1
Bachelor thesis
Success factors in lending-based crowdfunding in
The Netherlands
Date: 31-01-2018 Name: Jesper Stok
Student number: 10666133 Supervisor: Drs. P.V. Trietsch
Study program: Economics and Finance Track: Finance
2
Statement of Originality
This document is written by Jesper Stok 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
Abstract
This paper analyses the factors that contribute to a successful lending-based crowdfunding campaign in The Netherlands. Data from 1398 projects from Collin Crowdfund was analysed and a negative relation between the Euro amount of the funding goal and the success of a lending-based crowdfunding campaign was found. The funding goal is the target for the particular campaign. Also a positive relation between the Dun & Bradstreet credit score and success was found. Since firms care about the speed at which the funds are collected, a second regression shows the factors that influence the duration of the campaign. The same two variables were found to have a significant effect on the duration of the campaign.
4
Table of content
1. Introduction ... 5
1.1 Introduction and problem definition ... 5
1.2 Central question ... 5 1.3 Existing literature ... 5 1.4 Sub questions ... 6 1.5 Data ... 6 1.6 Method ... 6 1.7 Structure ... 6 2. Literature review ... 7
2.1 Why firms choose crowdfunding to finance their business ... 7
2.2 What forms of crowdfunding exist... 8
2.3 What is success of a crowdfunding campaign ... 8
2.4 The factors that lead to a successful crowdfunding campaign ... 9
2.4.1 Funding goal... 9
2.4.2 Time to maturity ... 10
2.4.3 Social media ... 10
2.4.4 Campaign video ... 10
2.4.5 Number of reward levels ... 11
2.4.6 Platform quality ... 11 3 Dataset description ... 12 4. Results ... 13 4.1 Correlations ... 13 4.2 Omitted variables ... 13 4.3 Regression analysis ... 13 5. Conclusion ... 15
6. Limitations and recommendations ... 16
Bibliography: ... 17
5
1. Introduction
1.1 Introduction and problem definition
Crowdfunding is a way to externally finance a firm using multiple investors. It is available in different forms and is offered on online platforms (Ahlers, Cumming, Günther, & Schweizer, 2015). When considering external finance obtained from private individuals, it is found to be difficult to find these individuals willing to finance a business (Mason & Harrison, 2015). An intermediary is needed to connect individuals with businesses in need of funds. This gap is filled by several platforms on the internet where firms can finance themselves by the public instead of other forms of traditional financing.
Crowdfunding has increased in popularity and contributes to €170 million in 2016 compared to €128 million in 2015 in The Netherlands (Douw & Koren, 2017). In the USA crowdfunding is much larger, with $10.1 billion in 2015 and $10.9 billion in 2016 (Massolution, 2016). Lending-based crowdfunding is the most used form in The Netherlands and contributes to 77% of total funding generated from crowdfunding (Douw & Koren, 2017). Lending-based crowdfunding is a bond like loan from the crowd to a firm and investors receive interest as reward (Belleflame, Lambert, & Schwienbacher, 2014). A question that arises is what the factors are that drive the success of crowdfunding.
1.2 Central question
What are the factors that contribute to a successful lending-based crowdfunding campaign in The Netherlands?
1.3 Existing literature
Crowdfunding appears to be the most suitable option for medium sized businesses who need a small amount of money for an unpredictable cash flow project (Lewis, 2015). A decision tree can help find the appropriate form of finance for a business. The amount of money needed and the turnover per year influence the decision using this tree (Lewis, 2015).
Different factors determine the chances of success in crowdfunding. A negative correlation between the funding goal in dollars and funding success was found (Calic & Mosakowski, 2016; Mollick, 2014). Moreover was found that campaign videos of different project categories should apply category specific strategies for making videos to improve success (Dey, Duff, Karahalios, & Fu, 2017). Mollick (2014) supported this and found that not including a video decreases success by 26%. Also social media was found to be important to the success of crowdfunding campaigns. Both Calic and Mosakowski (2016) and Mollick (2014) found a positive relation between funding success and the number of Facebook friends the founder of the project has. Lastly the platforms themselves could consider several things to improve chances of success. The speed at which the funds are collected are found to be important to firms performing crowdfunding as well as the simplicity and transparency (Maier, 2016). Being open about the success rate of the platform or limiting the administrative aspects increases chances of success. Being flexible and innovative towards investors might lead to the same result (Maier, 2016).
6
1.4 Sub questions
To form an answer to the central question, the following sub questions should be answered: - Why do firms choose crowdfunding as the most appropriate form of financing? - What forms of crowdfunding exist?
- How can the success of a crowdfunding campaign be measured? - What are factors that lead to a successful crowdfunding campaign?
1.5 Data
Data from 1398 loan requests of Collin Crowdfund will be analysed of which 430 are successfully funded. They are a Dutch based platform providing lending-based crowdfunding with a total investment of just over €89 million. Their success rate is 99.4% providing an average of 7.55% interest to investors (Collin Crowdfund, 2017). The data will be analysed on success, duration, funding goal, time to maturity of the investment, interest rate, credit score and Dun & Bradstreet credit score. The data consists of projects that finished between May 2014 till November 2017.
Other factors may influence success significantly. Social aspects like the number of Facebook friends the owner of the business has, are found to have a significant effect on success (Calic & Mosakowski, 2016). This social aspect will not be included in this research because the database used is not provided with private information of the firms.
1.6 Method
The dataset will be used to determine the variables that influence success of a lending-based crowdfunding campaign. To come to the desired results, a logistic regression analysis will be performed using success as dependent variable and funding goal and Dun & Bradstreet credit score as independent variables. Thereafter a second regression shows the factors that influence the duration of the campaign using duration as dependent variable and funding goal, time to maturity of the investment, interest rate, credit score and Dun & Bradstreet credit score as independent variables.
1.7 Structure
In section 2.1 is explained why firms choose crowdfunding to finance their business. In 2.2 all forms of crowdfunding are mentioned. Chapter 2.3 explains when a crowdfunding campaign is considered successful and 2.4 addresses all factors that lead to a successful crowdfunding campaign. Chapter 3 is about data and explains all variables used in the regression. Using the regression results of chapter 4 combined with existing literature forms a conclusion in section 5. Lastly in section 6 are some limitations and recommendations discussed.
7
2. Literature review
2.1 Why firms choose crowdfunding to finance their business
Firms with high capital expenditures and strong growth objectives are likely to seek external finance (Cosh, Cumming, & Hughes, 2009). The firm’s assets are the most important factor regarding obtaining the desired amount of bank finance. Firms without significant assets have difficulty obtaining bank finance and are more likely to obtain finance through venture capital funds like crowdfunding (Cosh et al., 2009). Also if firms have a lack of outside capital, a bank loan is usually not available (Cosh et al., 2009). Other possibilities are venture capitalists, private individuals, leasing, factoring, suppliers, customers or partner shareholders (Cosh et al., 2009).
As seen in figure 1, crowdfunding appears to be the most suitable option for medium sized businesses who need a small amount of money for a high risk investment (Lewis, 2015). In the finance decision tree, the turnover per year, the amount of risk and the amount of money needed are evaluated. The turnover is the revenue a firm earns during one year. The risk is measured in terms of whether the cash flows are perfectly predictable and the amount of money is either small or large, ranging from a few thousands to millions (Lewis, 2015).
8
2.2 What forms of crowdfunding exist
Four forms of crowdfunding exist: donation, reward-based, equity and lending-based crowdfunding. Donation crowdfunding means that an investor is not expecting a return for his investment (Mollick, 2014). Money is donated in mostly small amounts for charity purpose. This is the cheapest way of financing because hardly no costs are incurred for the firm (Belleflame et al., 2014).
Reward-based crowdfunding is a form of pre-ordering the product or service. This form is used mostly by starting entrepreneurs providing a product or service which is distributed to the investor at a later stadium (Calic & Mosakowski, 2016). According to Mollick (2014), investors in reward-based crowdfunding are threaded as early consumers, allowing them early access or receiving discount to the product. In 2014, $3.26 billion were transferred through this form of crowdfunding worldwide (Belleflame, Omrani, & Peitz, 2015).
Equity crowdfunding involves bond-like shares of the company that is invested in and provides dividends and gains in share price as rewards. (Ahlers et al., 2015). Rights of ownership varies among platforms. Crowdcube offers direct ownership to investors (Crowdcube, n.d.), Seedrs holds the shares as nominees (Seedrs, n.d.) and WeFunder is a proxy holder who holds the voting rights (Wefunder, n.d.). This form of crowdfunding is still small due to high levels of regulation (Mollick, 2015).
Lastly lending-based crowdfunding. This type of crowdfunding is included with interest paid to the investor in return for their investment. The loan is paid back at maturity as a whole, or in peaces during the holding period of the loan (Belleflame, et al, 2014). Dutch platforms like DuurzaamInvesteren and Collin Crowdfund are lending-based. They set entry requirements for loan requests and have a minimum investment of €500 (DuurzaanInvesteren, n.d.; Collin statistieken, n.d.). In 2016, €130 million were collected through this form of crowdfunding in The Netherlands, corresponding to 77% of the total amount collected (Douw & Koren, 2017). Also worldwide this is the most used form of crowdfunding, contributing to $11.08 billion or 68.4% of the total crowdfunding volume in 2014 (Belleflame et al., 2015).
2.3 What is success of a crowdfunding campaign
A crowdfunding campaign is successful when the funding goal is reached within the predetermined time. The predetermined time is a time limit in which the firm wants to receive the desired amount of funds and usually lies between 30-60 days depending on the platform (Mollick, 2014). If the funding goal is not reached within the time available, it depends on the terms and conditions of the platform whether the funds are transferred back to investors or transferred to the firm.
Three groups are involved in crowdfunding: the firm, the platform and the crowd. Success varies among these groups. The main goal for the firm is receiving the money they initially wanted within the time available (Mollick, 2014). Besides the main goal, several side effects occur. Crowdfunding can be a good marketability test for the product. If people are keen to invest, it might be a sign that the idea will work well in the market (Valencine & Jegeleviciute, 2013). This is especially important when products need complementary products, like for example a video game console needs video games developed by third parties (Mollick, 2014). Also by sharing the idea the amount of feedback about the product
9 will most certainly increase (Valencine & Jegeleviciute, 2013). Lastly potential press attention can be beneficial to firms (Mollick, 2014). Thus, besides funding, crowdfunding offers multiple factors that can potentially be beneficial to firms.
Crowdfunding platforms are independent profit companies that need to make a certain profit in order to exist (Belleflame et al., 2015). The gain that crowdfunding platforms make depends of the crowdfunding form that they offer and whether they pay out when a campaign is not fully fulfilled within the time limit.
Donation based crowdfunding platforms are profit organisations that take a cut of the campaign (Belleflame et al., 2015). For example GoFundMe, which collected more than $5 billion by December 2017. Their fee in The Netherlands is 8.55% plus €0.30 per donation. Their US fee is 2.9% plus $0.30 per investment. If the campaign is not fully fulfilled, all collected donations will still proceed to the campaign organizer and the platform will take its share (Gofundme, n.d.).
Equity-based crowdfunding platforms take a cut only when the funding is completely full (Belleflame et al., 2015). For example Crowdcube, a UK based platform charges 7.5% to 9.9%. They report a funding success rate of 60% for 2015. For the remaining 40%, the platform will not make revenue (Crowdcube, n.d.).
Reward-based crowdfunding platforms also take a cut only when the funding is completely full (Belleflame et al., 2015). For example Kickstarter, the biggest reward-based platform charges 8% plus $0.20 per investment. They report a success rate of 35.9% as of 31-12-2017 out of a total of 382.020 projects (Creator Questions, n.d.).
However, lending-based crowdfunding platforms report a success rate of close to 100% (DuurzaanInvesteren, n.d.; Collin statistieken, n.d.). So they take a share out of almost all projects that start their campaign. Fee structures differ among platforms.
The success for the investor depends on the needs and wants of the investor. If invested in donation-based crowdfunding, an investor is not expecting a return for its donation (Mollick, 2014). So success will be reached when its donations are being accepted by the fund raiser. If invested in any other form of crowdfunding, the crowd can be seen as investors who make investment decisions based on their expectations of success (Mollick, 2014). So crowdfunding success for the investor is reached when the campaign is fulfilled and the investor receives its product, share of the firm or interest on the loan.
2.4 The factors that lead to a successful crowdfunding campaign
2.4.1 Funding goal
The funding goal is the target for the particular campaign and is found to have a negative relation to success (Calic & Mosakowski, 2016; Mollick, 2014). Generally a higher funding goal results in a smaller chance of success. This relation was found by analysing 708 reward-based crowdfunding campaigns from April 2009 to July 2013 divided in 392 technology and 315 film and video projects (Calic & Mosakowski, 2016). The dataset was limited to projects with a funding goal greater or equal to $5000. At or above this funding level crowdfunding is supposed to compete with traditional forms of financing. The average funding goal was $19,732 for technology projects and $13.359 for film and video projects. 30% of technology projects and 39% of film and video projects were successful (Calic & Mosakowski, 2016).
Since both research is on data from reward-based crowdfunding, this paper will research if the funding goal has a significant effect on success also for lending-based
10 crowdfunding. Accordingly, the hypothesis formed is that the funding goal has a significant negative relation with success.
2.4.2 Time to maturity
The time to maturity is the time from the start of the loan towards the end of the loan (Sekkel, 2011). It is only relevant for lending-based crowdfunding since that is the only form of crowdfunding that offers a loan. Because most research has been on data from reward-based crowdfunding, literature on time to maturity is very scarce.
The amount of risk rises with the time to maturity because the longer the time to maturity, the longer time is available for interest rates to change on other investments (Sekkel, 2011). Usually a longer time to maturity is compensated with higher interest, therefore the hypothesis concerning time to maturity is that it does not have a significant effect on success due to the compensation of return versus risk.
2.4.3 Social media
The effect of social media on success can be measured by the amount of social media interaction the founder of the crowdfunding campaign has. A positive relation between funding success and the number of Facebook friends of the founder of the project was found, although only 46% of the observations were linked to Facebook (Calic & Mosakowski, 2016).
Also Mollick (2014) studied the effect of social media on success and found a positive relation between funding success and the number of Facebook friends the founder of the project has. He analysed 48,526 Kickstarter projects of which 48.1% were successfully funded. Projects on Kickstarter have a connection to Facebook so that makes it possible to determine the amount of Facebook friends the founder of the project has.
Also the number of tweets had been used to predict the success of campaigns. In their research, Etter, Grossglauser, and Thiran (2013) created an engine to predict the success of Kickstarter campaigns by using both direct information as the funding goal, duration and number of investors as well as social features. By adding the number of tweets about the campaigns, their prediction accuracy increased by 4% to 76% just 4 hours after the start of a campaign (Etter et al., 2013).
2.4.4 Campaign video
A video is a powerful communication channel to connect emotionally to the crowd (Clark & Sterwart, 2007). Campaign videos of different project categories should apply category specific strategies for making videos (Dey et al., 2017). This relation was found by recruiting 3,150 workers from Amazon to evaluate the campaign videos in three categories: technology, fashion and design. This allowed them to observe the persuasion effect of campaign videos. Technology campaign videos should focus on product related aspects and explain or demonstrate the product. Design and fashion campaigns should focus on audio-video quality to persuade the crowd (Dey et al., 2017).
Another study focussed on the role of preparedness as a signal of quality to investors and found that not including a video decreases chances to success by 26% (Mollick, 2014). Preparedness was determined by the degree to which firms took the time and effort to ensure that projects meet a certain level of quality. He mentioned in his research that not including a campaign video will seriously disservice the project quality. In
11 terms of project quality, he used the video as an indicator for a high quality project and 86% of the sample has a video (Mollick, 2014).
2.4.5 Number of reward levels
The number of reward levels is defined as the number of reward possibilities a project offers (Calic & Mosakowski, 2016). For example, one will offer a $10 investment where the investor gets one product or an offer to get 2 products for $18. More of these investment possibilities may fulfil the needs and wants of potential investors and therefore leads to more success. This relation was analysed by Calic and Mosakowski (2016). Their analysis on 708 reward-based crowdfunding campaigns found an average of 6.3 reward levels for technology an 9.2 reward levels for film and video projects. Similar to previous studies (Mollick, 2014), they found a positive relation between the number of reward levels and funding success (Calic & Mosakowski, 2016).
2.4.6 Platform quality
Since crowdfunding platforms act as intermediary between firms and investors, their fee structure make them dependent on actual investment and success of the campaigns (Maier, 2016). The speed at which the funds are collected are found to be important to firms performing crowdfunding as well as the simplicity and transparency (Maier, 2016). Being open about the success rate of the platform or limiting the administrative aspects included with crowdfunding are examples that they should consider to attract more firms to their platform (Maier, 2016). He also found that investors do not care about their relation with a financial institution, they care about flexibility and innovativeness. So to persuade a
potential investor, the focus must be on being flexible and innovative, by fulfilling the needs and wants of potential investors. For example by increasing the amount of investment possibilities or by including a mobile application of the platform (Maier, 2016).
12
3 Dataset description
Data from 1398 loan requests from Collin Crowdfund will be analysed of which 412 were successfully funded. In terms of money invested, Collin Crowdfund is the biggest lending-based crowdfunding platform in the Netherlands with a total investment of just over €89 million providing an average of 7.55% interest to investors (Collin statistieken, n.d.). The database consists of roughly 21% of all crowdfunding campaigns organised in The Netherlands in 2016 and 16% in 2015 (Douw & Koren, 2017; Collin statistieken, n.d.).
Table 1: Descriptive statistics, source: Collin Crowdfund
Variable Frequency Mean Median Std. Dev. Min Max Success 0 (no success) or 1 (success) 1398 0.31 0.00 0.46 0.00 1.00 Duration Days 431 3.38 0.30 7.60 0.30 30.00 Funding goal Euro's 1398 256,961.64 165,000.00 314,834.83 10,000.00 2,500,000.00 Time to maturity Months 569 52.82 60.00 14.87 1.00 180.00 Interest Percentage per
year 579 7.58 7.50 0.67 1.00 10.00 Credit score 1-100 517 54.93 50.00 16.83 3.00 100.00 Dun & Bradstreet score 1-100 915 34.46 26.00 24.99 1.00 100.00
The variables used are success, duration, funding goal, time to maturity, interest rate, credit score and Dun & Bradstreet score. Success indicates whether the loan request is successfully funded or not. This variable is either 0, indicating that the loan request is rejected by the platform or not fulfilled within the available time, or 1, indicating a loan request that is fulfilled within the time available. The time available is 30 days.
The duration measures the number of days until the campaign is fulfilled. As seen in table 1, only 431 campaigns feature a duration because the other part was rejected before the start of the campaign.
The funding goal is the Euro goal for the particular loan request. If the funding goal is not reached within the time available, the funds are transferred back to the investors and the firm gets no funding.
Time to maturity is the time from the start of the loan towards the maturity of the loan, measured in months. Interest is what investors receive in return for their investment and usually varies with the amount of risk involved in the investment (Maier, 2016).
Two credit scores are assigned to each loan request, one internal and one from outside. The internal credit score is created by the platform itself and the external one is comes from Dun & Bradstreet, an independent credit rating agency. They are licensed as leading indicator to predict company risk and to evaluate possible future liquidity problems (Kredietbeoordeling, n.d.). Both scores indicate the potential risk included with the investment and range between 1 and 100. A low score might indicate a liquidity problem and therefore should lead to a higher interest to compensate for the increased risk.
13
4. Results
4.1 Correlations
To ensure validity and reliability of the results, the variables should be tested on high correlations that might lead to multicollinearity. As seen in table 3, the highest absolute correlation is -0.55 between interest and credit score. This correlation is below 0.7, but is still quite high and might be disrupting the reliability of the results.
Table 2: correlations
Success Duration Funding goal Time to maturity Interest Credit score Dun & Bradstreet score Success 1 Duration -0.01 1 Funding goal -0.10 0.12 1 Time to maturity -0.04 -0.07 0.06 1 Interest -0.09 -0.03 -0.21 -0.04 1 Credit score 0.04 -0.02 0.20 0.01 -0.55 1 Dun & Bradstreet
score 0.10 -0.08 0.02 0.07 -0.20 0.20 1
4.2 Omitted variables
Omitted variable bias can occur when certain variables are not included in the dataset. As seen in the literature review, many aspects can influence success of a crowdfunding campaign. Social media features are excluded from the data since no connection with social media is made on the platform of Collin Crowdfund. Also campaign videos and reward levels are not included because those are only relevant for reward-based crowdfunding. Lastly platform quality was mentioned to have an impact on success. This aspect is difficult to measure and is therefore neither included.
4.3 Regression analysis
In this part, two regression analysis will be performed. First all 1398 loan requests from the database will be used to predict the factors that lead to success. Thereafter 430 successful campaigns are used to determine the factors that lead to fast fulfilment of the campaign as being the next level of success.
In the first analysis, a logistic regression will be done using success as dependent variable. By using this form of regression it becomes possible to use a binary dependent variable. However, if one or more variables of a certain campaign are missing, the whole campaign will be excluded from the analysis. Therefore if all variables are used in predicting the success of the campaign, only 65 of 968 unsuccessful campaigns can be used. Since those 65 unsuccessful campaigns contain more data than the remaining unsuccessful campaigns the results are likely biased (see Appendix for regression results). To be able to come to unbiased results, the least frequent variables are left out from the regression. The following regression uses 915 loan requests of which 430 were successfully funded. Success is used as binary dependent variable and funding goal and Dun & Bradstreet score as independent variables.
14
Table 3: logistic regression output using success as dependent variable
B S.E. Wald df Sig. Exp(B)
Funding goal -7.15 e-7 0.00 5.44 1.00 0.02 0.99 Dun & Bradstreet score 0.0083 0.00 9.50 1.00 0.00 1.01 Constant -0.25 0.13 3.53 1.00 0.06 0.78
Both the funding goal and Dun & Bradstreet score appear to have a significant effect on predicting the success of a campaign. The funding goal is significant with 0.02 and on average, a €100.000 increase in the funding goal decreases the chances of success with 7.15%.
Also the Dun & Bradstreet score appears to be significant with 0.00 in predicting the success of a campaign. On a scale of 1 to 100, a 1 unit increase of this score increases the chance of success with 8.3%.
In the next analysis the variable duration is used as dependent variable and funding goal, time to maturity, interest, credit score and Dun & Bradstreet score are the independent variables. Now the data consists of 430 successful projects to determine the factors that lead to fast fulfilment of the campaign.
Table 4: regression output using Duration as dependent variable
Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta (Constant) 13.91 7.32 1.90 0.06 Funding goal 6.04 e-06 0.00 0.14 2.88 0.00 Time to maturity -0.04 0.02 -0.08 -1.69 0.09 Interest -0.91 0.81 -0.07 -1.12 0.26 Credit score -0.03 0.03 -0.06 -1.06 0.29 Dun & Bradstreet score -0.03 0.02 -0.10 -2.01 0.05
As seen in table 4, the funding goal appears to be the most significant variable in predicting the amount of days a campaign needs to reach the funding goal. So increasing the funding goal leads to a longer campaign duration. In particular, a €100.000 increase in funding goal leads on average to a 0.6 days longer campaign duration.
Also the Dun & Bradstreet score appears to be significant with 0.05 in predicting the duration. On a scale of 1 to 100, a 1 unit increase of this score decreases duration by 0.03 days.
The regression does not show any evidence that the interest has a significant effect on the duration of the campaign. Although the regression output shows a negative relation with duration, this relation is significant with 0.26. The same is found for credit score, there is no significant relation between credit score and the duration of the campaign.
15
5. Conclusion
In this research a connection between the success of crowdfunding and the lending-based form of crowdfunding was made. To repeat, the central question was the following:
What are the factors that contribute to a successful lending-based crowdfunding campaign in The Netherlands?
In the literature review was found that a crowdfunding campaign is successful when the funding goal is reached within the predetermined time. The success of the firm, platform and investor depends on the form of crowdfunding used and the terms and conditions of the platform.
A negative relation between the funding goal in dollars and funding success was found (Calic & Mosakowski, 2016; Mollick, 2014). Also social media was found to be important to the success of crowdfunding campaigns. Both Calic and Mosakowski (2016) and Mollick (2014) found a positive relation between funding success and the number of Facebook friends the founder of the project has. Dey et al. (2017) found that campaign videos of different project categories should apply category specific strategies for making videos to improve success. Mollick (2014) supported this and found that not including a video decreases success by 26%. Lastly the platforms themselves could consider several things to improve chances of success. The speed at which the funds are collected are found to be important to firms performing crowdfunding as well as the simplicity and transparency (Maier, 2016). Being open about the success rate of the platform or limiting the administrative aspects increases chances of success. Being flexible and innovative towards investors might lead to the same result.
To form an answer to the central question, data from 1398 loan requests from May 2014 till November 2017 was analysed. A logistic regression analysis was performed and found a negative relation between the Euro amount of the funding goal and funding success for a lending-based crowdfunding campaign. On average, a €100.000 increase in funding goal decreases chances of success with 7.15%. This negative relation was also found in other research that focussed on the success of reward-based crowdfunding campaigns (Calic & Mosakowski, 2016; Mollick, 2014). A positive relation between the Dun & Bradstreet score and success was found. A 1 unit increase of this score increases the chance of success with 8.3%. No research this far has shown the existence of this relation. The reason that this relations was not yet shown is that most databases in other research contain data from reward-based platforms where no credit scores exist.
Time to maturity, interest and credit score were not found to be statistically significant and do not certainly contribute to success of a lending-based crowdfunding campaign in The Netherlands.
In the literature review was found that firms care about the speed at which the funds are collected (Maier, 2016). Therefore the second regression was used to determine the factors that lead to fast fulfilment of a lending-based crowdfunding campaign. We found that the funding goal is the most significant variable in predicting the amount of days a campaign need to get fulfilled. A €100.000 increase in funding goal leads on average to a 0.6 days longer campaign. Also the Dun & Bradstreet credit score was found to be effective in predicting the duration. On a scale of 1 to 100, a 1 unit increase of this score decreases duration by 0.03 days.
16
6. Limitations and recommendations
Although many factors were inserted in this research, some limitations are present. Other factors like social media or the segment in which the firm operates might influence the success of campaigns. Those were not included in the dataset. Further research is needed to include those variables and to extend towards a more dynamic model, including all aspects that contribute to the success of a lending-based crowdfunding campaign. Also this paper only addresses lending-based crowdfunding and one platform. Further research should include other forms of crowdfunding and multiple platforms as well.
Several recommendations to all parties involved with crowdfunding follow from this research. Firstly, firms should consider what form of financing is the most appropriate for them. Instruments like the finance decision tree can help them to find the right form of financing. Secondly, platforms should try to insert the needs and wants of both potential investors and firms. They should simplify the administrative aspects and they should be transparent to satisfy firms willing to perform crowdfunding. They also should be flexible and innovative towards investors. Including a mobile application or increasing the amount of investment possibilities are examples they should consider. Lastly, the investor should consider the form of crowdfunding that he or she is investing in. Returns vary among different forms of crowdfunding and so do chances of default.
17
Bibliography:
Ahlers, G., Cumming, D., Günther, C., & Schweizer, D. (2015). Signaling in Equity Crowdfunding. Entrepreneurship Theory and Practice, 39(4), 955-980.
Belleflamme, P., Lambert, Thomas, & Schwienbacher, Armin. (2014). Crowdfunding tapping the right crowd. Journal of Business Venturing, 29(5), 585-609.
Brodersson, M., Enerbäck, M., and Rautiainen, M. (2014). The Angel Investor Perspective on Equity Crowdfunding. Internationella Handelshögskolan, Högskolan i
Jönköping.
Bruton, G., Khavul, S., Siegel, D., & Wright, M. (2015). New Financial Alternatives in Seeding Entrepreneurship: Microfinance, Crowdfunding, and Peer‐to‐Peer Innovations. Entrepreneurship Theory and Practice, 39(1), 9-26.
Calic, G., & Mosakowski, E. (2016). Kicking Off Social Entrepreneurship: How A Sustainability Orientation Influences Crowdfunding Success. Journal of
Management Studies, 53(5), 738- 767.
Clark, T., & Stewart, J. (2007). Promoting Academic Programs Using Online Videos.
Business Communication Quarterly, 70(4), 478-482.
Collin statistieken, (n.d). Retrieved from https://www.collincrowdfund.nl/statistieken/ Cosh, A., Cumming, D., & Hughes, A. (2009). Outside Enterpreneurial Capital*. Economic
Journal, 119(540), 1494-1533.
Creator Questions: Getting started. (n.d.). Retrieved from https://www.kickstarter.com/help/faq/creator+questions
Crowdcube (n.d.). Retrieved from https://www.crowdcube.com/
Crowdsourcing LLC Massolution (2015). The crowdfunding industry report 2015cf.
The Crowdfunding Industry Report 2015, 2015.
Dey, S., Duff, B., Karahalios, K., & Fu, W. (2017). The Art and Science of Persuasion: Not All Crowdfunding Campaign Videos are The Same. Proceedings of the 2017 ACM
Conference on Computer Supported Cooperative Work and Social Computing,
755-769.
Douw, S., & Koren, G. (2017). Crowdfunding in Nederland 2016. Retrieved from http://douwenkoren.nl/wp-content/uploads/2017/08/DK-Crowdfunding-in-NL-2016.pdf
DuurzaamInvesteren (n.d.). Retrieved from https://www.duurzaaminvesteren.nl/
Etter, V., Grossglauser, M., & Thiran, P. (2013). Launch hard or go home: Predicting the success of kickstarter campaigns. Proceedings of the First ACM Conference on
Online Social Networks, 177-182.
Gofundme (n.d.). Retrieved from https://www.gofundme.com/
Kredietbeoordeling (n.d.). Retrieved from
18 Lewis, S. (2015). Raising finance for Europe’s small & medium-sized businesses. Retrieved from https://www.afme.eu/globalassets/downloads/briefing-notes/raising-finance-for-europes-small-medium-sized-businesses-english-version.pdf
Maier, E. (2016). Supply and demand on crowdlending platforms: Connecting small and medium-sized enterprise borrowers and consumer investors. Journal of Retailing
and Consumer Services, 33, 143-153.
Mason, C., & Harrison, R. (2015). Business Angel Investment Activity in the Financial Crisis: UK Evidence and Policy Implications. Environment and Planning C:
Government and Policy, 33(1), 43-60.
Mollick, E. (2014). The dynamics of crowdfunding: An exploratory study. Journal of
Business Venturing, 29(1), 1-16.
Seedrs (n.d.). Retrieved from https://www.seedrs.com/
Sekkel, R. (2011). International evidence on bond risk premia. Journal of Banking and
Finance, 35(1), 174-181.
Valencine, L., & Jegeleviciute, S. (2013). Valuation of crowdfunding: benefits and drawbacks. Economics and Management, 18(1), 39-48.
19
Appendix:
Table A1:
Case Processing Summary
Unweighted Casesa N Percent
Selected Cases Included in Analysis 495 35.4
Missing Cases 903 64.6
Total 1398 100.0
Unselected Cases 0 .0
Total 1398 100.0
a. If weight is in effect, see classification table for the total number of cases.
Table A2:
Omnibus Tests of Model Coefficients
Chi-square df Sig.
Step 1 Step 8.154 5 .148
Block 8.154 5 .148
Model 8.154 5 .148
Table A3:
Variables in the Equation
B S.E. Wald df Sig. Exp(B)
95% C.I.for EXP(B) Lower Upper Funding_goal .000 .000 .023 1 .881 1.000 1.000 1.000 Dun_Bradstreet -.004 .006 .485 1 .486 .996 .985 1.007 Time_to_maturity -.002 .010 .043 1 .836 .998 .979 1.017 Interest -.790 .286 7.608 1 .006 .454 .259 .796 CreditScore -.010 .010 1.067 1 .302 .990 .971 1.009 Constant 8.764 2.657 10.877 1 .001 6397.830
