Skip to content
          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

  1. Long-term Debt ($) 266,084 224,076 0.33 ns
  2. Revenue/Sales ($) 338,230 410,236 0.04 ns
  3. Cost o

    Goods Sold ($) 154,345 208,582 0.04 ns

  4. Taxes Paid ($) 5,719 2,249 1.53 ns
  5. Net Income ($) -169,249 -256,864 0.46 ns Project Category
  6. Apps/eCommerce/Plat

    orm 18 64 -

  7. Restaurant/Food/Beverage 5 47 -
  8. Entertainment/Recreation 1 34 -
  9. Exercise/Health 4 11 -
  10. Clothing/Fashion/Cosmetic 8 5 -
  11. Hardware/Electronics 0 19 -
  12. 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

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.

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