Student: Jesse Witkamp Student number: S3270513 Track: SB&E
Supervisor: Florian Noseleit Co-assessor: Samuele Murtinu Wordcount: 11.180
How to traverse the ‘Valley-of-Death’?
A Multiple Case Study Perspective Among
Abstract
Startups with innovative Renewable Energy Technology (RET) are crucial in accelerating the energy transition. Although there are myriad of startups with promising ideas, many fail to commercialize successfully and end up in the ‘Valley-of-Death’ (VoD). In order to reverse this trend, effective commercialization strategies are needed. Given the lack of adequate literature regarding commercialization strategies in the Renewable Energy (RE) sector, the purpose of this thesis was to enhance our understanding of the problem and to identify how founders/managers can increase their chances of commercialization success. By means of a literature review, expert interviews, and a comparative case study among three Dutch Renewable Energy Startups (RESs) this study found that specific properties of RET lead to complicated competitive, political, and financial challenges that complicate the commercialization process. On the other hand, the case study findings also found that by applying certain strategies, the founders/managers of the examined startups managed to overcome them and commercialized successfully. The identification and exploitation of a specific niche; Intellectual Property Rights (IPR); internationalization; active involvement of stakeholders; reserving sufficient financial resources; and synchronic Research and Development (R&D) and commercialization, are among the strategies that the founders/managers used to overcome these challenges. The results of this research contributed to extant literature by providing a deeper understanding of the challenges that are associated with RET commercialization. It also aimed to provide practitioners with insights in how to address these challenges adequately. Finally, it contributed by formulating propositions and suggesting recommendations for future research.
List with abbreviations and definitions
IPR: Intellectual Property Rights (refers to the general term for the assignment of property rights through patents, copyrights and trademarks. These property rights allow the holder to exercise a monopoly on the use of the item for a specified period.)
RE: Renewable Energy (energy that is collected from renewable sources, which are naturally replenished on a human timescale, such as sunlight, wind, rain, tides, waves, and geothermal heat.)
RET: Renewable Energy Technology (technologies that use renewable energy to provide light, heat, cooling, or mechanical or electrical energy for use in facilities or other activities.)
RES: Renewable Energy Startup (are those that develop technologies for energy generation, energy storage, advanced fuels, energy efficiency, energy software, and energy software appliances, as well as those that deploy or finance clean energy technologies)
R&D: Research and Development (refers to the activities that companies undertake to innovate and introduce new products to the market.)
VC: Venture Capitalist (an investor that provides capital to firms exhibiting high growth potential in exchange for an equity stake. This could be funding startup ventures that wish to expand but do not have access to equities markets. Venture capitalists are willing to risk investing in such companies because they can earn a massive return on their investments if these companies are a success. VCs experience high rates of failure due to the uncertainty that is involved with new and unproven companies.)
Table of contents
1. Introduction ... 6
1.1 Problem statement ...6
1.2 Research questions ...7
1.3 Contribution and significance ...8
1.4 Structure of thesis ...8 2. Theoretical background ... 9 2.1 Commercialization challenges ...9 2.1.1 Competitive challenges ...9 2.1.2 Political challenges ... 10 2.1.3 Financial challenges ... 11 3. Methodology ... 14 3.1 Introduction ... 14
3.2 Justification for research method ... 14
3.3 Research design... 15
3.3.1 Role of prior theory ... 15
3.3.2 Unit of analysis ... 16
3.3.3 Number of cases ... 16
3.3.4 Data collection techniques and procedures ... 17
3.3.5 Collecting, analyzing, and interpreting data ... 17
3.4 Trustworthiness and limitations ... 18
4. Research results ... 19
4.1 Introduction ... 19
4.2. Case study narratives... 19
4.2.1 Case A ... 19
4.2.2 Case B ... 22
4.2.3 Case C ... 24
4.3 Cross-case analysis findings and discussion ... 26
4.3.1 Patterns concerning competitive challenges ... 27
4.3.2 Patterns concerning political challenges ... 28
4.3.3 Patterns concerning financial challenges ... 28
5.1 Introduction ... 30
5.2 Summary of findings ... 30
5.3 Implications for practice ... 31
5.3.1 Startup founders and managers ... 31
5.3.2 Policy makers ... 31
5.4 Suggestions for further research ... 32
References ... 33
Appendix I Intellectual property expert ... 39
Appendix II Interview political expert ... 42
Appendix III – Interview private equity expert ... 45
1. Introduction 1.1 Problem statement
Startups have a crucial role in accelerating the energy transition (Hockerts and Wüstenhagen, 2010). An evaluation of RET readiness, measured on scalability and cost competitiveness, indicated that the world is not nearly physically equipped to displace fossil fuels as the main sources of the world’s base load energy requirements by mid-century (Green, Baksi, and Dilmaghani, 2007). More efficient RET are therefore desperately needed (IPCC, 2018). However, although there are myriad of startups with promising RET, only few are able to commercialize (i.e. to translate their ideas into innovation1 ) and strand in the ‘VoD’
accordingly (Balachandra, Nathan, and Reddy, 2009; Luthra, Kumar, Garg, and Haleem, 2015). For instance, Bluerise- a startup that developed cooling technology by using ocean temperature differences- is a recent case that failed to commercialize. Bluerise received several millions of funding and prizes (RTLZ, 2016). Still, they ran out of capital as costs for experiments were too high (RTLZ, 2019).
Commercialization is challenging as it requires new resources and capabilities for the R&D of the product concept, its successful launch, and interaction with potential buyers and other stakeholders (Pellikka and Virtanen, 2009; Aarikka-Stenroos and Sandberg, 2009). Although startup failure rates are generally high (Pellikka and Virtanen, 2009), they seem to be even higher in the RE sector. As Gaddy, Sivaram, and O’Sullivan (2016) explained, certain properties of RET cause the commercialization environment of startups in this sector to be extra challenging compared to other high-tech sectors, such as biotech or IT.
First, since energy is a commodity, success stems from being a low cost provider rather than from selling differentiated products where customers are willing to pay price premium for (Nanda, Younge, and Fleming, 2013). With increased price competition coming from ‘cheap’ RET (such as Chinese solar panels) and abundant oil and gas, pressure is put on the already razor-thin margins (Bloomberg, 2011). Investing in R&D while also operating a lean manufacturing operation is therefore even more difficult.
Second, most RET is not able to compete against conventional sources of energy (e.g. coal and oil) because fossil fuel investments are often already amortized. The profitability and diffusion of RET therefore depends for a large extent on government policy, such as subsidy provision and carbon taxation (Wiser and Pickle, 1998). Due to this dependency, investors have been wary in investing in the RE sector as policy tends to change frequently, making it difficult to accurately predict return on investment (Verbong, Geels, and Raven, 2008). In addition,
1 In innovation literature, invention is often referred to as the ‘idea’ sprouting from the mind of the inventor.
social protests against RE projects have caused major delays in the past, creating even more uncertainty amongst investors (Breukers, 2007).
Third, the little differentiation opportunities, high capital requirements, long development trajectories, and government dependency have led to a low availability of funding for RESs. Figure 1 shows the failure rate of 150 promising Silicon Valley startups that failed to return capital to investors compared to other high-tech sectors. As illustrated, this figure was over 95% after 2007 (only software company Nest generated high returns as it was acquired by Google in 2010). Consequently, whereas startups in other high-tech sectors can rely on VCs for funding, this is less apparent for startups in the RE sector (Gaddy et al., 2016).
1.2 Research questions
Due to the aforementioned, RESs fail to commercialize relatively often. Although some scholars sought to develop commercialization strategies for startups in the high-tech sector2, it
remained unclear how RESs can overcome these particular challenges and increase their chances for successful commercialization. Thus, the purpose of this thesis was to enhance our understanding of the problem by looking at the causes and effects of competitive, political, and financial challenges on RET commercialization. In addition, it wanted to identify how founders/managers can increase their chances of commercialization success under these conditions. The research questions of this thesis were: What are the effects of competitive, political, and financial challenges on RET commercialization success?; How can RES founders/managers address these challenges in order to overcome them? To answer these questions, it was first examined what these challenges actually encompassed. A literature review
2 In the past, scholars have argued that high-tech startups should form a partnership with third parties to overcome
the VoD (e.g Gans and Stern, 2003; Walsh, 2011; Shan, 1990). By means of collaboration, startups could use the complementary assets from partners to commercialize. In reality, however, startups might not be able or prefer to pursue a partnering strategy as collaboration comes with great risks. Startups can be expected to suffer from dominance of partners and potential ‘hold-up’, because incumbents can dictate the terms in an innovation project (Rosenbusch, Brinckmann, and Bausch, 2011). In addition, independent commercialization is often the only option for startups because incumbents do not always acknowledge the value of the innovation in the early seed phase (Klepper, 2009). Therefore, this thesis will take a wider perspective than partnering strategies only.
Figure 1 Cleantech startups fail to return capital to investors. Source: Gaddy et al.
and expert interviews were conducted to get a deeper understanding of the phenomena. Second, it compared three Dutch RESs who have successfully commercialized and tried to identify what strategic decisions helped in overcoming these issues.
1.3 Contribution and significance
The results of this research contributed to extant commercialization and entrepreneurship by providing a deeper understanding of the challenges that are associated with RET commercialization. It also aimed to provide entrepreneurs, policymakers, and other readers that are interested in the subject with insights in how to address these challenges adequately. According to Bygrave (2007), improving entrepreneurial practice and formulating implications for policymakers is a fundamental objective in entrepreneurship literature (Bygrave, 2007). Finally, it contributed by formulating propositions and suggesting recommendations for future research.
1.4 Structure of thesis
2. Theoretical background 2.1 Commercialization challenges
This chapter provides a theoretical background of the commercialization environment and properties of RESs. It covers the developments of RET in the last decades and explains why certain properties of RET hamper the pace of commercialization.
2.1.1 Competitive challenges
The first challenge lies in increased price competition that can be observed in the RE industry. Whereas the RE market was characterized by a market with only a few small players in the early 90s (McDermott and O’Connor, 2002), it has matured over the years and large multinationals now dominate the market (Lewis and Wiser, 2007). In the battle for efficiency, incumbents have produced solar panels on an unprecedented scale and built wind turbines larger than ever before. As a result, estimations are that the costs most sources of RE will undercut the cost price of conventional energy sources before 2030 (IRENA, 2018). Although this is beneficial for end users, it also undermines innovation and investments as the competitiveness of smaller firms is undercut (Volkskrant, 2018; Astapore, 2016). The solar industry provides a clear example. In 2018, the prices of solar panels decreased with 35% due to Chinese surpluses (Bloomberg New Energy Finance, 2018). SunCycle, a Dutch startup that had won several prices with its solar technology, was not able to survive under these conditions and went bankrupt (RTLZ, 2018). Thus, next to the competition from conventional sources of energy, RESs have to deal with fierce price competition from RET incumbents.
power over startups (Cozzolino and Rothermael, 2018). Protecting RET is therefore hard for startups and IPR can even work counterproductive as key technologies are in the hands of incumbents, complicating the competitive environment for RESs to a further extent.
2.1.2 Political challenges
When it comes to commercializing RET, firms are faced with additional challenges because of a double externality problem (Rennings, 2000). Similar to innovation other sectors, there is an externality in that technological spillovers prevent the innovator from appropriating the full value of the innovation (Auerswald, 2007). However, in the case of RET there is a second externality, namely the lack of internalization of environmental or social costs for incumbent technologies (such as oil and coal). As Hockerts and Wüstenhagen (2010) describe, the presence of external costs has two implications. First, it reduces the relative benefit of technology for customers. Therefore, they have to convince customers to pay price premium for the same commodity, namely energy. Second, the role of government policy affects commercializing RET to a greater extent, because it is the government’s role to internalize these external costs through taxation or other economic policies.
However, the effectiveness of the policies that nations imply diverge. During the COP21 in Paris, 194 countries agreed on counteracting climate change by rigorous measures (UNFCCC, 2016). The collective goal was to reduce greenhouse gasses by 95% in 2050 with the intermediate goal of 49% in 2030. While some countries are ahead of their targets, there are still many that fail to make progress3. According to several scholars (e.g. Balachandra et al., 2013;
Lewis and Wiser, 2007), the main constraint for the development and diffusion for RET was unstable policy. Failed RE projects and technological progress caused many governments to shift policy, almost every few years, which gave governments the image of unreliability (Verbong et al., 2008)4. As governments became more careful in supporting uncertain RET
projects, they generally favored matured technology with the lowest investment risks (Masini and Menichetti, 2013). Some scholars argued that this trend is at the expense of startups with technology options that may present greater efficiency and emissions reduction gains, and strengthens the position of matured technology (Byrnes, Brown, Foster, and Wagner, 2013; Menanteau, 2000). Considering that the profitability of most RET is still dependent on
3 According to Eurostat (2019), Sweden is on the first place when it comes to RE, with a share of 54.5% of its
final energy usage coming from RE sources in 2019. On the other end of the spectrum, the Netherlands is lacking behind their targets (14% in 2020) by having a share of 6.6% in 2019.
4 Verbong et al., (2008) conducted a multi-niche analysis of dynamics in Dutch RE policies. Their finding that
government policy5 (such as on feed-in tariffs or offset regulation), high levels of political
uncertainty hampers private investments (Verbong et al., 2008). Under such conditions, investors are unable to formulate accurate return on investment forecasts.
Next to problems pertaining financial support of RE companies, another issue is the neglect of social embedding of RET with policy focusing primarily on the technical side of innovation (Breukers, 2007). Although RE is widely supported among communities, the negative sides such as horizon pollution and bird shredding, cause a lot of frustration6. Contestation was often
exacerbated by the freedom institutional arrangements gave to stakeholders to protest. Therefore, environmentalists and local inhabitants can easily delay RE projects they dislike (Verbong et al., 2008).
2.1.3 Financial challenges
Raising sufficient capital for commercialization is a real challenge for RESs. Technology development in this sector is typically capital intensive, and requires a substantial degree of up-front debt and equity (Wiser and Pickle, 1998). As Fraser (2009) has shown, startups and other small businesses, in contrast to larger businesses, heavily rely upon internal sources of finance (e.g. personal savings and retained profits). Once they mature and grow in size, they are increasingly able to draw upon trade credit and the short- to medium-term loan market. Berger and Udell (1998) argued that this is because smaller businesses do not have access to the same sources of finance as larger firms. For instance, banks are often unwilling to invest in startups due to a lack of track record and collateral that banks require (Wiser and Pickle, 1998). Special grants or subsidized loans offered by a public organization could be a another source of finance. However, these are often unavailable or insufficient to cover the expenses of capital intensive RET (Masini and Menichetti, 2013).
In other capital intensive industries that are subject to uncertainty and high risks, such biotech or IT, startups rely on VCs to bridge the VoD. (Drover, Busenitz, Matusik, Townsend, Anglin, and Dushnitsky, 2017). However, VC funding for RESs seems to have dried up. At the start of the ‘cleantech-boom’ in 2006, sparked in part by Al Gore’s movie “An Inconvenient Truth” and John Doerr’s TED talk “Green technologies- going green- is bigger than the internet”, many investors agreed that RET represented a massive market opportunity and that
5 IRENA (2018) expects that 75% of onshore and 80% of solar PV capacity that is due to be commissioned by
2020 will produce power at lower prices than the cheapest new natural gas, oil, or coal solutions. Future RE projects will therefore be less dependent on government subsidy to be competitive.
6 For instance, in the Dutch provinces of Groningen and Drenthe, entrepreneurs that invested in wind parks were
the energy sector was ripe for disruption (Gaddy et al., 2016). Although investments in RET demonstrated a sharp increase, this came to a hold in 2011 as the cleantech sector shambled (partly due to the fracking revolution, financial crisis, and cheap Chinese solar panel exports (Forbes 2015; 2009). Estimations are that from the $25 billion dollar that US investors had plowed into the RE sector, over half was lost (Gaddy et al., 2016). As a consequence, funding has dropped substantial ever since and the fit between VCs and the RE industry was seriously questioned.
It is generally very hard for VCs to predict which startups are likely to be successful (Kerr, Nanda, and Rhodes-Kropf, 2013). Hall and Woodward (2010) and Sahlman (1990; 2010) report that approximately 60% of VC investments fail and the vast majority of returns are typically generated from the 10% that perform extraordinary. To reduce this risk, VCs often invest in stages- i.e. buying a series of real options, where data gained after initial investments justifies whether to continue or abandon investments (Gompers, 1995; Bergemann and Hege, 2005; Bergemann, Hege, and Peng, 2008; Guler, 2007). In that regard, staging helps in minimizing losses in failed projects, and in putting a larger amount in firms that eventually succeed. Hence, properties of startups that maximize the option value of their investment, enlarge their portfolio value (Nanda et al., 2013). Examples of these properties are: (1) capital efficient investments- i.e. where the costs of buying are low, (2) where experiments after initial investments reveal many information about the viability of the project, (3) when the experiments can be conducted in a short period of time.
Nevertheless, VCs do invest in other industries like biotechnology, semiconductors and IT hardware that also require vast initial investments. The difference here is that investors rely on established exit mechanisms to sell their early stage investments (Rai, Udwin, Funkhouser, and Livingston, 2015). For instance, the VC model for biotechnology firms has evolved in such a way that pharmaceutical incumbents buy promising ventures even before commercial viability is proven (Ghosh and Nanda, 2010). This vital element in the innovation eco-system reduces the risks for investors because they believe they can sell their shares well before the startup hits the VoD (Nanda and Rhodes-Knopf, 2014). In the RE industry however, incumbents have been less pro-active in acquiring promising RET firms (Figure 2). The reason lies again in that energy is a commodity and that differentiation opportunities are limited (Astapore, 2006). Whereas incumbents in other industries strive to acquire startups that contribute in meeting consumer demand, the typical energy consumer can hardly distinguish between different sources of energy, unless the government puts appropriate taxes on the externality costs of carbon. In the absence of supportive policy, there has been little incentive for incumbents to acquire new RESs (Nanda et al., 2013). As a result, the time to exit for the typical RET firm is much longer than the usual three to five year horizon that VCs define.
For those reasons, Gaddy et al., (2016) argue that betting on RESs is simply not sensible for VCs, who require a profile of risk and return from their investments that is better found in other sectors. Companies that develop new materials, hardware, chemicals, or processes are poorly suited for VC investment because they require significant capital, have a long R&D timeline, are uncompetitive in commodity markets, and are unable to attract corporate acquirers. Hence, investments within the RE category have shifted away from energy generation towards energy efficiency and software (Figure 3).
Figure 3 RESs are less likely to be acquired or exit through an IPO. Source: Gaddy et al., (2016)
3. Methodology 3.1 Introduction
As demonstrated by extant literature, fierce competitive pressures, government dependency, social contestation, and financial constraints at least partly explain why mortality rates in the RE sector are relatively high compared to other high-tech sectors. Still, there are cases of startups that did manage to commercialize their RET successfully. The remainder of this thesis aims to identify how startups are affected by these challenges and which strategic decisions founders/managers of RESs made in order to overcome them. This chapter outlines the research method.
3.2 Justification for research method
Research that is exploratory in nature is more suitable for qualitative than for quantitative research (Eisenhardt, 1989). However, the usual juxtaposition of these types of research makes it easy to miss the fact that qualitative research itself encompasses at least two paradigms: interpretivism and positivism. The difference between the two is that: “Positivist work seeks to identify qualitative data with propositions that can then be tested or identified in other cases, while interpretive work seeks to combine those data into systems of belief whose manifestations are specific to a case” (Lin, 1998, p:162). In other words, the positivist research aims to identify commonalities or patterns between cases, whilst interpretive research aims to understand what general concepts mean in their specific operation.
Although the two paradigms are often juxta positioned to each other, Lin (1998) argues that combining both logics is more effective. Positivist work can identify the existence of causal relationships that are present in data, with some degree of probability. What it cannot do is explain how the mechanism implied by a particular causal relationship works and why the link exists. Interpretivist work, on the other hand, can produce detailed examinations of causal mechanisms in the specific case, explaining how particular variables interact. Without positivist work, however, one does not know how widespread the existence of similar cases might be, which is a question considered important for business practitioners and policy makers (Bygrave, 2007). The combination of both allows the researcher to use the generalizing power of the positivist model to get a sense of the important variables and the scope of the problem; the intensity of the interpretivist model provides the explanations to conclude that a set of relationships is significant theoretically and substantively.
researcher to be more rigorous about defining specific relationships, provides the researcher with ready-made collection of alternative explanations, and keeps the definition of terms from being so situation specific that parallels to other situations are lost.
Hence, a comparative case study approach was reasoned to be most suitable for this thesis as it could answer the research questions adequately. It allowed the researcher to observe why and how the competitive, political, and financial challenges affected the commercialization process in their specific context. As turned out, the challenges were all perceived by the startup founders/managers but they enfolded in a different way. Taking the context into notion was therefore important before any generalizations could be made. In addition, although qualitative studies are generally not suitable for statistic generalizations (Yin, 2009), by incorporating multiple cases into the study it was still possible to compare the decision making of founders/managers and derive patterns and propositions from them, that can function as a starting point for future (quantitative) research.
3.3 Research design
The purpose of the research design is to link questions to the processes for empirical data collection in order to draw conclusions (Bloomberg and Volpe, 2008; Yin, 2009). As Eisenhardt (1989) argued, researchers that apply case study research should consider the role of prior theory, the unit(s) of analysis, the number and selection of cases, the techniques to be used for data collection, and the method(s) by which collected data will be analyzed. Decisions regarding these items in the case study protocol should help to ensure uniformity in research projects where data is collected over an extended period of time in multiple locations (Maimbo and Pervan, 2005).
3.3.1 Role of prior theory
Theme Expertise Function7 Appendix
Competition Intellectual property and valorization. Patent lawyer I Policy Economics of regulating networks, in
particular the energy sector.
Professor II
Finance VC finance, in particular investments in early-stage startups.
Portfolio manager III
Table 1 Overview interviews with experts
3.3.2 Unit of analysis
The unit of analysis were Dutch RESs which successfully commercialized- i.e. they introduced a working product to the market and sold these to consumers. Although case studies are context specific, comparable cases were selected based on age (younger than ten years) and on their properties. Meit and Bouwhuis (2015) identified four energy ‘domains’ that are each characterized by specific properties (table 2). For instance, companies in the energy generation domain (which is hardware related) have higher capital requirements than companies in the energy consumption domain (which is often software related). Considering time and space limitations, this thesis mainly focused on startups in the energy generation domain, because the aforementioned challenges apply more to this subsegment (Gaddy et al., 2016). Thus, startups were selected that commercialized RET related to energy generation8. Moreover, the
founders/managers were selected as appropriate interview candidates since they were likely to be involved in the commercialization process from the start. As Lybaert (1998, p: 188) states: “their strategic position serves as focal point around which all business activities are centralized.”
3.3.3 Number of cases
Generally, the number of participants in qualitative research is small. Yin (2009) advises to limit the number of cases to less than ten, because of the complexity of managing and analyzing the large volumes of data. Evidence from multiple cases is “often more compelling, and the
7 For ethical reasons, real names are not disclosed.
8 Energy generating startups were selected based on a list of StartupDelta (2018), which categorized all Dutch
‘cleantech’ companies according to their innovation domain.
Energy domain Characteristics
Energy Generation Activities related to processing energy into useful energy Energy Facilitation Activities which transfer and/or facilitate energy-products to users Energy Consumption Activities related to energy use, including energy efficiency Energy Storage Activities related to storing energy carriers
overall study is therefore regarded as being more robust” (Yin, 2009, p: 53). Since there were relatively few startups active in the energy generation domain that have successfully commercialized RET, seven startups were contacted initially. The request was sent by phone or email and included an explanation of the purpose of the study, the reason why participation is important, and the amount of time the interview would take. From the seven potential participants, three accepted the request.
3.3.4 Data collection techniques and procedures
One strength of the case study method is its adaptability and flexibility that allows multiple methods of data collection to research a problem (Cavaye, 1996). There are various data collection methods, including observations, interviews, and focus groups (Myers, 2009). Using multiple sources of data and multiple participants is preferable in order to triangulate data (Yin, 2009). A primary source of data in case studies is the interview as it generally gives rich insights. I designed a semi-structured interview (Appendix IV) to keep focus and facilitate a cross-case analysis (Carson, Gilmore, Perry, and Gronhaug, 2001). On the other hand, it allowed room for the exploration of new and relevant issues that emerged from the interviews. News articles, webpages, and social media were used to gather information about the background of the startups and their founders/managers.
3.3.5 Collecting, analyzing, and interpreting data
The interview data were analyzed according to Cope’s (2005) four levels of analysis. The interviews were conducted in a face-to-face setting over a five week period (April 2019-May 2019. The duration of each interview was approximately one hour. All interviews were conducted in Dutch, as all owners were either from the Netherlands or have been living here for several years. Each startup had a quiet room which allowed me to record the interview properly. The first five to ten minutes of each interview were dedicated to a short personal introduction from both sides. I asked permission whether I could record the conversation for transcription.
approach was applied to determine themes up-forehand based on extant literature. Chapter 5 used the three main commercialization challenges in that respect. Lastly, Level 4 concerns the development of a deeper understanding of the data by discussing them in the context of extant literature and the findings of the expert interviews. Therefore, chapter 5 elaborates on the findings by comparing the data with results from extant literature and expert interviews.
3.4 Trustworthiness and limitations
Standards to judge quantitative research are reliability, validity, and generalizability. Although some qualitative researchers use the same terminology, Bloomberg and Volpe (2008) remark that the emphasis of qualitative research is on judging how well the researcher is able to provide evidence that the descriptions and analysis represent reality. Therefore, some researchers suggest to use alternative terminology that may better reflect the different nature of qualitative research; these include dependability, credibility, and transferability. (Gorman and Clayton, 2005; Bloomberg and Volpe, 2008).
Dependability refers to the quality of the data collection and analysis- i.e. the research studied what it claimed to study (Carter, 1999). In line with the arguments of Gorman and Clayton (2005), the dependability in this thesis is demonstrated by outlining the subjective role of the researcher, explaining the data collection and research process in detail, and the usage of multiple sources of data wherever possible. However, it was not always possible to triangulate since there was often no further information available. This is a limitation of the study.
Credibility concerns the match between the researcher’s portrayal of participants with the participant’s perceptions (Bloomberg and Volpe, 2008). To increase the chance for a match, transcriptions and findings derived from the interviews were sent back to the participant for approval and multiple sources, such as news articles, were used scrutinize the findings. However, a limitation is that much of the data per case relies on the truthfulness of a single individual. It is possible that their responses do not match reality.
4. Research results 4.1 Introduction
This chapter presents the results of the second and third level of Cope’s (2005) analysis. These concern the case narratives, the cross-case analysis findings and the subjective interpretation thereof.
4.2. Case study narratives
This section presents the case study narratives. A brief description of each case is listed in table 3. Each case narrative includes a short background introduction of the owner or manager that has been interviewed followed by a description of their startup innovation and the challenges they encountered during the commercialization process.
Case9 Year of establishment Industry Number of employees
A 2012 Wind 25
B 2009 Solar 7
C 2012 Hydro 13
Table 3 Overview of cases
4.2.1 Case A
Startup A was established in 2012 by a group of four founders who all studied at a technical university in the Netherlands. One of the founder’s parents were farmers and had the idea to do something with RE. During that time, solar panels were upcoming and seemed the best option, however, they also had their limitations regarding sunlight availability. The idea to build wind turbines sprouted after one of the founders came back from a trip in Denmark, where he had seen many small scale wind turbines on the properties of farmers. Then, he decided to build the first prototype in his parents’ garage.
4.2.1.1 Competitive challenges
During the time of R&D, some large wind turbine manufacturers were active on the global market. In general, these were also the companies that could produce the largest and most efficient turbines. Considering the tender systems that most governments in Europe worked with, they were also in the competitive advantage regarding price. However, there were only few competitors active on the market that manufactured turbines suitable for small applications (i.e. less than 15 meters high), such as for farmers. Especially since solar energy mainly
generates electricity on the wrong moments (during the day), many farmers regard wind as a more consistent solution for their energy supply.
Still, the solar industry was one of their main competitors as solar panels are another option for small scale applications and had become relatively cheap over the last years (e.g. due to large scale production by Chinese incumbents). The increase in ‘prosumers’ – i.e. households with solar panels that deliver their electricity surplus back to the grid – caused a decrease in electricity price. In turn, this also affects mitigates the costs benefits that a customer can gain from purchasing one of startup A’s wind turbines.
Regarding the development of parts, startup A tried to develop the majority of the components in-house. Since most parts available on the market during the time of development were too big, everything had to be downscaled in order to fit. Although the threat of competitors in imitating the design and idea was real, company A did not patent anything of their technology. One reason is that it was too costly to do so and that they recognized it is not very difficult to work around it.
4.2.1.2 Political challenges
Startup A’s business model strongly depended on government Dutch policy in three ways. First, the benefits for a customer of installing one of their products in terms of costs depended on subsidies. Startup A needed to manage the subsidy application process strictly. Otherwise, projects were delayed with at least six months or their customer could not take advantage of the subsidy.
Second, whether it is allowed for customers to offset their surplus electricity and the height of the compensation they will receive per kWh affected the benefits derived from solar panels compared to wind turbines. With the current policy, prosumers had received the same price for the energy they deliver back to the grid. Especially because solar panels typically generate electricity during the day and in summer months (when we generally use the least electricity), they are particularly interesting when compensation for netting is high. However, the current offset policy lasts until 2023 and will then gradually be reduced until it ends in 2031 due to grid capacity limitations. Consequently, consumers need to find solutions to store energy (e.g. in large lithium-ion batteries such as the Powerwall from Tesla) or they need to make use of a source of energy that is more consistent during the day and seasons, such as wind. Therefore, as offset policy is less likely to be in favor for the solar panel industry, startup A expected that their sales will increase.
substantial market for startup A. Social protests and aversion against on-shore wind energy was not only eminent in Friesland, but also in other provinces where large wind turbine parks are about to be build. In similar ways as in Friesland, this could provoke restrictive laws pertaining wind turbines in these provinces as well.
4.2.1.3 Financial challenges
For product R&D and commercialization, startup A tried to be independent from VCs because it would dilute their autonomy. Banks were not interested to invest as they questioned the risks involved in lending capital to companies in an uncertain sector with limited available market data.
4.2.1.4 Commercialization strategy
In order to overcome the challenges discussed in the previous narratives, startup A made several decisions regarding their commercialization strategy.
Pertaining competition, startup A knew that competing with large incumbents directly is difficult: “Tender systems make it impossible for us to win from large competitors… We thought about working with larger institutions with the aim to install larger wind parks. However, we realized that these trajectories are too long and complicated.” Startup A therefore consciously targeted a niche market: “We choose to focus on small usage systems which are uninteresting for large incumbents.” Startup A also remarked that because they left the incumbents’ market ‘alone’, the chances for imitation were lower as well. Although their success had been noted by other small companies, there are none that developed similar wind turbines for this market.
Finally, startup A experienced major financial challenges to develop and commercialize their products. As they knew banks would not support them, and they were not willing to sell shares to VCs, other means had to be found to finance these capital intensive projects. By persuading early customers to make upfront payments in return for a ‘better deal’, they were able to purchase and develop the required parts without needing other capital. By funding projects this way, startup A could test each turbine, improve it for free, and build a better turbine for the next customer accordingly. Product R&D and commercialization were not separate processes, but simultaneous activities with minimized costs and financial risks.
4.2.2 Case B
The founder of startup B had worked for several years at a solar photovoltaic department of a large energy cooperation before he started his own company in 2009. Initially, the founder traded in the Netherlands with high quality solar panels for grid applications. After three years however, he came up with the idea to integrate lithium ion batteries in the frame of the panel, which allowed the panel to store electricity within the module. Combined with high-quality parts that can withstand high temperatures, startup B’s off-grid solar panels are also applicable for remote and extreme conditions, such as in the desert.
4.2.2.1 Competitive challenges
Competition in the solar industry was fierce, according to the founder: “We see that many startups in this industry are copied within six months by, for instance, Chinese developers. This even happens after long R&D trajectories and venture capital support.” Technology was hard to protect by patents. Although startup B has applied for patents in Europe and the United States, they would not extent these to the Chinese or African market because “…patents are less effective in these markets. Manufacturers will always find a way to work around it and applying for a patent will only week up sleeping giants.” In addition, due to excess capacity and surpluses, Chinese producers sold their panels for dump prices on the market which reduced margins.
4.2.2.2 Political challenges
4.2.2.3 Financial challenges
It took five years before the first commercial product of startup B was ready. Having limited financial resources was one of the main reasons: “I had to invest large amounts of my own capital and worked for other projects as well. If we had the resources of a large company, I think we could have had a fully operational model within two years.” Banks and VCs were also not interested in lending money to startup B because the risks were perceived too high: “I knocked on the door of banks and equity investors many times, but they were not willing to invest in a company that had not generated any revenues for the last three years and operated in such an uncertain environment.” Only after four years and many effort, two banks were willing to issue an subordinated loan to startup B as the market for solar PV showed significant growth signs.
4.2.2.4 Commercialization strategy
Startup B made the following decisions in order to overcome challenges during the commercialization process.
Pertaining competitive pressures, the founder knew, after four years of trading with ‘normal’ solar panels, the margins would go down further in the coming years as a result of (Chinese) surpluses. By developing an off-grid alternative, startup B targeted a niche market which was still in its infancy. The founder examined which customers in the solar market were willing to pay price premium for these solutions and found at that there were two potential interesting segments: “We did not focus on the complete off-grid market, but first on two subsegments: the army and telecom. The reason is that the army has purchasing power and mainly cared about the quality and reliability of products instead of price.” Since the army operated in remote and hot areas frequently, startup B developed panels with batteries that could withstand extreme temperatures (a ‘normal’ solar panel becomes less efficient at temperatures above 25 degrees) and can be installed modularly within hours. Chinese batteries have a bad reputation in the Middle-East and Africa, because these are not designed for hot conditions and have to be replaced often. Startup B tries to exploit this opportunity left open by incumbents.
Pertaining political challenges, startup B tried to be independent from the Dutch market and broadened its scope early on. By focusing on multiple countries around the world, they aimed to minimize risks in case of policy heads to a negative direction in a certain country: “We do not want to rely too much on subsidies and other favorable policy. Our technology is mainly sold in countries where there is not any subsidy for ‘green’ applications. This keeps us sharp and fosters our drive to be competitive.” Although the founder consciously tries to avoid dependency on government policy, he did remark that the startup was to some extent dependent on government subsidy to commercialize the first technologies.
capital, the founder received a subsidy from the European Space Agency (ESA) and from the government. However, these were only issued after the first prototypes were completed: “We could only start talking to investors after we had developed our first prototypes… they are not interested in startups that have nothing to show.” Therefore, startup B first made simple prototypes and gave these to the army in order to test its working. Although the prototype was sold for a small amount of money, they earned some income with other services, such as consultancy. With the feedback data the company received from its customer they was able to improve each prototype until it was good enough to sell. During the commercialization trajectory, it was important to keep costs low and to start early: “For us, starting as soon as possible in the cheapest way was crucial for our survival. That makes it possible to experiment with first prototypes without losing too much money and time.”
4.2.3 Case C
Startup C was established in 2010, after the founder had the idea to develop a small size, fish-friendly watermill. The watermill was designed to function in waters with minimal ground height difference, which means currents are relatively weak. Moreover, whereas existing turbines ‘kill’ at least 25% of fish that try to surpass, the watermill of startup C works with slats that opens and closes automatically as a result of different current flows and reduces fish mortality- according to the founder- near to zero. In contrast to existing turbines with a horizontal ax, the vertical ax design of startup C allows the system to operate completely underwater and is therefore invisible.
4.2.3.1 Competitive challenges
4.2.3.2 Political challenges
Startup B was highly dependent on governments in selling their products. Rivers are usually national property and controlled by governmental institutions. However, the founder of startup C experienced very little support from the Dutch government: “Hydropower is regarded as the stepchild of the RE sector in the Netherlands. Although there is a lot of attention for hydropower internationally, the Dutch government just states the applications for hydro are insufficient. They do not realize such technology can become an important export product.”
Next to political resistance, startup C also experienced fierce protests from the Dutch fish federation. The fish federation had a strong aversion against any form of hydropower, as conventional turbines are associated with high fish mortality and the disruption of eco-systems. Even though the technologies of startup C were proven by independent researchers to be fish-friendly, it took over two years to convince the fish federation to have a look with serious delays as a consequence.
4.2.3.3 Financial challenges
It took several years before the first commercial product installed. Not because of technological reasons, but because negotiations with national and international governments took longer than expected. As a result, developing and commercialization costs during the first couple of years had to be funded, whilst little revenue was generated.
Options for funding were personal savings and subsidies. Banks and VCs were not willing to invest due to a lack of track record: “We got zero response from banks. Even though we had won several prizes in competitions that were sponsored by large banks, none of them were willing to invest…. Seems like banks just cooperated to the events for publicity, a typical example of greenwashing.” Hence, the founder had to invest a substantial amount of his own capital to survive the first few years.
4.2.3.4 Commercialization strategy
In order to overcome the aforementioned challenges, startup C made the following decisions.
Dealing with political issues was most challenging for startup C. For each country they aimed to operate in, the government was the most important stakeholder as they have authority over national rivers. Long-term efforts were necessary in order to convince them to cooperate: “Involving prominent people from national institutions early in the process was important for us. We knew the haul would be long, so we started talking with governments while we were still building the first prototypes. Eventually, it took more than four years before we could start building the first real installation.” A similar strategy was applied pertaining the fish federation. It took over two years to convince the fish federation of their product’s ‘fish-friendliness’. Although the commercialization process was delayed, early attempts to involve the fish federation probably prevented worse, according to the founder.
As mentioned before, startup C was dependent on the founder’s personal wealth to a large extent. Especially in the first years, it was crucial to keep costs low: “It took a long time before we earned any income. Keeping costs low was crucial, we developed most of the prototypes by the two of us.” In addition, the founder also considered it important to commercialize as soon as possible, even though the product was not perfect yet: “We always make special arrangements with our customers. For instance, we are now building two installations for the same customer. First, we install one watermill and see how it works. Based on test data, we make a better second version and then we improve the first one accordingly.”
4.3 Cross-case analysis findings and discussion
The findings from the cross-case analysis (Level 3) are presented in this section and are discussed in the context of extant literature and insights from field experts (Level 4). In addition, the discussion includes the subjective interpretation of the researcher. As the case narratives in section 4.2 described, all startups were seriously challenged by competitive, political, and financial issues during the R&D and commercialization phase. Although each case is context specific, similar strategies were identified pertaining the manner they managed to overcome these particular challenges (table 4). The following subsections use the three challenges (competitive, political, and financial) to provide a more systematic analysis of the observed commercialization strategies.
Challenges Patterns in strategy
Competition 1. In order to avoid direct price competition, all cases tried to identify and target certain niches within the market they aimed to operate in. These niches were, by the time of commercialization, left unexploited by incumbents.
was to have the rights to trade with the product internationally, without the risks of being constrained by a competitor who does patent a similar technology.
Political 1. Dependency on the government for commercialization and delays caused by bureaucracy affected all cases. To mitigate these issues, all cases broadened their focus market internationally as soon as possible with the aim to be less affected by a single policy maker.
2. All cases, although one stronger than the other, experienced pressures from social groups. The chance that protests could hamper the commercialization process was recognized by all cases from the start; by involving all stakeholders early on in the R&D phase, they tried to prevent protests as much as possible.
Financial 1. Financial constrains were one of the major issues of the cases to survive the first years. Although some of their costs were covered by subsidies, it was far from sufficient to survive. At least in the early stages, none of the cases could find support from banks or VCs. On the other hand, they were also not very eager to sell company shares to VCs in the early stages, as it would dilute their autonomy. Survival therefore greatly depended on their founders’ personal wealth.
2. In order to shorten the R&D and commercialization phase, all cases tried to combine the R&D and commercialization process to one trajectory. After the first ‘simple’ prototypes were finished, they installed these at customers who were often willing to pay the cost price. By testing and experimenting with these prototypes, improved versions could be built. Meanwhile, a working prototype could be showed to stakeholders, such as investors and potential customers, in order to accelerate product diffusion.
Table 4. Summary of findings cross-case analysis
4.3.1 Patterns concerning competitive challenges
All startups recognized threats from industry incumbents regarding price competition. Instead of competing with incumbents directly, the previous narratives described that each startup identified a smaller ‘niche’ in the market with specific needs other than just price. All founders were able to identify these unexploited niches and differentiated their RET accordingly. Niches can only be identified if entrepreneurs have a deep understanding of their customers (Bellisario, 2009). Direct price competition was thereby avoided to some extent. That niche strategies are an important factor for startup success has also been demonstrated by other studies. Yap and Sourder (1994) compared US startups in the electronics sector and found that innovation success rates significantly increase when startups pursued a niche strategy.
important for them. For the first four to five years, this can cost between €100.000 and €150.000” (Appendix I). Startup A therefore did not choose to patent their technology. For startup B and C protection was just one aspect in their decision to apply for a patent. More importantly, it gave them the rights and time to export and expand internationally without having to fear from competitors that seek to quickly imitate their products.
4.3.2 Patterns concerning political challenges
Each case illustrated a degree of government dependency, either due to certain policy or because the government was an important stakeholder for the commercialization of their technologies. Since RE policy was regarded as unpredictable, all startups realized that dependency on one market should be minimized. Therefore, a more international approach was enacted to spread risk. Even if policy causes them to withdraw from a certain market, it was still possible to extract revenues from elsewhere.
Social contestation was another factor that delayed the commercialization of RET, especially in case A and C. In order to mitigate social objections, involvement of social groups during the R&D and commercialization phase was considered crucial for the startups. According to a political expert, the government has also learned from social protests: “The climate agreement now emphasizes the importance of regional cooperation, such as co-development of energy strategies… The idea is that when people are involved, resistance mitigates. In the preparation phase for solar and wind parks it is obligatory that at least 50% of civilians are involved” (Appendix II).
4.3.3 Patterns concerning financial challenges
High initial costs and long development trajectories characterized each of the cases. Corresponding to the findings of Nanda et al. (2013), banks and VCs were, especially in the early phase, not willing to lend capital. Reasons were, according to the founders/managers, uncertainty and the lack of a track record. Also, according to a financial expert the value of a startup is difficult to assess in the early stages of a startup for VCs: “Discussions about determining the value of a company and how many shares we receive when investing an x amount of capital is a complicated exercise. If they do a pilot after a month and it does not work, it is worthless. If it works better than expected, it could be worth millions” (Appendix III).
competition between technologies to be most efficient. More expensive technologies are therefore in the disadvantage” (Appendix II).
With limited external capital available, the founders of each case funded the majority of their projects with personal wealth or with the help of family. With an increase in personal risk for bankruptcy, the need for quick commercialization became even more relevant. In each of the narratives, the startups aimed to complete development and commercialization tasks simultaneously. In comparison to the general pattern of technology commercialization where a ‘perfect’ product is completed before selling it to customers, their strategy was to build a simple prototype, install it at a customer for a low price, and then test and experiment with it (figure 4). This allowed them to gather full-scale tests results which they could use to improve new versions, earn some revenue to cover some costs, and to create marketing exposure as they were able to show a operational model.
5. Summary, implications and further research 5.1 Introduction
This chapter addresses the implications for founders/managers and policymakers that can be derived from the findings and concludes with propositions and suggestions for future research.
5.2 Summary of findings
Startups play a major role in the energy transition as more efficient RET is needed in the pursuit for greenhouse gas reduction and electrification (Hockerts and Wüstenhagen, 2010). However, the likelihood of RESs to fail- and end up in the VoD- is relatively high compared to startups in other high-tech sectors, such as in biotech and IT. Certain properties of RET cause extra competitive, political, and financial challenges that startups have to deal with during the commercialization trajectory.
Given the lack of adequate literature regarding commercialization strategies in the RE sector, the purpose the purpose of this thesis was to enhance our understanding of the problem by looking at the causes and effects of competitive, political, and financial challenges on RET commercialization. In addition, it aimed to identify how founders/managers can increase their chances of commercialization success under these conditions. The research questions of this thesis were: What are the effects of competitive, political, and financial challenges on RET commercialization success?; How can RES founders/managers address these challenges in order to overcome them?
By means of a literature review and expert interviews, this study found that specific properties of RET complicate the commercialization process. Little differentiation opportunities, government dependency, high capital requirements, and long development horizon are amongst the most important ones. The findings of the case studies demonstrated that each startup was affected by these challenges albeit in a different way and gradation. On the other hand, it also found that by applying certain strategies, the founders/managers of the examined startups managed to overcome them and commercialized successfully. The identification and exploitation of a specific niche; patenting; internationalization; active involvement of stakeholders; reserving sufficient financial resources to overcome the first period when little income can be generated; and synchronic R&D and commercialization, are among the strategies that the founders/managers used to overcome these challenges.
5.3 Implications for practice
Qualitative research can either conclude with recommendations or implications (Wolcott, 1994). However, as limitations of qualitative research pertaining generalizability make recommendations perilous, the focus of this thesis is on the implications. The implications of the findings can be both directed to founders/managers of startups in the RE sector, and to policy makers who aim to realize an innovative RES climate in their country.
5.3.1 Startup founders and managers
The first important implication of our study is that RES founders/managers of should be aware that success in the energy market often stems from being a low costs provider rather than from offering price premium innovations. However, findings show that niches exist in the market where price competition, at least to a certain extent, can be avoided. In addition, imitation or free riding by others is a constant threat. Patents, although a costly exercise, can be used as a protection measure and to have the right to export globally.
A second implication that can be derived from our findings is frequently changing RE policy can severely affect the profitability of RET and that high risks are involved when betting on one country only. Moreover, the findings point out that social protests against all types of RET are a frequent event which can delay the commercialization process.
The third implication of our study shows that the properties of RET make it hard for founders/managers to receive capital from investors. Especially in the early stages, when no operational model can be showed and when there are no customers yet, the financial risk of the project is often perceived too high by investors. Based on these findings, founders/managers should critically reflect on the following questions (table 5).
Theme Questions
Competition 1. Is the market we serve with our RET sensitive to direct price competition from incumbents? 2. Is our RET properly protected against imitation?
Policy 1. Is our business model still viable if subsidy or other supportive RE policy stops? 2. Which social groups are affected by our RET and have we involved them sufficiently?
Financial 1. Are we able to survive (a couple of years), even when we cannot find external sources of finance? 2. Is our R&D and commercialization path designed effectively?
Table 5 Implications of findings for founders/managers of RESs
5.3.2 Policy makers
synthesized to a larger extent. Long term policy and the involvement of civilians in RE projects are therefore important.
Another issue that was emphasized in this study is the lack of funding opportunities for RESs. The many challenges that confront startups seeking to traverse the VoD have led to some to argue that a persistent innovation gap exists and that, in the absence of VCs, it is the government’s role to support early seed startups (Auerswald, 2007). According to Hepburn, Pless, and Popp (2018), governments can foster the RES environment in three ways: (1) Natural capital should be get a price. This means that energy companies that cause negative externalities need to pay compensation (e.g. carbon tax), which encourages them to invest in clean technologies. (2) Further R&D support for RET. Much of the regulations imposed today only ‘punish’ companies or tend to favor technologies that are already close to the market. Additional subsidies are required to ensure the development of new technologies. Indirect fiscal incentives (such as tax credits) have the advantage of not favoring a specific technology over another and being predictable in financial planning. The facilitation of government laboratories and research institutes are also have an important role to play in reducing startup costs, especially since the R&D and commercialization trajectory of RET tends to be relatively long. (3) Judiciously support of early phase development. Finding sufficient capital is especially difficult in the early startup phase because everything has to be made from scratch. Even if technology specific deployment policy is less cost effective than technology neutral policy, they can accelerate the development of more efficient RET solutions which are crucially needed. RE mandates are an example of such policy, which require energy companies to produce a certain percentage of electricity from RE sources.
5.4 Suggestions for further research
References
Astapore Books (2006). Energy Venture Capital Best Practices: Leading VCs on Spotting Opportunity, Assessing Risk, And Exiting the Investment.
Auerswald, P.E. (2007). The simple economics of technology entrepreneurship: market
failure reconsidered. In: Audretsch, D.B. Grilo, I. and Thurik, R. (eds.): Handbook of Research on Entrepreneurship Policy, Cheltenham: Elgar, p. 18-35.
Balachandra, P., Nathan, H. S. K., and Reddy, B. S. (2009). Commercialization of sustainable energy technologies. Renewable Energy, 35, 1842-1851.
Bellisario, A. (2009) Why serve a niche strategy? Retrieved on 06-06-2019 from: https://www.entrepreneur.com/article/203958
Berger, A.N., and Udel, G.F. (1998). The economics of small business finance: the roles of private equity and debt markets in the financial growth cycle. Journal of Banking and Finance, 22(6/8), 613-673.
Bergemann, D., and Hege, U. (2005). The Financing of Innovation: Learning and Stopping. Journal of Economics, 36, 719-752.
Bergemann, D., Hege,U., and Peng. L. (2008). Venture Capital and Sequential Investments. Working Paper.
Bloomberg, L. D., and Volpe, M. (2008). Completing your qualitative dissertation: A roadmap from begin-ning to end. Thousand Oaks, CA: Sage.
Bloomberg (2018). Solar power buyers loved 2018, but solar panel makers loathed it. [Press release] Retrieved on 17-05-2019 from: https://www.bloomberg.com/news/articles/2018-12-18/solar-power-buyers-loved-2018-but-panel-makers-loathed-it.
Bloomberg (2011) Chinese solar makers seen shrinking to 15 on supply glut. [Press release] Retrieved on 01-06-2019 from: https://www.bloomberg.com/news/articles/2011-11-29/chinese-solar-panel-makers-seen-shrinking-to-15-in-5-years-on-supply-glut.
Breukers, S. (2007). Changing institutional landscapes for implementing wind power. PhD thesis, University of Amsterdam.
Brown, E. (2008). Patents in energy and climate change mitigation technologies. Oak Ridge National Laboratory, 69-71.
Bygrave, W. D. (2007). The entrepreneurship paradigm (I) revisited. In: Handbook of qualitative research methods in entrepreneurship, Neergaard, H. and Parm Ulhøi, J. (Eds.) Cheltenham, UK: Edward Elgar Publishing, pp. 17-48.
Byrnes, L., Brown, C., Foster, J., and Wagner, L.D. (2013). Australian renewable energy policy: barriers and challenges. Renewable Energy, 60, 711-721.
