Crowdfunding : an empirical study of determining relevant factors of the level of financial contributions

CROWDFUNDING

,

An empirical study of determining relevant

factors of the level of financial contributions

Student ID: 6366864 / 10105697 Supervisor Ilko Naaborg

Bachelor Thesis

Faculty of Economics and Business Specialization Finance and Organization Field Finance

Amsterdam, 17 February 2014

Crowdfunding is a new concept in today’s economy. Many crowdfunding platforms have been established over the last years and often proved to be successful in collecting money. This is done through an open call to a large group of people – “the crowd” – in order to fund projects with a specific purpose. This study examines different factors that determine the level of financial contributions through crowdfunding. It will also investigate whether there is a relationship between those factors and the level of financial contributions. This is established by conducting a descriptive desk research and empirical research. With the dataset used from the ‘Nationaal Crowdfunding Onderzoek 2013’, this study finds evidence that there is a big difference in the importance of the different factors for the determination of the level of financial contributions. The level of financial contributions is highest when it is related to equity- and lending-based crowdfunding. This is not in line with the theory in the relevant literature. For the factors ‘the amount of money that is necessary for the crowdfunding project’ and ‘the level of financial contributions of others’ holds that when the importance increases the level of financial contributions increases. For all the other factors included which are ‘the return, the financial situation of contributors’, percentage that is already funded, minimum amount of money that can be fund’, the opposite is true. The most important factor for the contributors is the financial situation. ‘The amount of money that is necessary for the crowdfunding project’ has a big significance relevance for the level of financial contributions, while the return is not significant at any level. Keywords: Crowdfunding, crowdfunding platform, level of financial contributions

2

Table of content

2.2.1 Types of crowdfunding platform ... 11

2.3LEVEL OF FINANCIAL CONTRIBUTIONS ... 13

3. METHODOLOGY AND DATA ... 14

3.1METHODOLOGY ... 14

3.2DESCRIPTIVE DATA ... 16

4. EMPIRICAL RESULTS ... 18

4.1CHANGE IN LEVEL OF FINANCIAL CONTRIBUTIONS ... 18

4.2MARGINAL EFFECTS ... 20 4.3PREDICTED PROBABILITIES ... 21 5. CONCLUSION ... 22 REFERENCES ... 26 APPENDIX A ... 28 APPENDIX B ... 29

3 LIST OF FIGURES

Figure 1 - Crowdfunding: How it works 8

Figure 2 - The major forms of capital provision ranked by process complexity 10

Figure 3 - The importance of the factors which determine the level of financial contributors 17

Figure 4 - The different levels of financial contributions 18

LIST OF TABLES

Table 1 - Coefficients of the ordered probit model 21

4

1. Introduction

This chapter presents the background of crowdfunding and then the purpose of this thesis will be given. After presenting the research question, the structure is outlined.

Crowdfunding is a new upcoming financial concept, which is used to finance projects of entrepreneurs. Crowdfunding involves a large group of people, also called the ‘crowd’, who donate small amounts of money. Collecting small amounts of money from a large number of people has a rich history in many domains. One of the examples are Mozart and Beethoven, who financed their concerts and new music compositions with money from interested patrons (Hemer, 2011). The election campaign of President Barak Obama in 2008 was raised for the most part by small donations from all over the web. He raised over $750 million from almost four million donors (Bradford, 2012). Furthermore, the Statue of Liberty in New York was funded by small donations of Americans and French people (Hemer, 2011). In the last years, crowdfunding has become more important as a financial source. It allows entrepreneurs to raise money through an open call on the Internet (Belleflamme et al., 2013).

The purpose of this bachelor thesis is to describe and analyze the new financing phenomenon crowdfunding and to investigate different factors that determine the level of financial contributions. The different factors in this study vary from the different platforms to the importance of other factors such as the received return, the target amount of money for a project, the percentage that is already raised, minimum amount of money that can be donated and the contributions of others. It will also examines if there are any relationships between those factors and the level of financial contributions. There has been several research about crowdfunding. Most of them are qualitative exploratory like the studies of Gerber et al. (2012), Vliet (2011) and Kleverlaan et al. (2013) which all investigate the motivations of contributors in crowdfunding. While others are descriptive or empirical like the study of Agrawal et al. (201) about the geography of crowdfunding. Even though, none of them are focused on the level of financial contributions through crowdfunding. This makes this thesis unique with a great impact on the empirical knowledge of crowdfunding. The

5

data used is a selection from the database of ‘Nationaal Crowdfunding Onderzoek 2013’. An ordered probit model will be used, with the level of financial contributions as dependent variable. This leads to the research question of this thesis:

“Which factors are playing a role in the determination of the level of financial contributions through crowdfunding?”

The remaining part of this thesis is structured as follows. The second chapter will discuss existing literature that is relevant for this research. It also provides an overview of the hypotheses. In the third chapter the methodology that will be used is introduced and the data collection is presented. The results of this research will be presented in the fourth chapter and chapter five contains the conclusion.

2. Literature review

This chapter aims to present the theoretical framework of crowdfunding. First, crowdfunding is explained through web 2.0 and crowdsourcing. Thereafter, previous studies and theories will be presented to address the different crowdfunding platforms and the determination of the level of financial contributions. This is followed by the hypotheses.

2.1 What is Crowdfunding?

The emergence of crowdfunding can be explained by two main fundamentals. The concept of crowdfunding comes from the broader, more general concept of crowdsourcing. This involves using the “crowd” to obtain feedback, solutions and ideas to develop corporate activities. It has been made possible on a large scale by the emergence of ´Web 2.0’, which is a critical fundamental that has facilitated the access to the “crowd” (Lambert & Schwienbacher, 2010). The most important factor in crowdfunding is the ‘crowd’, being a group of individuals connected by the internet. The Internet and especially Web 2.0 has been an effective way to reach a large group of people at the same time. Web 2.0 is a prerequisite for the

6

rapid development of crowdfunding. It refers to the development that the internet has made since 2004. Web 2.0 provides a platform to the crowd to collaborate and combine own knowledge and resources. Furthermore, it offers an open opportunity to different crowdfunding projects which people would like to take part in and contribute freely. Collaboration and openness are two of the three main characteristics for Web 2.0 according to Lee et al. (2008). Participation is the third characteristic, which has increased thanks to the ease of access and use of computers and mainly the Internet. Due to the increasing participation, crowdfunding was able to prosper in the last years (Gerber et al.,

2012).

Several authors (including Brabham 2008, Kleemann 2008 and Schwienbacher & Larrade 2010) state that there is a relationship between the emergence of Web 2.0, crowdsourcing and crowdfunding. According to those authors its structure was mandatory for companies to be able to reach networks of consumers that easily. The emergence of Web 2.0 has created new forms of interactive communication through the internet and therefore has created new opportunities to bring knowledge of different people together. Users become actively involved and participate on the internet (O'Reilly 2007). Different organizations have started to develop a crowdfunding platform on the internet. Regarding the research of Lambert and Schwienbacher (2010) 80% of the platforms has been developed after 2007. International examples are Kickstarter,

IndieGoGo and RocketHub. A well-known Dutch platform is SellaBand. Mentioned

before, crowdfunding is originated from the broader concept of crowdsourcing. Kleemann et al., (2008, p.6) pointed out that

“crowdsourcing takes place when a profit oriented firm outsources specific tasks essential for the making or sale of its product to the general public (the crowd) in the form of an open call over the internet, with the intention

of animating individuals to make a [voluntary] contribution to the firm’s production process for free or for significantly less than that contribution is

worth to the firm.”

Jeff Howe and Mark Robinson first introduced the concept in the June edition of Wired Magazine in 2006 (Belleflamme et al., 2013). According to Lévy (1997) the

7

more individuals share, the more important knowledge becomes, because “no

one knows everything, everyone knows something, [and] all knowledge resides in humanity” which is namely the interpretation of ‘collective intelligence’.

Brabham (2008) states that the more diverse the crowd members are, the higher the efficiency is of crowds in solving problems. Crowdsourcing and crowdfunding can be used as an instrument to provide valuable signals on the market potential of a product they wish to launch (Schwienbacher & Larralde, 2010).

Crowdsourcing originates from a combination of the words “crowd” and “outsourcing” and can be used in both commercial and non-commercial environments. A well-known example and probably the earliest and clearest example of a non-commercial form of crowdsourcing is Wikipedia. Jimmy Wales and Larry Sanger created an online encyclopedia where every entry can be added by any web user and edited by any other (O'reilly, 2007). Schwienbacher and Larralde (2010) extend the definition of crowdsourcing provided by Kleemann et al. (2008), by describing crowdfunding as

“an open call, essentially through the Internet, for the provision of financial resources either in form of donation or in exchange for some form of reward

and/or voting rights in order to support initiatives for specific purposes”.

The definition stated above is purely based on financing crowdfunding projects, but crowdfunding can also be used for marketing purposes or to develop better and stronger relationships with customers and consumers (Belleflamme et al., 2013).

Schwienbacher and Larralde (2010) conclude that there are several key elements to understand why crowdfunding is embedded in the definition of crowdsourcing. The first common feature and most significant characteristic of both crowdfunding and crowdsourcing is tapping a general public, namely the crowd. In crowdfunding the crowd can only provide financial help, while the crowd in crowdsourcing can also provide voluntary contributions to the project initiators. The second common feature is that Web 2.0 is a critical instrument that has facilitated the access to the ‘crowd’ (Brabham, 2008; Kleemann et al. 2008). Last and third common feature is that there are many practical

8

differences between crowdsourcing and open-sourcing, some of which also exist between crowdfunding and open-sourcing, this has been confirmed by Brabham (2008).

In figure 1 below crowdfunding is simplified into tree steps. Small donations by a large amount of people will eventually give the power to achieve a financial goal of the project initiators.

Figure 1 - Crowdfunding: How it works

(Source: Singapore Entrepreneur, 2008) 2.2 Crowdfunding platforms

Alongside of the development of crowdfunding several “platforms” have emerged to act as an intermediary between the public and the funding projects (Schwienbacher & Larralde, 2010). This creates a two-sided market, which means that there is a two-way collaborative communication possible. People are able to easily connect with each other around the world (Belleflamme et al., 2013). In recent years crowdfunding platforms were launched in large numbers across the globe. The world dominant platform is Kickstarter. As of late January 2014, the total amount of dollars pledged for Kickstarter’s projects is $951 million and it succeeded in funding 55,297 projects. At Kickstarter the people

who support projects are called backers1_{, which are 5,504,614 up until now.}

VoordeKunst is the first Dutch platform that is focused art, where upcoming art projects from artists or art organizations are seeking funding from individuals

9

and businesses. Up until begin February 2014, 614 projects have been funded completely by 35,344 funders. The amount of the projects are summed to an amount of €3,664,238. Some other well-known platforms are Kiva, 1%Club, CrowdAboutnow and Share2Start.

An internet platform provides the opportunity for people to present their projects to the world, this breaks down the geographic restrictions of fundraising. According to Agrawal et al. (2011)

“Crowdfunding platforms are purposefully designed to overcome distance-related frictions”.

There are different platforms which are focused on different industries. Kickstarter and IndieGoGo focus on a broad creative industry, while SellaBand primarily focuses on music, Cinecrowd on movies, Tenpages on books and Sybid and CrowdAboutNow are focused on broader entrepreneurship. According to the study of Massolution (2012) there will be over 530 crowdfunding platforms by December 2012.

Crowdfunding platforms may be categorized according to: (i) Crowdfunder’s

participation rights in the project’s outcome, (ii) the fees that the project initiators have to pay to the crowdfunding platform; (iii) the services offered by the crowdfunding platform (Giudici et al.,2012). In the research of Massolution

(2012) the classification is based on the funders’ primary motivation for engaging in crowdfunding. There are four major types of crowdfunding platforms:

1. equity-based: when the project is fully funded, the crowdfunders have the right to share the residual income of the specific project;

2. lending-based: sometimes called peer-to-peer lending. Funds are paid back to the crowdfunders and they have the right to receive an interest payment which is contractually agreed;

3. donation-based: the contributors receive no compensation, for philanthropic or sponsorship proposal;

10

4. reward-based: funds are provided in exchange for non-monetary benefits. This can for example be a release of the product or citations (Giudici et al., 2012). According to Hemer (2011) there are different levels of complexity and uncertainty over the different business models of the crowdfunding platforms. As shown in figure 2 below, equity-based crowdfunding is the most complex from both a legal standpoint and with respect to information asymmetries.

Figure 2 - The major forms of capital provision ranked by process complexity

(Source: Hemer, 2011)

Bradford (2012) differentiates the crowdfunding platforms into five models, which relates to what investors are, if anything, promised in return for their contribution. The fifth model not mentioned before is the pre-purchase model. However, in this research the pre-purchase model and reward model will be combined into one mode, which is very common to see in different platforms. For example AngelList, Lending Club and Seedrs are platforms that are both debt- and equity-based crowdfunding platform.

Massolution is a unique research, advisory and implementation firm that specializes in crowdsourcing solutions. In one of their research the functionality of different crowdfunding platforms and their different models are explained. The study consisted of a survey, where the respondents were asked 30 detailed questions relating to the participants on their platforms, the functionality of their platforms and their fundraising activities for the calendar years 2009, 2010 and 2011. Further data was gathered via direct communications with 135 CFPs and significant secondary research. The participants were selected from

11

Crowdsourcing.org’s Directory of Crowdfunding Sites which contained 452 active crowdfunding platforms at the time the survey was conducted. As basis for their analyze, they selected 135 of the 170 submissions from crowdfunding platforms that were comprehensive and of high-integrity according to Massolution.

2.2.1 Types of crowdfunding platform

Massolution (2012) classified the equity-based model as:

“a model in which funders receive an interest in the form of equity in the venture they fund or, alternatively, revenue or profit-share arrangements.”

Equity-based crowdfunding would usually involve securities. The contributions are purchasing securities if the investors receives ordinary corporate stock in exchange (Bradford, 2012). According to Massolution (2012) equity-based is the fastest – growing category by net year-on-year growth and were, on average, much larger in size in terms of fund raised. Equity-based crowdfunding is especially suitable for digital goods such as software, film and music. Examples of crowdfunding platforms which offer contributors an interest in the venture in the form of equity or some sort of profit sharing agreement are SellaBand,

Symbid and Wefunder (Agarwal et al., 2011). Project initiators define a time

period and a target amount of money which serves as a threshold. The target is divided into thousands equal slices which are offered as equity shares (stocks) to the crowd for a fixed price for example €10,-. Then pledging begins until the threshold is reached (Hemer, 2011).

Lending- based is according to Massolution (2012):

“A crowdfunding model in which funders often receive fixed periodic income

12

Unlike other models, this type of crowdfunding does not involve a donation but it involves a loan. Investors provide the necessary amount of money and when the project is pledged they expect a repayment in return. In some cases, even interest is offered on the funds they provide. On the leading lending-based platform Kiva, the funders do not receive interest on their loans. Several platforms do offer interest, these are also known as peer-to-peer platforms (Bradford, 2012). In the research of Massolution (2012) it is shown that lending-based platforms have the highest funds pledged to funds paid out ratio out of all categories. Besides this, they state that this category is the smallest in terms of platforms.

Unlike the previous platforms, in which contributors expect some sort of financial return, donation- and reward-based platforms involve no monetary compensation for participation. Not even the return of the principal amount contributed (Bradford, 2012). These models are characterized by non-financial motivations for engaging in crowdfunding. Contributors expect no compensation because for example philanthropic or sponsorship based incentive. Donation-based crowdfunding has raised the most funds. Even though, they are the slowest-growing category (Massolution, 2012).

The reward and pre-purchase models are very common and, as mentioned above, tend to be used together. The contributors are given some sort of “reward” that is not interest or a percentage of the profit (Bradford, 2012). They gain a non-financial benefit in return (Massolution, 2012). It can vary from a “thank you” phone call to a meet-and-greet. It can also contain a product that the project initiator is making or the right to buy the product at a reduced price upon completion (Bradford, 2012). Reward-based crowdfunding performs best when the cause-based campaigns appeal to funders’ personal beliefs and passions for example environment, faith or community. From a category perspective, 50% of the overall composition of the crowdfunding marketplace represented reward-based platforms. This means it has the largest number of online platforms (Massolution, 2012). In Gerber et al.'s (2012) qualitative exploratory study it is shown that contributors who fund a reward-based crowdfunding project tend to give a higher amount of money, when they receive a reward in return. The study consisted of 11 informants, where 3 informants

13

have exclusively created projects, 6 informants have exclusively funded projects, and 2 informants have both created and funded projects, and 3 senior executives representative of each crowdfunding platform. Furthermore, there is a big variety in expertise and project type (Gerber et al., 2012).

2.3 Level of financial contributions

A previous study of Vliet (2011) has mainly focused on the motivation of the

crowdfunders. Vliet (2011) interviewed fourteen crowdfunders between 22nd _of

September and 10th _{of October in 2011. The participants for the interviews were}

sought by placing a request in both the newsletter of the platform VoordeKunst and the platform of CineCrowd. There are different motives for the crowd to determine the level of financial contributions. By interviewing those crowdfunders Vliet (2011) finds four reasons why people fund a certain amount of money through crowdfunding. The first reason is the target amount of money of a project. Some people indicate that the target was very important for them, while others say that the percentage of the target that was already collected was more important. If a project almost failed in collecting the pre-determined target amount at the end of campaign, they would donate a higher amount or even donate for the second time to ensure that the project would be fund successfully. Secondly, the importance of the type of the return can be a determining factor. The opinions amongst the crowd are extremely varied. While some people say that the return they receive is a determining factor, others say that the return is a inessential factor. There are contributors that don’t want to fund a project when there is no return involved. In the research of Vliet (2011) one of the contributors received a limited edition artwork due to his high level of financial contributions. For this contributor the return the determining factor.

There are different reasons given why risk is also important for the determination of the level of financial contributions. Often there is a link between the risk involved in the quality of the end product and amount of money. Because of this risk , contributors don’t want to fund a project with a large amount of money. The last reason that was given in the research is the financial situation of the funders. The level of financial contribution depends on the money that the contributors could miss.

14

This study will examine which factors determine the level of financial contributions. According to Massolution(2012) funders tend to give a higher amount when it is related to reward-based crowdfunding. This leads to the first hypothesis:

H1= The financial contribution in crowdfunding will be higher when it is associated with the reward-based crowdfunding model comparing to the other

models.

As mentioned above there are other factors that are important for the level of financial contributions. To examine this, other hypotheses are formed:

H2 = The factor return plays a more important role when it is related to a higher level of financial contribution comparing to lower financial contribution.

H3 = There is a negative relationship between the financial situation of funders and the level of financial contribution.

H4 = The level of financial contributions is the lowest when it is related to donation-based crowdfunding.

H5 = The level of financial contribution will be higher when the target amount of money of a crowdfunding project is seen as an important factor.

H6 = When the percentage of the target that is already been fund is an important factor, contributors are tend to give more money to crowdfunding projects.

3. Methodology and data

This chapter introduces the methodology used in this thesis, followed by the data collection. An overview of the descriptive data will be presented.

3.1 Methodology

For this empirical research an ordered probit model will be used to examine the importance of different factors that determine the level of financial contributions. An ordered probit model is used to estimate relationships

15

between an ordinal dependent variable and independent variables. In this research the ordinal dependent variable is ‘Level_financial_contributions’, this indicates the level of financial contributions of the respondents. It is categorized into 7 categories: less than €10; €11- €25; €26 - €50; €51 - 100; €101 - €250; €251 - €500; more than €500. The higher the category, the higher is the level of financial contributions. Below the ordered probit regression equation is given:

Level_Financial_contributions = β0+ β1(Ddonation)+ β2(Dequitylending)+ β3(Dunkown)+ β4(DFinancial_situation0)+ β5(DFinancial_situation1)+ β6(DFinancial_situation2)+ β7(DFinancial_situation4)+ β8(DFinancial_situation5) + β9(DImportance_return0)+ β10(DImportance_return1)+ β11(DImportance_return2)+ β12(DImportance_return4)+

β13(DImportance_return5)+ β14(DTarget_amount0)+ β15(DTarget_amount1) + β16(DTarget_amount2)+ β17(DTarget_amount4) + β18(DTarget_amoun5)+ β19(DPercentage_raised0)+ β20(DPercentage_raised1)+

β21(DPercentage_raised2)+ β22(DPercentage_raised4)+ β23(DPercentage_raised5)+ β24(DContributions_other0) + β25(DContributions_other1) + β26(DContributions_other2) +

β27(DContributions_other4)+ β28(DContributions_other5)+ β29(DMinimum_amount0) + β30(DMinimum_amount1) + β31(DMinimum_amount2) + β32(DMinimum_amount4) +

β33(DMinimum_amount5) + εi

Data for this study is used from the ´Nationaal Crowdfunding Onderzoek 2013´2_.

An online survey was made with different kinds of questions about crowdfunding. The questions that are used for this research are shown in Appendix A. In the relevant questions, the respondents could only choose one answer. For this reason, all the independent variables in the equation are dummy variables. The dummy variables can take the value 1 or 0. Per question one of the answers is left out of the equation. This prevents the model from multicollinearity, also called the dummy trap. When all the dummies are zero, the remaining answer is chosen by the respondent. When the dummy variable is 1, the dummy indicates the answer that the respondent has chosen. In the first question in Appendix A respondents were asked if they ever made a financial contributions through crowdfunding. If the answer is ‘yes’ they could give an indication of the level of financial contribution in question 2 in Appendix A. The third question in Appendix A is about whether the respondents received a return and if yes, what kind of return. Mentioned before, the return is different for every

2 _{For the use of the database, which has been provided by Peter van den Akker, Ronald}

Kleverlaan, Gijsbert Koren en Koen van Vliet , the following Creative Commons license is applied CC BY-NC-SA 3.0 NL.

16

crowdfunding platform. The dummies that are added in the equation are ‘Ddonation’, ‘Dequitylending’ and ‘Dunknown’. This indicates the different kinds of platforms. The reference dummy ‘Dreward’ is left out, so the model compares whether people gave a higher or lower level of financial contributions in respect to reward-based crowdfunding platforms. Reward-based crowdfunding is left out because this is the most important platform according to Massolution(2012).

All the following dummies ‘DFinancial_situation0’ till

‘DMinimum_amount5’are related to last question in Appendix A. This question is

about different factors that determine the level of financial contribution. The respondents were asked to give the factors a rate from 1 to 5, where 1 means very unimportant, 3 neutral and 5 means that the factor is very important. There is also a possibility to choose the value 0. When this is chosen the factor was not applicable. In order to see if the respondents tend to give more or less money when the importance of the factor increases, the dummy variable with a rate of 3 is left out. The variable εi is the error term which captures all the relevant variables not included in the model because they are not observed in the dataset. 3.2 Descriptive Data

In order to empirically analyze the different factors that determine the level of financial contributions, data is used from the ´Nationaal Crowdfunding Onderzoek 2013´. This has been one of the biggest research projects about the motivations of crowdfunders in Europe. They made an online survey including 54 question, 4 open and 50 multiple choice questions. 1277 Respondents took part in the research which has been held from the 21st _{of January till 21}st _of

March 2013. Of all the 1277 respondents who answered the survey, only 439 respondents did a financial contribution through crowdfunding. There were 19 respondents who never finished the entire survey. Since the answers of these respondents are not accurate, they are not included in this research.

The third question in Appendix A contains different factors that determine the level of financial contributions. Ignoring whether respondents received a return or not, the importance of the factor is given between the range of 1-5, 1 being very important and 5 being very unimportant. Figure 3 shows that

17

the most important factor is the financial situation and least important factor is the level of financial contributions of others funders. The return is the second most important factor when the respondents determine their level of financial contributions.

Figure 3 - The importance of the factors which determine the level of financial

contributors

(Source: Kleverlaan et al., 2013)

Figure 4 shows how many respondents have chosen a certain level of financial contribution. Most of respondents have fund a project with an amount between €11 and €25. Between the range of €11 and €500 the number of respondents decreases as the amount of money increases. A surprising appearance is that it subsequently increases when the amount of money is higher than €500. Another outstanding feature in figure 4 is when the level of financial contributions is less than €10. In the survey of ´Nationaal Crowdfunding Onderzoek 2013´ the respondents could also answer the question about the level of financial contributions with the option ‘unknown’. These 9 respondents are omitted from the regression.

18 Figure 4 - The different levels of financial contributions

(Source: Kleverlaan et al., 2013)

Furthermore, there is a big diversity in the received return between the respondents. Of the total respondents 22.14% have funded a project concerning donation-based crowdfunding, while 14.36% received a financial return. Reward-based crowdfunding has been the largest platform in this database with 51.58%, which is equal to 212 respondents.

4. Empirical results

In this chapter the results of the empirical study will be presented and analyzed. These results include the coefficients of the ordered probit model and their corresponding z-value. Furthermore, the marginal effects and the predicted probabilities will be shown in this section.

4.1 Change in level of financial contributions

In Appendix B the coefficients of the independent variables are presented. Since the coefficient of the donation-based dummy is negative, it shows that people tend to give a lower amount of money when it is related to donation-based

crowdfunding in comparison with reward-based crowdfunding. On the other

hand, funders tend to give a higher level of financial contributions when it is related to equity- or lending-based crowdfunding. Seeing the model as a whole,

0 20 40 60 80 100 120 Less than €10 €11 - €25 €26 - €50 €51 - €100 €101 - €250 €251 - €500 More than €500 N u m b e r o f r e sp o n d e n ts

19

the variable equity- and lending-based crowdfunding has the most significant influence on the level of financial contribution with a z-value of 3.33, which is significant at a 1% level.

Different factors, such as the ‘financial situation’ and ‘minimum amount of

money that can be fund’, also play a role in the determination of the amount of

money that will be funded. The level of financial contributions will differ, depending on their importance. Looking at the factor ‘the financial situation of

the contributors’, it shows that the coefficient of the ‘financial situation with a rate of 1’ (very unimportant) is higher than the coefficient of the ‘financial situation with a rate of 5’ (very important). This means that in comparison with a neutral

point of view, funders tend to give a lower amount of money when ‘financial

situation’ is seen as an important factor and a higher amount of money when

‘financial situation’ is seen as an unimportant factor. This also holds for the factor ‘minimum amount of money that can be fund.’ Both factors are significant at a level of 10%, and the factor ‘minimum amount of money that can be fund’ is also significant at a level of 5%.

The factor ‘the amount of money that is necessary for the crowdfunding

project’ is significant at a 1% significance level. Together with the variables about

the different crowdfunding platforms it has the highest significant influence on the level of financial contributions in the model. From a neutral point of view, which means a rate of 3 is given to the factor, the level of financial contributions increases. No matter whether the importance decreases or increases. However the level of financial contributions increases the most when ‘the amount of money

that is necessary for the crowdfunding project’ is seen as an important factor.

‘Percentage that is already funded’ is the next factor presented in Appendix B. When contributors find the factor unimportant (rate of 1) they tend to give a higher level of financial contributions comparing to when the factor is seen as an important factor (rate of 5). As the importance of the factor decreases the level of financial contributions increases. This also holds for the factor ‘the return’. On the other hand, for the factor ‘the level of financial contributions of others’ the opposite is true. As the importance of the factor increases the level of financial contributions decreases. The factor ‘percentage that is already funded’ is significant at a 5% level, while the factors ‘return’ and ‘the level of financial

20 contributions of others’ are not significant at all. This means both factors have no

impact on the determination of the level of financial contributions in this model. The boundaries of the different categories of the level of financial contributions are also calculated with the ordered probit model. The category cut offs are presented in Appendix B for each category. Since the level of financial contributions is divided by 7 categories, there are 6 thresholds.

4.2 Marginal effects

In this section the marginal effects on different independent variables will be discussed, which are presented in table 1 below. For all the different factors that play a role in determining the level of financial contributions, there is a common appearance in the marginal effects. The utmost difference is always in the category ‘€11 - €25’ or ‘more than €500’. Only the utmost differences will be covered since the marginal effects will only increase or decrease when the level of financial contributions changes from the category ‘€11-25’ to ‘more than €500’. For this reason the first category ‘less than €10’ is left out. In table 1 the marginal effects are given in percentage. First, the marginal effects on the different crowdfunding platforms in respect to reward-based crowdfunding will be discussed. Donation-based crowdfunding is 6% less likely to be in the category ‘more than €500’ and 11,4% more likely to be in the category ‘€11 and €25’. While equity- and lending-based crowdfunding are 10,1% more likely to have a level of financial contributions in the category ‘more than €500’ and 12,7% less likely to be in the category ‘€11 and €25’. It can be concluded that donation-based

crowdfunding is related to a lower level of financial contributions, while equity- and lending-based crowdfunding is related to a higher level of financial

contributions.

The remaining variables in the model will be compared from a neutral point of view, which means the factor has a rate of 3. Only rate 1 ‘very

unimportant’ and rate 5 ‘very important’ are covered since these are the utmost

rates that can be chosen in the question (the last question in Appendix A). One variable that stands out is the target amount of money, which shows the biggest difference in percentage of all variables shown in table 1. The factors ‘the

21 financial situation of the contributors’, ‘the return’ and ‘percentage that is already funded’ are most likely to be in the category ‘more than €500’ no matter the rate

of importance. When a rate of 5 is given, which means contributors find the factor important, the factors ‘percentage that is already funded’ and ‘minimum

amount of money that can be fund’ are more likely to be in a lower category of

financial contributions and less likely to be in a higher category. The opposite is true for the factor ‘the level of financial contributions of others’.

Table 1: Marginal effects in percentage

Variables € 11- € 25 >€500

Crowdfunding platforms

Donation-based crowdfunding +11,4 -6,0

equity- and lending-based crowdfunding -12,7 +10.1 When option 1 is chosen: 'very unimportant' The financial situation of the contributors -6.0 +4.2

The return -4.4 +2,9

The amount of money that is necessary for the crowdfunding

project -5.0 +3,3

Percentage that is already funded -0,9 +0,6

Level of the financial contributions of others +3.0 -1,7 Minimum amount of money that can be fund -8.7 +6,2

When option 5 is chosen: 'very important'

The financial situation of the contributors -4.4 +2.7

The return -2.2 +1.4

The amount of money that is necessary for the crowdfunding

project -13.6 +11.2

Percentage that is already funded +2.3 -1.3

Level of the financial contributions of others -9.9 +7.7 Minimum amount of money that can be fund +4.9 -2.7

4.3 Predicted probabilities

Each of the individual observation in the data has a prediction for the predicted probability of being in each category. The predicted probabilities are shown in table 4 below and are summarized across all the individuals. Comparing the predicted and actual probabilities with each other, it is shown that they are very close to what is observed in the data. The prediction is on average very good, but individual predictions can vary.

22 Table 2: predicted and actual probabilities

Variable Range

Predicted

probability Actual probability

Category 1 < €10 7,85% 8,03% Category 2 €11 - €25 27,33% 27,01% Category 3 €26 - €50 24,36% 23,84% Category 4 €51 - €100 16,24% 16,55% Category 5 €101 - €250 8,63% 9,00% Category 6 €251 - €500 4,43% 4,62% Category 7 >€500 11,15% 10,95%

5. Conclusion

This chapter summarizes a conclusion. The empirical results are discussed and compared to previous research and studies. It also proposes different insight for future research.

The purpose of this study was to investigate what factors determine the level of financial contributions and whether there is a relationship between those factors and the level of financial contributions. Furthermore, the importance of the different factors are examined as well. The data in this thesis is a selection of the database of ‘Nationaal Crowdfunding Onderzoek 2013’. The questions which were relevant for this research are presented in Appendix A. An ordered probit model is used with the dependent variable ‘Level_financial_contributions’. In order to see what determines the level of financial contributions the following factors are included in the model: ‘the financial situation of the contributors; the

return; the amount of money that is necessary for the crowdfunding project; percentage that is already funded; the level of financial contributions of others; minimum amount of money that can be fund’. In the survey of the ‘Nationaal

Crowdfunding Onderzoek 2013’ the respondents were asked about the importance of these different factors. The importance of the different factors are ranked from 1-5, 1 being very important and 5 being very unimportant. There are different crowdfunding platforms, which are also included in the model. The reward-based crowdfunding platform is left out of the model, to compare the equity-, lending- and donation-based crowdfunding in respect to reward-based

23

crowdfunding.

According to the study of Massolution (2012), the largest platform in crowdfunding is reward-based which represents 50% of the overall composition of the crowdfunding marketplace. For the database used in this research 22.14% of the respondents have funded a project concerning donation-based crowdfunding and 14.36% concerning equity- and lending-based crowdfunding. Evidence is found in this research that reward-based crowdfunding has been the largest platform with 215 respondents, which is equal to 51.58% of all respondents. This is approximately the same as what Massolution states, there is a small difference of 1.58%. In this thesis evidence is found that the level of financial contributions is the highest when it is related to equity- and lending-based crowdfunding. This is does not support the theory in Gerber et al. (2012), which state that contributors tend to give a higher amount of money when it concerns a reward-based crowdfunding project compared to other crowdfunding platforms. While the level of financial contributions in donation-based crowdfunding are most likely to be in a low category, equity-and lending-based crowdfunding are most likely to be in the highest category. Furthermore, evidence is found that the level of financial contributions of reward-based crowdfunding in this study is always lower than the level of financial contributions in equity- and lending-based crowdfunding. This is not in line with the theory in the related literature, which reject the first and the fourth hypotheses both at a significance level of 1%.

Besides the different crowdfunding platforms several factors, which are mentioned before, are examined whether they are important for the determination of the level of financial contributions. The first factor that will be discussed is the financial situation of the contributors. Figure 3 shows that the majority of the respondents have chosen the answer ‘very important’ for the factor ‘financial situation’. This is line with the finding in the research of Vliet (2012), where respondents pointed out that the financial situation was a important factor in determining the level of financial contributions. In this research it is found that there is a negative relationship between the importance of the financial situation and the level of financial contributions. When the importance of the financial situation increases people tend to give a lower

24

amount of money to crowdfunding projects. The results are only significant at the 10% level, which makes it not very strong. However, they are positive, so the third hypothesis can be rejected. In the research of Vliet (2011) it states that the contributors will first decide how much money they could miss looking at their financial situation in order to determine the level of financial contributions.

The factors ‘the amount of money that is necessary for the crowdfunding

project’ and ‘percentage that is already funded’ are seen as important factors in

the research of Vliet (2011). Contributors tend to give a higher amount of money when the importance increases and some contributors would even donate twice. For the database used, it is shown that the level of financial contributions increases when the factor ‘the amount of money that is necessary for the

crowdfunding project’ is seen as an important factor. It is significant at a 5% level,

which leads to rejection of the fifth hypothesis. Likewise, evidence is found that when the factor ‘the amount of money that is necessary for the crowdfunding

project’ is seen as an important factor the amount of money is 11.2% more likely

to be in the highest category of the level of financial contributions. This is all in line with the theory of Vliet (2011). On the other hand, in the research of Vliet (2011) state that the factor ‘percentage that is already funded’ is seen as an important factor which leads to a higher the level of financial contributions. In this research this does not hold. When the factor is important for the contributions they tend to give a lower amount of money instead of a higher amount of money. Besides that it is more likely to be in a low category of level of financial contributions. This leads to acceptance of the sixth hypothesis with a significance level of 10%.

‘The level of financial contributors of others’ is chosen as the least important factor and is not significant at any level. The factor has no significance relevance for the level of financial contributions, which supports the theory in the study of Vliet (2011). The determination of the level of financial contributions does not depend on ‘the level of financial contributors of others’. The factor ‘minimum amount of money that can be fund’ is chosen as the second least important factor to determine the level of financial contributions. In comparison with a neutral point of view, funders tend to give a higher amount of money when the factor is seen as unimportant factor, therefore it is most likely

25

to be in a high category of level of financial contributions. The factor is significant at a level of 5%.

The last important factor that will be discussed is the ‘return’. The level of financial contributions is higher when respondents find the factor ‘return’ unimportant. This means that the second hypothesis is accepted. Even though, there is a remarkable appearance. The factor is not significant at a significance level of 1%, 5% or 10%. This means that the factor ‘return’ has no significant influence on the level of financial contributions, which is very odd since it has been chosen as second most important factor to determine the level of financial contributions. Ignoring the significance, the second hypothesis is accepted. This is partly in line with the theory in the study of Vliet (2011), where the opinions amongst the crowd are extremely varied about the importance of the factor

‘return’. This can be seen as a explanation for its insignificance. Due to the small

size of data it is possible that the majority of the respondents find the factor

‘return’ unimportant to determine the level of financial contributions which

results in insignificance.

According this thesis it would be recommended to take into account the following factors for succeeding a crowdfunding project in funding. Firstly, the target amount of money is important. Contributors tend to give a higher level of financial contributions when the target is almost reached, which is also related to the factor ‘percentage that is already funded’. Furthermore, the return can be a determining factor. Even though the opinions about the importance amongst the ‘crowd’ are extremely varied it is always good to take this into account. Evidence in this thesis is also found that the biggest crowdfunding platforms are reward based compared to the other platforms.

It has to be noted that the data used is very small to provide conclusive evidence, so the results should be used as suggestive evidence as larger databases are needed to provide conclusive evidence. For further research it would be suggested to investigate other factors that can determine the level of financial contributions such as the relationship between the project initiator and the crowdfunders and the type of crowdfunding platform. Moreover, a valuable extensions of this study would be to examine how big the importance is of the target amount of money for a crowdfunding project in order to succeed.

26

References

Agrawal, A.K., Catalini, C., & Goldfarb, A. 2011. The geography of crowdfunding. National Bureau of Economic Research, Cambridge, MA, USA. Retrieved from

http://nber.org/papers/w16820

Belleflamme, P., Lambert, T., & Schwienbacher, A. 2011. Crowdfunding: Tapping the right Crowd. Corporate Finance. Retrieved from

http://www2.dse.unibo.it/dsa/seminari/610/Crowdfunding_BLS_13Feb2011.p df

Brabham, D. C. 2008. Moving the Crowd at iStockphoto: The Composition of the Crowd and Motivations for Participation in a Crowdsourcing Application. First Monday, 14(1), 1-90.

Bradford, C. S. 2012. Crowdfunding and the Federal Securities Laws Review. Columbia Business Law, 1, 51-67.

Gerber, E.M., Hui, J. S., & Kuo, P. 2012. Crowdfunding: Why people are motivated to post and fund projects on crowdfunding platforms. Northwestern University Creative Action Lab, Sheridan Drive, Evanston.

Giudici, G., Nava, R., Lamastra, C. & Verecondo, C. 2012. Crowdfunding: The New Frontier for Financing Entrepreneurship? Retrieved from

http://ssrn.com/abstract=2157429

Hemer, J. 2011: A Snapshot on Crowdfunding. Working papers firms and region, R1/2011.

Kleemann, F., Voß, G. G., & Reider, K. 2008. Un(der)paid Innovators: The Commercial Utilization of Consumer Work through Crowdsourcing. Science, Technology & Innovation Studies, 4(1), 5-26.

Kleverlaan, R., Klaver, F., Vliet, K. & Akker,P. 2013. Nationaal Crowdfunding Onderzoek 2013. Retrieved from http://www.crowdfundingonderzoek.nl/

Lambert, T. & Schwienbacher, A. 2010. An Empirical Analysis of Crowdfunding. Working paper, Université catholique de Louvain & University of Amsterdam Business School

27 Lee, S., DeWester, D., & Park, S. R. 2008. Web 2.0 and opportunities for small

businesses. Service Business, 2(4), 335-345.

Lévy, P. (1997). Collective Intelligence: Mankind’s Emerging World in Cyberspace. Cambridge, Perseus Books.

Massolution 2012. The Crowdfunding Industry Report: Market Trends, Composition and Crowdfunding Platforms. Retrieved from

http://www.crowdsourcing.org/document/crowdfunding-industry-report-

abridged-version-market-trends-composition-and-crowdfunding-platforms/14277

O’Reilly, T. 2007. What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software. Communications & Strategies, 65, 17-37

Schwienbacher, A. & Larralde, B. 2010. Crowdfunding of Small Entrepreneurial Ventures. Handbook of Entrepreneurial Finance, Oxford University Press, Forthcoming.

Singapore Entrepreneur, Initials. 2008. New source of money for start-ups:

crowdfunding. Retrieved from http://entrepreneur.com.sg/wordpress/?p=236 van Vliet, K. 2011. Crowdfunding: Waarom doen we mee? Master thesis University of

28

Appendix A

Relevant questions of the survey from the ‘Nationaal Crowdfunding Onderzoek 2013’

29

Appendix B

Coefficients of the ordered probit model

Dependent variable: Level_Financial_contributions

Log likelihood = -689.66201

Coefficients z-value

Donation-based crowdfunding -0.469 -3.17***

Equity-based crowdfunding 0.518 3.33***

Unknow crowdfunding platform -0.277 -0.89

Financial situtation rate 0 0.257 0.54

Financial situtation rate 1 (very unimportant) 0.238 1.02

Financial situtation rate 2 0.037 0.15

Financial situtation rate 4 0.260 1.69*

Financial situtation rate 5 (very important) 0.172 1.17

Importance return rate 0 -0.662 -1.35

Importance return rate 1 (very unimportant) 0.175 0.86

Importance return rate 2 -0.08 -0.45

Importance return rate 4 0.214 1.35

Importance return rate 5 (very important) 0.086 0.53

Target amount of money rate 0 1.116 1.6

Target amount of money rate 1 0.197 0.98

Target amount of money rate 2 -0.214 -1.23

Target amount of money rate 4 0.200 1.4

Target amount of money rate 5 (very important) 0.558 2.82***

Percentage raised rate 0 0.07 0.12

Percentage raised rate 1 (very unimportant) 0.036 0.17

Percentage raised rate 2 0.115 0.68

Percentage raised rate 4 -0.389 -2.66**

Percentage raised rate 5 (very important) -0.090 -0.37

Contributions others rate 0 -0.380 -0.72

Contributions others rate 1 (very unimportant) -0.118 -0.68

Contributions others rate 2 -0.228 -1.37

Contributions others rate 4 -0.023 -0.12

Contributions others rate 5 (very important) 0.399 1.32

Minimum amount rate 0 0.509 1.64

Minimum amount rate 1 (very unimportant) 0.347 2.05**

Minimum amount rate 2 0.141 0.86

Minimum amount rate 4 -0.123 -0.74

Minimum amount rate 5 (very important) -0.195 0.98

Category 1 -1.400 Category 2 -0.224 Category 3 0.482 Category 4 1.002 Category 5 1.358 Category 6 1.595 Number of observation 411 Prob > chi2 0.0000 Psuedo R² 0.0660

*, **, *** indicates significance levels of 10%, 5% and 1% respectively.