Erica Broadus, Ph.D. Joseph Cordes

Transcription

1 "Founding Social Enterprises in the U.S.: Points ofentry, and Barriers to Entry With a Special Focus on the Relative Roles ofmale and Female Social Entrepreneurs" Erica Broadus, Ph.D. Trachtenberg School of Public Policy and Public Administration The George Washington University Joseph Cordes Professor of Economics, Public Policy and Public Administration, and International Affairs Trachtenberg School of Public Policy and Public Administration The George Washington University Preliminary Draft

2 Social are businesses that intentionally pursue social value creation through earned income strategies. A growing body of research suggests that women are prone toward social enterprise due to their desire to make contributions to society; propensity toward altruism, care, and protection of disadvantaged groups; and attraction to mission driven initiatives. These findings align with empirical research that suggests that women are motivated toward, conduct, and measure business success differently than men with emphasis on social impact rather than profit alone. In the U.S., women owned commercial businesses have historically lagged men s in key economic indicators: ownership, revenue, and size. Deeper inquiry has revealed that issues related to gender not just biological sex help explain these disparities. The theories of entrepreneurial expectancy and social learning suggest that women owned businesses can perform as well as men s, but external feedback from people, personal experiences, and external forces (e.g., media, society, and industry demographics) undermine women s confidence and hinder them from achieving financial goals. This study is a first step toward understanding if the same patterns hold for for profit social. Our paper attempts to address the several questions about the participation of women entrepreneurs in the growing social enterprise sector. (1) Are there different patterns of entry and ownership male and female social entrepreneurs? (2) How,if at all, do these patterns compare with those in the more traditional for-profit sector? (3) Does the financial performance of female and male social enterprise start-ups differ? (4) Do female social entrepreneurs set different goals for their startups than do their male counterparts? To address these questions we proceed as follows. First, we briefly summarize analyses of why from the vantage point of both economics, and also of feminist theory women entrepreneurs are more likely to face challenges than male entrepreneurs. We then present tabulations from responses by female and 1

3 male social entrepreneurs to a recent survey of alumni of a social enterprise accelerator of social. Background: Female Vs. Male Entrepreneurs Issues related to gender help explain differences between the financial performances of womenand men-owned businesses in the commercial sector. Studies of female business ownership in economics, business, and finance have found: (1) that women are much less likely to be business owners than are men, and (2) to quote Fairlie and Robb (2008) female-owned businesses are less successful than male-owned businesses because they have less startup capital, and business human capital acquired through prior work experience in a similar business and prior work experience in family business. We also find some evidence that femaleowned businesses work fewer hours and may have different preferences for the goals of their business. Figures 1-3 provide some descriptive information about women-and men-owned firms in the United States The question to be taken up is whether similar patterns are observed in the case of women social entrepreneurs who seek to create and to sustain social. The data set that we use to examine this question is from the Global Accelerator Leadership Initiative (GALI The study uses survey data from the Global Accelerator Learning Initiative (GALI), a product of the Entrepreneurship Database Program (the Program) at Emory University in Atlanta, Georgia. The Program collected data from 13,495 worldwide that applied to accelerator programs between 2013 and The data includes demographics, legal status, operations, social motives, impact areas and beneficiaries, financial goals and performance, founders/owners backgrounds, and more. The list below is a description of the full dataset. A list of accelerators that participated in the data collection and the survey instrument are included in Appendices 3 and 4, respectively. 2

4 Figure 1: Number of U.S. women- and men-owned firms, 2012 Millions Figure 2: Annual receipts of U.S. women- and men-owned firms, 2012 Trillions $10 $8 $6 $4 $2 $- Women owned Men owned 3

5 Figure 3: Top 5 industries by number of women- and men-owned firms. 4

6 Geography: The 13,495 organizations operate in 159 countries that span six continents. The most prevalent countries are the United States of America (2,888; 21%), Mexico (1,612; 12%), and India (1,255; 9%). The most prevalent continents are North America (5,315; 39%), Africa (3,698; 27%), and South America (2,334; 17%). See Appendix 5 for a full list. Enterprise Age20: Nearly three quarters (9,626; 74.2%) of the organizations were founded in the past five years. About one fifth (2,648; 20.4%) were founded six to 10 years ago. Another 3.2% (411) were founded years ago, 1.2% (150) were founded years ago, and 1.1% (142) were founded more than 20 years ago. A small percentage (518; 3.8%) were unknown. Number of founders/owners: Almost half (5,887; 44%) of the entities are founded/owned by three individuals. About one third (4,538; 34%) have two founders/owners, and less than one quarter (2,986; 22%) are solely owned. A small percentage, 0.1% (84) are not known. 20 As of June 2018, when quantitative data was analyzed. Sex of founders/owners 1 2: The entities are owned (or majority owned) as follows: 21% by women (2,854), 64% by men (8,591); and 13% are equally owned by women and men (1,736). An additional 2% (314) did not disclose the sex of their founders. Legal status: Among the, 80% (10,804) are for profits, 10% (1,364) are nonprofits, 6% (807) indicated other legal sector, 4% (502) are undecided, and 0.1% (18) did not state a legal sector. Social motive: Among the organizations, 87% (11,801) have a social motive; 13% do not have a social motive, and very few (0.1%; 8) were unknown. The organizations primary operational sectors and models 2 are listed in Tables 3.1 and 3.2, respectively, from high to low frequency: 1 This study distinguishes women and men owned social by the sex of the majority of founders/owners (e.g., 1 of 1, or 2 of 3), similar to the Survey of Business Owners, which defines women and men owned businesses as entities with majority (51% or more) female or male ownership, respectivel 2 Firms could select more than one operational model, thus percentages exceed

7 Table 3.1: Operational sectors (full data set) Primary sector (industry) Number of firms Percentage of firms Other 2, % Education 2, % Agriculture 1, % Health 1, % Info and comm. Tech 1, % Financial Services 1, % Energy % Environment % Tourism % Artisanal % Supply chain services % Water % Culture % Housing development % Infrastructure/facilities dev % Technical asst. services % Not stated % Table 3.2: Operational models (full dataset) Operational models Number of firms Percentage of firms Services 8, % Production/manufacturing 4, % Distribution 3, % Wholesale/retail 2, % Processing/packaging 1, % Financial services 1, % Unsure % 6

8 Table 3.3 provides the firms areas of social impact, from most prevalent to least 23. Table 3.3: Social impact areas (full data set) Number of Percentage of Areas of social impact firms firms Employment generation 3, % Income/productivity growth 2, % Community development 2, % Access to education 2, % Health improvement 2, % Equality and empowerment 2, % Access to information 1, % Agriculture productivity 1, % Capacity building 1, % Other 1, % Access to financial services 1, % Food security 1, % Pollution prevention and waste management 1, % Support for women and girls 1, % Access to energy % Sustainable land use % Sustainable energy % Generate funds for charitable giving % Energy and fuel efficiency % Disease specific prevention and mitigation % Access to clean water % Efficiency % 23 Affordable housing % Biodiversity conservation % High impact % Water resources management % Natural resources conservation % Conflict resolution % Human rights protection or expansion % 7

9 Table 3.4 describes the degree to which the target certain demographic groups, from most prevalent to least. Table 3.4: Beneficiaries (full data set) Beneficiaries Impacted Number of firms Percentage of firms Other 5, % Children and adolescents 1, % Women 1, % Minorities 1, % Disabled % Not stated 3, % Exclusion Criteria Our study focuses on U.S. based for profit social from a gender perspective. Therefore, 11,869 observations were excluded from the full dataset (described above) based on the following criteria: Enterprises that do not operate in the United States. The data contains the variable Country of Operations, which indicates each entity s country of main operations. Of the 13,495 initial observations, I dropped 10,607 that operate outside the U.S. This yielded 2,888 entities that operate in the U.S. Enterprises that are not socially motivated. The data contains the variable Has Social Motives Y/N, which indicates whether the entity has a social motive; 1 for yes, and 0 for 8

10 no. Excluded observations (those that reported 0 ) totaled 271. An additional 90 were not stated. This decreased the total to 2,527 U.S. based social. Enterprises that do not have a for profit legal status. The data contains the variable Legal Status, which indicates whether the entity operates as a for profit or another type. Exclusions included observations whose legal status was nonprofit (379), other (164), or undecided (83). This yielded 1,901 U.S. based for profit social. Enterprises that are not founded/owned by women or men. 3 The data contains three variables, Fndr 1 Gender, Fndr 2 Gender, and Fndr 3 Gender, which indicate the biological sex of up to three founders/owners of each entity. The variables are coded F for female and M for male. From these data, I created a new variable called Gender of Ownership that reports the majority sex of each enterprise s founders/owners 24. Two hundred thirty one entities equally owned by one man and one woman were excluded, as were 44 that did not provide ownership information. This yielded a final sample of 1,626 U.S. based for profit social founded/owned by women (433; 26.6%) or by men (1,193; 73.4%). Summary of the Quantitative Sample Among the 1,626 U.S. based for profit social that are founded/owned by women or men, most (75.9%) are founded/owned by more than one person. The range from one to 58 years of age, but most (75.8%) are one to five years old. The average age is about four and a half years. The entities operate in a variety of sectors; most prevalent are Education (18.6%), Health (17.8%), and Other (17.1%), and their most frequent operational models are 3 This method of determining the sex of founders/owners is used by the U.S. Census Survey of Business Owners and the Annual Survey of Entrepreneurs. 9

11 Analysis Research Question 1, the distribution of women and men owned social, is answered through descriptive statistics analyzed in Excel and Stata. They include simple statistics (e.g., mean, median, etc.), cross tabulations, and tests for statistically significant differences between social founded/owned by women and by men in four areas: demographics (ownership and age), operations (sectors and models), financial performance (revenue, profit margin, and size), and social impact (areas of impact, beneficiaries, and rates of impact measurement). Research Question 2, whether gender matters to the financial performance of social enterprise, is answered through quantitative and qualitative analysis. Following is an overview of the quantitative portion. Tables 3.5 to 3.8 list each hypothesis (described in Chapter 2), related subquestions, components of each sub question, applicable variables, and methods of analysis. 10

12 Table 3.5: Hypothesis 1, Sub question A, Components of Sub Question A, Variables, and Analysis H1: Women and men owned social operate in different industries. Sub Q A: To what extent do women and men owned social operate in different industries? Components of Sub Q A Variables Analysis A1. What operational sectors and models 25 are most prevalent women owned social? A2. What operational sectors and models are most prevalent menowned social? A3. To what extent do women and men owned social operate in different sectors? A4. To what extent do women and men owned social use different operational models? GALI data: 4 Gender of Ownership Sector of Operations Operational Model GALI data: Gender of Ownership Sector of Operations Operational Model GALI data: Gender of Ownership Sector of Operations GALI data: Gender of Ownership Operational Model Cross tabulation of operational sectors and models by sex of founders/owners. Report the 3 most frequent operational sectors and models for women owned. Cross tabulation of operational sectors and models by sex of founders/ owners. Report the 3 most frequent operational sectors and models for men owned. Report which sex has the greater level of participation in each operational sector. T test results for statistically significant differences. Report which sex has the greater level of participation in each operational model. T test results for statistically significant differences. 4 The GALI data contains two variables that signify industry : operational sector and operational model. Both variables are used in analysis because the U.S. Department of Labor classifies many of the data s operational sectors (e.g., education, health) and operational models (e.g., manufacturing, wholesale trade) as industries. 11

13 le 3.6: Hypothesis 2, Sub question B, Components of Sub Question B, Variables, and Analysis H2: Women owned social are less likely than men owned social to operate in high revenue industries. Sub Q B: To what extent do women and men owned social operate in high revenue industries? Components of Sub Q B Variables Analysis B1: How do women and men owned social top 3 industries compare to high revenue industries? GALI data: Gender of Ownership Sector of Operations Operational Model Average Annual Revenue SBO data 5 Industries by receipts Compare the top 3 women s and men s social enterprise operational sectors and models to the SBO data, by annual receipts. Alternative 1: Calculate and report the top 3 social enterprise operational sectors and models by low and high revenue 6. Then, compare the percentage of women and men owned social that operate in each of these high revenue sectors and models. Alternative 2: Report the 3 most prevalent operational sectors and models women and men owned. Then, calculate and report the percentage of high revenue firms contained within these sectors and models. 5 As describe in Chapter 1, SBO data is the 2012 U.S. Census Survey of Business Owners and Self employed Persons. SBO data informed two methods of comparison: The Industry Differences by Gender: Top Five Industries by Average Receipts 2012 report issued by the National Women s Business Council in March 2016, and the Data from the 2012 Survey of Business Owners: Top 20 Industries by Sales report issued by the U.S. Small Business Administration s Office of Advocacy on May 31, $1 million plus is the highest revenue category tracked by the Census Survey of Business Owners. Thus, I defined high revenue as average annual revenue of at least $1 million, and low revenue as average annual revenue less than $1 million. 12

14 B2: To what extent are women owned social less likely than men owned social to operate in a high revenue sector? GALI data: Gender 3 high revenue 7 operational sectors Team Average Job Tenure High Ed Bachelor s degree or higher Team Any type of start up Y/N Team FP Experience Y/N Team NP Experience Y/N Logit regression to determine the odds that operating in a high revenue sector depends on gender (being woman owned), while controlling for variables related to founders/owners competency. B3: To what extent are women owned social less likely than men owned social to use a highrevenue operational model? GALI data: Gender 3 high revenue models 8 Team Average Job Tenure High Ed Bachelors or Higher Team: Any type of start up Y/N Team: FP Experience Y/N Team NP Experience Y/N Logit regression to determine the odds that using a high revenue operational model depends on gender (being woman owned), while controlling for variables related to founders/owners competency. 13

15 Table 3.7: Hypothesis 3, Sub question C, Components of Sub Question C, Variables, and Analysis H3: Women owned social have lower financial expectations and preferences than those owned by men. Sub Q C: To what extent do the financial expectations and preferences of social differ by sex of founders/owners? Components of Sub Q C Variables Analysis C1. To what extent do women and men owned social have profit margin expectations? C2. To what extent do the profit margin preferences of women and menowned social differ? C3. To what extent are women owned social less likely than men owned social to prefer high profit margin 9? GALI data: Gender of Ownership Has profit margin expectation Y/N GALI data: Gender of Ownership Profit margin preference Range GALI data: Gender Profit Margin Preference > 20 Team Average Job Tenure Team Any type of start up Y/N Team FP Experience Y/N Team NP Experience Y/N Cross tabulation of Has profit margin expectation by sex of founders/owners. T test for significance. Cross tabulation of Profit margin preferences by sex of founders/owners. T test for significance. Logit regression to determine the odds that preferring a high profit margin depends on gender (being womanowned), while controlling for variables related to founders/owners competency. 9 2 GALI survey respondents selected from five choices: 0 5%, 6 10%, 11 15%, 16 20%, and more than 20%, as described in Table Thus, I defined high profit margin preference as the highest category: more than 20%. 14

16 Table 3.8: Hypothesis 4, Sub question D, Components of Sub Question D, Variables, and Analysis H4: The financial performance (revenue, profits, and size) of women owned social is lower than men s. Sub Q D: To what extent does the financial performance of social differ by sex of founders/owners? Components of Sub Q D Variables Analysis D1. To what extent does the revenue of women owned social differ from men s? D2. To what extent do the profits of women owned social differ from men s? D3. To what extent does the size (number of employees) of women owned social differ from men s? GALI data: Gender of Ownership Average annual revenue GALI data: Gender of Ownership Profit margin in Year t 1 GALI data: Gender of Ownership Full time employees in Year t 1 Part time employees in Year t 1 Cross tabulation of average annual revenue by sex of founders/owners. T tests for statistical significance. Cross tabulation of profit margin in the previous year by sex of founders/owners. T tests for statistical significance. Cross tabulation of full time and part time employees by sex of founders/owners. T tests for statistical significance. Dependent Variables Sub question B analyzes the odds that operating in a high revenue industry (dependent variable) depends on sex of the founders/owners. As mentioned, scholars report that women are more likely to incorporate in industries that are less profitable (Allen and Minniti 2007), slower growing and more competitive than industries favored by men (Hisrich and Brush 1983, Miskin and Rose 1990). More recently, the U.S. Small Business Administration (2017) reported that the top 20 commercial industries by sales, women own a lower percentage of firms than men do in every industry. Conversely, 12 of the bottom 20 industries by sales are owned by more women 15

17 than by men. High revenue varies by industry. However, $1 million plus is the highest category tracked by the Census Survey of Business Owners 29. Thus, I defined high revenue as average annual revenue of at least $1 million, and categorized the as low revenue (below $1 million average annual revenue) and high revenue (at/above $1 million average annual revenue) for each operational sector and model, as described in Tables 3.9 and 3.10, respectively. Table 3.9: Operational sectors by high and low average annual revenue Low Revenue High Revenue Proportion of High Firms Firms Revenue Firms Agriculture % Artisanal % Culture % Education % Energy % Environment % Financial Services % Health % Housing development % Info and comm. Tech % Infrastructure/facilities dev % Other % Supply chain services % Technical assistance services 9 0 0% Tourism % Table 3.10: Operational models by high and low average annual revenue Low Revenue High Revenue Proportion of High Firms Firms Revenue Firms Production/manufacturing % Processing/packaging % Distribution % Wholesale/Retail % Services 1, % Financial Services % 16

18 Sub question C analyzes the odds that having a high profit margin preference (dependent variable) depends on the sex of founders/owners. Manolova et al. (2012) suggest that growth intentions and desired outcomes from the entrepreneurial process differ between women and men because women do not focus solely on financial performance, but seek to fulfill many goals (e.g., self realization, recognition) simultaneously. The GALI data contains the variable Profit Margin Preference Range. The corresponding survey question asks: What annual profit margin 30 would you be happy achieving on average? Respondents selected from five choices: 0 5%, 6 10%, 11 15%, 16 20%, and more than 20%, as described in Table Thus, I defined high profit margin preference as the highest category: More than 20%. Table 3.11: Profit Margin Preference Women owned Men owned All social N % at each level % women N % at each level % men N % all social 0 5% % 0.3% 0 0.0% 0.0% 1 0.1% 6 10%* % 1.9% 0 0.0% 0.0% 7 0.7% 11 15%* % 21.4% % 9.7% % 16 20%* 2 1.2% 0.6% % 26.6% % More than % 75.8% % 63.7% % 20%* * Statistically significant difference at the.05 level between women and men owned social. 17

19 Independent Variable The biological sex of founders/owners (labeled Gender ) is the independent variable of interest in this study. Previous research (U.S. Small Business Administration 2017, National Women s Business Council 2016, U.S. Census Survey of Business Owners 2012) indicates that women and men owned businesses in the commercial sector have different financial performances (i.e., prevalence, revenue, size). This study investigates whether similar patterns hold for social enterprise. Thus, gender was included as an independent variable in the Logit regression models of sub questions B and C. Control Variables Scholars suggest that founders/owners competency, such as knowledge, social roles, and skills, influence venture creation, growth, and success (Mitchelmore & Rowley 2010, Bird 1995). Thus, founders/owners who are more educated and have more experience (e.g., professional, startup) may achieve greater financial success than those who are less educated and experienced. Control variables that represent founders/owners competency are: Education: The GALI data reported the highest level of education completed by each founder/owner. Enterprises can have 1, 2, or 3 founders/owners. Thus, I created a new dummy variable titled, High Ed Bachelors or Higher that indicated whether at least one founder/owner of each enterprise had a bachelor s degree or higher. Among all social, more than half (53.9%) are led by at least one social entrepreneur who possesses a bachelor s degree. The same is true women owned (53.8%) and menowned (54.0%). And, all social, about one third (31.9%) are led by at least one social entrepreneur who possesses a master s degree. The same is true 18

20 women owned (34.2%) and men owned (31.0%). Table 3.12: Highest level of education completed by one or more founder/owner of each social enterprise Women owned Men owned All social N % of % N % of % N % all each response womeneach response mensocial owned owned None** 0 0.0% 0.0% % 0.8% 9 0.6% Primary School % 0.7% % 0.6% % Middle School % 1.8% % 2.3% % 9th Grade % 1.6% % 2.8% % High School % 9.0% % 11.1% % Associates Degree** % 3.9% % 2.3% % Technical/Vocational % 15.9% % 13.3% % Bachelors' Degree % 53.8% % 54.0% % Some Grad School % 9.7% % 12.4% % Master's Degree % 34.2% % 31.0% % PhD* % 2.8% % 6.9% % 19

21 Previous start up experience: The GALI data reports whether each founder/owner has started any previous. I created a new dummy variable titled, Team: Any type of start up; Y/N that indicates whether at least one founder/owner of each enterprise has previously started any type of entity (for profit, nonprofit, or other). Among all social, nearly two thirds (62.8%) are founded/owned by at least one social entrepreneur with previous start up experience. The same is true men owned (65.0%) and greater than half women owned (56.6%). Table 3.13: Previous Start up Experience: Start ups of any type by at least one founder/owner of each social enterprise Women owned Men owned All social N % of each response % womenowned N % of each response % menowned N % all social Yes* % 56.6% % 65.0% 1, % No* % 43.4% % 35.0% % 433 1,193 1,626 * Statistically significant difference at the.05 level between women and men owned social. Length of work experience: The GALI data reports the length of work experience for all founders/owners. I summed these values for each enterprise and divided by the number of founders/owners to create a new variable titled, Team Average Job Tenure. The data indicates that more than half (59.7%) of all social are founded/owned by those who average one to 11 years of work experience. The same is true women owned (61.2%) and men owned (59.2%) entities. 20

22 Table 3.14: Length of Work Experience: Average work experience (in years) for each enterprise s founders/owners Women owned Men owned All social N % of each response % women N % of each response % men N % all social 0 <1 year % 15.9% % 16.2% % 1 <11 years % 61.2% % 59.2% % 11 < % 15.2% % 16.8% % Years 21 < % 4.2% % 3.9% % Years 31 < % 2.3% % 2.3% % Years 41+ years % 1.2% % 1.7% % 433 1,192 1,625 Type of work experience: The GALI data indicates the sectors in which founders/owners have worked professionally. I created two new dummy variables titled, Team FP Experience Y/N and Team NP Experience Y/N that indicate whether at least one founder/owner has worked in the for profit or nonprofit sector, respectively. Among all social, most (87.8%) are founded/owned by at least one social entrepreneur who has worked in the for profit sector. The same is true women owned (87.3%) and men owned (88.0%). Among all social, nonprofit is the second most frequent (38.3%) sector of work experience. The same is true women owned (40.9%) and men owned (37.4%). 21

23 Table 3.16: Work experience: Legal sector of one or more founder/owner of each social enterprise Women owned 40 Men owned 41 All social 42 N % of each type % women N % of each type % men N % all social owned owned For profit % 87.3% 1, % 88.0% 1, % Nonprofit % 40.9% % 37.4% % Government % 21.5% % 18.1% % Other % 12.0% % 10.9% % This section provides the study s quantitative analysis and findings in three parts. First, I present descriptive statistics to answer Research Question 1. Then, I address four hypotheses and related sub questions to answer the quantitative portion of Research Question 2. Last, I summarize the findings. Descriptive Statistics Research Question 1 asks: What is the distribution of women and men owned social? This question is answered through descriptive statistics analyzed in Excel and Stata. Analysis includes simple statistics (e.g., mean, median, etc.), cross tabulations, and tests for statistically significant differences between social founded/owned by women and by men. Analysis and findings of descriptive statistics are presented in four sections: demographics, operations, financial performance, and social impact. Demographics In the U.S., men own nearly three times (73.4%) as many social as women own (26.6%). Women owned social average 2.1 founders/owners and men owned average 2.4 founders/owners. Yet, more than one third (36.0%) of women s entities are solely owned, while less than one fifth (19.8%) of men s entities are solely owned. This difference is statistically 22

24 significant. Conversely, men owned social are run by two (24.4%) or three (55.8%) individuals more often than women owned entities (16.2% and 47.8%, respectively). There is a statistically significant difference between the rates at which women and men owned social have two or three founders/owners. 23

25 Table 4.1: Frequency of women and men owned social (by majority ownership) Women owned Men owned All social N 433 1,193 1,626 Percentage 26.6% 73.4% 100.0% Table 4.2: Number of Founders/Owners of each Social Enterprise Statistics Women owned Men owned All social Mean Median Min Max Table 4.3: Number of Founders/Owners of each Social Enterprise by levels N Women owned Men owned All social % at each N N level % womenowned % at each level % menowned % all social 1 founder/ % 36.0% % 19.8% % owner* 2 founders/ % 16.2% % 24.4% % owners* 3 founders/ % 47.8% % 55.8% % owners* 433 1,193 1,626 * Statistically significant difference at the.05 level between women and men owned social. 24

26 Social are typically young, averaging almost four and a half years for both groups. The majority (75.8%) of social are five years old or less. This is true women owned (77.0%) and men owned (75.3%). Among all observations, the youngest enterprise is one year old for women and men owned. The oldest, which is owned by men, is 58 years; women s oldest firm is 34 years. Table 4.4: Age of Social Enterprises Statistics Women owned Men owned All social Mean Median Min Max Table 4.5: Age of Social Enterprises Percentages by levels Women owned Men owned All social N % at each level % women owned N % at each level % menowned N % all social 1 5 years % 77.0% % 75.3% 1, % 6 10 years % 21.6% % 21.6% % % 0.7% % 2.2% % years* years 0 0.0% 0.0% % 0.4% 5 0.3% > 20 years % 0.7% % 0.4% 8 0.5% 421 1,164 1,585 * Statistically significant difference at the.05 level between women and men owned social. 25

27 Operations Education (18.6%), Health (17.8%), and Other (17.1%) are the most prevalent operational sectors all social. The same is true men owned (18.5% each for Health and Education, Other 16.3%). Among women owned, Other (19.4%) was most prevalent, followed by Education (18.9%) and Health (15.9%). Statistically significant differences (at the.05 level) between women and men owned social exist in Housing Development and Supply Chain Services, as well as (at the.10 level) in Information/Communication Technology and Tourism. No such differences exist in other operational sectors. Table 4.6: Operational Sectors Women owned Men owned All social N % of % N % of % N % of all each sector womeneach sector mensocial owned owned Agriculture % 9.9% % 8.6% % Artisanal % 1.2% % 0.9% % Culture % 1.2% % 1.8% % Education % 18.9% % 18.5% % Energy % 4.2% % 5.8% % Environment % 2.1% % 3.2% % Financial Services % 13.9% % 12.3% % Health % 15.9% % 18.5% % Housing % 2.1% % 0.8% % Development* Information and % 6.7% % 9.3% % Communication Technologies** Infrastructure/facilities % 1.2% % 1.1% % Development Supply Chain Services* % 2.5% % 1.1% % Technical Assistance % 0.7% % 0.5% 9 0.6% Services Tourism** 1 5.9% 0.2% % 1.3% % Other % 19.4% % 16.3% % 433 1,192 1,625 * Statistically significant difference at the.05 level between women and men owned social. ** Weak statistically significant difference at the.10 level between women and men owned social. 26

28 Services (66.4%) is the most prevalent operational model all social, including those founded/owned by women (60.7%) and by men (68.5%). Financial Services (12.8%) is the least prevalent operational model overall, and those founded/owned by women (7.2%) and by men (14.8%). There is a statistically significant difference (.05 level) between womenand men owned social use of many operational models: Processing/Packaging, Wholesale/Retail, Services, and Financial Services. No such differences exist in Production/Manufacturing or Distribution. Table 4.7: Operational Models Women owned Men owned All social N % of each model % womenowned N % of each model % menowned N % of all social 55 Production/Mfg % 31.2% % 32.6% % Processing/Pkg.* % 21.0% % 15.1% % Distribution % 25.9% % 26.2% % Wholesale/Retail* % 29.6% % 20.6% % Services* % 60.7% % 68.5% 1, % Financial Services* % 7.2% % 14.8% % * Statistically significant difference at the.05 level between women and men owned social. 27

29 Financial Performance Overall, the social average annual revenue ranges from $0 to $1.3 billion. The overall annual mean is $1.4 million. Women average almost $2.2 million annually and men average almost $1.2 million. Yet, the majority of both groups earn very little. Nearly three quarters (73.5%) earn less than $5,000 annually. The same is true women (74.1%) and men (73.3%). And less than one percent (0.9%) of all social earn $1 million or more per year. The same is true for women owned (0.9%) and men owned (0.8%) entities. Table 4.8: Average annual revenue Statistics Women owned Men owned All social Mean $2,194,483 $1,166,125 $1,460,608 Median $0 $12 $25 Min $0 $0 $0 Max $900,000,000 $1,314,971,181 $1,314,971,181 Table 4.9: Average annual revenue by levels Women owned Men owned All social N % of each response % women N % of each response % men N % all social $0 4, % 74.1% % 73.3% 1, % $5,000 9,999** % 7.2% % 5.0% % $10,000 24, % 5.3% % 7.0% % $25,000 49, % 3.2% % 3.2% % $50,000 99, % 2.1% % 2.7% % $100, , % 2.5% % 3.4% % $250, , % 0.9% % 1.5% % $500, , % 0.9% % 0.8% % $1 million % 0.9% % 0.8% % 421 1,164 1,585 ** Weak statistically significant difference at the.10 level between women and men owned social. The GALI survey data reported rates of previous year profit margin in increments of five percent, up to 20, and negative return on investment (ROI) 58. The most frequent rate of previous year profit margin was 0 5% women (37.5%) and men (31.8%), followed by 28

30 negative ROI. Nearly one third of both groups experienced losses; 30.5% women and 37.9% men. Table 4.10: Profit margin 59 in the previous year N Women owned Men owned All social % of each N N response % women owned % of each respons e % menowned % all social enterprise 0 5%** % 37.5% % 31.8% % 6 10% % 6.8% % 7.6% % 11 15% % 5.2% % 5.7% % 16 20% % 8.9% % 7.6% % More than % 11.1% % 9.4% % 20% Negative ROI* % 30.5% % 37.9% % ,220 * Statistically significant difference at the.05 level between women and men owned social. ** Weak statistically significant difference at the.10 level between women and men owned social. Last, the social number of full time employees ranges from zero to over 1,000, and part time staff ranges from zero to 25,000. Yet, nearly half of all social do not employee full or part time staff. Among employer firms, most employ 1 19 workers. Table 4.11: Size Number of full time employees Statistics Women owned Men owned All social Mean Median Min Max 1, ,090 29

31 Table 4.12: Size Number of full time employees by levels 62 0 employees (besides fndrs/owners) Women owned Men owned All social N % at each level % women N % at each level % men N % all social % 46.0% % 49.3% % 1 19 employees % 50.1% % 48.8% % % 3.2% % 1.6% % employees* % 0.7% % 0.3% 6 0.4% Employees 500+ employees 0 0.0% 0.0% % 0.1% 1 0.1% 433 1,193 1,626 * Statistically significant difference at the.05 level between women and men owned social. 30

32 Table 4.13: Size Number Women owned Men owned All social of part time employees Statistics 63 Mean Median Min Max ,000 25,000 Table 4.14: Size Number of part time employees by levels 0 employees (besides fndrs/owners)* Women owned Men owned All social N % at each Level % women N % at each level % men N % all social % 51.0% % 56.7% % 1 19 employees* % 47.3% % 40.3% % employees % 1.4% % 2.8% % % 0.0% 0 0.0% 0.0% 0 0.0% Employees 500+ employees % 0.2% % 0.2% 3 0.2% 433 1,193 1,626 * Statistically significant difference at the.05 level between women and men owned social. Social Impact 31

33 Overall, the three most prevalent social impact areas women and men owned social are Employment Generation (27.6%), Income/Productivity Growth (23.6%), and Community Development (18.6%). Least prevalent impact areas are Natural Resources/Biodiversity (0.9%), Affordable Housing (1.2%), and Water Resources Management (1.2%). Among womenowned social, the most frequent impact areas are Employment Generation (25.4%), Income/Productivity Growth (24.0%), and Health Improvement (18.0%). Women s least frequent impact areas are Natural Resources/Biodiversity (0.5%) and Affordable Housing (0.7%). Among men owned social enterprise, the most prevalent impact areas are Employment Generation (28.3%), Income/Productivity Growth (23.5%), and Community Development (19.7%). Men s least frequent impact areas are Water Resources Management (0.9%) and Natural Resources/Biodiversity (1.1%). Weak statistically significant differences between women and menowned social impact areas exist (at the.10 level) in Access to Clean Water and Community Development. No other statistically significant differences exist. 32

34 Table 4.15: Social impact areas Women owned Men owned All social N % of each area % women 66 N % of each area % men 67 N % of all social 68 Access to clean % 3.7% % 2.1% % water** Access to education % 14.8% % 16.2% % Access to energy % 4.6% % 5.2% % Access to financial % 9.9% % 10.8% % Services Access to information % 14.3% % 13.2% % Affordable housing % 0.7% % 1.3% % Agriculture % 11.1% % 12.9% % Productivity Biodiversity % 2.3% % 1.8% % conservation Capacity building % 10.6% % 11.9% % 33

35 Community % 15.7% % 19.7% % development** Conflict resolution % 1.2% % 2.0% % Disease specific % 2.3% % 3.5% % prevention and Mitigation Employment % 25.4% % 28.3% % generation Equality and % 15.5% % 15.3% % empowerment Food security % 7.4% % 10.0% % Generate funds for % 3.5% % 4.5% % charitable giving.health improvement % 18.0% % 15.6% % Human rights % 1.8% % 1.6% % protection or Expansion Income/productivity % 24.0% % 23.5% % Growth Natural % 0.5% % 1.1% % resources/biodiversity Other % 13.9% % 11.0% % Pollution prevention % 7.2% % 7.8% % and waste management Sustainable energy % 9.5% % 8.4% % and fuel efficiency 69 Sustainable land use % 3.5% % 4.0% % Support for high % 2.1% % 2.2% % impact entrepreneurs Support for women % 8.5% % 8.3% % and girls Water resources management % 1.8% % 0.9% % ** Weak statistically significant difference at the.10 level between women and men owned social. The most prevalent beneficiary group all social is Other (54.3%), which includes the general population, youth, farmers, students, millennials, families, women and men, the elderly, and entrepreneurs 70. Other is the most prevalent beneficiary women owned (55.3%) and men owned (53.9%). Overall, the least prevalent beneficiary is the disabled (2.2%). The same is true women owned (3.4%) and men owned (1.7%). There is a weak statistically significant difference (at the.10 level) between women and men owned social 34

36 regarding benefiting the disabled. No other statistically significant differences exist. Table 4.16: Beneficiaries 71 Women owned Men owned All social N % of each type % womenowned N % of each type % menowned N % of all social Children and % 15.6% % 16.4% % Adolescents Disabled** % 3.4% % 1.7% % Minorities % 9.4% % 11.7% % Other % 55.3% % 53.9% % Women % 16.3% % 16.3% % ,192 ** Weak statistically significant difference at the.10 level between women and men owned social. Among all social, about one third (31.5%) measure social impact. The same is true women owned (34.2%) and men owned (30.6%) social. No statistically significant differences exist between the rates at which women and men measure social impact. Table 4.17: Rates of Social Impact Measurement Women owned Men owned All social N % of each response % womenowned N % of each response % menowned N % of all social Yes % 34.2% % 30.6% % No % 65.8% % 69.4% 1, % 433 1,193 1,626 Hypotheses and Sub questions Research Question 2 asks: Does gender matter to the financial performance of social enterprise? Four hypotheses, introduced in Chapter 2, are tested by answering four sub questions as described in Table Next, I present analyses and findings. 35

37 Table 4.18: Hypotheses and related sub questions Hypotheses 1: Women and men owned social operate in different industries. 2: Women owned social are less likely than men owned social to operate in high revenue industries. 3: Women owned social have lower financial expectations and preferences than those owned by men. 4: The financial performance (revenue, profits, and size) of women owned social is lower than men s. Related Sub questions A. To what extent do women and men owned social operate in different industries? B. To what extent do women and men owned social operate in high revenue industries? C. To what extent do the financial expectations and preferences of social differ by sex of founders/owners? D. To what extent does the financial performance of social differ by sex of founders/owners? Hypothesis 1 and Sub question A H1: Women and men owned social operate in different industries. Sub question A: To what extent do women and men owned social operate in different industries? Sub question A is answered in 4 parts: A1. What operational sectors and models are most prevalent women owned social? A2. What operational sectors and models are most prevalent men owned social? A3. To what extent do women and men owned social operate in different sectors? A4. To what extent do women and men owned social use different operational models? A1 asks: What operational sectors and models are most prevalent women owned social? I conducted two cross tabulations; one for operational sectors and one for operational models, by women owned entities. Then, I ordered results from most to least frequent and reported the three most prevalent of each. 36

38 A1 Findings: Per Table 4.19, the three most prevalent operational sectors women owned social are Other (19.4%), Education (18.9%), and Health (15.9%). Per Table 4.20, the three most prevalent operational models women owned social are Services (60.7%), Production/Manufacturing (31.2%), and Wholesale/Retail (29.6%). Table 4.19: Most to least prevalent operational sectors women owned social N % women owned Other % Education % Health % Financial Services % Agriculture % Information and Communication Technologies % Energy % Supply Chain Services % Environment 9 2.1% Housing Development 9 2.1% Artisanal 5 1.2% Culture 5 1.2% Infrastructure/facilities Development 5 1.2% Technical Assistance Services 3 0.7% 37

39 Tourism 1 0.2% 433 Table 4.20: Most to least prevalent operational models women owned social N % womenowned Services % Production/Mfg % Wholesale/Retail % Distribution % Processing/Pkg % Financial Services % A2 asks: What operational sectors and models are most prevalent men owned social? Like the previous analysis (A1), I analyzed the distribution of operational sectors and models this time by men owned entities then ordered results from most to least frequent and reported the three most prevalent of each. A2 Findings: Per Table 4.21, the three most prevalent sectors men owned social are Education (18.5%), Health (18.5%), and Other (16.3%). Per Table 4.22, the three most prevalent operational models men owned social are Services (68.5%), Production/Manufacturing (32.6%), and Distribution (26.2%). Table 4.21: Most to least prevalent operational sectors men owned social N % menowned Education % Health % Other % Financial Services % Information and Communication Technologies % Agriculture % Energy % Environment % Culture % 38

40 Tourism % Infrastructure/facilities Development % Supply Chain Services % Artisanal % Housing Development 9 0.8% Technical Assistance Services 6 0.5% 1,192 Table 4.22: Most to least prevalent operational models men owned social N % menowned Services % Production/Mfg % Distribution % Wholesale/Retail % Processing/Pkg % Financial Services % A3 asks: To what extent do women and men owned social operate in different sectors? First, I ordered the operational sectors from least to greatest difference between women and menowned social. Then, I reported which group has the greater proportion in each operational sector (W for women; M for men). Last, I conducted t tests and reported statistically significant differences between women and men owned for each operational sector. A3 Findings: Per Table 4.23, statistically significant differences exist between the operational sectors of women and men owned social at the.05 level in Housing Development and Supply Chain Services; and at the.10 level in Tourism and Information/Communication Technologies. Table 4.23: Operational sectors of women and men owned social : Least to greatest difference N % womenowned 39 N % menowned Sex of greater proportion Statistically significant difference

41 Infrastructure/facilities 5 1.2% % W No Development Technical Assistance 3 0.7% 6 0.5% W No Services Artisanal 5 1.2% % W No Education % % W No Culture 5 1.2% % M No Tourism** 1 0.2% % M Yes Environment 9 2.1% % M No Housing Development* 9 2.1% 9 0.8% W Yes Supply Chain Services* % % W Yes Agriculture % % W No Financial Services % % W No Energy % % M No Information and % % M Yes Communication Technologies** Health % % M No Other % % W No 433 1,192 * Statistically significant difference at the.05 level between women and men owned social. ** Weak statistically significant difference at the.10 level between women and men owned social. A4 asks: To what extent do women and men owned social use different operational models? Similar to the previous analysis (A3), I ordered the operational models from least to greatest difference between women and men owned social. Then, I reported which group has the greater proportion in each operational model (W for women; M for men). Last, I conducted t tests and reported statistically significant differences between women and menowned for each operational model. A4 Findings: Per Table 4.24, statistically significant differences exist between women and menowned social at the.05 level in Processing/Packaging, Financial Services, Services, and Wholesale/Retail. Table 4.24: Operational models of women and men owned social : Least to greatest difference 40