In many cases, there are only rare references even to interaction with political institutions of the Upper Austria state (Chatzichristos 2018b;Chatzichristos 2018e;Chatzichristos 2018j). This might have to do with the fact that 'other institutions do not intervene to the basic working plan' of the LAG (Chatzichristos 2018b), or that LAGs have developed mechanisms to absorb the pressures, such as having 'a mayor as a political mediator for keeping the struggles away from the fieldwork' (Chatzichristos 2018j). Even for those cases, that bureaucratic interaction raises political conflicts, a funding autonomy of the LAGS entails that they hardly have anything to do with the higher institutions (Chatzichristos 2018f). Overall, there is a broader sense that the locally based, regional development should be disconnected from the politicians and public funding in order to be autonomous (Chatzichristos 2018k).
Most importantly, mismatches with Upper Austria and Federal institutions do not work against the LAGs' guidelines (Chatzichristos 2018d) and remain theoretical and away from the regional field, especially since a national networking, manifested to a LEADER Forum, has managed to further 'empower the voice of the LAGs on a national scale' (Chatzichristos 2018i). The LEADER forum meets with the Upper Austria government every two months and they have a fairly good cooperation, with sporadic divergences only on the levels of funding. Nonetheless, even these debates are discussed 'in a positive way' (Chatzichristos 2018h). Eventually, despite their divergences, LAGs in general have managed to sustain a high level of autonomy from higher political levels.
This is not a one-sided interpretation since the higher institutions of the Upper Austria state seem to have a common perception about the interaction with the LAGs. The NUTS3 Regional Development Office (RDO) is one hundred percent publicly funded institutional agent and is considered 'the extended arm of the country Upper Austria' (Chatzichristos 2018m). The RDO members conceive of the LAGs as complementary of their work. The NUTS3 and the LAGs are conceived as belonging to a 'parallel structure' (Chatzichristos 2018m). This also has to do with the fact that the RDO of M€ uhlviertel is assigned to the ministry of economics, whereas LEADER regions to the ministry of agriculture. The structure is described as very decentralised, and this is something good because M€ uhlviertel is 'just too big' and thus LAGs 'bring a lot' (Chatzichristos 2018m;Chatzichristos 2018b).
The clear-cut division between the two levels is also reflected to the topics of the RDO, which remain away from the locally driven regional development. For the RDO there are two main regional themes: mobility challenges and intraregional disparities. The focus of the RDO is primarily on infrastructure in order to address disparities between rural and urban areas, linking regional development with issues of mobility (Chatzichristos 2018b). The existence of six LEADER regions inside M€ uhlviertel might be interpreted as a further indication of regional disparities, due to the 'imaginary border … that divides east and west' (Chatzichristos 2018m). Nonetheless, there is a high level of regional cohesion and identity, since 'the M€ uhlviertel knows where the M€ uhlviertel is' (Chatzichristos 2018m).
Further divergences were identified on the level of perceptions, since regions are described from the Upper Austria stakeholders again in terms of infrastructure. Indicatively, the economic success of Rorbach is described as an implication of the facts that 'enterprises were settled' and 'children's hospitals were developed'. At the same time challenges are identified in the level of wages, i.e. 'the disadvantage that I do not earn so much', the shortage of 'skilled workers', as well as the existence of a weak 'broadband internet' (Chatzichristos 2018m). This is often interpreted as an implication of the fact that RDO facilitates bigger projects, whereas LEADER regions bottom up initiatives (Chatzichristos 2018f). Alternatively, as the RDO manager puts it, RDO is responsible for 'the care of the communities and not for the citizens' (Chatzichristos 2018m).
Specifically for social innovation, the interviewees of the RDO have difficulties to express what it means for them, or whether the region is socially innovative (Chatzichristos 2018b;Chatzichristos 2018m). Innovation is linked with innovative working practices such as a 'four-day week'. At the same time, the region is described as highly social, with 'a dense social network' (Chatzichristos 2018m). Whereas the two notions of 'social' and 'innovation' are described, the link between them is missing. This seems to have more than a symbolic value. By defining innovation as strictly attached to the field of infrastructure, the connection to the social dimension remains in question. The institutions of the higher political levels seem to remain away from the social innovation topics, practices and discourse. Sometimes this distance is interpreted as an explicit top down approach of the Upper Austria state compared to the bottom-up LEADER processes (Chatzichristos 2018h).
These distinguishable perceptions, identified as a dichotomy between a promotion of the welfare of the communities to that of the citizens, appear to entrench the field of action for LAGs. Nonetheless, despite the different agendas, mismatches seem not to end up in substantial institutional struggles that would influence or block each other's actions (Moulaert and Nussbaumer 2005). Whereas autonomy is usually considered as the 'categorical imperative' of the local stakeholders, having a week institutional arena might imply an inability of innovative initiatives to disseminate and travel across the institutional space. Whether LAGs can independently accumulate sufficient power to promote social innovation, or the concordance with higher state innovations remains a precondition (Moulaert and Delatetsima 2000), had to be tested on the micro-environment of the organisations.
A micro-analysis: tracing the micro-dynamics of change By exploring the micro-dynamics of LAGs, spatial and regional characteristics seem at cases cardinal for the development of the M€ uhlviertel region, as well as of its challenges. Commencing with the proximity to the metropolitan centre of Linz, Urfhar West is described as the 'bacon belt of Linz' (Chatzichristos 2018f) and as such, it has been drifted to an intense regional development process (Chatzichristos 2018a). The linkages to the urban environment have shifted the attention from heavy agriculture projects to a more civic approach (Chatzichristos 2018f), reflected to the explicit focus of some LAGs on leisure and recreation tourism. On the other hand, this urban proximity has enacted negative implications that have to do with Labour and educational migration (Chatzichristos 2018f), as well as the sharp increase of land prices (Chatzichristos 2018a). An analogous situation seem to be at stake for Perg Strudengau, where a successful regional endeavour, came with the 'migration of youth' challenge (Chatzichristos 2018e). What seems to differentiate Perg Strudengau though, is the big size of the region, which renders the region susceptible to intraregional disparities. Thus, for Perg Strudengau as an entity, tourism is not a unifying theme, but rather a theme of further potential 'intraregional divisions' (Chatzichristos 2018e).
A similar situation seems to be at hand for Stengartl Gusental, where disparities between rural and urban areas enact a regional heterogeneity (Chatzichristos 2018j). Spatial characteristics appear also catalysts of regional development for Donau-Bohmerwald. Whereas the region is not so close to Linz, its development has grounded on its position 'in the crossroads between three rich metropolitan centres, Munich, Vienna and Prague' (Chatzichristos 2018b).This resulted to an exogenous development with good working conditions, high education and a growing population (Chatzichristos 2018b). Constant references to an international network, and crossborder cooperation confirms this interpretation. Overall, the geographical characteristics of the LEADER regions seem to have played a substantial role to their development.
Nonetheless, if geographical conditions would be adequate to fully explain success or failure of regional development, then we would reasonably expect a rather weakened process for the region of M€ uhlviertler Alm, namely a region far away from metropolitan centres, 'with poor municipalities, with a weak economy and less income' (Chatzichristos 2018g). On the contrary, M€ uhlviertler Alm has been a role model for regional development during the last thirty years (Chatzichristos 2018g), 'a constant point of reference' (Chatzichristos 2018k), as well as a 'positive influence for other regions' (Chatzichristos 2018f;Chatzichristos 2018j). During the interviews with stakeholders from M€ uhlviertler Alm there were rare references to external factors. Something rather more endogenous seems to have been facilitated, which is of major interest for the present research, as well as for the broader institutionalism literature.

Institutional embeddedness and mobility
A gradual loss of institutional legitimacy seems to have been the catalyst for the political initiative that started twenty-five years ago in the M€ uhlviertler Alm (Chatzichristos 2018k). The regional development process for M€ uhlviertler Alm has started, as a political initiative and has been the 'reply to a hitherto failed structure' (Chatzichristos 2018g). More specifically, 'a population drop was an awakening and an initial motivation to take initiatives' (Chatzichristos 2018g). Bounded by a significant lack of resources and regional disadvantages the initiative had to come via processes of enhancing the embeddedness of the political institutions to the civic society. Thus, an initial venture has started as a cooperation between the mayors of the region, in order to develop a trust building apparatus and a social network (Chatzichristos 2018g). The main axis around which regional development was activated was getting the people to work together and getting the municipalities to work together, reflected to the emblematic quote of the region 'not against each other, not beside each other, but only with each other' (Chatzichristos 2018g).
Mobilisation of the institution has been a vital precondition in order the region to trigger institutional change and empower its voice and autonomy. When the cooperation venture started, an underpinning culture of cooperation was already there but remained non-institutionalized. In virtue, of the institutionalisation of this cultural element, an extended bottom-up process was facilitated, where people were largely involved, 'by participating in policy teams and constantly expressing their needs' (Chatzichristos 2018g). Such small adaptations have led to substantial projects for the region of M€ uhlviertler Alm, such as the 'Johannesweg', namely a touristic hiking path across the region. Furthermore, a substantial change of status came with the LEADER region status in 1995 as a compensation for a long-term process of regional development, which was described as very intense (Chatzichristos 2018g). After the region gained its LEADER status there was a project boom and a strong regional development period. Eventually, the M€ uhlviertler Alm gradually developed a successful apparatus by merging previous associations and fragmented initiatives into a coherent network and by involving civic society in the process (Chatzichristos 2018l). In other words, an institutionalisation of this culture of cooperation changed the civic society, as well as the institutions per se.
The institutional mobilisation was also manifested to projects that had a pedagogical dimension. Such an educative initiative that travelled across the region of M€ uhlviertel was organic farming. Organic farming has been a major axis around which regional development programmes have been developed across M€ uhlviertel (Chatzichristos 2018f;Chatzichristos 2018j;Chatzichristos 2018b;Chatzichristos 2018k). Organic farming though, has not been approached merely as an alternative, advanced way of farming, but rather as a broader communal alternative and a mean of further integration. Farmers have been heavily involved in the communal cooperation and the regional development processes (Chatzichristos 2018g). &Omicron;rganic farming on the LAGs level is conceived as a way to empower small farming structures in order to serve 'a sustainable development goal' (Chatzichristos 2018j), as a 'mean to promote a more qualitative lifestyle through the creation of leisure time' (Chatzichristos 2018b), or even as a 'qualitative tool via which people disseminate their work' (Chatzichristos 2018f). In any case, it is described more as a mean rather than an end to the regional process. This mean has been a carrier of the cultural elements that M€ uhlviertler Alm introduced and contributed to the embeddedness of perceptions across the institutions.
The culture of cooperation between the local stakeholders, becomes also apparent from our quantitative data, with members of the institutions evaluating the cooperation between the institution and the civic society as well as between the institution and local partners as high (80% and 79% respectively). The themes of networking and citizens' involvement are common between the interviews with the managers of the LAGs. To address regional challenges the LAGs try to inform and involve people in the decision making process (Chatzichristos 2018d). Thus, bringing the stakeholders together and generating bottom-up processes are mains responsibilities of the LAGs Chatzichristos 2018f). Thus, a broader culture of regional cooperation has been cultivated (Chatzichristos 2018b), enriched by cross-regional partnerships (Chatzichristos 2018e) and cross-border interrelations, either through international sports events, such as the World Cup of rowing (Chatzichristos 2018f), or INTEREG programmes (Chatzichristos 2018j). A culture of institutional mobility and extroversion that even surpasses the regional and international borders is reflected to the assessment of the extroversion of the regions by the members of the institutions as rather high (81%).
The high level of embeddedness of the institutions to the civic society is finally closely related to the high identification of the members of the institutions with the region that they are located in (91%). The members of the institutions seem to have a high appreciation of their regional identity and they identify their interests with the interests of the region. This high appreciation of the regional identity emerges as a common element from the interviews, as that which 'brings us as a region on' (Chatzichristos 2018f), as the need to 'create common perceptions' (Chatzichristos 2018b), or even as the major theme, that due to interregional disparities has to be enhanced (Chatzichristos 2018j). Eventually its significance could be justified in comparison to the identification with the national identity, which is considerably lower (77%)

Institutional cohesion
The level of cohesion between the members of the LAGs is rather high. It is quite noticeable that although not always serving for a long time (46%), they share the ethical codes and values of the institution in high numbers (87%) and assess their cooperation with and the trust that they enjoy from other members similarly high (85% and 82% respectively). Additionally, their responsibilities seem to be very concretely defined (86%) and they have a strong feeling of creativity (83%). The above successful framework is combined with a moderate sense of routinisation (71%). This seems to be related with a 'time-consuming bureaucracy' that appears common across the M€ uhlviertel region (Chatzichristos 2018f;Chatzichristos 2018j) and is related to the LEADER mainstreaming period, which imposed high bureaucracy and mutated LEADER to a simple funding structure (Chatzichristos 2018i). This led to a restriction of projects that could be otherwise actualised, such as more INTEREG networking (Chatzichristos 2018j). Maybe even more important, it leads to a severe restriction of changes of roles and responsibilities as well as of practices (25% and 37% respectively). This seems to entail a practice-conservativism and a level of institutional stagnation reflected to the internal processes of training and evaluation of the members that are significantly low (52% and 61% respectively).
The mixed data of our research thus describe an institutional cohesion that doesn't seem to generate experimentation. Our results seem to follow up to the results of the extensive survey that was conducted by the European Network for Rural Development for all the European LAGs. There the Austrian LAGs appear highly receptive to innovation (37-29% EU average) and to finding innovative solutions (55% to 33%). Nonetheless, the favour the status quo in big proportions (46% to 20%) and appear sceptical on the positive effects of changes, such as increasing funding (29% to 46%). Eventually, what seems to be at stake is a controversy between sustaining an efficient apparatus and promoting an innovative orientation that might require sacrificing some of the successful blueprints of the past.
The very same controversy emerged during our semi-structured interviews. Thus, whereas the interviewees have reflected on their work rather passionately, at times, this has only been an epiphenomenon and aspects of a stagnation emerged indirectly. Indicatively, regional managers when asked about specific projects needed time and assistance to recall some of the important ones (Chatzichristos 2018e), there was a superficial reflection on the challenges of the region, by claiming that there is hard to find any challenges (Chatzichristos 2018b), or even an insistence to future challenges that are weakly (if at all) defined (Chatzichristos 2018d). These occurrences reveal indications of a stagnation that could be also linked to the moderate sense of routinisation.
Tracing the causes of the routinisation, bureaucracy appears again as one of the important factors, but not the only one. Often the success of the LAGs in facilitating multiple projects absorbs a lot of time and capital (Chatzichristos 2018f;Chatzichristos 2018j). In cases, even a premature absorption of funds (Chatzichristos 2018e) led to indications of a stagnation. Furthermore, the routinisation seems at cases to have come indirectly, as for example the outcome of a reposition of the LAG to the edge of the region to empower peripheral areas within the region (Chatzichristos 2018e). Eventually, in one way or another, successful regional development practices seem often to raise negative externalities to the activation and creativity of institutional members. In order to interpret this phenomenon we might have to look again to the most experienced LEADER region, to trace previous occurrences.
The experience of M€ uhlviertler Alm as the longest endeavour of regional development thus seems again to be very informative. According to one of the pioneers of regional development for the region, after the LEADER status of the region there was a project boom during the 1995-2002 period (Chatzichristos 2018g). The quote that 'the main topic was as many projects as possible' (Chatzichristos 2018g) is indicative of the developmental approach that prevailed during the period. Nonetheless, whereas this led to an enhanced regional development, it also absorbed a big part of the social capital and the spirit of cooperation seems to got lost at around the new millennium (Chatzichristos 2018g). Many people were participating in the projects, 'the egoism rise and the social contacts started missing' (Chatzichristos 2018g). A saturation soon came and everyone was dedicated in completing his small share of the process and the whole picture was lost (Chatzichristos 2018g). A steady successful development of the region seems to have come through a maturity period.
Overall, the M€ uhlviertel LAGs enjoy a high level of institutional cohesion. Thus, the dynamics and opportunities for institutional innovation remain rather strong. Nonetheless, indications of a growing routinisation seem to imply that the region appears in the edge of a maturity period similar to the one that the M€ uhlviertler Alm came through in the beginning of the twenty-first century. A long-term successful regional development has absorbed a part of the institutional innovation capital and has attached the LAGs to the rather successful status quo. What have been the impacts of this attachment, to the receptiveness of stakeholders to social innovation and social economy will be further analysed.

Receptiveness to social innovation and social economy
The analysis of discourse struggles has informed the attempts to apply the Gramscian analysis to how policy discourses become hegemonic (Fischler 2000;Jensen and Richardson 2000). The embeddedness of the emergent discourse around social economy and social innovation seems substantial in challenging the dominant forms of governance. Researching the familiarity of the institutional members with the emergent discourse and more specifically with the terms social innovation and social economy, the results were moderately high (79% and 76% respectively). Nonetheless, during the interviews, giving a definition for the term social innovation has been rather difficult for almost every interviewee. This is something reasonable since the term is very controversial even in a theoretical level. Nonetheless, a clear definition was not the main aim of the question. The primary intention was to trace the focus of the perception of the interviewees around the relative terminology. Thus, social innovation was related to external changes (Chatzichristos 2018b), alternative ways of tackling social problems (Chatzichristos 2018j), or more narrowly to specific projects of integration (Chatzichristos 2018f). A second dimension has been the communal aspect, as bringing people together (Chatzichristos 2018e), empowering communal life to face future challenges, or even empowering interregional cooperation to bring new practices to the region (Chatzichristos 2018k). Overall, we could say that the term was linked to a communal cooperation to tackle social problems.
The general feeling about social economy is moderately high (78%) and so are the assessments about the social economy potentialities of the region (71%). Additionally, the assessment of benefits that the development of the social economy sector might bring to the region are similarly high (77%). Nonetheless, all these assessments appear roughly grounded. This is reflected to the fact that the conditions of the social economy sector, as well as the social economy policies in the region are evaluated lower (67% and 63% respectively). Additionally, there seem to be inadequate educative initiatives around the field of social economy (60%) which raises concerns about the long-term sustainability of the endeavour. Finally, there is a wide perception that LAGs facilitate the important projects, whereas social enterprises merely act as trust mediators (Chatzichristos 2018d), or providers of open spaces (Chatzichristos 2018b).
Thus, whereas there is a broad positive evaluation of social economy, there is an identifiable divergence between the theoretical assessment and the assessment of the present situation. This could be interpreted as a critical stance towards the present situation and a quest for more actions. However, this is not the case, since no such arguments were raised during the interviews. On the contrary, members of the LAGs by considering their institutions as having a strong socially innovative perspective (81%), appear to consider the role of social enterprises as secondary (Chatzichristos 2018f;Chatzichristos 2018e;Chatzichristos 2018j). Eventually, the strong LAG network seems to perceive itself as the main agent of addressing communal challenges, as well as the main socially innovative agent. Furthermore, future potentialities for social economy policies remain unpromising, due to the rather moderate educational initiatives (60%), as well as a meso-term horizon of the policies (69%). The micro-dynamics of institutional change confirm the independence and the entrenched field of action of LAGs, in respect not only to the higher political institutions, but further now to the lower social economy organisations.

Discussion: a macro-micro fusion
The present article investigated the dynamics of political, institutional innovation at the regional level, as well as the influence that they have to locally embedded innovation, in the sense of social economy. The political institutional arena of the M€ uhlviertel region was used as a case example to apply a sociological institutionalism methodological tool. On the macro level, the political framework was approached as a network of institutional struggles and power relationships, to illuminate the institutional arenas upon which these struggles unravel. Field observations, and qualitative data, shifted the analytical attention to the LAGs and the network of interactions that is developed around them. An in-depth investigation of the institutional struggles of LAGs with the higher political levels, indicated an institutional arena of low intensity, with LAGs having entrenched their own socio-political field of action. Policies, practices and discourses of the higher political levels remain away from the social innovation and social economy agendas. LAGs constitute the main stakeholders that facilitate social innovation at the local level.
On the micro-level a mixed method analysis indicated a framework of strong institutional innovation, which nonetheless presents signs of saturation and stagnation. More importantly, the micro-research enhanced the argument of the empowered, local stakeholders. However, it also indicated a further delimitation of the practices and agendas of LAGs and a thin link with social economy. Even for the locally embedded institutions of LAGs, the institutional innovation that they generate is not channelled to lower levels. The case research suggests that institutional innovation is not a sufficient condition for social economy. On the contrary, the mixed data displayed cases where the relationship between the two might even be antagonistic. This might seem as a strong argument that requires further research, which remains outside the limited boundaries of this article.
What remains important for now, is that both macro and micro analysis illustrate on the one hand a severe lack of intermediate space to link an innovative agenda with higher-state political agendas, and on the other hand an inability to link institutional innovation to locally embedded innovation, in the sense of social economy. These empirical findings seem to enhance a line of thinking, which stresses that inside the European context, social economy lacks an intermediate space between the market, the state and civil society (Defourny and Nyssens 2013). Thus, by denying social enterprises the fundamental characteristic of belonging to that intermediate space, social enterprises' field of action is incorporated either to the public or to the private spheres (Defourny and Nyssens 2013). Specifically, in our case study, social economy's field of action is in cases embodied in the LAGs' agenda and in other cases severely undermined. 'Why is it problematic to have a low-level public agent that generates institutional innovation independently?' a state welfarist might ask. The answer is that it is not; at least in the short-term. What is actually problematic is the the long-term sustainability of the endeavour, due to the inability to relate to national-scale alliances or to attach itself to broader social movements (Fainstein and Hirst 1995). What the case example of M€ uhlviertler Alm suggests, is that this long-term sustainability is in question indeed, since the quote of 'not against each other, not beside each other, but only with each other' (Chatzichristos 2018g) seems to have an institutional appeal as well.
Notes 1. All data from Eurostat, https://ec.europa.eu/eurostat/web/regions/data/database 2. See https://www.leader.at/ueberblick.html 3. For more details see the full questionnaire and the results in Appendix 1. 4. On the 1/2018 the 'Federal Ministry of Agriculture, Forestry, Environment and Water Management (BMLFUW)' was renamed to the 'Federal Ministry for Sustainability and Tourism'