DUQUESNE UNIVERSITY
SCHOOL OF LAW LEGAL STUDIES RESEARCH PAPER SERIES
Female entrepreneurs and equity crowd — unding in the US: Receiving less when asking — or more Seth C. Oranburg Assistant Pro — essor o — Law and Mark Geiger Assistant Pro — essor o — Business 2018
Duquesne University School o --- Law Research Paper
No. 2018-20
Originally published in: JOURNAL OF BUSINESS VENTURING INSIGHTS
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 1 o — 23
Female entrepreneurs and equity crowd --- unding in the US:
Receiving less when asking --- or more
Mark Geiger a,*
Seth C. Oranburg b,*
Duquesne University
Note: When citing this research, please re
erence the published version o — the manuscript.
Re --- erence --- or this research (APA --- ormat):
Geiger, M., & Oranburg, S. C. (2018). Female entrepreneurs and equity crowd
unding in the US: Receiving less when asking — or more. Journal o — Business Venturing Insights, 10. https://doi.org/10.1016/j.jbvi.2018.e00099
a Duquesne University, Palumbo Donahue School o — Business, 600 Forbes Avenue, Pittsburgh, PA, 15282, United States. b Duquesne University, School o — Law, 600 Forbes Avenue, Pittsburgh, PA, 15282, United States.
*Author ordered alphabetically; author contributed equally as a
irst author.
E-mail addresses: geigerm1@duq.edu (M. Geiger), oranburgs@duq.edu (S. C. Oranburg).
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 2 o — 23
Female entrepreneurs and equity crowd --- unding in the US:
Receiving less when asking --- or more
ABSTRACT
In this paper, we explore the relationship between gender and
unding raised
through equity crowd
unding. Using data collected — rom the population o — US
equity crowd
unding campaigns, we — ind that campaigns receive signi — icantly less
unding when the primary signatory is — emale. Furthermore, we explore interactions
between gender and a campaign’s
unding target. The results suggest that
campaigns raise signi
icantly less — unding, as the target amount increases, when the
primary signatory is
emale. These results are the — irst to suggest a relationship
between gender and
unding among the population o — US equity crowd — unding
campaigns. Implications and
uture directions are discussed.
Keywords: gender; crowd
unding; equity crowd — unding.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 3 o — 23
INTRODUCTION
On April 5, 2012, the Jumpstart Our Business Startups (JOBS) Act was signed into law,
mandating that the Securities and Exchange Commission (SEC) create rules to allow equity
crowd
unding in the United States (US). When these rules became e —
ective on May 16, 2016,
equity crowd
unding in the US became a legitimate source o —
unding — or startups. While
startups have raised money via rewards-based crowd
unding on plat — orms like Kickstarter — or
over a decade, equity crowd
unding is di —
erent in that entrepreneurs raise money in exchange
or a pro — it interest in their company. While research has begun to examine equity crowd — unding
in Europe (Vulkan, Astebro, & Sierra, 2016) and Australia (Ahlers, Cumming, Günther, &
Schweizer, 2015), we still know very little about this phenomenon in the US.
Equity crowd --- unding in the US has generated excitement about its potential to make
capital available to more entrepreneurs. Some scholars believe that equity crowd
unding
provides a plat
orm — or underrepresented groups to raise — unding — or their companies. For
example, crowd
unding supporters argue that crowd — unding o — all kinds provides more accessible
unding — or women entrepreneurs (Mollick & Robb, 2016). In — act, research suggests that
rewards-based crowd
unding (e.g., Kickstarter) may be democratizing access to capital — or
women entrepreneurs (Marom, Robb, & Sade, 2016; Mollick & Robb, 2016). However, equity
crowd
unding is a di —
erent phenomenon, creating a big unknown concerning the democratizing
issue.
In this study, our goal is to examine the relationship between women entrepreneurs and
unding by exploring this relationship among the population o — US equity crowd — unding
campaigns. In the
orthcoming sections, we brie — ly review key concepts o — the study, provide an
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 4 o — 23
exploratory examination o
gender and — unding raised, and conclude with a discussion o — the
results.
LITERATURE AND CONCEPTS
Research suggests there is a strong interest in the relationship between gender and
unding (Greenberg & Mollick, 2017; Kanze, Huang, Conley, & Higgins, 2018; Mollick &
Robb, 2016). In
act, a paper by Mollick and Robb (2016) discussed the results o — working
papers concerning gender bias and rewards-based crowd
unding. For example, a working paper
by Marom et al. (2016) examined Kickstarter campaigns and
ound that women entrepreneurs
had higher success rates in meeting
unding targets, which was signi — icant regardless o — the
unding target. A working paper by Meek and Sullivan (c. — . Mollick & Robb, 2016), however,
examined Kickstarter campaigns and displayed no di
erence in — unding targets or — unding raised
between women and men. Given these limited, preliminary, and contrary results, there seems to
be a need
or more research on gender and crowd — unding. Moreover, prior studies’ results stem
rom the context o — rewards-based crowd — unding, a context that di —
ers — rom that o — equity
crowd
unding.
Crowd
unding
Crowd --- unding is a process whereby entrepreneurs seek --- unding --- rom a crowd o ---
contributors, o
ten through plat — orms on the Internet (e.g., Kickstarter). Unlike traditional
venture
inance, where an entrepreneur solicits large contributions — rom a — ew contributors, in
crowd
unding, the entrepreneur requests a large number o — relatively small contributions.
Crowd
unding can take di —
erent — orms, such as donations, rewards, pre-purchases, loans, or
equity investments (Oranburg, 2016a).
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 5 o — 23
Rewards-based crowd --- unding. The most --- amiliar --- orm o --- crowd --- unding is rewards-
based, where a
undraiser o —
ers something in return — or a contribution to a project. For example,
ilmmaker Matt Porter — ield rewards — unders who contribute $1000 by tattooing their initials in
his arm; but the law prohibits such
undraisers — rom o —
ering — inancial returns such as a percent
o
pro — its — rom sales o — the — ilm. Relatively speaking, rewards-based plat — orms like Kickstarter
o
er a donative-type o — crowd — unding, whereby the — under may also receive something tangible
in return
or the contribution (e.g., a sample o — the entrepreneur’s product). Evidence suggests
that donors o
rewards-based crowd — unding are primarily motived by intrinsic — actors. For
example, Gerber (2012)
ound that even when a reward is provided, — unders contribute to
crowd
unding campaigns to satis — y intrinsic motivations. Additionally, research by Ordanini,
Miceli, Pizzetti, and Parasuraman (2001) suggests that crowd
unding donors value their role as
co-creators, and they are primarily driven by intrinsic motivation.
Equity crowd --- unding. Equity crowd --- unding is the process whereby entrepreneurs seek
unding — rom a crowd — or an equity stake in their business. In contrast to rewards-based
crowd
unding, — unders o — equity crowd — unding are investors. As such, raising capital through
equity crowd
unding may di —
er — rom rewards-based crowd — unding in substantive ways. For
example, whereas donors o
rewards-based campaigns are mostly driven by intrinsic — actors,
investors o
equity crowd — unding campaigns may be primarily motivated by pro — its. In — act,
research by Cholakova and Clarlysse (2015: 147) showed that “the decision to invest in equity
[crowd
unding] was positively predicted only by — inancial return motivations.” I — the motivations
o
investors — or equity crowd — unding campaigns are mostly driven by pro — its, the patterns o —
unding with respect to gender may be similar to those o — more traditional — orms o — venture
inance.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 6 o — 23
Gender and Venture Finance
When seeking venture --- unding that involves elements o --- pro --- it and risk, discrimination
against
emale entrepreneurs is well documented. Indeed, a paper was dedicated to address the
question: “Why does the gender gap persist in obtaining new venture
inance?” (Leitch, 2018:
103). Research consistently shows that
emale entrepreneurs — ace greater challenges than their
male counterparts do with respect to venture
unding. For example, studies indicate that — emale
entrepreneurs receive less venture capital (Lins & Lutz, 2016), receive less
unding — rom banks
(Ste
ani & Vacca, 2013), pay more — or credit (Alesina, 2013), and are charged higher interest
rates on micro
inance loans (Dor — leitner, 2013). All o — these challenges women experience when
trying to raise venture capital
rom traditional sources has led some scholars to inquire, “Is
crowd
unding di —
erent?” (Barasinska & Sha — er, 2014)1.
Gender and crowd --- unding. While --- emale entrepreneurs are known to raise less than their
male counterparts
rom traditional investors, research examining Kickstarter has shown
preliminary evidence that
emale entrepreneurs may be more success — ul than males when using
rewards-based crowd
unding (Kanze et al., 2018). However, research suggests that the motives
o
unders may be di —
erent — or equity crowd — unding. As such, the ability o —
emale entrepreneurs
to raise
unds via equity crowd — unding may be quite di —
erent. In this research, using the
population o
equity crowd — unding campaigns in the US, we explored whether a relationship
exists between gender and
unding raised.
To examine the relationship between gender and equity crowd --- unding, we --- ocused on the
gender o
the primary signatory o — the campaign. The primary signatory is the individual who is
most responsible
or the company that — iled the equity crowd — unding campaign with the SEC.
1 Barasinkska and Sha — er (2014) examined loan-based crowd — unding in Germany and — ound no gender e —
ect on a borrower’s chance to raise — unds. They noted that lenders are protected — rom losses, which likely a —
ects lender behavior.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 7 o — 23
This individual is usually the key
ounder, president, or CEO o — the company. The primary
signatory is liable
or legalities (e.g., — raud or misconduct) and is generally the — ace o — the equity
crowd
unding campaign. As such, we provide the — ollowing research question:
Research Question: In the context o
equity crowd — unding campaigns, is the gender o — the primary signatory related to the amount o —
unding raised?
METHODS
Sample and Procedure To examine the relationship between gender and the amount o —
unding raised through
equity crowd
unding campaigns, we used the population o — US equity crowd — unding campaigns
through mid-March 2018. Using custom so
tware, we scraped publicly available in — ormation
available on the SEC’s Electronic Data Gathering, Analysis, and Retrieval (EDGAR) system.
Based on this data, 773 startups
iled an o —
ering statement — or an equity crowd — unding campaign
with the SEC. O
those that — iled an o —
ering statement, 276 — iled a progress update. O — those that
iled a progress update, 243 reported a — unding amount raised.
For gender coding, we --- ollowed established procedures in crowd --- unding research and
used the genderize.io tool to code primary signatories o
equity crowd — unding campaigns as
emale or male (Greenberg & Mollick, 2017; Marom et al., 2016). When the automated gender
tool returned an unknown
or a primary signatory’s gender, we manually searched the respective
equity crowd
unding campaign — or gender identi — iers (he/she). Moreover, we collected the
unding targets o — the campaigns and in — ormation on — irm characteristics that are reported to the
SEC (e.g., assets, revenue, debt, number o
employees, etc.). Funding targets and — irm
characteristics are relatively objective
actors that may in — luence a startup’s ability to raise
unding. In all, we collected data on a sample o — 243 equity crowd — unding campaigns (N = 243)
or our analyses.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 8 o — 23
Sample bias checks. To assess the extent to which the sample is representative o --- the
population we conducted sample bias checks. First, we used the raw genderize.io coding to
examine the gender ratio o
signatories o — the population and the sample. For the population o —
773, results returned 108
emale (~14%), 648 male (~84%), and 17 unknown (~2%). For the
sample o
243, results returned 40 — emale (~16%), 198 male (~81%), and 5 unknown (~2%).
Based on this observation, the gender representation o
the sample was similar to that o — the
population. Second, we conducted a logistic regression to compare the 530 campaigns not
reporting
unding amount raised (coded 0) with the 243 campaigns reporting — unding amount
raised (coded 1). We conducted this analysis
or the — unding targets and — irm characteristics.
Neither
unding targets nor — irm characteristics were signi — icant in predicting campaigns
reporting versus not reporting
unding amount raised.
Project category (supplemental data and analysis). In addition to --- irm characteristics,
the type o
business or product — or which a startup is seeking — unding may in — luence a
campaign’s ability to raise
unds. As such, in a supplemental — ashion, we reviewed the 243
campaigns to assess the extent to which the campaigns could be organized into categories. A
ter
reviewing the campaigns, we identi
ied some consistencies with respect to the markets in which
the startups were operating. Based on our review, we developed seven categories, which we
labeled: (a) Apps/eCommerce/Plat
orm, (b) Restaurant/Food/Beverage, (c) Entertainment/
Recreation, (d) Exercise/Health, (e) Clothing/Fashion/Cosmetic, (
) Hardware/Electronics, and
(g) Other. The number o
project categories by gender are reported at the bottom o — Table 1.
Analyses and Results
First, we conducted a preliminary examination comparing the means o ---
unding amount
raised,
unding targets o — the campaign, and — irm characteristics by gender. We conducted a one-
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 9 o — 23
way ANOVA to compare the means o
these variables between — emale and male signatories. As
displayed in Table 1,
emale signatories received signi — icantly less — unding — rom their equity
crowd
unding campaigns ($152,918 vs. $258,098, p < .05). However, no signi — icant di —
erences
between
emale and male signatories were — ound — or — unding targets or — irm characteristics.
------------------------------------------------
Insert Table 1 about here.
------------------------------------------------
Second, we examined the correlations. As displayed in Table 2, there was a signi --- icant
negative relationship between
emale signatories and — unding raised (r = -.14, p < .05). There
were also signi
icant correlations between — emale signatories and project categories:
Clothing/Fashion/Cosmetic (r = .28, p < .01); Entertainment/Recreation (r = -.15, p < .05);
Hardware/Electronics (r = -.13, p < .05). However, these categories showed no signi
icant
correlations with
unding raised. With respect to — unding targets, as expected, both the target
o
ering amount and maximum o —
ering amount had a positive correlation with — unding raised (r
= .23, p < .01; r = .41, p < .01, respectively). Moreover,
irm characteristics, such as a — irm’s
assets (r = .38, p < .01) and revenues (r = .29, p < .01), among others, had strong correlations
with
unding raised.
------------------------------------------------
Insert Table 2 about here.
------------------------------------------------
To examine our research question, “In the context o --- equity crowd --- unding campaigns, is
the gender o
the primary signatory related to the amount o —
unding raised?,” we conducted OLS
regression analyses. We examined the extent to which gender o
the primary signatory predicts
unding raised, beyond all other variables. Additionally, we explored how gender moderates the
relationship between a campaign’s
unding targets and the amount o —
unding raised. As
displayed in Model 1 o
Table 3, we — irst ran a regression with — irm characteristics and project
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 10 o — 23
categories predicting
unding raised. This model accounted — or 32% o — the total variance o —
unding raised (R2 = .32). In Model 2, we added the — unding targets, which accounted — or an
additional 12% o
the total variance o —
unding raised (R2 = .44).
To test the direct relationship between --- emale signatory and the amount o ---
unding raised,
we added the
emale variable to Model 3. The results showed a signi — icant negative relationship
between
emale signatory and the amount o —
unding raised (B = -.11, p < .05). Moreover, the
emale signatory variable alone accounted — or an additional 1% o — the total variance o —
unding
raised (R2 = .45). In Model 4, we entered the interaction between
unding targets and — emale
signatory predicting the amount o
unding raised. The results showed a signi — icant negative
relationship with
unding raised — or the interaction between target o —
ering amount and — emale
signatory (B = -.31, p < .01) and
or the interaction between maximum o —
ering amount and
emale signatory (B = -.20, p < .01). These interactions accounted — or an additional 5% o — the
total variance o
unding raised (R2 = .50).
------------------------------------------------
Insert Table 3 about here.
------------------------------------------------
To examine the signi --- icant interactions, we --- irst explored the magnitude o --- the e ---
ect o —
target o
ering amount on — unding raised as a — unction o — gender by plotting a two-way
interaction. By exploring the signi
icant interaction in this manner, we improved our ability to
interpret the e
ects (Preacher, Curran, & Bauer, 2006). As displayed in Figure 1, campaigns
with a low target o
ering had little di —
erence in — unding raised when the primary signatory was
emale versus male. However, as the target o —
ering increased, campaigns raised signi — icantly
less
unding when the primary signatory was — emale. We also explored the interaction — or the
e
ect o — maximum o —
ering amount on — unding raised as a — unction o — gender. As displayed in
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 11 o — 23
Figure 1,
unding raised was relatively stable — or male signatories regardless o — maximum
o
ering amount. In contrast, campaigns received increasingly less as the maximum o —
ering
amount increased when the primary signatory was
emale.
------------------------------------------------
Insert Figure 1 about here.
------------------------------------------------
DISCUSSION
In this research, we explored the relationship between gender ( --- emale) o --- primary
signatories o
equity crowd — unding campaigns and the amount o —
unding raised. Prior research
suggests that crowd
unding o — all types might bene — it women (Mollick & Robb, 2016). Yet, we
ound signi — icant evidence that crowd — unding campaigns that have a — emale primary signatory
receive less
unding. This relationship exists even when controlling — or other — actors related to
the amount raised including the target amounts o
the campaign, — irm characteristics, and project
categories.
Moreover, we --- ound a signi --- icant interaction between a campaign’s o ---
ering amount and
gender (
emale) o — the primary signatory in predicting — unding raised. Treating gender as a
moderator o
the relationship between o —
ering amount and — unding raised, we — ound that — emale
signatories received increasingly less
unding than male signatories as the o —
ering amount
increased. Furthermore, we
ound that maximum o —
ering had little in — luence on — unding raised
or male signatories, whereas, as maximum o —
ering increased — or — emale signatories, the amount
o
unding raised decreased substantially. These — indings may have important implications — or
research and practice.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 12 o — 23
Implications
or Research
The results o --- this study may have several implications --- or research. First, to the best o ---
our knowledge, this research is the
irst to provide evidence o — a relationship between gender and
unding raised through US equity crowd — unding. These — indings support previous research that
suggests
emale entrepreneurs raise signi — icantly less — unding than their male counterparts
(Kanze et al., 2018). This is contradictory to the
indings o — research on the relationship between
gender and
unding in the context o — rewards-based crowd — unding (Marom et al., 2016; c — .
Mollick & Robb, 2016).
One explanation --- or the mixed results between those o --- rewards-based crowd --- unding and
this study may reside in a donation versus investment distinction. For example, women
entrepreneurs may receive support
rom other women when using a donative approach to raising
unds — or their venture. This argument is supported by research that used Kickstarter data to
show that campaigns
ounded by women had signi — icantly greater odds at success — ully raising
unds via rewards-based campaigns and theorized that this e —
ect is due to “activist choice
homophily” (Greenberg & Mollick, 2017). Additionally, a study by PwC (2017)
ound that
women are 32% more likely than men to raise money on Kickstarter, while a Pew Research
Center survey (Smith, 2016)
ound that women are 5% more likely than men to contribute to a
crowdsourced
undraising project. Thus, data and theory suggest that women are more likely
than men to succeed in donative crowd
unding.
Conversely, when it comes to investment --- unding, women entrepreneurs may be
disadvantaged at raising
unds compared to men. For example, while 17% o — startups are
women-led, and 7% o
venture capital partners are women, less than 3% o — venture capital
unding goes to women-led startups (First Round Capital, 2015). Additionally, a study using a
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 13 o — 23
sample o
startups raising capital through a venture capital — unding competition — ound that
emale entrepreneurs received signi — icantly less — unding (Kanze et al., 2018). As such, consistent
with investment research and the results presented in our study, activist choice homophily may
not explain equity crowd
unding behavior as well as it explains other — orms o — crowd — unding.
In general, our results suggest that equity crowd --- unding is more similar to venture capital
unding than to rewards-based crowd — unding concerning — unding raised as a — unction o — gender.
This perspective would support our argument that the context o
equity crowd — unding may di —
er
drastically
rom that o — rewards-based crowd — unding. As such, — indings stemming — rom research
on Kickstarter, and similar plat
orms, should not be generalized to the context o — equity
crowd
unding.
Implications
or Policy and Practice
Equity crowd --- unding was signed into law in order to democratize access to capital so a
more diverse range o
entrepreneurs could start up — irms (Oranburg, 2016b). However, our data
show that it is not having this democratizing e
ect, at least with respect to gender. These
indings raise critical policy questions about whether Congress needs to — ix the JOBS Act so it
ul — ills its intended purpose. Indeed, the Fix Crowd — unding Act (114th Congress H.R. 4855) was
introduced soon a
ter the SEC promulgated its complex — inal rules to make it simpler — or a wider
range o
entrepreneurs to use equity crowd — unding, but its — orm and passage has been the source
o
much debate.
Our data suggest that the crowd --- unding regulations may indeed need to change i --- equity
crowd
unding is to provide more equal access to capital, although some proposed changes such
as raising the total contribution limit
rom $1 million to $5 million may not primarily have a
democratizing impact (Oranburg, 2016c). Indeed, adjacent studies on gender and campaign
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 14 o — 23
inance show that the average size o — donations to — emale candidates is smaller than to male
candidates; thus, “By simply increasing the individual contribution limit
rom $1,000 to $2,000,
[the Bipartisan Campaign Re
orm Act o — 2002], in e —
ect, exacerbated the — emale candidates’
disadvantage in each o
the three a — orementioned — acets o — gender speci — ic — undraising” (Baker,
2006: 20). Our data likewise suggest that, in equity crowd
unding, women might raise more
when asking
or less. There — ore, to make equity crowd — unding more equitable, policymakers
should consider making it easier to solicit smaller individual donations
rom a larger number o —
people.
For practitioners who are seeking --- unding, our --- indings might help an entrepreneur
decide whether to pursue that
unding via rewards-based or equity crowd — unding or by seeking
venture capital investment. Based on preliminary data, women experience a higher likelihood o
raising
unds via a rewards campaign, a signi — icantly lower chance o — raising — unds via equity
crowd
unding, and a substantially lower likelihood o — raising — unds via venture capital than men
do. Meanwhile, our data also show that company characteristics such as higher assets and
revenue also increased the likelihood o
a success — ul campaign, regardless o — gender. There — ore,
a women entrepreneur might choose to maximize
undraising potential by engaging in a rewards-
based campaign
irst to generate revenue and assets; then, second, leveraging that success to
increase the likelihood o
success in a — ollow-on equity crowd — unding campaign. Theoretically,
completing a success
ul equity crowd — unding campaign would also improve the likelihood o —
success in venture-capital
undraising as a third step (Oranburg, 2016c).
Limitations and Future Directions
While the present research provides evidence o --- a relationship between gender and
unding raised through equity crowd — unding, limitations and — uture directions should be
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 15 o — 23
discussed. For example, this is an exploratory study on a sample o
243 equity crowd — unding
campaigns in the US. Over time, the number o
equity crowd — unding campaigns in the US will
increase, and the relationships
ound in our research may change. As such, our — indings should
be considered preliminary evidence. Follow-up studies will be needed concerning the
relationships examined in this study.
Furthermore, research may need to explore the underlying reasons --- or why campaigns
with
emale primary signatories receive less — unding. One avenue o — research may need to
examine the quality o
the campaigns. For example, using data collected — rom Kickstarter,
Mollick and Robb (2016)
ound that signals o — quality, such as, outside endorsements, evidence
o
prototypes, and past success in — luence — unding decisions. On the other hand, there is evidence
that men and women raise money di
erently in campaign — inance (Baker, 2006), public — inance
(Palmer, 1995), and personal
inance (Shin, 2015), so it is important to determine whether
crowd
unding and other venture — inance laws create or maintain any structural bias against the
ways that women raise venture capital.
Another avenue may be to explore the characteristics o --- those who are investing in the
campaigns. For example, research by Greenberg and Mollick (2017) showed that women are
likely to support other women when making decisions. They argued that “activist choice
homophily, which entails support
or — ellow members o — disadvantaged groups,” (364) may
explain di
erences in — unding based on demographics such as gender. That is, — emale-run
campaigns may seek and receive less
unding due to a limited pool o —
emale investors. On the
other hand, activist choice homophily may not apply when
unding decisions are an investment.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 16 o — 23
Conclusion
This paper provides empirical evidence that gender has an e ---
ect on the amount o —
unding raised through US equity crowd — unding. There is much to learn about equity
crowd
unding in the US and we hope this paper is help — ul — or — uture research.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 17 o — 23
REFERENCES
Ahlers, G. K., Cumming, D., Günther, C., & Schweizer, D. 2015. Signaling in equity
crowd --- unding. Entrepreneurship Theory and Practice, 39(4): 955-980.
Baker, A. 2006. Reexamining the gender implications o
campaign — inance re — orm: How higher
ceilings on individual donations disproportionately impact --- emale candidates, The
Modern American, 2: 18-23.
Barasinski, N., Scha
er, D., 2014. Is crowd — unding di —
erent? Evidence on the relation between
gender and --- unding success --- rom a german peer-to-peer lending plat --- orm, German
Economic Review. 15(4): 436-452.
Cholakova, M., Clarysse, B. 2014. Does the possibility to make equity investments in
crowd --- unding projects crowd out reward-based investments?, Entrepreneurship Theory
and Practice, 39(1): 145-172.
Dor
leitner, G., Leidl, M., Priberny, C., von Mosch, J. 2013. What Determines Micro-Credit
Interest Rates?, Applied Financial Economics, 23: 1579-1597.
First Round Capital. 2015. 10 year project. <http://10years.
irstround.com/>
Gerber, E., Hui, J., Kuo, P. 2012. Crowd
unding: Why people are motivated to post and — und
projects on crowd --- unding plat --- orms, Proceedings o --- the International Workshop on
Design, In --- luence, and Social Technologies: Techniques, Impacts and Ethics.
Greenberg, J., & Mollick, E. 2017. Activist choice homophily and the crowd
unding o —
emale
ounders. Administrative Science Quarterly, 62(2): 341-374.
Kanze, D., Huang, L., Conley, M. A., & Higgins, E. T. 2018. We ask men to win and women not
to lose: Closing the gender gap in startup --- unding. Academy o --- Management Journal,
61(2): 586-614.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 18 o — 23
Leitch, C. 2018. Women entrepreneurs’
inancing revisited: taking stock and looking — orward,
Venture Capital, 20(2): 103-114.
Lins, E., Lutz, E. 2016. Bridging the gender
unding gap: do — emale entrepreneurs have equal
access to venture capital? International Journal o --- Entrepreneurship and Small Business,
27(2-3): 347-365.
Marom, D., Robb, A., Sade, O. 2016. Gender dynamics in crowd
unding (Kickstarter): Evidence
on entrepreneurs, investors, deals and taste based discrimination. SSRN Working Paper.
Available at: http://dx.doi.org/10.2139/ssrn.2442954
Meek, W. R., & Sullivan, D. M. Kickstarting new businesses: exploring the dynamics o
gender
and crowd --- unding. In Diana International Research Con --- erence, Wellesley,
Massachusetts. Working paper.
Mollick, E., & Robb, A. 2016. Democratizing innovation and capital access: The role o
crowd --- unding. Cali --- ornia Management Review, 58(2): 72-87.
Ordanini, A., Miceli, L., Pizzetti, M., Parasuraman, A. 2011. Crowd-
unding: trans — orming
customers into investors through innovative service. Journal o --- Service Management,
22(4): 443-470.
Oranburg, S. C. 2016a. Start-up
inancing. In F. Pacheco-Torgal, E. Rasmussen, C-G. Graqvist,
V. Ivanov, A. Kaklauskas, & S. Makonin (eds.), Startup creation: The smart eco-e ---
icient
built environment: 57-70. Woodhead Publishing: Cambridge, United Kingdom.
Oranburg, S. C. 2016b. Democratizing startup. Rutgers University Law Review, 68: 1013-1066.
Oranburg, S. C. 2016c. Bridge
unding: Crowd — unding and the market — or entrepreneurial
inance. Cornell Journal o — Law & Public Policy, 25(2): 397-452.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 19 o — 23
Palmer, I. 1995. Public
inance — rom a gender perspective, World Development, 23(11): 1981-
1986.
Preacher, K. J., Curran, P. J., & Bauer, D. J. 2006. Computational tools
or probing interactions
in multiple linear regression, multilevel modeling, and latent curve analysis. Journal o ---
Educational and Behavioral Statistics, 31(4): 437-448.
PwC. 2017. Women unbound, unleashing
emale entrepreneurial potential, <https://www.pwc.
com/gx/en/diversity-inclusion/assets/women-unbound.pd --- >
Shin, L. 2015. New survey reveals how men and women bank di
erently, Forbes,
<https://www. --- orbes.com/sites/laurashin/2015/03/25/new-survey-reveals-how-men-and-
women-bank-di ---
erently/#e7c77b613 — c3>
Smith, A. 2016. Shared, collaborate, and on demand: The new digital economy. Pew Research
Center, <http://assets.pewresearch.org/wp-content/uploads/sites/14/2016/05/
PI_2016.05.19_Sharing-Economy_FINAL.pd --- >
Vulkan, N., Åstebro, T., & Sierra, M. F. 2016. Equity crowd
unding: A new phenomena. Journal
o --- Business Venturing Insights, 5: 37-49.
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521
Page 20 o — 23
TABLE 1 Firm and Funding Characteristics by Gendera Variable Female (41)c Male (202)c F-statisticb Funding Outcome 1. Funding Raised ($) 152,918 258,098 5.07* Funding Targets 2. Target O —
ering ($) 78,415 71,373 0.27 ns 3. Maximum O —
ering ($) 578,823 639,909 0.82 ns Firm Characteristics 4. Firm Age (days) 1,308 1,326 0.00 ns 5. Number o — Employees 4 7 1.50 ns 6. Assets ($) 269,595 382,882 0.39 ns 7. Cash & Cash Equivalents ($) 70,775 92,670 0.18 ns 8. Accounts Receivable ($) 10,270 22,986 0.60 ns 9. Short-term Debt ($) 118,747 142,821 0.17 ns
- Long-term Debt ($) 266,084 224,076 0.33 ns
- Revenue/Sales ($) 338,230 410,236 0.04 ns
-
Cost o
Goods Sold ($) 154,345 208,582 0.04 ns
- Taxes Paid ($) 5,719 2,249 1.53 ns
- Net Income ($) -169,249 -256,864 0.46 ns Project Category
-
Apps/eCommerce/Plat
orm 18 64 -
- Restaurant/Food/Beverage 5 47 -
- Entertainment/Recreation 1 34 -
- Exercise/Health 4 11 -
- Clothing/Fashion/Cosmetic 8 5 -
- Hardware/Electronics 0 19 -
- Other 5 22 - a Gender o — the primary signatory o — the equity crowd — unding campaign. b F-statistic o — one-way ANOVA test — or signi — icant di —
erence between Female and Male. c Means — or each variable are reported; total number o — campaigns are reported in parentheses. ns p > .10
-
p < .05
Electroniccopy Electronic copyavailable available at: at: https://ssrn.com/abstract=3263521 https://ssrn.com/abstract=3263521Page 21 o — 23
TABLE 2 Correlations 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 1 Female 2 Funding Raised -.14 3 Target O —
ering .02 .23 4 Maximum O —
ering -.06 .41 .13 5 Firm Age -.01 .04 -.10 .12 6 Number o — Employees -.08 .33 -.02 .16 .36 7 Assets -.05 .38 .00 .19 .33 .48 8 Cash & Cash Equivalents -.04 .32 .03 .20 .20 .32 .55 9 Accounts Receivable -.06 .15 -.01 .18 .32 .26 .51 .48 10 Short-term Debt -.02 .19 -.01 .24 .43 .43 .38 .22 .40 11 Long-term Debt .03 .20 .00 .14 .25 .35 .49 .19 .18 .30 12 Revenue/Sales -.02 .29 .01 .22 .44 .67 .61 .48 .68 .66 .29 13 Cost o — Goods Sold -.02 .14 -.02 .16 .39 .52 .48 .38 .62 .75 .19 .87 14 Taxes Paid .07 .21 .02 .04 .20 .42 .24 .12 .12 .15 .16 .40 .22 15 Net Income .05 -.32 -.01 -.24 -.17 -.35 -.38 -.35 -.17 -.45 -.54 -.26 -.29 -.05 16 Apps/eCommerce/Plat — orm .10 -.04 -.19 .07 -.09 -.07 -.13 -.02 -.02 .04 -.03 -.06 .01 -.10 .01 17 Restaurant/Food/Beverage -.10 .03 .10 -.16 -.05 -.01 .00 -.04 .04 -.10 -.05 -.03 -.03 .11 .14 -.37 18 Entertainment/Recreation -.15 .12 -.01 .17 .05 .16 .08 .08 .05 -.01 -.02 .06 .04 -.05 -.05 -.29 -.21 19 Exercise/Health .07 -.04 .10 -.15 .03 -.03 .05 -.03 -.06 -.04 -.06 -.05 -.04 -.03 .04 -.18 -.13 -.11 20 Clothing/Fashion/Cosmetic .28 .03 .02 .05 .04 .02 .11 .06 .04 .11 .04 .16 .08 .18 -.05 -.17 -.12 -.10 -.06 21 Hardware/Electronics -.13 -.05 -.02 .01 .01 -.06 -.06 -.02 -.07 -.02 -.01 -.07 -.04 -.05 -.05 -.21 -.15 -.12 -.08 -.07 22 Other .02 -.06 .09 -.01 .09 -.01 .04 -.02 .00 .06 .16 .04 -.01 -.01 -.10 -.25 -.18 -.15 -.09 -.08 -.10 N = 241-243. Pairwise deletion. Correlations ≥ .13 signi — icant at p < .05 level; Correlations ≥ .17 signi — icant at p < .01 level.
Electronic
Electronic copy available copy available at: https://ssrn.com/abstract=3263521
at: https://ssrn.com/abstract=3263521
Page 22 o — 23
TABLE 3 Regression Analyses: Dependent Variable = Funding Raised Model 1 Model 2 Model 3 Model 4 Firm Characteristics Firm Age -.13* -.11 -.10 -.09 Number o — Employees .01 .05 .04 .01 Assets .24** .24** .24** .23** Cash & Cash Equivalents .10 .06 .06 .04 Accounts Receivable -.12 -.11 -.12 -.12 Short-term Debt .15 .07 .06 .03 Long-term Debt -.13 -.13 -.11 -.11 Revenue/Sales .65** .49** .48** .48** Cost o — Goods Sold -.64** -.48** -.47** -.43** Taxes Paid .05 .06 .07 .10 Net Income -.27** -.23** -.23** -.22** Project Categorya Apps/eCommerce/Plat — orm .16 .14 .14 .17* Restaurant/Food/Beverage .17 .14 .13 .10 Entertainment/Recreation .17* .11 .09 .09 Exercise/Health .05 .03 .03 .06 Clothing/Fashion/Cosmetic -.01 -.02 .01 .00 Hardware/Electronics .05 .03 .01 .01 Funding Targets Target O —
ering .22** .23** .42** Maximum O —
ering .27** .26** .31** Primary Signatory Gender Female -.11* .13 Interactions Target O —
ering X Female -.31** Maximum O —
ering X Female -.20*
Model R2 .32 .44 .45 .50 Δ R — or step 2 .12** .01* .05** N = 241-243. Listwise deletion. Standardized coe —
icients are reported. Δ R2 — or step indicates change — rom preceding model. a Other is the excluded category.
-
p < .05 ** p < .01
Electroniccopy Electronic copyavailable available at: at: https://ssrn.com/abstract=3263521 https://ssrn.com/abstract=3263521Page 23 o — 23
FIGURE 1 E —
ect o — Target O —
ering and Maximum O —
ering on Funding Raised as a — unction o — Gender
Male
Female
Funding Raised
$10,000 $1,000,000
Target O ---
ering
Male
Female
Funding Raised
$10,000 $1,000,000
Maximum O ---
ering
Electroniccopy
Electronic copyavailable
available at:
at: https://ssrn.com/abstract=3263521
https://ssrn.com/abstract=3263521