The Influence of Monetary Funding on Firm´s Innovation-Award Winning Chances: A Resource-Based Perspective

(1)

The Influence of Monetary Funding on Firm´s Innovation-Award

Winning Chances: A Resource-Based Perspective

Philipp Metzner

S2576708

June 22, 2015

Until now, no research has considered the factors that could predict the chances of winning innovation awards. This study acts as a first step in filling this gab by examining financial factors. The goal of the study is therefore to examine the influence of monetary funding on the chances of innovation award winning. The paper takes a Resource-Based View in the analysis of influencing financial factors. Data from the Accenture Innovation Awards have been analysed using regression analysis. The predictive capabilities of financial resources have been unveiled by showing the positive relationship between invested capital and successful participation. Results also showed the importance of the sources of capital. The study provides valuable information about what the chances of success for a contestant are and when to enter a competition.

Keywords: Innovation Award, Investment, Sources of Capital, Financial Barriers, Resource-Based View

University of Groningen

Faculty of Economics and Business

MSc Business Administration

Specialization: Strategic Innovation Management

Word Count: 17726

(2)

2

Executive Summary

Innovation awarding may stimulate technological and societal development by fostering innovation and creativity (Murray, Stern, Campbell & MacCormack, 2012). Additional positive impacts of innovation awards for the awarding institutions, the participants and especially the winners (Locke, Shaw, Saari & Latham, 1981; Murray et al., 2012) underlines the need of understanding the factors that influences award winning chances. The study aims to set a first step by examining the factor monetary funding on innovation award winning chances.

Previous studies have shown the importance of investments in innovation for future success (Cohen & Levinthal, 1990; Coleman, 2013), and this paper analyses whether the importance of

investments also holds true in an innovation award context. Several sub-aspects of funding were

analysed and related to success chances in an innovation award competition to examine the influence of (1) the amount of the investment, (2) the type of funding, (3) the experienced financial

barriers, and (4) possible other factors that influence the relationship between funding and award

winning chances. All financial factors are defined as resources according to the Resource-Based View (Barney, 1986a) and therefore it is attempted to determine the value of each resource for an organization in an innovation awarding context.

Hypotheses are tested by analysing data from the Accenture Innovation Awards using regression analysis. Additionally two interviews were held in order to support or challenge the claims of this study, and to enrich the discussion and conclusion.

The most important finding of this study is the positive impact of the initial investment in an innovation on award winning chances. Results also show that not every source of financial capital is equally important for award winning chances, but that bank loans and venture capital enhance success chances most, which could be explained with the counselling and goal setting a company receives when accessing capital from such sources, as well as with the signalling capability of external investors. Experienced financial barriers do not influence the relationship between the amount of investment and AWC as it was expected, but it has a direct positive impact on success chances.

(3)

1

1. Introduction

Awarding is considered a very effective tool used to stimulate technological and societal development (Murray, Stern, Campbell & MacCormack, 2012). The purpose of awarding is to stimulate certain behaviour in order to develop a field in a preferred direction (Chuan & Soon, 2000). This can be important for many reasons, which depends on the field and the awarding institution. For the awardee, receiving an award creates a feeling of appreciation and motivation (Azadegan & Pai, 2008; Kouzes & Posner, 1999). Even working towards possibly receiving an award can stimulate a focused, effective, and energetic working behaviour; because wanting to receive an award can create a goal one can strive towards (Murray et al., 2012). One of the fields in which awards are considered an important tool of field development is innovation, which is one of the drivers of the modern economy and technological progress. Awards in this field are given to stimulate creativity and innovation by honouring these behaviours (Azadegan & Pai, 2008). Therefore, various different institutions make use of innovation awards (Bhidé, 2009). This type of awarding is applicable on several different levels, such as the personal, team, strategic business unit, or the firm level. Innovation awards are usually governed by a jury of various experts from different industrial or educational backgrounds and different expectations (Azadegan & Pai, 2008). Participation in innovation award competitions is especially attractive to start-ups, as they usually have only one innovation that they would like to promote. The survival of the start-up often depends upon the success of that particular innovation (Cefis & Marsili, 2005), and winning an award can help to find partners or financiers who will support it. It is important to further develop the field of innovation award research, in order to further optimize the benefits that can be reaped from awarding.

(6)

4 enhance the chances of winning. Unfortunately, an analysis of which factors influence the chances of award winning, thus an analysis of potential success predictors is missing.

There are many potential factors that could influence award winning chances, as for example the team experience, the size or constellation of the team, existing partners and suppliers, or the access to technical and market information. One of the major factors that has been identified to be responsible for companies to survive and grow is monetary funding (Cooper, Gimeno-Gascon & Woo, 1994). For instance, Hellmand & Puri (2002) have shown that attracting venture capital has an enormous impact on the professionalization and success of start-ups. Recently, new ways of financing, such as crowdfunding, have become more and more important, which promise not only capital, but also publicity (Ordanini, Miceli, Pizzetti & Parasuraman, 2011). There are two main reasons why it should be expected that funding is also an essential factor that influences innovation award winning chances. First, when an inventor is willing to invest his private money into his idea, it shows commitment to the idea and confidence of its success. Second, when external parties such as venture capitalists or banks invest their capital in an idea, it is underlined that independent parties are convinced of the idea’s success as well (Mason & Stark, 2004). Therefore, it seems that monetary funding is one of the most important factors that could predict award winning chances.

Several factors of monetary funding will be examined in more detail such as the height of the investment, sources of capital and the effects of constraint access to capital. In order to determine the value of these factors in an innovation awarding context, this paper takes a Resource-Based perspective. This means that all investigated factor are defined as resources, which implies that either the amount, or the valence of the concerned resource determine its influence on the firm’s award winning chances. Therefore, it is investigated which financial resources provide added value or valence in the face of an innovation awarding competition.

1.1 Research Aim and Research Questions

(7)

5 The main research question of the current study is:

How does monetary funding influence innovation award winning chances (AWC)?

The following sub-questions are addressed:

 How does the amount of invested capital influence award winning chances?

 How does the type of funding (venture capital, crowdfunding, subsidy, bank loan, private capital) influence the relationship between funding and AWC?

 How do experienced financial barriers to innovation influence the relationship between funding and AWC?

 Are there other factors such as size of the organizations or type of organization that influence the relationship between funding and AWC?

1.2 Research Scope and Domain

Premises and delimitations of the current study are as follows:

Premises:

 The aim of the current study is to examine how monetary funding influences innovation award winning chances.

 There is limited research about predicting innovation award winning chances. Therefore the Resource-Based View as an economic theory is applied to the awarding context and innovation context in order to derive hypotheses about the effect of the financial factors on award winning chances.

 All different factors of monetary funding that are analysed are therefore defined as resources (independent of the nature of the resource; money, value of a network, information or signals of quality). In this way the value of each resource in an awarding context can be examined.

 The capability of these financial resources to predict an innovation award contestant’s success is examined.

 The financial factors that will be included are the height of the investment in an innovation, the sources of capital that finance an innovation and the financial barriers that the innovator experiences. Financial factors are the focus because they are regarded as generalizable, objective, not prone to bias and they are available from most organizations.

(8)

6

 The size of the sample is large, in total 429 companies are included in the sample of the study.

 A regression analysis is used to test the hypotheses and therefore to analyse the effects of financial prerequisites on the success of contestants. This way, the applicability of theories in an awarding context is analysed and grounds for the use of financial data as success predictors are set.

Delimitations:

 This research will not consider contextual variables related to the composition of a firm with respect to any sociological or psychological variables like: team member gender, experience level, abilities, motives, personal preferences, level of team member familiarity or age.

 This research will not study the time range after winning an innovation award, thus the predictions of a success of a firm based on investments are only applicable until an award is received or not received.

1.3 Research Outline

(9)

7

2. Literature Review

The aim of the current study is to examine the influences on monetary funding on innovation award winning chances. The literature review section gives an overview of the relevant literature and key constructs related to this research aim. The first part focuses on innovation in general, and why it is such an important field. The second part of the literature review provides an overview of the current awarding literature, the innovation prize used in this paper and its link to innovation. In the third part theories are introduced with the aim to apply them to awarding theory, and therefore to expand the knowledge in the field, and to answer the research question. The main theory used in this paper is the Resource-Based View (Barney, 1986a), as it defines all examined factors as resources. Furthermore, the key constructs, such as sources of monetary funding and financial barriers to innovation, used in this paper are explained. An overview of the constructs and theories that are used can be found in table 1.

Table 1

Overview of used constructs and theories

Theory /

Construct Definition Authors

Innovation "Innovation is the embodiment, combination, and/or synthesis of knowledge in novel, relevant, valued new products, processes, or services.”

e.g. Dawar & Frost, 1999; Garcia & Cantalone, 2002; Freel, 2000; Leonard et al., 1999; Teece, 1986

Innovation Awards

A prize, aimed at honouring innovators for their innovations, organized as a competition between contestants.

e.g. Brunt et al., 2012; Gemser et al., 2008; Kay, 2011; Murrey et al., 2012; Van der Eijk et al., 2012

Signaling Theory

Signaling is a way of communication between two or more partners, in which one of them wishes to reduce information asymmetries.

e.g. Conelly et al. 2011; Clancy & Moschini, 2013; Dewally & Ederington, 2006; Mina et al. 2013; Reuer & Ragozzino, 2012

Resource- Based View

A management tool that is used to determine the availability of a firms strategic assets.

e.g. Barney, 1986a; Dierickx & Cool, 1989; Mahoney & Pandian, 1992; Peteraf, 1993; Rothaermel, 2012

Funding Providing financial resources to finance a need, program or project.

e.g. Bharadwaj, 2000; Bhide, 1992; Casey & Bartczak, 1985; Clarysse, 2004; Weill, 1992

Financial Barriers

Encountered difficulties or problems with the access to capital.

(10)

8

2.1 Innovation as a Driver of Society

To explain the increasing relevance and popularity of innovation awarding, it is essential to understand the significance of innovation to society and organizational development. Innovation has increased productivity and improved technology for hundreds of years (Cefis & Marsili, 2005; Dawar & Frost, 1999; Freel, 2000; Zhang, Yu, & Xia, 2012). Hence, innovation has been one of the most discussed and important topics in the field of strategy research (i.e. Abernathy & Clark, 1985; Chiaroni, Chiesa & Frattini, 2011; Teece, 1986), as well as among managers. Innovation takes place with the introduction of each new product or service that is brought to the market; as a consequence it is essential to almost any globally operating company. Due to faster and cheaper communication and transportation of goods, competition in almost every market has increased. Because of these globalization effects, firms must continually focus on differentiation, fast response times and gaining a lead in technological development. To survive in the long term, organizations must constantly increase their innovativeness (Cefis & Marsili, 2005; McAdam & Keogh, 2004).

There are several common definitions of “Innovation” (Garcia & Cantalone, 2002; Pol & Carroll, 2006; Schilling, 2010), most of which include a process of invention, development and commercialization or implementation. A common mistake is that innovation is confused with invention. The essential difference is that an innovation must be (successfully) introduced into the market or improve the innovating organization in some way, and therefore must not only be invented, but also developed and implemented. One could say that an invention is only the starting point of many innovations. Furthermore, differentiations can be made for instance between product, service, process and business model innovations (Garcia & Cantalone, 2002).

(11)

9 In sum, innovation is of importance to society and organizational development and should therefore be promoted. One way of doing this are innovation award competitions.

2.2 Awarding as a Driver of Innovation

The most important aspect of this section is showing how awarding can function as a driver of innovation. The section about awarding consists of three parts: The first gives an overview over awarding practices in general, over different ways to judge and to evaluate projects, over the benefits for the institution organizing the competition, as well as over the benefits of the participants in such a competition. The second part focuses on the Accenture Innovation Awards (AIA), as the data of the current study were gathered during these awards. The last part introduces the signaling theory, as it is one of the most important concepts in the field of awarding.

2.2.1. Awarding practices in general

Different kinds of awards are bestowed everywhere in the world and award competitions can be found in almost every industry. For example, there are famous movie awards such as the Oscars, there are awards aimed at scientific breakthroughs or special commitment to bringing peace to the world like the Nobel Prizes and there are awards aimed at the advancement of technology and innovation such as the Xprize. Different awards competitions use different ways to judge and rate its participants. Three main methods have been identified: The first is expert assessment; a jury of knowledgeable and trustworthy people assesses the candidates. The second method is peer assessment, meaning that a group of people assesses and chooses among themselves, like it is done at the Oscars. Assessment by the public is the third method (vox populi), this means that a broad public can vote or that commercial success (in most sales in records, like at the American Music Awards) determines the awardees (Lampel, Jha & Bhalla, 2012).

Awards are essential in so many industries because they are a useful tool to foster societal, economic and technological development (Anand & Watson, 2004; Kay, 2011; Williams, 2012). Next to the general improvement of society, there are very specific benefits for the awarder; therefore many reasons for different institutions to organize award competitions, especially in the field of technological development and innovation. Examples of these benefits for the organizers are media attention and publicity, recruiting talents, identifying trends, accentuating certain topics or problems and getting access to capital and markets (Murray et al., 2012; McKinsey, 2009). In addition, awarding grants the awarder a certain amount of power, as he partly takes control of the information regime in the field (Anand & Peterson, 2000).

(12)

10 towards receiving an award will give the participant a challenging goal (Locke, Shaw, Saari & Latham, 1981; Murray et al., 2012). The candidates receive a full evaluation and in case of a good performance also validation from an external party, which can be very valuable especially for young and inexperienced entrepreneurs (Zhang et al., 2012). The possibility to improve the network of the entrepreneurs and therefore the network of the participating firm is another considerable benefit, as collaboration and networking has become an essential part of innovating. Participants have the opportunity to meet firms from similar industries, and will therefore be able to form ties. This could allow them to get access to new ideas, complementary assets or additional financial resources (Gulati, Nohria & Zaheer, 2000; McKinsey, 2009; Schilling, 2010; Teece, 1986). Of all participants, the award winners are likely to benefit the most. They presumable receive most of the publicity and media attention, therefore they are most likely to build a strong network and send the strongest signals of quality and innovativeness to potential partners and financiers. Additionally, many innovation prizes reward their winners with money or other benefits (Lampel et al., 2012; Murray et al., 2012). One of the largest Dutch innovation prizes doing so are the Accenture Innovation Awards, which provide the data of the current study.

2.2.2 Accenture Innovation Awards

The Accenture Innovation Awards (AIA) are a good example of an innovation prize that embodies all the benefits for the awarder and the awardees named in the previous section. It was first held in 2006 and became one of the largest innovation prizes in the Netherlands with 750 participants and winners in 10 categories in 2014 (Accenture, 2014). Innovators from any company can apply for these awards, as long as the product or service will be launched in the Benelux states. After an official application, the procedure is as follows: First, the participants are sorted into one or multiple categories. The categories in 2014 were: (E-) Retail, Consumer Products & Services, Travel & Transportation, Health, High Tech, Communications, Media & Entertainment, Financial Services, Energy & Chemicals and Sustainability. The categories represent the trends of this year’s innovation topics. After a pre-screening, the contestants have the opportunity to present their innovations in several rounds. The exact number of rounds depends on the number of contestants in each category. Twenty judges, who have a professional background in the category they are judging for, rate the contesting innovation by their innovativeness, their success potential and overall impression. Most members of the jury are innovators and also experts in their field, indicating that both, an expert assessment and a peer assessment, are used (Lampel et al., 2012).

(13)

11 financiers in the future, which is one of the most beneficial effects of winning the award (Gemser et al., 2008; Murray, 2012; Reuer & Ragozzino, 2012).

2.2.3 Signaling theory

Signaling theory has evolved into one of the larger research streams of management literature in recent years, with more than 140 publications in 2009 (Conelly, Certo, Ireland & Reuzel, 2011). The process of signaling is as follows: One party intends to send information to one or multiple other parties to overcome “information asymmetries” (Dewally & Ederington, 2006). The receiver then observes this signal and feedback is sent back (Conelly et al., 2011). Generally, quality of the information is difficult to transfer, as the receiver needs a valid reason to believe in the accuracy of the information. In the case of awarding, a signal can either be the participation in a prestigious awarding competition, or even winning one of the competitions (stronger signal). Therefore, an awarding competition provides the receiver with a trustworthy indication that the signalers innovation is of high quality.

This type of signaling is especially important for Start-ups, since they still need to build up credibility and cannot yet rely on their brand image to signal quality (Clancy & Moschini, 2013). But also most Small and Medium Enterprises (SMEs) will find innovation awards to be a useful signaling tool, as many of them are not required by law to publish annual reports (Mina, Lahr & Hughes, 2013), and therefore cannot make use of their annual reports to signal quality and reliability. Large firms can use awards as quality signals to strengthen their brand, as brand reputation is often signaled from large firms (Reuer & Ragozzino, 2012). In the case of the AIA, there is not only signaling between a participating firms to external parties, but there is also signaling within the competition; from the participating firms to the jury. Firms have to signal the innovativeness of their product, their future potential and make a good general impression by making pitches, demonstrating their ideas or presenting their data and analysis. Accenture itself is also making use of signals. By hosting the AIA, the company signals expertise in the field of innovation to its customers and by presenting an expert jury, the company signals fairness and credibility to contestants.

In sum, awarding has several different benefits for the awardee, the awarder and the society. Also, there are benefits of competing in awarding competitions for signaling, as it can signal high quality of an innovation. Predicting and increasing the chances of success in an innovation award competition is therefore important.

2.3 Funding

(14)

12 financial data. It has been shown that monetary funding is an important resource to almost any company, which leads to a high level of generalizability in the literature (e.g. Bharadwaj, 2000; Casey & Bartczak, 1985; Weill, 1992). The Resource-Based View is used as a framework to explain the effects of financial factors on award winning chances. Different aspects of financial resources that could relate monetary funding and award winning chances are investigated in the following.

2.3.1 Resource-Based View

One of the most well-known and widely used theories in economic research is the Resource-Based View (RBV), which views a firm as a specific set of resources (Wernerfelt, 2006). In the RBV, certain resources or a combination of multiple resources are seen as the key to superior firm performance (Rothaermel, 2012). The RBV further states that for a firm it is important to focus on its key resources in order to develop a “sustained competitive advantage”. These key resources must fulfil the VRIN (valuable, rare, in-imitable, non-substitutable) conditions (Barney, 1986a; Dierickx & Cool, 1989; Mahoney & Pandian, 1992; Peteraf, 1993) in order to be considered sustainable. There is also criticism among the scientific community towards the RBV. Priem & Butler (2001) for instance criticized that a value creating strategy could not be based on resources that are by definition valuable, because this would lead to a self-verifying or “tautological” argumentation. Nonetheless, the RBV is one of the most widely used theories and it has proven a useful tool to analyse organizational behaviour in the past (Mahoney & Rajendran, 2006).

Financial resources usually do not fulfil the VRIN conditions as they are in most cases not unique to an organization and thus do not lead to a sustained competitive advantage. However, despite being common to most organizations, financial resources are regarded a key resource by most firms as they are a fundamental building block that is needed to build up or manage any other resource or capability. Previous study underlined the importance of capital as a key resource (Cohen & Levinthal, 1990; Coleman, Cotei & Farhat, 2013). For example, Coleman et al. (2013) identified “adequate levels of start-up financial capital” as the key resource for new firm survival and growth. Furthermore, Cohen & Levinthal (1990) showed the importance of investments in R&D to improve innovative capabilities. The RBV therefore indicates that more financial capital should predict successful innovation award performance. In a broader perspective however, financial factors such as the source of capital can lead to a competitive advantage, as they may provide benefits to a company that other firms cannot access. This study therefore evaluates the resource value that each of the financial sources provides.

(15)

13 will perform better in innovation awards than those without, especially small and young firms, that are dependent on start capital. Different types of resources might have different effects on contestants in an awarding context, therefore they are more thoroughly investigated in the following.

2.3.3 Source of funding

The aim of the current study is to examine the influence of financial resources on the chances of winning an innovation award. It is important to differentiate between sources of funding, as each of these sources of capital are resources with differing value to a firm. Companies, especially start-ups, may differ in their ways of funding their innovation process (Clarysse, 2004). In this paper, it is differentiated between five different types sources of capital: (1) Own capital, (2) venture capital, (3) bank credit (4) crowdfunding and (5) subsidies. All of these different types of funding might have different influences on award winning chances: (1) When inventors do not attract foreign capital, they either belong to a firm that can provide the necessary funding or they are investing their own private capital. This may be a sign that they are very committed to their idea, however the risk of being undercapitalized is high (Bhide, 1992). (2) If a participant has attracted venture capital, it is a sign of quality. It means that other investors already like and support the idea and that they have confidence in the future growth of the company (Hellman & Puri, 2002; Jell, Block & Henckel, 2011; Reuer & Ragozzino, 2012). (3) A bank credit might indicate that the creditors are trustworthy and that they have a solid business plan (Bhide, 1992). (4) A company that has attracted crowdfunding capital might not be evaluated by professionals as venture capital backed companies are (Hellman & Puri, 2002), but it has already been proven that there might be a potential market for it and the innovation has likely received more publicity then others have (Ordanini et al., 2011). (5) R&D and innovation subsidies are a type of funding that is provided by governments or other institutions in order to stimulate technological development and innovative behaviour. However, a large part of subsidies go to firms that are investing in R&D anyway (Almus & Czarnitzki, 2012). It can therefore be assumed that subsidies have a positive effect on innovative behaviour, but since there are no additional benefits such as expert advice, increased publicity or signaling of quality, the positive effect may be weaker in comparison with venture capital, crowdfunding or bank loans (Reuer & Ragozzino, 2012; González, Jaumandreu & Pazó, 2005), thus own capital and subsidies are expected to be less valuable resources than crowdfunding, venture capital or bank loans.

(16)

14 award winning chances, the origin of the funding has to be taken into account as different sources may affect firm´s award winning chances differently.

2.3.4 Financial barriers

Research has identified a number of different common barriers to successful innovation (D’Este, Iammarino, Savona & Tunzelmann, 2012), which are often a constraint in the access to desired resources. Different types of organizations typically experience different barriers. While larger companies are more likely to suffer from internal barriers such as slow communication or the inability to react to changes in the environment (Christensen & Overdorf 2000), SMEs and start-ups are facing external and internal barriers. In the modern globalized economy, competition has increased significantly, which forces companies to quickly adapt to changes in technologies, and to effectively exploit its own innovations in order to capture as much value as possible. SMEs are often not able to do that as specialists are needed for each of these problems (Christensen & Overdorf, 2000; Porter, 1985). Additionally, SMEs are likely to have a less sophisticated market research program and can at the same time rely less on their brand name and good reputation to sell their products than larger companies. This leads to competition, information, marketing, human resource and financial barriers, that are especially strong for SMEs and start-ups.

Financial barriers are especially important to overcome for start-ups, as their survival often depends on securing their initial investments and overcoming the limited access to deseired financial resources. Most start-ups cannot rely on their reputation and past experiences, therefore a missed initial investment by an external party can quickly lead to liquidity problems (Buse, Tiwari & Herstatt, 2010). Innovative and new ideas are often regarded speculative, which is why financial barriers are especially present when dealing with new and unproven technology (Intrachooto & Horayangkura, 2007). Tourigny & Le (2004) argue that barriers are problematic, because firms spend a lot of time and effort on dealing with them, instead of focusing on their core business. In sum, it can be assumed that firms which recognize financial barriers will be less successful in innovating than those who did not recognize them, as they do not have access to the resources they require to function as planned.

2.4 Summary of the Literature Review

(17)

(18)

16

3. Conceptual Model

The theories and constructs explained in the literature review are combined to formulate hypotheses about the research questions. The research questions of the current study are attempted to be answered by testing the hypotheses.

3.1 Initial Investment & Competition Success

The existing literature (Cooper et al., 1994; Hellman & Puri, 2002; Ordanini et al., 2011) as well as the Resource-Based View suggest that a sufficient financial basis is important for future survival and growth of a company, especially start-ups. Even though financial resources are imitable and therefore cannot lead to a sustainable competitive advantage (according to the RBV), they are an essential prerequisite for almost any firm to innovate successfully (Coleman et al., 2013; Rothaermel, 2012). Having arranged financial resources prior to participating in an innovation awards competition should therefore be an advantage. Therefore it is hypothesized that:

H1: The higher the initial investment in an innovation, the more successful a firm will be in an innovation awards competition.

3.2 Source of Funding

It has been shown that the type of funding may impact future survival, growth and performance of a firm (Hellman & Puri, 2002; Jell et al., 2011; Ordanini et al., 2011; Reuer & Ragozzino, 2012). Attracting venture capital and bank loans is seen as a sign of quality, as external forces have approved the firm’s innovation and planning (Hellman & Puri, 2002). A successful crowdfunding campaign is an indicator for the market’s acceptance of the innovation, while owner funded innovations have a higher likelihood of undercapitalization (Bhide, 1992). When the sources of capital are defined as resources, the benefits of each of the sources determine the value of the resource. As crowdfunding, venture capital and bank loans combine more additional benefits than subsidies and own capital, they are predicted to be more valuable resources, that lead to a competitive advantage. It is therefore hypothesized that:

H2: The relationship between the initial investment and AWC will be influenced positively when using venture capital, bank loans or crowdfunding.

(19)

17 access to these valuable resources as well. Furthermore, the different external financial sources (crowdfunding, venture capital, bank loan) offer different additional benefits to the firm that acquired them; therefore it can be argued that combining multiple benefits will lead to in improved innovation performance (Lee et al., 2001). Therefore it is hypothesized that:

H3: The relationship between monetary funding and AWC will be influenced positively by the number of different external financial sources.

The combination of rare valuable resources may lead to a competitive advantage that cannot be replicated by other contestants. For the duration of the awarding competition, contestants with access to multiple sources of capital may fulfil VRIN conditions (Dierickx & Cool, 1989; Mahoney & Pandian, 1992). The moderation effect for both hypotheses is assumed because of the increased involvement of the financing parties with an increased investment (Colin & Harrison, 2002).

3.3 Financial Barriers to Innovation

As explained in the literature review section, it can be assumed that firms who experience financial barriers to innovation are likely to perform worse in an innovation awards competition, as they are lacking required resources (Intrachooto & Horayangkura, 2007; Tourigny and Le, 2004). This can be due to liquidity problems that hinder the execution of a project or because a firm focuses too much on its limitations, instead of its core business. Furthermore a moderating effect is assumed because firms that have access to large amounts of financial resources will typically also enjoy better lending conditions and fewer constraints in the way the capital is accessed (Batra & Kaufmann, 2003), thus their resources are even more valuable. Therefore it is hypothesized that:

(20)

18

3.4 Research Model

Figure 1 shows the research model to visualize the above stated hypotheses. It shows the proposed direct influence of the initial investment in an innovation on the success chances of award winning, as well as the moderating effects of the source of external capital, the number of capital sources and of experienced financial barriers.

Figure 1. Research model. All hypotheses about the relationship between monetary funding and award

(21)

19

4. Methodology

4.1 Research Approach

Based on the nature of the research questions, a quantitative research approach is chosen (Creswell, 2014). A quantitative approach allows to compare a large number of participants, and to gather statistically valid results from the research model. This approach is used mostly when theory is already developed, but not yet validated in the field, which is the case in this study. Approved and developed theories, such as the Resource-Based View and the signaling theory, are applied to an innovation award context, in order to support or reject their applicability in this field. Theoretically this means focusing on the second half of the empirical cycle, which includes the steps of deduction, testing and evaluation (based on Blumberg et al., 2011; de Groot, 1969; Saunders et al., 2007). This study follows the classic theory testing process by Aken et al. (2012) shown in figure 2.

Figure 2. Theory testing process by Aken et al. (2012)1

4.2 Selection of Sample and Data Sources

Data was collected by Accenture Netherlands and was obtained in the Accenture Innovation Awards 2014 by means of the Accenture Innovation Award Questionnaire 2014 (AIAQ-2014). Companies that were included in the current study had to enter the Accenture Innovation Awards, for which three perquisites had to be fulfilled: (1) The innovation had to be younger than three years, (2) the innovation had to be launched in the BeNeLux states (Belgium, the Netherlands and Luxembourg) and (3) the innovation had to have a certain level of newness, which was determined by the jury. All companies that fulfilled the prerequisites were asked to fill out the AIAQ-2014. The questionnaire had to be filled in prior to the judging process, controlling for a possible bias of winning or no winning (table 2).

1

(22)

20

Table 2

Stages of the Accenture Innovation Awards. Indicated is the moment of completion of the AIAQ-2014.

Registration Companies register for the awards by completing questions concerning their innovation

Pre-selection Prerequisites are checked based on the questions, companies fulfilling them go through to the judging events

Filling in AIAQ-2014 All companies fulfilling the prerequisites are asked to fill in the AIAQ-2014

Judging events Companies present their innovation which is judged by a jury, finalists are chosen

Final Finalists present their innovation again (in 2014: 55) Winners Winners are chosen (in 2014: 10)

In total, 726 participants returned the AIAQ-2014, however, 297 had to be excluded due to missing data. This left 429 participating companies in the dataset. Of the 429 participants, 31 were finalists and 7 were award winners. Table 3 presents descriptives of the sample, separately for the successful and non-successful participants. Note that variables belonging to the category entrepreneur included in the descriptives of the sample have missing values. This is also the reason why these variables were not included in further statistics.

The sample contains a surprisingly low number of female constants (4%), none among the successful candidates. Also there are not many mixed groups (17%) in the whole sample, which leaves a very high number of purely male contestant groups (85% successful, 78% non-successful). It is interesting to note that there are 32% of entrepreneurs who only have a high school diploma, while it is only 2% among the non-finalists. Surprisingly, entrepreneurs with a master´s degree are not well represented among the successful contestants, while PhD´s are much more frequent in the successful group than in the non-successful group.

(23)

21

Table 3

Percentage or average and standard deviation of descriptives of the sample for successful, non-successful and total number of participants.

4.3 Accenture Innovation Award Questionnaire 2014

(24)

22 The survey questions of the AIAQ-2014 have been based on the Community Innovation Survey (CIS) survey, which indicates generalizability on the European level. Furthermore, the reliability and the content and construct validity of the questionnaire was assured by Accenture experts, therefore no further testing with regard to the reliability and validity of the questionnaire was conducted.

4.4 Variables

All variables included in the analysis, will be explained in the following. An overview of all included variables is provided in table 4.

4.4.1 Independent variable

The initial investment in an innovation is the independent variable in the regression model. One question of the AIAQ-2014 asked how many financial resources a company had invested in the participating innovation. The investments reach from €0,00 to €20.000.000,00. The distribution of the dataset was highly skewed towards the lower investment side as shown in figure 3. In order to transform the dataset towards a normal distribution, the following log transformation was used:

=LOG(investment+1)

Afterwards, the independent variable was transformed into Z values.

4.4.2 Dependent variable

The dependent variable is a binary variable, displaying the success of a participant in the AIA competition. A one (1) indicates that the firm reached the finals of the competition and is therefore seen as successful participant. The seven winning companies are also included in the group of successful participants. A zero (0) indicates that the firm participated, yet did not reach the finals. These firms are considered to be not successful.2 Of the 429 firms in the dataset, 31 are considered successful, while 398 are considered not successful.

4.4.3. Moderators

The factors assumed to moderate the relationship between the independent and the dependent variable are the financial barriers a firm encounters, the number of capital sources and the sources themselves (table 4). The sources are own capital, subsidies, bank loans, crowdfunding and venture capital.

2_{In the process of data analysis,}

(25)

23

Table 4

Overview of variables included in the analysis

Type Name Reference Scale Measure Questiona,b

Control (e)Retail Category - Nominal Dummy; 1 = yes, 0 = no - Consumer Products & Services Category - Nominal Dummy; 1 = yes, 0 = no 14.4 Travel & Transport Category - Nominal Dummy; 1 = yes, 0 = no -

Health Category - Nominal Dummy; 1 = yes, 0 = no 14.8

Public Services Category - Nominal Dummy; 1 = yes, 0 = no 14.9

High Tech Category - Nominal Dummy; 1 = yes, 0 = no 14.3

Communication Category - Nominal Dummy; 1 = yes, 0 = no 14.1

Media & Entertainment Category - Nominal Dummy; 1 = yes, 0 = no 14.2 Financial Services Category - Nominal Dummy; 1 = yes, 0 = no 14.5 Chemistry & Energy Category - Nominal Dummy; 1 = yes, 0 = no 14.6, 14.7 Sustainability Category - Nominal Dummy; 1 = yes, 0 = no 14.10 Firm Size (Number of Employees) Cohen & Klepper, 1996 Interval Log transformation 17 Type of Innovation Garcia & Cantalone, 2002 Nominal Dummy; 1 = start-up, 0 = no st. 4 Dependent Success Score Zhang et. al, 2012 Nominal Dummy; 1 = finalist, 0 = no fin. - Explanatory Initial Investment (independent variable) Bharadwaj, 2000 Interval Log transformation 25

Financial Barriers (internal & external) Buse et al., 2010 Interval 0-3 in 0,5 steps 40 Number of financial sources Lee et al., 2001 Interval Dummy; 1 = yes, 0 = no 29

Own Capital Bhide, 1992 Nominal Dummy; 1 = yes, 0 = no 29.5

Subsidy Almus & Czarnitzki, 2012 Nominal Dummy; 1 = yes, 0 = no 29.1 Crowdfunding Hellman & Puri, 2002 Nominal Dummy; 1 = yes, 0 = no 29.4 Venture Capital Hellman & Puri, 2002 Nominal Dummy; 1 = yes, 0 = no -

Bank Loan Bhide, 1992 Nominal Dummy; 1 = yes, 0 = no 29.6

(26)

24 Financial barriers are represented by a scale value between 0 and 3 in steps of 0,5. The number of financial sources a firm accessed and the moderating effects of each of these sources were tested separately. The sources are all coded as dummy variables, with 1 representing the use of the source and 0 not using it.

In order to perform a moderation analysis, cross-products of the independent variable and the moderating variables were developed. For that purpose, all moderators were standardized and Z values were created of all moderators, which were multiplied by the Z value of the independent variable, in order to display the moderating effect. The moderators were also added as independent variables to the regression model in order to display their direct effect on the dependent variable, and not as moderators.

As the number of used financial sources and the sort of financial sources themselves are per definition influencing each other (if a source becomes 1, the number of sources increases by 1), separate regression models will be displayed that show isolated effects, as well as the combined effects.

4.4.4 Controls

In the AIA, different categories were build based on innovation topics. The most important control variables used to improve the regression model are a set of dummy variables that indicate the categories in which a contestant participated, because the categories have different sizes. It is much easier for a contestant to reach the final of the competition in a small category (few contestants) than it would be in a large category (many contestants). Furthermore, it is controlled for the size of the participating organization by counting the number of employees. As the distribution of this measure was similarly skewed as the initial investment measure, a logarithmic transformation was used here as well. A visualization of the transformation can be found in figure 3. It is also controlled for the type of innovation by using a dummy variable that indicate whether an innovation is a product or not. Service and Business Model Innovations were combined as non-products in order to avoid collinearity problems. Several potential control variables were tested but then excluded from the model, as they did not contribute to the significance or goodness-of-fit of the data.3

3

(27)

25

Figure 3. Log transformation of initial investment (top) and number of employees (bottom).

4.5 Data Analysis

Models relating the success score to the independent predictors were developed by logistic regression, using the statistical software SPSS version 20.0 (IBM, Armonk, New York), with PROCESS as plugin (by A.F. Hayes).

(28)

26 using the enter method (p-of-entry < 0.1). The complete model, containing control variables, independent variables and moderators is estimated as follows:

Probability (reaching the final in the awarding competition = 1)

= β0 + β1 Cat. eRetail + β2 Cat. Consumer Products & Services+ β3 Cat. Travel & Transport + β4 Cat. Health + β5 Cat. Public Service + β6 Cat. High Tech+ β7 Cat. Communication + β8 Cat. Media & Entertainment + β9 Cat. Financial Services + β10 Cat Chemistry & Energy + β11 Cat. Sustainability + β12 Number of Employees + β13 Type of Innovation + β14 Initial Investment (IV) + β15 Cat. Financial Barriers (FB) + β16 Number of Financial Resources (NFR) + β17 Own Capital (OC) + β18 Subsidy (S) + β19 Crowdfunding (CF)+ β20 Venture Capital (VC) + β21 Bank Loan (BL) + β22 (IV x FB) + β23 (IV x NFR) + β24 (IV x OC) + β25 (IV x S) + β26 (IV x CF) + β27 (IV x VC) + β28 (IV x BL) +εi

εi = random error for each observation Cat. = Category

4.6 Research Quality Criteria

In order to assess the regression model and the generalization of the model to other samples, several research quality criteria were calculated. Also, interviews with an expert panel were conducted to fulfil the triangulation criteria, which was done to increase construct validity and reliability.

4.6.1 Assessing the regression model

The goodness of the fit of the regression model was assessed by the log-likelihood and the R-statistics. In order to assess whether the model fitted the data, the observed and the predicted values of the outcomes were compared using the log-likelihood. Large values of the log-likelihood indicate poorly fitting statistical models. Thus, the -2 Log-likelihood must be minimized with each change of the model, reaching the smallest possible value. To examine how much the model improves as a result of the inclusion of the predictor variables, the Hosmer and Lemeshow´s R-square, the Nagelkerkes R-square and the Cox & Snell R-square were calculated.

Wald statistics were used to assess the contribution of the individual predictors, which indicate whether the b coefficient for the predictor is significant different from zero. A significant contribution of the predictor to the outcome was set at a significant level of p<0.1.

4.6.2 Generalization of the regression model

(29)

27

4.6.2.1 Checking underlying assumptions of logistic regression

Three major assumption must be checked when performing a logistic regression in order to make the model generalizable, which are linearity, independence of errors and multicollinearity (Field, 2009). The assumption of linearity in logistic regression assumes that there is a linear relationship between the continuous predictors and the logit of the outcome variable. In this study, this was the linear relationship between initial investment and the success score. Linearity was tested by examining whether the interaction term between the predictor and its log transformation was significant.

The assumption of independence of errors assumes that the residual terms of two observations should be uncorrelated. This was tested with the Durbin-Watson test, which tests for serial correlations between errors and which is according to Field (2009) applicable in logistic regression analysis. A value of 2 is considered in the current study to indicate that the residuals are not correlated. Values below 1 or above 3 are taken as boundaries to indicate that the assumption had been violated (Field, 2009).

The assumption of multicollinearity states that the predictor variables should not correlate too highly, thus multicollinearity exists when there is a strong correlation between two or more predictors in a regression model. In the current study, possible multicollinearity was identified by scanning the correlation matrix of all of the predictor variables. Multicollinearity was indicated when the correlation was higher than 0.45. The assumption was violated when the correlation between predictors was higher than 0.8 (Field, 2009). Pearson correlation coefficients were calculated. The variance inflation factor (VIF) and the tolerance statistics (tolerance is 1 divided by VIF) were also examined to see whether a predictor had a strong linear relation with another predictor. Myers (1990) suggests that a VIF value higher than 10 indicates multicollinearity. A tolerance value below 0.1 indicates that the assumption of multicollinearity has been violated and a tolerance value below 0.2 indicates a potential violation (Menard, 1995).

4.6.2.2 Cross-validation of the model

Assessing the accuracy of a model across different samples is known as cross-validation. If a model

(30)

28 𝑎𝑑𝑗𝑢𝑠𝑡𝑒𝑑 𝑅2_{= 1 − [(} 𝑛 − 1 𝑛 − 𝑘 − 1) ( 𝑛 − 2 𝑛 − 𝑘 − 2) ( 𝑛 + 1 𝑛 )] (1 − 𝑅 2₎

R2 is the adjusted value, n is the number of participants and k is the number of predictors in the model. The Nagelkerkes R-square was taken as the R2.

4.6.3 Triangulation

In order to improve construct validity and reliability, a triangulation approach was chosen (van Aken, Berends & van der Bij, 2012). In this paper this construct was used to combine two research strategies to study the same phenomenon, one being the statistical analysis of the questionnaire and the other being interviews with experts in the field of awarding and finance. This strategy was not only used to improve construct validity and reliability, but also to enrich the discussion of this paper. Surveys and interviews have different advantages, combining them can therefore return complementary results (van Aken et al., 2012). The method chosen to analyse the interviews is the “Qualitative Data Matrix”, which will display the most relevant quotes of the interviewees linked to the construct to be analysed (Bijlsma-Frankema & Droogleever Fortujn, 1997).

(31)

29

5. Results

5.1 Descriptive Statistics

The descriptive statistics of all variables included in the regression analysis for the finalists (successful) and non-finalists (non-successful) are shown in table 5. Regarding the control variables, the descriptive statistics show that successful participants have a higher mean in the categories Chemistry & Energy, Consumer Products & Services, eRetail, Health, High Tech, Sustainability and Travel & Transport. Only in the categories Communication and Media & Entertainment were on average more non-finalists. This shows that winners tend to be more present in multiple categories, which could indicate that competing in more categories could lead to an increase in successfulness. Especially interesting is the High Tech category, because the mean of the successful participants is almost double the mean of the non-successful participants (0,32 to 0,14). This reveals that a lot of successful companies are competing in the High Tech category. Furthermore, successful companies have on average more employees, and have developed product innovations.

Descriptive statistics of the main effects show that successful firms invested on average more than non-successful firms, indicating that the initial investment could play an important role in the prediction of success. The investments of finalists are not only higher, they are also less variable (Lower S.D.) than those of non-finalists. The descriptives also show that successful contestants experience higher financial barriers and use more financial sources compared to non-successful contestants. The mean moderation effects are much higher for each of the moderators for successful firms, with the exception of the moderator own capital, which has a negative mean for finalists.

5.2 Correlations of Predictors

In order to better understand the data and its interrelations, a bivariate correlation analysis was performed, including all variables that are included in any of the models. An overview of all correlations between all variables can be found in table 6. This analysis focusses solely on significant relationships witch a Pearson Correlation Coefficient above 0,45 or below -0,45 as these are of most interest.

Results of the Pearson correlation show a positive correlation of 0,56 between the categories Chemistry & Energy and Sustainability (table 6). This can be explained by the nature of these categories; participants who developed an innovation related to energy are very likely also suitable for the Sustainability category.

(32)

30 while all else is coded as 0. As start-ups are likely to have fewer employees than other firms, this interaction is not surprising. For further analysis, it has been decided to focus on number of employees as control for company size, and exclude type of organization.

High correlations were found between the number of financial sources and the bank loan predictor (R=0,55), the venture capital predictor (R=0,42) and the subsidy predictor (R=0,64) (table 6). Similar effects have also been found between the same factors when treated as moderating effects. This was expected as each of the financial sources directly influences the value of the number of sources.

Interestingly, a negative correlation is observed between the crowdfunding moderator and the own capital moderator (R=-0,47). This could be explained by the increased use of crowdfunding campaigns when own capital is scarce.

Table 5

(33)

29

Table 6

Pearson Correlation Coefficients of all variables included in the regression model

(34)

30

5.3 Prediction model

The earlier stated hypotheses have been tested using a logistic regression analysis. Following the Resource-Based View (Barney, 1986a), hypothesis 1 states that a higher initial investment in an innovation has a positive impact on the participants success in the AIA competition. The independent variable “initial investment” is highly significant and displays positive B values from 0,829 to 1,286 across all models (Table 7)4. Hypothesis 1 is therefore supported.

Hypothesis 2 states that the relationship between the initial investment and the competition success will be positively influenced by the access to additional external financial sources such as bank loans, venture capital or crowdfunding. Table 7 shows different constellations of the regression model, with model 7, 8, 9 and 10 displaying the moderation effects of the financial sources. It can be noted that the B values of the moderators bank loan, crowdfunding and venture capital are consistently more positive or less negative than the moderators own capital and subsidy. Subsidies influence the relationship between the initial investment and the competition success negatively in model 8 and 10. Bank loans show a positive moderation effect in model 7 and 9 (B values between 0,613 and 1,115), and venture capital shows a positive moderation effect in model 9 (B = 2,371). However all of these findings are only significant on the p < 0,1 level. Only the own capital shows a p < 0,05 significance in model 8 (B = -1,009) and a p < 0,1 level significance in model 10 (B = -1,998). Hypothesis 2 can therefore be declared partially supported.

Hypothesis 3 states that a greater number of financial sources moderates the relationship between the initial investment and competition success positively. Most models display a positive influence of this moderator, yet only model 8 shows a p < 0,1 significance (B = 1,458). Surprisingly, model 9 even displays a non-significant negative B value. Hypothesis 3 can therefore not be supported.

Hypothesis 4 states that the relationship between monetary funding and AWC is influenced positively when firms do not experience financial barriers. Throughout all models there is no significant support for an interaction effect between financial barriers to innovation and the relationship between the initial investment and the success logistic, even though the B values were consistently negative. There is therefore no support for hypothesis 4. Surprisingly, models 4, 5, 6, 7, 9 and 10 showed a significant (p < 0,05) positive direct impact of experiencing financial barriers on the dependent variable (B between 0,575 and 0,802).

The control variables show that successful contestants where focusing significantly more on product innovation than non-successful participants (B between 1,942 and 2,186) among all models. It is also shown that contestants that were assigned to the awarding categories Consumer Products & Services (B between 0,91 and 1,212 among all models), High Tech (B between 0,1,177 and 1,178 in

4

(35)

31 models one and two) and Financial Services (B between 1,145 and 1,269 among all models) performed better than those in other categories.

5.4 Results of research quality analysis

5.4.1 Assessing the regression model

All research quality criteria stay within their as acceptable defined range, with the exception of a significant Hosmer-Lemesow test in model 3. This model is therefore not considered in the analysis. Furthermore, the analysis shows that all model descriptives level off in the models 7 to 10. This gives these last models equal meaningfulness in the analysis. The Omnibus test shows model significance for the models 7 to 10 (p < 0,001). The -2 Log Likelihood levels off at 152,576, and the R² lies between 0,151 (Cox & Snel) and 0,372 (Nagelkerke), showing that between 15% and 37% of the variance can be explained by the model.

5.4.2 Generalization

The generalization of the regression model has been tested checking the underlying assumption of logistic regression and by determining the cross-validation of the model.

5.4.2.1 Checking assumptions

The assumption linearity of the logit was tested by examining the interaction terms between the continuous predictor initial investment and the outcome variable success score. The interaction has a significance value greater than 0.05 (p = 0.43) which indicates that the assumption of linearity of the logit is not violated (Field, 2009). Concerning the independence of errors, the results of the Durbin-Watson statistics show that the value of the test statistic is 2.06, which indicates that the assumption of independence of error has been met. Multicollinearity was assessed by analysing Pearson Correlation Coefficients which are shown in table 6. Results show that there are no correlations higher than 0.8, showing that the assumption of no multicollinearity was not violated. This is also underlined by the values of the VIF (ranging from 1.06 to 1.78) and the tolerance statistics (values of the tolerance statistics ranged from 0.51 to 0.94).

5.4.2.2 Cross-validation of the model

(36)

32

Table 7

(37)

33

5.5 Summary of results

Table 8 gives an overview of the results by showing the hypotheses and whether they have been supported.

Table 8

(38)

34

6. Expert Panel

To support and challenge the results and therefore to increase the construct validity, interviews with experts from the fields of awarding and finance were held, in which the hypotheses and results were discussed. The results from these interviews are used to find explanations for unexpected effects, and to add different points of view to the discussion. By using a triangular approach (van Aken, Berends & van der Bij, 2012), a more complete set of findings can be achieved than with the administration of one of the methods alone (Bryman, 2011). In the following, results of the interviews are presented, categorized in the main topics of this study, which are in correspondence with the hypotheses.5 A qualitative data matrix is chosen as method of analysis, it can be found in table 9.

Table 9

Qualitative data matrix, containing relevant quotes of interviewees

Constructs Paul van Renslaar6 Arne Brix7

Influences of investments on AWC

“Generally, high investments should lead to a better performance in the competition; however it is important to take the type of innovation into account,

different types may need capital at different points in time.”

“Yes, I would assume that those who have invested more, will also be more successful. I would assume that those who invested more capital are also more

credible.”

Sources of Capital “Firms that have secured the support of a venture capitalist tend to be more successful because most of them will have

well defined goals and ambitious growth plans, as it is required by their investor.

The VC backed firms tend to professionalize quickly”

“Even though all of our clients are consulted and supported by us the best

we can, of course there will be more resources available for us to support them if they are larger clients or if more

money is at stake for us.”

Number of Financial Sources

“Yes, I would have assumed that the combination of different sources will lead

to a better performance. The more investors, the bigger the network, and

thus more valuable knowledge and opportunities for the firm.”

“I think that there could be advantages to multiple investors, but in many cases,

one strong partner will be the better option, because you will be able to build

a relationship, with more trust and a more efficient work flow than with more

partners” Financial Barriers “Your outcome is very interesting, I

proposed it myself when researching financial barriers, however I found the

opposite effect in my work using a different database.”

“I think limited access to capital will negatively influence the performance,

these kinds of restrictions tend to dampen firms abilities to develop.”

5

For access to a complete transcript of the interviews please contact the author at philipp.metzner@hotmail.de 6

The statements have been translated from Dutch into English 7

The Influence of Monetary Funding on Firm´s Innovation-Award Winning Chances: A Resource-Based Perspective