Understanding the strategic ‘black hole’ in regional innovation coalitions: reflections from the Twente region, eastern Netherlands

ABSTRACT Active coalitions of regional stakeholders are at the heart of contemporary regional economic development policies, such as Smart Specialisation or Constructing Regional Advantage. These coalitions consist of actors from various organizations such as regional authorities, companies and higher education institutions that come together to achieve common agendas and advance their region. Accordingly, the numerous stakeholders are expected to work together seamlessly, build and implement strategies and thereby deliver regional development. However, by assuming that strategy formulation and implementation is straightforward, the challenges that lie within partnerships and the tensions that may arise between stakeholders can become neglected. Therefore, it is vital to understand tensions that drive towards situations in which strategy-building is not successful and ‘black holes’ of strategy-building emerge. By identifying the tensions between regional partners in the Twente region of the Netherlands, the aim is to understand how such stakeholder tensions affect regional development. It is assumed that by easing or resolving these tensions, stakeholder partnerships can contribute to the successful advancement of their region. The data for this qualitative case study are drawn from both research interviews and secondary sources.


INTRODUCTION
There is a common contemporary understanding that regional innovation policies and strategies, systematic and goal-oriented activities in a regional environment are developed and executed by a set of key regional stakeholders (OECD, 2010;Sotarauta & Beer, 2017). In this context of emergent strategy making through a bottom-up approach, the term 'regional innovation coalition' (RIC) has been introduced, describing broad-based coalitions of stakeholders from heterogeneous organizations such as regional authorities, firms and universities who work together on the basis of shared common interests (Benneworth, 2007;Lester & Sotarauta, 2007). Diverse terms, such as regional development coalitions (Thorkildsen, Kaulio, & Ekman, 2015), inter-institutional partnerships (Silva, Teles, & Rosa Pires, 2016) and multilevel partnerships (Morgan & Nauwelaers, 2003), describe broadly similar ideas. In an ideal world, such stakeholder groups agree on a long-term vision for their region which involves various short-term exercises (Nieth & Benneworth, 2018). Accordingly, within current regional policy ideasthe most relevant example being Smart Specialisationit is intuitively presupposed that stakeholders within RICs work together seamlessly, build and implement strategies and as a result promote regional development.
Nevertheless, assuming that strategy-building within RICs is straightforward disregards the reality that stakeholders have varying, often even competing, regional visions and that the balancing act between shared/collective and private/individual interests can be remarkably challenging. The various classes of actorsranging from institutions as diverse as companies and government entities to a variety of higher education institutions and research and development (R&D) laboratoriesthat interact within regional economies are complex and compete according to specific criteria within their own 'markets'; as a result, coordinating activities between them is extremely difficult (Lagendijk & Oinas, 2005). Just as the region is complex due to the aggregation of diverse actors, the actors themselves can be multifaceted. Therefore, the actors within a coalition are typically overwhelmed when faced with the need to formulate and implement unified regional strategies.
Factors such as the various priorities and interests of stakeholders can impair the effective and joint development of strategies. As stakeholders try to fulfil their more urgent demands (simply put: a company 'wants' to sell, a university 'wants' to publish, regional politicians 'want' less unemployment), they often fail to agree on strategic long-term priorities. Instead, the actors find easy, win-win activities that they can agree on and that are expected to bring short-term results. Sotarauta (2016) describes this as regional stakeholders falling into 'black holes' of strategy-building if they accord preference to facile and interim objectives/activities, instead of focusing on inventing and executing long-term visions for their region (see Literature Review).
In this paper the research question is formulated as: What causes regional actors to fall into 'black holes' of strategy-building? It therefore seeks to understand those aspects that drive regional stakeholders to fall into strategic black holes and create a deeper understanding of the processes related to that. The author uses a qualitative case study looking at the Twente region in the eastern Netherlands that is based on interviews with key stakeholders and document analysis. To address the research question, the paper presents a typology of the various factors that drive regional actors to choose suboptimal strategic outcomes. Finally, the paper concludes by arguing that more consideration is needed for the processes that lead those drivers of strategic suboptimality to emerge. It is proposed that one needs to develop a deeper understanding of the aspects that can reduce negative drivers, help the stakeholders to focus on long-term strategic outcomes and thus bring theories and policies forward.

LITERATURE REVIEW
Regional innovation systems (RISs) and regional innovation coalitions The diverse regional actors and groups of stakeholders that interact in a region and create regional innovation strategies are part of a RIS. The RIS approach departs from the notion that actors in the knowledge application subsystem and the knowledge generation subsystem interact actively and thereby facilitate a constant exchange of knowledge, resources and human capital (Cooke, Gomez Uranga, & Etxebarria, 1997). While the RIS concept has been widely praised, critics claim it tends to provide a 'static snapshot of "usual suspect" actors and institutions, reducing the analysis to an inventory-like description of "the system"' (Edquist, 2010;Uyarra & Flanagan, 2010, p. 683). In other words, the systematic RIS approach is criticized for presenting the current situation, without offering heuristic tools to help understand how to construct change in the region.
It is exactly here that regional partnerships come into play, providing a form of dynamism that the strategic RIS approach does not engender. One of the identifiers of regions that have been able to overcome static situations and developed capacities that create new regional futures is an active interplay and cooperation between stakeholders as a contribution to regional advancement. Often cited examples are the Emilia-Romagna region in Italy (Cooke & Morgan, 1994) and the Tampere region in Finland (Lester & Sotarauta, 2007), where stakeholders were able to build successful alliances and achieved shared objectives such as the development of a local innovation environment. Indeed, stakeholder partnerships have become increasingly central to the way that one thinks about regional innovation, have been responsible for the development and implementation of innovation strategies, and hence bring a form of dynamic agency into static systems (Benneworth, Pinheiro, & Karlsen, 2017).
The fact that current regional innovation theory assumes that partnerships can develop longterm strategies to drive change is acknowledged in regional innovation policies, whose success crucially depends on the cooperation within dynamic and enthusiastic RICs. For instance, the regional innovation policy model based on the idea of Constructing Regional Advantage (CRA), launched by the European Union's Directorate-General for Research and Innovation (European Commission, 2006), assumes that public-private partnerships will use and apply their existing knowledge in new ways in order to create regional economic advantages (Asheim, Boschma, & Cooke, 2011). Thus, regional advantage is to be proactively constructed by local actors in coalitions within the existing regional contexts. Another example is Smart Specialisation, where actors are expected to engage in entrepreneurial discovery processes, recognizing and determining those sectors, technologies or overall activities which offer significant future potential for the region (Foray, David, & Hall, 2009;McCann & Ortega-Argilés, 2015).

Regional strategies and the risk of falling into black holes
In both policy concepts, actors within RICs are expected to reflect on the development potential of their region and steer regional trajectories, thereby helping the region to escape from lock-in situations or to develop novel pathways. In other words, the success of the strategies crucially depends on active transformative activities conducted by motivated stakeholders. Nevertheless, whether regional partnerships are actually equipped to deliver what is expected and how regional stakeholders can be mobilized into action are matters that have as yet received little consideration within policy formulation (Capello & Kroll, 2016;Sotarauta, 2018).
Coalition-building and strategy formation are not as easy as assumed by the above-mentioned policies. Rather than being a bureaucratic procedure, strategies emerge from dynamic processes and discursive mechanisms, with stakeholders communicating and negotiating on priorities while trying to agree on future pathways for their region. Developing a regional innovation strategy is therefore an evolving and ever-changing process (OECD, 2009) and building consorted collective action, finding synergies, as well as creating a supportive setting for the heterogeneous stakeholders can present a major challenge .
In line with this thought, Sotarauta (2018, p. 190) argues that Smart Specialisation Strategies are not only about 'policy formulation, implementation and evaluation but also [about] pooling scattered resources, competencies and powers to serve both shared and individual ambitions'. Thus, it is vital for RICs that a 'pooling' of knowledge and resources takes place, serving both collective and also the individual ambitions of stakeholders. To develop the capabilities of regional stakeholders, a shared learning process between the actors is required. Actors need to learn how to take joint decisions, prioritize and consolidate in order to build effective and targeted regional innovation strategies that have the potential to contribute to the region's development and competitiveness. In an environment that is not favourable for common strategy creation, fierce competition between partners, disagreement on priorities and fragmentation might emerge and undermine otherwise constructive activities.
Regional stakeholders do not all have the same levels of rationality and their visions for the future of their region may diverge reflecting their individual/institutional interests and priorities. Sotarauta's metaphorical concept of strategic 'black holes' is an attempt to articulate why regional actors fail to develop coherent long-term strategies given the emphasis on cooperation and collaboration within the contemporary 'multi-actor' and 'multi-value' world of regional development (Sotarauta, 2004, p. 14). In his vision, stakeholders may fall into black holes if the level at which they are able to agree on visions and strategies remains overly abstract and vague. This vagueness may arise as an attempt to resolve the individual aims and competing endeavours of the involved stakeholders, consequently creating strategies that are '"nice and easy to support" because they exclude almost nothing' (Sotarauta, 2016, p. 113). This failure to create a concrete common vision typically leads to an excessive and repeated focus upon short-term objectives and fragmented activities. Thus, cooperation remains short term and ad hoc, and does not build up towards delivering the overall vision over the long term ( Figure 1).

Why do stakeholders fall into black holes?
Having outlined the complexity of cooperation and the risk of strategic black holes, the factors that might explain why actors fall into them will be examined in this subsection. More particularly, the paper examines those aspects that might undermine the long-term and effective working of partnerships and thus lead the stakeholders to agree on suboptimal compromises.
One of the most apparent factors mentioned in the literature is the lack of competency to strategically couple the diverse actors. Although stakeholders cooperate, they fail to merge their interests successfully into a coherent long-term strategy. Yeung (2006, p. 14) explains that combining diverse interests and priorities into long-term regional strategies depends on intentional intervention, time and space. Additionality, interpersonal contacts and continuous dialogue, aspects that eventually lead to trust among stakeholders, have been highlighted as an important prerequisite for strategic partnerships to work successfully (Gertler & Wolfe, 2004). Similarly, Pike, Rodríguez-Pose, and Tomaney (2006, p. 18) argue that 'the risks of failing to identify the correct assets [of regional strategies] are high', referring, again, to the necessary knowledge about how and when to connect which stakeholders and their respective competences.
Thus, a lack of trust between partners or missing experience in working together and coupling interests can lead to the inadvertent acceptance of suboptimal strategies with a focus on short- term goals that do not meet the strategic priorities of the region. With this in mind, another driver for strategic suboptimality can be the tendency to 'parish-pump' politics. Skelcher (2003, p. 2) explains that 'parish-pump' governance is based on the 'parochial' priorities of the close community based on 'small-minded and self-interested individuals'. When parish-pump tendencies cannot be overcome and local governance is principally focused on local interests and accountability to local constituencies (Hospers, 2014), strategic regional partnerships with a focus on a common vision barely have a chance. In the same way, short-term planning horizons based on aspects such as election cycles or annual turnover goals can put an additional burden on strategic partnerships that aim to create a regional vision focused on long-term outcomes.
Related to the above-mentioned factors, the need for intermediaries to negotiate or possibly translate between regional stakeholders has been introduced as an important factor in finding common ground. In this sense, Wright, Clarysse, Lockett, andKnockaert (2008, p. 1208) argue that the choice of the right boundary spanner is crucial as they need to be able to communicate 'the perceptions, expectations and ideas of each to the other'. Kuhlmann (2001, p. 970) points out that intermediaries are necessary for the effective functioning of heterogeneous partnerships, not only because of their ability to facilitate the exchange of knowledge and information but also because they can oversee '"mediated contestation" between representatives of conflicting interests'. The drivers that lead actors to fall into black holes might thus be overcome through the activity of an intermediarypossibly an institutional entrepreneurspanning boundaries between actors that cannot agree on optimal and longterm strategies. In relation to this, Benneworth et al. (2017, p. 237) argue that 'institutional entrepreneurs mobilise resources and actionable knowledge to create/transform "institutions" […] to address RIS inefficiencies'.
The following sections will therefore explore the causes of RICs falling into black holes, aiming to understand the factors that prevent the successful design and implementation of long-term strategic objectives. Delving deeper into the reasons for the appearance of strategic black holes will help one to understand how regional partners can be saved from falling into them and instead build long-term strategies based on activities conducted jointly by enthusiastic and motivated stakeholders.

CASE STUDY AND METHODOLOGY
Since this paper intends to examine which factors can cause regional stakeholders to fall into black holes, an exploratory case study design was adopted. The Twente region is an interesting case for this study because it showcases, on the one hand, regional success stories of cooperation and, on the other, tensions and struggles in the joint design and implementation of regional strategies (see the Organisation for Economic Co-operation and Development (OECD) Peer Review Report of Garlick, Benneworth, Puukka, & Vaessen, 2006). The objective is to deduce general knowledge about difficulties between regional stakeholders that impact on strategy-building and implementation in regional contexts from this case.
The region of Twente, in the eastern Netherlands, unites 14 municipalities comprising the municipal areas around the main towns (Enschede, Hengelo and Almelo) and their rural hinterlands. Twente, being formed by its past textile industry and its quite peripheral position, accounts for about 3.6% of the Dutch population and shares a border with Germany to the east (Figure 2).
The key stakeholders in the Twente region involved in the domain of regional innovation policy and in university-regional cooperation are diverse (Table 1). Although active cooperation between regional stakeholders can be found in Twente (e.g., in the Twente Technology Circle and on the Twente Board), this research will focus on the diverging interests and expectations that affect the region. Specific problems that have been identified in the past are complications: (1) between the markedly different municipalities; (2) between competing sectors; and (3) with respect to the roles of the local university and other HEIs (Garlick et al., 2006).
This case study is based on data collected between April and May 2017 through semi-structured, open-ended interviews and the analysis of key documents. The 14 interviewees recruited (out of 20 approached) were key regional stakeholders from various institutions involved in regional innovation policy (Table 1). For the purpose of identifying a wide range of relevant stakeholders, a snowball-sampling technique was applied, consisting of: (1) investigation of key documents to reconstruct the roles people play in the respective institutions; and (2) recommendations of interview partners and other specialists within the Twente region. To engage actively with the interviewees and stimulate their memories, a narrative interview technique was adopted.
Interviewees were asked to tell 'their' story of working within the Twente region chronologically and answer complementary questions about regional collaboration and engagement, with a particular focus on the University of Twente (UT) and its role in the region. The author was one of two researchers who conducted the interviews jointly.
Additional information was taken from academic and policy literature as well as from documents that included: press releases, strategic agendas, strategic programmes and the mission statements of regional institutions. The author used thematic analysis combined with a framework approach to analyze the empirical material. In other words, a matrix based on the central and recurring themes was created, thereby sorting and synthesizing the data (Bryman, 2012;Ritchie, Spencer, & O'Connor, 2003). Those main themes and subthemes that served as 'the basis for a theoretical understanding of [the] data' (Bryman, 2012, p. 580) were identified through the help of a thorough reading of the interview transcripts and the researchers' notes.

INSIGHTS INTO THE DRIVERS FOR STRATEGIC SUBOPTIMALITY IN TWENTE
The following subsections present the initial findings, each describing one of the three categories of drivers for strategic suboptimality that became apparent throughout the data-collection period. These drivers for strategic suboptimality seem to increase substantially the amount of work that has to be done for the coalitions to operate successfully and inhibit the development of common long-term strategies (often resulting in black hole situations). The first category, misaligned stakeholders, concerns conflicts around unclear and ambiguous roles and tasks of the diverse stakeholders. Second, the absence of the 'right' intermediaries or unattainability of the existing ones has been perceived as particularly challenging for the Twente region. Finally, a dependence on key individuals, their experience, knowledge and individual networks has been identified.
References to the interviewees' names are not disclosed because they explicitly requested anonymity. Therefore, the paper refers only to their functions in: (1) regional government bodies (RG); (2) organizations within the region (for instance, companies, development agency, chamber of commerce, etc.) (RO); or (3) university, including academia (UA) and management (UM). Understanding the strategic 'black hole' in regional innovation coalitions Stakeholder misalignment Interview partners persistently reported that actors in Twente were not suitably aligned. This reflects on the reality that although actors aimed to collaborate, personal and institutional interests within their 'own systems' were often perceived to be of higher value and consequently more important. Throughout the interviews, this discrepancy between priorities was exemplified when discussing past strategies and previous regional boards intended to develop common and streamlined goals for Twente. Accordingly, while one engaged academic (UA2) stated that the regional strategy, namely the Twente Agenda, 'confirms a regional ambition that indeed brings all the regional partners closer together', other interview partners accused the boards of not considering all relevant stakeholders, therefore leaving important actors out of the strategy-building processes. Indeed, the boards were often quite homogeneous, including the same individuals in various constellations, and a representative of the local government claimed that 'a complete base of people who all together have the same goal' (RG1) cannot be found in any strategic body operating in Twente.
A highly relevant aspect in Twente is that the 14 municipalities are particularly diverse and, therefore, the smaller or more rural ones often felt 'left out' and/or resisted ideas initiated by the more urbanized municipalities around the main towns. One interviewee from a regional knowledge transfer organization characterized the situation in Twente as towns and municipalities 'just fighting over and over again, [asking] "Am I visible enough?"' which he claimed is a difficult initial position if one aims to 'create an ecosystem for innovation and high-tech' (RO4).
Another aspect significant here is the fact that interviewees claimed that individual stakeholders in Twente either did not have clearly defined goals and tasks or failed to communicate those effectively to their counterparts within the region. This aspect was particularly evident around the university: while on the one hand some interviewees regarded the UT as an actor that should be focusing much more on the region and its priorities, others thought that this would be in the interests of neither the region nor the UT, arguing that 'for universities, it is quite important to be an international main player; if you are only regional as a university, you are nothing' (UM1). One regional stakeholder felt that this made a joint projection of the region difficult since 'we are one region, the Twente region, but we do not tell the same story' (RO2).

Missing/unsuitable intermediaries
When talking about communication and information/knowledge exchange the second category, namely missing or unsuitable intermediary organizations, became evident. According to information provided by the interview partners, this was particularly evident in relation to interaction/exchange between the university and other regional actors, such as companies and municipalities. One regional stakeholder exclaimed that there is a 'really big wall around the university with big signs [saying] Don't enter! It's our! [sic]' (RO1), highlighting the need for intermediary bodies/individuals who can break down this wall. While many actors in Twente did not know what the mission of the UT was or what it has to offer, they were also uncertain about how to approach the university, or more particularly, its staff. One interview partner pointed out that 'People in Twente don't understand what the university is doing. They are too far away from it' (RG1). Actors within RICs need to be able to exchange knowledge and communicate successfully to accomplish high-order strategy making.
The complexity concerning the role and expectations towards the main and most widely known knowledge transfer institution, namely Novel-T, was highlighted by almost all interviewees. An employee of Novel-T described Novel-T as 'not the university and not the outside, but something in between' (RO3). While on the one hand, interviewees explained that Novel-T was an important actor in bridging the various worlds and necessities of stakeholders in Twente, on the other, it was criticized for focusing on one specific niche of intermediary support. Indeed, it was said to focus entirely on those firms developing from within their start-up support programme and those focusing on technology entrepreneurship.

Dependence on individuals
The final category identified was the strong dependence on knowledgeable, experienced and well-connected individuals from the various key institutions within Twente. As stated above, interviewees claimed that such individuals continually moved within the same circles meeting each other again and again in the various constellations/boards. One interviewee from a regional organization asserted that his networks are based on personal contacts and not on his position, recognizing that 'personal networks are extremely important, which is positive, absolutely positive, but from a quality and consistency point of view, it's a risk' (RO3). Several interview partners stated that the diverse institutions failed to formalize and protocol their experience and networks, resulting in a loss of information should a person leave. A representative of the regional government highlighted that: People are in place for some years, then they take another step, and they are gone. And then you see mostly [that] all the things you have built up [are] gone … there is not a knowledge system that keeps the knowledge. (RG1) A programme manager at the UT confirmed this, stating that because of this movement of people, there is also a discontinuity of projects between partners (UM1). Accordingly, considerable effort, time and money had to be invested in continually rebuilding networks and experiences, while it presented a significant challenge to find new people who were willing to build on and reinforce what had been built by their predecessor.
Although not all the hindering factors discussed above seem to be directly linked to strategy formulation, it is argued that one needs to look beyond the obvious in order to understand this link: the functioning of communication, interaction and exchange of knowledge between various actors in day-to-day business is a prerequisite for them to formulate and implement strategies. RICs consist of partners that work together by pooling knowledge and resources, allow failures to be absorbed collectively and are in constant flux (as a region evolves, the partners evolve). If actors involved in such a bottom-up strategy making process fail to work with each other on general day-to-day terms, they are likely to fail in the 'the higher order strategy-making' that defines them.

DISCUSSION
The evidence introduced above has demonstrated that there are diverse factors that hinder regional partners from building long-term strategies. These factors have an impact on the flow of knowledge, resources and human capital, as well as on the interaction between stakeholders. Indeed, they limit or even inhibit key actors from building long-term strategies, and instead trigger the appearance of black holes. The following subsections will discuss the classes of explanations for suboptimal strategy formulation that were found.

Do actors fall into the black holes because stakeholders fail to align?
The point about misalignment can be seen as indicative of parish-pump attitudes in Twente. As discussed above, Skelcher (2003, p. 2) highlights that the language in strategic partnerships 'is one of leveraging resources, outcome targets and networking'. In direct contrast are the parish-pump attitudes that can be observed in Twente: the continual explanations for not being able to align show a language of justification and defensiveness. Thus, the problem is not of misalignment per se, but can be understood as a possible excuse for stakeholders to favour their own interests and priorities. By continuously blaming misalignment on their regional counterparts, stakeholders from very diverse backgrounds have shown a dismissive attitude to one another.
In that sense, the fact that tasks and missions are not clearly aligned is not the main problem because actors with differing missions could still collaborate effectively. In its place, the problem centres on the unwillingness of stakeholders to 'go an extra step' and the appearance of a kind of satisficing behaviour. In other words, the various regional partners seem to be concerned with rationalizing why they cannot align and accept suboptimal/ordinary outcomes, instead of thinking about cooperation with an entrepreneurial mindset and exerting the serious effort that is needed to transform the region.
For instance, accusing past regional boards of not being able to deliver outcomes is an easy justification for why the diverse stakeholders have not expended an extra effort on cooperation. The fact that they seem to be justifying why things have not gone well is indicative of an ex post facto rationalization of why actors have not worked together. In short, regional partners in Twente find it easier to rationalize why they have not worked together, than to start working together. This factor can therefore be understood in terms of misalignment. It is not known whether there really is misalignment, but the constant talk about distanced aims and actors in Twente thereby converts this argument into the justification for poor/non-transformative performance. In the end, it is the claim that misalignment is predominant in Twente that acts as a barrier and encourages regional actors to accept suboptimal outcomes such as short-term goals instead of finding new ways to align and shape the region over the long term.

Do actors fall into the black holes because of unsuitable intermediaries?
The argument around the absence and unsuitability of regional intermediaries can be understood as a lack of coupling experience and a lack of opportunity recognition. Despite strategic good intentions to build common and long-term strategies and connect actors, considerable experience is necessary to link them. If this experience is absent, actors end up developing short-term strategies (strategic black holes) that might not represent the strategic priorities of the region, risking a failure to identify regional competitive advantages and structural bottlenecks (Pike et al., 2006, p. 18). Since the diverse organizations in Twente (or indeed in any other region) are complex, actors or intermediary institutions who understand each other and have the experience of connecting each other are essential. Thus, when actors say they do not understand what the university is doing and how to approach it, they seem unaware that strategic coupling is a 'process [that] does not happen without active intervention and intentional action on the part of the participants' (Yeung, 2006, p. 14). This active and intentional action is not an easy process but relies on the involvement of experienced actors or intermediaries.
Another relevant aspect is that actors within the diverse institutions in Twente do not get the chance to recognize possible opportunitiespossibly because they do not have experience of doing soand therefore do not start the processes of change. There are potential change makers who could mobilize processes of regional transformation, but they: (1) do not seem to be aware of such opportunities; and (2) do not have the ability to take action due to a lack of experience. It seems that there is a problem of lack of experience and activation, centring on the fact that the change agents/intermediaries who are needed to deliver change are not properly engaged or fail because they do not have the experience to identify the relevant strategic assets. As a consequence of this, regional actors fall into black holes, setting only short-term objectives and pursuing activities that diverge from long-term regional visions.
Do actors fall into the black holes due to the mobility of individuals? As in the first two categories, this factor is not about the mobility of and dependence on individuals per se, but refers to the absence of trust in new stakeholders and the long process of dialogue and interaction that is needed to build new, trusting relationships that are the basis for all long-term strategy-building. The repeatedly reported fear and challenge of losing interpersonal partners with whom connections, experience andmost significantlytrust, have been built up is thus related to the long and complex process of creating mutual understanding among new partners. Gertler and Wolfe (2004, p. 51) claim that 'Building trust among economic actors in a local or regional economy is a difficult process that requires a constant dialogue between the relevant parties so that interests and perceptions can be better brought into alignment.' In this sense, when actors in Twente claim that knowledge is lost and projects are discontinued due to the mobility of people, this is related to not only having to communicate with new partners, but more particularly having to achieve mutual understanding and acceptance (Storper, 2002). In return, this difficulty leads to actors favouring the design of easy and short-term strategies within partnerships, often falling into strategic black holes along the way.
An additional aspect that needs to be considered here is that individuals seem to be required to draw on all their networks (personal, institutional and even networks from former positions), instead of only drawing on their direct links ('I talk to who I am supposed to talk to in my position'). The development of strategies, coalitionsand to a larger extent even regionsthus depends on regional actors that are willing to build up institutional capacity over time, by having and using stable social networks, and by performing bridging functions between each other. Thus, in a process that is heavily dependent on entrepreneurial, active and interconnected actors, it can be particularly damaging if such actors leave and take their knowledge and connections with them. Therefore, if a continuous shuffling of those 'change makers' is taking place, the process not only becomes unstable, but even redundant.

CONCLUSIONS
This paper has analyzed the drivers of strategic suboptimality, focusing on those factors that hinder regional stakeholders from formulating and implementing long-term regional development strategies. Since the development of regions is influenced by an almost limitless set of socio-political-economic forces (Storper, 2013), understanding such factors helps in evaluating and/or forecasting the success of coalitions and strategies, which in return affect the achievement of long-term development goals. The data presented above reveal that diverse partners within the Twente innovation system are facing cooperation hurdles and that strategy formulation does not happen seamlessly.
To help address the research question, the author developed a first typology of the various classes of reasons that might explain why actors fall into strategic black holes. The first factor that makes actors more likely to agree on suboptimal compromises is the parish-pump problem. In short, regional stakeholders show a clear prioritization of their own, immediate needs and interests, not being willing to compromise in pursuit of a common, regional strategy. Second, actors tend to lack the experience of connecting their ideas and building common strategies. This pushes them towards more facile short-term activities that are not in line with the longterm strategies. The third category identified is related to the complexity of engendering trust when new actors come together. Due to the high mobility of stakeholders, new relationships have to be repeatedly built up. As this is a complicated and long process, stakeholders are tempted to design easy solutions to their problems instead of building up trust so as to help design longterm strategies.
Whether the factors identified may be particular to the Twente case or applicable more generally needs to be discovered through further research. Naturally, these dimensions need more explanation as the basis for understanding what drives these black hole problems. What is known is that all regions face challenges and that it is vital to be aware of those before modifying policies or implementing change. Regional strategies thus need to be rooted within the existing structures and competences of a region. Policy-makers aiming to improve their regions should therefore understand the specific regional challenges and develop ways to overcome or ease those. Having developed this first typology, politicians and practitioners can better understand what is happening in a region, ease malfunctions and thereby find a 'way out' of the strategic black hole. By being aware of the challenges and minimizing their existence, RICs can develop into regional bodies that have a high potential to contribute to the successful advancement of their region through long-term strategy formulation.
Current policies and strategies tend to focus on 'just' the shared ambitions, while not giving enough attention to the interaction of individual ambitions. Therefore the 'pooling' of ambitions needs to be considered in more detail throughout this process. Additionally, there is a relatively weak treatment of individual and collective actorhood in regional strategies. The idea that individual change agents and coalitions of those agents are going to drive the region forward is implicit within CRA and Smart Specialisation, but what becomes clear is that the individual change agents and coalitions face tensions that inhibit them from exerting that change agency. Hence, they do not deliver what is expected. The ways that  and Sotarauta (2016) have used institutional entrepreneurship as well as inter-institutional and 'soft' place leadership might be first steps to further develop these points.
This study will thus have implications for both theory and practice. It suggests that, specifically in policy formulation, there is a need to enhance the understanding regional actors have of each other's differences. Furthermore, policy makers need to learn how to identify, empower and mobilize change agents. Having a better understanding of the change agents that are available within a region, recommendations about which problem-solvers should be 'sitting together on one table' in order to develop and implement the 'right' strategy can be made. The findings also suggest that policy designed to encourage regional engagement and strategy-building needs to consider possible drivers for strategic suboptimality and find ways to reduce the occurrence of those.
The conceptual contribution of this paper is first, the identification of this lacuna in the literature and second, filling this gap by recognizing and categorizing the reasons for the frequent failure of RICs. The analysis has shown that although actors are willing to develop common and long-term strategies, there are factors that lead them towards falling into black holes of strategy formulation and implementation. Therefore, what has been described as an ordinary process (coalition building and strategy formulation) in policy agendas such as Smart Specialisation and CRA, is very complex. Indeed, what has tended to be dealt with as an everyday, additive, bureaucratic process is in reality a constructive, creative, innovative, uncertain and transformative process, and the actors included in the definition and implementation of it have been underresearched. Therefore, those actors, their willingness, seriousness and commitment to their region need to be better considered, and the understanding of the factors that lead towards strategic black holes needs to be further developed.