Crowdfunding : a study about the composition of investors in crowdfunding projects on CrowdAboutNow

(1)

Crowdfunding,

A study about the composition of investors in crowdfunding projects on CrowdAboutNow

Student: Philip Pril Student ID: 10150390 Supervisor: Dr. Joeri Sol

Bachelor Thesis

Faculty of Economics and Business Specialization Business Studies

Amsterdam, 5 may 2015

Abstract

This study examined 29 successful crowdfunding projects on de platform CrowdAboutNow. Specifically, I tested whether the composition of investors was correlated with changes in target amount in crowdfunding projects. Relations from investor to initiator change with changes in target amount: Friends become a less important source of funding as targets go up, while ‘professional’ investors become more prevalent. Moreover, the amount of invested capital of individual investors increases with interest rates. On top of that, the composition of investors with respect to gender also changes as target amount varies, with men being more prevalent in projects with high interest rates. Over all, we find enough evidence to conclude that changes in target amount leads to significant changes in the composition of investors in crowdfunding projects on CrowdAboutNow.

(2)

2

Statement of originality

This document is written by Student Philip Pril who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is 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 supervision of completion of the work, not for the contents.

(3)

3 Contents Preface v 1 Introduction 4 2 Literature review 6 2.1 What is crowdfunding? . . . . . . 6 2.2 Advantages of crowdfunding . . . . . . 7 2.3 Types of crowdfunding . . . . . . 8

2.4 Reason for crowdfunding platform: CrowdAboutNow . . . . . . 10

2.5 Composition of investors . . . 10

3 Methodology and Data 12

4 Results 4.1 Descriptive data . . . 16

4.2 Correlations . . . 17

4.3 Hypothesis testing . . . 18

5 Conclusions, limitations and recommendations 5.1 Conclusions . . . . . . 24

5.2 Limitations and recommendations . . . 25

6 Bibliography 26 7 Appendix A (pie charts from all individual projects) 28

(4)

4

List of figures and tables

Figure 1: The complexity of crowdfunding involving intermediaries . . . 9 Figure 2: Selected projects per range in target amount . . . 13 Figure 3: 29 Selected projects . . . 14 Figure 4: Example of data collection of one individual investor ‘De Olijfboom’ . . . . 15

Table 1: Mean of Male and Female investors and all dummy variables.. . . 16

Table 2 : Average invested, investors and interest (minimum, maximum) . . . 17

Table 3: Correlations between Target amount and all relations from investor to initiator . 18

Table 4: Correlation between Interest and Average amount invested. . . . 21

(5)

5

Chapter 1 Introduction

Raising financial capital is an obstacle encountered by many entrepreneurs. On top of that, the credit crisis in 2008 has put pressure on bank lending to startups and small and medium sized enterprises. A novel and popular way of raising capital emerged: crowdfunding.

Crowdfunding is the use of an external source, such as internet, to raise money through small contributions from a large number of investors (Bradford, 2012). Collecting small amounts of money from a large number of people has a rich history in many domains. For example, Mozart and Beethoven financed their concerts and publications of new music compositions with money that they raised via advanced subscriptions from people that where interested (Hemer, 2011). Even the election campaign of President Barack Obama in 2008 was crowdfunded. For this campaign almost 4 million people raised together $750 million to support his campaign (Bradford, 2012). Furthermore, the Statue of Liberty in New York is partly crowdfunded through small financial contributions from French and American people because the US government had been unable to raise $250.000 (nowadays $6.3 million) for a granite plinth for the statue.1 More recently, a video game named Star Citizen has funded $65 million from 700,000 backers for production of the game and is so the most successful crowdfunding project thus far. 2 The amount funded for the Star Citizen game indicates the popularity of crowdfunding nowadays.

The recent increase in popularity and awareness to use crowdfunding to obtain capital is in favor of entrepreneurs, who traditionally have had great difficulty obtaining financial capital. In this way, they have access to anyone in the world with a computer and spare cash to invest. Besides raising capital, crowdfunding can also serve as a tool for testing

marketability (Jegeleviciute and Valancience, 2013). Popularity of crowdfunding is also owed to intrinsic motivation among funders, they enjoy the possibility to contribute to crowdfunding projects they believe in, even if they can invest only small amounts. Also the government appreciates crowdfunding activities due its positive effects on the economy where crowdfunding creates jobs, fostering economic recovery and contributes to innovation. The purpose of this bachelor thesis is to describe and analyze the new financing

1

Source: http://www.bbc.com/news/magazine-21932675 (retrieved on May 19, 2015) 2

Source: http://www.theguardian.com/technology/2014/dec/03/star-citizen-crowd-funding-record-65m-game (retrieved on May 19, 2015).

(6)

6 phenomenon crowdfunding and to search for differences in composition of investors in crowdfunding projects on the platform: www.Crowdaboutnow.nl. The different factors in this study vary from changes in return (interest), relation of investor to entrepreneur, amount of individual investment of investor and gender differences of investors. But given the novelty of the concept, the literature on crowdfunding is rather small. Little research has been done to understand the assumptions of crowdfunding and the effects of the concept in the long run. Also basic academic knowledge of its dynamics is still lacking (Mollick, 2014). This thesis is unique, with great impact of crowdfunding because till now, no research has been done on the composition of investors in different target amount crowdfunding projects. Outcomes of this thesis will be in importance for entrepreneurs who want to start a crowdfunding project. This research will give initiators of a project a better insight on which investors to expect and which Target amount is most appropriate for achieving their target goal. The data for this thesis is generated from the crowdfunding website of CrowdAboutNow. I will examine the correlation between the composition of investors and change in variables, gender differences of investors, amount of individual investment of investors, relation of crowdfunder to

crowdseeker and differences in return, of crowdfunding projects. Based on the research purpose, the research question of this thesis can be formulated as follows:

‘’Does the composition of investors change with changes in target amount in crowdfunding projects?’’

The remaining part of this thesis is structured as follows. The second chapter contains the literature review where existing literature on this research will be discussed and all

hypotheses are listed. The methodology that is going to be used and the data collection will be presented in chapter 3. The results of this thesis will be presented in the fourth chapter. Finally, chapter 5 contains the conclusion.

(7)

7

Chapter 2 Literature review

2.1 What is crowdfunding?

‘’Crowdfunding’’ is derived from the broader concept of ‘’crowdsourcing’’, which describes the process of outsourcing specific tasks to a ‘’crowd of people’’, generally through an open source like the internet (Kleemann Vos & Rieder, 2008). Firms also use crowdsourcing to obtain ideas, services, solutions and feedback through the ‘crowd’ for developing and improving their activities (Belleflamme et al., 2013, p. 3). In reaction of crowdsourcing and the need for external financial capital in society, a new phenomenon emerged: crowdfunding. The main goal of crowdfunding is to obtain funds from the ‘crowd’. Or in other words, the objective of crowdfunding is to obtain money (Hemer, 2012).

Practioners in the crowdfunding business use different definitions for crowdfunding, concentrating on different (parts of) aspects of crowdfunding. For example, Lin and

Viswanathan (2013, p. 17) restrict their rather short definition of crowdfunding to internet-based internet-based resource gathering:

“Internet-based peer-to-peer lending.”

Whereas Kleeman et al. (2008, p. 6) only use the term internet in their definition as example of an open source:

“Crowdfunding involves an open source, like internet, for the provision of financial

resources either in the form of donation, in exchange for the future product or some other form of reward to support initiatives for specific purposes.”

The definition of crowdfunding by De Buyser et al. (2012, p. 9) is more focusing on the socially constructed part of funding:

“Crowdfunding is a joint effort of several individuals who are part of the network and pool their resources to support projects initiated by other people or organizations. This activity is usually done through or with the help of Internet. Thus, certain projects are financed by small contributions from a large number of individuals, allowing entrepreneurs to use their social networks to raise capital.”

And where The Europoean Crowdfunding Network (2012-2013) give a more process-based description of crowdfunding:

(8)

8

investments from a large audience of supporters in exchange for equity or liabilities carrying financial returns or other non-financial rewards, where supporters are people or

organizations who network, usually via the internet, to jointly support other people or organizations.”

As stated above, there is no one correct definition of crowdfunding. All these definitions shed light on the phenomenon crowdfunding from different angles and are rather compulsory than elusive. For the simplicity, I will use the following definition of crowdfunding in the present thesis:

“Crowdfunding is a mechanism for raising financial capital through a joint effort of several individuals via an open source like internet.”

2.2 Advantages of crowdfunding

Raising financial capital is the main but not the only goal and advantage of crowdfunding. For example, crowdfunding can provide the entrepreneur valuable signals about the market potential of the product/project that they want to run (Schwienbacher and Larralde, 2010, p. 7), potentially leading to having many potential customers/consumers. Also De Buyser et al. (2012), argue that entrepreneurs use crowdfunding for market research and marketing, next to using it for financing their project. For financing their project but also for market research and marketing. The whole mechanism of crowdfunding is based on attracting a big number of individuals, who find an idea interesting, worth their investment and time (Jegeleviciute and Valancience, 2013). If many individuals believe in a project, leading it to be crowdfunded, this project is likely to succeed. Contributions of investors or an audience are valuable, and it seems that they are sometimes even more effective in solving problems than the entrepreneur (Schwienbacher and Larralde, 2010). Furthermore, according to Gajda and Walton (2013), the main advantage of crowdfunding is reflected in the fact that financiers of the projects are also potential customers and ambassadors of the project and will indirectly assist in its promotion through their own networks.

Next to these additional goals of crowdfunding, there are also positive side effects that emerge from crowdfunding a project. Funding a project through an open source like internet removes geographic barriers to invest. Belleflamme et al. (2013) investigated the

geographical origin of the consumers who invested in a crowdfunding platform SellaBand. They measured an average distance of almost 5,000 km between entrepreneurs and investors,

(9)

9 suggesting that geographical distance in crowdfunding plays not a significant role for

investment. However, does not always have to be geographically limitless. In contrast, crowdfunding is sometimes also used as a tool for people to invest in their own communities (According to Kitchens and Torrence, 2012) or “own geographical area” (Belleflamme et al., 2013). In this way, investors specifically focus on projects within their geographical

community.

2.3 Types of crowdfunding

According to (Giudici et al., 2012) there are four different forms of crowdfunding: reward-based, donation-reward-based, equity-based and loan-based crowdfunding. These different types of crowdfunding concepts are based on the type of return the investor wants for his or her financial capital.

Funders who choose for supporting a crowdfunding project that provides a pre-ordered product in return, want the product in return of their financial capital. According to Mollick (2013), pre-ordering of the product is a form of financial reward-based

crowdfunding. Funders receive a product rather than a share in a company in return for their financial contributions (Ahlers, Cumming, Gunther, & Schweizer, 2012. p. 8). This, because, reward-based crowdfunding offers a non-financial return in exchange for investment. The investors in pre-ordered based crowdfunding projects get in this way the product before they enter the market, and so the investors are treated as early consumers.

With donation-based crowdfunding, funders donate their financial capital to the project. This is in contrast with the pre-ordered based crowdfunding where investors want something in return for supporting the project. Investors in donation-based crowdfunding projects do not want and do not expect anything in return (Ahlers, Cumming, Gunther & Schweizer, 2012, p. 8).

The third type of crowdfunding is through equity-based crowdfunding. With equity crowdfunding, entrepreneurs make an open call to sell specified amounts of equity or bond-like shares in a company, hoping to attract a large group of investors (Ahlers, Cumming, Gunther & Schweizer, 2012, p. 1). In this way of crowdfunding, the investor is bonded to the company, also when the project is finished.

The last type of crowdfunding is the lending model or loan-based crowdfunding model. In this type of crowdfunding, funds are offered as a loan (Mollick, 2013). Funders

(10)

10 will receive an interest payment, which is contractually agreed. In this way, higher amount of interest payments indicate that there are more investors willing to invest in the project

(Massulution, 2013). The crowdfunding platform “CrowdAboutNow” is a platform that uses such a loan-based crowdfunding model.

The process behind these different forms of crowdfunding concepts may be complex if large numbers of investors have to be managed (see figure 1 below). Many initiators of ventures are inexperienced, not capable or simply have not time for managing a

crowdfunding project themselves and prefer a so-called ‘intermediary’.

Figure 1: The complexity of crowdfunding involving intermediaries

(Source: Hemer et al., 2011)

In accordance with the origin of the crowdfunding, these intermediary services tend to be Web- and software-based and therefore call themselves ‘’Crowdfunding platforms’’ (Hemer, 2011, p. 10). And as stated above, use different types of crowdfunding concepts. Figure 1 illustrates how Crowdfunding platforms coordinate with all interested parties. As most initiators of crowdfunding projects start a project only once, or a couple of times in their lifetime, it is not likely that crowdseekers will generate experience in starting projects. However, crowdfunding platforms do gain experience and professionalism through the

(11)

11 routine of many projects. These intermediaries act as neutral facilitators for both,

crowdseekers and crowdfunders and provide the opportunity for people to present their projects to the whole world, without a geographical restriction.

2.4 Reason for crowdfunding platform: CrowdAboutNow

CrowdAboutNow is a Dutch loan-based crowdfunding platform. As stated above, funders on this platform will receive an interest payment on the invested capital. The reason for the platform CrowdAboutNow is twofold. First, on this platform the relation between the

investor and entrepreneur is feasible. The investor can choose if he or she makes her relation with the entrepreneur feasible. The investor can choose for the following relation:

anonymous, friend, family, customer/investor, colleague, fan or neighbor. In this way, the composition of investors is testable within and between successful projects, and research can be done between different target amounts of projects. Next to this, other variables like returns (interest rates), amount of individual investments and gender of investor are feasible and therefore testable. Secondly, the size of the platform fits perfect within the aim of this

research. There are enough successful projects to test for correlation between compositions of investors and to test if this changes as target amount of projects change.

2.5 Composition of investors

This study will examine which factors determine the correlation between the compositions of investors and change in variables, as stated above, in different target amounts of

crowdfunding projects. The relationship between funders and founders varies by context and the nature of the funding effort (Belleflamme et al., 2012). These variations in funding effort can be addressed to the four types of crowdfunding concepts but also in variables like return on investment, gender differences or differences in target amounts of projects.

A previous study of Vliet (2011) has mainly focused on the motivation of the crowdfunders. That is, Vliet found out that there are four different motives for the crowd to determine to invest and the level of investment. One of the reasons is the target amount of the project. Some people indicate that the target amount was very important to them for the choice of investing in the project. This is the reason why the first hypothesis is as follow:

(12)

12

H1 = Composition of investors changes with an increase in the target amounts in crowdfunding projects; specifically, I expect the share of (a) friends, (b) family, (c)

colleagues, and (d) neighbours to decrease with higher target amounts, while the share of (e) businesspeople, (f) fans, and (g) anonymous investors is expected to rise with higher target amounts.

Friends and family of the entrepreneur disproportionately invest early in the funding cycle, generating a signal for later funders through accumulated capital. The asymmetry between friends and family and others in terms of funding behavior is strongest for the first investment decision but subsequently fades when funders are able to monitor the progress of the project (Agrawal, Catalini, and Goldfarb, 2011). Therefore, friends and family play a bigger role in composition of investors in projects with low target amounts. More general, my first

hypothesis follows from the notion that the entrepreneur her social network is fixed, and with an increasing target amount, the entrepreneur should increasingly appeal to members outside of her network. Moreover, the importance of the type of the return can be a determining factor. Whereas some people say that the return they receive is an inessential factor, others say that the return, for them, is the determining factor. In other words, there are contributors that do not want to fund a project when there is no return involved (Vliet, 2011). Therefore the second hypothesis is as follow:

H2 = The higher the interest payment, the higher the invested amount of investors.

Several studies have found that women invest more conservatively than men (Bajtelsmit and VanDerhei, 1996) and that woman are more risk averse (Jianakoplos and Bernasek, 1996). Also Schubert, Brown, Gysler and Brachinger (1999) state that women are more risk averse than men. Therefore, the fourth and last hypothesis is stated as follow:

H3 = In projects with high amount of interest on return (higher risk to invest in) is dominated by men.

(13)

13

Chapter 3 Methodology and descriptive data

3.1 Methodology and descriptive data

To search for differences in composition of investors with varying target amounts in crowdfunding projects, I use the loan-based crowdfunding platform CrowdAboutNow. Difference in Target amount is not the only variable that plays a role in this study. Other quantitative variables include changes in return (interest), amount invested of individual investor, relation of investor to initiator of the project and gender of investor play a role in the composition of investors in crowdfunding projects.

I am going to test the hypothesis through inspection of the correlations between the variables in SPSS. First, I am going to collect the appropriate 29 projects for this study according to the Target amounts. Second, I will select all general information from these selected projects through a Crawler. Third, I will collect all variables for testing the hypothesis, and insert these in Microsoft Excel by hand. Fourth, I will open the Excel file in SPSS, the program that I use to analyze the data.

As stated above, I start with randomly selecting 29 successful projects with varying target amounts (see figure 2). I will do that by randomly selecting 3 successful projects for each of the following ranges of target amounts: €0,- to €9.999; €10.000,- to €19.999; €20.000,- to €29.999…€70.000,- to €79.999; > €80.000,-. When groups do not have 3 successful crowdfunding projects in a particular range, I select all projects.

(14)

14 Figure 2: Number of selected projects per range in target amount

After selection of all 29 appropriate projects, I will collect all general data of every project (i.e. name; start-date; end-date; target goal; number of investors; return on investment; duration; and success ratio) through a single-linkage Crawler. With this Crawler, webpages are instantly turned into data (see figure 3).

3 x (successful) projects with target amount from €0,- to €9.999

4 x (successful) projects with target amount from €10.000,- to €19.999,- 5 x (successful) projects with target amount from €20.000,- to 29.999,- 3 x (successful) projects with target amount from €30.000,- to €39.999,- 4 x (successful) projects with target amount from €40.000,- to €49.999,- 3 x (successful) projects with target amount from €50.000,- to €59.999,- 2 x (successful) projects with target amount from €60.000,- to €69.999,- 2 x (successful) projects with target amount from €70.000,- to €79.99,- 3 x (successful) projects with target amount > €80.000,-

(15)

15 Figure 3: Information from 29 projects

Project Name Project started Ended Target goal Investors Return on investment % duration/

months success%

30MLESPRESSO 14/08/2013 01-Oct € 6,000.00 16 15% yield (5% / year) 36 100%

Zon Lambertusschool 14/04/2015 14-Jun € 7,200.00 30 10% yield (0,9% / year) 132 100%

Junk Food 10/11/2014 31-Jan € 7,500.00 50 10% yield (5% / year) 24 200%

De Olijfboom 28/11/2013 27-Jan € 15,000.00 19 15% yield (5% / year) 36 100%

Vonnies Rijsttafel 16/05/2014 31-Jul € 15,000.00 19 12% yield (4% / year) 36 167%

Piogg 19/10/2011 20-Dec € 16,000.00 52 12% yield (4,2% / year) 34 138%

TOET 26/06/2013 24-Jul € 19,000.00 49 13.5% yield (4,3% / year) 38 100%

Eat with bravoure 01/05/2014 12-Jun € 20,000.00 40 15% yield (5% / year) 36 100%

VROG 16/07/2014 16-Oct € 25,000.00 34 15% yield (5% / year) 36 101%

BsaB 28/03/2014 01-Jul € 25,000.00 42 15% yield (5% / year) 36 130%

Theekamer PIEN 19/11/2013 14-Feb € 25,000.00 65 15% yield (5% / year) 36 100%

Rhythm & Music

Muziekschool 16/10/2014 16-Jan € 27,500.00 60 17.5% yield (3,5% / year) 60 112%

CleverCoin 03/09/2014 04-Oct € 34,000.00 36 12% yield (8% / year) 18 100%

In De Aap 07/10/2013 17-Jan € 35,000.00 31 12% yield (4% / year) 36 100%

Queens Royal Bakery 23/05/2014 01-Aug € 37,500.00 67 25% yield (5% / year) 60 126%

Iaddress 25/06/2014 25-Sep € 40,000.00 60 19.5% yield (6,5% / year) 24 112%

Het Massagehuys 07/03/2014 21-May € 45,000.00 64 15% yield (5% / year) 36 103% Bake & Moore 14/11/2014 30-Dec € 49,500.00 105 27.5% yield (5,5% / year) 60 136%

Savvy 22/05/2014 01-Aug € 50,000.00 25 15% yield (5% / year) 36 100%

Lakeside Zwolle 24/10/2014 24-Jan € 50,000.00 80 9% yield (3% / year) 36 100%

Lola 06/08/2013 01-Nov € 50,000.00 71 18% yield (6% / year) 36 127%

Yoghurt Barn 31/05/2013 12-Jul € 50,000.00 48 15% yield t (4,7% / year) 38 100%

Brasserie Midi 29/07/2014 30-Sep € 55,000.00 88 22.5% yield (5% / yea) 54 100%

Buro Lima 16/05/2014 17-Jul € 66,600.00 39 33% yield (3,3% / year) 120 105%

Sodafabriek 02/09/2013 01-Dec € 75,000.00 96 12% yield (4% / year) 36 100%

Nijmegse Boekhandel 28/02/2014 15-May € 75,000.00 512 25% yield (5,1% / year) 59 200% Jacketz Oud-West 16/05/2014 01-Aug € 98,000.00 103 15% yield t (4,3% / year) 42 124%

Tramrestaurant

Hoftrammm 01/12/2014 01-Mar € 100,000.00 182 25%v yield (5% / year) 60 106%

Stichting Administratiekantoor Crowdfunding Anno12

(16)

16 After retrieving all general data from the projects, variables of every individual crowdfunding project the following variables are collected: name of investor; amount invested in the project by investor; relation of investor to initiator; and gender of investor.

Whereas the variable ‘invested amount of the investors’ is always public, other

variables such as name of investor, relation of investor to crowdseeker and gender of investor are optical and so not always known. As an investor you have a couple of options to choose from when it comes to the relation from investor to initiator. Options to choose are:

Anonymous (1), Friend (2), Business/Consumer (3), Fan (4), Colleague (5), Neighbour (6) and Family (7). I collect the variable gender through linking the name of the investor to the gender man or woman. If the name of an investor is unclear of its gender, I will report the gender as ‘unknown’.

Example: See figure 4 (below) for an example of how the data from all 2359 investors, spread over 29 projects, is collected. In this example the project ‘De Olijfboom’ is used. As you can see in figure 3, the project has the following data: started 28/11/2013; ended in 27-Jan; target goal of €15.000; 19 investors; 5% yield a year; duration of 36 months; and a success ratio of 100%. For this example I illustrate the data collection with one investor: Jacqueline Nijhof (Name investor). She invested €250,- (Amount €) in the project and is a friend of the crowdseeker (Relation* = 2 = friend). The name Jacqueline Nijhof indicates that the investor is a woman, leading to a ‘0’ in the last column (men =1 and woman = 0).

Figure 4: Example of data collection of one individual investor ‘De Olijfboom’

After collecting all the data, I calculated the averages of amounts invested in percentages. I did this by creating dummy variables of the relation form investors to initiator and gender of investors, whereafter I averaged the amount invested per variable. Then, I opened the Excel data-file in SPSS, which I used to display a correlation matrix.

De Olijfboom Name investor Amount (€) Relation* m/v (1/0)

1 Jacqueline Nijhof € 250.00 2 0 *1 = unknown (anonymous) *2 = friend *3 = businesspeople (klant/belegger) *4 = fan *5 = colleague *6 =neighbour *7 = family

(17)

17

Chapter 4 Results

4.1 Descriptive Data

The aim of the present study was to determine whether the composition of investors varies with change in target amounts of crowdfunding projects on CrowdAboutNow. To do this, I selected data of 29 crowdfunding projects (with varying target amounts) with a total of (N=)2319 investors and.tested for correlation between target variables.

The selected sample consisted of 568 males and 363 females. The remaining

percentage of investors is not displayed on CrowdAboutNow and therefore unknown (59%).

Table 1: Mean of Male and Female investors and all dummy variables.

Statistics

Male Female Anonymous Friend Business Fan Colleague Neighbour Family

N 29 29 29 29 29 29 29 29 29

Mean .245 .156 .367 .206 .183 .107 .023 .025 .079

Std. Deviation .159 .179 .157 .157 .134 .101 .038 .069 .130

As can be seen in figure 4, there are 7 types of relationship a crowdfunder can have with the initiator of the crowdfunding project. The most common relation from crowdfunder to crowdseeker (modus) in our sample of investors is Anonymous (36,75% of the sample), followed by Friends (20,64%), Businesspeople/Costumers (18.38%), Fan (10,77%), Family (8,00%), Neighbour (2,50%) and Colleague (2,40%).

In this study I investigated 2319 investors in 29 projects. With an average amount of 79,96 ≈ 80 investors, the highest amount of investors in a project is 512 and the lowest amount of investors in a project is 16. The average amount invested is €806,64. The mean of the average amount of interest for all 29 projects is 4,66%, with the highest interest payment of 8% a year and the lowest 1% a year.

(18)

18 Table 2: Average invested, investors and interest (minimum, maximum).

Statistics

Average Invested Investors Interest

Mean 805.008 79.966 .046

Minimum 240.00 16.00 .01

Maximum 2355.93 512.00 .08

4.2 Correlations

The main focus of this study is the composition of investors in different crowdfunding projects. Correlations with a significance level of =.05 are displayed in Table 3. As can be seen, the composition of investors differs with variance in target amount (dependent

variable); relation from crowdfunder to crowdseeker; average invested amount of money of individual investors; amount of interest and gender of investors (independent variables). As can be seen in table 3, there is a positive relationship between crowdfunder and initiator of the crowdfunding projects. Thus, when Target amount is high, crowdfunders mostly are Anonymous and Fans. The relationships between Target amount and Friend, Businesspeople, Colleagues, Neighbours and Family, are negative. This means that the higher the target amount is, the less likely it is crowdfunded by friends, buisinesspoeple, colleagues, neighbours, and family. As you can see in the table, not all positive and all negative

correlations are equivalent. Correlations with higher amounts are stronger correlated than correlations with lower amounts.

Similar relations and correlations can be found among the other variables. In the following section, I will explain interactions between the variables more detailed. That is, in the following section, the hypotheses will be tested statistically.

(19)

Table 3: Correlations between Dummies (relation from crowdfunder to initiator of project) and Target amount. Total

collected

Anonymous Friend Business Fan Colleague Neighbour Family Average invested

Total collected Pearson Correlation 1 .166 -.347* .353* .123 -.098 -.146 -.131 .003 Sig. (1-tailed) .194 .033 .030 .263 .306 .225 .250 .495 N 29 29 29 29 29 29 29 29 29 Anonymous Pearson Correlation .166 1 -.300 .030 -.413* -.211 -.202 -.346* -.161 Sig. (1-tailed) .194 .057 .438 .013 .136 .147 .033 .202 N 29 29 29 29 29 29 29 29 29 Friend Pearson Correlation -.347* -.300 1 -.521** -.277 -.006 .050 -.101 -.016 Sig. (1-tailed) .033 .057 .002 .073 .487 .399 .302 .468 N 29 29 29 29 29 29 29 29 29 Business Pearson Correlation .353* .030 -.521** 1 -.035 -.096 -.259 -.255 .348* Sig. (1-tailed) .030 .438 .002 .428 .310 .087 .091 .032 N 29 29 29 29 29 29 29 29 29 Fan Pearson Correlation .123 -.413* _-.277 _-.035 ₁ _.464** _.106 _-.114 _.057 Sig. (1-tailed) .263 .013 .073 .428 .006 .291 .278 .384 N 29 29 29 29 29 29 29 29 29 Colleague Pearson Correlation -.098 -.211 -.006 -.096 .464** ₁ _-.111 _-.228 _.088 Sig. (1-tailed) .306 .136 .487 .310 .006 .284 .117 .325 N 29 29 29 29 29 29 29 29 29 Neighbour Pearson Correlation -.146 -.202 .050 -.259 .106 -.111 1 -.130 -.290 Sig. (1-tailed) .225 .147 .399 .087 .291 .284 .251 .064 N 29 29 29 29 29 29 29 29 29 Family Pearson Correlation -.131 -.346* -.101 -.255 -.114 -.228 -.130 1 -.084 Sig. (1-tailed) .250 .033 .302 .091 .278 .117 .251 .332 N 29 29 29 29 29 29 29 29 29 Average Invested Pearson Correlation .003 -.161 -.016 .348* .057 .088 -.290 -.084 1 Sig. (1-tailed) .495 .202 .468 .032 .384 .325 .064 .332 N 29 29 29 29 29 29 29 29 29

*. Correlation is significant at the 0.05 level (1-tailed). **. Correlation is significant at the 0.01 level (1-tailed).

(20)

4.3 Hypothesis testing

The first hypothesis, composition of investors changes with an increase in the target amounts in crowdfunding projects, I expect that the share of variable (a) Friend investors decreases with higher target amounts. The relationship between Friends and Target amounts is yielded significant (r=-.347, p<0.05) and therefore hypothesis 1(a) is confirmed. A moderate negative relationship between Friends and Target amounts can be seen as follows: when Target amount of crowdfunding projects are raising, the composition of Friend investor’s decreases. So for ’low’ Target amounts, more Friends are investing in the project. For projects with ‘higher’ Target amounts, fewer Friends are investing. An explanation of this outcome could be that the relationship can be seen in public and so everyone can see that you as an investor have contributed to the project.

The second hypothesis is that composition of investors changes with an increase in the target amounts in crowdfunding projects; specifically, I expect that (b) Family investors, decrease with higher Target amounts. Family investors and Target amount are, according to this study, not correlated to each other (r=-.131, p>0.05). According to this outcome, the variable of Family investors in all the projects stays invariably linear as target amounts raises. Possibly, investors that are Family with the initiators want to support the project and the initiator, rather than making a financial profit of it (Agrawal, Catalini, and Goldfarb, 2011). This is also in line with Lee and Persson (2013) who state that family and friends do not want to make money on initiator. So for that reason, the variable Family investor stays relatively constant over the projects. For these reasons, hypothesis 1(b) is not confirmed.

The third hypothesis, composition of investors changes with an increase in the target amounts in crowdfunding projects; specifically, I expect the share of (c) Colleagues, to decrease with higher target amounts. Investors that are Colleagues of the crowdseeker on CrowdAboutNow are according to this study not correlated with Target amount (r=-.098,

p>0.5). This outcome can be explained: colleagues are basically co-owners of the project.

They have, together with the initiator of the project, similar interests in the project and want the project to succeed. Colleagues start together with the entrepreneur, the project and for that reason it would be strange that there would be a correlation between Colleagues and Target amount. Colleagues and imitators of the project start a crowdfunding campaign for raising financial capital, not for investing into their own projects. For that reason, hypothesis 1(c) is not confirmed.

(21)

21 The fourth hypothesis, the composition of investors changes with an increase in the target amounts in crowdfunding projects; specifically, I expect the share of (d) Neighbours to decrease with higher target amounts. The composition of Neighbour investors (to

entrepreneur) does not change witch change in Target amounts of crowdfunding projects on CrowdAboutNow (r=-.146, p>0.05). It can be explained why there is no correlation. Besides the fact that the term Neighbour of the initiator is a vague concept, investors that call

themselves Neighbour are dependent to the location where the project is going to be placed. Differences between project have to be made on behave of industries/products of the project. Or as (Agrawal, Catalini, and Goldfarb, 2011, p. 1) state: ‘Distance and local nature of projects play a role in crowdfunding’. For instance when you start a crowdfunding project for a sport article or you start a project for a restaurant in Amsterdam, for the second project can be expected that there are more Neighbour investors in the project. Linking this information to this study, hypothesis 1(d) is not confirmed.

The fifth hypothesis, composition of investors changes with an increase in the target amounts in crowdfunding projects; specifically, I expect the role of (e) Businesspeople to rise with higher target amounts. Analysis yielded a significant positive correlation between Businesspeople/Consumers and Target Amount (r=.353, p<0.05). As explained in the first paragraph (but vice versa) this positive relationship can be seen as follows: when Target amount of crowdfunding projects on CrowdAboutNow raises, composition of

Businesspeople/Consumers investors also (moderate) increases. This effect can be explained through the fact that projects with ‘high’ Target amounts, attract much investors and thus also much consumers. Businesspeople see the project as an investment rather than a contribution to the project for helping to attain the goal of the initiator. Therefore, the positive correlation can be explained by the negative correlation between Businesspeople and Friends (r=-.521), indicating that businesspeople do not invest into the project because of granting (i.e. what Friends do according H1(b)) but because of making money out of it. All this proves hypothesis 1(e).

The sixth hypothesis, composition of investors changes with an increase in the target amounts in crowdfunding projects; specifically, I expect the share of (f) fans investors is expected to rise with higher target amounts. This hypothesis is not supported by the statistical tests.The relationship between investors that are Fan of the project and Target amount are unexpected not correlated (r=.123, p>0.05). Although there is a correlation, this (r=.123) can be seen as no or a negligible relationship between Fan investors and Target amount and so hypothesis 1(f) is not confirmed. The relationship of Fan is pretty vague, because it is not

(22)

22 clear when an investor considers itself as being a fan. This could be one of the reasons why there is no correlation from Fan to Target amount. According to the literature, many investors include themselves as a Fan investor although they have an anonymous or Business

relationship with the initiator of the project (Agrawal, Catalini, and Goldfarb, 2011).

Accordingly, a fan can be any other contributer, leading to a non-significant relation between Fan and Target amount.

The seventh hypothesis, composition of investors changes with an increase in the target amounts in crowdfunding projects; specifically, I expect that the variable anonymous investor rises with higher target amounts. Unexpectedly it turns out that Anonymous investors (in relation to entrepreneur) and Target amounts of projects in this study are not correlated with each other(r= .166, p>0.05). A correlations between Anonymous and Target amount of r=.166 can be seen as no or a negligible relationship and therefore hypothesis 1(g) is not confirmed. According to this outcome, the variable of anonymous investors does not change over different amounts of target amounts. This outcome can be explained by the concept of registration the relationship from investor to entrepreneur on CrowdAboutNow. Investors can choose to publish the relationship they have with the entrepreneur. It might be that an investor does not want the initiator to see that they have invested in the project. Accordingly, his/her contribution is stated as Anonymous, whereas the relation between him and the crowdseeker can fall under one of the other categories (e.g. friend, family, neighbor).

Table 4 – Correlation between Interest and Average amount invested.

Correlations between Interest and Average amount invested (individual investor)

Interest Average_Invested Interest Pearson Correlation 1 .519** Sig. (1-tailed) .002 N 29 29 Average_Invested Pearson Correlation .519** 1 Sig. (1-tailed) .002 N 29 29

(23)

23 Table 4, above, shows the correlation between Interest of the 29 projects and Average amount invested per individual investor of all projects. I expected (H2) that higher interest payments (return on investment) would lead to higher invested amounts of money from investors. As can be seen in table 5, the correlation between Interest and Average amount invested is correlated (r=.519, p<0.01), significant. These findings are in line with Tobin (1956) who states that higher interests attract investors to invest higher amounts of money. A highly significant correlation of r=.519 can be seen as follows: when Interest of a project raises, it will strongly associate with raise in Average amount invested raises and vice versa. This study made clear that when initiators of a project want high average invested amounts of money from investors, it has to have high returns on investments (compared with low interest payments). In this way it will attract investors that are willing to invest more into the project than projects with lower interest payments. All this proves that hypothesis 2 is confirmed.

Table 5 – Correlation between Interest and Male investors.

Correlations Interest and Male investors

Interest Male Female

Interest Pearson Correlation 1 .312* -.222 Sig. (1-tailed) .050 .124 N 29 29 29 Male Pearson Correlation .312* 1 -.135 Sig. (1-tailed) .050 .243 N 29 29 29 Female Pearson Correlation -.222 -.135 1 Sig. (1-tailed) .124 .243 N 29 29 29

*. Correlation is significant at the 0.05 level (1-tailed).

Table 5, above, shows the outcome of the correlation between Interest and Male investors. I hypothesized that projects with high amount of interest on return (higher risk to invest in) is dominated by Male investors. As you can see in table 6, there is a correlation between

(24)

24 Interest and Male investors (r=.312, p<0.05), significant. A moderately significant

correlation of r=.312 indicates that projects on CrowdAboutNow have more Male investors as Interest rises. The opposite, when Interest is decreasing less Male investors are investing, is also true. For this reason, as expected, hypothesis 3 is according to this study confirmed. Therefore, the composition of investors on CrowdAboutNow differs with varying Target amounts, according to the fourth hypothesis. This is in line with the findings of Also Schubert, Brown, Gysler and Brachinger (1999) who state that women are more risk averse than men. According to Croson and Gneezy (2009) has this phenomenon to do with

experience and profession, which is not taken into account in the present study.

Hypothesis 1, which stated that composition of investors change with changes in target amounts, is partly supported. For more accuracy I divided the first hypothesis into seven sub-hypothesis, which I tested through observing statistical correlations between the target amount and the relation investors had with the initiator (i.e. 1a = Friends, 1b = Family, 1c = Colleague, 1d = Neighbour, 1e = Business people, 1f = Fan, 1g = Anonymous). I found a significant negative correlation between Friend investors and Target amount and a

significant positive correlation between Businesspeople and Target amount. For hypothesis H1(b), H1(c), H1(d), H1(f) and H1(g) I did not found enough evidence for a supporting a significant correlational relation.

(25)

25

Chapter 5 Conclusions and directions for further research

5.1 Conclusions

This study examined 29 successful crowdfunding projects on the platform CrowdAboutNow. Specifically, I tested whether the composition of investors had a relation with changes in target amount in crowdfunding projects. In this study, I divided the composition of investors in the seven relationships (i.e. Friend, Family, Neighbour, Colleague, Business people, Fan and Anonymous) an investor can have with the initiator; difference in interest payments that investors prefer; and gender differences of investors. As expected, almost all hypothesis were confirmed.

From all sub-hypothesis of H1, only the relationships Friend investors and Businesspeople investors were significant. Therefore, I only found enough support for a correlation between those variables and Target amount. For all other relations in H1 (i.e. Anonymous, Fan, Colleague, Neighbour and Family) are not found enough evidence for correlation with changing Target amounts. Moreover, for the second hypothesis (higher interest payments leads to higher invested amounts of money from investors), I found a strong positive relation in support of the hypothesis. This means that composition of investors do change as interest payments changes. Also for the last hypothesis H3 (projects with high amount of interest on return is dominated by men), I found a moderate positive correlation in support of the hypothesis. This indicates that projects with higher interest rates, are more likely to be invested by male investors. So, the composition (gender) of investor’s changes with changes in target amount.

Altogether, this study made clear that there is difference in composition of investors with varying target amounts on CrowdAboutNow, but only to a certain extent. Interest and gender differences in projects with varying target amounts do significantly impact the composition of investors. Differences in composition of investors when looking at the relation the investor has to the initiator of the project, does only have impact on Friends and Business investors and therefore not all relations have strong impact on the composition of investors. When looking at the outcomes of this study, it is plausible that the next most successful crowdfunding project ever, defeating Star Citizen (assuming high interest and high target amount) is dominated by male Business people investors who invest relatively a high amount of capital.

(26)

26 5.2 Limitations and recommendations

This research study has several limitations. Firstly, all data and information is acquired from the crowdfunding platform CrowdAboutNow, which is a Dutch, rather small, loan-based crowdfunding platform. This study would have been more reliable when other, non-Dutch, platforms also would have been investigated. Therefore, the outcomes from this study are not generalizable to all crowdfunding platforms and other countries, because these projects on CrowdAboutNow are all Dutch. In other words, the results of this study cannot be interpreted in a cross-cultural way.

Moreover, correlational research like this study only uncovers relationships; it cannot provide a conclusive reason for why there is a relationship. This study does not reveal which variable influences the other. How is it possible that Friends investors are correlated with Target amount but Family investors are not correlated with Target amount? Further research has to make clear why variables act in a way it presented in this study. This can be done through surveys or other qualitative research.

It could be that the data that is generated from CrowdAboutNow is not reliable. The investor relation is filled in by the investor self. In this way, anonymous investors can be stated as friends, or vice versa. This can lead to biased correlations.

Additionally, I did research on 29 successful crowdfunding projects, which is a relatively small sample. One of the recommendations for this study is to do further research with a bigger sample size. Another recommendation is to take differences of industries of projects into account. Also more variables, for instance the location where the project is going to take place, have to be included for a more reliable result.

(27)

27

Bibliography

Ahlers, G. K., Cumming, D., Günther, C., & Schweizer, D. (2015). Signaling in equity crowdfunding. Entrepreneurship Theory and Practice.

Agrawal, A. K., Catalini, C., & Goldfarb, A. (2011). The geography of crowdfunding (No. w16820). National bureau of economic research.

Bajtelsmit, V. L., & Bernasek, A. (1996). Why do women invest differently than men?.

Financial Counseling and Planning.

Belleflamme, P., Lambert, T., & Schwienbacher, A. (2014). Crowdfunding: Tapping the right crowd. Journal of Business Venturing, Vol. 29, 585–609.

Belleflamme, P., Lambert, T., & Schwienbacher, A. (2010). Crowdfunding: An industrial organization. Working paper.

Bradford, S.C. (2012). Crowdfunding and the federal securities laws. Columbia Business Law

Review, Vol. 1, 1–150.

Croson, R., & Gneezy, U. (2009). Gender differences in preferences. Journal of Economic

literature, 448-474.

European Crowdfunding Network (2012-2013). Retrieved from: www.europeancrowdfunding.com

Freedman, S., and Jin, G. Z. 2014. “The Signaling Value of Online Social Networks: Lessons from Peer-to-Peer Lending,” Working Paper No. 19820, National Bureau of Economic Research.

Gajda, O. & Mason, N. (2013). Crowdfunding for impact in Europe and the USA. European

(28)

28 2015.

Giudici, G., Nava, R., Rossi Lamastra, C., and Verecondo, C. 2012. “Crowdfunding: The New Frontier for Financing Entrepreneurship?,” Available at SSRN 2157429.

Hemer, J. (2011). A snapshot on crowdfunding (No. R2/2011). Working papers firms and region.

Kleemann, F., Voß, G.G., & Rieder, K. (2008). Un(der)paid innovators: the commercial utilization of consumer work through crowdsourcing. Science, Technology & Innovation

studies, Vol. 4, No. 1, 5–26.

Schubert, R., Brown, M., Gysler, M., & Brachinger, H. W. (1999). Financial decision-making: are women really more risk-averse?. American Economic Review, 381-385.

Schwienbacher, Armin and Benjamin Larralde. 2010. „Crowdfunding of Small

Entrepreneurial Ventures”. HANDBOOK OF ENTREPRENEURIAL FINANCE. Oxford University Press. Forthcoming. Accessed October 14, 2013. DOI:

http://dx.doi.org/10.2139/ssrn.1699183.

Tobin, J. (1956). The interest-elasticity of transactions demand for cash. The Review of

Economics and Statistics, 241-247.

Valanciene, Loreta, and Sima Jegeleviciute 2013. „Valuation of Crowdfunding: Benefits and Drawbacks”. Economics and Management, Vol 18, No 1 (2013): 39-48. Accessed October 15, 2013. doi: http://dx.doi.org/10.5755/j01.em.18.1.3713.

(29)

29

Appendix A (pie charts from individual projects)

Project 1,

30MLESPRESSO

Zon Lambertusschool

Target amount €6.000,-

Investors 16

Interest rate (year) 5%

Project 3,

Project 4,

Junk Food

De Olijfboom

Investors 50

Investors 30

Interest rate (year) 0.9%

Investors 19

(30)

30

Project 5,

Project 6,

Vonnies Rijsstafel

Piogg

Investors 19

Project 6 - Composition

Project 7,

Project 8,

TOET

Eat with bravoure

Investors 49

Interest rate (year) 4,3%

Investors 16

Investors 40

(31)

31

Project 9,

Project 10,

VROG

BsaB

Investors 34

Project 11,

Project 12,

Theekamer PIEN

Rhythm & Music

Investors 65

Investors 42

Investors 60

(32)

32

Project 13,

Project 14,

CleverCoin

In de Aap

Investors 36

Project 15,

Project 16,

Queens Royal Bakery

Iadresses

Investors 67

Investors 31

Investors 60

(33)

33

Project 17,

Project 18,

Het Massagehuys

Bake & Moore

Investors 64

Project 19,

Project 20,

Savy

Lakeside Zwolle

Investors 25

Investors 105

Investors 80

(34)

34

Project 21,

Project 22,

Lola

Yoghurt Barn

Investors 71

Project 23,

Project 24,

Brasserie Midi

Buro Lima

Investors 88

Investors 48

Investors 39

(35)

35

Project 25,

Project 26,

SodaFabriek

Nijmegse Boekhandel

Investors 96

Project 27,

Project 29,

Jacketz Oud-West

Tramrestaurant Hoftramm

Investors 103

Investors 512

Investors 182

(36)

36

Project 29,

Administratiekantoor Crowdfunding Anno 12

Investors 236