drivers of fundraising success in equity...

1

Drivers of fundraising success in equity crowdfunding

Sandra Correia

School of Economics and Management, University of Porto, Porto, Portugal [email protected]

Miguel Sousa


Elísio Brandão


Version: March 2019

Abstract:

Using an unique hand-collected database of 1,256 campaigns from the two major

platforms of equity crowdfunding in the United Kingdom, between 2015 and 2018, we

find that the factors that most influence the campaign success are: (i) the quality of the

project, signalized by equity retention and the presence of a large investor; (ii) the

information disclosure and (iii) the early investments. There is no evidence economic

potential of the firm are relevant for investor decisions. Investors probably use other

signals (than growth and profitability perspectives) to evaluate the economic potential of

firms and project quality.

Keywords: equity crowdfunding, alternative finance, herding behavior

Electronic copy available at: https://ssrn.com/abstract=3354210

2

1. Introduction

Start-ups have distinct characteristics from older firms. They are riskier and it takes longer

to get the return on such investment. Many start-ups fail in the first years (risk of capital

loss), they are private (with no secondary market - illiquidity risk), the exit strategies are

usually by acquisitions by other companies or by going public (long term investment),

and rarely pay dividends. This way, start-ups have usually more difficulties in accessing

capital, which leads some to argue that there is a financing gap for start-ups (OECD,

2014). They frequently use informal sources of capital like family & friends, as well

venture capital and business angels. However, even these sources are not always available

for start-ups, in particular for smaller businesses. Additionally, given the lack of positive

cash-flows and the high risk of start-ups, its capital structure should be mainly financed

by equity rather than debt. This way, the equity crowdfunding is a recent financing

alternative for start-ups that shows a remarkable growth in recent years (WorldBank,

2013).

Crowdfunding allows entrepreneurs to fund their ventures through small contributions

from a large number of funders, without standard financial intermediaries. Different

models can be used (reward, donations, lending, equity) but our focus is on equity

crowdfunding. This is a more complex structured model (Bruton, Khavul, Siegel, &

Wright, 2015) and the decision of entrepreneurs is essentially a financing one. In other

crowdfunding models (e.g. reward) it could be also an operating decision that affects sales

and production levels, which is more suitable for low funding needs (Belleflamme,

Lambert, & Schwienbacher, 2014).

Much of the literature about crowdfunding is concerned about the static determinants of

success of crowdfunding campaigns. However, it is essentially focused on reward model


3

(e.g. Colombo, Franzoni, & Rossi-Lamastra, 2015; E. Mollick, 2014; E. R. Mollick &

Kuppuswamy, 2014), where monetary return isn’t the main motivation of investors.

The literature about equity crowdfunding is still scarce, due to the novelty of the theme

and the lack of data (Hornuf & Schwienbacher, 2015) and it is concentrated on two main

topics: the differences between this new source of financing and the traditional ones

(venture capital and business angels) and the determinants of funding success.

The first studies on equity crowdfunding focused on the static determinants of funding

success. Ahlers, Cumming, Günther, & Schweizer (2015) find that the main effective

signals about project quality are the equity retention and the existence of detailed

information about risks of the projects, while the social and intellectual capital (patents)

don’t seem to be relevant in this context. Belleflamme, Lambert, & Schwienbacher (2013)

highlight the relevance of the organization form (profit versus non-profit) to fundraising

success and Hornuf & Schwienbacher (2015) suggest that the typical pattern of project

support is dependent on the mechanism used by the equity crowdfunding platform. While

in a first-come, first-serve mechanism the typical pattern of project support is L-shaped

(higher number of crowd investors in the first days), under an auction mechanism the

pattern of project support is U-shaped (higher number of crowd investors in the first and

in the last days of the campaign). The authors also suggest that investor’s decisions are

influenced by the information provided by the entrepreneur during the campaign and by

the investment behavior and comments of other investors. Others authors suggest that

drivers of success in equity crowdfunding are different from the criteria usually used by

venture capitalists and business angels to decide their investments (Lukkarinen, Teich,

Wallenius, & Wallenius, 2016). Vismara (2016) underline the role of equity retention and

social networks to the campaigns’ success, while for (Vulkan, Åstebro, & Sierra, 2016)

to being successful, the campaigns need to have a strong start, many backers and, at least,


4

has one investor who provides a large pledge. Other recent papers concentrate its analysis

in a particular driver of fundraising success. For instance, Piva & Rossi-Lamastra (2018)

focus their analysis on the relevance of the human capital.

In addition to scarcity, the literature about equity crowdfunding has some limitations,

essentially related to the sample size, which is obtained from only one platform, but also

because they cover a period (and country) when (where) the equity crowdfunding market

was less developed. For instance, Ahlers et al. (2015) analyze an Australian platform

between 2006 and 2011, during a period when equity crowdfunding was still unknown

by the general public. Belleflamme et al. (2013) use data from individual crowdfunding

practices (without using crowdfunding platforms as intermediaries). Hornuf &

Schwienbacher (2015) study the German market that has different characteristics from

United Kingdom (UK). For instance, equity crowdfunding platforms in Germany don´t

offer common shares on a PPL company, but mezzanine financial instruments, and the

main mechanism of equity crowdfunding in the UK doesn’t limit the investments after

the funding goal is achieved. Lukkarinen et al., (2016) use a sample of only 60 campaigns

from the platform Invesdor in Finland. Vulkan et al. (2016) use a larger sample (636

campaigns) but limited to one platform (Seedrs).

More recently, the literature on crowdfunding is also concerned to the dynamics of the

campaign, suggesting the existence of a herding behavior of investors (Hornuf &

Schwienbacher, 2015). This herding behavior is found in microloans platforms (Zhang &

Liu, 2012) and in reward crowdfunding (Colombo et al., 2015; Kuppuswamy & Bayus,

2014). However, we can argue that, in equity crowdfunding, the motivations to pledge a

campaign are someway different (primarily related to potential monetary returns, not a

pre-sale of a product) and that investors are probably more sophisticated and more

cautious given the high investment risks, whereby the herding behavior may be different


5

from other crowdfunding models. There are just a few papers (Agrawal, Catalini, &

Goldfarb, 2015; Hornuf & Schwienbacher, 2015; S. Vismara, 2018; Vulkan et al., 2016)

that already analyze this question in an equity crowdfunding environment, and they also

have some limitations. All these papers use only one platform for their analysis. Agrawal

et al. (2015) use a revenue sharing platform limited to musical artist; Hornuf &

Schwienbacher (2015) assess the herding behavior in German equity crowdfunding

platforms (with different characteristics from usual equity platforms in UK), but suggest

that such effect depends on the mechanism of equity crowdfunding used. Vismara (2018)

and Vulkan et al. (2016) are the only that, to our knowledge, that gives some insights

about the separation between rational and irrational herding in equity crowdfunding, but,

as the authors recognize, more research on this topic is needed.

As some of the previous papers on equity crowdfunding, our research studies the (statics

and dynamics) determinants of campaigns success. However, our paper differs from

previous ones and contribute to a better understanding of equity crowdfunding for several

reasons. First, we use an unique hand-collected database of 1,256 campaigns, between

2015 and 2018. As Vismara (2016), our data is from the two biggest equity crowdfunding

platforms in the UK. However, we use a larger and more recent sample (1,256 campaigns

from 2015 to 2018, while Vismara uses a sample of 271 campaigns from 2011 to 2014).

The relevance of sample size is stated by Vismara himself, “future studies will benefit

from larger samples” (p. 588), and in the last years the equity crowdfunding market has

grown strongly and became more mature1. Second, we collected a more detailed

information about the campaigns, the firms and project team, which allowed us to carry

1 According to “The 3rd European Alternative Finance Industry Benchmarking Report” (Ziegler et al,

2018), this market grew by 41% in 2016 on Europe (101%, excluding the United Kingdom), reaching 7,671

millions of euros. The UK market continues to absorb the largest share (73%), followed by France,

Germany and the Netherlands.


6

a very detailed characterization of equity crowdfunding in UK over the last three years.

Finally, our paper evaluates, for the first time to our knowledge, the effect of economic

potential of firms, as well the effect of competitive campaigns in equity crowdfunding

campaigns.

2. Literature Review and Hypotheses Development

Given the still scarce literature about crowdfunding, we construct some of our hypothesis

on literature about venture capital and business angels. Like in venture capital and

business angels, investors in equity crowdfunding are, in general, searching for good

projects (or firms) to invest their money and to get a financial benefit from it in future

(dividends and/or capital gains), but they are probably less informed than the previous

ones (many of them are amateurs instead of professional investors).

According to the literature, there are many factors that may influence the success of an

equity crowdfunding campaign. From the static point of view, those factors are related to

the project quality, team quality, social capital and information disclosure, among other

motivations, such as tax reliefs and nonfinancial rewards. From the dynamic point of

view, some authors also suggest the existence of a herding behavior of investors, arguing

that early days’ dynamics (amount and number of investors) influence the campaign’s

success. This paper will analyze both static and dynamic determinants of equity

crowdfunding campaign.

2.1. Project Quality

In general, investors face problems of information asymmetry that imply difficulties in

selecting high quality projects. This problem is exacerbated in equity crowdfunding, in


7

the presence of small investors to whom due diligence and monitoring costs are relatively

higher. This way, investors need to find out signals of project quality in order to reduce

such costs and be able to select the best projects.

As predicted by the signaling theory (Busenitz, Fiet, & Moesel, 2005; Leland & Pyle,

1977), the entrepreneurs’ willingness to invest in his own project is perceived as a signal

of project quality. In fact, entrepreneurs own private information that allows them to

better evaluate the project quality. In the presence of high-quality projects, the

entrepreneur will try to keep the highest share of capital as possible. Previous literature

on equity crowdfunding already confirm that equity retention is a key factor to the success

of the campaigns (Ahlers et al., 2015; Silvio Vismara, 2016).

Hypothesis 1A – The highest equity retention, the highest the probability of success

of the equity crowdfunding campaign.

Venture capital and business angels’ investors play a relevant role in start-ups, providing

mentoring, support services (such as helping in the development of business plan,

facilitating strategic partnerships, building the internal organization of the firm, accessing

other financial intermediaries) and certification of firm quality (Denis, 2004; Hellmann

& Puri, 2002). Thus, the less informed and small investors may benefit from the quality

certification of institutional and bigger investors (Megginson & Weiss, 1991).

Additionally, some authors also highlight the ability of venture capitalists to identify start-

ups with high growth potential (Baum & Silverman, 2004); while others suggest that the

probability of failure is lower for firms financed by venture capital (Puri & Zarutskie,

2012). Hence, we can argue that small investors may benefit from the advantages of the

presence of large institutional investors in selection, monitoring and management support

(Hornuf & Schwienbacher, 2015).


8

Hypothesis 1B – The presence of large institutional investors has a positive impact

on equity crowdfunding campaign success.

Taking into account the lack of history and the absence of credible reputation

(Huyghebaert & Van de Gucht, 2007), start-ups have higher information asymmetries and

agency costs than older firms. However, younger firms invest more on R&D (Czarnitzki

& Kraft, 2004), and young innovative firms are more able to grow than mature firms,

even that innovations undertaken by young firms are riskers, and less predictable than

those from older firms (Coad, Segarra, & Teruel, 2016). This way, we argue that the

impact of firm age on the success of equity crowdfunding campaign is an empirical

question. It can be positive (the probability of success increases with the age) because

younger firms are riskier. Or it can be negative, given the lower growth perspectives of

mature firms. The result varies according to the risk aversion of investors.

Hypothesis 1C – Firm age influences the success of equity crowdfunding

campaigns.

Many papers emphasize the importance of monetary potential of the project/firm in the

decision of venture capitalists and business angels to invest. For instance, Landström

(1998) argues that informal investors attach considerable importance to the economic

returns of the investments (they are not philanthropists), and during the investment

process, they assess the market and product potential. Hsu, Haynie, Simmons, &

McKelvie (2014) confirm that economic potential of the new venture positively affects

the investment decisions in both venture capital and business angels, in terms of expected

returns (above of market) and in expected time to break even. Puri & Zarutskie (2012)

suggest that actual or potential firm dimension (in terms of sales and employment) as well

growth perspectives are used as criteria by venture capitalists to select firms where to


9

invest. In fact, some argue that monetary potential is one of the most relevant investment

decision factors to venture capitalists. For instance, Tyebjee & Bruno (1984) find that

firm’s profit margin is a relevant factor of expected return for venture capitalists.

Accordantly, Macmillan, Siegel, & Narasimha (1985) find that the key financial concerns

for venture capitalist are with high upside potential and high investment liquidity. This

way, we argue that, as well for venture capitalists and business angels, equity

crowdfunding investors prefer to invest in firms with high economic potential.

Hypothesis 1D – The economic potential of firm positively influences the success of

equity crowdfunding campaigns.

2.2. Team quality

It is broadly recognized the importance of top management team to the performance and

success of firms, in particular on new ones (Eisenhardt, 2013), because management team

is who defines firm’s strategy, hires employees, decides investments, i.e., they have the

power over the critical factors to growth and success of firms.

According to the learning theory, prior experience positively influences the success of the

firm. In fact, general business experience provides relevant skills to entrepreneurs (related

to negotiation, leading, decision-making and problem solving), which can improve the

new business success (S. A. Shane, 2003). So, in the investment decision process,

business angels usually reject projects when the entrepreneur has no previous experience

(Carpentier & Suret, 2015), and they rely on the competence and trustworthy of

entrepreneurs to reduce market risk of the new ventures (Fiet, 1995). Some empirical

studies observed a significant and positive relationship between the probability of getting

funding from business angels investors and the industry experience of entrepreneurs

(Carpentier & Suret, 2015), as well the years of experience with young firms (Hsu et al.,


10

2014). For instance, according to Baum & Silverman (2004), identity and background of

top management are widely regarded as important signals of a start-up’s future potential,

increasing its chances of obtaining financing from venture capital. Also, Nofsinger &

Wang (2011) find that prior start-up experience of entrepreneur signals managing skills,

which could influence the ability to get external financing from institutional investors.

So, according to the previous literature on venture capital and business angels, the team

quality is a critical factor in the investment decision process for those investors. For small

investors in equity crowdfunding, with low ability to influence firm management in the

future, the team’s quality is probably even more relevant to firm success and to the

expected return on investment.

In the crowdfunding context, Piva & Rossi-Lamastra (2018) found that entrepreneurs’

business education and entrepreneurial experience contribute to the success of equity

crowdfunding campaigns. According to Bernstein (2015), the team quality is relevant not

only for pure signaling reasons, but also for the operational capabilities and expertise of

the founders.

Hypothesis 2A – Team quality (measured by team qualifications and previous

experience) positively influences the success of equity crowdfunding campaigns.

Large teams have more capabilities to process information and provide more viewpoints

(conflicting views reduces the probability of costly mistakes), so some authors suggest

that large teams increase firms’ performance (Haleblian & Finkelstein, 1993). For

instance, in a study of U.S. semiconductor ventures, Eisenhardt & Schoonhoven (1990)

find that top management team (specifically larger teams, teams with heterogeneity in

industry experience and teams with a prior experience together) contributes to a higher

growth of start-ups. Additionally, larger boards have more difficulties in achieving


11

consensus, reducing the probability of taking extreme decisions, resulting that firm

performance is less variable in firms with larger boards (Cheng, 2008). There is empirical

evidence that the number of founders can be a proxy of human quality, and there is

empirical evidence that start-ups with larger top management teams obtain more venture

capital financing (Baum & Silverman, 2004). This way, we can argue that the team size

can be a proxy of human capital, and, as in venture capital, it has a positive influence on

success of crowdfunding campaign.

Hypothesis 2B – The higher the team size the higher the probability of success of

equity crowdfunding campaigns.

Entrepreneurs highly committed with the project put more effort (time and energy) to

achieve the venture success. In this context, several studies state the importance of

commitment (perceived passion) of the entrepreneur as a relevant decision criterion for

both venture capitalists and business angels (Chen, Yao, & Kotha, 2009; Hsu et al., 2014;

Mitteness, Sudek, & Cardon, 2012). This way, we argue that if the entrepreneur will

decide to be in full-time in the project, he/she is signaling his/her commitment to the

project. This can also be interpreted as a signal of project quality, because entrepreneur

only be in full-time in the project if he/she believes in firm success.

Hypothesis 2C – The higher the entrepreneur commitment, the higher the probability

of success of equity crowdfunding campaigns.

2.3. Information disclosure

The literature provides empirical evidence that investments in start-ups tend to be local

because of the importance of reputation and trust of entrepreneurs, as well the distance

costs (related to investment opportunities search, due diligence, and monitoring), but


12

crowdfunding platforms seem to reduce such frictions (Agrawal et al., 2015). In fact,

social networks, as well interactions with investors during the campaign, may diminish

such local funders advantages by reducing information asymmetries between

entrepreneurs and investors (S. Shane & Cable, 2002).

Some studies emphasize the role of information disclosure during the campaigns, in both

form of updates (new information provided by the entrepreneurs during the campaign)

and Q&A posted during the campaign (questions raised by investors and answered by the

entrepreneurs). These interactions contribute to reduce information asymmetries between

investors and entrepreneurs and they can have a positive influence on investor’s decisions.

In fact, there are already some empirical evidence about it. For instance, Hornuf &

Schwienbacher (2015) suggest that investors decisions take in account the updates as well

the comments of others investors during the campaigns in the German crowdinvesting

market, and Kuppuswamy & Bayus (2014) also find that, in the reward-crowdfunding

platform Kickstarter, additional backer support is positively influenced by project

updates.

Hypothesis 3: The success of an equity crowdfunding campaign is positively

influenced by the number of interactions between entrepreneurs and investors.

2.4. Other motivations

In addition to the financial motivations (dividends and capital gains), other factors may

influence the investment decisions of investors, such as tax reliefs and nonfinancial

rewards. In UK, investors may access to substantial tax reliefs when they acquire new

shares of small qualifying companies, according to Enterprise Investment Scheme (EIS)

and Seed Enterprise Investment Scheme (SEIS), but companies must satisfy some

requirements to be eligible for these tax reliefs. The SEIS scheme is more favorable to


13

investors than EIS. It is expected that projects of firms qualified for SEIS or EIS tax

scheme have a higher probability of success of their equity crowdfunding. However, that

is not empirically confirmed by previous studies on equity crowdfunding (Silvio Vismara,

2016; Vulkan et al., 2016).

Hypothesis 4A: The existence of tax reliefs has a positive impact on the success

of equity crowdfunding campaigns.

Future potential financial returns (dividends and capital gains) are usually the main

motivations of investors when they decide invest in start-ups. However, in equity

crowdfunding campaigns the firms sometimes also offer other nonfinancial rewards to

investors, such as product discount on purchases, exclusive series of products or services,

etc. The existence of these nonfinancial rewards can positively contribute to investment

decisions even that, like in the case of tax reliefs, the empirical evidence do not confirm

it (Cholakova & Clarysse, 2015; Silvio Vismara, 2016).

Hypothesis 4B: The existence of nonfinancial rewards has a positive impact on the

success of equity crowdfunding campaigns.

2.5. The competition effect

As the campaigns compete to each other for funding, if we consider that the available

funds from investors are relatively fixed, thus we can argue that the probability of funding

from each investor depends on the number (and quality) of other competitive campaigns

that are active during the campaign. Previous empirical evidence shows a negative effect

between the number of active campaigns and the probability of success for other

crowdfunding models, such as reward (Kuppuswamy & Bayus, 2014), microfinance (Ly

& Mason, 2012) and donation (Meer, 2014).


14

Hypothesis 5: The higher the number of competitors for funding (active campaigns),

the lower the probability of the campaign being successful.

2.6. Early investments

Previous research also finds that fundraising success is positively influenced by the early

investments, suggesting the existence of a herding behavior of investors.

Hornuf & Schwienbacher (2015), using a hand-collected data from four German

platforms of equity crowdfunding, find a decrease in the probability of the next

investment when it takes longer to conclude the previous ten investments. The importance

of early investments is also stated by Vulkan et al. (2016). Using a sample from Seedrs

platform, the authors find that one of the factors that are more strongly associated with

the probability of success is the capital share accumulated in the first week of the

campaign.

Vismara (2018) explores the role of information cascades among individual investors in

equity crowdfunding. He finds that the number of early investors positively influences

the campaign’s success and the higher percentage of investors with public profile, the

higher number of early investors, suggesting that undecided investors observe the

behavior of early investors to make their investment decisions (observational learning),

i.e., early investors transmit a relevant and positive signal to late investors. However, the

author recognizes that other explanations may justify his results: (i) attracting many early

project may have a positive payoff externality, because such campaigns have a higher

probability of being successful thus the investors avoid to waste time and resources in

investing in failed campaigns; (ii) large early investors reduce uncertainty but also

contribute to the accumulated amount raised and consequently increases the probability

of success; (iii) many early investors may leverage the effect of “word of mouth”, (iv)


15

some investors may prefer to invest later after seeing how the campaign develops and (v)

other behavioral causes, such as procrastination or conformity, may explain the late

biddings. In particular, the last explanation may justify the increased number of investors

in the last days of the successful campaigns (after they have already reach their target).

Even so, the author argues that herding can be a rational decision. Given that investors

are able to invest small amounts, such as 10 pounds, the costs of evaluation properly each

firm can be disproportionally high. Additionally, as many of the investors in equity

crowdfunding are not qualified, they cannot have the tools to distinguish good from weak

projects. So, it can be rational for investors to observe and to use the information about

the behavior of previous investors to make their investment decision.

The influence of early days in the success of crowdfunding campaigns is also empirically

observed in other models of crowdfunding, as in the reward model (Colombo et al., 2015;

Kuppuswamy & Bayus, 2014) or in the lending model (Zhang & Liu, 2012).

Colombo et al. (2015) use data from Kickstarter (a reward-based crowdfunding platform)

and evaluate the impact of early days of the campaign (between day one and one sixth of

its total period) by introducing in the analysis the variables “early bakers” (number of

investors in the early days) and “early capital” (amount raised in the early days). Both

number of backers as well amount of capital raised in the first days increase the

probability of crowdfunding success. According to the authors, the early investments

reduce the campaign uncertainty in three ways: (i) observational learning and pledge an

unsuccessful project can be a wasted time; (ii) generate word-of-mouth and (iii) a large

number of early backers potentially give more suggestions and feedback that help to

modify and consolidate the projects, increasing their success probability.


16

Kuppuswamy & Bayus (2014) also use data from Kickstarter to analyze the dynamics of

project support. They include in the model (conditional fixed effect Poisson model for the

number of backers added in a day) binary variables for the first seven days and last seven

days of project funding cycle and they find that these variables are positive and

significant, revealing that the propensity of the backers to pledge a campaign is higher in

the first and last weeks of the campaign. Additionally, the magnitude of its coefficients

decreases over the first week and increases over the last week, confirming the U-shaped

pattern of backer support. Despite they find evidence for positive effects of social

influence related to the percentage already raised (in percentage of its target), according

to their results, this pattern is not explained by the previous number of backers, so they

do not confirm the effect of the herding behavior of investors.

Hypothesis 6: The fundraising success in equity crowdfunding campaigns is

positively influenced by the early investments.

3. Sample and Methodology

3.1. Sample selection

The most active equity crowdfunding market in the world is in the United Kingdom.

According to Ziegler et al. (2018), the UK captured 73% of alternative finance European

market in 2016. Then, we selected the two biggest equity crowdfunding platforms from

UK - Crowdcube and Seedrs - to construct our sample. Crowdcube was the first equity

platform created in UK, in 2011, and it is (one of) the biggest platform of equity

crowdfunding in the world. This platform has currently an investor community of around

518,000 individuals and more than £404 million have been successfully invested through


17

the site2. Seedrs was the first equity platform in UK obtaining authorization to develop

equity crowdfunding activity from Financial Conduct Authority (in 2012) and, according

to its blog, since its launch, £310 million were invested through Seedrs Platform and more

than 590 deals were funded3.

There are no databases about crowdfunding campaigns and equity crowdfunding

platforms don’t provide other information beyond the available on their websites. Thus,

we use (daily) hand-collected information about campaigns, firms and teams from those

two platforms during three years. Additionally, some complementary information of the

companies (such us industry NACE4 codes, number and nationality of directors,

incorporation date and firm name) were obtained from Companies House website5,

Infoempresa.com6, Amadeus database and, in some cases, companies’ websites.

In our empirical analysis we use a sample that includes all the campaigns that started and

finished in both platforms between 7th April 2015 and 6th March 2018. After excluding

the 43 campaigns with other forms of crowdfunding (bonds, convertibles and funds), our

sample comprises 1,256 campaigns of equity crowdfunding (704 from Crowdcube and

552 from Seedrs).

The information hand-collected include: (i) campaign characterization variables (target

amount, percentage of equity offered, type of shares offered, tax reliefs, other rewards;

highest investment amount, duration of the campaign, number of updates and Q&A;

number of active campaigns in the day of campaign launching; (ii) Firm and team

2 As publicity in the site https://www.crowdcube.com/, accessed on April, 7 2018. 3 https://www.seedrs.com/learn/blog/seedrs-most-successful-year-2017-year-in-review, accessed on April,

7 2018. 4 Classification of economic activities used in the European Union. 5 Companies House is an executive agency from UK, sponsored by the Department for Business, Energy

& Industrial Strategy. See https://www.gov.uk/government/organisations/companies-house. 6 According to its website, https://www.infoempresa.com, this is a web service that provides financial and

economic information on Spanish companies, their directors and business administrators.


https://www.crowdcube.com/

https://www.seedrs.com/learn/blog/seedrs-most-successful-year-2017-year-in-review

http://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:European_Union_(EU)

https://www.gov.uk/government/organisations/companies-house

https://www.infoempresa.com/

18

information variables (country, industry, incorporation date; financial information of the

firm, number of team members, team commitment, qualifications and previous

experience of the first team member, and (iii) Daily evolution of the campaign (raised

amount and number of investors in each day while the campaign is active). The variables

are described in Appendix Table 1.

3.2. Descriptive analysis

3.2.1. Campaigns characterization and evolution

Table 1 shows the descriptive statistics7. From the 1,256 campaigns in our sample, 50.4%

(633) were successful (achieved the target amount). The target amount of each campaign

ranges from £17,000 to £5,000,000 and the stake of equity offered is always smaller than

50%. In Seedrs the shares offered are ordinary shares held by a nominee structure (N

shares) while in Crowdcube the most frequent is offering two types of shares, depending

on investment value. For small investments it is offered B investment shares (without pre-

emption and voting rights) and from a certain value are offered A ordinary shares (with

pre-emption and voting rights)8.

The effective highest amount raised for one campaign was almost £7.3 million (note that

campaigns can accept a higher amount of funding than the target). The campaign that

proportionally gets more funded, raised more than seven times the target amount. Around

50% of the campaigns have a large investor (who invest at least 20% of the target

amount).

7 The pairwise correlation matrix is presented in Appendix Table 2. 8 The Appendix Table 3 shows the distribution of campaigns by share type offered.


19

The average number of investors is 223, but 45 campaigns have more than 1,000 investors

(maximum of 3,709). In both platforms the minimum investment amount is, in general,

£10. Each campaign is usually active (online) for 30 days (or 45, until July 2015) in

Crowdcube and 60 days in Seedrs. However, some campaigns close sooner and others

extend the campaign for more days. The campaigns in our sample were active for 44 days

on average (between 1 and 132 days).

During the campaign, the team may interact with investors by giving updates of the

campaign or answering the questions posted by investors. Some teams are very active and

maintain a constant updating of the campaign, while others don’t use this faculty. On

average, each campaign has 8 updates and 12 questions posted by investors. The average

of the number of active campaigns in each platform is 29, and it ranges between 9 and 53

active campaigns in each day.

Table 1

Descriptive statistics (all campaigns). All variables are described in appendix Table 1.

Variable Unit N Mean Median SD Min Max

Campaigns Characterization

Target amount GBP ('000) 1,256 313.43 200.00 351.00 17.00 5,000.00

Equity retention % Equity 1,256 88.49 90.00 6.49 50.18 99.29

Raised amount GBP ('000) 1,256 319.24 137.55 541.33 0.06 7,289.56

Raised (%) % Target 1,256 92.69 100.00 0.76 0.00 7.32

Funded Dummy

(Yes=1/No=1) 1,256 0.504 1,00 0,50 0,00 1,00

Investors Nº 1,256 223.27 114.00 351.83 4.00 3,709.00

Highest amount GBP ('000) 1,050 88.64 35.02 185.44 0.02 2,859.58

Highest investment % Target 1,050 25.56 20.00 29.72 0.00 501.68

Large investor Dummy

(Yes=1/No=1) 1,050 0.50 1.00 0.50 0.00 1.00

Campaign duration Days 1,256 43.93 44.00 20.21 1.00 132.00

Updates Number 970 7.82 6.00 7.12 0.00 43.00

Q&A Number 974 12.31 8.00 13.25 0.00 140.00

Competitive campaigns Number; first day 1,176 29.16 29.00 9.09 9.00 53.00

Tax reliefs Dummy

(Yes=1/No=1) 1,252 0.89 1.00 0.32 0.00 1.00

Nonfinancial rewards Dummy

(Yes=1/No=1) 1,244 0.51 1.00 0.50 0.00 1.00


20

Team and Firm Information

Firm age Years 1,256 3.18 2.32 3.54 -0.23 46.39

Financial information Dummy

(Yes=1/No=1) 1,256 0.27 0.00 0.45 0.00 1.00

Firm value GBP ('000) 1,256 3,864.94 1,661.59 7,634.79 80.00 95,000.00

Sales GBP ('000) 323 1,311.88 475.59 2,561.34 0.08 29,616.20

Assets GBP ('000) 338 1,020.74 341.94 2,181.94 -18.73 26,639.31

Sales growth % (AAGR) 324 480.34 191.28 1,747.25 7.25 29,334.10

EBITDA margin % Sales 323 -1,854.06 -28.38 29,166.93 -524,102.60 138.90

Qualification/Experience Dummy

(Yes=1/No=1) 1,256 0.76 1.00 0.43 0.00 1.00

Team size Number 1,249 4.77 4.00 2.38 1.00 19.00

Commitment Dummy (Full-

time=1/Part-time=0) 551 0.92 1.00 0.27 0.00 1.00

Team members in full-time Number 550 2.88 2.00 1.75 0.00 13.00

Investors in the United Kingdom may benefit from tax reliefs when investing in start-ups.

There are two tax breaks: EIS (Enterprise Investment Scheme) and SEIS (Seed Enterprise

Investment Scheme). Both tax schemes aim to help small and high-risky firms raising

finance by offering tax relief on new shares. In our sample, almost all the firms are

qualified for one of these tax reliefs (89%). In order to attract more investors, some

campaigns offer other rewards to investors. In fact, in our sample around half of the

campaigns (50.72%) offer other rewards.

3.2.2. Firm and team information

Almost all the firms with campaigns in Crowdcube are from United Kingdom, however

in Seedrs the campaigns from other countries are more frequent9.

The firms with equity crowdfunding campaigns in both Crowdcube and Seedrs platforms

are in general young, small and with negative profitability. On average, the firms are three

years old but some are very young. More than 62% of the firms are young (less than 3

9 In our sample, there are campaigns from 9 different countries (Appendix Table 4) and they are

concentrated mainly in three industries (Appendix Table 5): information and communication wholesale and

retail trade and manufacturing.


21

years) and 16% are older than 5 years. The average pre-money valuation of the firms is

£3.86 million.

The financial information is only available for some firms in Crowdcube platform10, so

only 27% of firms in the all sample provide financial information. As expected for start-

ups, the firms in the sample are small (average sales of £1.3 million, 67% have lower

sales than £1 million and 76% of them have less than £1 million of assets), have negative

profitability (average EBITDA margin is -1854%) and have exponential expected growth

rate (the mean of the AAGR of sales is 480%).

The firm teams have between 1 and 19 members, but the maximum number of team

members committed in full-time with the project is 13 and 5 firms have no members in

full-time. In 76% of the campaigns the information about qualifications and/or experience

of the first member of the team is available.

3.2.3. Comparison between successful and unsuccessful campaigns

Table 2 compares the descriptive statistics between successfully and unsuccessfully

campaigns. As expected, the raised amount, the percentage of the raised amount and the

number of investors is much higher for successful campaigns. With exception of

campaign duration and tax reliefs, the other variables of campaigns characterization are

significantly different between successfully and unsuccessfully campaigns. However,

among the variables about the team and firm information, only financial information, firm

value, team size and team members in full-time reveal the existence of significant

differences between the two groups.

10 Until June of 2017, many of the Crowdcube campaigns provided historic and forecast financial statements

(profit and loss statement, balance sheet and cash-flow statement). However, since then, such information

is no longer available to the general public, except in some cases upon request directly to the founders.


22

Table 2

Descriptive statistics (successfully versus unsuccessfully campaigns). This table reports the descriptive

statistics (number of observations and mean) for successfully and unsuccessfully campaigns and the t-tests

on the differences between the two groups of campaigns.

Variable Unit Successful Unsuccessful Mean

Difference

(t-test) N Mean N Mean


Target amount GBP ('000) 633 345.10 623 281.20 63.859***

Equity retention % Equity 633 89.59 623 87.38 2.209***

Raised amount GBP ('000) 633 536.90 623 98.07 438.834***

Raised (%) % Target 633 153.10 623 31.35 121.716***

Investors Nº 633 373.10 623 71.01 302.102***

Highest amount GBP ('000) 581 125.90 469 42.49 83.403***

Highest investment % Target 581 35.65 469 13.05 22.601***

Large investor Dummy (Yes=1/No=1) 581 71.94 469 23.24 48.704***

Campaign duration Days 633 44.59 623 43.25 1.345

Updates Nº 533 9.77 437 5.45 4.321***

Q&A Nº 532 16.34 442 7.46 8.879***

Competitive campaigns Nº; first day 590 28.30 586 30.03 -1.729***

Tax reliefs Dummy (Yes=1/No=1) 632 0.90 620 0.88 0.02

Nonfinancial rewards Dummy (Yes=1/No=1) 633 0.54 611 0.48 0.061**


Firm age Years 633 3.16 623 3.20 -0.0390

Financial information Dummy (Yes=1/No=1) 633 0.30 623 0.25 0.047*

Firm value 1,000 GBP 633 4,812.00 623 2,902.00 1,910***

Sales 1,000 GBP 178 1,351.00 145 1,264.00 86.52

Assets 1,000 GBP 185 1,150.00 151 876.40 273.4

Sales growth % (AAGR) 179 372.60 145 613.30 -240.7

EBITDA margin % Sales 178 -178.70 145 -3,900.00 3700

Qualif/Experience Dummy (Yes=1/No=1) 633 0.75 623 0.77 -0.0180

Team size Nº 633 5.12 616 4.41 0.714***

Commitment Dummy (Yes=Full-

time/No=Part-time) 259 0.94 292 0.91 0.031

Team members in full-time Nº 259 3.23 291 2.57 0.661***

3.3. Methodology

Success on equity crowdfunding can be measured in different ways: (i) Funded; (ii)

Raised amount (in percentage of the target); (iii) Number of investors; (iv) Speed of

investment (number of days needed to complete financing). As both Crowdcube and

Seedrs platforms use the “all or nothing” model, the firms don’t receive anything if they


23

don’t get the target amount, regardless if they reached 1% or 99% of the target, or if they

have few or many investors. Additionally, the firms only receive the raised amount after

the campaign is closed. So, what really matters for firms is to reach the target (in fact,

except for team ego, it is not a big difference to reach the target sooner or later). Thus, we

use Funded as measure of success, i.e., a campaign is successful if it raises an investment

amount at least equal to its target.

To analyze the static drivers of fundraising success, we use a logistic regression where

Funded is the dependent variable. According to literature review, the factors that explain

the probability of success of the equity crowdfunding campaign are related to (i) Χ -

Project Quality (equity retention, presence of a large investor; firm age; firm value,

growth rate of sales; EBITDA margin); (ii) Ζ - Team Quality (information about

qualifications and/or previous experience; team size; first team member commitment);

(iii) Π - Information Disclosure (interactions between the team member and investors

during the campaign); (iv) Ώ - Other Motivations (tax reliefs and nonfinancial rewards)

and (v) Ψ - Competition Effect. The model also includes other Φ - Control Variables,

such as dummies for industry, duration of the campaign, dummy for platform (Crowdcube

or Seedrs), dummies for the type of shares.

The model for static drivers of success in equity crowdfunding campaigns can be

expressed in equation 1:

𝛄𝑖 = 𝚾′𝛽𝑖 + 𝚭′𝛾𝑖 + 𝚷′𝜂𝑖 + 𝛀′𝜑𝑖 + 𝚿′ϕ𝑖 + 𝚽′𝜃𝑖 + 𝜇𝑖 (1)

In order to test the effect of early investments on fundraising success, as Vulkan et al.

(2016), we start by adding the variable “percentage funded until the end of the first week”

(𝐘𝒊) on our analysis.

𝛄𝑖 = 𝐘𝒊 + 𝚾′𝛽𝑖 + 𝚭′𝛾𝑖 + 𝚷′𝜂𝑖 + 𝛀′𝜑𝑖 + 𝚿′ϕ𝑖 + 𝚽′𝜃𝑖 + 𝜇𝑖 (2)


24

4. Results

Table 3 reports the estimation results of logistic regressions for static drivers of

fundraising success (Equation (1)). Models (1) to (9) differ by the sample and variables

used.

4.1. Project Quality

To test the influence of project quality on the success of the equity crowdfunding

campaigns, we use four variables: equity retention, the presence of a large investor, firm

age and economic potential of the project/firm (firm value, growth sales rate, and

EBITDA margin11).

The coefficient of equity retention variable is positive and statistically significant in all

models. Our results are in line with previous research on equity crowdfunding (Ahlers et

al., 2015; Silvio Vismara, 2016; Vulkan et al., 2016) and confirm the hypothesis 1A, that

equity retention signals the project quality and positively influences the success of the

equity crowdfunding campaign.

As expected, the presence of a large investor12 increases the probability of the equity

crowdfunding campaign being successful, confirming hypothesis 1B. As Hornuf &

Schwienbacher (2015) and Vulkan et al. (2016), our results suggest that the presence of

11 In order to limit the influence of outliers, in the regression analysis, we use the winsorized variables of

growth sales rate and EBITDA margin. 12 We define a large investor as the one who invests at least 20% of the target amount. In our sample, 44%

of the campaigns have a large investor, according to this definition. Although the 20% trigger can be seen

as an aleatory value, the main conclusions remain valid when we use other triggers (ex: 30%, 40%, 50%).


25

a large investor (usually a sophisticated or, at least, a more informed investor) is a

valuable signal to undecided investors.

As previous research (Ahlers et al., 2015; Belleflamme et al., 2014), our results are not

conclusive about the relevance of firm age on the success of equity crowdfunding

campaigns (hypothesis 1C).

We don’t find evidence that economic potential of the project/firm influences the

investment decision (hypothesis 1D). Although the coefficient of the variable firm value

is significant in almost all models, its sign (negative) is not aligned with our expectations,

and there is no evidence that growth rate of sales or EBITDA margin significantly

influence the success of equity crowdfunding campaign. A probable explanation for this

evidence is that investors may suspect that firms with a higher valuation are overvalued

(and consequently the shares price is too high) discouraging them from investing in such

campaigns. Additionally, investors do not seem to attach much importance to the

forecasts of sales provided by founders. In fact, given the fact firms are relatively young

(around 42% of firms are less than two years old) and many of projects are related to new

products or services, the forecasts of sales have a high level of uncertainty. Additionally,

(almost) all the campaigns don’t provide details of assumptions used to develop such

forecasts. This way, we can argue that even if the economic potential of firms is relevant

to investment decisions, investors are cautious about the financial forecasts provided by

the founders and use other signals besides the growth and profitability projected by firm’s

founders to evaluate the economic potential of firms.


26

Table 3 Logistic regression models for static determinants of success in equity crowdfunding campaigns. This table reports logistic regressions for static drivers of fundraising success. The dependent variable is one if the

campaign is successful (funded) and zero, otherwise. Models (1) to (9) differ by the sample and variables used. Models (1) to (5) use the entire sample. Model (1) uses as independent variables only those related to the

project quality (equity retention, presence of a large investor, firm age and firm value). Model (2) adds the variables related to team quality (qualifications/experience, team size and first team member commitment), the model (3) adds the variables related to information disclosure (number of updates and number of Q&A), model (4) adds the variables related to other motivations (tax reliefs and nonfinancial rewards) and model (5)

include all variables, including the number of competitive campaigns. Models (6) and (7) use only the subsample of campaigns from Crowdcube platform, but model (7) includes new variables, only available in some

campaigns from this platform, about the economic potential of firm/project, (Sales growth and EBITDA margin). Models (8) and (9) use only the subsample of campaigns from Seedrs platform, but model (9) includes the variable, only available on Seedrs, about first team commitment (full-time/part-time). Control variables (industry, campaign duration, platform and shares type) are included in all models but not reported. All the

variables are defined in Appendix Table I. Standard errors are in parentheses. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively. The Hosmer–Lemeshow goodness-of-fit test (results not

reported) confirm that our model fits reasonably well in all logistic regressions.

(1) (2) (3) (4) (5) (6) (7) (8) (9)

PROJECT QUALITY

Equity retention 0.058*** 0.059*** 0.081*** 0.081*** 0.081*** 0.054** 0.079** 0.139*** 0.143***

(0.014) (0.014) (0.018) (0.018) (0.018) (0.023) (0.039) (0.034) (0.035)

Large investor (0/1) 0.023*** 0.023*** 0.023*** 0.023*** 0.023*** 0.025*** 0.029*** 0.023*** 0.025***

(0.002) (0.002) (0.002) (0.002) (0.002) (0.003) (0.005) (0.004) (0.004)

Firm age -0.049** -0.050** -0.020 -0.021 -0.020 -0.023 -0.142* -0.018 0.004

(0.024) (0.024) (0.027) (0.027) (0.027) (0.031) (0.086) (0.060) (0.060)

Firm value (ln) 0.067 -0.030 -0.551*** -0.547*** -0.545*** -0.476*** -0.785** -0.545** -0.545**

(0.091) (0.097) (0.130) (0.131) (0.131) (0.176) (0.356) (0.223) (0.226)

Growth sales -0.000

(0.000)

EBITDA margin 0.000

(0.000)

TEAM QUALITY

Qualifications/Experience

-0.170 -0.142 -0.119 -0.146 0.132 0.949 -0.089 -0.147

(0.212) (0.274) (0.275) (0.278) (0.558) (0.856) (0.357) (0.365)

Team size (n)

0.113*** 0.108** 0.105** 0.106** 0.153** 0.263** 0.080 0.070

(0.037) (0.044) (0.044) (0.044) (0.060) (0.107) (0.072) (0.073)

First team member commitment (0/1)

1.600**

(0.645)

INFORMATION DISCLOSURE

Updates (ln)

0.816*** 0.824*** 0.832*** 1.420*** 0.766** 0.546*** 0.569***

(0.137) (0.138) (0.138) (0.225) (0.367) (0.209) (0.214)

Q&A (ln)

1.073*** 1.054*** 1.055*** 1.114*** 1.755*** 1.109*** 1.103***

(0.133) (0.133) (0.133) (0.188) (0.371) (0.216) (0.220)


27

(1) (2) (3) (4) (5) (6) (7) (8) (9)

OTHER MOTIVATIONS

Tax reliefs (0/1)

-0.062 -0.041 0.147 -0.771 -0.344 -0.141

(0.326) (0.327) (0.467) (0.805) (0.512) (0.520)

Nonfinancial rewards (0/1)

0.002 -0.003 -0.193 0.041 0.502 0.514

(0.236) (0.236) (0.301) (0.516) (0.449) (0.457)

COMPETITION EFFECT

Competitive campaigns (n) 0.007 0.027* 0.041 -0.007 -0.001

(0.012) (0.015) (0.028) (0.020) (0.021)

Observations 1,019 1,017 816 811 811 528 227 283 283

Sample All All All All All Crowdcube Crowdcube Seedrs Seedrs

Pseudo R2 0.2398 0.2471 0.3328 0.3330 0.3335 0.3808 0.4385 0.3394 0.3558

Log likelihood -532.19 -525.86 -366.83 -363.94 -363.70 -220.71 -86.64 -124.91 -121.81


28

4.2. Team Quality

The team quality is measured, in our model, by three variables: availability of information

about qualifications or previous experience; team size and first member commitment

(full-time/part-time), but only the first two variables are available in both platforms. The

commitment (full-time/part-time) of the first team member (usually the founder and/or

CEO of the firm) is only available on Seedrs platform, so is only tested on model (9).

According to our hypothesis 2A, the team quality, measured by team qualifications and

previous experience, influences positively the success of equity crowdfunding

campaigns. The information about team qualifications and previous experience is not

always available, because the team members have the option to disclose or not such

information and, even when they do it, that information is not always displayed in the

same way, which makes difficult to use it for statistical purpose. So, as they probably

only decide to disclosure such information if they have high qualifications and/or

previous experience (in the industry or as entrepreneurs) is relevant, we use the

availability of such information as proxy of team quality. However, contrary to previous

literature (Ahlers et al., 2015; Piva & Rossi-Lamastra, 2018), that find a significant

empirical evidence that human capital is positively related to success of equity

crowdfunding campaigns, we don’t find empirical evidence that the availability of

information about qualifications and/or previous experience influences the success of

equity crowdfunding campaigns. This can be related to the fact that only 27% of the

campaigns do not disclose this information, which may bias the statistical results. So,


29

further investigation is need, using for instance other proxies for team qualifications or

previous experience13.

The coefficient of the variable team size is positive, as expected, but became statistically

insignificant for the subsample of Seedrs campaigns (models (8) and (9)). Thus, we can

only partially confirm the hypothesis 2b, i.e., the higher the team size the higher the

probability of success of equity crowdfunding campaigns, suggesting that the team size

is used as a sign of firm’s ability to deal with market uncertainty. Previous empirical

research is not conclusive too. Some find a positive relationship between the number of

team members and the success of the equity crowdfunding campaigns (Ahlers et al., 2015;

Silvio Vismara, 2016), while others don’t find evidence of such relationship (Hornuf &

Schwienbacher, 2015; Vulkan et al., 2016).

According to our hypothesis 2c, the higher the entrepreneur commitment, the higher the

probability of success of the equity crowdfunding campaign. We use the commitment of

the first person of the team (full-time or part-time) as proxy for entrepreneur commitment.

According to our empirical evidence, the coefficient of the variable commitment of the

first person of the team is significant and positive, as expected.

4.3. Information Disclosure

To evaluate the influence of information disclosure in the success of equity crowdfunding

campaigns we use two variables related to the interactions between team members and

investors during the campaign - the number of updates and the number of Q&A. The

13 The previous literature uses as proxies of human capital other variables, such as the percentage of MBA

graduates among executive board members of a founding team (Ahlers et al., 2015) or entrepreneurs’

business education and entrepreneurial experience (Piva & Rossi-Lamastra, 2018). However, unluckily

such team information in Seedrs and Crowdcube is not always available, and even if available it is dispersed

and difficult to standardize. Thus, given the sample size of our paper, it would be impracticable to hand

collect such information from other sources, such us LinkedIn Profiles of each team member of the 1,256

campaigns (remember that the average team size is 4.77).


30

coefficients associated with both variables are positive (as expected) and statistically

significant. This way, consistently with prior research (Hornuf & Schwienbacher, 2015;

Kuppuswamy & Bayus, 2014; E. Mollick, 2014) we find empirical evidence that the

success of an equity crowdfunding campaign is positively influenced by the number of

interactions between entrepreneurs and investors during the campaigns, confirming our

hypothesis 3b. Two reasons may contribute to this result. First, both types of interactions

(updates and Q&A) contribute to reduce the information asymmetry. Second, the

campaigns that catch more attention from potential investors (for example, because there

are focused on innovative products) are more likely to be able to be funded. In addition,

investors like to get effective and fast responses, and if that happens, as the dates and

responses are visible on the platform, more investors are likely to raise questions and

therefore have interest in investing.

4.4. Other Motivations

The success of an equity crowdfunding campaign seems not be significantly influenced

by the existence of tax reliefs or nonfinancial rewards. These results are aligned with

previous research on equity crowdfunding about the impact on investors decisions of tax

incentives (Silvio Vismara, 2016; Vulkan et al., 2016) and nonfinancial motives

(Cholakova & Clarysse, 2015; Silvio Vismara, 2016).

However, while our sample is balanced in order to the other nonfinancial rewards (50.7%

of the campaigns have other rewards), the same is not verified for the variable tax reliefs.

Investors may benefit from tax reliefs (EIS or SEIS) in almost 90% of the campaigns in

our sample. This way, the evidence on the irrelevance of tax reliefs for investment

decisions is not very robust.


31

The existence of other nonfinancial rewards also seems not having a significant effect on

the probability of success of the equity crowdfunding campaigns. Additionally, the

coefficient is not positive in all regression models, as expected, according to our

hypothesis 4b). This way, we cannot confirm that investors decide their investments

depending on the existence of tax reliefs or other nonfinancial rewards.

4.5. Competition Effect

If we analyze the probability of a campaign being successful according to the number of

active campaigns (Table 4), it looks that such probability is quite higher when there is a

small number of active campaigns (63.8% for less than 15 active campaigns in the day

that the campaign is launched) than when there is a large number of active campaigns

(38.6% for more than 45 active campaigns).

Table 4 – Probability of success according to the number of active campaigns (first day)

Number of active campaigns (camp)

TOTAL <15 16-25 26-35 36-45 >46

Unsuccessful 29 156 238 136 27 586

Successful 51 192 209 121 17 590

Total 80 348 447 257 44 1176

% successful campaigns 63.8% 55.2% 46.8% 47.1% 38.6% 46.10%

However, in the logistic regression model, the coefficient associated with the number of

active campaigns (in the launching day) is not statistically significant. So, we cannot

confirm our hypothesis 5, neither previous research on crowdfunding (E. Mollick, 2014;

Silvio Vismara, 2016). This result may be related to the fact that equity crowdfunding can

be classified as a two-sided market according to the definition of Rysman (2009)14. In this

market, the platform is the intermediary and the two sets of agents are the investors and

14 According to the author, “a two-sided market is one in which 1) two sets of agents interact through an intermediary or platform, and 2) the decisions of each set of agents affects the outcomes of the other set of agents, typically through an externality”. (p. 125)


32

the entrepreneurs/companies. A platform is attractive to investors if many and quality

projects are raising capital there, and it is attractive to entrepreneurs if it has a lot of

registered investors. Neither investors nor entrepreneurs will be interested in using the

platform if the other part is not. Recognizing this market structure, platforms make an

effort to attract good projects, as well many investors. For instance, Seedrs pays to anyone

that refers a new investor, 50% of the fees they charge to the business on any investment

the investor makes over two years.

Additional, as many of the investors in each campaign are from the team’s network

(family & friends) (Agrawal et al., 2015), each new campaign usually helps platforms

attracting new investors, who can also invest in other campaigns. So, a higher number of

active campaigns is probably associated with a higher number of new investors in the

platform, with probable positive effect on campaigns success, i.e., the opposite of the

competition effect previously described. These opposite effects may explain why the

coefficient of the number of active campaigns in the launching day is not statistically

significant. So, further investigation on this issue is needed.

4.6. Early investments

In this chapter, we analyze the evolution of equity crowdfunding campaigns during the

period the campaign is active and evaluate the impact of such evolution on fundraising

success.

As shown in graphs 1 and 2, the investment pattern in equity crowdfunding in our sample

is U-shaped. The campaigns are more active in the firsts and last days – these two periods

capture the higher percentage of investors and amount raised. This effect is more

pronounced for successful campaigns, suggesting that first days are relevant to determine

the campaign’s success.


33

In order to test the effect of early investments on fundraising success, as Vulkan et al.

(2016), we start by adding the variable “percentage funded until the end of the first week”

on our analysis. Table 5 reports the estimation results for regressions including the impact

of first days investments (according to equation (2)).

As stated by other empirical studies on equity crowdfunding (Hornuf & Schwienbacher,

2015; S. Vismara, 2018; Vulkan et al., 2016), our results confirm that fundraising success

is positively influenced by the early investments, suggesting the existence of a herding

behavior of investors. Not only the coefficient of the variable “% Funded on the first

week” is positive and statistically significant, but also the quality of the model (measured

by the pseudo r-squared) increases substantially.

54,8%

19,6%17,6%

1,0%

0%

20%

40%

60%

10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Time elapsed (% campaign duration)

Graph 1 - Investment raised (%target) during campaign

Successful Unsuccessful

27,0%

13,3%

51,3%

3,0%0%

20%

40%

60%

10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Time elapsed (% campaign duration)

Graph 2 - Percentage of total investors during campaign

Successful Unsuccessful


34

Table 5

Logistic regression models for determinants of success in equity crowdfunding campaigns. This table

reports logistic regressions for fundraising success, including a variable about early investments. The

dependent variable is one if the campaign is successful (funded) and zero, otherwise. Models (10) to (12)

differ by the sample and variables used. Model (10) use the entire sample and therefore excludes the

variables only available on Crowdcube or on Seedrs platform. Model (11) includes the variable, only

available in some campaigns of Crowdcube platform, about the economic potential of firm/project, (Sales

growth and EBITDA margin). Model (12) includes the variable, only available on Seedrs, about team

commitment (First team member commitment). Control variables (industry, campaign duration, platform

and shares type) are included in all models but not reported. All the variables are defined in Appendix Table

I. Standard errors are in parentheses. ***, **, and * indicate significance at the 1%, 5%, and 10% levels,

respectively. The Hosmer–Lemeshow goodness-of-fit test (results not reported) confirm that our model fits

reasonably well in all logistic regressions.

Model (10) (11) (12)

EARLY INVESTMENTS % Raised on first week 0.067*** 0.095*** 0.067***

(0.007) (0.020) (0.011)

PROJECT QUALITY Equity retention 0.085*** 0.088 0.179***

(0.022) (0.055) (0.043)

Large investor (0/1) 0.015*** 0.015*** 0.016***

(0.002) (0.006) (0.005)

Firm age -0.014 -0.055 -0.010

(0.034) (0.096) (0.067)

Firm value (ln) -0.932*** -1.383*** -0.895***

(0.169) (0.471) (0.284)

Sales growth -0.000

(0.000) EBITDA margin 0.001

(0.000) TEAM QUALITY

Qualifications/Experience 0.089 0.720 -0.047

(0.325) (1.120) (0.437)

Team size (n) 0.094* 0.254* 0.033

(0.053) (0.139) (0.089)

First team member commitment (0/1) 1.353*

(0.702)

INFORMATION DISCLOSURE Updates (ln) 0.813*** 0.823 0.445*

(0.168) (0.512) (0.252)

Q&A (ln) 0.884*** 1.523*** 1.101***

(0.159) (0.461) (0.261)

OTHER MOTIVATIONS Tax reliefs (0/1) -0.180 -1.143 -0.595

(0.388) (1.193) (0.578)

Nonfinancial rewards (0/1) 0.169 -0.187 0.574

(0.279) (0.669) (0.526)

COMPETITION EFFECT Competitive campaigns (n) 0.012 0.079** 0.027

(0.014) (0.038) (0.028)

Observations 811 227 283

Sample All Crowdcube Seedrs

Pseudo R2 0.5176 0.6353 0.5194

Log likelihood -263.22 -56.27 -990.88


35

4.7. Robustness check

Given the different definitions of success, we replicate the analysis using as dependent

variables the number of investors and the percentage raised. For testing the determinants

of the number of investors is used the negative binomial regression model (NBRM)15 and

for testing the determinants of the percentage of target raised is used ordinary least

squares model (OLS). The results are very similar, as can be seen in Table 6. Even so, we

can identify some differences between the drivers of funded (see the results of the logistic

model in Table 3) and the drivers of the number of investors (NBRM model). While the

results for the determinants of the percentage of target raised (OLS model) are very

similar to the previous ones, we see some significant differences to the determinants of

the number of investors. In the NBRM model, (i) the coefficient of the variable equity

retention is negative and no more statistically significant when we introduce the variables

related to economic potential of firm/project or first team commitment; (ii) the coefficient

associated to the presence of a large investor is not significant on Seedrs sample, when

we introduce the variable related to the commitment of first team member, while (iii) the

coefficient of firm value is now positive (negative in the others models) (iv) in Seedrs

sample, the existence of other nonfinancial reward has a significant positive impact on

the increase of number of investors. These apparently contradictory results are probably

related to the existence of different motivations of large investors (who invest large

amounts that are critical to the campaign reach the target, but they are in a small number)

from small investors. However, further research is needed about the differences between

the investment motivations for large and small investors.

15 In our sample, the Poisson model is not adequate because of the presence of over dispersion. The mean

number of investors is 223 and its variance is 123,781, so the variance is 555 times higher than the mean.


36

Table 6 Robustness Check for measures of success. This table reports the robustness check for different regression models and different measures of success (number of investors and

percentage raised). Models (13), (15) and (17) uses the Negative Binomial Regression Model (NBRM) and the number of investors as dependent variable. Models (14), (16)

and (18) use the Ordinary Least Squares Model (OLS) and the percentage raised as dependent variable. Models (13) and (14) use the entire sample and therefore excludes the

variables only available on Crowdcube or on Seedrs platform. Models (15) and (16) include the variables only available in some campaigns of Crowdcube platform, about the

economic potential of firm/project, (Sales growth and EBITDA margin). Models (17) and (18) include the variable, only available on Seedrs, about team commitment (First

team member commitment). Control variables (industry, campaign duration, platform and shares type) are included in all models but not reported. All the variables are defined

in Appendix Table I. Standard errors are in parentheses. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

(13) (14) (15) (16) (17) (18)

Model NBRM OLS NBRM OLS NBRM OLS

Dependent Variable Nº Inv. % Raised Nº Inv. % Raised Nº Inv. % Raised

EARLY INVESTMENTS

% Raised on first week 0.010*** 0.890*** 0.011*** 0.918*** 0.009*** 0.886***

(0.001) (0.045) (0.001) (0.093) (0.001) (0.085)

PROJECT QUALITY

Equity retention -0.012*** 2.014*** -0.010 2.156*** -0.011* 2.940***

(0.003) (0.290) (0.007) (0.582) (0.006) (0.514)

Large investor (0/1) 0.001*** 0.318*** 0.002** 0.370*** 0.001 0.293***

(0.000) (0.034) (0.001) (0.067) (0.001) (0.063)

Firm age 0.005 0.024 0.011 -0.222 -0.011 -0.283

(0.005) (0.421) (0.009) (0.797) (0.012) (1.063)

Firm value (ln) 0.191*** -10.637*** 0.153*** -14.745*** 0.210*** -12.208***

(0.025) (2.062) (0.052) (4.515) (0.039) (3.483)

Sales growth -0.000 -0.004

(0.000) (0.006)

EBITDA margin 0.000 0.005

(0.000) (0.003)


37

TEAM QUALITY

Qualifications/Experience -0.108** 0.527 -0.157 12.841 -0.137** 0.077

(0.055) (4.488) (0.139) (12.222) (0.067) (5.931)

Team size (n) 0.034*** 1.594** 0.029* -0.045 0.038*** 1.961*

(0.009) (0.699) (0.016) (1.432) (0.013) (1.092)

First team member commitment (0/1) 0.042 2.301

(0.123) (10.593)


Updates (n) 0.211*** 12.799*** 0.205*** 10.325** 0.128*** 9.340***

(0.027) (2.242) (0.059) (5.202) (0.039) (3.457)

Q&A (n) 0.452*** 18.844*** 0.505*** 28.566*** 0.376*** 16.944***

(0.024) (2.101) (0.050) (4.701) (0.036) (3.299)

OTHER MOTIVATIONS Tax reliefs (0/1) 0.013 -2.375 0.019 -6.733 -0.029 -12.691

(0.063) (5.231) (0.126) (11.277) (0.095) (8.358)

Nonfinancial rewards (0/1) 0.006 -0.474 0.031 -3.454 0.157* 0.731

(0.047) (3.966) (0.085) (7.586) (0.081) (7.354)

COMPETITION EFFECT Competitive campaigns (n) -0.006** -0.115 -0.010** 1.070** -0.003 -0.560

(0.002) (0.198) (0.004) (0.415) (0.004) (0.375)

Observations 818 818 236 236 283 283

Sample All All Crowdcube Crowdcube Seedrs Seedrs

R-squared/ Pseudo-R-squared 0.1132 0.651 0.114 0.677 0.100 0.623


38

In sum, according to our empirical evidence the factors that most influence the campaign

success are (i) the quality of the project, signalized by equity retention and the presence

of a large investor; (ii) the information disclosure (interactions between entrepreneurs and

investors during the campaign) and (iii) the early investments. There is no evidence that

firm age and economic potential of firm are relevant for investor decisions. Investors

probably use other signals (than growth and profitability perspectives of firms) to evaluate

the economic potential of firms and project quality. The table 7 summarizes the

comparison between the expected sign of each variable and the empirical evidence.

Table 7

Expected sign and empirical evidence sign of each variable. This table summarizes the comparison

between the expected sign of each variable and the empirical evidence.

Hypothesis Variable Expected Sign Empirical Evidence

Project Quality

1a Equity retention (+) (+)

1b Large investor (+) (+)

1c Firm Age (+/-) (?)

1d Economic potential of firm (+) (?)

Team Quality

2a Qualifications/previous experience (+) (?)

2b Team size (+) (?)

2c Entrepreneur commitment (+) (+)

Information Disclosure

3 Interactions during campaign (+) (+)

Other motivations

4a Tax reliefs (+) (?)

4b Nonfinancial rewards (+) (?)

Competition effect

5 Competitors (-) (?)

Early investments

6 % Raised on first week (+) (+)

We also check if the determinants of fundraising success are different between the

successful campaigns and the unsuccessful campaigns (see Table 8). The results remain

similar.


39

Table 8

Robustness check – comparing the fundraising factors for successful and unsuccessfully campaigns.

This table compares successful and unsuccessful campaigns using the number of investors and percentage

raised as measure of success16. Models (19) and (20) use the Ordinary Least Squares Model (OLS) and the

percentage raised as dependent variable. Models (21) and (22) use the Negative Binomial Regression Model

(NBRM) and the number of investors as dependent variable. Control variables (industry, campaign

duration, platform and shares type) are included in all models but not reported. All the variables are defined

in Appendix Table I. Standard errors are in parentheses. ***, **, and * indicate significance at the 1%, 5%,

and 10% levels, respectively.

(19) (20) (21) (22)

OLS_S OLS_U NBRM_S NBRM_U

Dependent Variable % Raised % Raised Nº Investors Nº Investors

EARLY INVESTMENTS

% Raised on first week 0.523*** 0.815*** 0.004*** 0.016***

(0.055) (0.057) (0.001) (0.002)

PROJECT QUALITY

Equity retention 2.352*** 0.325** -0.024*** -0.013***

(0.394) (0.143) (0.004) (0.005)

Large investor (0/1) 0.178*** 0.089*** -0.002*** -0.002**

(0.046) (0.021) (0.000) (0.001)

Firm age 0.387 0.078 0.006 0.013**

(0.632) (0.182) (0.006) (0.006)

Firm value (ln) -7.799*** 0.870 0.318*** 0.154***

(2.702) (1.142) (0.025) (0.038)

TEAM QUALITY

Qualifications/Experience 3.011 0.938 -0.048 -0.100

(5.768) (2.316) (0.054) (0.076)

Team size (n) 1.348 -0.090 0.022*** 0.022*

(0.879) (0.380) (0.008) (0.013)


Updates (n) 7.453** 2.941*** 0.060** 0.157***

(3.219) (1.064) (0.029) (0.035)

Q&A (n) 18.227*** 2.370** 0.364*** 0.354***

(2.936) (1.027) (0.025) (0.034)

OTHER MOTIVATIONS

Tax reliefs (0/1) -3.177 3.880 0.000 0.051

(6.816) (2.698) (0.062) (0.090)

Nonfinancial rewards (0/1) -3.955 4.490** -0.012 0.121*

(5.038) (2.075) (0.045) (0.070)

COMPETITION EFFECT

Competitive campaigns (n) -0.256 -0.013 -0.006** -0.006

(0.246) (0.109) (0.002) (0.004)

Observations 494 324 494 324

Sample

Sucessfull

campigns

Unsucessfull

campigns

Sucessfull

campigns

Unsucessfull

campigns

R-squared/ Pseudo R-squared 0.412 0.696 0.1155 0.1039

16 Note that this comparison is not possible for logistic regression because it uses “funded” as dependent

variable (the variable assumes the value 1 if the campaign is successful and 0, otherwise).


40

5. Conclusions

This paper explores the drivers of fundraising success in equity crowdfunding, using an

unique and large hand-collected database from the two major platforms of equity

crowdfunding in the United Kingdom (Seedrs and Crowdcube) for the period between

2015 and 2018. According to the literature review, we analyze the relevance of the project

and team characteristics, information disclosure, other motivations, competition and early

investment on the probability of success of the equity crowdfunding campaigns.

In addition to contributing to a better understanding of equity crowdfunding in general,

our study provides some useful indications to entrepreneurs and platforms in order to

ensure a higher probability of success of their campaigns. We find that the factors that

most contribute to the success of equity crowdfunding campaigns are the equity retention,

the presence of a large investor, the entrepreneur commitment with the project,

maintaining an intensive interaction with the investors during the campaign (providing

updates of the project and answer to the Q&A posed by investors) and a ensuring a good

start of the campaign (through, for instance, capturing some relevant early investments

before the campaign is active). These results suggest that campaign’ success depend not

only on the intrinsic quality of the project but also on the promoter's commitment before

and during the campaign.

However, our empirical results are not conclusive about the relevance of firm age,

qualifications/previous experience, team size and competition. We don’t find evidence

that the existence of other motivations (tax reliefs and nonfinancial rewards) influence

the investment decision, as well the economic potential of the project/firm. Although the

coefficient of the variable “firm value” is significant, its sign is not aligned with our


41

expectations, suggesting that investors may suspect that a high valuation may signal the

overvaluation of the firm. Additionally, investors do not seem to attach much importance

to the forecasts of sales and profitability provided by founders. Thus, we can argue that

even if the economic potential of firms is relevant to investment decisions, investors are

cautious about the financial forecasts provided by the founders and use other signals

besides the growth and profitability projected by firm’s founders to evaluate the economic

potential of firms.

Although our study contributes to a better understanding of the equity crowdfunding, we

recognize that it has some limitations and we can identify some avenues for future

research.

First, our sample includes only public campaigns available to the general investors.

However, beyond both Crowdcube and Seedrs platforms do not accept all the campaigns

that apply for funding, they also provide the possibility to the campaign launches privately

first. Thus, as the sample only includes the pre-selected campaigns, our results could be

skewed, depending on which criteria are used by the platforms to reject the campaign

before being publicized on the public site. However, we believe this is not a major

problem, since that, according to information provided by the platforms, they just confirm

some basic legal, financial and compliance information before accepting the campaigns,

as well give some suggestions to improve the success of the campaign. Even so, further

research can focus on the analysis of which factors that most influence the pre-selection

of the campaigns by the platforms.

The competition effect wasn´t confirmed by our study what, as previously said, may be

due to the fact that equity crowdfunding can be classified as a two-sided market. As many

of the investors in each campaign are from the team’s network (family & friends)


42

(Agrawal et al., 2015), each new campaign usually helps platforms attracting new

investors, who can also invest in other campaigns. So, a higher number of active

campaigns is probably associated with a higher number of new investors in the platform,

with, contrary to the competition, could have a positive effect on campaigns success.

These opposite effects may explain why the coefficient of the number of active campaigns

in the launching day is not statistically significant, but further investigation on this issue

is needed.

Finally, we do not distinguish the projects (campaigns) that are searching for equity

crowdfunding for the first time from those that are in subsequent rounds. This raises the

questions whether the probability of success and the drivers of fundraising success remain

the same or are different in the subsequent rounds. This question should be raised in future

research.


43

Appendix Table 1 List and definition variables.

Label Type of variable Description


Funded Dummy

A campaign is funded if the raised amount is equal (or higher) to

target. The variable assumes 1 if the campaign is funded, and 0

otherwise.

Target amount Numerical (>0) Target funding amount.

Equity retention Numerical (0-100) Percentage of equity retained [(1- equity offered) * 100].

Raised amount Numerical (>0) Raised amount until the last day of the campaign.

Raised (%) Numerical (>0%) Ratio between the raised and target amount*100.

Investors Numerical (>0) Number of investors.

Highest amount Numerical (>0) Highest investment amount.

Highest investment Numerical (>0%) Ratio between the highest investment and target amount*100.

Large investor Dummy

Presence of a large investor (with an investment higher than 20% of

target). The variable assumes 1 if the campaign has a large investor, and

0 otherwise.

Campaign

Duration Numerical (>0) Effective duration of the funding campaign (number of days).

Updates Numerical (≥0) Number of updates about the project during the campaign.

Q&A Numerical (≥0) Number of Q&A during the campaign.

Competitive

campaigns Numerical (≥0) Number of active campaigns when the campaign is launched.

Tax reliefs Dummy Qualification of firm for tax relief (EIS or SEIS). The variable

assumes 1 if the firm is eligible for tax relief, and 0 otherwise.

Nonfinancial

rewards Dummy

Presence of other rewards (beyond the shares of the company). The

variable assumes 1 if there is another reward, and 0 otherwise.


Firm age Numerical (≥0) Age of the firm (years).

Firm value Numerical (>0) Pre-money value of firm.

Sales Numerical (≥0) Expected sales in the current year. (*)

Assets Numerical (≥0) Assets in current year. (*)

Sales growth Numerical (%)

Expected average annual growth rate (AAGR) of sales between year

n+2 and n-1 (when sales of year n-1 are available, it is used sales of

year n). (*)

EBITDA margin Numerical (>0%) Expected EBITDA margin (EBITDA/Sales) in current year*100. (*)

Qualification/

Experience Dummy

Availability of information about qualifications and/or previous

professional experience of the first team member. The variable

assumes 1 if such information is available, and 0 otherwise.

Team size Numerical (>0) Number of project team members.

Commitment Dummy Commitment of the first person of the team with the project. The

variable assumes 1 if it is in full-time, and 0 if it is in part-time. (**)

Team members in

full-time Numerical (≥0) Number of project team members in full-time in the project. (**)

Other control variables

Industry Categorical Industry code according to NACE- Rev. 2.

Platform Categorical Platform where the campaign is launched.

Type of shares Categorical

Type of shares offered to investors [A - Ordinary shares (with pre-

emption and voting rights); B - Investment shares (without pre-

emption and voting rights); N- Ordinary shares held by a nominee

structure; O - Other shares (with limited pre-emption or voting rights)]

(*) Only available for campaigns in Crowdcube. (**) Only available for campaigns in Seedrs.


44

Appendix Table 2 Pairwise correlation matrix. Variables: (1) Funded; (2) Target amount; (3) Equity retention; (4) Raised amount; (5) Raised (%); (6) Investors; (7) Highest investment; (8) Presence

of a large investor; (9) Campaign duration; (10) Updates; (11) Q&A; (12) Competitive campaigns; (13) Tax reliefs; (14) Nonfinancial rewards; (15) Firm age; (16) Firm value; (17)

Sales growth; (18) EBITDA margin; (19) Qualifications/Experience (20) Team size; (21) Commitment of first team member. Number of observation in brackets. * p<0.1.

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21)

(1) 1.000

(1256)

(2) 0.091* 1.000

(1256) (1256)

(3) 0.170* 0.011 1.000

(1256) (1256) (1256)

(4) 0.406* 0.788* 0.144* 1.000

(1256) (1256) (1256) (1256)

(5) 0.799* 0.107* 0.257* 0.511* 1.000

(1256) (1256) (1256) (1256) (1256)

(6) 0.430* 0.542* 0.183* 0.782* 0.518* 1.000

(1256) (1256) (1256) (1256) (1256) (1256)

(7) 0.378* 0.071* 0.161* 0.296* 0.567* 0.113* 1.000

(1050) (1050) (1050) (1050) (1050) (1050) (1050)

(8) 0.484* 0.070* 0.148* 0.227* 0.489* 0.108* 0.547* 1.000

(1050) (1050) (1050) (1050) (1050) (1050) (1050) (1050)

(9) 0.033 -0.061* 0.102* -0.064* -0.031 -0.093* -0.013 0.052* 1.000

(1256) (1256) (1256) (1256) (1256) (1256) (1256) (1050) (1256)

(10) 0.279* 0.149* -0.010 0.214* 0.254* 0.266* 0.003* 0.058* 0.029 1.000

(872) (872) (872) (872) (872) (872) (850) (850) (872) (872)

(11) 0.416* 0.380* 0.168* 0.493* 0.471* 0.592* 0.116* 0.145* -0.064* 0.363* 1.000

(957) (957) (957) (957) (957) (957) (930) (930) (957) (860) (957)


45

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21)

(12) -0.095* -0.074* 0.052 -0.109* -0.135* -0.138* -0.054* -0.039 0.271* -0.242* -0.117* 1.000

(1176) (1176) (1176) (1176) (1176) (1176) (1027) (1027) (1176) (970) (974) (1176)

(13) 0.031 -0.120* -0.088* -0.079* 0.009 -0.065* -0.089* -0.044 -0.043 0.077* -0.035 -0.063* 1.000

(1252) (1252) (1252) (1252) (1252) (1252) (1047) (1047) (1252) (870) (954) (1173) (1252)

(14) 0.061* 0.097* -0.107* 0.085* 0.064* 0.149* -0.066 -0.050 -0.292* 0.290* 0.181* -0.294* 0.066* 1.000

(1244) (1244) (1244) (1244) (1244) (1244) (1046) (1046) (1046) (869) (954) (1242) (1166) (1240)

(15) -0.006 0.197* 0.111* 0.185* 0.031 0.174* -0.002 -0.025 -0.001 0.110* 0.078* -0.066* -0.058* 0.098* 1.000

(1256) (1256) (1256) (1256) (1256) (1256) (1050) (1050) (1256) (872) (957) (1176) (1252) (1244) (1256)

(16) 0.125* 0.658* 0.338* 0.632* 0.185* 0.571* 0.074* 0.062* -0.065* 0.126* 0.387* -0.024 -0.217* 0.022 0.187* 1.000

(1256) (1256) (1256) (1256) (1256) (1256) (1050) (1050) (1256) (872) (957) (1176) (1252) (1244) (1256) (1256)

(17) -0.069 -0.045 0.06 -0.050 -0.028 -0.054 0.075 0.073 0.061 0.080 -0.077 -0.028 -0.001 -0.129* -0.050 -0.016 1.000

(324) (324) (324) (324) (324) (324) (281) (281) (324) (245) (249) (291) (320) (322) (324) (324) (324)

(18) 0.064 -0.014 -0.039 0.001 0.010 0.020 -0.086 -0.061 -0.036 -0.072 0.059 0.059 -0.014 0.114* -0.008 -0.006 -0.926* 1.000

(323) (323) (323) (323) (323) (323) (280) (280) (323) (244) (248) (290) (319) (321) (323) (323) (322) (323)

(19) -0.021 -0.005 -0.071* -0.020 -0.021 0.011 -0.094* -0.0234 -0.238* 0.199* 0.059* -0.087* 0.097 0.285* 0.044 -0.016 0.027 -0.015 1.000

(1256) (1256) (1256) (1256) (1256) (1256) (1050) (1050) (1256) (872) (957) (1176) (1252) (1244) (1256) (1256) (324) (323) (1256)

(20) 0.150* 0.282* 0.184* 0.265* 0.179* 0.232* 0.090* 0.098* 0.077* 0.095* 0.158* 0.006 -0.044 0.024 0.104* 0.218* -0.019 0.028 -0.1213 1.000

(1249) (1249) (1249) (1249) (1249) (1249) (1048) (1048) (1249) (870) (955) (1171) (1245) (1242) (1249) (1249) (322) (321) (1249) (1249)

(21) 0.057 -0.005 0.151* 0.033 0.048 0.042 -0.036 -0.039 -0.009 0.011 0.065 -0.085* -0.038 0.045 -0.062 0.052 . . -0.033 0.046 1.000

(551) (551) (551) (551) (551) (551) (440) (440) (551) (316) (388) (518) (551) (545) (551) (551) (0) (0) (551) (551) (551)


46

Appendix Table 3

Frequency distribution of shares types. A shares are ordinary shares (with pre-emption and voting rights),

B shares are investment shares (without pre-emption and voting rights), N shares are ordinary shares held

by a nominee structure and O Shares are other shares types.

Shares Type Freq. Percent Cum.

A 122 9.74 9.74

A,B 435 34.72 44.45

A,B,O 1 0.08 44.53

A,N 11 0.88 45.41

A,O 3 0.24 45.65

B 101 8.06 53.71

B,O 5 0.4 54.11

N 562 44.85 98.96

N,O 2 0.16 99.12

O 11 0.88 100

Total 1,253 100.00

.

Appendix Table 4

Frequency distribution of crowdfunding campaigns by firm country

Country Frequency %

United Kingdom 1,188 94,74

Portugal 14 1,12

Spain 13 1,04

Germany 10 0,80

Ireland 8 0,64

Netherlands 6 0,48

France 3 0,24

Isle of Man 2 0,16

Switzerland 2 0,16

United States 2 0,16

Australia 1 0,08

Belgium 1 0,08

Czech Republic 1 0,08

Luxembourg 1 0,08

Malta 1 0,08

Malta, United Kingdom 1 0,08

Total 1,254 100.00


47

Appendix Table 5

Frequency distribution of crowdfunding campaigns by industry

Industry Freq. Percent

Information and communication 347 28.30

Wholesale and retail trade; repair of motor vehicles and

motorcycles 214 17.46

Manufacturing 168 13.70

Administrative and support service activities 106 8.65

Professional, scientific and technical activities 87 7.10

Accommodation and food service activities 71 5.79

Others activities 233 19.00

Total 1,226 100.00


48

References

Agrawal, A., Catalini, C., & Goldfarb, A. (2015). Crowdfunding: Geography, Social Networks,

and the Timing of Investment Decisions. Journal of Economics & Management Strategy,

24(2), 253-274.

Ahlers, G. K. C., Cumming, D., Günther, C., & Schweizer, D. (2015). Signaling in Equity

Crowdfunding. Entrepreneurship Theory and Practice, 39(4), 955-980.

Baum, J. A. C., & Silverman, B. S. (2004). Picking winners or building them? Alliance,

intellectual, and human capital as selection criteria in venture financing and performance

of biotechnology startups. Journal of Business Venturing, 19(3), 411-436.

Belleflamme, P., Lambert, T., & Schwienbacher, A. (2013). Individual crowdfunding practices.

Venture Capital, 15(4), 313-333.

Belleflamme, P., Lambert, T., & Schwienbacher, A. (2014). Crowdfunding: Tapping the right

crowd. Journal of Business Venturing, 29(5), 585-609.

Bernstein, S. a. K., Arthur G. and Laws, Kevin. (2015). Attracting Early Stage Investors: Evidence

from a Randomized Field Experiment. Journal of Finance, Forthcoming.

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.

Busenitz, L. W., Fiet, J. O., & Moesel, D. D. (2005). Signaling in Venture Capitalist—New

Venture Team Funding Decisions: Does It Indicate Long-Term Venture Outcomes?

Entrepreneurship Theory and Practice, 29(1), 1-12.

Carpentier, C., & Suret, J.-M. (2015). Angel group members' decision process and rejection

criteria: A longitudinal analysis. Journal of Business Venturing, 30(6), 808-821.

Chen, X.-P., Yao, X., & Kotha, S. (2009). Entrepreneur Passion and Preparedness in Business

Plan Presentations: A Persuasion Analysis of Venture Capitalists' Funding Decisions. The

Academy of Management Journal, 52(1), 199-214.

Cheng, S. (2008). Board size and the variability of corporate performance. Journal of Financial

Economics, 87(1), 157-176.

Cholakova, M., & Clarysse, B. (2015). Does the Possibility to Make Equity Investments in

Crowdfunding Projects Crowd Out Reward-Based Investments? Entrepreneurship

Theory and Practice, 39(1), 145-172.

Coad, A., Segarra, A., & Teruel, M. (2016). Innovation and firm growth: Does firm age play a

role? Research Policy, 45(2), 387-400.

Colombo, M. G., Franzoni, C., & Rossi-Lamastra, C. (2015). Internal Social Capital and the

Attraction of Early Contributions in Crowdfunding. Entrepreneurship: Theory &

Practice, 39(1), 75-100.

Czarnitzki, D., & Kraft, K. (2004). An empirical test of the asymmetric models on innovative

activity: who invests more into R&D, the incumbent or the challenger? Journal of

Economic Behavior & Organization, 54(2), 153-173.

Denis, D. J. (2004). Entrepreneurial finance: an overview of the issues and evidence. Journal of

Corporate Finance, 10(2), 301-326.

Eisenhardt, K. M. (2013). Top management teams and the performance of entrepreneurial firms.

Small Business Economics, 40(4), 805-816.

Eisenhardt, K. M., & Schoonhoven, C. B. (1990). Organizational Growth: Linking Founding

Team, Strategy, Environment, and Growth Among U.S. Semiconductor Ventures, 1978-

1988. Administrative Science Quarterly, 35(3), 504-529.

Fiet, J. O. (1995). RISK AVOIDANCE STRATEGIES IN VENTURE CAPITAL MARKETS.

Journal of Management Studies, 32(4), 551-574.

Haleblian, J., & Finkelstein, S. (1993). Top Management Team Size, CEO Dominance, and Firm

Performance: The Moderating Roles of Environmental Turbulence and Discretion. The

Academy of Management Journal, 36(4), 844-863.


49

Hellmann, T., & Puri, M. (2002). Venture Capital and the Professionalization of Start-Up Firms:

Empirical Evidence. The Journal of Finance, 57(1), 169-197.

Hornuf, L., & Schwienbacher, A. (2015). Funding Dynamics in Crowdinvesting. Available at

SSRN: http://ssrn.com/abstract=2612998.

Hsu, D. K., Haynie, J. M., Simmons, S. A., & McKelvie, A. (2014). What matters, matters

differently: a conjoint analysis of the decision policies of angel and venture capital

investors. Venture Capital, 16(1), 1-25.

Huyghebaert, N., & Van de Gucht, L. M. (2007). The Determinants of Financial Structure: New

Insights from Business Start-ups. European Financial Management, 13(1), 101-133.

Kuppuswamy, V., & Bayus, B. L. (2014). Crowdfunding creative ideas: The dynamics of project

backers in Kickstarter. UNC Kenan-Flagler Research Paper(2013-15).

Landström, H. (1998). Informal investors as entrepreneurs: Decision-making criteria used by

informal investors in their assessment of new investment proposals. Technovation, 18(5),

321-333.

Leland, H. E., & Pyle, D. H. (1977). Informational asymmetries, financial structure, and financial

intermediation. The Journal of Finance, 32(2), 371-387.

Lukkarinen, A., Teich, J. E., Wallenius, H., & Wallenius, J. (2016). Success drivers of online

equity crowdfunding campaigns. Decision Support Systems, 87, 26-26-38.

Ly, P., & Mason, G. (2012). Competition Between Microfinance NGOs: Evidence from Kiva.

World Development, 40(3), 643-655.

Macmillan, I. C., Siegel, R., & Narasimha, P. N. S. (1985). Criteria used by venture capitalists to

evaluate new venture proposals. Journal of Business Venturing, 1(1), 119-128.

Meer, J. (2014). Effects of the price of charitable giving: Evidence from an online crowdfunding

platform. Journal of Economic Behavior & Organization, 103, 113-124.

Megginson, W. L., & Weiss, K. A. (1991). Venture Capitalist Certification in Initial Public

Offerings. Journal of Finance, 46(3), 879-903.

Mitteness, C., Sudek, R., & Cardon, M. S. (2012). Angel investor characteristics that determine

whether perceived passion leads to higher evaluations of funding potential. Journal of

Business Venturing, 27(5), 592-606.

Mollick, E. (2014). The dynamics of crowdfunding: An exploratory study. Journal of Business

Venturing, 29(1), 1-16.

Mollick, E. R., & Kuppuswamy, V. (2014). After the campaign: Outcomes of crowdfunding.

UNC Kenan-Flagler Research Paper(2376997).

Nofsinger, J. R., & Wang, W. (2011). Determinants of start-up firm external financing worldwide.

Journal of Banking & Finance, 35(9), 2282-2294.

OECD. (2014). Financing SMEs and Entrepreneurs - An OECD Scoreboard: OECD Publishing.

Piva, E., & Rossi-Lamastra, C. (2018). Human capital signals and entrepreneurs’ success in equity

crowdfunding. Small Business Economics, 51(3), 667-667 - 686.

Puri, M., & Zarutskie, R. (2012). On the life cycle dynamics of venture‐capital‐and non‐venture‐

capital‐financed firms. The Journal of Finance, 67(6), 2247-2293.

Rysman, M. (2009). The Economics of Two-Sided Markets. Journal of Economic Perspectives,

23(3), 125-143.

Shane, S., & Cable, D. (2002). Network Ties, Reputation, and the Financing of New Ventures.

Management Science, 48(3), 364-381.

Shane, S. A. (2003). A general theory of entrepreneurship: The individual-opportunity nexus:

Edward Elgar Publishing.

Tyebjee, T. T., & Bruno, A. V. (1984). A model of venture capitalist investment activity.

Management Science, 30(9), 1051-1066.

Vismara, S. (2016). Equity retention and social network theory in equity crowdfunding. Small

Business Economics, 46(4), 579-590.

Vismara, S. (2018). Information Cascades Among Investors in Equity Crowdfunding.

Entrepreneurship: Theory and Practice, 42(3), 467–497.

Vulkan, N., Åstebro, T., & Sierra, M. F. (2016). Equity crowdfunding: A new phenomena.

Journal of Business Venturing Insights, 5, 37-49.


50

WorldBank. (2013). Crowdfunding’s Potential for the Developing World. 2013. Retrieved from

Washington, DC:

Zhang, J., & Liu, P. (2012). Rational Herding in Microloan Markets. Management Science, 58(5),

892-912.

Ziegler, T., Shneor, R., Garvey, K., Wenzlaff, K., Yerolemou, N., Rui, H., & Zhang, B. (2018).

Expanding Horizons: The 3rd European Alternative Finance Industry Report.


drivers of fundraising success in equity...

Documents