‘FIT’ TO COLLABORATE?
An exploratory study to the driving factors of public
sector and startup collaboration in grant/incubation
inducement prizes
MASTERS THESIS
Name: Vince Vermeiren (s1285475)
Supervisor: Dr. J. van der Voet
Second reader: Dr. S. Giest
Date: June 11th, 2019
TABLE OF CONTENTS
CHAPTER 1………..2
1.1 INTRODUCTION………....2
1.2 PROBLEM DEFINITION AND RESEARCH QUESTION………...4
1.3 SCIENTIFIC AND PRACTICAL RELEVANCE………...5
1.4 READER’S GUIDE (THESIS STRUCTURE)………....6
CHAPTER 2………..7
2.1 COLLABORATION ……….…..………...7
2.2 PUBLC SECTOR COLLABORATION AND THE INDUCEMENT PRIZE……….8
2.3 THE STARTUP………...10 2.4 THEORETHICAL FRAMEWORK………....12 CHAPTER 3………19 3.1 EXPLORATORY RESEARCH………...19 3.2 CASE SELECTION….………...20 3.3 RESEARCH METHOD…….………...21 3.4 RESEARCH STRATEGY………..23
3.5 STRENGTHS & WEAKNESSES………..25
CHAPTER 4………....27 4.1 CASE DESCRIPTION………...………....27 4.2 CONSEQUENTIAL INCENTIVES………...30 4.3 UNCERTAINTY ………...……….32 4.4 INTERDEPENDENCE…..………...32 4.5 LEADERSHIP………35 CHAPTER 5………....39 5.1 DISCUSSION………...39 5.2 PRACTICAL IMPLICATIONS ………42 5.3 STUDY LIMITATIONS………43
5.4 FUTURE RESEARCH DIRECTIONS………...44
REFERENCE LIST ………...45
APENDIX A ………....52
APPENDIX B………53
CHAPTER 1 1.1 Introduction
As a society, we face problems that are unprecedented in their scope such as a depleted environment and an aging population, problems that cannot be solved with our current methods and approaches (Harris & Albury, 2009). Public sector organizations realize that innovative solutions are necessary and as a result, ‘In house’ innovation strategies can be complemented by multi-actor engagement as a tool to tackle societal problems more effectively (Hartley, Sorensen & Torfing, 2013). A collaboration between public and private actors can yield a range of positive results such as mutual learning and the development of daring new ideas (Hartley, Sorensen & Torfing, 2013). It is therefore not surprising that a pluralization of public governance with forums of deliberation and decision-making involving various actors outside of the traditional locus of bureaucracy is becoming more profound (Torfing, 2016).
Due to the need for innovative social solutions and the opportunities the Internet offers to reach out to problem solvers anywhere, institutions are increasingly resorting towards ‘challenge prizes’ as a way to connect to people and businesses with possible solutions (Nesta, 2014). A challenge prize also known as an inducement prize is a system that incentives citizens to contribute to a problem by providing the best solution with a reward (Nesta, 2014). In such a challenge prize, one or multiple problems are defined and institutions ‘source’ this with the aim of finding a better and more effective way to tackle it. These innovation inducement prizes do work as they can spark competition, which enhances the quality of innovation and can lead to a higher market size due to increased public awareness (Brunt et al., 2008).
The characteristics and merits of innovation inducement prizes have not gone unnoticed in policy implementation. This is especially true in the United States where the 2009 ‘strategy for American innovation’ introduces the challenge prize as a prominent open innovation strategy and policy mechanism to promote innovation. In this strategy, prizes are seen as a powerful way to contribute to find answers to complex challenges by making use of the ideas and skills of the people both inside the government and outside (National Economic Council and Office of Science and Technology, 2009). The resulting America Competes Reauthorization Act (2010) subsequently gave government agencies greater liberties to held prize competitions by giving the heads of agencies the right to ‘awards prizes competitively to stimulate innovation that has the potential to advance the mission of the respective agency’ (section 24B). A pinnacle
of this use of prizes challenges is challenge.gov, an online platform that allows government agencies to post specific challenges to which anyone with an idea of a solution can react. Since its creation in 2010, more than 2000 challenges have been resolved using this method (Challenge.gov, 2018). The Dutch government, although more implicitly, also outlined several policies that stimulate the use of prizes in the public sector. The government promotes public agencies to experiment more with ways to include external actors to improve the public service delivery, which includes adding small parties and start-ups to collaboration initiatives (Rijksoverheid, 2018a).
An implementation of a challenge prize can be in the form of a ‘grant/incubation hybrid’, which in addition to rewarding the outcome also offers additional support to the actors involved through acceleration (Nesta, 2014). In this process, an actor such as the local government invites people or companies to target a challenge in a programme in which they will receive support through collaboration to accelerate their idea, after which the best ideas are awarded a prize at the end of the trajectory. A prominent example of this grant/incubation hybrid, which is implemented in various cities in different countries, is the ‘Startup in Residence’ project. This programme is targeted at solving societal challenges by connecting startups and public organizations. This programme has now been run in several American and Dutch cities, and has recently been introduced in the central government with the Ministry of Internal Affairs of the Netherlands launching is own Residence programme (Rijksoverheid, 2018b). Another programme that uses this structure is the Dubai Future Accelerators Program, a yearly programme bringing over 40 companies and startups together with government institutions in Dubai in a 9-week accelerators contest with the aim to tackle the big challenges of the 21st century (Dubai Future Accelerators, 2019).
With multi-actor engagement on the rise, it is not surprising that the academic community has developed paradigms for this phenomenon of collaboration. Scholars have devoted time and effort to theorizing collaborative governance, exploring the drivers and limitations (Harley, Sorensen & Torfing, 2013) and providing a theoretical model (Ansell & Gash 2008; Emerson, Nabatchi & Balogh, 2011). These efforts are however more geared to strengthen and develop an overarching paradigm or research the outcome then to provide concrete insights into collaboration processes itself, and the do’s and don’ts of collaborating with different external actors.
1.2 Problem Definition and Research Question
Cross-sectoral collaboration has become a central topic of Public Administration research over the last several decade (Crosby, Bryson & Stone, 2015). Yet limited academic attention has gone out to a particular use of cross-sectoral collaboration, the grant/incubation inducement prize. Given the increasing prominence of cross-sectoral collaborative innovation initiatives in the public sector and the involvement of a new actor in the public arena, the startup, could be problematic.
Public organizations and public managers are increasingly focused on managing collaboration and building up competences that are necessary for successful collaborative undertakings (Wendel De Joode, Kaats & Opheij, 2013). In order to effectively manage collaboration in grant/incubation inducement prizes it is helpful to understand how different actors work together and the effect this has on collaboration processes.
Governments are often working together with startups in grant/incubation inducement prizes. These are relatively new actors in the public arena with other aims and operating models then established companies and other Small-Medium Enterprises (SMEs). Startups differ from established companies and other Small-Medium Enterprises in their business model, as it is designed to be repeatable and scalable (Blank, 2010b). As well as in their targeting of disruptive innovation which lots of established companies struggle with due to fixed sets of values and rigid processes (Christensen, 1997). This raises questions as to if startups exhibit different dynamics and how the collaboration process can best be managed within the confines of new forms of collaboration such as grant/incubation inducement prizes.
This thesis addresses the problem of limited knowledge of the collaboration process and startup dynamics in cross-sectoral collaboration initiatives, especially in processes such as grant/incubation inducement prizes. To fully understand the process and manage this effectively, it is important to know more about the perceived drivers to collaboration of the startup and the government. As limited research has been conducted this study focuses on collecting data using an exploratory design. By doing so, it aims to answer the following research question:
‘What are factors that drive the collaboration process between startups and the municipal government in grant/incubation inducement prizes?’
1.3 Academic and Practical Relevance
This research will help expand the current research to inducement prizes and the broader fields of public management and Public Administration in several ways.
Within the field of public sector collaboration, multi-actor engagement has been researched but often in a broad theoretical manner with the exploration of paradigms such as New Public Governance, Open innovation and Collaborative Governance. Research by De Vries, Bekkers & Tummers (2014) indicates, on the basis of a review of 181 books and articles in the top 5 public administration journals, that much of the literature focusses predominantly on intra-organizational process innovations and that as such other forms of innovation have often not been adequately investigated. This is especially the case for startup and public sector collaboration, as little emphasis has been placed on startup inclusion in specific venues of collaboration. This study will broaden public administration scholarship by moving beyond paradigms as a description of public-private engagement and reform towards the collaboration process between specific actors in specific venues. In doing so, it adds a dimension to the literature by shedding light on the collaborative process of public agencies with startups.
Within the field of research to inducement prizes, a surge of research has been visible but particularly in legal and economic and discussions about intellectual property, with much publications discussing how intellectual property can be replaced with prizes (Roin 2014). Yet, the research into the actual real-world applications of innovation inducement prizes is rare. Empirical investigations to actual challenge prizes are limited, with just 11 studies researching government prizes over the last 15 years (Kay, 2018). Next to this, scholars also still have to examine the factors that explain whether a prize is success or not (Liotard, Revest 2017). In moving away from discussions about ownership this thesis aims to add a new layer to the collaborative process of challenges prizes, which has not yet been researched between startups and public organizations, by exploring the conditions that can explain its success. The additional value of this thesis for the literature as regards to inducement prizes is therefore to get a better understanding of the dynamics between actors in the process of collaboration as opposed to a mere theoretical evaluation of the ownership of outcomes.
This thesis furthermore makes a more practical contribution, which can be of value to real world practitioners that are managing public sector grant/incubation inducement prizes. Current collaborative public management research has led to disconnect between the academic
community and the world of practice due to, among others, having a limited effect and connection on and to the public managers (O’Leary, Vij 2012). The authors attribute this to the lack of coherence and aggregating in collaborative public management research. In aiming to provide insights in real concrete and new collaborations in the public sector, this research would combine and expand upon existing collaboration research and add a new layer to this understanding. This knowledge is necessary, as there is necessity to learn more about how prize contests can be structured in an effective way (Liotard & Revest 2017). This new knowledge about the drivers and barriers of public sector inducement prizes could help guide public managers in establishing effective collaborative relationships and aims to contribute to successful collaboration management in grant/incubation inducement prizes. The thesis resulted in several practical implications of the drivers of the collaborative process between startups and municipalities in challenge prizes which can be used to enhance the effectiveness of these processes.
1.4 Readers Guide
This thesis is designed in various chapters and subchapters in order to provide the research in a clear manner. In the next chapter, chapter 2, the theoretical framework will be presented. This will form the basis for the empirical investigation that will be outlined in chapter 3 with the research design, case selection, research method and research strategy. In Chapter 4, the conducted research will be provided and analyzed. Chapter 5 offers a summary of the research and a discussion of the implications of this study for collaborative public management as well as the broader domain of public administration. The appendix includes the reference list and additional information from the research design.
CHAPTER 2
2.1 Collaboration
Collaboration has become a central topic in discussions about innovation and modern public sector service delivery. In the field of Public Administration, there is interest in how collaboration works and the consequences it entails.
Collaboration in public services is however not an entirely new phenomenon, with collaborative public management tracing decades back in intergovernmental relations and policy implementations (McGuire 2006). This involvement is predominantly visible in terms of various layers of bureaucracies and agencies interacting in the coordination and execution of state activities. The interest in collaboration currently focuses more on horizontal collaboration involving several often private sector actors, outside the locus of bureaucracy. In governments in the past this engagement with private sector actors was often seen as a source of competition whereas more recent governments have viewed this as a source of innovation and improvement (Entwistle & Martin 2005). This change is visible in the surge of collaborative governance and current collaborative management scholarship.
As mentioned in the introduction, the focus of this thesis will be to research the drivers and barriers of startup and public sector engagement in grant/incubation inducement prizes. The following theoretical framework will explore various themes such as the properties of the startup, the challenge prize and the starting conditions in relation to start-ups, in order to provide a substantial theoretical background for the empirical analysis. With a topic that crosses boundaries between public administration and business administration, literature from both entrepreneurial studies and collaborative public management is used to explain the concepts and implications in a clear manner. This chapter will proceed by outlining the various concepts and conclude by exploring possible drivers to collaboration.
2.2 Public Sector Collaboration and the Inducement Prize
Various different conceptualizations to describe collaboration exist. In a general sense, collaboration can be seen as the management of differences between actors with the aim of finding solutions to joint problems (Gray, 1989). As such, collaboration is a tool policy makers can use in a world of differences to overcome complexity and produce new solutions. Wood & Gray (1991) looked into the commonalities between existing definitions in an attempt to sharpen the definition of collaboration. The authors conclude by defining collaboration as a process that ‘occurs when a group of autonomous stakeholders of a problem domain engage in an interactive process, using shared rules, norms and structures, to act or decide on issues related to that domain’ (Wood & Gray, 1991, p. 146). Public service collaboration can specifically be regarded as a generally voluntary but contractualized two-way support relationship between at least two partners that share responsibilities and a common objective to provide a public service (Prefontaine et al., 2000).
Collaboration is distinct from other forms of working together such as cooperation, coordination and service integration. Selden, Sowa & Sandfort (2002) reflected on the difference between those collaborative arrangements. On the one end cooperation describes the least formalized and least encompassing dimension where the service arrangement is characterized by close relationships between various actors. A slightly more structured form of an arrangement is coordination, where organizations retain their autonomy but nevertheless try to align certain processes. Next up is collaboration, where actors share their resources and the benefits related to the outcome of the arrangement. The most formalized and far-reaching in implications is service integration in which actors work provide a service together to customers they have in common. As such, when talking about collaboration this thesis refers to an interactive process with shared rules, norms and structures in which resources and benefits are shared.
The private sector can be involved in the various stages of the innovation process, being idea generation, selection and conversion. Systematic idea generation involves attempts to define a certain problem and searching for the solution that best fits the specific problem at hand. Eggers & Singh (2009) note that external partners can do this though involvement in networks, selling innovations and providing knowledge. In terms of selection, businesses can help with the creation of metrics and defining common goals. For conversion and implementation, the public
sector can benefit from the culture the businesses bring, form partnerships and/or use businesses as a provider (Eggers & Singh, 2009).
Inducement prizes are a way for the public sector to reach out to the private sector for this idea generation, selection and conversion. Inducement prizes do not just involve a brainstorm and a concept, but also generally involve a process in which the idea is developed further in order to meet the predetermined challenge (Kay, 2018). In a grant/inducement prize format a public sector organization provides support to the participating parties in the programme in the form of possible financial support for development, tutoring skills to advance the ideas, opportunities for feed-back, formats for peer-learning and input from possible buyers (Bannerjee, Bielli & Haley, 2016). The start-up offers new knowledge and ideas, which can be beneficial for the public sector organization. The support in an inducement prize is often a predetermined and fixed process with a certain life cycle that can be classified as an accelerator trajectory.
Accelerators are ‘fixed-term, cohort-based programs, including mentorship and educational
components that culminate in a public pitch event or demo-day’ (Cohen & Hochberg, 2014, p.
4)
Accelerator programs usually start with a selection of the start-ups that meet the entry requirements, followed by an educational program and end with a so-called ‘demo-day’, where the proposed final product/service is pitched (Wright & Drori, 2018).
Wright & Drori (2018) list six phases: - Candidate sourcing
- Cohort selection process - Accelerator program - Demo day
- Exit to next stage
- Assessment of Performance
The previously mentioned Dubai Future Accelerators programme is a good example of the scope of collaboration endeavors. In the concept/initiation phase companies and startups are introduced to government partners and expertise, in the development phase both work together to develop proposals for piloting, which are pitched to and evaluated by the government
partners in the delivery and feedback phases of collaboration after which a possible agreement is negotiated (Dubai Future Accelerators, 2019).
This practice can be seen as a sharing of resources and benefits related in order to attain an outcome and can therefore be conceptualized as collaboration. The creation of a support system make inducement prizes distinct from coordination where only processes are aligned. Ownership of the service remains at the start-up level, which therefore implies that the outcome is not a joint service and therefore not service integration.
The collaboration in challenge prizes can also manifests itself at the end of the programme as the public sector organization can function as a customer for the proposed solution. This can be seen as sub-contracting or contracting-in. This is along with outsourcing one of the main vehicles of Public-private collaboration (Prefontaine et al., 2000). In the case of sub-contracting the public sector organization hires a private partner to complete a specific task, which the internal organization is often not capable of. In outsourcing, there is a capability to do it but the public sector chooses to hire external actors to complete the task. As the solution, here is something that remains at the startup level and requires the startup employees as executioners, this practice can be marked as sub-contracting.
2.3 The Start-up
Categorizing ‘the start-up’ is not that straightforward with a plethora of used conceptualizations that and range in scope from broad dictionary definitions of entrepreneurial activity to narrow definitions focussing on a range of qualities, highlighting different aspects of operations. A handful of different definitions are listed below:
- ‘A fledgling business enterprise’ (Merriam Webster Dictionary, 2018)
- ‘A business or undertaking that has recently begun operation’ (American Heritage Dictionary, 2018).
- ‘An organization formed to search for a repeatable and scalable business model’ (Blank, 2010a)
- ‘A human institution designed to create a new product or service under conditions of extreme uncertainty’ (Ries, 2011, p. 37).
The dictionary definitions place emphasis on the novelty of the enterprise whereas the entrepreneurial definitions of Steve Blank and Eric Ries note scalability, innovativeness and uncertainty as defining features. This thesis will combine these elements and refer to the entity
of the start-up as ‘a young enterprise with the intention to grow that provides an innovative product and/or service and operates under conditions of extreme uncertainty’.
The literature of public-private cooperation does not often make a distinction between small businesses in general and the ‘start-up’ as an own entity of research. Most of the literature regarding smaller businesses uses the term Small Medium Enterprises (SME), which is highly contested in its meaning. Various countries, international institutions and statistical agencies all have their own particular definitions and categorizations of what would constitute a Small Medium Enterprise (Berisha & Pula 2015). A prominent SME categorization of the European Commission (2003/361) lists amount of staff and turnover rates as main determining factors, with all enterprises with a staff of fewer than 250 employees and a turnover of less than 50 million falling in the category of SMEs. Conceptualizations like these would include most start-ups and a broad range of companies differing in terms of organizational structure, innovativeness and business model. This is problematic as the start-up is a distinct actor.
The elements of fundamental operational uncertainty, scalability, novelty and the provision of innovative goods and services are what make start-ups different from other small businesses. Several authors have reflected on the difference:
- Blank (2014) notes that a traditional company mainly focuses on a business model that works and is hence repeatable whereas start-ups focus on innovation and therefore change.
- Hurst & Pugsley (2011) researched motivations to start a business in the US and found that companies are not necessarily concerned with scalability as enterprises are created for other reasons such as being independent and your own boss. Next to this, they note that roughly two thirds of US-based small-businesses provide ‘relatively standardized goods or services’ to just 40 industries (Hurst & Pugsley, 2011, p.1).
These findings illustrate that most companies are not typically concerned with disruptive innovation and operate under conditions of more certainty instead of fundamental uncertainty.
Research has indicated that start-ups favour to collaborate with other actors. Mercandetti et al., (2017) conducted interview and survey research to start-up connections in Switzerland and found that over half of the start-ups had collaborated with other actors, mostly with a positively perceived outcome and with most of the efforts taking place in the first year of being into business. The researchers note that this collaboration can, in part, be explained because of
difficulties of networking using their own network, and an unclear idea where the innovation could be implemented. The authors also note that the quest for a suitable partner is difficult and that the surveyed start-ups refrained to collaborate with small and like-minded companies due to the low stability it entails.
2.4 – Theoretical framework
Scholars of Organizational Theory and Public Administration have undertaken systematic efforts to grasp the dynamics of the collaborative process in order to understand what conditions are of fundamental importance to a successful collaboration. The study of collaborative governance offers two interesting frameworks in this regard, the collaborative governance model of Ansell and Gash (2008) and Emerson, Nabatchi & Balogh (2011). These models are well cited overviews of the collaborative governance literature, come from reputable journals and emphasize the factors that are of importance in the collaborative process between external stakeholders and public agencies. As this research concerns the collaboration between startups (external actor) and public agencies, these frameworks are interesting to consider.
Both models list the factors that are perceived to be of importance to collaborative processes in which non-traditional stakeholders engage with public agencies. These models both make a distinction between a collaborative process in which certain factors provide a loop that drives this process and factors that influence the existence of such a collaboration. These factors are several starting conditions, the institutional design and the facilitative leadership in the model of Ansell & Gash (2008), and a system context and several drivers in the model of Emerson, Nabatchi & Balogh (2011).
The model of Emerson, Nabatchi & Balogh (2011) is best suited for the purposes of this study as it lists several concrete drivers of the collaborative process and is broader in scope, In the model of Ansell and Gash (2008) collaborative governance is seen as ‘A governing
arrangement where one or more public agencies directly engage non-state stakeholders in a collective decision-making process that is formal, consensus-oriented, and deliberative and that aims to make or implement public policy or manage public programs or assets’ (P. 545),
whereas the collaborative governance is seen by Emerson, Nabatchi & Balogh (2011) as ‘the processes and structures of public policy decision making and management that engage
people constructively across the boundaries of public agencies, levels of government, and/or the public, private and civic spheres in order to carry out a public purpose that could not
otherwise be accomplished’ (p. 2) . The latter places less emphasis on formality and can involve
a wide range of processes and structures as its primary emphasis is on an engagement of people across different spheres to work on a certain public goal that is not to be attained in another way. This fits the notion of the public sector grant/incubation inducement prizes as startups are involved and invited to come forward with a new solution to help address a public goal. This is not to say that this fits collaborative governance perfectly, but many similarities can be drawn making the listed system context and drivers interesting to consider and use as characterization to see if these are also of importance for the setting of the inducement prize.
The general systematic context factors in this framework are noted to be different to starting conditions and are more seen as a setting or background, which influences the process dynamics. This thesis research is interested in the conditions that play a role in the facilitation of this process and therefore these systematic context factors will not be explicitly researched. The drivers that are listed in the model are of importance here, because they are regarded to drive the collaborative process between public agencies and external stakeholders to unfold. As this relation is also present between the startup and the municipality, these drivers could be relevant in the collaboration process. The drivers of this framework will therefore be outlined in the subsequent sections and will serve as a categorization to structure the research. These drivers are ‘consequential incentives’, ‘uncertainty’, ‘interdependence’ and ‘leadership’ (Emerson, Nabatchi & Balogh, 2011).
Consequential incentives
Incentives are seen to be possible drivers of a collaborative process in the framework of Emerson, Nabatchi & Balogh (2011). The authors divide these incentives into two categories, being ‘internal incentives (problems, resource needs, interests, or opportunities)’ and ‘external incentives (situational or institutional crises, threats or opportunities)’ incentives, but do not give concrete examples of what such incentives could be for public agencies and external actors. Whereas, it is difficult to pinpoint the exact incentives for startups and public agencies to engage in a challenge prize format prior to the research, it is helpful to zoom-in as to what these external and internal incentives that drives actors to collaborate could be for both the public sector and the start up.
External incentives to collaborate with other actors in a prize format can be related to possible benefits that a prize can have in achieving policy goals. Prizes can be especially relevant as
external innovation policy instruments, as they can be used for a range of economic and social objectives, including fostering innovation amongst participants, improving the job market and educating citizens (Liotard & Revest, 2017). These prizes have particularly been used to stimulate innovation in society and promote technological development (Kay, 2018). As the use of prizes can have possible external benefits, it can create a strong driving factor to use such a format. It can therefore be expected that external innovation incentives play a major role in grant/incubation inducement prizes. As startups are committed to innovation, particularly the drive to promote technology can be expected as an external incentive of a public agency to commit to collaboration.
Internal incentives to collaborate can also be present as collaboration with private actors can have positive effects for the public sector (Donahue & Zackhauser, 2011). The literature regarding public private partnership and strategic alliances provides an overview account of internal rationales for the public sector to engage with the private sector. Donahue & Zackhauser (2011) provide several specific incentives for public organizations:
- Resource scarcity: Private organizations could provide solutions to resource scarcity of public agencies. It could be that the public budget is not big enough to pursue a specific public goal through public means alone and/or more spending in that specific area cannot be accounted for. -Performance and Productivity: Private engagement could improve performance and productivity.
- Information: Private engagement may help overcome problems of lack of information, where it is not sure how to best target a specific problem.
- Legitimacy: External actors can provide legitimacy especially when support from citizens is low.
The challenge prize could be driven by internal incentives due to the search for a solution to a particular problem/challenge it entails. It is therefore likely that collaboration with startups is at least a bit driven by the problem of lack of information about how to best target the challenge. The problems of tackling the wicked problems could likely be an important incentive for collaborating in grant/incubation inducement prizes.
For startups, incentives to collaborate include enhanced visibility of the startup and its product/service, the ability to gain more knowledge of – and access to a certain market
(Bannerjee, Bielli & Haley 2016). Partnerships and alliances may also lower the financial risks of innovation, enhance the image of the startup and provide the startup with expertise and channels to bring their ideas to the market (Neyens, Faems & Sells 2010). It therefore thus provides the startup with tools to overcome problems that might be associated and common for enterprises of that scope and novelty. It is therefore expected that startups predominantly participate in collaborations due to an internal need to get more information about a market, the mitigation of external threats and a perceived opportunity in working together.
A necessary requirement for an incentive to be a driver of a collaboration process is a certain consequentiality (Emerson, Nabatchi & Balogh, 2011). This means that an incentive needs to be very important to the participants in order for it function as a driver. In this case, it is not entirely clear what the incentives of public agencies to engage specifically with startups in a grant/incubation inducement prize. Therefore, the core focus off the research will be placed on the discovery of what these incentives could be and clearly look for an expressed need of collaboration in order to find the certain incentives that play a particular driving role in this collaboration.
Uncertainty
Uncertainty can also act as a driver of the collaborative process, as a persistent uncertainty in the ability to tackle difficult problems can spark effects to ‘reduce, diffuse and share risk’ (Emerson, Nabatchi & Balogh, 2011, p. 10). In this sense, uncertainty arises when it is unclear how to deal with the complexity of a given problem and an unknown in how your action would contribute to it. Whereas uncertainty in this sense is specifically related to the uncertainties in managing difficult social problems, it can still be rather vague as an analytical construct. Koppenjan et al., (2004) differentiate among the following uncertainties that can arise when dealing with these difficult problems or ‘wicked problems’.
- Substantive uncertainty, which is an uncertainty that can arise when there is an unknown due to a lack of information and knowledge about the nature of the problem as well as differences in how to interpret this information. This can cause difficulties in identifying the problem as well as well as what information to act on.
- Strategic uncertainty, which encompasses uncertainty that develops when there are many strategies to target a wicked problem and it is difficult to establish how different strategies influence the problem and what strategies are chosen by other actors.
- Institutional uncertainty, which arises due to the involvement of different stakeholders with a unique institutional context in a certain problem area. This can cause clashes between institutional regimes that differ substantially and result in an uncertainty regarding the interaction between the actors.
These uncertainties in managing the problems can create an inability to target these problems effectively and as a result a dependency between actors, which can subsequently drive them to engage in cooperation and learning endeavours with other stakeholders (Koppenjan et al., 2004). Given the increasing complexity of social problems, it can well be that government agencies experience at least some of these uncertainties in dealing with these problems. As, a challenge prize involves a public sector agency to put out a certain difficult challenge for other actors to solve, a degree of uncertainty as to the solution of the problem can be expected. Due to the search for innovations in a challenge prize there could well be a strong substantive uncertainty that drives the collaboration, as new information about possible problem solutions is actively being searched for. There could also be strategic uncertainty that drives the collaboration as there could be doubts in which strategy to choose to target the problem or even a strategic need to collaborate with startups in addressing the problem.
If and what kinds of uncertainties actually prevail is not certain but due to the possible importance in driving collaboration over wicked problems these are included in the framework. In the analysis, attention will be paid to substantive, strategic and institutional uncertainties that might be expressed in the relation between the startup and the public agency over these challenges.
Interdependence
Emerson, Nabatchi & Balogh (2011) note that actors can experience a failure to address a certain issue alone, which can drive them to collaborate to solve their problems. Often times, public sector organizations first try to solve their problems staying in their own sector, but when this is not adequate to meet the problem or is expected to not be adequate, it can serve as an important mechanism in reaching out to other organizations (Bryson, Crosby & Stone 2006). A failure in solving a problem can therefore create a certain interdependence on actors beyond their own sector to solve problems. However, this need needs to be acknowledged in order to constitute an interdependence and spur the collaboration. For a cross-sectoral collaboration to
start it is important that the problem solution is seen as being dependent on the involvement of other actors, because if it is not organisations may not collaborate (Longsdon, 1991).
Gray (1985, p. 921) notes that ‘collaborative problem solving efforts do not make sense’ when there is not an acknowledged interdependence between the actors. We can therefore expect there to be at least some interdependence between the start-up and the government agency in a challenge prize. The invitation of startups to contribute to the problem solving could suggest a certain difficulty in the government’s organizations ability to counter the problems, whereas for the start-up there might be a dependency, as the startup might need additional resources due to its scope in order to contribute to solving the problem. However, if this interdependence is actually there between the startup and government agency and how this precisely looks remains unclear. Nevertheless, the literature is clear in the likelihood of interdependence as a driver for the collaborative process, and it is hence worth to take into account.
Leadership
In the model of Emerson, Natbatchi & Balogh (2011) leadership can be both an external driver as well as part of the collaborative governance process. For the purposes of this study, the role of leadership in driving collaboration efforts as an external driver will be central to the analysis. The authors regard leadership as an important driver for the initiation of a collaborative process. The presence of ‘an identified leader who is a position to initiate and help secure resources and support’ is deemed to crucial for the collaborative process to unfold (Emerson, Nabatchi & Balogh, 2011, p. 9). In this view of leadership as a driver, the leader needs to be acknowledged and able to help mobilize resources. As such, not all leadership is sufficient as certain requirements are added to the functioning of leadership as an external driver of collaborative processes. The focus on the ability of the leader to secure resources fits with the perception of the leader as an individual catalyst or as a boundary spanner. In this light, two leadership roles are helpful to distinguish being ‘the champion’ and ‘the sponsor’. A ‘sponsor’ can be distinguished as a person that is not often involved in the collaboration itself, but can use required resources to support it due to their access and stature. A ‘champion’ is a person that is close to the actual collaboration and can use their leadership skills to help the collaboration move forward (Bryson, Crosby & Stone 2006). If these roles are perceived to be there, the chances of leadership as a driver of the collaboration process increases.
One might expect a programme coordinator of the public agencies side to exert a certain leadership role as being part of the large initiating organisation. As startups are relatively small and young, it is expected that this leading role is a required driver for the collaborative process to unfold. It should however be noted, that it is not clear how the leadership driver materializes in collaborative processes between the start up and public agency. However, due to the possible driving effect it can have on a collaboration, it is also helpful to keep in mind.
CHAPTER 3
3.1 Exploratory Research
The collaborative governance literature provides insights as to what conditions are expected to be of importance in collaborative processes with public agencies and external stakeholders, but it is not entirely clear how that relates to new ways of collaboration that focus on working with startups. It is unclear to the literature what drivers are key to grant/incubation inducement prizes, and how and if this is different from the conditions of collaborative governance. As such, the literature can only go as far as to provide substantial information about the nature of collaboration and general drivers and barriers in the process. This is problematic from the point of a deductive research approach, as this entails a precise formulation of theory, which can be tested empirically in order to either confirm or refute a certain hypothesis. It is difficult, if not impossible, to conduct this type of research for collaboration between startups and public sector actors in challenge prizes given the fact that only a few empirical studies have been conducted to government prizes, with zero to none theory building. This lack of knowledge warrants an exploratory study design to contribute to theorizing this gap in academia.
Along with the novelty of the area of interest, this research is suited for an exploratory design due to its explorative aims. The goal of this research is to provide valid insights into factors that could function as a driver in the collaborative process between startups and municipalities and by doing so to shed light on this relatively new venue of collaboration. Deductive testing is not necessarily required in order to attain this research goal. Therefore, an exploratory design is selected, in which methodical exploration will form the basis for the generation of valid insights that could contribute to the advancement of knowledge of collaborative processes.
This explorative design will be conducted systematically to contribute to theory building. It should be noted that academic exploration is contrary to what the name would suggest, not an entirely free endeavor but a ‘systematic and prearranged undertaking’ (Stebbins, 2001, p. 3). A completely open exploration devout of any theory or hypothesis is impossible as researchers are always connected to the physical world, which therefore also makes it necessary to start an explorative inquiry from the point of theory and hypotheses (Reiter, 2013). This research therefore made the conscious decision to briefly reflect on the drivers that are currently known to play a role in collaborative processes. It should however be noted that an exploratory design is not geared towards testing these theoretical elements but rather checks if these elements could
make sense or how well the theory can explain the reality (Reiter, 2013). As such, these elements are used to check if they hold up to the reality of the challenge prize, and in doing so also providing a structure for a systematic academic exploration.
3.2 Case Selection
Case selection for explorative research is different from what a comparative case study in deductive research would entail because only the most telling information dense cases regarding the relationship between variables are selected. The selection of cases has the objective to be ‘a representative sample’ and to be a ‘useful variation on the dimensions of theoretical interest’ (Seawright & Gerring, 2008). As a case needs to be dense in information to provide sufficient insights it should contain a high degree of representativeness regarding the plurality of startups while maintaining the ability to derive useful variation.
This research selected the Startup in Residence programme as the basis for the research. As discussed briefly in the introduction this is project run by various public actors focusing on solving societal challenges through forging connections between startups and the municipality over several themes. This is a particularly suitable case as the general set-ups of the Startup in Residence programmes closely mirror the phases of a grant/incubation inducement prize structure as indicated by Wright & Drori (2018). As such, the set-up of these programmes are typical of a grant/incubation inducement prize.
-Information meetings & applications (candidate sourcing)
- The selection process, in which the larger pool is narrowed down into a cohort of a few suitable startups per challenge (Cohort selection process)
- The actual programme itself, consisting of mentoring, workshops and testing (accelerator programme)
- A demo day, where the startups pitch their progress (demo day)
- An evaluation of the program and the possibility to further develop the services and products of the selected startups. (Exit to new stage, assessment of performance).
As of now, 12 different public sector organizations have been set up or are in the process of starting a Startup in Residence programme. Among those 12 are several cities (The Hague, Amsterdam, Groningen and Utrecht), ministries (Defense, Justice and Security, Interior and Kingdom Relations) and provinces (Zuid-Holland, Noord-Holland, Overijssel, Gelderland) and
a greater metropolitan areas (Amsterdam). Time and resource constraints prevent researching all of these programmes. However, given the observation that these programs in general all follow the same process it would not necessarily attribute to more information density about startup and public sector collaboration in challenge prizes. It is therefore decided is to perform an in-depth single case study.
Not all of these residence programmes would be suitable for a case study to the drivers and barriers of the collaborative process. In order get insights into the entire process from start to end it is a necessity for a program to be at least fully conducted once. The ministries of Defense, Interior and Kingdom relations, the provinces of Overijssel, Gelderland and the city of Groningen are hence excluded as a suitable case. The remaining startup in residence programmes would, in theory be good cases given their typicality of the grant/incubation inducement prize. However most of these programmes follow the example of the first two programmes in the Netherlands, The Hague and Amsterdam. These two were the first to demo in the Netherlands and could therefore also give insights in the struggles of adopting this accelerator set-up to the context of the Netherlands and Dutch startup and public sector relations. Given the fact that these programmes have now been run for several times there is also a lot information regarding the collaboration to be discovered, plenty of variation, and ample opportunity to connect to the many involved participants and employees. It is chosen to select Residence trajectory of the municipality of The Hague as this is the programme with the most variation in challenges, accounting for a large variety of observable startups that differ substantially from each other.
3.3 Research Method
Explorative research to collaboration often involves participatory fieldwork or some sort of interviewing method. Recent examples include among others Dipankui’s (2016) study to the collaboration between practitioners and researchers in health technology assessment in which the researcher conducted semi-structured interviews with both groups alongside participatory fieldwork. In another study Marlier et al., (2015) researched which factors of cross-sectorial partnerships build capacity at the partnership and organizational levels of community sport programs by conducting 44 face-to-face semi-structured interviews in four areas in Belgium. This use of qualitative interviewing in exploration is unsurprising as this method can provide ‘rich and detailed’ answers about the interviewee’s perspective (Bryman, 2012). As such, this
research used an interviewing method in order to gain knowledge and understanding into the collaborative process.
The research used a largely unstructured interview approach, as the question were not strictly defined in advance. This method is chosen, as it is appropriate in situations where research is conducted in underexplored areas and depth is of importance (Firmin, 2008). Unstructured interviews allow for flexibility to ask follow-up questions and deeply dive into a topic that is brought up by the participant, which fits well with the aim of this research to explore factors that could function as drivers and provide insights. However, it should be noted that this interview was not completely unstructured as several topic were outlined in a topic list. This will be discussed in detail in section 3.4
The interviews were conducted with involved employees of the municipality and involved personnel of the startups that participated. Involved employees of the municipality included the personnel that were responsible for the design and execution of the programme, such as the programme managers. Involved personnel of the startups are regarded as any person that represented the startup in the programme. To account for variation in experience not just the ‘winners’ were interviewed, but also representatives that did not make the top 5 as well as representatives of the startups that did not pass the validation stage. This was done with the aim of getting a complete picture and the most information density. As such, all the participating startups as well as the involved programme coordinators of the latest cohort of the programme received an Email invitation to participate in an interview.
The call was received well and resulted in the participation of representatives of 5 different startups (2 winners, 3 ‘losers’) and two municipality coordinators. These interviews were conducted between April 17th and may 3th 2019 and lasted approximately 15 till 40 minutes’ dependent on the responses of the interviewees as well as time constraints. Four of these interviews were conducted face-to-face and three by phone. Face-to-face interviews were preferred, as these types of interviews yield better results in interviews where the topics are complex and the interviewing time is relatively long (Shuy, 2012). As this research focuses on complex processes such as the dynamics of collaboration and requires a lot of feedback this method of interviewing was always the first choice. This was however not always possible and due to the small cohort of eligible participants the upper hand was given to a willingness to participate.
Prior to the start of the interviews, all of the participants were explicitly asked to give their consent by signing a consent form. All of the participant agreed to participate, but concerns were expressed about confidentiality of the research data. Given the sensitive nature of the information provided, the decision was made to anonymize all the participants, and ensure that their participation in this research would not be revealed to others. To ensure the full confidentiality of participation, also information that could directly point to a specific coordinator or startup was left out.
In qualitative research it is customary to record and transcribe the interviews as the manner in which people state their thoughts is of fundamental importance (Bryman, 2012). As such, the interviews were recorded, transcribed and added to this thesis. This is done to allow for other researchers to evaluate the analysis, illustrate the absence of bias and overall contribute to the openness and transparency of the research. In this regard it should be noted that the interviews were conducted in the native language of the participants as to give them the ability to express themselves in language they feel comfortable in. As all of the participants were Dutch citizens, the interviews were all conducted in Dutch. The conscious decision was made here to sacrifice English readability for free expression. Excerpts of the transcripts that were used in the thesis were translated as closely as possible to their original statements in Dutch. The information that could directly point to the identity of the startup or their solution was made unreadable.
3.4 Research strategy
This research used a general interview guide with several topics as the basis of the interviews. The guide comprised a document in which several the key concepts of the theoretical framework were outlined in advance, which served as a reminder to touch-upon the key areas during the interviews. In addition, this document also listed several of the main components of the programme, which were included as a note to ask about the experiences in the various parts of the programme. The guide was used as a ‘living document’ as possible drivers that were mentioned in the earlier interviews, outside of the scope of the theoretical framework, were added to the list. The goal of the interview guide was not to test the framework of Emerson, Nabatchi & Balogh (2011) but to initiate talking, therefore this guide was centered around the topics of interest and did not include strictly defined questions.
Different types of questions were asked in the interviews with the startup representatives, being broad overview questions regarding their motivations to participate, questions to their experiences in the programme, and questions to initiate a conversation about the drivers of the theoretical framework. The first, was asked explicitly to allow the respondents to bring up their motivations and perceived drivers and barriers in their own words. The second, was undertaken to learn more about their experiences in working with the municipality and in the process also learn more about the case. The rationale for this type of question, was that in talking about what the participants like and do not like, relevant aspects regarding the drivers of the collaborative process could come to light. The last type of questions were geared towards facilitating a discussion about the factors that the literature to collaborative governance describes as influencing factors of the collaborative process. This included, among others, inquiries about the perceived interdependence and incentives. It should however be noted that in each of the interviews, the core emphasis was placed on the discovery of the factors that were relevant to the participants, not on asking about all factors that could play a role. As such, the questions should be seen as a method that maximizes the ability of the participant to tell their own experience. The interviews with the programme coordinators generally followed the same type of questions, however, emphasis was also placed on the set-up of the programme and the choices the municipality made in this regard.
The resulting interview transcriptions were analyzed to find commonalties and in doing so, this research used a coding analysis to structure the raw data. In an exploratory study it can be appropriate to limit the coding to the first several phases of the coding process, when there is sufficient attention to validity of the study (Robson, 2002). The authors distinguish these phases as ‘familiarizing yourself with the data’, ‘generating initial codes’, and ‘identifying themes’ (p. 476). This research proceeded according to these steps by first carefully reading the interview transcripts several times in order to get a good grasp of the data. After completion of this step, the research loosely coded the text into groups on the basis of the drivers outlined in the theoretical framework: Incentives, uncertainty, interdependence and leadership, by comparing the text with the descriptions presented in the theoretical chapter. As a general rule of thumb, the data that could express a driving effect or the contrary effect on the collaboration process was coded in this step. This step was followed by a phase in which the coded data was re-examined and themes were distinguished on the basis of repetition. The themes that were witnessed in several instances were re-read, and included in the analysis chapter. It should be
noted however, that the coding was predominantly used to figure out themes that were common among participants and therefore not every piece of text was coded.
3.5 Strengths and weaknesses
The design of this study has several implications that are important to distinguish before outlining the results. A particular strong aspect of using an exploratory design relates to the goal of the research in uncovering ‘hidden facets of reality’ with a certain plausibility (Reiter, 2013). This could provide a fresh point of entry allowing (new) explanations of- and insights into a certain phenomenon and views from a different angle. Given proper conduction this entails that exploratory studies have value in providing an explanation, but it should however be noted this value is not a given, proven ‘fact’ which can solely explain a relationship. As such, this study is limited by its design that permits it to hint at plausible explanation about which drivers are important and not statistical proof.
This study addressed the plausibility and trustworthiness or the research by taking concrete steps in accounting for reliability and validity. The concept of reliability as it prevails in quantitative research entails an ‘exact replicability’, which is not viable in qualitative research (Leung, 2015). Reliability as a measure in qualitative research is distinct and especially entails the researcher to be ‘thorough, careful and honest in carrying out the research, but also being able to show others that you have been’ (Robson, 2002). The research accounts for this by explaining the expectations prior to the research in the theoretical chapter, carefully describing the analyzing process and transparently keeping a record of the interview process through transcriptions that are added to this thesis. This provides the reader with a detailed account of the thought process and the actual data.
This research addresses validity by accounting for common threats in interview research, and using techniques that are regarded to improve the validity of qualitative research. In accounting for common threats, measures were taken to lower the possibility of respondents’ bias and researchers’ bias when interviewing the participants of this study. This research counters respondents’ bias, the type of bias that occurs when participants are not honest in their expression, by clearly stating prior to the interview that the resulting data would not reveal the identities of the participants in any way, and by clearly expressing them the right to decline answering the questions. As some of the startups still
have contact with the municipality in one way or another, this confidentially made it likely easier to be honest about their experiences, whereas the right to decline to answer would make the truthfulness of the answers they were willing to give be possible higher. Researchers’ bias was mitigated by clearly stating the researcher’s perceptions and expectations in the theoretical framework, and by not solely asking questions about the theoretical drivers of collaboration.
This study also focused on persistent observation and the gathering of rich detailed information about the drivers of interest and the case to improve the validity of the research. Persistent observation entails the identification of the elements that are central to the research problem and focusing on them extensively (Lincoln, Guba 1985). This was done by focusing on the issues that the participants described to matter to them and asking follow-up questions when they appeared to be central to the research. In doing so, the participants were urged to tell as much as possible, which resulted in rich descriptions and information density. In addition, this research was careful by only including findings that were consistent among several participants. Whereas, this study was limited by a small sample size, it could not account for full data saturation. Yet, several consistencies among participants were witnessed, which together with the rich descriptions contributed to gathering of valid insights.
CHAPTER 4 4.1 Case description
In 2015, the town council of the municipality of The Hague approved the ‘The Hague Impact Economy’ programme, a policy framework geared towards improving the business climate/ecosystem of start-ups in The Hague. This programme calls for investments in six strategic areas that are thought to create and facilitate an impact ecosystem/economy. These six areas are ‘capital and infrastructure’, ‘visibility’, ‘talent’, ‘networking and events’ and ‘experimentation’. (Dienst Stedelijke Ontwikkeling, 2015).
The Start-up in Residence programme is a part of this Impact Economy programme and has been introduced in 2016 as part of the pillar of experimentation. That year, civil servants of the municipality encountered the program during a work visit to San Francisco and its Residence Programme as part of a Dutch delegation. The programme sparked enthusiasm among the delegation as it was witnessed as a ‘good example of collaboration between government and start-ups’ (Programme coordinator #1) and a ‘nice concept’ (Programme coordinator #2). Subsequently, the programme was proposed to- and introduced in the municipality of The Hague. The programme started small as a pilot of municipality trainees to test if and how it would work in the city (Programme coordinator #1). This pilot proved to be successful and resulted in the full implementation of the program into the department and Impact Economy programme. The programme has, at the time of writing this thesis, been completed two times since the pilot experiment in 2016.
The process itself starts each year with the sourcing of the challenges. The municipality coordinators visit several departments to collect possible urban problems of which several are selected and posted online. The 2018 edition of the Startup in Residence programme featured 10 challenges, covering a broad range of topics. Examples include:
Easy pay for parking: ‘Develop an innovative and easy solution for paid parking’.
No lunch to waste: ‘Find a way to reduce food waste at organized lunches, starting at the municipality’.
Uber-fantastic transportation: ‘Develop an uber-like solution for transportation options for residents with disabilities’ - Dienst Stedelijke Ontwikkeling (2018)
After the challenges are selected and posted online, the Startup in Residence trajectory can begins. This consists of five phases that build on each other being: concept, intensive validation, action & pilot, further development and implementation. The section below will briefly explain what each of these phases entailed for the programme of 2018.
1. Concept (April 10th – June 28th)
The programme started in the concept phase, a time-frame in which interested parties could enrol in one or more of the sourced challenges if they were convinced that their concept or solution could contribute to the challenge (P1). In the Startup in Residence programme 2018, the application process was open to established entrepreneurs as well as people with not yet fully formalized concepts. Interested parties could apply by filling in an online application form after which a jury, consisting of a representative from the Impact Economy Programme and experts on innovation and entrepreneurship, selected a maximum of three applicants per challenge.
2. Intensive validation (June 28th – September 4th)
The following stage is a three-week intensive validation programme in which the participants figure out if their proposed solution fits the sourced challenge. In 2018 this stage included ‘about 8’ mandatory days, in which the participants received training and validated their solution. (Programme coordinator #1). During these weeks, the startup participants met their ‘challenge owner’, the civil servant that represents the municipality in that specific challenge. The weeks culminated in a pitch-day in which the participants presented their business plans to the jury.
3. Action & Pilot (September 5th – End of December)
The best participant(s) per challenge received 7000 euros to work towards a minimum viable
product (MVP) in the action and pilot stage. In this stage, the participants were presented with
workspace and a set number of vouchers to be spent on the help of an experts. The stage ends in a pitch day, which served as the end of the Residence programme.
4. Further Development (Mid December- n.a.)
The startups that were selected to continue to the further development phase after the programme received additional funding to do a pilot and further develop their prototype. This phase of the programme does not follow a set number of days or week, and does not necessarily lead to a lauching customership.
5. Implementation (n.a. – n.a.)
After the further development phase, the municipality can decide to become launching customer (buyer) of the start-up. This means that the municipality will agree by means of contract, to formally work together.
The five interviewed start-ups differ in both their challenge areas as well as their ‘success’ in the programme. The five startups participated in four different challenges and include two startups that ‘won’ and were invited to further develop their prototypes, and three startups that participated but were not selected after the ‘intense validation’ phase. This made the group of interviewees range substantially in experiences, which account for diverse insights regarding the drivers of the collaboration process. Both the programme coordinators as well as the participating startup representatives were eager to discuss their experiences, which resulted in a broad range of information regarding many facets of the collaboration process and its drivers.
The following sections outline the factors that were found to be of important as drivers of the collaborative process in this particular grant/incubation inducement prize. These factors will be presented using the overarching structure of drivers presented by Emerson, Natbatchi & Balogh (2011) as all of the factors that serve as drivers can be discussed in light of these categories. In doing so, the analysis will touch upon all of the categories, as the interviews yielded interesting insights in all of them. In presenting the drivers, the decision was made to not present a list of quotations but to bundle information together in an easy to read text. Whenever some information is presented that is directly related to one of the participants or programme coordinators, this will be referred to the corresponding person. As the confidentially of the participants is important in this research, the corresponding person will be mentioned either as Startup participant (S) or Programme coordinator (P), followed by the number that indicates the order in which they participated in the research. For example, S5 indicates the 5th interviewed startup and P2, the second programme coordinator to be interviewed. As previously
