• No results found

The influence of the crowdfunding environment on the crowdfunding performance : A comparative study of the performance of Dutch and German projects launched on the Kickstarter crowdfunding platform

N/A
N/A
Protected

Academic year: 2021

Share "The influence of the crowdfunding environment on the crowdfunding performance : A comparative study of the performance of Dutch and German projects launched on the Kickstarter crowdfunding platform"

Copied!
52
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

The influence of the crowdfunding

environment on the crowdfunding

performance

A comparative study of the performance of Dutch and German projects launched on the Kickstarter crowdfunding platform

Master’s thesis

Joint master VU/UvA: Entrepreneurship

Name: Abe Hendriks

Student number: 10416587 (UvA) / 2508557 (VU)

Date: August 16, 2017

(2)

2

Table of contents

Preface ... 3 Abstract ... 4 Chapter 1. Introduction ... 5 1.1 Social relevance ... 5 1.2 Academic relevance ... 5 1.3 Objective………. ... 6

Chapter 2. Literature review ... 7

2.1 The concept of crowdfunding ... 7

2.2 Kickstarter…….. ... 10

2.3 Emerging crowdfunding markets ... 15

Chapter 3. Theoretical framework ... 20

3.1 Influencers of the success of a campaign ... 20

3.2 Crowdfunding environment ... 21

3.2.1 Social ... 22

3.2.2 Cultural ... 23

3.2.3 Economic ... 25

3.2.4 Political ... 27

Chapter 4. Research design ... 29

4.1 Data Collection ... 29

4.2 Data Preparation ... 30

4.3 Operationalization of the variables ... 32

Chapter 5. Results... 37

5.1 Descriptive data ... 37

5.2 Correlations………… ... 39

Chapter 6. Discussion ... 42

6.1 Discussion………… ... 42

6.2 Limitations and future research ... 43

Chapter 7. Conclusion ... 45

References ... 46

(3)

3

Preface

This document is written by Abe Hendriks who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it. The faculty of Economics and Business is responsible solely for the

(4)

4

Abstract

Despite the rising popularity of crowdfunding in both academic research and practice, there is a research gap in cross-country comparisons, although several studies suggest the influence of country-specific factors. Thus, in this research is tried to answer the following research question: Is the difference in crowdfunding performance between the Netherlands and Germany explained by the crowdfunding environment? To answer this question, this study investigates data from campaigns in Europe on the largest crowdfunding platform Kickstarter. The crowdfunding environment of a country is determined as a combination of factors: social (trust), cultural (contribution of friends and family), economic (GDP per capita) and political (perceptions of regulations). The crowdfunding performance of a country is measured by using the average rate of successful campaigns and the amount of money pledged per capita per year. There is found no support from the data to claim the presence of a relationship between the determinants of the crowdfunding environment and the crowdfunding performance per country.

(5)

5

Chapter 1. Introduction

1.1 Social relevance

Raising external capital is one of the main obstacles for entrepreneurs given the absence of sufficient cash flow and the presence of information asymmetry amongst investors (Cosh, Cumming, & Hughes, 2009). Since the financial crisis in 2008, traditional financial institutions have been less keen on investing their money in projects perceived as riskier, including new venture development (Bruton, Khavul, Siegel, & Wright, 2015). There are several scholars, including Giudici, Nava, Rossi Lamastra, and Verecondo (2013), Mollick (2014) and

Schwienbacher and Larralde (2010), who recognize the use of the internet, and crowdfunding in particular, as the answer to the gap created by the financial crisis. The concept of crowdfunding is rapidly developing and it is becoming more popular amongst entrepreneurs as an alternative venture financing source (Mollick, 2014). Wardrop, Zhang, Rau, and Gray (2015) examine and compare the differences in crowdfunding performance between countries. Unfortunately, their article does not look into the drivers of these differences between countries, which could be of interest for both investors and entrepreneurs who wish to acquire financial resources for the development of new ventures. In continental Europe, both the Netherlands and Germany are doing well in terms of overall crowdfunding performance (Wardrop et al., 2015). In terms of funding, Germany has almost twice as much alternative finance transactions in 2014, while the Netherlands performs better when looking into the alternative finance volumes per capita. This article provides insight into the differences between the Netherlands and Germany. The results of this paper provide both investors and entrepreneurs with a framework for successfully starting a Kickstarter campaign.

1.2 Academic relevance

As mentioned above, it is becoming popular amongst entrepreneurs to use crowdfunding as a way to finance their ideas. More scholars have also begun to research crowdfunding (Agrawal, Catalini, & Goldfarb, 2011; Giudici et al., 2013; Kuppuswamy & Bayus, 2015; Mollick, 2014; Wardrop et al., 2015). Different publications have recognized that crowdfunding is still

developing and has not yet matured as a mainstream financing technique (Ordanini, Miceli, Pizzetti, & Parasuraman, 2011; Wardrop et al., 2015). The majority of the current literature focusses on the internal characteristics of campaigns (Cumming, LeBoeuf, and Schwienbacher,

(6)

6

2014; Mollick, 2014; Schwienbacher & Larralde, 2010). Other scholars have investigated the underlying reasons why supporters choose to invest in crowdfunding ventures (Agrawal et al., 2011, Agrawal, Catalini, & Goldfarb, 2013; Ordanini et al., 2011). Only a few scholars have examined the country-specific environment in which crowdfunding takes place (Zheng, Li, Wu, & Xu, 2014; Cho & Kim, 2017). This study aims to fill the gap in the literature on the

crowdfunding environment by conducting a cross-country comparison of crowdfunding performance. Germany and the Netherlands have several economic, political, and cultural similarities. They are both innovation- and export-driven economies (Global Entrepreneurship Monitor Consortium, 2017; OECD, 2017). They are the founding fathers of the European Union and active participants in international political organizations such as NATO, OECD, and others. However, as mentioned above, there are differences in their crowdfunding performance.

Therefore, in this paper, we take a closer look at the crowdfunding environments of these countries to gain additional insights.

1.3 Objective

Little scholarly attention has been paid to the influence of country-specific factors as

determinants of crowdfunding performance. In related areas of study, including entrepreneurial finance and entrepreneurship, this type of study is more common and provides insights for both academicians and professionals. For example, the Global Entrepreneurship Monitor Consortium (2017) recognizes differences towards entrepreneurship in countries. The aim of this study is to develop insights into the role of the crowdfunding environment on the crowdfunding

performance of a country. This study contains a combination of the literature review and a

correlation analysis on cross-country data from the crowdfunding platform Kickstarter in order to answer the main question: Is the difference in crowdfunding performance between the

(7)

7

Chapter 2. Literature review

2.1 The concept of crowdfunding

History and development of crowdfunding

The term crowdsourcing implies the use of the crowd to develop business activities and obtain human resources such as ideas, feedback, and solutions (Schwienbacher & Larralde, 2010). There are several examples of organizations and projects that are based on the concept of crowdsourcing, such as Wikipedia and the open-source operating system Linux (Howe, 2006). While crowdsourcing is about obtaining human resources, crowdfunding is used by the

entrepreneur to obtain financial resources from a group of people that can be used as an

investment (Schwienbacher & Larralde, 2010). Social networks like LinkedIn and Facebook play a huge role in reaching such groups of investors (Hui, Greenberg, & Gerber, 2014; Lu, Xie, Kong, & Yu, 2014). This type of fundraising is different from the more conventional method of raising large investments from a small group of investors. The idea behind crowdfunding is that the money for a project comes from a larger group of individuals, each of whom contributes a smaller amount of money (Belleflamme, Lambert, & Schwienbacher, 2014).

Different models of crowdfunding

The different models of crowdfunding differ mainly in what investors will receive for their donation (Hemer, 2011). The different ways to establish agreements between investors and entrepreneurs can mainly be categorized in four models: 1. Lending model; 2. Equity model; 3. Donation model; 4. Reward model. These different types of crowdfunding differ in what

investors will receive for their donation (Hemer, 2011). These different models include a specific role for the investors, whether they are actively involved in the development of the project or passively.

1. Lending model

Harrison (2013) explains that the lending model is focussed on business development over product development. In the lending model, the investors expect a return on investment, but the type of interest depends on how the model is interpreted (Harrison, 2013). The interest rate can be determined by project owners, investors, or even the crowdfunding platform. The lending model has many similarities with the concept of microfinance because the project owners are commonly unable to connect with other types of finance (Schwienbacher & Larralde, 2010).

(8)

8

Whether this model can be divided into an active or passive investment depends on the conditions applied by the project owners, investors, and mainly the crowdfunding platform.

2. Equity model

The equity model entails the project owners offering the investor a stake in the business or a stake in the future earnings of the project (Harrison, 2013). The behaviour of investors participating in projects that follow the equity model can be expected to be quite similar to rational investors because they have the same goal. Crowdfunding platforms that adhere to the equity model are subject to many regulations, such as traditional laws, because they sell shares as a financial product (Mollick, 2014). Use of the equity model is increasing rapidly, and many different scholars predict that this development will continue (Belleflamme, Lambert, &

Schwienbacher, 2013; Mollick, 2014; Schwienbacher & Larralde, 2010). One of the similarities between traditional investing and this type of crowdfunding is the active involvement of

investors in the project, as seen in the framework of Schwienbacher and Larralde (2010).

3. Donation model

The donation model is the most simplistic model. In this case, investors get the opportunity to support a project by offering a (small) amount of money without the expectation of receiving a return on the investment (Harrison, 2013; Schwienbacher and Larralde, 2010). These projects are mainly, but not exclusively, non-profit and charity projects. Therefore, this type of crowdfunding is also called charity crowdfunding because it relies completely on charitable donations

(Harrison, 2013). The donors help by willingly giving the money as support, while the investors take the position of philanthropists and bear little resemblance to traditional investors (Mollick, 2014). Another branch of the donation model is civic crowdfunding. In comparison to the

donation model, in this model, investors get something in return, namely something that is useful for their community. This method is mainly used by a combination of citizens and local

governments when the governments lack the funds to realize a project of public interest (Stiver, Barroca, Minocha, Richards, & Roberts, 2015). Another branch within the donation model is social crowdfunding, which is similar to civic crowdfunding, although the investors’ return is intangible and linked to a solution or project that solves a social problem (Lehner, 2013).

(9)

9 4. Reward model

This reward-based type of crowdfunding is the most popular model; one of its main perceived advantages is that it allows the entrepreneur to maintain full ownership of the company (Mollick, 2014). The main difference between the reward model and the donation model is the (tangible) reward that investors get for their investment in the reward model (Schwienbacher & Larralde, 2010). This type of investing is a passive investment, because the investor is not actively

involved in the process. The difference between passive investments and donations is that in the reward model, investors receive rewards for passive investments.

The two largest reward-based crowdfunding platforms are Kickstarter and Indiegogo. In this research, the focus is on the data of Kickstarter, thus the details of this model will be explained further. The reward model is often perceived as the pre-purchase models, so the investors are perceived as customers. From this point of view, the investors pay money to the project owner in order to ensure the project as one of the first or with a special discount in return (Mollick, 2014). In relation to laws and policies, reward-based crowdfunding platforms are therefore often

perceived as web shops, instead of platforms offering a financial transaction. In the reward model, investors provide monetary support in return for a reward that does not any type of stake in the company. As the projects do not always include products, the type of rewards can take many different forms. For example, the rewards for an investment could exist out of a meeting with the project owners, or providing input for the project. This could lead to a situation where the investors play a key role in the development.

All-or-nothing vs. Keep-it-all

Different crowdfunding websites, like Indiegogo and Kickstarter, use different structures with regards to their supporters. Cumming et al. (2014) explain the difference between all-or-nothing (AON) and keep-it-all (KIA) structures. There are important differences between these structures that influence the decisions made by the project owner and how they behave with respect to their supporters.

In the AON model, also known as the threshold pledge model, the investors make a pledge to invest a specific amount of money in the project. No actual payment is made to the project owner right away, but a signed contract allows the project owner to withdraw the money when the financial goal of the project is achieved within a certain amount of time (Cumming et al., 2014,

(10)

10

Tomczak & Brem, 2013). If the project owner fails to reach the financial goal, the money is not sent to the project owner and is deposited back to the investors by the crowdfunding platform. If the project owner exceeds her goal, she could have more opportunity to add more features to the product or conduct organizational renovations because she has to apply the extra money and provide the promised returns (Hemer, 2011; Tomczak & Brem, 2013). The AON model is mainly associated with donation and reward models (Hemer, 2011).

In the KIA model, the target amount does not have to be reached by the project owner in order to receive the investments. In this case, the risk is not taken on by the entrepreneur, but is shifted to the investors, who are locked into investing once they make the pledge (Cumming et al., 2014; Tomczak & Brem, 2013). According to Cumming et al., (2014) this model is more successful when the financial goal is low, as opposed to AON projects.

2.2 Kickstarter

Kickstarter was established in April 2009 and is considered the largest reward-based platform

based on the amount of capital raised by their campaigns1. As Kickstarter claims on its website,

its goal is to help different kinds of creators find the resources and support they need to shape their creative ideas into reality1. These different kinds of creativity are divided into 15 categories,

e.g. fashion, art, music, and technology.1 Since 2009, more than $3.1 billion has been pledged by

13 million different backers on Kickstarter for more than 125,000 projects.1 Kickstarter project

owners need to specify the duration of their campaign. When the goal is met, the owner is legally obligated to complete the project and distribute the rewards. When the goal is not met, the

project owner is obligated to refund their backers.2 Therefore, Kickstarter’s business model is

based on multiple fees; For example, the company collects 5% of the total pledged amount if the goal is met or exceeded.2 Back in 2009, it was only possible for residents of the United States to participate and create projects, but since October 2012, residents of a select group of other countries have also been allowed to participate. The UK was the first country to be added, as shown in Table 1.3

1https://www.kickstarter.com/about?ref=nav (Retrieved on August 5, 2017). 2

https://www.kickstarter.com/terms-of-use?ref=footer (Retrieved on August 5, 2017)

3

https://www.kickstarter.com/blog/welcoming-austria-belgium-italy-luxembourg-and-switzerland-to-th (Retrieved on August 5, 2017).

(11)

11

Table 1

Kickstarter launch dates of European countries

Country Date Austria 16-6-20153 Belgium 16-6-20153 Switzerland 16-6-20153 Germany 12-5-20154 Denmark 21-10-20145 Spain 2-6-20156 France 27-5-20157 United Kingdom 31-10-20128 Ireland 21-10-20145 Italy 16-6-20153 Luxembourg 16-6-20153 The Netherlands 29-4-20149 Norway 21-10-20146 Sweden 21-10-20146 Characteristics

As stated by Schwienbacher and Larralde (2010), on platforms like Kickstarter, the projects need to possess certain characteristics in order to be considered crowdfunding projects. The project characteristics are there to inform the crowd about the project and the project owners.

Number of backers

4

https://www.kickstarter.com/blog/hallo-deutschland (Retrieved August 11, 2017)

5https://www.kickstarter.com/blog/kickstarter-in-scandinavia-and-ireland (Retrieved August 11, 2017) 6https://www.kickstarter.com/blog/kickstarter-ya-esta-disponible-en-espana (Retrieved August 5, 2017 7https://www.kickstarter.com/blog/kickstarter-est-arrive-en-france (Retrieved August 5, 2017)

8

https://www.kickstarter.com/blog/kickstarter-in-the-uk (Retrieved on August 5, 2017)

9

https://www.kickstarter.com/blog/kickstarter-in-the-netherlands-0 (Retrieved on August 5, 2017)

(12)

12

The number of backers is the total number of individuals who have supported a specific project (Cumming et al., 2014). This number is not restricted to a specific maximum, and every backer is allowed to fund each project at whatever perk level they desire. Kuppuswamy and Bayus (2015) claim that backers are more likely to support new and low funded projects, because backers think that the projects that have already achieved 80% of their goal, will meet their goal also without their support. Besides this, Kuppuswamy and Bayus (2015) also argue that backers feel less responsible in the early stages of a project, so while it is more likely that the backers will support the project, the investment is more likely to be less.

Campaign success

Campaign success depends on the ratio of the pledged amount to the campaign goal. Cumming et al. (2014) claim that AON projects usually aim for higher amounts of money and have higher campaign success rates compared to KIA projects. In contrast, other scholars claim that larger projects are less successful (Belleflamme et al., 2013; Mollick & Kuppuswamy, 2014). Mollick (2014) suggests that it is the quality of the project that is linked to the success rate. Success is defined as reaching the financial goal and receiving the investment.

Financial goal

The financial goals of crowdfunding projects can vary greatly depending on the projects’ ultimate goal and magnitude. There are both small campaigns for art projects and non-profit projects and large campaigns that employ crowdfunding as an alternative to venture capitalist funding (Schwienbacher & Larralde, 2010). The financial goal is defined as the amount of money the entrepreneur determines he wants to raise before the start of the project (Mollick & Kuppuswamy, 2014).

Additionally, other characteristics such as perk levels, comments, project category, project duration, reward type and title length are perceived as important characteristics of a project (Cumming et al., 2014; Kuppuswamy & Bayus, 2015; Mollick, 2014; Mollick & Kuppuswamy, 2014). Within these characteristics, the importance for this research is less than the mentioned and described characteristics.

(13)

13 Phases of a campaign

There are three different phases to a crowdfunding campaign (Ordanini et al., 2011), starting with friend funding. This first phase consists of investments made by people who are personally related to the project owner and usually accounts for half of the projected financial goal. The entrepreneur’s main focus in this phase is to reach out to people she knows to create a safe basis for the remaining part of the process. The second phase is called “getting the crowd” by Ordanini et al. (2011). In this phase, investment growth threatens to slow down. This is one of the most important phases of a crowdfunding campaign because it typically determines whether the project will be successful. Therefore, it is crucial to motivate and involve people in this phase to reduce the slowing of investment. Inability to motivate and involve people is a common reason for failure (Ordanini et al., 2011). When funding is not increasing, the project faces multiple problems and becomes a higher risk investment. Many projects stagnate in this phase. At the end of the second phase, the development of the funding determines whether a project reaches the “the engagement moment” according to Ordanini et al. (2011, p.457), as illustrated in Figure 1. If a project reached this moment, it will enter the last phase: “the race to be ‘in’ ”. This phase is characterized by a chain reaction that leads to rapid growth and the reaching of the financial goal. Investors in this phase are mainly people without a direct connection to the project. Therefore, Ordanini et al. (2011) claim that the type of investors in this stage act similarly to regular investors in financial markets; they do not want to miss the opportunity.

(14)

14 Figure 1. Typical path of consumer investment via a crowdfunding platform (Ordanini et al., 2011)

(15)

15 2.3 Emerging crowdfunding markets

The focus of this paragraph is on the growth of crowdfunding in Europe. First, the financial growth of the concept will be described, where after the development of a number of successful campaigns and active backers will be examined. Finally, the Alternative Finance Maturity Index in the Crowdfundinghub report: ‘Current State of Crowdfunding’ (Roodink et al., 2016) will be described in order to provide some insights before constructing the theoretical framework.

Growth of crowdfunding in Europe

Since the onset of crowdfunding, several scholars have conducted research on the development of crowdfunding in specific regions, but there is a lack of research to the crowdfunding market in the European Union (De Buysere et al., 2012). In 2014, Cambridge University and EY decided to combine their strengths, and Wardrop et al. (2015) conducted extensive research on

crowdfunding in Europe. According to these scholars, there has been serious growth in the European crowdfunding market over the past five years. Wardrop et al. (2015) claim that €487 million was raised through crowdfunding in 2012. In the next year, the size of the market increased by 149% to a total of €1.221 billion funded by crowdfunding in Europe. This

spectacular growth rate persisted through 2014, when the size of the market increased to €2.957 billion. Wardrop et al. (2015) expect this growth to continue. An interesting trend emerges when comparing the growth in Europe with that in other regions. According to Massolution (2015), although the market as a whole is rapidly developing, Europe has lost a serious share, mainly due to the extensive growth in Asia.

The researchers from Cambridge University and EY concluded that the United Kingdom is far ahead of continental Europe with regards to crowdfunding, as depicted in Figure 2 (Wardrop et al., 2015, p.16).

(16)

16 Figure 2. Volume of alternative finance transactions in 2014

Due to the geographic locations of the Netherlands and Germany, the number of successful campaigns and the number of active backers rapidly developed, as illustrated in Table 2 and 3, respectively (Wardrop et al., 2015). These two variables, successful campaigns and active backers, could be viewed as signals of the rapid development of the concept of crowdfunding as a whole. An interesting side note made by Wardrop et al. (2015) is the absence of accurate data, thus the numbers in Table 2 and 3 could be somewhat inflated. Even though these statistics may be inflated, there is no doubt that there has been an enormous evolution on both the demand side and the supply side in the crowdfunding market, such that the market seems to be far from maturing (Wardrop et al., 2015).

2 3 3 7 1 5 4 1 4 0 1 0 8 7 8 6 2 2 2 1 7 1 2 8 .2 4 3.6 2.5 2.5 2 1

VOLUME OF ALTERNATIVE

FINANCE TRANSACTIONS IN 2014

(17)

17

Table 2

Number of successful campaigns in continental Europe

Year Successful campaigns Growth rate

2012 75.583 -

2013 206.704 123%

2014 348.241 68%

Note. Adapted from: “Moving mainstream: The European alternative benchmarking report.”, by Wardrop, Zhang, Rau, & Gray, 2015, p.21-23.

Table 3

Number of active backers in continental Europe

Year Active backers Growth rate

2012 421.741 -

2013 893.330 119%

2014 1.510.000 69%

Note. Adapted from: “Moving mainstream: The European alternative benchmarking report.”, by Wardrop, Zhang, Rau, & Gray, 2015, p.21-23.

Looking closely at the statistics provided by Wardrop et al. (2015) and interpreting the numbers on another scale, namely volume per capita per country, there is a recognizable difference between the countries, as presented in Figure 3 (Wardrop et al., 2015, p.16). From the

perspective of total money raised, Germany has raised €140 million and is 3rd in Europe, and the

Netherlands has raised €78 million and is 5th in Europe. However, when comparing countries

with respect to volume per capita, the Netherlands moves up to 4th place with €4.6 per capita,

and Germany drops to 8th place with €1.7 per capita.

Summarizing, it must be noted that the amount of pledged money is higher in the Netherlands than in Germany when looking into per capita, while the total amount of money as pledged is higher in Germany than in the Netherlands.

(18)

18 Figure 3. Alternative finance volume per capita

According to the Alternative Finance Maturity Index (AFMI) of Crowdfundinghub (Roodink et al., 2016), both the Netherlands and Germany perform well in crowdfunding. This index makes no distinction between the different crowdfunding models, but functions mainly as an overview of the crowdfunding environment per country. In order to rank performance, the researchers used a three-point Likert scale from positive to negative to compare each country to other countries. The AFMI divides the research areas into two categories: general and regulations. The

Netherlands is ranked 2nd by the AFMI, and Germany is ranked 5th. The Netherlands and

Germany differ in three topics in the regulatory research area: degree of organization, volumes, and registration obligations. The Netherlands’ performance is perceived more positively than Germany’s (Roodink et al., 2016).

Degree of organization

In the majority of countries there is no data available on the total volumes of the crowdfunding industry, mostly because there is a low level of organization in the industry. In the United Kingdom, the Netherlands, and France, data on the total volume of the industry are collected, structured, aggregated, and reported on a yearly basis. In some countries, e.g. Germany,

aggregated data are only available with respect to volumes in certain subsectors of the industry.

3 6 1 6 .7 1 0 .9 4 .6 3 .1 2 .4 1 .9 1 .7 1 .5 1 .3 0 .4 0.4 0 .2 0.2 0.2 0.2

ALTERNATIVE FINANCE VOLUME

PER CAPITA (IN EURO)

(19)

19

It is important to note that there is increasing disagreement about precise volumes in countries because reporting is often done by sources closely linked to the industry.

Volumes

Because of the lack of aggregated data in the majority of countries, it is difficult to create an exact ranking of countries based on volumes. Many countries have reported that volumes are growing fast, in some cases more than doubling each year. Based on 2014 volumes, Wardrop et al. (2015) have reported that the UK ranked 1st, and France ranked a distant 2nd. In 2015, the UK and France were indisputably still 1st and 2nd, respectively, in terms of volume. Germany,

Sweden, and the Netherlands ranked 3rd, 4th, and 5th. We have no reason to believe that in 2015, these countries did not rank in the top 5, but because of the lack of data (in Germany and Sweden); we cannot provide a precise ranking. In terms of volume per capita, Germany and France rank lower than Sweden and the Netherlands, and Estonia has on average a high volume per capita, although most of it derives from P2Pconsumer lending.

Registration obligations

France and the Netherlands recently made implemented a registry and made it compulsory for lending- and equity-based crowdfunding platforms to register at the financial regulatory authorities. More than 60 platforms are registered in both countries.

(20)

20

Chapter 3. Theoretical framework

3.1 Influencers of the success of a campaign

The success of crowdfunding is often perceived as reaching a financial goal within a

predetermined timeframe—the campaign duration—, although there are also other measures of success, like number of backers, total amount of funding, and speed of funding (Ahlers,

Cumming, Günther, & Schweizer, 2015).

Most of the commonly referenced success measures are directly related to the characteristics of the project itself, thus these factors could be perceived as the internal factors of the campaign (Belleflamme et al., 2013; Mollick, 2014). Previous literature commonly cites duration of a campaign and goal amount as the most important drivers of the success of a crowdfunding campaign. Several studies have stated that the optimal duration of a campaign is about 30 days (Belleflamme et al., 2013; Frydrych, Bock, Kinder & Kock, 2014; Mollick; 2014) and laid out the specifics of appropriate goal amounts (Beaulieu, Sarker, & Sarker, 2015; Belleflamme et al., 2013).

Besides these internal factors, some scholars have recognized the influence of the crowdfunding environment on the success of a campaign. These factors seem to be less useful as tools for identifying differences between individual campaigns, but are better for observing patterns amongst different environments (Dushnitsky et al., 2016; Belleflamme, Omrani, & Peitz, 2015). Scholars that have written about the crowdfunding environment, like Agrawal et al. (2011), Agrawal et al. (2013), claim that it is composed of several aspects including: participants— including backers and project owners—, the digital platform, the regulatory framework, the economy, trust, social network, and information quality. In addition to the aforementioned aspects, a report by infoDev (2013) commissioned by the World Bank posits that social, cultural, economic, and technological factors are the main drivers behind crowdfunding. In their view, economic drivers are mainly influenced by laws and regulations. For a long time, socio-cultural, economic, and political drivers have been considered to have an influence on the entrepreneurial environment (GEM, 2016; Spigel, 2015). Scholars including Feller et al. (2013) and Peng and Nunes (2007) have linked the entrepreneurial environment and the crowdfunding environment in order to compare the crowdfunding environments of different countries. Therefore, we use the

(21)

21

social, cultural, economic and political factors to develop insight into the differences between the crowdfunding environments of the Netherlands and Germany.

3.2 Crowdfunding environment

The following paragraphs describe various ways to measure the factors that influence the crowdfunding environment, as visualised in Figure 4. Firstly, the social factor is described according to the current literature on the relationship between social capital and crowdfunding. Secondly, the cultural factor is examined using current literature on the contribution of friends and family and Hofstede’s cultural dimensional framework. Thirdly, the economic factor is described in terms of country-specific economic factors, specifically the gross domestic product purchase power parity (GDP PPP). Lastly, the political factor is described in terms of the perceptions on laws and regulations.

Figure 4. Conceptual model

Crowdfunding environment Social Cultural Economic Political H1 H2a/b H3 H4

(22)

22 3.2.1 Social

“Crowd funding works best if there is engagement, which means that relational and social capital is being transformed into financial and innovation capital” (Iske, 2010, p.15). Scholars have argued for a long time that entrepreneurship is closely related to the social environment because the phenomenon of entrepreneurship is social as well as economic (Berger, 1991,

Shapero & Sokol, 1982; Steyaert & Hjorth, 2008). Zheng et al. (2014) argue that the attraction of financial resources for entrepreneurial ventures is especially influenced by social capital. Several studies have examined the relationships between signals of social capital and crowdfunding success (Giudici et al., 2013, Kuppuswamy & Bayus, 2015; Mollick, 2014) and concluded that crowdfunding backers are more than just a random crowd. They are a specific group of

customers with similar social and cultural values that are reflected in their crowdfunding behaviour (Calveri & Esposito, 2013). In order to derive backers from the crowd, entrepreneurs must use social networks. The number of people in a social network is recognized as a positive influencer of successful projects (Mollick, 2014). Therefore, both funders and project owners play a vital role in the crowdfunding process that involves social networks, giving feedback, and providing help and financial resources (Hui, Greenberg, & Gerber, 2014). However, including social networks in one’s business model is not easy and relies mainly on trust (Gerber & Hui, 2013), especially because of the risk of crowdfunding. Therefore, trust can be considered a key dimension of the social factor.

When comparing the Netherlands and Germany, Knack and Keefer (1997) found that the Netherlands performs better in terms of social capital and economic performance. According to Knack and Keefer (1997), one of the main drivers of the Netherlands’ economic success is trust. Trust potentially can influence the economic performance of a country through two channels: “micro-economic” and “macro-political”. At the micro-economic level, the interpersonal trust can “reduce transaction costs, enforce contracts, and facilitate credit at the level of individual investors” (Knack & Keefer, 1997, p.1). In more recent research, trust has re-emerged as a parameter of social capital in crowdfunding (Zheng et al., 2014; Giudici et al., 2013; Agrawal, Catalini, & Goldfarb, 2015). According to Eurostat (2013), people in the Netherlands are more trusting than people in Germany; The Netherlands scores 6.9 on a 10-point Likert scale

measuring trust, and Germany scores only 5. Therefore, we investigate the relationship between trust, social capital, and crowdfunding performance in greater depth.

(23)

23 H1: If the level of trust in a country is higher, the crowdfunding performance of this country is

better.

3.2.2 Cultural

According to infoDev (20133), crowdfunding is influenced by the entrepreneurial culture in a country, which in turn is influenced by the national culture of a country (GEM, 2016). Therefore, national culture may have an influence on a country’s crowdfunding performance. Cho and Kim (2017) argue that national culture influences the crowdfunding environment and examine the relationship between Hofstede’s dimensional framework of national culture and crowdfunding performance.

Hofstede’s dimensional framework

Hofstede (2011) claims that it is possible to identify a group by looking at the interaction between commodities and the environment. Since 1980, Hofstede has conducted research on different national cultures and has set up a framework with which to measure and quantify culture. Hofstede (1984; 1991; 2011) has identified six dimensions according to which cultures can be measured and understood: power distance (PDI), uncertainty avoidance (UAI),

individualism (IDV), masculinity (MAS), long-term orientation (LTO), and indulgence (IVR). Hofstede’s dimensional framework has been used by numerous scholars to compare national cultures (i.e. Hayton, George, and Zahra [2002]). Hayton et al. (2002) claim that cultures that value risk-taking and independent thought, for example, are more likely to develop and introduce radical innovation. In contrast, cultures that value conformity and control of the future are less likely to exhibit independent thinking and risk-taking behaviours.

It must be noted that Hofstede’s framework has been subject to criticism. McSweeney (2002) prefers to use qualitative techniques to assess the abstract subject of culture rather than

quantitative techniques. Javidan et al. (2006) focus their critique on the bias in Hofstede’s data sample. They argue that because the data are derived from the specific context of IBM

employees in the United States, they cannot be generalized. Despite these and other criticisms, Hofstede’s framework is still the dominant model used in cross-cultural research.

Cho and Kim (2017) use Hofstede’s dimensions to compare the United States and South Korea, but they exclude the dimensions used in this research: masculinity, indulgence, and long-term

(24)

24

orientation. Cho and Kim (2017, p.314) argue that they exclude masculinity because “gender roles are rarely explicated in crowdfunding and fundraising contexts, and there are some overlaps between the power distance and masculine/feminine dimensions.” However, we argue that the masculinity dimension pertains to more than just gender roles and therefore have not excluded it from this research. Cho and Kim (2017) also exclude long-term orientation and indulgence because Fang (2003) challenged these dimensions due to their philosophical flaws and lack of practicability. We agree and see little difference between these dimensions, as presented in Figure 5. Thus, we have also excluded them here.

Figure 5. Hofstede’s dimensional framework—The Netherlands compared to Germany (Hofstede, 2011).

The degree of masculinity reflects a society’s preference for “achievement, heroism, assertiveness, and material rewards for success” (Hofstede, 1991 in Cumming, LeBoeuf, & Schwienbacher, 2017, p. 26). A country that is more feminine focusses more on balancing life, friends, family and quality of life. The roles of males and females in the business context are also included in this masculinity value (De Mooij & Hofstede, 2011). It is expected that the

masculinity value influences how friends and family behave in the context of crowdfunding. The 38 80 14 53 67 68 35 67 66 65 83 40 0 10 20 30 40 50 60 70 80 90 100

Power distance Individualism Masculinity Uncertainty avoidance

Long-term orientation

Indulgence

(25)

25

Netherlands scores 14 (low) on this dimension. Thus, we predict that Dutch family and friends would be more likely to support a crowdfunding effort than German family and friends. Germany scores 66 (high) on the masculinity dimension. In summary, the current literature suggests that because the Netherlands is a more feminine society, the Dutch are more likely to support friends and family in their crowdfunding campaigns.

H2a: The crowdfunding performance in a country is better when there is a higher contribution of

friends and family.

H2b: The contribution of friends and family for crowdfunding is higher in countries with a lower

level of masculinity

3.2.3 Economic

One of the most important and most used measures of social and economic well-being is GDP per capita. GDP per capita is used by the World Bank and Eurostat (Eurostat, 2017; World Bank, 2017). GDP per capita: the gross domestic product divided by the total number of inhabitants is seen as an objective economic measure (i.e. Barro, 1991). In this study, a derivative of the GDP per capita is used: purchase power parity. This measure is used in this study to investigate the relationship between the economic well-being of a country and crowdfunding performance. The definition of purchase power parity is usually credited to Gustav Cassel (1918, in Reinert, Rajan, Glass, & David, 2009) who stated: “when measured in the same unit, the monies of different countries should have the same purchasing power and command the same basket of goods. Otherwise, international arbitrage should bring about adjustments in prices, exchange rates, or both, which will ultimately restore parity.”

The gross domestic product (GDP) purchase power parity (PPP) per capita (PC) in the

Netherlands and Germany are quite different, as the average GDP PPP PC of the Netherlands is between 4.5% and 6.5% higher than in Germany (International Monetary Fund, 2017; World Bank, 2016; Central Intelligence Agency, 2016). The following table presents economic data that clarify the differences between the Netherlands and Germany.

(26)

26

Table 4

Economic data related to the Netherlands and Germany

2016 Source Netherlands Germany

Population (million) * 17,02 82,67

GDP ($, billion) * 770,845 3466,75688

GDP per capita (EU-28 = 100) ** 128 123

GDP per capita ($) * 40,948 37,912

GDP PPP per capita ($) * 50,898 48,730

Economic growth (GDP, annual variation, %) ** 0,9 1,2

Consumption (% of GDP) ** 43,6 51,8

Exports of goods and services (% of GDP) * 82,4 46

Imports of goods and services (% of GDP) * 71,4 38,4

Note. From “World Development Indicators database”, by World Bank, 2016 and “Europe in Figures: Eurostat Yearbook 2016”, by Eurostat, 2016.

Burtch, Ghose, and Wattal (2013), among others, have considered GDP PPP PC as a determinant of crowdfunding performance. Their research focusses on Kiva.org, a lending platform, where there seems to be a relationship between GDP PPP and platform activity. Kickstarter, on the other hand, is a reward-based platform, as mentioned above, that contains often a pre-purchase model, because the majority of the rewards are products of the campaign (Bretschneider, Knaub, & Wieck, 2014). Therefore, the relationship between purchasing power and pre-purchasing seems to apply here. This leads to the following hypothesis.

H3: The crowdfunding performance of a country is better when the country has a higher GDP

(27)

27 3.2.4 Political

The current laws and regulations have to be taken into account when examining the political aspect of the crowdfunding environment. Numerous studies have focussed on regulatory norms that affect crowdfunding (i.e. De Buysere et al., 2012; Cuesta et al., 2015). Since both the

Netherlands and Germany lack regulations on reward-based crowdfunding, the main focus in this section is on general crowdfunding regulations, alternative financing, and the perceptions of the regulations.

The Netherlands and Germany are European Union members and therefore have to implement the directives of the European Union. The European Commission (2015) came up with a report in order to harmonize the capital markets: “The action plan for building a capital markets union”. As a result of this report, in 2015, the European Union established the Market in Financial

Instruments Directive (MiFID) and the Alternative Investment Fund Manager Directive. These European directives must be implemented by the member countries, but according to European Crowdfunding Network (2014), the rules are not that strict because the European Union fears that the benefits of harmonized markets are outweighed by the costs of implementing strict regulations. Both the MiFID and AIFMD directives have implications for crowdfunding, but only for equity-based and loan-based crowdfunding. Therefore, donation-based and reward-based crowdfunding platforms are not considered part of the financial system, but as channels for donations and pre-purchasing. Nevertheless, there are several regulations in both Germany and the Netherlands, including EU directives and national regulations, that do influence these types of crowdfunding. However, as De Buysere et al. (2012) note, these regulations are almost the same and should not be taken into account as a major driver of reward-based crowdfunding. Juredieu and Mayoux (2016) claim that it will be hard for crowdfunding to develop without an adequate specified legal framework and because it is dissociated from the standard financial regulations.

Wardrop et al. (2015) provide a useful view on the adequacy of the regulatory landscape of both countries. In their article, the authors state (p.24) that “the regulatory landscape of the European alternative finance market is fluid and multifaceted. In some countries, existing regulations have been ‘stretched’ to accommodate online alternative finance. In other countries, new regulations

(28)

28

have put clear boundaries around the industry; in yet others, there has been little regulation at all.”

They have also found that over 40% of the surveyed platforms in the Netherlands perceive the regulations as adequate, while in Germany, over 58% platforms perceive that the (proposed) regulations are too strict. Thus, there seem to be significant differences between the countries in this regard. The Alternative Finance Maturity Index (Roodink et al., 2016) also indicates that overall, crowdfunding market regulations are perceived slightly less positively in Germany than in the Netherlands.

It seems consistent that if the inhabitants of a specific country perceive the crowdfunding regulations in their country to be adequate, it is more likely that they will participate in crowdfunding.

H4: The better the regulatory framework is perceived to be in a country, the better the country’s

(29)

29

Chapter 4. Research design

In this section, the research design is described. Firstly, the data collection method is presented. Secondly, the data preparation method is presented. Then, the operationalization of the variables is discussed.

4.1 Data Collection

The data for this study were obtained using Webrobots.io (Webrobots, 2017). The raw data

included 179,032 cases between April 24, 2009 and April 15, 2017.10 Every month, Webrobots

collects data from Kickstarter in five steps. In the first step, scraper robots target the Kickstarter website and capture data. Webrobots logs all of the scraper robots’ actions and takes snapshots in order to check the scraped data. In the second step, the scraped data are validated using a

predefined scheme, like a numeric product ID, a decimal number price greater than 0, etc. The records that fail to pass this validation step are excluded from the original data set and stored in the ‘bad records’ table with an explanation for their failure. In the third step, the data are mapped in order to look for matches. This is a great tool for matching various data sets, but it is not used in this study. In the fourth step, the data are tested for statistical validation through comparison with previous scraped data. This check includes verifying the number of records, the averages of numeric fields, etc. In the final step, the data are exported to a customer database so that the customer can use it. In this case, the data were exported to a CSV file. The method of data collection follows that of Mollick (2014), Kuppuswamy and Bayus (2015), and Agrawal et al. (2011).

According to official Kickstarter reports, a total of 365,076 campaigns has been launched since

April 2009.11 Thus, about half of the total cases are included in this study. When using

web-scraping software, it is not uncommon for there to be discrepancies between the sample and the total number of campaigns. There is no reason to assume that the discrepancy between the sample and the total number of campaigns is caused by a specific attribute of the data. However, the size of the data set is considered adequate here, and the data set includes cases from the entire period and from all different countries.

10

https://s3.amazonaws.com/weruns/forfun/Kickstarter/Kickstarter_2017-04-15T22_21_18_122Z.zip

(30)

30 4.2 Data Preparation

Incomplete cases

First of all, incomplete cases were removed. Some cases in the file are lacking crucial data and could not be used in the analysis. This crucial data included, but is not limited to: the country of origin of the project, the pledged amount of money, and the unique ID. Therefore, these cases were excluded from the data set. After excluding incomplete data, the data size was 163,846.

Country

All of the data from countries outside of Europe were discarded. The focus of this study is the Netherlands and Germany, and only other European countries were used as references or

controls. The majority of the other campaigns are from the United States, but several campaigns from Australia, Canada, Mexico, and Singapore were also discarded. After excluding campaigns outside of Europe, there were still 19,636 campaigns. The majority of remaining campaigns are based in the United Kingdom (GB). Luxembourg has only 19. The United Kingdom probably has so many more campaigns than other countries because Kickstarter was launched two years earlier there (as described in Section 2.3) than in any other European country. Luxembourg probably has so few campaigns because it is such a small country. Thus, these countries are outliers in the data and excluded from the data.

Incomplete campaigns

Only completed campaigns were included in the sample. Thus, the suspended, cancelled, and live campaigns were discarded from the sample. In total, 1,247 cases were removed in this step. When a project is suspended by Kickstarter, it means that the project owner has violated

Kickstarter regulations.12 When a project is cancelled, all of the pledges from the backers are immediately returned to the backers, but the backers and founders can still comment on the project. One of the main reasons that project owners cancel projects is that they made a mistake in the process of creating the campaign, i.e. indicating the wrong duration, goal, or title

(Greenberg & Gerber, 2014). Finally, when projects are still live, they can still receive donations; thus, they are not reliable sources of data for our purposes. The steps of preparation are presented in Table 5 and the amount of cases per country are visualised in Figure 6.

(31)

31

Table 5

Data preparation

State of preparation Number of cases included

Raw data 179,032

After excluding the incomplete cases 163,846

After excluding countries outside of Europe 19,636

After excluding ‘UK’ and ‘LU’ 7,346

After excluding the incomplete campaigns 6,199

Figure 6. Number of cases (per country)

1 3 9 1 8 0 2 0 0 1 1 1 7 3 8 6 6 0 6 8 0 7 2 5 9 7 4 1 9 3 4 2 6 0 5 7 0 0 200 400 600 800 1000 1200

NUMBER OF CASES (PER

COUNTRY)

(32)

32 4.3 Operationalization of the variables

Crowdfunding performance

In order to measure a country’s performance, we included two variables. The first is the total amount of money pledged since the launch of Kickstarter in the country. The second is the success rate of the Kickstarter campaigns in the country.

Pledged

The financial support of the backers per case is reflected in the variable pledged. As the amount of pledged money in the data set is reflected in local currency, there is also included the value USD_rate in order to make it possible to compare the different currencies. Therefore, with the combination of the variables pledged and USD_rate, the variable pledged_USD is established. In this study, the most important aspect of this variable is the amount of pledged money per capita per year. Thus, the combination of the sum of the variable pledged_USD per country, divided by the population per country and the number of years Kickstarter is active in a country, there is derived a value per country: pledged_percapita_peryear. From the included countries, as

visualised in Figure 7, the highest amount pledged per capita per year is in Sweden with $0.20, in contrast with Italy with only $0.03. Project owners from the Netherlands ($0.14) have pledged more money per capita than project owners from Germany ($0.09).

Figure 7. Amount of money pledged per capita per year (per country)

$ 0 .1 2 $ 0 .0 6 $ 0 .1 2 $ 0 .0 9 $ 0 .1 5 $ 0 .0 5 $ 0 .0 9 $ 0 .1 4 $ 0 .0 3 $ 0 .1 4 $ 0 .1 0 $ 0 .2 0 A T B E C H D E D K E S F R I E I T N L N O S E

PLEDGED PER CAPITA PER

YEAR (PER COUNTRY)

(33)

33 Success rate

To determine the success of a project, a new value had to be created. Cumming et al. (2014) describe the completion value as the ratio of ‘total amount pledged’ to ‘project goal’. In this data set, pledged was divided by goal (both written in local currency). A value of 1 or higher

indicates a successful campaign, and a value below 1 indicates an unsuccessful campaign. When the success rate is >1, the variable remains >1 (success), but when the rate is <1 the variable is 0 (unsuccessful). Since Kickstarter uses an AON-model, all of the projects with a rate below 1 were considered 0. The average_success_rate per country was used to compare the countries as presented in Figure 8. Italy has the highest average success rate, with 1.57, while Austria has the lowest average success rate with only 0.51. Germany has a higher average success rate (0.81) than the Netherlands (0.55).

Figure 8. Average success rate (per country)

A closer look into these two determinant variables of the crowdfunding performance reveals some interesting differences. For instance, Italy and Spain have the lowest amount of pledged money per capita per year, while crowdfunding campaigns on Kickstarter from Italy and Spain are in the top three of the highest average success rate. Furthermore, the high score for Sweden on both the amount of pledged money per capita per year and the average success rate must be noticed. 0 .5 1 0 .5 6 0 .5 9 0.8 1 0 .6 9 1 .0 5 0 .9 1 0 .7 5 1 .5 6 0 .5 5 0 .5 9 1 .0 9 A T B E C H D E D K E S F R I E I T N L N O S E

AVERAGE SUCCESS RATE (PER

COUNTRY)

(34)

34 Crowdfunding environment

The variables related to the crowdfunding performance are compared to the variables related to the crowdfunding environment (trust, friends and family, GDP per capita, and perceptions of regulations) to determine whether there is a relationship between the crowdfunding environment and the crowdfunding performance.

Social capital

With respect to H1, ‘trust’ was used to compare countries and was linked to their social capital. The parameter ‘trust’ was derived from the Eurostat survey (Eurostat, 2013) which includes the question: “Would you say that most people can be trusted?” The answers were provided on a 10-point scale, from low-trust to high-trust. Since several factors influence crowdfunding, the relationship was determined by comparing all of the European countries and their performance. In the Nordic countries; Denmark (8.3), Norway (7.3) and Sweden (6.9) is the level of trust perceived as the highest, while the lowest level of trust is perceived in Germany (5.5). The level of trust in the Netherlands is also perceived as one of the highest (6.9). In Figure 9 is the level of trust presented (Eurostat, 2013, p.9).

Figure 9. Level of trust (per country(

5 .9 5 .7 6 .4 5 .5 8 .3 6 .3 5 6 .4 5 .7 6 .9 7.3 6 .9 0 1 2 3 4 5 6 7 8 9 10 A T B E C H D E D K E S F R I E I T N L N O S E

(35)

35 Friends and family

To address H2, we focus on the contribution of friends and family. As described in the literature, the first $20,000 in funding is often provided by family and friends and rarely by unknown individuals on crowdfunding platforms (Agrawal et al., 2013). Therefore, successful campaigns with pledged amounts between $0 and $20,000 were assigned ‘1’ in the dummy FriendsFamily, while unsuccessful campaigns between $0 and $20,000 were assigned ‘0’. In order to compare the countries, the average of the variable FriendsFamily is used and called mean_FF. This variable represents if countries have per project on average a higher contribution of friends and family, as presented in Figure 10. In Norway is the mean_FF the highest with 0.85, while the lowest value is in Switzerland with only 0.66. In the Netherlands, 73% (0.73) of the cases between $0 and $20,000 are successful, while 69% of these cases were successful in Germany,

Figure 10. Mean_FF

GDP PPP PC

In order to examine H3, we made the same type of comparison as when we tested the first hypothesis. All European countries where Kickstarter is active were included in the sample in order to investigate the relationship between GDP PPP PC (World Bank, 2016) and the

crowdfunding performance. The GDP PPP PC varies in the included countries from $36,415.00 in Spain up to $69,249.00 in Norway, as presented in Figure 11.

0 .6 8 0 .8 2 0 .6 6 0 .6 9 0 .8 4 0 .7 4 0 .7 0 .8 2 0 .7 1 0 .7 3 0 .8 5 0 .7 9 A T B E C H D E D K E S F R I E I T N L N O S E

(36)

36 Figure 11. GDP PPP PC per country

Perceptions of regulations

In order to examine H4, we focussed on the perceptions of regulations of the inhabitants of the European countries where Kickstarter is active. Wardrop et al. (2015) have compared different countries based on their crowdfunding performance. In their extensive study, they conducted a survey with regards to the perceptions of regulations. Thus, the perceptions were divided into different categories. Firstly, there was made a distinction in the answer of the participant of the survey between the current and upcoming regulations. Secondly, there was made a distinction between the necessity of regulations. Finally, it was considered whether the inhabitants perceived the regulations to be adequate. Unfortunately, these data were only available in France,

Germany, and the Netherlands. Wardrop et al. (2015) use the same data for all Nordic countries: Denmark, Norway, and Sweden, while the regulations differ per country . Due to the lack of data related to perception, we decided not test this hypothesis.

$ 4 8 ,0 0 4 .0 0 $ 4 5 ,0 4 6 .0 0 $ 5 9 ,5 6 0 .0 0 $ 4 8 ,1 1 0 .0 0 $ 4 7 ,9 8 5 .0 0 $ 3 6 ,4 1 5 .0 0 $ 4 2 ,3 1 3 .0 0 $ 6 9 ,2 3 0 .0 0 $ 3 6 ,8 3 3 .0 0 $ 5 1 ,0 4 9 .0 0 $ 6 9 ,2 4 9 .0 0 $ 4 9 ,8 3 6 .0 0 $10,000.00 $20,000.00 $30,000.00 $40,000.00 $50,000.00 $60,000.00 $70,000.00 $80,000.00 A T B E C H D E D K E S F R I E I T N L N O S E

GDP PPP PC (PER COUNTRY)

(37)

37

Chapter 5. Results

In this chapter, we took a closer look in the collected data in order to answer the main question. In the comparison of the variables, all included countries are examined in order to find a relation. Firstly, the descriptive data from the two determinant variables of crowdfunding performance are presented. Subsequently, the data is analysed using correlation analysis to determine if there is a linear relation between the variables. These two steps lead to results that help answer the main question: Is the difference in crowdfunding performance between the Netherlands and Germany explained by the crowdfunding environment?

5.1 Descriptive data

In the data set, there are 6.199 cases included. There are 1.117 cases from Germany and 934 cases from the Netherlands. The value for the amount of pledged money per capita per year performance per country is in the variable “pledged_percapita_peryear” and the value for the average success rate per country is in the variable “average_success_rate”. As stated in the previous chapter, the variable “pledged_percapita_peryear” incorporates the total amount of money pledged per capita per year and is derived from the data set variable: pledged_USD. In the Table 6, the descriptive statistics of this variable are presented. The mean amount of money pledged per crowdfunding project on Kickstarter is between $5,251 in Italy and $14,309 in Austria. In Germany, the mean amount of money pledged is $13,607, where this mean is much lower in the Netherlands with $8,119. When interpreting the standard deviation (Std. Deviation) of this variable, there must be noted that there are no projects with less than $0,00 pledged. The standard deviation varies from $17,322 in Denmark between $67,539 in France. The sum of the projects is closely related to the mean and the number of cases, where the lowest sum can be found in Belgium with $1,308,656.95 and the highest sum in Germany with $15,199,501.27. The median of the amount of pledged money in the investigated countries varies from $62.68 in Italy up to $679.33 in France. As mentioned before, the minimum amount of pledged money per project is $0.00, and this occurs in all included countries. The maximum amount of pledged money per project varies between $181,144 in Denmark and $1,370,333 in France. When testing the hypotheses, the total amount of pledged money in a country will be used (sum).

(38)

38

Table 6 Pledged (in $)

country N Mean Std. Deviation Sum Median Min. Max.

AT 139 14,309.34 55,439.60 1,988,997.57 343.37 0 444,748 BE 180 7,270.32 33,058.03 1,308,656.95 244.59 0 333,529 CH 200 10,383.42 39,076.12 2,076,684.59 242.11 0 419,054 DE 1117 13,607.43 61,803.73 15,199,501.27 304.17 0 1,072,256 DK 386 5,687.64 17,322.87 2,195,428.94 400.83 0 181,144 ES 606 7,784.09 36,925.50 4,717,156.98 103.82 0 520,918 FR 807 14,691.77 67,539.30 11,856,262.24 679.33 0 1,370,333 IE 259 6,690.79 19,137.36 1,732,915.19 383.56 0 206,830 IT 741 5,251.20 23,971.29 3,891,140.40 62.86 0 437,799 NL 934 8,119.56 27,973.81 7,583,668.47 247.37 0 363,820 NO 260 5,411.44 24,448.39 1,406,973.28 147.33 0 353,574 SE 570 9,147.44 39,022.29 5,214,040.67 361.30 0 578,552

The second variable used for the crowdfunding performance is the average_success_rate, and this variable is derived from the success_rate. In Table 7 the descriptive statistics of the success rate are presented. The mean of the success rate of the included cases varies between .51 in Austria and 1.58 in Italy. The mean success rate in Germany is .81, while the mean success rate in the Netherlands is .55. The standard deviation of these variables differs strongly between the countries, from 1.13 in Austria up to 30.39 in Italy. As the success rate is a ratio, the sum of this variable does not have any further implications. Just as with the previous variable pledged, there are failed cases without any funding in all countries, thus the minimum for this variable is in all countries: 0. The highest success rate for a case in the sample is 826 in Italy, while the lowest maximum is 8.20 in Austria.

(39)

39

Table 7 Success rate

country N Mean Std. Deviation Sum Min. Max.

AT 139 0.51 1.13 70.51 0 8.20 BE 180 .56 1.40 101.01 0 15.00 CH 200 .59 1.59 118.11 0 13.59 DE 1117 .81 2.80 905.27 0 45.48 DK 386 .69 1.51 266.94 0 22.92 ES 606 1,05 8.77 638.64 0 205.00 FR 807 .91 2.62 734.94 0 39.54 IE 259 .76 2.04 196.05 0 18.17 IT 741 1.58 30.39 1169.10 0 826.00 NL 934 .55 1.55 515.24 0 22.13 NO 260 .59 1.32 154.43 0 11.69 SE 570 1.09 3.07 623.12 0 37.46 5.2 Correlations

The correlation coefficient matrix is used to compare the data from the crowdfunding

environment countries with the variables related to crowdfunding performance. As the values are interval and ratio variables, the Pearson R correlation coefficient is used in order to determine whether there are linear relations between the variables.

Before testing the hypotheses, the data from Wardrop et al. (2015, p.16) is compared with the data set of this study. The results from the study from Wardrop et al. (2015) are the main driver reasons to investigate the relation between the crowdfunding performance of a country and the crowdfunding environment. The volume of alternative finance transactions in 2014 (figure 2) and the alternative finance volume per capita (figure 3) are compared to its equivalents of the data set of this study with a correlation coefficient matrix. The data presented in figure 2 is

(40)

40

translated into the variable wardrop_sum and the data presented in figure 3 is translated into the variable wardrop_percapita. These variables are compared with the sum of the variable pledged: sum_pledged and the pledged_percapita_peryear. As shown in table 8, the variable

wardrop_sum has a significant positive relation with the variable sum_pledged (r = .906; p < 0.01). The variable wardrop_percapita has a significant positive relation with the variable pledged_percapita_peryear (r = .711, p < 0.05).

After this comparison, the data related to the different hypotheses are compared with the determinants of the crowdfunding performance of a country.

H1: If the level of trust in a country is higher, the crowdfunding performance of this country is

better.

In order to examine H1, the variable trust has been compared with both the

pledged_percapita_peryear and the average_success_rate. The degree of trust has a

non-significant correlation with pledged_percapita_peryear with r = .559 and p = 0.59. There is also no signification relation between trust and the average_success_rate, with r = -.254 and p = .425, so there is found no support for H1.

H2a: The crowdfunding performance in a country is better when there is a higher contribution of

friends and family.

In order to examine H2a, the variable mean_FF has been compared with both

pledged_percapita_peryear and the average_success_rate. The level of mean_FF has a non-significant correlation with pledged_percapita_peryear with r = .248 and p = .437. There is also no signification relation between mean_FF and the average_success_rate, with r = -.150 and p = .641, so there is found no support for H2a.

H2b: The contribution of friends and family for crowdfunding is higher in countries with a lower

level of masculinity

In order to examine H2b, the level of masculinity has been compared with the mean_FF. The level of masculinity has a significant negative relationship with mean_FF, with r =.-583 at the p < 0.05 level. Thus, there is found support for H2b.

(41)

41 H3: The crowdfunding performance of a country is better when the country has a higher GDP

PPP per capita.

In order to examine H3, the variable GDP_PPP_PC has been compared with both

pledged_percapita_peryear and the average_success_rate. The level of GDP_PPP_PC has a non-significant correlation with pledged_percapita_peryear with r = .478 and p = .116. There is also no signification relation between GDP_PPP_PC and the average_success_rate, with r = -.531 and p = .075, so there is found no support for H3.

Since we decided not to test H4 due to the lack of data, there is found no support for this hypothesis.

Besides the hypotheses, the social value trust has a significant positive correlation with the mean_FF, r = .608 and trust has a significant negative correlation with masculinity, r = -.667. Both correlations are significant at the p < 0.05 level.

Referenties

GERELATEERDE DOCUMENTEN

This model was used to predict change in the natural frequency, thus estimating fatigue life, using only frequency domain information. Execution of the model required only the

decline in public funding/ high tuition fees; pressures to increase research productivity or improve teaching to attract more

Bij het onderzoek werden enkel relatief recente grachten aangetroffen die volgens de oude percelering lopen (zie uittreksel uit de Poppkaart), verder werden geen

The social network of the project leader will be tested via their Facebook friends and LinkedIn connections, while the quality of preparation will be tested via the business plan

The goal of this paper is to investigate how crowdfunding can be a suitable alternative financing option for small and medium car dealers in the Dutch automotive

Crowdfunding Success, Founder’s Social Network, Social Capital Theory, Social Network Theory, Social Network Analysis, Number of Facebook Friends, Number of Collaborators, Number

“To what extent do individual and geographical related motives of donors influence the amount of money donated on creative crowdfunding projects in The Netherlands

We experimentally demonstrate selective addressing any location (symbol) in a 9072 size grid (alphabet) to achieve 10.5 bit of mutual information between the sender and receiver