Enterprising Women: TOWARD EQUAL BUSINESS OPPORTUNITY IN SOUTHEAST ASIA Acknowledgements This report was produced by the World Bank’s East Asia and Pacific Gender Innovation Lab (EAPGIL). It was prepared by a team led by Hillary C. Johnson and Elizaveta Perova, and including Ervin Dervisevic, Jhon Jair Gonzalez Pulgarin, Forest Jarvis, Yulia Krylova, Sundas Liaqat, Akaravuit Pancharoen, Alexander Spevack, José Daniel Trujillo, and Yue Wu. The team received guidance from Andrew Mason, Aaditya Mattoo, and Sudhir Shetty. Inputs for the box on the impacts of COVID-19 were provided by Leonardo Iacovone, Jesica Torres Coronado, and Trang Thu Tran. The team gratefully acknowledges the comments and advice of the peer reviewers, who included Francisco Campos, Smita Kuriakose, Amy Lunistra, David McKenzie, and Anna O’Donnell. The team also thanks Keiko Inoue, Jonathan Lain, and Ririn Purnamasari for their helpful comments and suggestions. The team benefitted from the guidance and support of Manuela Ferro and Victoria Kwakwa, Regional Vice Presidents, East Asia and the Pacific. The report was edited by Jennifer McNally and designed and laid out by Samantha DeMott-Anderson, Katie Fielding, Kendra McHugh, and Ben Roberts at Six Half Dozen. EAPGIL is supported by the Umbrella Facility for Gender Equality (UFGE) in partnership with the Australian Department of Foreign Affairs and Trade. UFGE has received generous contributions from Australia, Canada, Denmark, Finland, Germany, Iceland, Ireland, the Netherlands, Norway, Spain, Sweden, Switzerland, the United Kingdom, the United States, the Bill and Melinda Gates Foundation, and the Wellspring Philanthropic Fund. Photos used in this report come from the following sources: Cover photo and photo on page 13: © Kanawa_Studio / Getty Images. Used with the permission of Kanawa_Studio / Getty Images. Further permission required for reuse. Inside Front Cover and photos on pages 5, 14, 17, 24, 35: © World Bank. Further permission required for reuse. Photos on pages 5, 7, 9-10, 16, 24, 30, 41, 53, 55, 61, 63, 66, 97, 111, 122: © Pingann Oung / World Bank. Further permission required for reuse. Table of Contents Acknowledgements ii SECTION 4: 91 List of Figures iv Do opportunities, constraints, and needs of male- and female-owned SMEs differ? List of Tables v List of Boxes v 4.1 Perceived constraints 92 Executive Summary 1 4.2 Networks 94 Introduction 25 4.3 Access to finance 96 SECTION 1: 35 SECTION 5: 101 Gender gaps in participation Toward a targeted, gender-informed and performance of MSMEs entrepreneurship policy 1.1 Stylized facts 37 5.1 Interventions to enhance skills 104 1.2 Conceptual framework 40 5.1.1 Rationale and targeting 104 5.1.2 Evidence of promising skills- 105 SECTION 2: 43 enhancing interventions To be or not to be an entrepreneur, 5.1.3 Ensuring gender-inclusive skills- 110 that is the question enhancement interventions 2.1 What factors are linked with women’s 47 5.2 Interventions to improve access to 112 engagement in microenterprises? capital 2.1.1 Domestic work 47 5.2.1 Rationale and targeting 112 2.1.2 Availability and relative attractiveness 52 5.2.2 Evidence of promising interventions 113 of other labor market opportunities to improve access to capital 2.1.3 Access to capital 56 5.2.3 Ensuring gender-inclusive 119 2.2 What characteristics are linked with 59 interventions to improve access to capital women’s entry into SME ownership? 5.3 Interventions to support MSMEs as 123 employers SECTION 3: 65 5.3.1 Rationale and targeting 123 What contributes to the gender gap in performance of microenterprises? 5.3.2 Evidence of promising labor-related 124 interventions 3.1 Access to labor 67 5.3.3 Ensuring gender-inclusive labor- 126 3.1.1 Hired labor 67 related interventions 3.1.2 Own labor 71 5.4 Interventions to alleviate time 127 3.2 Skills 75 constraints related to domestic work 3.2.1 Formal education 75 5.4.1 Rationale and targeting 127 3.2.2 Business skills 78 5.4.2 Evidence of promising interventions to 128 reduce women’s domestic workload 3.2.3 Access to information 78 5.5 The path forward 131 3.3 Access to capital 79 iii Appendix A: Details of data sources used 136 List of Figures in the report Appendix B: Methods and econometric 137 Figure 1: Gender gaps in participation 37 specifications used in the report exist among SMEs but not microbusinesses Overview of methods used 137 Figure 2: Gender gaps in business 39 performance of microenterprises exist, Methods and specifications used in 138 but less evidence of performance gaps Section 1 among SMEs Model 1.1: OLS regressions 138 Figure 3: Framework for understanding 40 Methods and specifications used in 142 gender gaps in entrepreneurial outcomes Section 2 Figure 4: Women with young children are less 48 Model 2.1 Linear probability model with 142 likely to work, but the relationship between gender interactions children and entrepreneurship is more complex Model 2.2 Linear probability model with 144 Figure 5: After the birth of their first child, 49 individual fixed effects women in Indonesia are more likely to be Model 2.3 Events study 144 entrepreneurs and less likely to be wageworkers Methods and specifications used in 145 Figure 6: There is more heterogeneity in the 51 Section 3 number of hours that female entrepreneurs work than female wageworkers Model 3.1 OLS regressions with gender 145 interaction terms Figure 7: Higher education levels are 54 associated negatively with entrepreneurship Model 3.2 OLS regressions with individual 146 and positively with wage work for both men and fixed effects women, but the relationship varies by gender Model 3.3 Stepwise regressions 147 and country Model 3.4 Oaxaca-Blinder decompositions 147 Figure 8: Access to capital is positively 57 Model 3.5 Cobb-Douglas production function 149 associated with engagement in entrepreneurship Model 3.6 Analysis of female entrepreneurs 150 Figure 9: Most SMEs had five or more 59 operating in male-dominated sectors employees at start-up Methods and specifications used in 151 Figure 10: SMEs in Vietnam have different 62 Section 4 characteristics than microbusinesses Model 4.1 OLS regressions with gender 151 Figure 11: Women’s microbusinesses have 68 interaction terms fewer paid employees than men’s in all countries, but gender differences in unpaid workers vary by Model 4.2 Cobb-Douglas production function 151 country Appendix C: Distribution of business 152 Figure 12: Women’s microbusinesses have 69 performance variables for male and female increasing returns to labor, whereas men’s microenterprises and SMEs microbusinesses have decreasing returns to labor References 154 in Vietnam Figure 13: Accounting for differences in labor 70 reduces the gender gap in microbusiness performance Figure 14: Time spent on domestic tasks is 72 negatively associated with profits of female microentrepreneurs in Cambodia, and time-saving domestic infrastructure is associated with higher profits in Cambodia and Indonesia iv Figure 15: Female microentrepreneurs are 74 Figure 29: Only a small percentage of female- 96 more likely to operate their businesses from owned SMEs report being well-served by home financial institutions Figure 16: Operating a microbusiness from 74 Figure 30: Many female-owned SMEs are not 98 residential property is negatively associated well integrated in the formal financial system; with business performance however, there is heterogeneity across countries Figure 17: Female entrepreneurs have lower 76 Figure 31: Female-owned SMEs in Vietnam 99 education levels than male entrepreneurs have larger marginal returns to financial Figure 18: Education levels are positively 76 capital and lower marginal returns to capital associated with business performance of stock than male-owned SMEs female microentrepreneurs Figure 19: After accounting for differences 77 List of Tables in education levels, the gender gap in microenterprise performance shrinks in all Table 1: Top three constraints for male-owned 93 countries except Vietnam and female-owned SMEs Figure 20: Women’s microbusinesses in 80 Table 2: Summary of global evidence of 109 Indonesia had lower start-up capital than men’s programs to improve entrepreneurs’ skills Figure 21: Female-owned microenterprises 80 Table 3: Summary of global evidence of 119 in Vietnam have lower assets than men’s programs to improve entrepreneurs’ access microenterprises to capital Figure 22: Access to capital is positively 82 Table 4: Summary of global evidence of 120 associated with women’s microbusiness programs to support MSMEs as employers performance Table 5: Summary of global evidence of 126 Figure 23: Female-owned microbusinesses 84 programs to alleviate women’s time have higher returns to financial capital and constraints related to domestic work lower returns to capital stock than male-owned microbusinesses in Vietnam Figure 24: Female entrepreneurs in male- 85 List of Boxes dominated sectors have higher profits than female entrepreneurs in female-concentrated Box 1: Existing evidence documents 26 sectors in all countries except Cambodia gender inequality in endowments and agency in Southeast Asia Figure 25: COVID-19 lowered growth in 86 sectors with higher concentrations of female Box 2: The economic structure and regulatory 29 entrepreneurs environment vary across the countries included in the report Figure 26: Female entrepreneurs had larger 88 declines in their revenues linked with the Box 3: Data sources 31 COVID-19 pandemic than male entrepreneurs Box 4: What to expect (and not to expect) 32 Figure 27: Declines in revenues linked with the 88 from this report COVID-19 pandemic were largest for female- Box 5: Definitions 36 owned microenterprises Box 6: How norms, preferences, opportunities, 45 Figure 28: Female SME owners in Vietnam 95 and constraints shape entrepreneurial choices have similar numbers of individuals in their Box 7: How does this report analyze factors 46 networks as male SME owners but rely more on associated with women’s engagement in their networks for support and have a different entrepreneurship? network composition v Box 8: How does this report analyze 67 factors limiting the performance of women’s microbusinesses Box 9: What is the role of infrastructure 84 in supporting the productivity of women’s businesses? Box 10: How is sector choice linked with the 85 productivity of women’s businesses? Box 11: Gendered impacts of COVID-19 on 86 MSMEs in Southeast Asia Box 12: How does this report gender-specific 92 constraints among SMEs? vi Executive Summary This report explores gender differences in entrepreneurship for micro, small, and medium enterprises (MSME) in Southeast Asia. It analyzes data from five Southeast Asian countries: Cambodia, Indonesia, Lao People’s Democratic Republic (Lao PDR), Timor-Leste, and Vietnam, and reviews existing evidence from across the region. Using data from household and firm-level surveys, the report establishes stylized facts about gender gaps in participation in entrepreneurship and in business performance for two types of businesses: microbusinesses and small and medium enterprises (SME). Microbusinesses include businesses with less than 5 employees, and SMEs include businesses with between 5 and 100 employees. This study focuses on MSMEs because they account for 97 percent of all enterprises in Asia-Pacific Economic Cooperation countries1 and because the owner’s gender may have a greater influence on the outcomes of MSMEs than large businesses, which are more likely to have multiple owners or be part of a larger corporation. After establishing stylized facts about gender gaps in entrepreneurship, the report employs various econometric 1 Enterprising Women | Executive Summary Removing techniques to understand better the factors that are associated barriers to female with these gaps. Finally, the report makes a case for policy action entrepreneurship and presents a review of global evidence on how these factors are best addressed. can unleash untapped potential Understanding how and why female entrepreneurs lag their for economic male counterparts is critical to promote two key development growth and objectives: equity and growth. Gender-based inequalities of foster equity of opportunity can impede women from making work choices aligned with their goals, interests, and skills. Understanding opportunity. and addressing the gender-specific challenges that hinder women’s entrepreneurship can thus foster equity of opportunity. Removing barriers to female entrepreneurship can also unleash untapped potential for economic growth. When some women are excluded from becoming entrepreneurs, the average talent of entrepreneurs is lower, leading to an estimated per capita income loss of seven percent in East Asia and the Pacific.2 Reducing barriers to women’s labor market participation and occupational choices has also been shown to make large contributions to economic growth in the United States.3 This report finds significant gender gaps in entrepreneurship in Southeast Asia, which differ by the scale of the enterprise. Gender gaps are linked to lower levels of entrepreneurial and other inputs in women-led enterprises due to inhibiting gender norms, legal inequities, and market failures. Women are as likely as men to run microbusinesses, but there are gender gaps in microenterprise performance. These gaps are associated with lower levels of key inputs, including hired labor, time for own labor, skills, and capital. Women are less likely than men to own SMEs, but gender gaps in SME performance are not statistically significant in most countries. Women’s lower levels of skills, access to information, and capital help explain the gender gap in SME ownership, as these inputs are correlated with owning an SME. A comprehensive policy agenda is needed to address the multiple constraints that female entrepreneurs face, and policies need to address the heterogeneous needs of women operating businesses of different sizes. 2 Enterprising Women | Executive Summary Gender gaps in Women are less likely than men to own SMEs; entrepreneurial outcomes however, there are not statistically significant in Southeast Asia depend gender gaps in performance among SMEs in most countries. In all countries except Timor- on firm size Leste, less than half of SMEs are owned by Although women run about half of women, with the largest gaps in SME ownership microenterprises, women’s microenterprises in Indonesia and Vietnam (Figure O.1). Gender make lower profits and sales than those differences in business performance of SMEs of men. As shown in Figure O.1, the share are only statistically significant in Indonesia, of microbusinesses run by women ranges and this difference shrinks and loses statistical from 49 percent in Indonesia to 62 percent in significance when comparing men and women Timor-Leste. Nevertheless, there are gender in the same sector of activity (Figure O2: Panel gaps in business performance, and these gaps B). In Lao PDR and Timor-Leste, the gender remain or become wider when comparing gap in business performance reverses to favor men and women operating in the same female-owned SMEs when comparing men sector of activity (Figure O2: Panel A). Female and women with similar personal and business microentrepreneurs have profits or sales that characteristics. are between 11 percent and 41 percent lower than male microentrepreneurs in similar sectors of activity. Figure O.1 Gender gaps in participation exist among SMEs but not microbusinesses Percentage of microenterprises and SMEs managed or owned by women Percentage of microenterprises and SMEs managed or owned by women 70% 62% 60% 60% 54% 52% 53% 49% 50% 44% 40% 30% 23% 20% 20% 16% 10% 0% Cambodia Indonesia Lao PDR Timor-Leste Vietnam % microbusinesses female-managed % SME female-owned Dotted red line indicates gender parity Source: Calculations using CSES, IFLS, LECS, TLSLS, VARHS for microbusinesses and the WBES for SME 3 Enterprising Women | Executive Summary Figure O.2 Gender gaps in business performance of microenterprises exist, but there is less evidence of performance gaps among SMEs Panel A. Percentage difference in business performance between male and female microentrepreneurs Panel A: Percentage difference in business performance between male and female microentrepreneurs Raw gender gap in performance Gap after including only sector controls Gap after including all controls 0% -1.6% -5% -10% -9.9% -9.6% % women earn less than men -11.4% -11.4% -12.2% -15% -20% -17.4% -20.1% -25% -24.3% -30% -27.8% -35% -34.2% -35.2% -35.1% -40% -40.4% -40.5% -45% Cambodia Indonesia Lao PDR Timor - Leste Vietnam Significant at 1% Significant at 5% Not statistically Significant Panel B. Percentage difference in business performance between male and female SMEs Panel B: Percentage difference in business performance between male and female SMEs Raw gender gap in performance Gap after including only sector controls Gap after including all controls 80% 68.5% more than men % women earn 60% 40% 20.0% 20% 4.3% 0% -1.8% -2.6% % women earn less than men -20% -13.8% -19.7% -25.2%-27.0% -27.3% -28.0% -40% -37.5% -39.1% -41.6% -46.8% -60% Cambodia Indonesia Lao PDR Timor - Leste Vietnam Significant at 1% Significant at 5% Not statistically Significant Source: Panel A: Calculations using CSES, IFLS, LECS, TLSLS, VSMES ; Panel B : Calculations using WBES for Cambodia, Indonesia, Lao PDR, Timor-Leste, VSMES Note: In Panel A, business performance is defined as log profits for Cambodia, Indonesia, Timor-Leste, and Vietnam and log sales for Lao PDR. In Panel B, business performance is defined as log sales for Cambodia, Indonesia, Lao PDR and Timor-Leste, and log profits for Vietnam. See Model 1.1 in Appendix B for technical details and the list of control variables included. 4 Enterprising Women | Executive Summary Gender gaps in entrepreneurship are linked to lower levels of inputs in women-led enterprises, which are shaped by gender norms, legal distortions, and market failures. Existing evidence demonstrates that women- Women in Southeast Asia seem led enterprises have lower levels of key able to open microenterprises entrepreneurial inputs, including the time and with comparable ease to men skills of the entrepreneur, hired labor, and but face gendered barriers to capital. Women in Asia and the Pacific spend four SME ownership. times more time on unpaid care work than men,4 which limits the amount of time available to them Female microentrepreneurs for market activities, like running a business. underperform compared to male Although the gender gap in school enrollment microentrepreneurs, but gender has been closing and even reversed in some gaps in business performance countries over the past two decades, there are are less salient among SMEs. still gender differences in educational attainment and literacy among adults5 as well as differences in access to business-specific knowledge.6 Women’s businesses have fewer workers than men’s businesses.7 In addition, women have unequal access to inputs like land and credit.8 5 Enterprising Women | Executive Summary As shown in the conceptual framework a microbusiness. As such, women’s lower developed for this report (Figure O3), levels of entrepreneurial inputs make it more gender gaps in entrepreneurial inputs challenging for them to open an SME than for explain the observed gender gaps in SME men, contributing to observed gender gaps ownership and microbusiness performance. in SME ownership. Women who succeed in New analysis for this report shows that opening an SME have different characteristics female microentrepreneurs have lower and greater access to inputs than women levels of entrepreneurial inputs than male who are unable to open SMEs, and these microentrepreneurs, and input levels characteristics and access to inputs enable are associated with the gender gap in them to have sales and profits that are on par microenterprise performance. Opening an SME with those of male-owned SMEs. requires higher levels of inputs than opening Figure O.3 Framework for understanding gender gaps in entrepreneurial outcomes 6 Enterprising Women | Executive Summary Women-led enterprises do not have the same simultaneously care for young children, likely entrepreneurial inputs as men due to gender due to greater time and mobility constraints.10 norms, market failures, and legal distortions Market failures refer to missing or inefficient that constrain their choices and influence provision of goods or services that arise due their preferences of how to allocate their to issues such as externalities—benefits or available time and resources. Gender norms costs that are not specific to a producer or are unwritten rules that define acceptable and consumer—or informational barriers. For appropriate actions for women and men in example, one reason entrepreneurs under-invest a group or society.9 Gender norms, such as in business training is the lack of information those that emphasize women’s role in domestic about the potential returns of training and tasks or that dictate appropriate types of work information asymmetries about the quality for men and women, shape women’s choices of training providers.11 Female entrepreneurs’ and preferences for entrepreneurship. These networks are often smaller and less formal norms also affect what entrepreneurial inputs than those of men, which can exacerbate these are available to women. For example, female informational barriers. Legal distortions refer to entrepreneurs in Uganda who brought their explicit or implicit gender discrimination in laws young children to their retail businesses with or policies. For example, gender discrimination them were more likely to experience stock in property laws limit women’s ability to access shortages than women who did not have to capital.12 7 Enterprising Women | Executive Summary Female microentrepreneurs may also contribute to gender gaps in business make lower profits than male size. Women may hire fewer workers if they microentrepreneurs because have less information than men about the potential returns to hiring additional workers,14 they have lower levels of key and the size and composition of their networks inputs, including hired labor, may make identifying qualified recruits time for own labor, skills, more challenging. Finally, gender norms that and capital. emphasize women’s role in domestic work can drive women to maintain a smaller scale to Hired labor is associated with higher sales better balance their business operations with and profits, yet women’s microbusinesses household responsibilities.15 have fewer paid workers than men’s and rely more on unpaid labor. Evidence from Vietnam Women’s greater role in managing domestic suggests that women are operating below the and care work limits the amount of time optimal size of their firms. Specifically, female they can dedicate to their businesses, and microbusinesses in Vietnam have increasing entrepreneur’s own labor is critical for the returns to labor—for example, doubling the business performance of microenterprises. number of workers would more than double Because most microbusinesses—ranging the value of production. A greater proportion from 74 percent in Lao PDR to 97 percent in of women’s workers in Indonesia and Lao Timor-Leste—do not have any paid workers, PDR are unpaid compared to men, which can the entrepreneur’s own labor is an essential affect the quality and reliability of their labor input. Women’s ability to invest time in their supply. After accounting for women’s lower businesses is limited by gender norms that levels of paid and unpaid workers, the gender emphasize women’s role in domestic and care gap in business performance shrinks by 19 work and a lack of supportive infrastructure, percent, 10 percent, 21 percent and 4 percent such as affordable, quality childcare or in Cambodia, Indonesia, Lao PDR, and Timor- time-saving domestic appliances. Female Leste, respectively. Similarly, the size of the entrepreneurs who spend less time collecting labor force explains 15 percent of the gender fuel and water and those who have time-saving gap in profits in Vietnam. cooking appliances have higher sales and profits than those who spend more time on Several underlying factors may contribute to these activities and do not have appliances. women’s lower levels of hired labor. Negative Even if higher sales and profits enable women stereotypes about women’s ability as managers to invest in better household infrastructure, can make it more difficult for women to hire these investments signal female entrepreneurs’ and retain male employees13 and can lower desire to reduce the time they spend on their confidence in their own ability to manage domestic work. personnel effectively. Informational constraints 8 Enterprising Women | Executive Summary 34% Women may have difficulties In Lao PDR, 34 percent of Female microenterprise hiring additional labor if they female microentrepreneurs owners in Vietnam who lack the necessary financing identify primarily as adopt new technology get or if negative stereotypes homemakers in at least one less of a boost to their about women’s ability as month over the past year, and businesses compared managers make it more their revenues are 18 percent to men, despite the high difficult for them to hire and lower than those who were dividends for firms using retain male employees. primarily working all year. innovative technologies. <50% Gendered social norms can Female microentrepreneurs in Male-owned microbusinesses make it more challenging for Indonesia had less than half seem comparatively more women to build and diversify the amount of start-up capital constrained in physical their networks, but networks than men, and women’s capital, while female- are particularly important microbusinesses in Vietnam owned microbusinesses for women’s business have 18 percent less physical seem comparatively performance in Vietnam. assets and 39 percent less more constrained in financial assets than men.* financial capital.* *Physical assets or capital refer to buildings, machinery and equipment. Financial assets or capital refer to cash and receivables, such as money owned by customers. 9 Enterprising Women | Executive Summary Gender gaps in knowledge are associated with the gender gap in microbusiness performance: market failures and gender norms can make it challenging for female entrepreneurs to close these knowledge gaps on their own. Formal education can enable entrepreneurs to gain critical literacy and numeracy skills, and formal schooling is positively linked with business performance.16 Female entrepreneurs in Southeast Asia have lower levels of education than men. Accounting for women’s lower levels of education shrinks the gender gap in microenterprise performance by 39 percent in Cambodia, 11 percent in Indonesia, 43 percent in Lao PDR, and 8 percent in Timor-Leste. In Vietnam, female entrepreneurs also have less knowledge of business-related laws and regulations and are less likely to adopt innovative business practices including introducing new product groups. Women’s lower levels of business knowledge explain part of the gender gap in profits of microenterprises in Vietnam. Women may face challenges investing in training to address these skill gaps due to a shortage of training programs that are aligned with their needs and a lack of knowledge of the potential of investing in skill development programs. In addition, gender norms that emphasize women’s role in the household constrain the time women can invest in skill development programs. 10 Enterprising Women | Executive Summary Female microentrepreneurs have lower levels of business Without gender- assets than male microentrepreneurs and face gender- sensitive recovery related challenges in accessing capital. Female entrepreneurs policies, the in Indonesia have lower levels of start-up capital than male COVID-19 entrepreneurs. Compared to men’s microbusinesses in Vietnam, pandemic may women’s microbusinesses have lower levels of both physical capital—including land, buildings, equipment, and machinery—and widen gender gaps financial capital—including cash and receivables. Discriminatory in entrepreneurial property laws can make it more challenging for women to access outcomes. credit because they limit women’s access to assets that can be used as collateral.17 Even when laws provide for equal land rights or ban gender discrimination in access to credit, gender norms that dictate men’s role in administering assets can lead to gender differences in what is required in practice.18 Perhaps reflective of these challenges, female microentrepreneurs are less likely to have loans, invest capital in the business, or own the business location than male microentrepreneurs in Southeast Asia. Nevertheless, these three factors are only positively associated with the business performance of female entrepreneurs, and there is no statistically significant relationship for male entrepreneurs. The COVID-19 pandemic has exacerbated the challenges facing female microentrepreneurs. Microbusinesses, where female entrepreneurs are concentrated, experienced larger declines in their revenues associated with the COVID-19 pandemic than small, medium, or large enterprises. Moreover, the decline in revenues was larger for female microentrepreneurs than male microentrepreneurs in Cambodia and Vietnam. Such challenges may be linked to three pre-existing gender disparities. First, prior to the pandemic, women spent four times more hours on unpaid care work than men.19 The COVID-19 pandemic increased the amount of domestic work for many households, due to school and daycare closures, the need to care for sick family members, additional cleaning, and more meals being consumed at home. Although both men and women in East Asia and the Pacific have increased the amount of time they spend on caretaking and domestic activities,20 global evidence shows that women have increased their hours more than men.21 Second, female entrepreneurs are concentrated in sectors that have been hardest hit by the pandemic. For example, female microentrepreneurs 11 Enterprising Women | Executive Summary in Indonesia are concentrated in hotels, restaurants, and light manufacturing, sectors that experienced large declines in sectoral growth in 2020. Third, prior to the pandemic, female microentrepreneurs had less education and lower access to capital, which can limit the coping mechanisms that female entrepreneurs can adopt. Box O.1 Although there is gender parity in running microenterprises, there are gender differences in factors that orient men and women’s choices to open a microenterprise. Women’s greater role in domestic work, and women with higher levels of education are including childcare, can both hinder and more likely to engage in wage work and are motivate women to run microbusinesses. less likely to be entrepreneurs. However, the For example, in Cambodia, working-aged strength of the relationship between education women with young children are less likely to be and entrepreneurship is different for men and entrepreneurs than those without young children, women, and these trends vary by country. but considering only women in the labor force, Access to capital is essential for both women with young children are more likely to be male and female entrepreneurs to open entrepreneurs than other women. Motherhood businesses; however, it appears to matter can lower the likelihood of entrepreneurship more for Southeast Asian women. There is because it lowers the likelihood of participating a stronger correlation between indicators of in the labor force at all. Nevertheless, access to capital and the likelihood of being a entrepreneurship offers more flexible hours and microentrepreneur for women in most countries location of work than wage work, which makes in Southeast Asia. For example, both women and it an attractive option for working mothers. men in households that have received loans are The availability and conditions of wage work more likely to engage in entrepreneurship than make entrepreneurship a comparatively more those who have not received loans; however, this or less attractive labor market choice, but relationship is more than three times stronger men and women face different wage work for women than men in Timor-Leste. Similarly, opportunities. Men and women tend to sort into having assets that can be used as collateral is different occupations and educational tracks and more strongly associated with the probability thus face different wage work opportunities.22 of being a Cambodian entrepreneur for women Gender norms influence the type of work that than men. Savings are also an important source is deemed appropriate for men and women, of start-up capital for entrepreneurs, and which can further affect relative access to Indonesian women are more likely than men available jobs. In Southeast Asia, both men to rely on savings to start their businesses. 12 Enterprising Women | Executive Summary Gender gaps in SME ownership are linked The most common with women’s lower levels of skills, access way to become to information, and capital. an SME owner is to open an SME It is more common for successful SME owners in Southeast directly, rather Asia to have opened an SME directly rather than scaling up a than growing a microbusiness. More than half of male- and female-owned SMEs had at least five employees at start-up, and many firms that were microbusiness into considered microbusinesses at start-up had at least three or four an SME. employees when they began operations. Although both male and female SME owners tend to directly open an SME, men are more likely to own SMEs. This implies that women face obstacles to opening an SME. SME owners have greater skills, access to information, and access to capital than microentrepreneurs. Both male and female SME owners in Vietnam have greater knowledge of laws and regulations and use better and more formal business practices than their microenterprise counterparts. Although it is possible that larger business owners acquire these business specific skills and information by necessity due to their more advanced operations, basic skills are also correlated with SME 13 Enterprising Women | Executive Summary Networks provide ownership. Vietnamese SME owners—both male and female— critical information are more likely than microentrepreneurs to have completed and support for higher secondary or university level education. Networks provide a critical source of business information and support,23 and female SME- female SME owners have larger networks and more men in their owners; however, networks than female microentrepreneurs. Female SME owners gender norms are also more leveraged and are more likely to have applied restrict women’s for formal credit and to have borrowed informally than female ability to network. microbusiness owners in Vietnam, suggesting the importance of access to capital for SME ownership. Gender gaps in skills, access to information, and access to capital thus make it more difficult for women than for men to open SMEs. Women in Southeast Asia have lower levels of educational attainment and literacy than men24 as well as lower access to business-specific knowledge.25 Gender norms restrict the time that women have available for networking and limit the types of social relations and networking activities in which women can engage.26 In addition, women have unequal access to capital inputs including land and credit.27 Because skills, access to information, and access to capital are associated with the likelihood of owning an SME instead of a microenterprise, women’s lower levels of these key inputs contribute to observed gender gaps in SME ownership. 14 Enterprising Women | Executive Summary Box O.2 Although gender gaps in the performance of SMEs are not statistically significant in most countries, female-SME owners do face unique challenges. Female entrepreneurs perceive different Nevertheless, female SME owners in Vietnam constraints than male entrepreneurs report receiving assistance from contacts in and are more likely to face regulatory or their networks more frequently than male SME transport-related challenges. Customs owners. Aligned with this finding, the size of and trade regulations are among the top women’s networks is more strongly associated three cited issues for women-owned SMEs with SME profits for women than men. in Indonesia and Lao PDR; however, they While neither male- nor female-owned SMEs are less frequently cited among men-owned are well integrated into the formal financial SMEs. Similarly, women-owned SMEs in system, female-owned SMEs face greater Vietnam cite transport as their main business constraints in land ownership and financial challenge more frequently than men-owned assets than male-owned SMEs. The share of SMEs. These differences remain after female-owned SMEs with a checking or savings accounting for the fact that men and women account ranges from 38 percent in Cambodia to often operate in different sectors of activity. 87 percent in Timor-Leste, and the share with an Gender norms may shape the experiences that overdraft facility ranges from 2 percent in Lao female SME-owners have when interacting PDR to 30 percent in Timor-Leste. In all countries with regulators or the transport sector. studied, less than a quarter of female-owned Women’s networks are less formal than men’s SMEs have a line of credit or loan from a formal networks, yet female SME owners rely more on financial institution. In Lao PDR and Vietnam, the their networks for support. Female SME owners share of women’s SMEs with a line of credit or are less likely than male SME owners to belong loan is 28 and 20 percentage points, respectively, to business associations or to pay membership lower than the share of men’s SMEs with a line dues for business associations, suggesting of credit or a loan. In Vietnam, compared to women’s networks are less formal than those men-owned SMEs, women-owned SMEs have of men. Gender norms about appropriate social greater marginal returns to land and financial relations and time constraints linked to women’s capital, including liquid assets and cash, which greater roles in domestic work may influence signals that women’s SMEs are comparatively the type and composition of their networks. more constrained in these factors of production. However, in all countries studied, The share of 38% less than a third female-owned in Cambodia to SMEs with a of female-owned SMEs have an overdraft facility, and checking or 87% savings account in Timor-Leste less than a quarter ranges from have a line of credit or loan from a financial institution. 15 Enterprising Women | Executive Summary Table O.1 Summary of barriers to women’s microenterprise performance and participation in SMEs GENDER GAP IN TYPES OF GENDER GAP IN SME BARRIER MICROENTERPRISE INTERVENTIONS PARTICIPATION PERFORMANCE INDICATED SKILLS, KNOWLEDGE, AND INFORMATION GAPS   Skills-enhancing LOWER LEVELS OF CAPITAL COMPARED TO MEN   Improving access to capital LOW, AND IN SOME CASES SUB- OPTIMAL, LEVELS OF HIRED LABOR  ? Supporting hiring and managing of workers TIME CONSTRAINTS DUE TO DOMESTIC WORK  ? Alleviating time constraints due to domestic work Note: Check mark indicates evidence that the barrier is associated with the observed gender gap in entrepreneurship. Question mark indicates that the data do not enable testing the relationship between the barrier and the observed gender gap. 16 Enterprising Women | Executive Summary A comprehensive policy agenda is Gender-informed needed to address women’s multiple MSME policy constraints and the heterogeneous contributes to needs of female entrepreneurs operating equitable, inclusive at different scales. development and maximizes the This report builds the case for tailored MSME policies that effectiveness of address women’s specific needs and are designed to reach limited resources. both microbusinesses and SMEs. Women have access to lower levels of entrepreneurial inputs than men, which affects their ability to engage and compete in entrepreneurship. These gender differences in inputs are linked to the underperformance of women’s microbusinesses compared to those of men and women’s lower representation among SME owners. The challenges that female microbusiness and SME owners face relative to men are different. Gender-informed MSME policies can promote both equity and growth objectives and maximize the efficiency of public investment by ensuring that some individuals and businesses are not excluded on the basis of gender. Two types of policies should be ideally adopted in parallel: policies that directly address input gaps and those that influence the underlying drivers of gender gaps. As described in the conceptual framework, gender gaps in entrepreneurial inputs—including hired labor, time for own labor, skills, and capital—are linked with gender gaps in entrepreneurial outcomes, including participation in entrepreneurship and business performance. Gender norms, market failures, and legal distortions are underlying factors that constrain women’s access to inputs and inform their preferences of how to allocate limited resources. Because addressing these underlying drivers is a 17 Enterprising Women | Executive Summary slow process, policies that can address gender entrepreneurs in Southeast Asia: skills gaps in inputs in the short term can help. In interventions, promoting access to capital, some cases, policies that directly address supporting hiring and management of workers, current gender gaps in inputs can also help and alleviating time constraints related to shift underlying drivers of gender gaps, creating domestic work. Many of these policies, such as a virtuous cycle. For example, programs to those supporting skills development, improving facilitate female entrepreneurs’ access to access to finance, and facilitating hiring and bank loans directly boost beneficiaries’ input managing workers, could also support male levels and simultaneously enable more bank entrepreneurs. Additionally, other types of officers to interact with female entrepreneurs. policies, such as improvements to the business Such interactions may counteract the environment, may support both male and influence of gender norms and shift bank female entrepreneurs. This report focuses on officers’ perceptions of female entrepreneurs’ policies that have the potential to address the managerial capacity, ability to repay, and gender gaps identified in this report. Table O.2 passion for their businesses. summarizes the global evidence of specific interventions for each of the four types This report reviews global evidence on four of policies. types of policies that can support female 4 ways to support female entrepreneurs in Southeast Asia: 1 Promote skill development 2 Improve access to capital 3 Support to hire and manage workers 4 Reduce time women spend on domestic work Enterprising Women | Executive Summary Table O.2 Summary of the effectiveness of different policy options based on existing, global evidence TYPE OF INTERVENTION RELEVANCE SUMMARY OF EVIDENCE EFFECTIVENESS INTERVENTION DESCRIPTION MICRO SME Standard business training programs can have a positive impact on firm performance. Standard business However, alternative training approaches that       training programs use heuristics, highly tailored content, or psychological principles have proven more effective on average. Business training that uses basic rules of Alternative thumb, content that is highly tailored to the business training local context, or is based on psychology       programs is more effective than standard business training at boosting business performance. SKILL ENHANCING Business consulting can lead to sustained Personalized positive impacts on SME performance. Micro guidance Evidence is somewhat mixed for from business     microenterprises and may be more effective consultants or for those with more experience, formal trained mentors education, or slightly larger businesses. SME Social learning interventions have improved Peer-to-peer the business performance of both SMEs and learning and       microenterprises and can be cost-effective network formation solutions. Microfinance can generate small, positive changes but are unlikely to lead Microfinance       to transformative impacts for female entrepreneurs. Although there is a paucity of rigorous Overcoming evidence, psychometric testing, cash flow collateral       loans, and credit guarantees hold potential to constraints support female entrepreneurs. Subsidized loans, While these programs can help credit- directed lending, constrained firms access credit, design and       and blended implementation are critical to ensure proper IMPROVING finance targeting and avoid distorting the market. ACCESS TO CAPITAL Cash grants have overall not succeeded in boosting the performance of women’s Cash microbusinesses, often due to issues of Unconditional diversion of funds from their businesses.     grants In-kind grants may hold more potential for female microentrepreneurs, but more evidence is needed. In-Kind Business plan competitions offering cash prizes have been proven to increase high Business plan growth entrepreneurship for both men and       competitions women; however, the design should include specific targeting to women, such as female- only competitions. 19 Enterprising Women | Executive Summary RELEVANCE TYPE OF INTERVENTION SUMMARY OF EVIDENCE EFFECTIVENESS INTERVENTION DESCRIPTION MICRO SME IMPROVING Savings interventions have helped increase ACCESS TO Savings promotion investment and business performance of       CAPITAL women’s microbusinesses. Improving knowledge of labor Limited rigorous evidence is available, but laws and good       existing evidence seems promising. human resources practices SUPPORTING HIRING AND MANAGING OF A few studies show that wage subsidies may WORKERS have impacts if they are tailored and targeted to specific types of firms; however, more Wage subsidies evidence is needed on how they can impact       female-owned firms. Combining subsidies with support to identify qualified workers may be more impactful. There is a solid evidence base that childcare can increase women’s labor force participation; however, more evidence is Improving access needed to understand the impacts on female to affordable,       entrepreneurship. More evidence is also quality childcare needed on the ideal type of childcare to support female entrepreneurs in different country contexts. Evidence shows that improved household infrastructure, such as in-house drinking water, time-saving cooking technologies, Supporting and other time-saving appliances reduce access to time- time spent on domestic tasks. However, saving, improved       more evidence is needed to understand the household impact of these time savings on female infrastructure ALLEVIATING entrepreneurs and whether the amount of TIME time saved is sufficient to translate into CONSTRAINTS meaningful impacts for female entrepreneurs. RELATED TO DOMESTIC WORK Well-designed parental leave policies that include sufficient, highly paid, nontransferable leave to both parents can boost men’s Parental leave involvement in childcare. However, this may policies not be a policy priority to support female       entrepreneurs in contexts where leave protections are not universal to all workers, such as those in the informal sector. Programs that engage men in discussions related to gender-related issues have shown Engaging men increases in joint decision-making and an and shifting increase in men’s involvement in domestic intra-household       tasks in addition to other development allocation of outcomes affecting women. More evidence is domestic work needed on how such programs can support female entrepreneurs more specifically. RELEVANCE Strong relevance Somewhat relevant Not relevant EFFECTIVENESS  emonstrated effectiveness D  ixed or limited evidence M  inimally or not effective M 20 Enterprising Women | Executive Summary In addition to specifically targeting women, clear protocols to mitigate risks of harassment broad MSME policies need to be designed with or gender-based violence and a robust gender-specific constraints in mind to ensure grievance redress mechanism to address any that women can equitably access and benefit concerns. Moreover, programs addressing from them. Accessing MSME programs may capital constraints must be designed to be a challenge for female entrepreneurs as account for gender imbalances in intra- they face greater time and mobility constraints household bargaining power and normative and are at a higher risk of harassment. To pressures for women to divert capital flows ensure that both men and women can equitably toward the needs of their households or other access all MSME programs, outreach should members of their family. For example, financial target women, enrollment procedures should information should remain confidential and accommodate women with greater time or accessible only to the female entrepreneur, and mobility constraints, and hours of program products such as in-kind grants or commitment operations should be compatible with women’s savings devices may enable women to better schedules. In addition, programs need to have shield these inputs from external demands. Checklist for Ensuring Equitable Access to MSME Policies: I  s the program’s content relevant for  re there mitigation measures in place A female entrepreneurs and is it tailored to protect female entrepreneurs from to their needs? the risk of gender-based violence, harassment, and discrimination? D  o the program’s logistical arrangements make it more difficult  ould gender imbalances in intra- C for women to access or participate in household bargaining affect the the program? effectiveness of the program for female entrepreneurs, and can a  ould social norms or stereotypes C gender-smart design overcome lower the likelihood that women such impediments? participate in the program, and can those risks be mitigated? 21 Enterprising Women | Executive Summary A comprehensive A comprehensive, gender-informed, MSME policy agenda is strategy and strong needed because addressing one constraint in isolation may not coordination be effective if others are binding. For example, an entrepreneur who is able to better identify opportunities after a skill among intervention will only reap the benefits if they are able to find the stakeholders financing to implement their idea. Conversely, an entrepreneur can advance who receives funding will make better use of that funding if they more equitable have solid managerial and entrepreneurial skills. Indeed, recent entrepreneurship evidence shows that intervention packages that address multiple policies for barriers to female entrepreneurship, such as those providing more equitable both skills development and financing, are more effective than outcomes. those that address only one constraint.28 Thus, a multiplier effect from complementary interventions may justify their increased cost. In a resource constrained environment, it may be more impactful to address multiple binding constraints for a smaller group of entrepreneurs than to address one constraint for a larger group of entrepreneurs. Nevertheless, some underlying drivers, such as gender norms, take time to change and may limit the extent to which certain input gaps, such as time for own labor and skills, can be effectively closed in the short term. The design and expected return to costly interventions may need to be conditioned on the extent to which the targeted or complementary inputs can be addressed. Because action is needed on multiple fronts, strong coordination is essential to achieve the goal of MSME policies that equitably benefit men and women with different sized firms. Interventions that address the varied challenges that women’s MSMEs face likely fall under the purview of more than one ministry or institution. Ideally, a high-level champion can push for a comprehensive strategy that convenes the various stakeholders to develop and adhere to a shared vision. Although the path to equitable entrepreneurship is long, the intrinsic value of equitable opportunities and the potential for economic growth make it a worthwhile endeavor. 22 Enterprising Women | Executive Summary Endnotes 1  sia-Pacific Economic Cooperation Working Group A on Small and Medium Enterprises, https://www.apec. org/ Groups/SOM-Steering-Committee-on-Economic- and- Technical-Cooperation/Working-Groups/Small-and- Medium-Enterprises.aspx 2 Cuberes and Teignier 2016 3 Hsieh et al. 2019 4 International Labor Organization 2018: 56 5 World Bank 2012b 6 Ellis, Kirkwood, and Malhotra 2010 7 World Bank 2016b; International Finance Corpration 2017 8 Ellis, Kirkwood, and Malhotra 2010 9 Cislaghi and Heise 2020 10 Delecourt and Fitzpatrick 2020 11 McKenzie 2020 12 Demirgüç-Kunt et al. 2008; Goheer 2013 13 International Finance Corporation 2008 14 Hanna, Mullainathan, and Schwartzstein (2014) demonstrate that entrepreneurs with little experience with an input, such as labor, do not learn that hiring an additional worker could be profitable. Given women have fewer workers than men, they are less likely to have had experience hiring additional workers and may be more likely to underestimate the potential benefits of doing so. 15 Asia Foundation 2013b Brush and Hisrich 1991; Casson 1982; Schutjens and Wever 16  2000 17 Demirgüç-Kunt et al. 2008; Goheer 2013 18 Ellis et al 2010; International Finance Corporation 2017 19 International Labor Organization 2018 20 UN Women 2020a 21 UN Women 2020b 22 World Bank 2012 23 Ács et al. 2017; Álvarez, Marin, Fonfría 2009; Szarka 1990 24 World Bank 2012b 25 Ellis, Kirkwood, and Malhotra 2010 26 Ellis, Kirkwood, and Malhotra 2010 27 Ellis, Kirkwood, and Malhotra 2010 28 Revenga and Dooley 2020 23 Enterprising Women | Executive Summary 24 Introduction This report identifies gendered barriers to opening and growing micro, small, and medium enterprises (MSMEs) in Southeast Asia and makes policy recommendations for addressing them. It combines a review of the existing evidence with new analysis from five Southeast Asian countries: Cambodia, Indonesia, Lao People’s Democratic Republic (Lao PDR), Timor-Leste, and Vietnam.1 The report measures gender gaps in entrepreneurial participation and performance for microbusinesses and SMEs. It defines microbusinesses as those with fewer than five employees and SMEs as those with between five and 100 employees. It then identifies factors associated with these gaps for both types of enterprises and finally reviews global evidence of what works to close them. Not all individuals are equally placed to open and successfully run an MSME, and gender influences one’s ability to engage and compete in entrepreneurship. To both open and successfully grow a business, both men and women must have the necessary skills, 25 Enterprising Women | Introduction finance, and time. However, as shown in Box Addressing gender inequalities that make it 1, women in Southeast Asia have lower levels harder for women to engage and compete of all three of these endowments, putting them in entrepreneurship has intrinsic value and at a disadvantage. It is also critical to have promotes a human rights agenda. The ability agency—the ability to set goals in line with one’s to live the life of one’s choosing and to be values and act in a way to achieve these goals— spared from deprivation is a basic human to develop and implement a business plan. right that should be equal for everyone, Southeast Asian women have lower levels of regardless of gender.2 Removing obstacles agency than men, which further impedes their that leave individuals with little choice or ability to engage in careers aligned with their opportunity to exercise their reasoned agency skills and preferences and to independently is both the principal means and the primary conduct business activities. ends of development.3 Promoting equitable opportunities for women’s entrepreneurship is thus an important development objective in itself. Box 1 Existing evidence documents gender inequality in endowments and agency in Southeast Asia Despite progress made, women in Southeast Women continue to lack voice and influence in Asia continue to have more limited both public and private spheres, even if East endowments, including skills, assets, and time, Asia performs better than other regions on than men. Although the gender gap in school many indicators of agency.9 In 2019, women enrollment has been closing and even reversed only held 20.5 percent of seats in national in some countries over the past two decades, parliaments and 11.3 percent of ministerial- there are still gender differences in educational level positions in East Asia and the Pacific.10 attainment and literacy among adults4 as In the private sphere, qualitative work reveals well as differences in access to business- that men in several Southeast Asian countries specific knowledge.5 In addition, women face continue to have more influence over household challenges in accessing credit due to a lack of decisions. For example, despite strong social assets that can be used as collateral, which is norms and respondent preferences for linked with discriminatory property laws6 and consultative decision-making in the Philippines, norms that regard husbands as responsible for wives often defer to their husbands’ opinions capital-related transactions or land.7 Moreover, because he is the head of the household or to women often face more demands on their keep peace in the household.11 Moreover, men time, as social norms often dictate that women in Vietnam reported allowing their wives to should be primary caregivers for the family, decide on small and routine purchases to avoid even when they work outside the home.8 housework and protect their masculinity but insisted they were always the decision-makers on major investments or important family decisions, even when their wives disagreed.12 26 Enterprising Women | Introduction Removing barriers to female entrepreneurship Gender-informed MSME can also unleash untapped potential for policies are critical because economic growth. If innate entrepreneurial gender-neutral policies talent is evenly distributed across men and can have gendered impacts women, when women are excluded from if men and women have managerial positions, the average talent of managers is lower. This in turn leads to lower different needs, face different innovation, slower technology adoption, and challenges, or encounter eventually to lower output.13 Cuberes and gendered barriers to fully Teignier (2016) quantify per capita income benefitting from public losses due to gender gaps in labor force programs. participation and entrepreneurship at between 15 percent and 16 percent in East Asia and advantage of public programs. In some cases, the Pacific. Gender gaps in entrepreneurship small tweaks may be needed to ensure that alone lead to losses in income per capita of 7 both men and women can benefit from general percent in the region. Removing barriers that MSME policies and programs. For example, prevent women from entering the occupation of because of women’s greater time constraints their choice can boost economic growth. Hsieh due to competing domestic responsibilities, et al. (2019) find that improved allocation of offering business training programs during talent due to a reduction in racial and gendered hours when children are typically in school, barriers to labor force participation and close to women’s place of businesses, and that occupational choice between 1960 and 2010 in offer childcare arrangements can facilitate the United States explained between 20 percent women’s participation. In other cases, it may and 40 percent of growth in aggregate market be necessary to offer tailored programs, output per person. The same study found that services, or policies to specifically reach the reduction in earnings gaps during this female entrepreneurs and address the unique period explained 8 percent of per capita GDP challenges they face. For example, financial growth. Aligned with these findings, Cavalcanti products that do not have strict physical and Tavares (2015) estimate that a 50 percent collateral requirements, such as cash flow increase in the gender wage gap leads to a 35 loans or movable collateral, can help female percent decrease in income per capita. While entrepreneurs, who often lack collateral, existing studies on earnings gaps focus on to access the finance they need for their wage workers, gender gaps in entrepreneurial businesses. earnings could constrain growth through the This report aims to help policy makers same mechanisms. design gender-informed MSME policies Policy makers in Southeast Asia need to by filling knowledge gaps about female ensure that MSME policies are meeting the entrepreneurship in Southeast Asia. Most needs of both men and women. Ostensibly existing studies of female entrepreneurship gender-neutral policies may have gendered in Southeast Asia focus on either impacts if male and female entrepreneurs have microbusinesses or SMEs, include only existing different needs, face different challenges, or entrepreneurs, and focus on one country. encounter gendered barriers to fully taking This report examines both microbusinesses 27 Enterprising Women | Introduction and SMEs and demonstrates that female entrepreneurship given the high-economic entrepreneurs running businesses of different growth that characterizes many Southeast sizes have different characteristics, challenges, Asian economies. All the countries included in and needs. Gender-informed MSME policies this report with the exception of Timor-Leste must not only consider how women’s needs have experienced sustained, high economic may differ from men’s but how to tailor growth over the past few decades, with growth programs for female entrepreneurs operating rates averaging over 5 percent prior to the at different scales. Another novelty in this COVID-19 pandemic. This high growth has report is the study of both current and potential been accompanied by substantial poverty entrepreneurs, which enables an exploration reduction prior to the pandemic. Such a context of factors that facilitate and impede women’s can provide fertile soil for various economic engagement in entrepreneurship. Finally, opportunities, including an increase in the by comparing across several countries with availability of wage jobs and entry points for heterogeneous contexts (see Box 2), this report opportunity driven entrepreneurs. Nevertheless, offers perspective on regional versus country- in a context where inequalities have been rising specific trends. in recent decades, it is necessary to examine the extent to which women have been able to The report focuses on MSMEs because equitably benefit from these new economic most entrepreneurs are concentrated in opportunities. this segment, and gendered constraints to entrepreneurship may be particularly binding This report first demonstrates the existence among MSMEs. MSMEs account for more than of gender gaps in entrepreneurial outcomes, 97 percent of all enterprises, employing more then explores drivers of these gaps, and than half of the workforce and contributing finally provides evidence of how to address between 20 percent and 50 percent of GDP them. The first section examines the extent in most Asia-Pacific Economic Cooperation to which gender gaps in entrepreneurial countries.14 Because most enterprises are outcomes, including participation and MSMEs, addressing gendered barriers to MSME business performance, exist in Southeast ownership and growth will alleviate constraints Asia. After establishing stylized facts on the for most female entrepreneurs. In addition, existence of gender gaps in participation and gendered constraints that MSME owners face performance, the report explores factors that may be more binding than for larger firms shape women’s choices and ability to engage where co-owners or parent corporations may be in entrepreneurship in Section 2. The next two able to provide support to overcome barriers. sections examine drivers of gender gaps in Approximately a quarter of SMEs have more microenterprise performance and gendered than one owner, compared with more than constraints to SME performance, respectively. three-quarters of large enterprises, and MSMEs Finally, section 5 presents recommendations are much less likely than large firms to have for how MSME policy can better address the any foreign ownership or to be part of a larger diverse needs of different entrepreneurs: corporation.15 effective policies must adopt a gender lens and also be adapted to challenges men and women Southeast Asia provides a unique context face when operating businesses of different for exploring gendered constraints to sizes. 28 Enterprising Women | Introduction Box 2 The economic structure and regulatory environment vary across the countries included in the report Cambodia a high reliance on commodities, with limited integration in global value chains and moderate Cambodia has had sustained average real exposure to international trade.21 Since 2015, on growth of 7.7 percent per year between average, there have been approximately 2 million 1998 and 2019, making it one of the fastest jobs created per year, resulting in relatively growing economies in the world. Three sectors, low unemployment. However, the majority of including tourism, manufacturing exports, jobs created are in agriculture or low-value- and construction, accounted for more than added services, with less than a quarter of 70 percent of growth and almost 40 percent jobs created in manufacturing or higher-value- of paid employment in 2019.16 The economy added services.22 As a large archipelago nation, is largely dependent on a narrow base of there are large differences in opportunities and manufactured exports, with garments and service delivery across different regions.23 footwear exports accounting for approximately 80 percent of total merchandise exports.17 Lao PDR Entrepreneurial activity in Cambodia is Lao PDR has averaged 7.8 percent GDP relatively recent, with the oldest businesses growth per year over the past decade, making in Cambodia founded in 1979. Economically, it the 13th fastest growing economy in the private ownership was banned under the world. This growth has been fueled by the Democratic Kampuchea regime governed use of natural resources, a gradual opening by the Khmer Rouge from 1975 to 1979, and of the economy, and regional integration. entrepreneurship disappeared.18 Although However, growth has not fueled job creation, households could engage in small trade or and 82 percent of the labor force work handicraft activities during the following regime in agriculture or self-employment. 24 of the People’s Republic of Kampuchea from 1979 to 1989, Cambodia did not transition back Timor-Leste to a market-based economy until after 1989, and a rise in entrepreneurship only occurred Timor-Leste’s economy is highly dependent on after 1993 when the Kingdom of Cambodia petroleum: at the peak of petroleum production, began to more fully liberalize the economy.19 it comprised 95 percent of government revenues. However, Timor-Leste’s largest oil and gas field is Indonesia approaching depletion. Other economic activity is driven by public spending or second round Indonesia is the world’s fourth most populous effects of public spending, such as consumption nation and 10th largest economy in terms of increases from cash transfer programs. purchasing power parity.20 Prior to the COVID-19 Manufacturing is not developed; however, private pandemic, Indonesia experienced consistent investment, notably in services, has started economic growth, with growth averaging 5 to contribute to growth since 2007. Most job percent per year since 1990. The economy has creation in recent years has been in the public 29 Enterprising Women | Introduction sector or linked to businesses with public- sector experienced substantial productivity sector contracts. Despite progress made since gains in the 1990s. Following the Enterprise Timor-Leste became an independent nation in Law in 2000, which legalized private firm 2002, the legacy of long, devastating conflict creation, private sector firm growth increased has left Timor-Leste with weak infrastructure dramatically. Today, there are over 650,000 and weak policy and institutional frameworks.25 registered domestic, private enterprises, compared to only 40,000 in 1999 and virtually Vietnam none in 1990. Strong job creation and growth in the service sector and export-oriented Vietnam has experienced remarkable growth manufacturing followed trade liberalization and over the past 30 years, going from one of the ascension to the WTO in 2007. Manufacturing world’s poorest nations to a thriving middle- exports including garments, footwear, and income economy. The Doi Moi reforms electronics have grown at an extremely fast of the late 1980s and continued reforms rate, with higher value products such as phones, have contributed to rapid growth. After the computers, and related components growing decollectivization of agriculture and creation to almost a third of exports in 2016 from less of individual land usage rights, the agricultural than 5 percent of exports 10 years prior.26 30 Enterprising Women | Introduction Box 3 Data sources This report uses existing panel and cross- income and consumption, they mainly contain sectional data from Indonesia and Vietnam information on household businesses, which and existing cross-sectional data from are predominately microbusinesses, and only Cambodia, Lao PDR, and Timor-Leste, to have basic information about the businesses. analyze microenterprises and SMEs. Appendix Firm surveys, on the other hand, provide rich A describes the data sources used in the information on businesses; however, they report’s analysis. Two types of data are used: typically have limited information on the household surveys and firm-level surveys. business owner or manager or their household. Both types of data have advantages and Because firm surveys only include information disadvantages. The different datasets used in on people who have decided to engage in this report provide complementary information entrepreneurship, they cannot address issues to create a more comprehensive picture of of selection into entrepreneurship. This report female entrepreneurship in Southeast Asia. uses two types of firm-level surveys: the World Household surveys include rich data on Bank Enterprise Surveys and a panel survey of the characteristics of different types of MSMEs in Vietnam. The World Bank Enterprise workers—including entrepreneurs—and Surveys focuses on formal businesses. Although their households and are typically nationally the panel survey of MSMEs in Vietnam is not representative. This type of data enables restricted to formal businesses, over 80 percent an analysis of the factors linked with the of the SMEs in the survey are formal businesses. choice to engage in entrepreneurship versus Because this report relies upon existing data other labor market choices. The household sources, the available information varies by business modules of these surveys provide country. If no results for a topic of interest basic information on microbusinesses that are mentioned for a specific country, readers can be linked with the rich information on should assume that data constraints impeded the business owners, their households, and analysis for that country, unless otherwise in some cases, their communities. Because indicated. Statistically insignificant results household surveys are typically designed to are either noted in the text or in footnotes. assess household-level outcomes, such as 31 Enterprising Women | Introduction Box 4 What to expect (and not to expect) from this report This report systematically analyzes factors male and female entrepreneurs and may associated with gender differences in play a larger role in determining enterprise entrepreneurial outcomes in Southeast Asia. performance than gender-specific factors. It measures gender gaps in participation in However, it is outside the scope of this report entrepreneurship and in business performance to conduct a comprehensive diagnostic for both microenterprises and SMEs in five of factors constraining entrepreneurship Southeast Asian countries. It explores factors in addition to identifying gender-specific that are linked with women’s decisions to factors that affect female entrepreneurs. As engage in entrepreneurship, the size of such, this report is focused on how women’s businesses they operate, and their business experience with entrepreneurship differs from performance. It compares how women’s men due to the gender-specific constraints decisions to enter entrepreneurship and they face. This analysis can complement factors linked with their productivity are broader diagnostics to ensure that resulting similar or different from men, which can MSME policies are addressing the specific provide suggestive evidence of gender- needs that female entrepreneurs face and specific opportunities, preferences, or that female entrepreneurs are able to equally constraints that may drive gender differences benefit from broader MSME policies. in participation in entrepreneurship or business Because the report relies on existing data performance. By conducting similar analysis sources for the new analysis, there are two main from several countries, the report can shed limitations. First, despite efforts to harmonize light on both convergent and divergent methodology and the inclusion of different trends across countries in Southeast Asia. variables in the analysis across countries This report does not provide a comprehensive and firm size, this was not always possible diagnostic of all constraints to entrepreneurship due to the different survey methodologies in Southeast Asia. It focuses on how women’s and questionnaires. Second, the analysis experiences with entrepreneurship are conducted for this report is descriptive different from men’s. Aspects of the business in nature. Although the correlations shed environment or resource constraints affecting light on policy-relevant trends, the analysis both men and women can constrain both cannot identify causal relationships. 32 Enterprising Women | Introduction Endnotes 1  mong Southeast Asian countries, the five focus countries A for new analysis were selected due to the availability of data needed to study both microenterprises and SMEs. 2 World Bank 2012a 3 Sen 1999 4 World Bank 2012b 5 Ellis, Kirkwood, and Malhotra 2010 6 Demirgüç-Kunt et al. 2008; Goheer 2013 7 Ellis et al., 2010: 21 8 International Labor Organization 2018a 9 World Bank 2012b 10 World Bank Gender Data Portal: https://www.worldbank. org/en/data/datatopics/gender/country/East%20Asia%20 &%20Pacific 11 Arugay et al. forthcoming 12 Ha 2020 13 Esteve-Volart 2004 Asia-Pacific Economic Cooperation Working Group 14  on Small and Medium Enterprises, https://www.apec. org/ Groups/SOM-Steering-Committee-on-Economic- and- Technical-Cooperation/Working-Groups/Small-and- Medium-Enterprises.aspx 15 Calculations using the most recent World Bank Enterprise Surveys for the countries included in this report. 16 https://www.worldbank.org/en/country/cambodia/overview 17 World Bank, 2016 18 Chhair and Ung, 2016 19 Chhair and Ung, 2016 20 https://www.worldbank.org/en/country/indonesia/overview 21 World Bank, 2020b 22 World Bank and International Finance Corporation, 2019 23 World Bank, 2020b 24 World Bank, 2017 25 World Bank, 2018 26 Eckardt, Demombynes, and Chandrasekharan Behr, 2016 33 Enterprising Women | Introduction 34 Section 1: Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) This report examines two important areas of entrepreneurship in which gender gaps can emerge: participation and performance. If men or women own or manage businesses at a higher rate than individuals of the opposite gender, there is a gender gap in participation in entrepreneurship. This type of gap may signify underlying gender- specific constraints or opportunities that either hinder or motivate individuals of a particular gender to engage in entrepreneurship. Gender-specific constraints and opportunities can also affect the performance of existing businesses and can lead to gender gaps in business performance in terms of profits, turnover, or other indicators of business performance. Existing evidence from Southeast Asia suggests that the extent to which a gender gap in participation in entrepreneurship exists depends on the size and characteristics of the businesses considered. Generally, trends show that among smaller businesses, men and women participate at roughly similar levels, or women 35 Enterprising Women | Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) are more engaged in business ownership medium enterprises (SMEs) owned by men and and management. However, as the size and women had similar average annual revenues.4 sophistication of the businesses increase, Analysis of rural, non-farm businesses in six women participate at lower levels than men. kabupaten in Indonesia showed that female- For example, in Indonesia, more women than owned firms have lower sales but similar sales men run microenterprises,1 but only 23 percent per worker compared to male-owned firms.5 of firms with between 10 and 200 employees This report helps fill knowledge gaps are owned by women.2 In Cambodia, 65 concerning gender differences in MSME percent of MSMEs are classified as women-led, performance while distinguishing between but women only own one third of registered firms of different sizes. Understanding how businesses and less than 1 percent of firms gender gaps in performance differ based on the with more than 10 employees.3 size of the firm is important, as global evidence Evidence on gender gaps in business shows that women’s performance relative to performance in Southeast Asia is thinner. men varies not only by region but also by the Notable exceptions include analysis from size of the firm.6 Vietnam and Indonesia. The 2015 enterprise census in Vietnam showed that small and Box 5 Definitions This report defines microenterprises as either their primary or secondary job is self- having less than five employees and SMEs employed or employer. To consider gender as having between five and 100 employees, gaps in productivity, the logarithm of annual which is aligned with the definitions used by profits is used when available.8 In cases in the International Finance Corporation and the which the annual profits is not available in World Bank Enterprise Surveys. The definition the data or cannot be reliably calculated, the of a woman-owned or led business depends on logarithm of annual revenues is used with the dataset structure. For firm-level datasets, controls for sector of activity.9 Profits are used the gender of the owner is considered, and for for microbusinesses in Cambodia, Indonesia, firms with multiple owners, at least 51 percent and Timor-Leste and for both microbusinesses of ownership must be female. For household and SMEs in Vietnam. Revenues are used for surveys, the gender of the primary person microbusinesses in Lao PDR and for SMEs in responsible for the business is considered.7 Cambodia, Indonesia, Lao PDR, and Timor-Leste. To explore selection into entrepreneurship, an individual is considered an entrepreneur if 36 Enterprising Women | Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) 1.1 Stylized facts Consistent with existing evidence, gender these countries. The gap is particularly large in gaps in participation in entrepreneurship vary Indonesia, Vietnam, and Lao PDR where only 16 by the size of businesses considered, with percent, 20 percent, and 23 percent of SMEs, women participating at similar levels to men respectively, have majority female ownership. in micro entrepreneurship but being under- In contrast, Timor-Leste has a reverse gender represented among SMEs (Figure 1). Among gap in participation in entrepreneurship, with microenterprises, men and women participate women running 62 percent of microbusinesses at similar levels, with between 49 percent and and owning 60 percent of SMEs. This difference 54 percent of microenterprises managed by may be linked with the overall lower levels of women in Cambodia, Indonesia, Lao PDR, and entrepreneurial activity in Timor-Leste, the Vietnam. However, significant gender gaps in prolonged conflict in recent decades, or the participation exist when considering SME in structure of Timor-Leste’s economy. Figure 1 Gender gaps in participation exist among SMEs but not microbusinesses Percentage of microenterprises and SMEs managed or owned by women Percentage of microenterprises and SMEs managed or owned by women 70% 62% 60% 60% 54% 52% 53% 49% 50% 44% 40% 30% 23% 20% 20% 16% 10% 0% Cambodia Indonesia Lao PDR Timor-Leste Vietnam % microbusinesses female-managed % SME female-owned Dotted red line indicates gender parity Source: Author calculation using CSES, IFLS, LECS, TLSLS, VARHS for microenterprises and WBES for SMEs. 37 Enterprising Women | Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) Although women are equally or more likely precise and comparable measures of profits than men to run microenterprises, women’s for both microenterprises and SMEs, there are microbusinesses perform worse than men’s statistically significant gender gaps in profits microbusinesses on average. Figure 2: Panel for microbusinesses but the gaps for SMEs A shows that female microentrepreneurs are both insignificant and close to zero. In earn less than male entrepreneurs in almost Lao PDR and Timor-Leste, the gender gap in all specifications. The raw gender gap in business performance remains statistically performance is not statistically significant insignificant but reverses to favor women when in Cambodia, but a statistically significant comparing male- and female-owned SMEs with gender gap emerges when comparing men similar owner and business characteristics. and women operating in the same sector of Nevertheless, in Indonesia, there is a activity. In other countries, the raw gender gap statistically significant gender gap in revenues ranges between 10 percent in Vietnam and of SMEs. Although it loses significance when 40 percent in Indonesia. Comparing women comparing male- and female-owned SMEs with and men with similar individual, household, similar owner and business characteristics, and business characteristics, the gender the magnitude of the gap is similar to that of gap in business performance widens in microenterprises. Cambodia and Lao PDR, shrinks somewhat in Indonesia, and remains fairly constant in Timor- Leste and Vietnam. After taking individual, household, and business characteristics into account, female microentrepreneurs earn between 10 percent and 35 percent less than male microentrepreneurs on average, with differences statistically significant at the 1 percent level. It is worth noting that while female microentrepreneurs have lower business performance than male microentrepreneurs on average, not all female microentrepreneurs perform worse than male microbusinesses. Appendix C shows the distribution of business performance for both microenterprises and SMEs by gender. For SMEs, gender gaps in business performance vary by country but in most cases are not statistically significant and do not persist after accounting for business and personal characteristics (Figure 2: Panel B). Gender gaps in business performance of SMEs are not statistically significant in Cambodia, Lao PDR, Timor-Leste, or Vietnam. In Vietnam, where detailed economic accounts data enable 38 Enterprising Women | Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) Figure 2 Gender gaps in business performance of microenterprises exist, but less evidence of performance gaps among SMEs Panel A. Percentage difference in business performance between male and female microentrepreneurs Panel A: Percentage difference in business performance between male and female microentrepreneurs Raw gender gap in performance Gap after including only sector controls Gap after including all controls 0% -1.6% -5% -10% -9.9% -9.6% % women earn less than men -11.4% -11.4% -12.2% -15% -20% -17.4% -20.1% -25% -24.3% -30% -27.8% -35% -34.2% -35.2% -35.1% -40% -40.4% -40.5% -45% Cambodia Indonesia Lao PDR Timor - Leste Vietnam Significant at 1% Significant at 5% Not statistically Significant Panel B. Percentage difference in business performance between male and female SMEs Panel B: Percentage difference in business performance between male and female SMEs Raw gender gap in performance Gap after including only sector controls Gap after including all controls 80% 68.5% more than men % women earn 60% 40% 20.0% 20% 4.3% 0% -1.8% -2.6% % women earn less than men -20% -13.8% -19.7% -25.2%-27.0% -27.3% -28.0% -40% -37.5% -39.1% -41.6% -46.8% -60% Cambodia Indonesia Lao PDR Timor - Leste Vietnam Significant at 1% Significant at 5% Not statistically Significant Source: Panel A: CSES, IFLS, LECS, TLSLS, VSMES ; Panel B : WBES for Cambodia, Indonesia, Lao PDR, Timor-Leste, VSMES Note: In Panel A, business performance is defined as log profits for Cambodia, Indonesia, Timor-Leste, and Vietnam and log sales for Lao PDR. In Panel B, business performance is defined as log sales for Cambodia, Indonesia, Lao PDR and Timor-Leste and log profits for Vietnam. See Model 1.1 in Appendix B for technical details and the list of control variables included. 39 Enterprising Women | Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) 1.2 Conceptual framework The conceptual framework developed for have lower levels of entrepreneurial inputs this report (Figure 3) posits that gender than male microentrepreneurs, and input gaps in entrepreneurial inputs explain the levels are associated with the gender gap in observed gender gaps in SME ownership and microenterprise performance. Opening an SME microbusiness performance. As described requires higher levels of inputs than opening in Box 1, women in Southeast Asia have a microbusiness. As such, women’s lower lower levels of endowments—including time, levels of entrepreneurial inputs make it more skills and capital—and agency than men. challenging for them to open an SME than for These endowments are critical inputs for men, contributing to observed gender gaps entrepreneurs, and women’s limited agency in SME ownership. Women who succeed in can hinder their ability to use their available opening an SME have different characteristics endowments as they see fit. New analysis and greater access to inputs than women for this report described in the following who are unable to open SMEs, and these sections also supports the important characteristics and access to inputs enable relationship between entrepreneurial them to have sales and profits that are on par inputs and gender gaps in entrepreneurial with those of male-owned SMEs. outcomes. Female microentrepreneurs Figure 3 Framework for understanding gender gaps in entrepreneurial outcomes 40 Enterprising Women | Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) Women-led enterprises do not have the same due to greater time and mobility constraints.11 entrepreneurial inputs as men due to gender Market failures refer to missing or inefficient norms, market failures, and legal distortions provision of goods or services that arise due that constrain their choices and influence to issues such as externalities—benefits or their preferences of how to allocate their costs that are not specific to a producer or available time and resources. Gender norms consumer—or informational barriers. For are unwritten rules that define acceptable and example, one reason entrepreneurs under-invest appropriate actions for women and men in in business training is the lack of information a group or society.10 Gender norms, such as about the potential returns of training and those that emphasize women’s role in domestic information asymmetries about the quality of tasks or that dictate appropriate types of work training providers.12 Female entrepreneur’s for men and women, shape women’s choices networks are often smaller and less formal and preferences for entrepreneurship. These than those of men, which can exacerbate these norms also affect what entrepreneurial inputs informational barriers. Legal distortions refer to are available to women. For example, female explicit or implicit gender discrimination in laws entrepreneurs in Uganda who brought their or policies. For example, gender discrimination young children to their retail businesses with in property laws limit women’s ability to access them were more likely to experience stock capital.13 shortages than women who did not have to simultaneously care for young children, likely 41 Enterprising Women | Gender gaps in participation and performance of micro, small, and medium enterprises (MSMEs) Endnotes 1  World Bank, 2016 2  Asia Foundation, 2013 3  Asia Foundation and Asian Development Bank, 2018 4 IFC, 2017 5  Costa and Rijkers, 2011 6  Bardasi et al., 2011 7  In Timor-Leste, the data does not distinguish between primary and secondary person responsible for the business. In this case, businesses with multiple people responsible for them are dropped from the analysis. 8 Microbusiness profits are reported directly in the Indonesia  dataset and are calculated by subtracting costs from revenues in datasets from Vietnam, Cambodia and Timor- Leste. 9  The exception is for microbusinesses in Lao PDR, for which the question pertained to monthly revenues instead of annual revenues. 10 Cislaghi and Heise 2020 11 Delecourt and Fitzpatrick 2020 12 McKenzie 2020 13 Demirgüç-Kunt et al. 2008; Goheer 2013 42 Section 2: To be or not to be an entrepreneur, that is the question Working-age men and women face several inter-related and simultaneous choices—whether to engage in the labor market, the type of work to do, and the number of jobs to do. For example, one may choose between refraining from the labor market, engaging in wage work, entrepreneurship, supporting a family business as an unpaid worker, or combining any number of these activities. When considering entrepreneurship, individuals can opt into microentrepreneurship or choose to open a larger-scale business that requires more employees. These choices are shaped by the opportunities that individuals perceive are available to them, social norms about the appropriateness of different options, individual preferences, and expected benefits and costs associated with each choice. Gender influences all these factors. Gender inequalities in endowments and agency determine the availability, feasibility, and relative attractiveness of different labor market options for women and men. These gender inequalities determine the availability, feasibility, and relative attractiveness of 43 Enterprising Women | To be or not to be an entrepreneur, that is the question different options for women and men. For and contribute to gender gaps in performance. example, if opening a small and medium Understanding why men and women choose enterprise (SME) requires significant capital to become entrepreneurs can therefore shed investment and women have lower financial light on factors influencing the gender gap in endowments than men, starting an SME may performance of microenterprises. Section 1 not be an option for many women. Alternatively, also showed that women are underrepresented if women do not have the skill levels that most among SME owners in almost all countries, employers require, wage work may be difficult suggesting gender-related barriers to opening to find. Agency can also play an important role an SME. Learning which factors seem to be in choosing economic activities. For example, key for women to open SMEs could shed light mobility constraints can make it difficult for on potential policies to help women overcome women to access wage jobs that are farther gender-specific obstacles to opening an SME. from home or to run businesses operations that require greater mobility. Understanding how gender inequalities influence the decision to engage in entrepreneurship can shed light on the stylized facts of gender gaps in entrepreneurial outcomes presented in Section 1 of this report. Although women are equally represented among microenterprises, they have lower business performance on average. If different types of women and men select into entrepreneurship due to the way gender influences their choices, these differences in characteristics and/or motivation can contribute to performance gaps. Due to a lack of outside opportunities and a need to earn a living, in many economies, women are more likely than men to engage in necessity entrepreneurship—motivated by a need to earn a living, the scarcity of jobs, and/or the need to balance labor market with domestic and care work.1 If business ownership is not a passion or if it is seen as secondary to one’s role as a caretaker for the family, one may be less driven to seek out new opportunities for business growth and development. Thus, even if there is gender parity in owning a business, if men and women own businesses for different reasons, gender gaps in both monetary and time resources invested in the business may arise 44 Enterprising Women | To be or not to be an entrepreneur, that is the question Box 6 How norms, preferences, opportunities, and constraints shape entrepreneurial choices Just as the choice of whether to engage in These various aspects of the entrepreneurial entrepreneurship is not determined in isolation, ecosystem shape preferences for the factors influencing this choice cannot entrepreneurship as well as opportunities be considered in isolation. There is a range and constraints to engage in it. The extent to of complicated interdependencies between which men and women view entrepreneurship preferences for self-employment, opportunities, as a promising and socially acceptable and constraints associated with starting, running, employment option influences the desire and growing a business. Gender norms and to become an entrepreneur. Moreover, the aspects of the entrepreneurial ecosystem shape availability and characteristics of alternative these preferences, opportunities, and constraints. labor market opportunities for men and women influence the relative attractiveness The term entrepreneurial ecosystem refers to of entrepreneurship compared to other multiple stakeholders (individuals, organizations, options. In addition to a desire to engage in and institutions) and their policies that are entrepreneurship, the availability of resources either conducive to or inhibitive of preferences such as financing and human capital can for self-employment and business choices. determine how feasible it is to start a business. For example, Isenberg (2010) identifies six domains within the entrepreneurial ecosystem Because all these factors are interconnected, that encourage individuals to start a business: it can be challenging to disentangle whether a conducive culture, enabling policies and observed gender differences in entrepreneurship leadership, availability of appropriate finance, are driven by the different opportunities that men quality human capital, venture friendly markets and women face, gender-specific constraints, for products, and a range of institutional support. or preferences, all of which may be affected The entrepreneurial ecosystem is not gender- by underlying social norms. However, because neutral, as Terjesen and Lloyd (2015) showed these factors are interconnected, shifting one in an analysis of gender-specific conditions of can also help in shaping and changing the entrepreneurial ecosystems in 77 countries. others. For example, policies that alleviate They found that family-related institutions, gender-specific constraints contribute to a equal legal rights, access to education, more balanced entrepreneurial environment. networks, technology, capital, supportive social This signals that female entrepreneurship norms and values are important determinants is valued in society, which can influence of high-potential female entrepreneurship women’s preferences. Rather than focusing development. On the contrary, gender stereotypes, on whether underlying drivers of gender biases against women in entrepreneurship, differences stem from opportunities, constraints, limited access to resources, social capital, preferences, or norms, this report focuses on and infrastructure limit both entrepreneurial policy recommendations that can foster an opportunities and perceptions of women related entrepreneurial ecosystem that is conducive to to starting and growing their businesses.2 female entrepreneurship in Southeast Asia. 45 Enterprising Women | To be or not to be an entrepreneur, that is the question Box 7 How does this report analyze factors associated with women’s engagement in entrepreneurship? To understand the factors that are associated correlate with one’s likelihood of engaging in with women’s engagement in microenterprises, different labor market activities while controlling we use household survey data that includes for unobservable time-invariant characteristics.4 information on all working-aged individuals, Rich panel data is available for Indonesia, which regardless of their employment status. We run allows us to conduct event studies to understand several linear probability models to understand how the birth of a woman’s first child affects her the extent to which individual and household labor market choices. The event study compares characteristics are associated with different labor her likelihood of engaging in entrepreneurship market choices, and because these patterns may in the years after her first child is born to her be different for men and women, we interact the likelihood of engaging in entrepreneurship characteristics of interest with the individual’s in the year prior to the birth of her child.5 gender.3 The results cannot be considered causal but can reveal interesting patterns about how Household surveys rarely capture SMEs, as likely men or women with a certain characteristic they typically ask about household businesses of interest are to engage in a specific activity that tend to be smaller operations. Most compared to those without that characteristic, information on SMEs is only available for while controlling for confounding factors. For existing businesses, which makes it difficult to example, it compares whether women with a child analyze selection. To still shed light on factors under age 3 in the household are more or less facilitating women’s SME ownership, we use likely to engage in entrepreneurship than women simple descriptive statistics to compare the without a child under age 3 in the household, characteristics of male and female SME owners while taking factors like age or education into with those of male and female microbusiness account. In Indonesia and Vietnam where panel owners for the country where we have a dataset data is available, we also run these models that includes a sizeable share of both. using individual fixed effects, which compares how changes in these characteristics over time 46 Enterprising Women | To be or not to be an entrepreneur, that is the question 2.1 What factors are linked with women’s engagement in microenterprises? Three main factors seem to influence women’s Women’s greater contributions to domestic decision to open a microenterprise in Southeast work may affect women’s selection into Asia. First, domestic responsibilities influence entrepreneurship in two ways. On one hand, women’s labor market choices, both in terms of women who spend more time on domestic work whether to engage in labor market activities and may be less likely to engage in labor market the type of work, including entrepreneurship. activities at all, including entrepreneurship.8 On Second, the enabling environment and other the other hand, women may choose to engage opportunities available in the labor market in entrepreneurship because it can afford more orient the type of work that women do. Finally, flexible working hours and location and can access to capital seems to facilitate women’s facilitate combining childcare with productive entry into entrepreneurship. work.9 There is evidence supporting both hypotheses for different countries in the region. 2.1.1 Domestic work Domestic workload is associated with a lower likelihood of participation in entrepreneurship Women in Southeast Asia, as in many parts of due to its link with lower labor force the world, are seen as the primary caretakers participation in Cambodia and Lao PDR, but in the household and do a disproportionate not in Indonesia, Timor-Leste, or Vietnam. In amount of unpaid, domestic work.6 According Cambodia and Lao PDR, women with young to the 2018 report on Care Work and Care Jobs, children are less likely to engage in the labor women spend 4.1 times more time on unpaid force (Figure 4: Panel A) and in entrepreneurship care work than men in Asia and the Pacific.7 (Figure 4: Panel B). This suggests that in Gender imbalances in time spent on domestic Cambodia and Lao PDR, women who likely work is interlinked with cross-cutting gender spend more time on unpaid care work are inequalities. The more time women spend on less likely to engage in entrepreneurship, domestic work, the more time constrained they since they are less likely to engage in labor are for participating in labor market activities. market activities at all. However, this trend Although time spent on domestic and care work is not universal in the region. In Timor-Leste can be aligned with women’s goals and values, and Indonesia, although women with children greater time on domestic and care work may under age 3 are less likely to be in the labor also reflect some women’s lack of agency to force (Figure 4: Panel A), they are more likely decide how they spend their time. The hours in Indonesia and just as likely in Timor-Leste to during which women perform certain domestic engage in entrepreneurship (Figure 4: Panel B). activities, such as caring for children outside We do not find any relationship between having of school hours, can limit which economic young children and labor force participation or opportunities are available to them. entrepreneurship in Vietnam. 47 Enterprising Women | To be or not to be an entrepreneur, that is the question Figure 4 Women with young children are less likely to work, but the relationship between children and entrepreneurship is more complex Panel A: Conditional Panel A. Conditional between between correlation correlation having under age 3 having children children under 3 and labor force participation and labor force participation for working-age women for working - aged women Cambodia Indonesia Lao PDR Timor - Leste Vietnam -0.16 -0.14 -0.12 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 Panel B. Conditional between having correlationcorrelation Panel B: Conditional between having under age 3 children children under and engaging 3 and engaging in in entrepreneurship entrepreneurship for working-age women for working - aged women Cambodia Indonesia Lao PDR Timor -Leste Vietnam -0.05 -0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04 Conditional between Panel C:correlation Panel C. Conditional correlation between having having children under age 3 and children under 3 and engaging in entrepreneurship, engaging in entrepreneurship, limited to women in the labor force limited to women in the labor force Cambodia Indonesia Lao PDR Timor - Leste Vietnam -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 Source: Author calculations using CSES, IFLS, LECS, TLSLS, VARHS Note: The blue dots show regression coefficients from Model 2.1, as described in Appendix B, and the bars show the 95 percent confidence intervals. 48 Enterprising Women | To be or not to be an entrepreneur, that is the question Evidence from Cambodia and Indonesia supports the After Indonesian hypothesis that women who need to balance childcare women give birth responsibilities and income generating activities may to their first child, choose entrepreneurship. As discussed above, women they are more likely in Cambodia with children under age 3 are less likely to to be engaged in be entrepreneurs because they are less likely to work. However, those who choose to work are more likely to entrepreneurship engage in entrepreneurship (Figure 4: Panel C). Women and less likely to be in Indonesia who have young children are also more working in wage likely to be entrepreneurs, whether considering the entire employment than working age population or only those who are active in before giving birth the labor force (Figure 4: Panel B and C). (Figure 5). Even After Indonesian women give birth to their first child, after their children they are more likely to be engaged in entrepreneurship reach school age, and less likely to be working in wage employment than women remain in before giving birth (Figure 5).10 Even after their children entrepreneurship reach school age, women remain in entrepreneurship and do not switch and do not switch back to wage work. Nevertheless, back to wage work. working women with young children are not more likely to be entrepreneurs in Lao PDR, Timor-Leste, or Vietnam. Figure 5 After the birth of their first child, women in Indonesia are more likely to be entrepreneurs and less likely to be wageworkers Panel A. Event study of first childbirth Panel B. Event study of first childbirth on female entrepreneurship on women’s wage employment Source: Author calculations using the IFLS. Note: See Model 2.3 in Appendix B for more details. 49 Enterprising Women | To be or not to be an entrepreneur, that is the question In all countries Entrepreneurship can offer more flexible hours, which studied, can enable women to more easily combine market entrepreneurship work with domestic work, such as childcare.11 The average number of hours that female entrepreneurs seems to offer work compared to female wageworkers depends much more on the country. In Cambodia and Lao PDR, female flexibility in the entrepreneurs work fewer hours per week on average amount of time than female wageworkers, whereas in Timor-Leste the that women spend opposite is true. In Indonesia, female entrepreneurs and working. wageworkers spend a similar number of hours working. However, in all countries studied, entrepreneurship seems to offer much more flexibility in the amount of time that women spend working. Figure 6 shows there is much more variability in the number of hours that female entrepreneurs work compared to female wageworkers, as the distribution is represented by a much flatter curve. The majority of female wageworkers tend to work a specific number of hours, which suggests women may have less choice about the number of hours they work when they are employed by someone else rather than self-employed as entrepreneurs. 50 Enterprising Women | To be or not to be an entrepreneur, that is the question Figure 6 There is more heterogeneity in the number of hours that female entrepreneurs work than female wageworkers Hours women spend working per week Hours women spend working per week kernel density estimate of hours worked in kernel density estimate of hours worked in entrepreneurship and wage jobs in Cambodia entrepreneurship and wage jobs in Indonesia Hours women spend working per week Hours women spend working per week Kernel density estimate of hours worked in entrepreneurship and wage jobs in Cambodia Kernel density estimate of hours worked in entrepreneurship and wage jobs in Indonesia .03 .1 Entrepreneurship Entrepreneurship Wage job Wage job .08 .02 .06 Density Density .04 .01 .02 0 0 0 20 40 60 80 100 0 50 100 Hours women spend working per week Hours women spend working per week Hours women spend working per week Hours women spend working per week kernel density estimate of hours worked in kernel density estimate of hours worked in entrepreneurship and wage jobs in Lao PDR entrepreneurship and wage jobs in Timor-Leste Hours women spend working per week Hours women spend working per week Kernel density estimate of hours worked in entrepreneurship and wage jobs in Lao PDR Kernel density estimate of hours worked in entrepreneurship and wage jobs in Timor−Leste .04 Entrepreneurship Entrepreneurship .06 Wage job Wage job .03 .04 Density Density .02 .02 .01 0 0 0 50 100 150 0 20 40 60 80 100 Hours women spend working per week Hours women spend working per week Source: Author calculations using CSES, IFLS, LECS, TLSLS Note: These graphs show the kernel density function, which depicts the underlying distribution of a continuous variable. In other words, it shows the extent to which female entrepreneurs (represented by the yellow lines) and female wageworkers (represented by the blue lines) report working different numbers of hours per week. A flatter line, such as the yellow lines, show that there is more variability in the number of hours worked, whereas a distribution with one or several peaks, such as the blue lines, demonstrate that a large share of women work a certain number of hours. 51 Enterprising Women | To be or not to be an entrepreneur, that is the question In addition, entrepreneurship can offer 2.1.2 Availability and relative more flexibility in terms of work location. attractiveness of other labor In Indonesia and Lao PDR, women are more market opportunities likely to operate their businesses from home than men, either partially or entirely. Running As discussed above, the decision of businesses from home can make it easier whether to engage in entrepreneurship is for women to combine caretaking or other not made in isolation: some people turn to domestic work with their business operations. entrepreneurship due to the lack of available Although we do not find significant correlations wage jobs. Entrepreneurship is one of between business location and the number several choices individuals make regarding of children under age 3 or primarily being a whether and how to engage in the labor homemaker in Indonesia or Lao PDR, because market, and these decisions are influenced business location can be difficult to change by the availability and conditions of different over time, past or expected future childcare labor market opportunities. According to the constraints and domestic responsibilities may Global Entrepreneurship Monitor, the share of still influence the current business location. entrepreneurs who started their business due to necessity varies by a country’s income level, Although entrepreneurship offers some with 35 percent of entrepreneurs in low-income flexibility for both men and women, global economies, 28 percent in middle-income evidence suggests that women are more economies, and 18 percent in high-income likely to take advantage of this flexibility to economies stating that they opened their simultaneously manage earning and caring business because they had no better option for responsibilities. For example, Delecourt and work.14 Fitzpatrick (2020) find that 37 percent of female microentrepreneurs in their sample in Uganda Men and women may not have the same bring small children to work, compared to 0 access to available job opportunities because percent of men. Nevertheless, entrepreneurs gendered social norms shape the formation who open businesses to resolve conflict between of men and women’s human capital and competing earning and caring roles may not interactions with perspective employers. be as well-equipped to compete with those Even though gender gaps in school enrollment who open a business out of passion or after have been closing in recent decades,15 due to identifying a promising business opportunity. historical gender gaps in access to education, Global evidence suggests that women who are working-age women have lower education pushed to open businesses to balance market levels than working-age men in all countries and domestic work are more likely to run smaller, included in this report. In addition, sorting less growth-oriented enterprises.12 Conversely, into fields of education and occupations is when institutional policies are supportive of also influenced by gender,16 which leaves men balancing market and domestic work, such as and women with skills and experience that those offering paid leave, publicly subsidized can determine their access to available job childcare, and part-time employment, women in opportunities. Even when skills and experience industrialized countries are more represented may not predetermine which opportunities are among growth entrepreneurs, even if they are available for men and women, strong social less likely to engage in entrepreneurship overall.13 norms about which jobs are appropriate for 52 Enterprising Women | To be or not to be an entrepreneur, that is the question women may influence women’s choice to apply In all countries studied, higher levels and their likelihood of receiving the jobs. In a of education are associated with lower study of gender social norms across 20 low- engagement in entrepreneurship and a higher and middle-income countries, Muñoz-Boudet probability of wage work for both men and et al. (2013) show that women’s choices women (Figure 7). However, the strength of were linked with whether a job was deemed this relationship is different for women and appropriate for women or not, and in many men in Cambodia, Lao PDR, and Vietnam, cases, inappropriate jobs involved a real or which may reflect differences in available perceived danger from interacting with men. wage jobs for men and women with different Even when women choose to apply to a job, skill levels. In Cambodia and Lao PDR, the firms may consciously alter their hiring choices negative relationship between education and if employees or customers prefer a gender- entrepreneurship is stronger for men than for segregated workplace,17 and unconscious bias women, whereas in Vietnam, the opposite may also make it more difficult for women to is true. This may reflect the differences in be hired for positions more strongly associated available wage jobs for women and men with with men.18 When men and women do not face different skill levels: for example, if there are the same wage work opportunities, it influences more opportunities for low-skilled women than their decisions of whether or not to engage in low-skilled men in a country, one would expect entrepreneurship by changing the number, type, that the relationship between education and and relative attractiveness of their other labor entrepreneurship would be weaker for women market options. than men. 53 Enterprising Women | To be or not to be an entrepreneur, that is the question Figure 7 Higher education levels are associated negatively with entrepreneurship and positively with wage work for both men and women, but the relationship varies by gender and country Panel A. Conditional correlation between education and A: Conditionalfor entrepreneurship Panel working-age correlation men and between womenand education entrepreneurship for working - aged men and women Men Women Cambodia Indonesia Lao PDR Timor - Leste Vietnam -0.025 -0.02 -0.015 -0.01 -0.005 0 Panel B. Conditional Panel B.correlation between Conditional education correlation and wage between work and education for working-age men and women wage work for working - aged men and women Men Women Cambodia Indonesia Lao PDR Timor - Leste Vietnam -0.005 0 0.005 0.01 0.015 0.02 0.025 0.03 Source: Author calculation using CSES, IFLS, LECS, TLSLS, VARHS Note: This figure shows the coefficients for men and women taken from the regressions in equation 2.1 in Appendix B. Panel A refers to specification 2, and Panel B refers to specification 4. 54 Enterprising Women | To be or not to be an entrepreneur, that is the question Beyond the skill level required, other Infrastructure19 that can facilitate small- characteristics of available wage jobs, such scale entrepreneurship may make it more as paid leave, hours of employment, or feasible or attractive for women to open employer-supported childcare, can shape their own businesses rather than supporting women’s labor market decisions. Chowdhury et those of other family members. Labor market al. (2019) shows that women in Vietnam sort choices are not limited to only wage work into occupations with shorter hours and better and entrepreneurship. For example, 4 percent, leave policies, which may enable them to better 24 percent, 33 percent, 31 percent, and 19 balance market and domestic work. Similarly, percent of women engage in unpaid family Thébaud (2015) finds that in 24 high-income work supporting family farms or businesses in countries, women are less likely to engage in Cambodia, Indonesia, Lao PDR, Timor-Leste, necessity entrepreneurship when there are and Vietnam, respectively.20 New analysis policies that can reduce work-family conflicts, for this report shows that access to markets such as paid leave, publicly subsidized is linked with a higher level of women’s childcare, and part-time employment. If the engagement in entrepreneurship and to a lower enabling environment does not mandate paid level of their involvement in unpaid family leave or subsidized childcare and the local labor work, even when accounting for women’s time- market does not have available jobs with these invariant preferences and characteristics.21 types of characteristics, some women may turn to self-employment instead. 55 Enterprising Women | To be or not to be an entrepreneur, that is the question 2.1.3 Access to capital more likely to be entrepreneurs in Cambodia, although this relationship is stronger for men Access to capital is positively linked to than women (Figure 8: Panel B).25 both men and women’s participation in entrepreneurship. Nevertheless, it is not possible to distinguish whether individuals with greater access to capital open businesses or whether entrepreneurs have a greater demand for and use of financial services. There are three aspects of access to capital captured in the different surveys used for this analysis: loans, availability of assets that can be used as collateral, and savings.22 The use of loans and savings is not only influenced by the availability of formal and informal financial services, but also by an individual’s financial literacy and knowledge of how to use and access available services. Because it is not possible to disentangle the relative contributions of availability and financial literacy, additional analysis is needed to determine how binding each constraint is and whether interventions targeting both access and knowledge are needed. Women in Cambodia, Timor-Leste and Vietnam are more likely to engage in entrepreneurship when they are in households that have received loans.23 Although there is also a positive and statistically significant relationship between being in a household that received loans and engaging in entrepreneurship for men in Cambodia and Timor-Leste, in Timor- Leste the relationship is more than three times stronger for women than men (Figure 8: Panel A). Even when accounting for entrepreneurs’ unobserved, time-invariant characteristics and preferences in Vietnam, there is a positive relationship between receiving loans and entrepreneurship for women but not men.24 Linked with access to loans, individuals with assets that can be used as collateral are 56 Enterprising Women | To be or not to be an entrepreneur, that is the question Figure 8 Access to capital is positively associated with engagement in entrepreneurship Panel A. Conditional correlation between loans and entrepreneurship Panel for working-age A. Conditional correlation men between loans and women and entrepreneurship for working - aged men and women Men Women Cambodia Timor -Leste Vietnam -0.01 0.01 0.03 0.05 0.07 0.09 0.11 0.13 0.15 Panel B. Conditional correlation between having assets that can be used as Panel B. Conditional correlation between having assets that can be used as collateral and entrepreneurship collateral and for working-age entrepreneurship men for working - aged and men women (Cambodia) and women (Cambodia) Men Women -0.05 0 0.05 0.1 0.15 0.2 Panel C. Conditional correlation between savings and entrepreneurship Panel for working-age C. Conditional correlation men between savings and and women entrepreneurship for working - aged men and women Men Women Cambodia Timor -Leste Vietnam -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 Source: Author calculations using CSES, IFLS, TLSLS, VARHS Note: This figure shows the coefficients for men and women taken from linear probability models as defined in equation 2.1 in Appendix B for Cambodia and Timor-Leste and from linear probability models with individual fixed effects as defined in equation 2.2 in Appendix B for Indonesia and Vietnam. All panels refer to regression specification 2, showing the correlation between capital and the likelihood of engaging in entrepreneurship among the working-age population. 57 Enterprising Women | To be or not to be an entrepreneur, that is the question Participation in Aligned with global evidence,26 savings can provide arisans may enable an important source of seed capital for female Indonesian women microentrepreneurs in Southeast Asia and support their ability to open their own businesses. The majority to save enough of female microentrepreneurs in Indonesia use savings funds to engage in to finance their start-up capital, and a higher share of their own activity female microentrepreneurs than men rely on savings to rather than support start their businesses. Moreover, Indonesian women who the activities participate in rotating savings and credit associations, of other family known as arisans, are more likely to be self-employed members. than to work as an unpaid family worker in a family farm or business, even after taking time-invariant preferences and characteristics into consideration (Figure 8: Panel C). 27 Participation in arisans may enable Indonesian women to save enough funds to engage in their own activity rather than support the activities of other family members. Similarly, in Vietnam, both men and women who have informal savings are more likely to engage in entrepreneurship. Interestingly, formal savings in Vietnam are not positively associated with entrepreneurship when observing how changes in an individual’s savings over time is linked with their labor market outcomes (Figure 8: Panel C).28 However, if relatively few individuals begin saving through formal channels, it would be more difficult to technically capture such a relationship. Further evidence of the importance of the availability of family resources to fund seed capital for entrepreneurs in Vietnam comes from the fact that men and women in wealthier households (proxied as those with higher food consumption) are more likely to be entrepreneurs.29 58 Enterprising Women | To be or not to be an entrepreneur, that is the question 2.2 What characteristics are linked with women’s entry into SME ownership? There is evidence that most entrepreneurs who women’s firms in Timor-Leste. This trend is successfully run an SME directly opened an not limited to female SME owners, as male- SME rather than scaling up a microbusiness. owned SMEs also overwhelmingly had five or However, available data do not provide more employees when they began operations. information on the number of microbusinesses In many countries, firms that were considered that attempt to grow their businesses into microbusinesses at start-up were just below an SME. Figure 9 shows the vast majority of the threshold for the definition of an SME. For female-owned SMEs were already SMEs at example, in Cambodia, 23 percent of women- start-up, ranging from 55 percent of women- owned firms had four employees when they owned firms in Cambodia to 92 percent of opened.30 Figure 9 Most SMEs had five or more employees at start-up % of SMEs who had five or more employees at start-up % of SMEs who had 5 or more employees at start - up 100% 93% 92% 90% 84% 82% 79% 80% 73% 73% 71% 70% 61% 60% 55% 50% 40% 30% 20% 10% 0% Cambodia Indonesia Lao PDR Timor-Leste Vietnam Male Women Source: Author calculations using World Bank Enterprise Surveys, recent years 59 Enterprising Women | To be or not to be an entrepreneur, that is the question Entrepreneurs who are able to open larger male SME owners are more likely than male scale operations likely have different microbusiness owners to have received loans motivation and/or characteristics than those and are more leveraged. Taken together, this who either choose to open microbusinesses suggests that compared to microbusiness or have no other choice. Indeed, analysis from owners, both male and female SME owners are a panel dataset in Vietnam shows that women more growth-oriented, with plans to develop who own microbusinesses have different their businesses that require more capital. characteristics than those owning SMEs.31 Microbusiness owners may be less likely to However, in many cases, there are similar have concrete plans to develop their businesses patterns for men, suggesting that some factors if they opened a business due to the lack of linked with owning larger businesses are not a wage job33 or if they prefer to remain small necessarily specific to gender. Nonetheless, if to better balance running a business with women face specific disadvantages in terms of domestic duties.34 access to collateral or levels of human capital, Nevertheless, there seems to be a financing a gender focus in policies to help entrepreneurs gap for both male and female SME owners. transition to running larger businesses would be Only one-third of female SME owners have necessary. applied for formal loans, and over half use Growth-oriented enterprises are firms that informal loans to meet their financing needs. have a strong desire and capacity to grow over Although higher than for women, these rates time, whereas necessity entrepreneurs refer are also fairly low for male SME owners: 40 to entrepreneurs who have been pushed into percent of male SME owners applied for formal self-employment and are running a business loans and 58 percent use informal loans for as an alternative to formal employment. The their businesses (Figure 10: Panel A). Moreover, size of the business is not synonymous with 22 percent of female SME owners and 24 whether the entrepreneur is growth-oriented or percent of male SME owners report that capital necessity-driven. Microentrepreneurs comprise is the main factor constraining their ability to both necessity and growth-oriented segments, implement new projects, and 32 percent and whereas there are also both necessity and 36 percent of male and female SME owners, growth-oriented SME owners.32 respectively, report that capital is the main constraint to their business growth. Although both microentrepreneurs and SME owners can be driven by either growth or In addition to capital, labor issues also seem to necessity, analysis for this report suggests affect SMEs more than microbusinesses with that male- and female-owned SMEs are SME owners reporting more labor turnover more growth-oriented than microbusinesses and more difficulties in hiring workers. The in Vietnam. While female SME owners in differences between microbusinesses and Vietnam are more likely to cite capital as SMEs are larger for women than men (Figure a main constraint for new projects and 10: Panel B). growth, they are also more likely than female Although both male- and female-owned SMEs microbusiness owners to have applied for are more likely to have road and rail access formal credit, borrowed informally, and are than microbusinesses, the correlation between more leveraged (Figure 10: Panel A). Similarly, infrastructure and firm size is stronger 60 Enterprising Women | To be or not to be an entrepreneur, that is the question for women than men.35 This suggests that and social insurance for employees than infrastructure is a stronger enabling factor for microbusiness owners (Figure 10: Panel D). women who would like to open an SME than As with infrastructure, it is not clear whether for men. In Vietnam, there is a relationship knowledge and business practices are what between infrastructure, including road and rail enable entrepreneurs to grow their businesses access, and business size (Figure 10: Panel C). into SMEs or whether it reflects the needs and However, it is not clear whether the presence lessons learned in running a larger operation. of infrastructure enables entrepreneurs to open There is suggestive evidence that knowledge larger firms, or whether those who decide to and skills facilitate SME ownership for both open an SME gravitate toward locations with men and women. SME owners are more likely better infrastructure. to have completed higher secondary or college There is also a positive relationship between education than microentrepreneurs, who are knowledge of laws and regulations and more likely to have a lower-secondary education use of good business practices and SME or technical education (Figure 10: Panel E). ownership for both men and women. Both While it is possible that SME owners may go male and female SME-owners have greater back to school to further develop their skills, knowledge of laws and regulations, are it is more likely that the education level they more likely to be formal, to keep and audit achieved earlier in life has prepared them to run accounting books, innovate, and provide health more sizeable businesses. 61 Enterprising Women | To be or not to be an entrepreneur, that is the question Figure 10 SMEs in Vietnam have different characteristics than microbusinesses Panel A. Percentage of firms who Panel D. Percentage of firms applied for or received loans Panel A: Percentage of firms who applied for or using business practices Panel D: Percentage of firms using business received loans practices 70% 60% 55% 52% 58% 60% 55% 50% 50% 45% 40% 36% 42% 34% 40% 32% 40% 34% 30% 27% 27% 24% 30% 19% 18% 22% 20% 20% 17% 10% 5% 7% 6% 3% 4% 4% 2% 3% 1% 1% 10% 0% 0% Keeps accounting Accounting books Introduced new Firm pays social Firm pays health Applied for formal loans Borrowed informally books audited technology insurance insurance % of men's microbusinesses % of women's microbusinesses % of men's microbusinesses % of women's microbusinesses % of men's SMEs % of women's SMEs % of men's SMEs % of women's SMEs Panel B. Percentage of firms with labor Panel E. Percentage of firms with challenges different levels of owner’s education Panel E: Percentage of firms with different levels of Panel B: Percentage of firms with labor challenges owner's education 20% 19% 18% 80% 18% 68% 65% 70% 16% 60% 14% 49% 11% 50% 42% 12% 40% 39% 9% 40% 10% 8% 30% 25% 23% 23% 25% 20% 6% 4% 4% 10% 6% 5% 4% 3% 2% 2% 0% Completed lower secondary Completed higher secondary Completed college (technical 0% (general educ) (general educ) educ) Labor turnover Difficulties in hiring workers % of men's microbusinesses % of women's microbusinesses % of men's microbusinesses % of women's microbusinesses % of men's SMEs % of women's SMEs % of men's SMEs % of women's SMEs Panel C. Percentage of firms with Panel F. Average share of individuals in access to infrastructure the owner’s network who are women Panel F: Average share of individuals in the owner's Panel C: Percentage of firms with access to infrastructure network who are women 86% 45% 42% 90% 82% 40% 80% 72% 69% 35% 70% 31% 30% 28% 60% 24% 48% 51% 25% 50% 41% 38% 20% 40% 15% 30% 10% 20% 5% 10% 0% 0% Women in network (ratio) Access to road Access to rail % of men's microbusinesses % of women's microbusinesses % of men's microbusinesses % of women's microbusinesses % of men's SMEs % of women's SMEs % of men's SMEs % of women's SMEs Source: Author calculations using the VSMES 62 Enterprising Women | To be or not to be an entrepreneur, that is the question Finally, networks may play a role in Whereas 42 percent of individuals in female supporting entrepreneurs to grow their microentrepreneurs’ networks are women, businesses, as both men and women who only 31 percent of those in female SME own SMEs have larger networks than owners’ networks are women (Figure 10: Panel those who own microbusinesses. However, F). Networks can be important for sharing network composition may be more salient information about customers or suppliers,37 for women than men. Both women and connecting with potential business partners,38 men who own SMEs have a larger share of improving production technologies,39 or men in their network than those who own learning about investment opportunities.40 microbusinesses,36 but the differences between Male role-models or mentors have also been SME and microbusiness owners is particularly shown to help female entrepreneurs break into stark for women. traditionally male-dominated sectors.41 63 Enterprising Women | To be or not to be an entrepreneur, that is the question Endnotes 1  Global Entrepreneurship Monitor 2019-2020 26  upas and Robinson, 2013; Campos et al., 2015 ; Suri and D 2  Manolova et al., 2017 Jack, 2016 ; Buvinic et al., 2020 3  Please see equation 2.1 in Appendix B for more technical 27 This relationship is stronger for women than men using details. OLS regressions (Model 2.1 in Appendix B) but is similar for 4  Please see equation 2.2 in Appendix B for more technical women and men when including individual fixed effects to details. control for time invariant preferences and characteristics 5  Please see Equation 2.3 in Appendix B for more technical (Model 2.2 in Appendix B). details. The model used is a fixed effects model, as described in 28  6 OECD, 2017 Model 2.2 in Appendix B. 7  International Labour Organisation, 2018: 56 29 This holds true for both equations 2.1 and 2.2 in Appendix 8 In Asia and the Pacific, 52.6% of the 770 million inactive  B. women are out of the labor force due to unpaid care work Smaller firms are more likely to have started as micro firms 30  responsibilities (UNESCAP, 2019: p.18). and grown in size. However, even considering firms that 9 Bahramitash & Kazemipour, 2011; Chen, 2001; Manning,  have between 5 and 10 employees, the majority of firms 1998; Vanek, 2013; Campana, Giménez-Nadal, Molina, 2020; had at least 5 employees at start-up. Of firms with between Bayudan-Dacuycuy 2019 5 and 10 employees, 55% in Cambodia, 70% in Indonesia, 10 This result comes from an event study analysis, as 56% in Lao PDR, 86% in Timor-Leste, and 74% in Vietnam described in Model 2.3 in Appendix B. This analysis was had at least 5 employees at start-up. only possible in Indonesia due to data limitations. This 31 Due to data constraints, this analysis was not possible for type of event study requires both panel data and detailed the other countries studied. pregnancy histories. 32 World Bank, 2016b Giminez-Nadal, Molina, Ortega, 2012; Lim, 2019 11  33 See Section 2.1.2 for more details. Buttner and Moore, 1997; Cliff, 1998; Loscocco and Smith- 12  34 See Section 2.1.1 for more details. Hunter, 2004; Buding 2006; Morris et al., 2006; Loscocco 35 It is possible that infrastructure is correlated with other and Bird, 2012 factors such as higher levels of education; however, 13 Thébaud, 2015 the relationship remains when controlling for the 14 Bosma and Kelley, 2018 characteristics of the owner. 15 World Bank, 2012 Defined as the number of people the owner has regular 36  16 World Bank, 2012 contact with at least once every three months, which is 17 Miller et al., 2019 useful for business operations Moss-Racusin et al., 2012; Reuben et al., 2014 ; Azmat and 18  McMillan and Woodruff, 1999; Greif, 1993 37  Petrongolo, 2014 38 Casella and Rauch, 2002 19 Infrastructure variables examined in this report include: Parente and Prescott, 1994; Conley and Udry 2001, 2010 39  access to roads, access to rail, availability of a bus stop, 40 Patnam, 2013 availability of non-motor transportation, and availability of a Campos et al., 2015; Cucagna et al, 2020 41  public marketplace. 20 World Development Indicators 2019 Result from a model including individual fixed effects, as 21  described in equation 2.2 in Appendix B. The only other country for which we have information on infrastructure that could enable business is in Timor-Leste, where there is a significant and positive relationship between proximity to vehicle roads and women’s participation in entrepreneurship; however, the magnitude of that relationship is small. 22 Each survey only contains information on one or two of the aspects discussed. Both significant and insignificant results are discussed, so omission reflects differences in data availability. Results that only discuss certain types of finance, such as loans or savings, are only a reflection of data constraints and not of the importance of such financing sources in the countries studied. 23 In Cambodia, the question focuses on any type of loans. In Timor-Leste, the question focuses on loans from institutions. In Vietnam, the question asks about any loans received in the past 2 years. 24 This result comes from a regression model that uses individual fixed effects, as described in Model 2.2 in Appendix B. 25 The only dataset that explicitly asks about assets that can be used as collateral is Cambodia. Owning land is not statistically significantly associated with being an entrepreneur for men or women in Lao PDR or Timor- Leste; however, it is unclear whether the owned land meets requirements for being used as collateral (for example, formal documentation). 64 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Section 3: What contributes to the gender gap in performance of microenterprises? Section 1 showed sizeable gender gaps in the performance of microenterprises in Southeast Asia, and Section 2 demonstrated that gender-related constraints shape the profiles and motivation of female microentrepreneurs. Women’s differential selection into microentrepreneurship can contribute to observed gender gaps in the performance of microbusinesses. For example, if women open microbusinesses due to a lack of appropriate wage work opportunities, they may be less driven to grow their businesses and instead invest more time in seeking other employment opportunities. Nevertheless, gender differences in the choice to open a business does not explain all of the gap in microenterprise performance. Gender gaps in microenterprise performance remain statistically significant when comparing men and women with similar characteristics. 65 Enterprising Women | What contributes to the gender gap in performance of microenterprises? This section explores how women’s lower may have an idea to expand her business but levels of inputs correlate with the gender may not be able to act on this idea without the gap in microenterprise performance. permission of her husband. Gender norms may Compared to male microentrepreneurs, female also constrain the performance of women›s microentrepreneurs have lower levels of hired microbusinesses beyond their influence labor, own labor, skills, and capital. Gender on entrepreneurial inputs. For example, if differences in inputs are associated with the customers perceive women’s businesses gender gap in business performance. These as less reliable or female managers as less gender differences in inputs are shaped by capable, women may face more obstacles gender norms, market failures, and legal to growing their market share and accessing distortions. new market opportunities. Due to data constraints, this section focuses on how gender Factors other than entrepreneurial inputs differences in inputs relate to the gender gap may also contribute to gender gaps in in microenterprise performance. Nevertheless, microenterprise performance; however, because gender gaps in business performance available data do not enable empirical analysis persist even after comparing men and women of these constraints. Women with lower with similar personal, household, and business levels of agency are less able to set objectives characteristics, it is likely that these other aligned with their goals and values and to work unobserved factors are contributing to gender toward these objectives, which can impede differences in performance. Future research their ability to run their businesses as they should explore this topic in more detail. see fit. For example, a female entrepreneur 66 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Box 8 How does this report analyze factors limiting the performance of women’s microbusinesses? We test the association between different are possible. First, we use Oaxaca-Blinder characteristics of interest and business decompositions, a technique that breaks down performance using Ordinary Least Squares (OLS) the parts of an observed difference in outcomes regressions and interact the manager’s gender to understand which relationships are most with the characteristics to assess whether this strongly associated with the gap. This technique association is different for male and female sheds light on the extent to which gender gaps entrepreneurs.1 Where panel data is available in in business performance are due to men and Indonesia and Vietnam, we also run these models women’s different levels of endowments, the using individual fixed effects that take time extent to which they are due to differences in the invariant characteristics that cannot be observed relationship between endowments and business into account.2 In addition, for all countries performance, and the extent to which they are due included in the analysis, we use stepwise to the interaction between endowment levels and regressions that examine how the gender gap relationships with performance.4 Second, we use in firm performance changes after accounting a Cobb-Douglas production function to examine for a specific characteristic.3 The results of this the returns to factors of production, including analysis do not reveal causal relationships. capital, land, and labor, for men and women.5 The However, they show interesting patterns about results of these additional techniques cannot factors correlated with business performance. prove causality, but provide additional nuance for understanding what factors are linked with Due to the detailed nature of the data in Vietnam, the gender gap in business performance. two additional descriptive analytical techniques 3.1 Access to labor 3.1.1 Hired labor in Timor-Leste to 53 percent in Vietnam (Figure In all countries studied, female 11: Panel A). Considering only businesses that microentrepreneurs have fewer paid workers have any paid workers, women still have fewer than men (Figure 11: Panel B). The percentage paid workers than men in Indonesia and Lao of female-owned businesses that have any PDR. In Indonesia, men have 3.7 paid workers paid workers ranges from 2 percent in Timor- on average compared to women’s 2.1 paid Leste to 37 percent in Vietnam, whereas the workers, and in Lao PDR men have 1.8 workers percentage of male-owned businesses that compared to women’s 1.4 paid workers (Figure have any paid workers ranges from 6 percent 11: Panel C). 67 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Figure 11 Women’s microbusinesses have fewer paid employees than men’s in all countries, but gender differences in unpaid workers vary by country Panel A. Share of microbusinesses with any workers Panel A: Share of microbusinesses with any workers 93% 97%97% 100% 92% 80% 55% 53% 60% 42% 42% 34% 37% 40% 31% 26% 22% 20% 10% 11% 5% 6% 2% 4% 7% 0% Paid Unpaid Paid Unpaid Paid Unpaid Paid Unpaid Paid Unpaid Cambodia Indonesia Lao PDR Timor-Leste Vietnam Male Female Panel B. Average number of workers in microbusinesses Panel B: Average number of workers in microbusinesses 2 1.75 1.81 1.49 1.5 1.36 0.96 0.97 1 0.75 0.7 0.58 0.63 0.55 0.47 0.5 0.31 0.24 0.05 0.09 0 Unpaid Paid Unpaid Paid Unpaid Unpaid Paid Unpaid Cambodia Indonesia Lao PDR Timor-Leste Vietnam Male Female Panel C: Panel C. Average Average number number of workers of workers in microbusinesses, in microbusinesses, conditional on having workers conditional on having workers 4 3.74 3.5 3 2.5 2.14 1.78 1.83 1.87 1.81 1.88 2 1.41 1.48 1.42 1.62 1.46 1.37 1.36 1.36 1.19 1.5 1 0.5 0 Unpaid Paid Unpaid Paid Unpaid Unpaid Paid Unpaid Cambodia Indonesia Lao PDR Timor-Leste Vietnam Male Female Source: Author calculations using CSES, IFLS, LECS, TLSLS, VSMES Note: The number of paid workers is not available for Cambodia or Timor-Leste; however, whether or not there are paid workers is approximated in these countries by whether or not the firm has salary costs. The number of unpaid workers in Timor-Leste may be an underestimate because the data do not distinguish which household member is primarily responsible for the business, so analysis was limited to firms where all family workers were of the same gender. 68 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Moreover, female microbusinesses in be benefits to expanding the size of their Vietnam are operating below their optimal businesses. Because only female-owned size. The Cobb-Douglas production function businesses have increasing returns to labor, shows the marginal return to labor for the it is likely that gender-related constraints or value of production is 1.07 for female-owned preferences affect women’s ability and/or microbusinesses and 0.34 for male owned- desire to expand their businesses. microbusinesses (Figure 11).6 A marginal More research is needed to understand return to labor that is greater than one signifies why women have fewer workers than men; that there are increasing returns to scale for however, either gender-specific constraints labor: in other words, doubling the number of or preferences may contribute to this gap. On workers would more than double the value of the one hand, women might have difficulties production. In absence of constraints, most hiring additional labor if they lack the necessary businesses would continue to hire workers financing or if negative stereotypes about until constant returns to scale have been women’s ability as managers makes it more reached, and the efficiency of the firm has been difficult for them to hire and retain male maximized. employees.7 On the other hand, women may Increasing returns to labor for female- prefer to maintain a smaller scale to better owned microbusinesses in Vietnam suggest balance their business operations with they either face challenges hiring qualified household responsibilities.8 workers or they make choices to remain in smaller-scale operations, even if there could Figure 12 Women’s microbusinesses have increasing returns to labor, whereas men’s microbusinesses have decreasing returns to labor in Vietnam Marginal returns to labor for value of production in Vietnam Marginal returns to labor for value of production in Vietnam Male Female 1.2 1.07 1 0.8 0.6 0.4 0.34 0.2 0 Full-time workers Source: Author calculations using VSMES Note: The marginal productivity of labor is calculated using the Cobb Douglas production function as described in Model 3.5 in Appendix B. Gender differences are statistically significant at the 5 percent level. 69 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Women’s lower levels of paid labor contribute Beyond only the number of workers, challenges to the gender gap in business performance related to hiring and retaining workers in Southeast Asia. Unsurprisingly, the number partially explain gender gaps in microbusiness of paid workers is positively correlated with productivity in Vietnam. Labor turnover (the business performance in all countries,9 which share of a firm’s total employees who leave means that having fewer paid employees the company during a year) contributes to puts female entrepreneurs at a disadvantage. the gender gap in value-added.11 Moreover, Indeed, the size of the labor force in Vietnam accounting for gender differences in the explains 15 percent of the gender gap in number of workers, amount of employee profits.10 After accounting for women’s lower turnover, and reported difficulties in hiring levels of paid and unpaid workers, the gender workers shrinks the gender gap in profits of gap in business performance shrinks by 19 microenterprises by 11 percent in Vietnam percent, 10 percent, 21 percent, and 4 percent (Figure 13). in Cambodia, Indonesia, Lao PDR, and Timor- Leste, respectively (Figure 13). Figure 13 Accounting for differences in labor reduces the gender gap in microbusiness performance Changes in the gender gap in microbusiness Changes in the gender gap in microbusiness performance performance foraccounting after after accounting for gender differences gender in labor differences in labor Gap without controlling for workers Gap after controlling for workers 0% -5% -10% -10.1% -11.4% -15% -13.7% -20% -17.4% -19.6% -25% -24.3% -30% -35% -33.0% -34.2% -40% -36.6% -40.5% -45% Cambodia Indonesia Lao PDR Timor-Leste Vietnam Significant at 1% Significant at 5% Not statistically Significant Source: Author calculations using CSES, IFLS, LECS, TLSLS, VSMES Note: Results are from stepwise regressions as described in Model 3.3 in Appendix B. Labor-related variables include the number of unpaid workers in all countries, the number of paid workers in Indonesia, Lao PDR, and Vietnam, whether the firm has paid workers in Cambodia and Timor-Leste, and employee turnover and difficulties hiring workers in Vietnam. 70 Enterprising Women | What contributes to the gender gap in performance of microenterprises? In addition to having fewer paid workers, female 3.1.2 Own labor microentrepreneurs have a greater reliance on unpaid labor, which in turn can affect the When considering labor supply availability quality and reliability of their labor supply. for microenterprises, it is important to also A greater proportion of women’s workers in consider the amount of time the business Indonesia and Lao PDR are unpaid compared owner or manager spends working in the to men. Relying on unpaid labor may mean business. The vast majority of microbusinesses accepting the support of whoever is available, in all countries do not have any paid workers rather than finding individuals with the most (the percentage ranges from 74 percent in relevant qualifications and experience for Lao PDR to 97 percent in Timor-Leste), and the job. Moreover, unpaid workers may have in Indonesia and Cambodia the majority also less incentive to exert the highest levels of do not have any unpaid workers.13 In these effort and commitment to the job and may cases, the only source of labor comes from temporarily or permanently leave the job on the person responsible for the business. Even short notice if paid work becomes available. for businesses with paid or unpaid workers, Despite relying more on unpaid labor in terms microbusinesses do not have many workers of the proportion of workers, women in Lao PDR on average (Figure 11), which means that the also have fewer unpaid workers than men.12 entrepreneurs’ own labor still represents an If women have less bargaining power in the important source of labor. household or if their businesses are considered Women’s greater role in managing domestic as secondary activities, they may face more and care work limits the amount of time challenges in seeking the help of household that female entrepreneurs dedicate to their members for their businesses. Because businesses, and this can is associated with the number of unpaid workers is positively the gender gap in business performance. correlated with business performance in Lao In Cambodia, the more time women spend PDR, the gap in the number of unpaid workers fetching water during the dry season, the lower could contribute to gender gaps in business their profits, whereas there is not a statistically performance. significant relationship for men (Figure 14: Panel A). In addition, women in Cambodia who have access to a cooking source that requires less time to collect and ignite, such as gas or electric, or who have firewood or charcoal brought to their house have higher profits (Figure 14: Panel B). Similarly, women in Indonesia who have water inside their household and less time-consuming cooking sources also have higher profits (Figure 14: Panel C).14 Spending less time cooking and collecting fuel and water may enable women to spend more time on their businesses, thus boosting their productivity.15 71 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Figure 14 Time spent on domestic tasks is negatively associated with profits of female microentrepreneurs in Cambodia, and time-saving domestic infrastructure is associated with higher profits in Cambodia and Indonesia Panel A. Conditional correlation between time spent fetching water in Panel A: the dry season Conditional and correlation profits between time in Cambodia spent fetching water in the dry season and profits in Cambodia Men Women -0.014 -0.012 -0.01 -0.008 -0.006 -0.004 -0.002 0 0.002 0.004 0.006 Panel B. Conditional correlation between time-saving domestic infrastructure Panel B: between in and profits Conditional correlation Cambodia time - savings domestic infrastructure and profits in Cambodia Men Women Easy cooking source Firewood or charcoal brought to the house 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Panel C. Conditional correlation between time-saving domestic Panel C:infrastructure between in and profits Conditional correlation Indonesia time - savings domestic infrastructure and profits in Indonesia Men Women Easy cooking source Water source in the household -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Source: Author calculation using CSES and IFLS Note: Regression coefficients come from Model 3.1, as described in Appendix B. 72 Enterprising Women | What contributes to the gender gap in performance of microenterprises? In Lao PDR, 34 percent of female between operating a business from home and microentrepreneurs identify primarily as business performance reflects the choices homemakers in at least one month over the or ability of more profitable entrepreneurs past year, and their revenues are 18 percent to move their businesses to a commercial lower than those who were primarily working location. Regardless of the direction of all year.16 Likewise, in Indonesia, women causality, if women are more likely than men whose primary activity is entrepreneurship to operate their businesses from home in have profits that are 28 percent higher order to align with expectations of combining than those who are not fully focused on business and domestic work rather than only their businesses, even when taking both an economic calculation, it can help explain observed and unobserved characteristics into gender differences in business performance. account.17,18 Though many women combine Indeed, women’s greater likelihood of working entrepreneurship with carrying out domestic from a residential location in Vietnam explains tasks, these examples indicate many women 9 percent of the gender gap in microbusiness identify primarily as homemakers and report profits.20 that their business is their secondary activity. It is likely women who identify their business as a secondary activity may face more time constraints that limit their own labor in the business. While operating the business from home can facilitate balancing market and domestic work, women who run their businesses from home are likely subject to more interruptions during the workday, limiting their labor supply and lowering their performance. In Lao PDR, Vietnam, and Indonesia, for example, women are more likely than men to operate their businesses from home (Figure 15), which is associated with lower business performance. Women who operate their businesses from home have revenues that are 24 percent lower in Lao PDR and profits that are 16 percent lower in Indonesia than those who have a separate business location. Although a negative trend between operating from a residential location and business performance also exists in Vietnam, it is not statistically significant (Figure 16).19 While home-based businesses may be less profitable due to interruptions during the workday or distance from customers, it is also possible that the negative correlation 73 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Figure 15 Female microentrepreneurs are more likely to operate their businesses from home Share of microbusinesses located in the residence Share of microbusinesses located in the residence 70% 63% 59% 60% 55% 50% 47% 40% 34% 36% 30% 20% 10% 0% Indonesia Lao PDR Vietnam Male Female Source: Author calculations using IFLS, LECS, VSMES Note: Differences between men and women are statistically significant at the 1 percent level. Figure 16 Operating a microbusiness from residential property is negatively associated with business performance Conditional correlation between operating the business out Conditional correlation between of the home business out of the home operating the performance and microbusiness and microbusiness performance Men Women Indonesia Lao PDR Vietnam -0.500 -0.400 -0.300 -0.200 -0.100 0.000 0.100 0.200 Source: Author calculations using IFLS, LECS, VSMES Note: Regression coefficients come from Model 3.1 in Appendix B. 74 Enterprising Women | What contributes to the gender gap in performance of microenterprises? The finding that domestic work limits women’s in self-employed income in the US21 and 22.5 business performance is aligned with evidence percent of the gender gap in self-employed from other countries and regions. Arráiz (2018) earnings in Germany.22 Combining childcare shows that time spent on household chores, with business activities can also lower women’s childcare, and care for the elderly explains 33 profits. Delecourt and Fitzpatrick (2020) percent of the gender gap in profits of micro document that female-owned businesses in and small businesses in Ecuador. Another Uganda where children are present earn 48 58.9 percent of the gender gap is explained percent lower profits than other female-owned by business size, entrepreneurial motivation, businesses that do not have children present. and sector of activity, which may also be Prices and product quality are similar among indirectly affected by the time available to these two groups of businesses, but women entrepreneurs. Similarly, studies show that time who simultaneously care for their children and spent on childcare, housework, and market operate their businesses are more likely than work explain 61 percent of gender differences those that do not to run out of stock.23 3.2 Skills 3.2.1 Formal education Nevertheless, controlling for education in Vietnam does not change the gender gap in As discussed in Section 2, both men and business performance (Figure 18).27 women with lower levels of formal education sort into microentrepreneurship. This can have important implications for business performance, as previous studies have shown that formal schooling is positively linked with business characteristics and performance.24 Business formation, operation, and management require skills that can be acquired through formal education and training.25 In all countries studied, female entrepreneurs have lower education levels than men (Figure 17),26 which contributes to the gender gap in performance of microenterprises. Indeed, taking gender gaps in education into account shrinks the gender gap in performance by 39 percent in Cambodia, 11 percent in Indonesia, 43 percent in Lao PDR, and 8 percent in Timor-Leste (Figure 18). After controlling for education, the gender gap in revenues is not statistically significant in Lao PDR. 75 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Figure 17 Female entrepreneurs have lower education levels than male entrepreneurs Average years of education of male and female entrepreneurs Average years of education of male and female entrepreneurs 9 8 7.7 7.1 7 5.8 6.0 6 5.6 5.0 5 4.7 4.5 4 3 2 1 0 Cambodia Indonesia Lao PDR Timor-Leste Men Women Source: Author calculations using CSES, IFLS, LECS, TLSLS Figure 18 After accounting for differences in education levels, the gender gap in microenterprise performance shrinks in all countries except Vietnam Changes in the gender gap in business performance after accounting Changes in the for gender gap ingender differences business in education performance after accounting for gender differences in education Gap without controlling for education Gap after controlling for education 0% -5% -10% -9.9% -11.4% -11.7% -15% -14.8% -20% -17.4% -25% -24.3% -30% -35% -31.5% -34.2% -36.0% -40% -40.5% -45% Cambodia Indonesia Lao PDR Timor-Leste Vietnam Significant at 1% Significant at 5% Not statistically Significant Source: Author calculations using CSES, IFLS, LECS, TLSLS Note: This graph shows the gender gap in business performance before and after controlling for education, according to the stepwise regressions described in Model 3.3 in Appendix B. 76 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Nevertheless, female microentrepreneurs for men’s businesses in Lao PDR and Timor- in some countries benefit more from formal Leste. Even in Cambodia where both men and education than their male counterparts. women’s education are positively related to Aligned with existing evidence,28 years of their business performance, the association education are positively correlated with the is stronger for women than men (Figure 19). performance of female microbusinesses in Evidence from other countries and regions all countries studied. However, contrary to also suggests that female entrepreneurs have expectations, there is no relationship between greater returns to their formal education than years of education and business performance male entrepreneurs.29 Figure 19 Education levels are positively associated with business performance of female microentrepreneurs Conditional correlation Conditional correlation between betweeneducation educationand business and performance business performance Men Women Cambodia Indonesia Lao PDR Timor-Leste -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 Source: Author calculations using CSES, IFLS, LECS, TLSLS Note: This graph shows the coefficients from regressions as described in Model 3.1 in Appendix B. 77 Enterprising Women | What contributes to the gender gap in performance of microenterprises? 3.2.2 Business skills counterparts,36 despite having fewer benefits or being treated differently than men who are In addition to formal education, evidence running formal businesses.37 from Vietnam shows that gender differences in business knowledge and use of good Digital skills and other technological business practices contribute to the gender innovation are important for business gap in microenterprise profits.30 Aligned with success, and the COVID-19 pandemic has evidence in the literature,31 knowledge of increased their relevance.38 Although the laws and regulations and implementation of results from Vietnam suggest that female- good business practices are correlated with owned microbusinesses have lower returns higher profits for both male and female-owned to introducing new technology than male- microenterprises.32 However, gender differences owned microbusinesses, in other countries, in formal education, business knowledge, women can reap greater benefits from new and adopting innovative business practices— technologies. A recent study in Indonesia namely levels of higher-secondary education, found that female entrepreneurs were more knowledge of laws and regulations, and likely to use and benefit from digitalization.39 introducing new product groups—explains part of the gender gap in profits of microenterprises. The profit gap would be even larger if female 3.2.3 Access to information microentrepreneurs were less likely to keep Networking can provide a range of benefits accounting books or to be formally registered.33 to entrepreneurs, including enhancing their In addition, gender differences in the skill development.40 It is positively linked relationship between these business skills to entrepreneurial development in general and performance contribute to the gender and, in particular, female entrepreneurship.41 gap in microenterprise profits in Vietnam. Entrepreneurs who have well-developed Female microentrepreneurs have a lower networks are more successful in their association between their business practices businesses, identify more viable opportunities, and performance than men, which contributes and have greater access to resources.42 to gender gaps in profits. Specifically, female New analysis for this report confirms that microenterprise owners who introduce networks are particularly important for the new technology get less of a boost to their profitability of female-owned microbusinesses businesses compared to men.34 This is despite in Vietnam. The number of individuals the strong relationship between intensive use in the owner’s network43 is positively of innovative technology and higher productivity associated with profits, and this relationship dividends for firms in Vietnam compared to is stronger for female entrepreneurs than firms with less-intensive technology use.35 for male entrepreneurs. For female-owned Differential knowledge about how to use the microenterprises, an additional person in the technology or to maximize its benefits may owner’s network is associated with 0.49 percent contribute to this gap. Nevertheless, female higher profits. This result does not appear microenterprise owners who formalize driven by an individual’s personality: comparing their businesses get more of a premium to the same firm over time, an additional person in that formalization compared to their male a female entrepreneur’s network is linked with 78 Enterprising Women | What contributes to the gender gap in performance of microenterprises? a 0.36 percent increase in profits.44 In contrast, work and other domestic work limits women’s an additional person in a male owner’s network time to engage in networking activities. is only correlated with a 0.15 percent increase Female entrepreneurs surveyed in Indonesia in profits, and this relationship disappears when and Malaysia underscored the importance of examining how changes in the owner’s network networking, however, many of them noted that over time correlates with a specific firm’s domestic work significantly restricted their time performance.45 to expand their business contacts, participate in formal associations and networks, and Social norms can make it more challenging for communicate with potential mentors.47 women to build and diversify their networks as These same time constraints can make it well as to participate in programs to enhance more challenging for women to participate their skills more broadly. For example, in in programs to build entrepreneurial skills, in Cambodia, it is considered inappropriate particular if entrepreneurship support programs for women to interact with male business are not designed with hours, location, and owners and government officials.46 In addition availability of support services like childcare to to gender norms related to mobility and facilitate women’s participation. interacting with men in the community, social norms related to women’s role in providing care 3.3 Access to capital Aligned with evidence from other regions,48 gender gaps exist at every level of financial female microentrepreneurs in Indonesia assets (Figure 21: Panel B). and Vietnam have lower levels of business Although existing evidence from Southeast assets than male microentrepreneurs. Female Asia shows that women are concentrated in microentrepreneurs in Indonesia had less sectors that are less capital intensive than than half of the amount of start-up capital those dominated by men,49 gender gaps in as men, and a gender gap in start-up capital asset levels remain after accounting for exists all along the distribution (Figure 20). In gendered sectoral segregation. Female Vietnam, female-owned microbusinesses have microentrepreneurs in Indonesia have start-up physical assets, including the value of land, capital that is much lower than that of men, buildings, equipment and machinery, that are even when comparing men and women within 18 percent lower than the physical assets of the same sector of activity. Similarly, gender male-owned microbusinesses. As shown in gaps in both physical and financial assets in Figure 21: Panel A, the gender gaps in physical Vietnam remain even after accounting for the asset levels are particularly large at the lower fact that men and women operate in different end of the distribution. In addition, female sectors of activity. Women’s lower levels of microentrepreneurs in Vietnam have financial capital compared to men operating in the same assets—nonphysical assets such as cash or sector of activity can hinder their ability to receivables (ex. money owed to the firm by grow their businesses, harness technology, and customers)—that are 39 percent lower than generate income. those of male microentrepreneurs, and these 79 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Figure 20 Women’s microbusinesses in Indonesia had lower start-up capital than men’s Distribution of start-up capital for microbusinesses in Indonesia Kernel density estimate: Log of start-up capital Distribution of start-up capital for microbusinesses in Indonesia Kernel density estimate: Log of start-up capital .2 .15 Density .1.05 0 0 5 10 15 20 Log of start-up capital Female Male kernel = epanechnikov, bandwidth = 0.3295 Source: IFLS Household and Community survey representing year 2000, 2007 and 2014 Source: Author calculations using IFLS Note: This graph shows the kernel density function, which depicts the underlying distribution of a continuous variable. In other words, it shows the extent to which women (as represented by the yellow line) or men (as represented by the blue line) in the sample are concentrated at different levels of start-up capital. Because the yellow line lies to the left of the blue line, the graph shows that women are concentrated among lower levels of start-up capital. Given that the entire curve (not just a portion of it) lies to the left of the blue line, it demonstrates that women have lower levels of start-up capital than men, regardless of whether compared to other women they have lower or higher levels of initial capital. Figure 21 Female-owned microenterprises in Vietnam have lower assets than men’s microenterprises Panel A. Distribution of microbusinesses’ Panel B. Distribution of microbusinesses’ physical assets in Vietnam, by gender financial assets in Vietnam, by gender Kernel Density Estimate Kernel Density Estimate (Land, Buildings, Equipment, Machinery) Panel A: Distribution of microbusinesses? physical assets in Vietnam (Cash, Receivables) Panel B: Distribution of microbusinesses? financial assets in Vietnam Kernel Density Estimate (Land, Buildings, Equipment, Machinery) Kernel Density Estimate (Cash, Receivables) .25 .4 .2 .3 .15 Density Density .2 .1 .1 .05 0 0 −5 0 5 10 −5 0 5 10 Log of physical assets Log of financial assets Female Female Male Male kernel = epanechnikov, bandwidth = 0.3409 kernel = epanechnikov, bandwidth = 0.2903 Source: Author calculations using VSMES Note: This graph shows kernel density functions, which depict the underlying distribution of a continuous variable. In other words, they show the extent to which women (as represented by the blue line) or men (as represented by the red line) in the sample are concentrated at different levels of physical capital (Panel A) or financial capital (Panel B). Because the blue line lies to the left of the red line, the graph shows that women are concentrated among lower levels of capital. For physical capital in Panel A, the gender differences are largest at the lower part of the distribution because the blue line lies to the left of the red line at lower levels, but is similar to the red line at higher levels of capital. For financial capital in Panel B, because the entire curve (not just a portion of it) lies to the left of the blue line, it demonstrates that women have lower levels of financial capital than men, regardless of whether compared to other women they have lower or higher levels of capital. 80 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Gender norms and discriminatory laws and Indonesia, the Philippines, and Vietnam, a institutions can limit women’s access to significant proportion of female entrepreneurs capital. In Indonesia and Timor-Leste, the law use informal social networks for capital does not prohibit discrimination in access to acquisition.56 In Cambodia, women also credit based on gender.50 However, even when typically raise capital through personal or family the law prohibits discrimination, social norms funds.57 Similarly, analysis conducted for this can inhibit women’s access to credit. For report shows that female microentrepreneurs example, in the Philippines, although women in Indonesia are more likely than men to rely have not needed spousal consent for applying on savings and on family members for their for credit since the passage of a 1992 law, this start-up capital, and less likely to receive law is not always enforced due to social norms formal financing. Although this may provide that make husbands responsible for capital- more flexibility and perhaps lower interest related transactions.51 rates, it also likely limits the amount of funding available to invest in the business. Discriminatory property laws also contribute to women’s limited access to loans because their Access to capital is essential for low ownership of assets negatively impacts the productivity of female-owned their ability to offer collateral when seeking microbusinesses, and gender gaps in access credit.52 Indonesian law does not grant equal to capital contribute to the gender gap in rights to inherit parental assets for sons and microbusiness performance.58 Although daughters, and female spouses do not have the women have lower levels of assets and capital same rights to inherit assets as male spouses.53 than men, the relationship between capital As with laws prohibiting discrimination, social and business performance is stronger for norms can also impede women from fully female microentrepreneurs than for men. exercising their property rights, even when they In both Cambodia and Timor-Leste, access are inscribed in the law. For example, despite to capital in terms of loans is linked with the 2001 Land Law in Cambodia that sets out higher profits for female-owned businesses, equal land rights, women’s enjoyment of their but there is not a statistically significant rights is lessened by customary norms that relationship for male-owned businesses in regard men as the heads of household who are Timor-Leste (Figure 22: Panel A). Yet, female responsible for land.54 In Vietnam, many banks entrepreneurs in both countries are less likely do not view female entrepreneurs as a priority than male entrepreneurs to live in households due to perceptions that they lack knowledge of that receive loans. Aligned with the results financial products and are unable to balance on receiving loans, investing capital in the professional and family responsibilities.55 business in Cambodia is positively associated with women’s profits but not men’s (Figure 22: Challenges accessing credit due to Panel D), but women are less likely to have discriminatory laws or norms affecting their invested capital. Owning the business location implementation can not only limit women’s reflects both higher levels of capital as well as ability to access financing for their business, ownership of an asset that may help secure a but also the sources of financing they loan. In both Cambodia and Lao PDR, owning use and the conditions of that financing. the building for the business is associated According to a recent study conducted in with better business performance for women’s 81 Enterprising Women | What contributes to the gender gap in performance of microenterprises? businesses, but not men’s (Figure 22: Panel relationship between receiving transfers and B). Receipt of government transfers may also business performance for either men or women enable entrepreneurs to invest in their business. (Figure 22: Panel C). For microbusinesses in In Timor-Leste, receiving government transfers Vietnam, women’s lower levels of capital, land, is correlated with higher business performance and labor jointly explain all the gender gap in only for female entrepreneurs; however, in value added and value of production.59 Cambodia, there is not a statistically significant Figure 22 Access to capital is positively associated with women’s microbusiness performance A. Conditional PanelPanel correlationbetween correlation A. Conditional loans between loans and and business business performance performance Men Women Cambodia Timor-Leste -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Panelcorrelation Panel B. Conditional B. Conditional between owning correlation between building owning for for building business and business and business performance business performance Men Women Cambodia Lao PDR -0.1 0 0.1 0.2 0.3 0.4 0.5 82 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Panel C. Conditional correlation between receiving transfers Conditional Panel C. from correlation between the government receiving and business transfers from the performance government and business performance Men Women Cambodia Timor-Leste -400 -300 -200 -100 0 100 200 300 400 500 Panel D. Conditional correlation between investing capital in the Panel D. Conditional correlation between investing capital in the business and business performance in Cambodia business and business performance in Cambodia Men Women Men Women -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 Source: Author calculations using CSES, LECS, TLSLS Note: These graphs show the coefficients from regressions as specified in Model 3.1 in Appendix B. Evidence from Vietnam suggests the type of higher return than men to financial capital, capital constraints female microentrepreneurs which includes cash and receivables such as face are different from those men face. Women money owned by customers. This suggests that are more constrained when it comes to financial the relative importance of different financial capital, whereas men are more constrained products for male and female entrepreneurs are when it comes to physical capital. A Cobb- likely different, with male entrepreneurs needing Douglas production function estimates the longer maturity loans that can enable purchases returns to different factors of production. Firms of machinery or equipment, whereas women with a higher return to a factor of production may need products tailored to working capital are considered relatively more constrained in needs. The formal financial system seems to be that factor than those who have comparatively reaching female entrepreneurs in Vietnam less, lower returns. As shown in Figure 23, men as female microentrepreneurs in Vietnam are have a higher return than women to physical less likely than men to have applied for formal capital, which includes buildings, machinery, loans and are less leveraged, with a lower debt to and equipment. Conversely, women have a equity ratio. 83 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Figure 23 Female-owned microbusinesses have higher returns to financial capital and lower returns to capital stock than male- owned microbusinesses in Vietnam Marginal Marginal returns returns to capital to capital for value for valuefor of production production for in Vietnam of microbusinesses microbusinesses in Vietnam Male Female 0.2 0.18 0.15 0.11 0.1 0.05 0.01 -0.02 0 -0.05 Capital stock Financial capital Source: Author calculations using VSMES Note: Marginal returns are calculated using a Cobb-Douglas production function, as described in Model 3.5 of Appendix B. Gender differences are statistically significant at the 1 percent level. Box 9 What is the role of infrastructure in supporting the productivity of women’s businesses? Multiple studies that show that infrastructure a study of cross border traders in Cambodia and is positively linked to the development Lao PDR shows on average and per crossing, of entrepreneurship,60 including female female traders pay more than double what entrepreneurship.61 Analysis undertaken for this male traders pay in transportation costs, likely report supports the relevance of developing because time-constrained women cope with different types of infrastructure and suggests weak transportation infrastructure by hiring that doing so may have particularly strong effects transporters or marketing smaller quantities.64 on female entrepreneurs. The presence of a While necessary, improved infrastructure alone marketplace in the community is positively related may not be sufficient to boosting women’s with female microbusiness profits in Indonesia, businesses. Beyond infrastructure constraints, even taking unobserved characteristics of the social norms that limit women’s physical mobility entrepreneur into account, whereas the same and interactions with strangers can block women’s is not true for men.62 Similarly, in the case of access to markets.65 Although new rural roads Vietnam, access to rail is found to be significantly in Vietnam have been shown to contribute to correlated with profits only for women-owned the development of local markets and improved microenterprises. Furthermore, in Timor-Leste, opportunities for self-employment,66 men and access to roads is found to be significantly and women may not benefit equally. Both male-headed more strongly correlated with women’s profits and female-headed agricultural households compared to men.63 This might be explained by benefitted from increased agricultural trade after the fact that better roads and rail access both rural road improvement in Vietnam; however, only increase access to consumer markets and lower male-headed households increased production. the associated transportation costs and travel Capital and labor constraints seemed to prevent time. Given the greater demands on women’s female-headed households from being able to time, improved transportation infrastructure may fully benefit from the improved infrastructure.67 particularly benefit female entrepreneurs. Indeed, 84 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Box 10 How is sector choice linked with the productivity of women’s businesses? As in many countries around the world,68 analysis for this report shows that women and men in Southeast Asia are concentrated in different sectors of activity. Similar to findings from other regions,69 women who operate in male dominated sectors70 have profits that are 41 percent, 55 percent, and 70 percent higher than those operating in female concentrated sectors in Indonesia, Lao PDR, and Vietnam, respectively (Figure 24). The opposite is true in Cambodia: women operating in female concentrated sectors outperform those in male- dominated sectors, likely due to Cambodia’s economic, social, and political history.71 Figure 24 Female entrepreneurs in male-dominated sectors have higher profits than female entrepreneurs in female- concentrated sectors in all countries except Cambodia Percent difference Percent in business difference performance in business of women performance in male-dominated of women in male - sectors compared dominated to women sectors in female-concentrated compared to women in femalesectors - concentrated sectors 80% 60% 40% 20% 0% -20% -40% Cambodia Indonesia Lao PDR Vietnam Source: Author calculations using CSES, IFLS, LECS, VARHS Note: See Model 3.6 in Appendix B for details. Nevertheless, contrary to findings from other regions,72 sector of activity explains little, if any, of the gender gap in microenterprise profits in Indonesia, Lao PDR, and Vietnam. Moreover, in Cambodia, a gender gap in business performance only exists when comparing men and women within the same sector, and the gender gap in performance increases when controlling for sector in Lao PDR and Vietnam (Figure 3). While sector choice may help women operating in male- dominated sectors outperform their peers in female-concentrated sectors in some countries, it explains little, if any, of the gender gap in profits. In addition, male-dominated sectors are not systematically more profitable than female-concentrated sectors: only in Vietnam do men in male-dominated sectors have higher profits than men in female-concentrated ones. 85 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Box 11 Gendered impacts of COVID-19 on micro, small, and medium enterprises (MSMEs) in Southeast Asia The COVID-19 pandemic has posed multiple experienced a 49 percent reduction in their sales.74 challenges for MSMEs. Many firms have either had Although the COVID-19 shock is not gender- to temporarily or permanently close their doors specific, pre-existing gender inequalities and or modify their operations due to government gender norms have left female entrepreneurs containment measures, health concerns among more vulnerable to the effects of the pandemic employees, or changes in customer demand. than their male counterparts. As shown in Figure Globally, less than 30 percent of firms remained 25, female microentrepreneurs in Indonesia are open during the most intense periods of restricted concentrated in certain industries, including mobility related to the pandemic, and while hotels and restaurants and light manufacturing, many re-opened, approximately 25 percent of both of which have been among those hardest firms remained temporarily or permanently hit by government restrictions and shrinking closed after a peak in the COVID-19 outbreak.73 demand. While transportation and construction, A study examining the impacts of COVID-19 which are heavily male-dominated, have on businesses in 51 countries shows that 84 also been greatly affected by COVID-19, they percent of firms reported a reduction in sales in represent a very small proportion of the overall the 30 days preceding the interview compared population of microenterprises in Indonesia. to the same period in 2019, and the average firm Figure 25 COVID-19 lowered growth in sectors with higher concentrations of female entrepreneurs Impact of COVID-19 on male- and female-dominated sectors in Indonesia % sectoral growth from Q2 to Q4 2020 compared to the 10% Gender Parity Agriculture 5% Financial Education 0% same period in 2019 -5% Construction -10% Manufacturing Services -15% -20% Hotels and restaurants Transportation -25% 0% 10% 20% 30% 40% 50% 60% 70% 80% % of firms in the sector run by women Source: Data on sectoral growth is from Haver Analytics. Data on the number of male and female microbusinesses in each sector is taken from the IFLS. Note: The x-axis shows the share of female entrepreneurs operating in each sector. The further to the left the sector appears on the graph, the more male-dominated the sector is for microbusinesses, and the further to the right on the graph the sector appears, the more female-concentrated the sector is for microbusinesses. Sectoral growth compares the output of the sector during the period of Q2 to Q4 in 2020 with the output of the sector during the period of Q2 to Q4 in 2019. Negative growth indicates a contraction in the sector. The size of the bubbles is proportional to the number of microbusinesses in the sector. 86 Enterprising Women | What contributes to the gender gap in performance of microenterprises? In addition to disproportionately affecting different gender gap in microbusiness performance. sectors of activity, COVID-19 has led to an Prior to the COVID-19 pandemic, female increase in the amount of domestic work for many microentrepreneurs were also disadvantaged in households due school and daycare closures, the terms of other endowments including education need to care for sick family members, additional and access to collateral to obtain loans.79 These cleaning due to more time spent at home and differences can affect the coping mechanisms concerns about the virus, and more meals being that businesses can use to respond to pandemic- consumed at home. Prior to COVID-19, women in related issues. Unequal intra-household bargaining Asia and the Pacific contributed more than four or perceptions about the relative importance times the number of hours that men spent on of women’s or men’s businesses to household unpaid care work.75 Although both women and wellbeing may influence the extent to which men in East Asia and the Pacific have reported household coping mechanisms disproportionately an increase in the amount of time spent on affect women. For example, the Asian Financial caretaking and domestic activities due to the Crisis lowered Indonesian women’s business pandemic,76 global evidence shows that women assets in both urban and rural areas but did not are taking on more additional hours than men. In a affect the business assets of their husbands.80 study of 16 countries, women added an extra 5.2 hours on average per week of childcare compared Data from the Business Pulse Surveys in to 3.5 extra hours per week for men.77 Given Cambodia and Vietnam confirm the theoretical baseline inequalities, this suggests the gender expectations that the COVID-19 pandemic has gap in unpaid care work may be widening, even if had a particularly strong impact on women’s men are contributing more than previously. These businesses. Entrepreneurs were asked how their increases may make it particularly difficult for sales in the past 30 days compared to the same women to continue operating their businesses period in 2019, prior to the COVID-19 pandemic. or to work as many hours as they did previously. As shown in Figure 26, both male and female- Because there is a negative correlation between owned firms reported large declines in revenues. time spent on domestic work and business The declines were larger for female-owned firms performance for female microentrepreneurs in both countries although the gender difference in Southeast Asia,78 increased domestic work is not statistically significant in Vietnam. due to the pandemic may further widen the 87 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Figure 26 Female entrepreneurs had larger declines in their revenues linked with the COVID-19 pandemic than male entrepreneurs Change in sales in past 30 days compared to the same period in 2019 for male and female business owners in Cambodia and Vietnam 0 percentage change in sales Average predicted -20 -40 -45 -45 -48 -60 -54 Men-led Women-Led Men-led Women-led Cambodia Vietnam Source: Business Pulse Surveys Note: Computations control for sector, region, and calendar month of interview. Microbusinesses and small businesses have been particularly hard hit by the pandemic (Figure 27), and women are more likely to own microbusinesses and to have fewer employees if it is a small and medium enterprise.81 Moreover, the differences between how male- and female-owned firms have been affected are largest for microbusinesses: the revenues of female microentrepreneurs have shrunk by 52 percent, compared to a reduction of 47 percent for male microentrepreneurs (Figure 27). This suggests that without additional policy support targeting female microentrepreneurs and addressing their specific needs, the COVID-19 pandemic may contribute to widening the gender gap in microenterprise performance. Gender-sensitive recovery policies should address the specific challenges that the pandemic has created for female entrepreneurs and ensure that their implementation is done in a way that is inclusive. Figure 27 Declines in revenues linked with the COVID-19 pandemic were largest for female-owned microenterprises Change in sales in past 30 days compared to the same period in 2019, by gender and firm size in Cambodia and Vietnam -5 percentage change in sales Average predicted -25 -32 -32 -38 -38 -45 -43 -44 -47 -52 -65 Micro (0-4) Small (5-19) Medium (20-99) Large (100+) Men-led Women-led Source: Business Pulse Surveys Note: Computations control for sector, region, and calendar month of interview. The estimates pool data from Cambodia and Vietnam and use weights equal to the inverse of the number of observations in each country. 88 Enterprising Women | What contributes to the gender gap in performance of microenterprises? Endnotes 1  Please see Model 3.1 in Appendix B for more technical 24  asson, 1982; Brush & Hisrich, 1991; Schutjens & Wever, C details. 2000 2  Please see Model 3.2 in Appendix B for more technical 25 Ács et al., 2017: 28 details. 26 It is worth noting that women’s lower education levels are 3  Please see Model 3.3 in Appendix B for more technical not limited to entrepreneurs: among those who are not self- details. employed, women also have lower education levels than 4  Please see Model 3.4 in Appendix B for more technical men. details. 27 Education was asked as a discrete variable in Vietnam, so 5  Please see Model 3.5 in Appendix B for more technical it was not included in Figure 17 since it cannot be directly details. compared to other datasets. 6  See Model 3.5 in Appendix B for technical details about this Brush & Hisrich, 1991; Kobeissi, 2010; Van Der Sluis et al., 28  analysis. 2008 7  International Finance Corporation, 2008 Centindamar et al., 2012; Van Der Sluis et al., 2008 29  8  Asia Foundation, 2013b 30 Information on business knowledge and use of business 9  As captured by the regressions in equation 3.1 in Appendix practices was only available for Vietnam. B. McKenzie and Woodruff, 2016; Bloom and Van Reenen, 31  10 This result is calculated by comparing the endowment 2007 effect component to the gender gap in performance using As calculated using regressions specified in Model 3.1 in 32  the Oaxaca-Blinder decomposition as described in Model Appendix B. 3.4 in Appendix B. See Box 6 or the overview of Appendix B These findings come from the endowment effect 33  for an intuitive explanation of this technique. component of the Oaxaca-Blinder decomposition, as 11 Calculated using an outcome variable of value-added and described in Model 3.4 in Appendix B. the Oaxaca-Blinder decomposition methodology described This finding comes from the differences in coefficients 34  in Model 3.4 in Appendix B. Result comes from the component of the Oaxaca-Blinder decomposition, as endowment effect component. described in Model 3.4 in Appendix B. 12 In Cambodia and Indonesia, the opposite is true, and 35 Xirera et al, 2021: 88 in Timor-Leste and Vietnam, there are not statistically This finding comes from the differences in coefficients 36  significant differences in the number of unpaid workers in component of the Oaxaca-Blinder decomposition, as men and women’s businesses. described in Model 3.4 in Appendix B. 13 The majority in the analysis sample for Timor-Leste also This finding comes from the interaction of endowments 37  do not have unpaid workers; however, this may be in part and coefficients component of the Oaxaca-Blinder linked with the way the data was constructed. Because decomposition, as described in Model 3.4 in Appendix B. the data do not distinguish which household member is 38 Xirera et al, 2021 primarily responsible for the business, analysis was limited 39 UN Women, 2020 to firms where all family workers were of the same gender. Álvarez, Marin, Fonfría, 2009 ; Szarka, 1990 40  14 In Lao PDR and Timor-Leste, the point estimate for distance Ács et al., 2017; Aldrich, 1999; Aldrich & Zimmer, 1986; Allen 41  to water source was very close to 0, and the correlation et al., 2007; Minniti, 2010 between cooking source and business performance was 42 Ács et al., 2017: 28 not statistically significant. Defined as the number of people the owner has regular 43  15 As the results denote only correlations, it is also possible contact with at least once every three months, which is that women with higher profits are more able to purchase useful for business operations these cooking appliances and fuel services. Even if The first statistic comes from OLS regressions with gender 44  the direction goes in this way, it shows that female interaction terms as discussed in Model 3.1, and the entrepreneurs are seeking ways to lower the amount of second statistic comes from a model including firm fixed time spent on domestic tasks, freeing time for either their effects as described in Model 3.2 in Appendix B. businesses, leisure, or other activities. The gender differences are statistically significant. 45  16 The correlation between being a homemaker and revenues 46 Asian Devolvement Bank, 2013: 23 is measured by equation 3.1 in Appendix B. 47 United Nations Economic and Social Commission for Asia 17 Measured by equation 3.2 in Appendix B. and the Pacific, 2013: 20 18 In Cambodia, although the relationship between being 48 World Bank, 2019 primarily a homemaker and profits is negative, it is not 49 World Bank, 2012 statistically significant. 50 Women, Business, and the Law database 2020 19 Operating out of the home or residential property is also 51 Ellis et al., 2010: 21 negatively correlated with business performance for men in Demirgüç-Kunt et al., 2008; Goheer, 2013 52  these countries, and gender differences are not statistically 53 Women, Business, and the Law Database 2020 significant. Correlations calculated using equation 3.1 in 54 Ellis et al., 2010: 21 Appendix B. 55 International Finance Corporation, 2017: 48 20 This calculation comes from comparing the endowment 56 Women’s World Banking, 2018: 5 effect of the Oaxaca-Blinder decomposition to the gender 57 International Finance Corporation, 2008: 21 gap, as described in Model 3.4 in Appendix B. See Box 6 58 As discussed in Section 2.1.3, existing data sources only and Appendix B Overview for a non-technical explanation include some variables related to capital for each country. of this technique and Model 3.4 in Appendix B for more As such, the results do not suggest that certain types of technical details. finance are more important in some countries than others. 21 Hundley, 2001 This finding comes from the results of a Oaxaca-Blinder 59  22 Lechmann and Schnabel, 2012 decomposition using the method described in Model 3.4 23 Delecourt and Fitzpatrick, 2020 in Appendix B with outcome variables of value-added and value of production. 89 Enterprising Women | What contributes to the gender gap in performance of microenterprises? 60  cs et al., 2017; Agénor & Canuto, 2012; Grady et al., 2018; Á Khandker et al., 2006; Lokshin & Yemtsov, 2005 Blackden & Wodon, 2006; Dinkelman, 2011; Ilahi & Grimard, 61  2000; Porter, 1995; Sewell & Desai, 2016 62 Results from Model 3.2 in Appendix B. 63 Results from Vietnam and Timor-Leste come from Model 3.1 in Appendix B. 64 World Bank, 2016 65 Asian Development Bank, 2018 66 Mu and van der Walle, 2011 67 Mannava, Perova, Tran, 2020 Klapper and Parker, 2010; Bardasi et al, 2011; Hallward, 68  2013; Anna et al, 1999 Goldstein et al, 2019; Campos et al, 2015; Bardasi et al, 69  2011 ; Hardy and Kagy, 2020 Defined as sectors in which 70% or more of businesses are 70  male-owned. 71 Men in female concentrated sectors also outperform men in male-dominated sectors in Cambodia. World Bank, 2019; Hardy and Kagy, 2020; Costa and Rijkers, 72  2012 73 Apedo-Amah et al. 2020 74 Apedo-Amah et al. 2020 75 International Labor Organization, 2018 76 UN Women, 2020a 77 UN Women, 2020b 78 See section 3.1.2 for more details. 79 See sections 3.2 and 3.3 for more details. 80 Dong, 2018 81 See Section 1 and Section 4 of this report for more details. 90 Section 4: Do opportunities, constraints, and needs of male- and female- owned small and medium enterprises (SMEs) differ? Although gender gaps in business performance of SMEs are only statistically significant in Indonesia, understanding the specific challenges and opportunities that men and women face in small and medium entrepreneurship is policy relevant. As discussed in previous sections, barriers that women face to start an SME may lead to a very distinct group of women choosing to and succeeding in becoming entrepreneurs at the SME level. However, they may continue to face different constraints compared to men. Without taking gender-specific constraints into account, gender-neutral SME policies may not meet the needs of all types of entrepreneurs or could generate inequities by only addressing the needs of certain SME owners. If male and female SME owners face different challenges and opportunities, gender-specific policies may be needed or policy makers should ensure that SME policies address the varied obstacles that men and women face. Moreover, understanding barriers to SME performance may feed back into the discussion of facilitating entry into small and medium entrepreneurship for more women. This section explores the 91 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? extent to which men and women face different challenges and opportunities in running SMEs in Southeast Asia and details obstacles that female-owned SMEs confront. Box 12 How does this report identify gender-specific constraints among SMEs? For most countries, we rely on descriptive characteristics to assess whether the analysis of declared constraints and relationship is different for male and female indicators of integration into the financial business owners.1 The results of this analysis system due to data constraints. Using a do not reveal causal relationships. However, detailed survey of manufacturing SMEs in they show interesting patterns about factors Vietnam, we use two additional techniques. correlated with business performance. First, we test the association between Second, we use a Cobb-Douglas production different characteristics of interest and function to examine the returns to factors of business performance using Ordinary Least production, including capital, land, and labor, Squares (OLS) regressions, interacting for men and women.2 the manager’s gender with the different 4.1 Perceived constraints Both the types of constraints that SME owners their business. However, the relative rankings cite and the extent to which male-owned of some obstacles are different for men and and female-owned SMEs report different women. For example, both access to financing constraints varies by country (Table 1).3 In and practices of competitors in the informal Cambodia and Timor-Leste, men and women sector are the main challenge for large shares perceive similar obstacles to their businesses, of both male and female-owned SMEs in and the relative importance of these obstacles Vietnam; however, access to finance is the is also similar for male-owned and female- most widely cited constraint for men, whereas owned SMEs. In all countries, there are at practices of competitors in the informal sector least some obstacles that sizeable shares is the most widely cited constraint for women. of male- and female-owned SMEs cite as Finally, in some countries, there is evidence the main constraint to their business. For that men and women face different constraints. example, in Indonesia, both men and women In Indonesia and Lao PDR, customs and trade cite practices of competitors in the informal regulations are one of the three most widely sector and political stability in large numbers, cited constraints for female-owned SMEs, and in Lao PDR, both men and women cite whereas it does not figure among the top three access to finance as the biggest obstacle to constraints cited by men in any country. 92 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? Table 1 Top three constraints for male-owned and female-owned SMEs PERCENT PERCENT BIGGEST BIGGEST OF FEMALE- OF MALE- OBSTACLE OBSTACLE FOR OBSTACLE OBSTACLE FOR COUNTRY OWNED SMES OWNED SMES RANK FEMALE-OWNED RANK MALE-OWNED CITING THIS CITING THIS SMES SMES OBSTACLE OBSTACLE Practices of Practices of 1 competitors in the 33% 1 competitors in the 26% informal sector informal sector CAMBODIA 2 Political instability 14% 2 Political instability 17% Inadequately Inadequately 3 9% 3 15% educated workforce educated workforce Practices of Practices of 1 competitors in the 22% 1 competitors in the 41% informal sector informal sector INDONESIA 2 Political instability 17% 2 Political instability 12% Customs and trade 3 14% 2 Tax rates 12% regulations 1 Access to finance 36% 1 Access to finance 40% Practices of Customs and trade 2 19% 2 competitors in the 18% regulations informal sector LAO PDR 3 Electricity 9% Inadequately 3 13% educated workforce Inadequately 3 9% educated workforce 1 Political instability 33% 1 Political instability 40% TIMOR-LESTE 2 Corruption 22% 2 Corruption 18% 3 Access to finance 14% 3 Access to finance 16% Practices of 1 competitors in the 20% 1 Access to finance 25% informal sector Practices of 2 Transport 19% 2 competitors in the 16% VIETNAM informal sector 3 Access to land 10% 3 Access to finance 16% Inadequately 3 10% educated workforce Source: Author calculations using WBES 93 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? Factors associated with the business customs and trade regulations are cited by 19 environment pose an important barrier to percent and 14 percent of female-owned SMEs both male-owned and female-owned SMEs in Lao PDR and Indonesia, respectively, and 19 throughout the Southeast Asian countries percent of female-owned SMEs in Vietnam cite studied. As shown in Table 1, political transport as their main obstacle. This difference instability is cited by large numbers of both does not seem to be driven by gender-based male-owned and female-owned SMEs in sectoral segregation, as women are more likely Cambodia, Indonesia, and Timor-Leste. Political to cite these obstacles than men even when instability may affect the business environment comparing men and women within the same by increasing the risks of innovation due to sector of activity. Perhaps women experience uncertainty over future economic conditions.4 the business environment differently due to Interruptions (or expected interruptions) in the size or composition of their networks or government programs meant to support knowledge of how to navigate regulations and entrepreneurs due to instability may also stifle connect to transportation services. Aligned innovation or remove important support for with this hypothesis, female SME owners self- SMEs.5 Practices of competitors in the informal report lower knowledge of laws and regulations sector is the main challenge for large shares of than male SME owners in a panel survey of both men and women in Cambodia, Indonesia, manufacturing SMEs in Vietnam. Moreover, and Vietnam and for male-owned SMEs in Lao a study of cross-border traders in Cambodia PDR. and Lao PDR showed that women paid a larger share of their gross profits in taxes and were Other aspects of the business environment less successful than men in negotiating taxes only figure among the top three most cited and fees.6 obstacles for female-owned SMEs. Namely, 4.2 Networks Networks are particularly important for owners’ networks are comprised of women on female entrepreneurs, but their composition average. Female SME owners are less likely is different than for male entrepreneurs. Data to belong to business associations, to pay from a panel survey of SMEs in Vietnam show membership dues for business associations, that female-owned and male-owned SMEs or to receive advocacy support from business have similar numbers of individuals overall associations than male SME owners (Figure 28: in their networks (Figure 28: Panel A), but the Panel B). Among those who are in associations, composition and relationship with networks women are still less likely than men to pay vary along gender lines. Female-owned SMEs membership dues, which suggests that women have larger shares of women in their network: may belong to different types of associations on average, 31 percent of the individuals that than men. Nevertheless, women are just as female business owners report having contact likely as men to receive advocacy support from with for their businesses are also women, an association, conditional upon belonging to whereas only 24 percent of male business an association. 94 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? Despite differences in network composition, times per year on average (Figure 28: Panel women tend to rely more frequently on their A). The number of individuals in the owner’s networks for support. Perhaps linked with network7 is positively associated with profits, this greater reliance for support, networks are and this relationship is stronger for female particularly important for the profitability of entrepreneurs than for male entrepreneurs. female-owned SMEs in Vietnam. On average, For female-owned SMEs, an additional person female SME owners receive assistance from in the owner’s network is associated with 0.55 their contacts 111 times per year for their firm, percent larger profits, compared to 0.22 percent whereas men report receiving assistance 103 higher profits for men.8 Figure 28 Female SME owners in Vietnam have similar numbers of individuals in their networks as male SME owners but rely more on their networks for support and have a different network composition Panel A. Gender differences in networks of SME owners in Vietnam Panel A: Gender differences in networks of SME owners in Vietnam Female Male 120 111.1 102.6 100 80 60 40 33.1 32.3 20 0 Number of people in the network Number of times received assistance from network Panel B. Gender differences in networks of SME owners in Vietnam Panel B: Gender differences in networks of SME owners in Vietnam Female Male 35% 31% 31% 30% 28% 25% 24% 20% 15% 10% 10% 8% 8% 6% 5% 0% % who pay membership fees for% who receive advocacy support % in network who are female % who are member of at least one business association least one association from at least one association Source: Author calculations using VSMES Note: Gender differences in number of people in the network are not statistically significant. Gender differences in percent in network who are female, percent who are a member of at least one business association, percent who pay membership fees for at least one association are statistically significant at the 1 percent level. Gender differences in the percent who receive advocacy support are statistically significant at the 5 percent level. Gender differences in the number of times received assistance from network are statistically significant at the 10 percent level. 95 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? 4.3 Access to finance Various sources suggest that access to SMEs that were unserved or underserved by finance poses a constraint to female-owned financial institutions (Women’s World Banking, SMEs in Southeast Asia, even if it was only 2018: 4).9 As Figure 29 shows, the number the most widely cited constraint for female- of unserved and underserved women-owned owned SMEs in Lao PDR according to the SMEs by financial institutions ranges from 46 World Bank Enterprise Surveys data (Table in the Philippines to 61 percent in Indonesia. 1). A study of women’s access to finance in Furthermore, only a small proportion of female- four Southeast Asian countries (Cambodia, owned SMEs reported being well-served by Indonesia, the Philippines, and Vietnam) found financial institutions, ranging from 3 percent in a significant proportion of female-owned Cambodia to 21 percent in Vietnam. Figure 29 Only a small percentage of female-owned SMEs report being well-served by financial institutions Access of women-owned SMEs to formal financing in Cambodia, Indonesia, the Philippines, and Vietnam (%) Access of women- owned SME to formal financing in Cambodia, Indonesia, the Philippines, and Vietnam (%) 0 10 20 30 40 50 60 70 80 90 100 Cambodia 41 8 3 Indonesia 42 19 8 Philippines 24 22 16 Vietnam 29 26 21 Do not need credit Unserved Underserved Well-served Source: Women’s World Banking (2018: 4). 96 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? Data from the World Bank Enterprise Surveys For most indicators, gender differences are confirms that many female-owned SMEs are not statistically significant. Notable exceptions not well incorporated into the formal financial include having an overdraft facility in Indonesia, system; however, there is a fair amount of where female-owned SMEs are less than half as heterogeneity across countries (Figure 30). likely as men to have one, and having a line of The share of female-owned SMEs that have a credit or loan from a financial institution in Lao checking or savings account ranges from 38 PDR and Vietnam, where the share of female- percent in Cambodia to 87 percent in Timor- owned SMEs with a line of credit or loan is 28 Leste, and the share that have an overdraft and 20 percentage points, respectively, lower facility ranges from 2 percent in Lao PDR to than the share of male-owned SMEs. In Lao 30 percent in Timor-Leste. Moreover, the vast PDR, female-owned SMEs are just as likely to majority of female-owned SMEs do not have a have applied for a loan, suggesting that women line of credit or loan from a financial institution. may face higher rates of rejection or receive For example, in Timor-Leste, only 10 percent loans that are much shorter in term. In contrast, of female-owned SMEs have a line of credit in Vietnam, female-owned SMEs are less likely or loan from a financial institution, while in to have applied for a loan than male-owned Vietnam, 23 percent of them do. SMEs and are more likely than male-owned SMEs to say they did not apply for a loan Although female-owned SMEs overall are not because they did not need credit. very well integrated into the formal financial system, in most cases, this trend is not limited to female-owned SMEs (Figure 30). 97 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? Figure 30 Many female-owned SMEs are not well integrated in the formal financial system; however, there is heterogeneity across countries Panel Panel A: Percentage A. Percentage of of SMEs SMEs with with aor a checking checking or savings account savings account 100% 87% 78% 80% 73% 61% 60% 62% 60% 54% 52% 38% 40% 32% 20% 0% Cambodia Indonesia Lao PDR Timor-Leste Vietnam % male-owned SME % female-owned SME Panel Panel B. Percentage ofB: Percentage SMEs of with a line of SMEs credit with or loan a line from of a financial institution credit or loan from a financial institution 50% 43% 39% 40% 28% 30% 22% 23% 21% 20% 17% 16% 11% 10% 10% 0% Cambodia Indonesia Lao PDR Timor-Leste Vietnam % male-owned SME % female-owned SME PanelC: Panel Percentage C. Percentage of SMEs of SMEs with with an an overdraft overdraft facility 35% 32% 30% 30% 25% 20% 15% 13% 10% 10% 10% 7% 8% 6% 5% 3% 2% 0% Cambodia Indonesia Lao PDR Timor-Leste Vietnam % male-owned SME % female-owned SME Source: Author calculations using WBES Note: The only gender differences that are statistically significant are having a line of credit or loan in Lao PDR and in Vietnam (Panel B) and having an overdraft facility in Indonesia (Panel C). 98 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? Evidence from Vietnam shows the importance and machinery, suggesting both male and of considering the type of capital that male female-owned SMEs face similar levels of and female-owned SMEs need. Female-owned constraints in accessing physical capital. SMEs in Vietnam appear to face greater Policies to boost SMEs access to finance need constraints than men in terms of land and to consider the types of financial products financial capital, which includes liquid assets offered, as a focus on capital that can be used and cash, as they have greater marginal returns to purchase machines and equipment will to these factors of production (Figure 31). not address women’s greater constraints for However, women have statistically similar financial capital and land. returns to physical capital, such as equipment Figure 31 Female-owned SMEs in Vietnam have larger marginal returns to financial capital and lower marginal returns to capital stock than male-owned SMEs Marginal Marginal returns returns to factors to factors of production of production for SMEs’ for in Vietnam production value SME's production value in Vietnam 0.50 0.38 0.39 0.40 0.30 0.20 0.20 0.08 0.08 0.10 0.05 0.03 0.02 0.00 Capital stock Full-time workers Land value Financial capital Male Female Source: Author calculations using VSMES Note: Gender differences in returns to land and financial capital are statistically significant at the 1 percent level, whereas gender differences in returns to capital stock and full-time workers are not statistically significant. Estimated using a Cobb-Douglas production function, as described in Model 3.2 in Appendix B. 99 Enterprising Women | Do opportunities, constraints, and needs of male- and female-owned small an medium enterprises (SMEs) differ? Endnotes 1 Please see equation 4.1 in Appendix B for more technical details. 2 Please see equation 4.2 in Appendix B for more technical details. 3 Respondents were asked what the biggest obstacle affecting the operation and growth of the establishment was, and there were 15 answer options: 1. Access to finance (availability and cost); 2. Access to land; 3. Business licensing and permits; 4. Corruption; 5. Courts; 6. Crime, theft and disorder; 7. Customs and trade regulations; 8. Electricity; 9. Inadequately educated workforce; 10. Labor regulations; 11. Political instability; 12. Practices of competitors in the informal sector; 13. Tax administration; 14. Tax rates; 15. Transportation of goods, supplies and inputs 4 Varsakelis, 2006 5 Allard, Martinez and Williams, 2012. 6 World Bank, 2016 7 Defined as the number of people the owner has regular  contact with at least once every three months, which is useful for business operations. 8 This estimate comes from Model 4.1 in Appendix B. 9 The Women’s World Banking report defines SME as women-  owned if the sole proprietor, or at least one joint owner, is female. 100 Section 5: Toward a targeted, gender-informed entrepreneurship policy This report provides evidence of gender-specific barriers to entrepreneurship in Southeast Asia. Section 1 of this report shows that women are just as likely as men to enter microentrepreneurship; however, they are less likely than men to own SMEs in almost all countries. This suggests gender-specific constraints hinder women from opening an SME or growing their microbusinesses into an SME. Women’s microbusinesses have profits that are lower than men’s profits. These gender gaps in entrepreneurial outcomes are linked with gender gaps in entrepreneurial inputs, including hired labor, time for own labor, skills, and capital. Gender norms, market failures, and legal distortions are underlying factors that constrain women’s access to inputs and inform their preferences of how to allocate limited resources. Two types of policies should be ideally adopted in parallel: policies that directly address input gaps and those that influence the underlying drivers of gender gaps. Addressing underlying drivers of gender gaps can lead to more sustainable changes in the long-term but can be a slow process. As such, policies that address gender gaps 101 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy in inputs can help in the short-term. In some and enable their participation in broader cases, policies that directly address current initiatives for four types of interventions: skills gender gaps in inputs can also help shift enhancement, improving access to capital, underlying drivers of gender gaps, creating support for hiring and managing workers, and a virtuous cycle. For example, programs to alleviating time constraints related to domestic facilitate female entrepreneurs’ access to work. The types of interventions reviewed in bank loans directly boost beneficiaries’ input this section are not exhaustive, and additional levels and simultaneously enable more bank types of interventions, such as improvements officers to interact with female entrepreneurs. to the business environment, may also support Such interactions may counteract the both female and male entrepreneurs. However, influence of gender norms and shift bank this section focuses on interventions that officers’ perceptions of female entrepreneurs’ can address the gender gaps identified in this managerial capacity, ability to repay, and report. passion for their businesses. A one-size-fits-all approach to supporting Addressing these gendered barriers through female entrepreneurs would be insufficient tailored policy actions can unleash untapped because of the heterogeneous needs of women growth potential and promote equity. As operating at different scales. This report has discussed in the introduction, gendered barriers shown that female microentrepreneurs have to opening a business lead to a misallocation different characteristics and face different of talent and lower per capita income, and challenges than female SME owners. As such, removing these barriers can lead to substantial policies should be tailored to different types of economic growth.1 Earnings gaps also weaken MSMEs. This section provides evidence of how economic growth.2 Thus, addressing gendered policies in the four domains can be adapted for barriers to participation and performance in women’s microbusinesses and SMEs. entrepreneurship is a smart economic decision. To maximize the effectiveness of MSME Moreover, removing barriers to women’s ability policies, policymakers should simultaneously to engage and compete in entrepreneurship address multiple constraints rather than ensures that women are free to live the life of tackling constraints in isolation. A recent their choosing, which is a basic human right.3 literature review and meta-analysis of Development entails removing obstacles that interventions to support female entrepreneurs leave individuals with little choice or opportunity notes that intervention packages that address to exercise their agency.4 Promoting equitable multiple barriers, such as providing both skill entrepreneurship is hence critical to promote development and financing, are more effective development and has a strong intrinsic value. than those that address only one constraint.5 A gender-informed entrepreneurship Intuitively, an intervention that helps firms policy should tackle challenges that female identify qualified workers will only be effective entrepreneurs face and ensure that female if they have enough liquidity or access to entrepreneurs can equitably benefit from finance to pay their salaries or to purchase broader MSME policies. This section provides additional equipment for the new employees evidence-based examples of how policies to use. Conversely, a finance intervention can both address women’s specific needs that enables firms to purchase additional 102 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy equipment and pay new workers may not jobs.6 Policies that facilitate women’s wage generate returns if firms cannot identify, hire, work can enable women to choose from more and retain qualified workers. A multiplier effect options about how to engage in the labor from complementary interventions may justify market and reduce the need for necessity- their increased cost. In a resource constrained oriented entrepreneurship. Nevertheless, environment, it may be more impactful to programs to increase access to women’s wage address multiple binding constraints for work may take time to enable all women to a smaller group of entrepreneurs than to access the jobs most aligned with their skills address one constraint for a larger group of and goals. As such, it is critical to invest in entrepreneurs. Nevertheless, some underlying boosting the performance of all types of female drivers of gender gaps, such as gender norms, entrepreneurs in parallel to unlocking new labor take time to change and may limit the extent market opportunities. to which certain input gaps, such as time for own labor and skills, can be effectively closed in the short-term. The design and expected return to costly interventions may need to be conditioned on the extent to which the targeted or complementary inputs can be addressed. Identifying the right package of policies should be highly tailored to the country context, target firms and their binding constraints. Because action is needed on multiple fronts, it is critical to create a coordinated policy agenda among the various relevant ministries, institutions, and stakeholders. A high-level champion can build consensus, track progress, and facilitate coordination on the agenda. Such a champion should ensure that MSME policy agenda addresses the needs of both women and men as well as female entrepreneurs operating at different scales. Finally, promoting gender equality in terms of endowments, agency, and economic opportunities more broadly can help women engage in labor market activities that are most aligned with their goals and interests. Due to a lack of outside opportunities and a need to earn a living, in many economies, women are more likely than men to engage in necessity entrepreneurship, motivated by a need to earn a living and the scarcity of 103 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy 5.1 Interventions to enhance skills 5.1.1 Rationale and targeting from these exchanges could bolster their businesses since female SME owners have Programs to support entrepreneurs’ skill different network composition than their male development have the potential to bolster counterparts and use their networks in slightly female entrepreneurship in Southeast Asia. different ways. Given differences in education As shown in Sections 2 and 3, lower-skilled and skills, programs should adapt content of workers tend to sort into entrepreneurship; programs to the different baseline levels and however, education and business needs of microenterprises and SMEs. knowledge are positively associated with Entrepreneurs may underinvest in developing microentrepreneurs’ profits. Although female their business skills due to market failures, microentrepreneurs have lower education which can be one justification for public levels than their male counterparts, the positive support.7 Entrepreneurs may not know the relationship between education and business potential benefits of business training, and performance is stronger for women than men, this may be particularly the case for firms with and skills gaps contribute to the gender gap in the worst business practices who may lack microenterprise profits. As such, while skills- knowledge of what good business practices enhancing interventions would likely benefit include. Even if entrepreneurs think a training both male and female enterprises, women’s could be beneficial, they may lack information businesses may reap greater benefits. Global about the magnitude of the return on their evidence sheds light on several promising types investment or have concerns about the quality of interventions to develop entrepreneurs’ skills, of the training offered. Financially constrained which have the potential to address obstacles firms may also underinvest, as it can be to women’s entrepreneurship that this report challenging to borrow funds for intangible has identified. assets like training. Moreover, in some contexts, Skills-enhancing interventions are especially training is not readily available in the market. relevant for female microentrepreneurs, Equity concerns also justify public intervention who have lower education levels and to support the skill development of female knowledge of business practices than both entrepreneurs. As discussed in previous their male counterparts and female SME sections, female entrepreneurs have lower owners. However, because SME ownership is education levels than male entrepreneurs, and associated with greater levels of both education these gender gaps in endowments put female and business-specific knowledge, programs to entrepreneurs at a disadvantage. Closing boost the skills of prospective business owners gender gaps in access to formal education hold potential for supporting women who would is critical to promote equitable opportunities like to break into the SME sector rather than among the next generation, and gender gaps in opening microbusinesses. Although addressing school enrollment have closed or reversed in skill gaps of female microentrepreneurs is recent decades.8 However, such policies will not critical, supporting female SME owners to enable adult women who have completed their expand their networks and reap more benefits 104 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy education to equitably engage and compete training but were in the same markets as in entrepreneurship. Targeted interventions to those who did were small and not statistically support the business and entrepreneurial skills significant, and the sales volume of the entire of adult women can help level the playing field. market where women were trained increased.12 Alternative types of business training 5.1.2 Evidence of promising Although traditional business training may skills-enhancing interventions be one possible solution, several variations in content, delivery, or adaptation of training Traditional business training materials have been shown to be more A recent meta-analysis shows that managerial effective for promoting the development of business training programs can positively women’s businesses. One such promising impact firm performance, boosting profits on approach is to use heuristics, or rules of thumb, average by 10 percent and sales by 5 percent. to help entrepreneurs improve the management These effect sizes are reasonable given the of their businesses. For example, Drexler et al. relatively short duration of most programs and (2014) compared traditional business training expected return on investment.9 A growing to a training that focused on simple heuristics body of rigorous evidence testing business to help entrepreneurs—90 percent of whom training programs typically involve a trainer were women—separate household and business teaching groups of 15 to 40 potential or existing finances in the Dominican Republic. They found entrepreneurs in a classroom setting over three that the simple rule of thumb training was more to 12 days. Common modules for potential effective than traditional accounting training at entrepreneurs include identifying a business improving entrepreneurs’ business practices idea, developing a business plan, permits, and revenues, especially for firm owners who costing, pricing, and budgeting, while common were less educated and had lower initial use modules for existing entrepreneurs include of business practices. Arraíz et al. (2019) record-keeping and accounting, marketing, conducted a similar experiment in Ecuador and human resources (HR) management, inventory found that a four-hour training providing rules of management, planning and operations thumb to improve finances increased the daily management.10 While costs of training profits of firms by 8.1 percent one year later, and programs vary depending on the country the training was particularly effective for women context and implementation arrangements, and entrepreneurs with lower cognitive scores. several studies have found that the benefits In another example, informing microbusinesses outweigh the costs under reasonable of the importance of keeping correct change hypotheses and time periods.11 to avoid losing customers led to an increase in profits three months after the intervention in Profit increases do not necessarily come at Kenya.13 the expense of firms in the same location who were not trained, according to emerging Although most business training programs evidence. A study that offered training to some undergo some form of adaptation to the local businesses in select markets in Kenya showed context and population, some promising that impacts on firms who did not receive interventions have done much more adaptation 105 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy of materials than is typically done in a more based training program. Alibhai et al. (2019) standard program. For example, the Get Ahead showed that a personal initiative training training program studied in Bulte et al. (2016) for female entrepreneurs in Ethiopia did not and McKenzie and Puerto (2021) teaches have a statistically significant impact on entrepreneurial skills from a gender perspective profits, likely due to the profile of trainers who and includes topics specifically designed to delivered the training. However, an alternative overcome gender constraints. Although the psychology-based business training program in profits variable was too noisy in the Bulte et al. Ethiopia, which aimed to boost entrepreneurs’ (2016) study to detect impacts, McKenzie and self-esteem and entrepreneurial spirit, was Puerto (2021) show that after three years, this effective at promoting entrepreneurial success, program increased profits by 15 percent, which boosting average monthly profits by 30 is greater than average impacts in traditional percent.18 Another example of a psychology- programs. Using a different approach, Dalton et based training for entrepreneurs includes an al. (2019) developed a highly tailored training agency-based training that draws on principles program for entrepreneurs in Indonesia by first of positive psychology and aims to boost self- studying the specific practices that were most knowledge and actionable growth strategies. strongly correlated with business success This type of training improved the ability of in the project areas. The program included a entrepreneurs to sell improved cookstoves handbook that detailed the local best practices, in Kenya, with the greatest improvements for corrected common misperceptions about female entrepreneurs.19 them, and provided relevant examples of using Overall, these alternative approaches have these practices in the local context. Combining proven more effective for supporting business this handbook with two 30-minute sessions to development than more traditional business help firms understand and use them increased training approaches and content. A meta- profits by 35 percent one year later.14 analysis examining the average impacts of Another promising alternative to traditional business training programs using a variety business training uses psychology to help of alternative approaches found a 15 percent entrepreneurs develop a more entrepreneurial increase in profits on average and an 11 percent mindset. One specific type of psychological increase in sales on average, greater than the training teaches personal initiative, or the average effect sizes found in more traditional development of a proactive, persistent, training programs.20 Another meta-analysis future-oriented mindset.15 In a small-scale that focuses only on female entrepreneurs also pilot in Uganda, Glaub et al. (2014) showed shows that enhanced training programs are that a personal initiative training boosted more effective for boosting the performance of entrepreneurs’ personal initiative behavior, women’s businesses.21 Most individual studies and this increased entrepreneurial success. that have compared traditional business A similar training in Togo increased firms’ training with a more innovative approach profits by 30 percent on average16 and was have also found alternative approaches equally effective for women with different to be more effective than traditional ones, initial levels of human capital.17 Nevertheless, even if the difference between the two is not the characteristics of the trainers may be systematically statistically significant.22 essential to the success of a psychology- 106 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy Consulting and mentoring consulting can be costly, so Iacovone et al. (forthcoming) tested two types of consulting Personalized guidance from business interventions among auto parts manufacturers consultants or trained mentors provides a with more than 10 workers in Colombia. Some promising alternative to classroom-based entrepreneurs received individual consulting, group training; however, this approach which cost approximately US$30,000, while seems to be more effective for SMEs than others received group consulting services that for micro firms. One advantage of consulting cost approximately US$10,000 per firm. Group types of interventions for SMEs is that it can consulting was clearly more cost-effective, help owners address the specific needs of increasing use of good business practices their businesses. Bruhn et al. (2018) showed by between 8 and 10 percentage points, that Mexican SMEs received management employment by between six and 15 workers, consulting services for various types of sales by between 28 and 33 percent, profits by managerial practices, and that no single between 5 and 26 percent, and value-added by practice seemed to be a silver bullet to unleash 43 percent. growth. As such, highly tailored interventions can ensure that skills enhancement programs Consulting also supported firms with between meet the specific needs of each SME. Such two and 15 workers in Nigeria; however, programs can also introduce entrepreneurs insourcing and outsourcing interventions to market-based solutions to grow their performed equally well at half of the cost. businesses, helping overcome informational Insourcing and outsourcing interventions barriers surrounding the quality of available enabled entrepreneurs to hire skilled service providers and the returns to business professionals to conduct certain tasks rather development services. Indeed, a study in than providing the entrepreneur with the skills Nigeria showed that firms that received to conduct the tasks themselves.25 Market- consulting services or support to insource based solutions that help entrepreneurs hire or outsource specialized tasks to trained skilled workers or contractors can thus be a professionals were more likely to return to the promising alternative to upskill SMEs. business service market for additional support Nevertheless, interventions offering after the subsidized intervention period.23 personalized guidance from business Consulting interventions have been shown to consultants or trained mentors have had boost the productivity of SMEs. For example, mixed results for boosting the productivity of in Mexico, SMEs who were offered four hours microenterprises. Karlan et al. (2015) found per week over the course of one year to work that while consulting services offered to with a consultant on needs identified through microenterprise Ghanaian tailors led to modest a thorough firm diagnostic increased their increases in business practices in the short total factor productivity and return on assets run, these impacts faded over time, and there by 0.2 standard deviations and profits by were no impacts on profits during any time 0.1 standard deviations one year after the period. While in the short run, microbusinesses intervention. Effects appear long-lasting, as in a Kenyan slum that received mentoring from treatment firms had 50 percent more workers profitable business owners in their area saw even five years after the program.24 One-on-one an increase in profits of 20 percent, the effect 107 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy dissipated by 12 and 17 months.26 Similarly, Social learning interventions in Peru, the impacts of combined training Promoting peer-to-peer learning and network and technical assistance were stronger for formation is another promising approach to female microentrepreneurs approximately 10 support the development of SMEs. In a light- months after the program, but by approximately touch intervention in Ethiopia, Tanzania, and two years after, the training alone and Zambia, Fafchamps and Quinn (2018) found the combination of training and technical that peer networks formed through a business assistance had similar impacts, suggesting plan competition boosted knowledge of VAT the individualized assistance may not be registration and the need for bank account worth the additional cost.27 In other cases, ownership. Building off this work, Cai and Szeidl mentoring add-on interventions to business (2018) tested a more intensive peer-learning training did not boost the effectiveness intervention in a sample of 2,820 SMEs in of programs for microenterprises in the China. Small groups of managers were asked to Dominican Republic,28 Kenya,29 or Pakistan.30 hold monthly meetings for a year during which Nevertheless, some studies have shown more members would meet at a member’s firm, tour promising impacts of mentoring interventions the firm, and discuss relevant business issues. for microentrepreneurs. For example, volunteer Even one year after the end of the intervention, marketing coaches from across the world those who had been invited to the meetings worked with Ugandan entrepreneurs whose had revenue that was 8.1 percent higher, as well businesses employed on average 1.7 paid staff. as higher profits, inputs, number of partners, Over a two-year period, firms matched with borrowing, and management practices. international marketing experts experienced a Managers shared business information, 36 percent increase in profits and 16 percent especially more general information with firms increase in employees which seemed driven by that were not direct competitors, and learning adopting product differentiation.31 Moreover, from peers seems to be one channel for the individualized consulting, whether delivered in success of the intervention since those with the business or in the classroom after a training better performing peers benefitted more.34 In program, boosted income by approximately another example, working with high-growth 15 percent one year after the intervention in technology firms in India, Chatterji et al. (2018) Chile, although an alternative intervention using found that entrepreneurs who received advice role models was able to produce a similar from peers with a formal approach to managing impact at one-tenth of the cost.32 In line with their employees experienced employment studies showing positive impacts of consulting growth of 28 percent and were 10 percentage interventions on SMEs, some studies have points less likely to fail two years after the shown that personalized assistance may be intervention than those who received advice more effective or have more sustained impacts for peers with an informal human resource for microentrepreneurs with more experience, management approach. formal education, or comparatively larger businesses.33 Social learning interventions also hold promise for microentrepreneurs. In Chile, inviting a successful alum of a business training program to speak at training sessions boosted the 108 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy income of training participants by 15 percent experiment suggested that ties to family and after one year. These role models seem to shift community are positively linked with women’s participants’ attitudes toward entrepreneurship entrepreneurial activity and profit.37 and investment decisions.35 In India, women who were invited to attend a business training Summary of promising interventions with a friend experienced much bigger impacts Global evidence suggests several types of on borrowing, business volume, and business interventions that policy makers in Southeast plans than those who were invited to attend Asia may want to consider, which address on their own. They also had higher household the skills gaps that contribute to the gender incomes and expenditures and were less likely gap in microenterprise profits, support SME to state their occupation as housewife. Access development, and differences in networks to networks seems to contribute to these among male and female SME owners. Table impacts, as impacts of attending the training 2 provides a summary of the global evidence with a peer were concentrated among women described above. belonging to groups with more restrictive social norms.36 Also in India, a seven-year field quasi- Table 2 Summary of global evidence of programs to improve entrepreneurs’ skills INTERVENTION RELEVANCE SUMMARY OF EVIDENCE EFFECTIVENESS DESCRIPTION MICRO SME Standard business training programs can have a positive impact on firm performance. However, alternative training STANDARD BUSINESS approaches  that use heuristics, highly tailored content,     TRAINING PROGRAMS or psychological principles have proven more effective on average. Business training that uses basic rules of thumb, content ALTERNATIVE that is highly tailored to the local context, or is based on BUSINESS TRAINING   psychology is more effective than standard business PROGRAMS training at boosting business performance. PERSONALIZED Business consulting can lead to sustained positive impacts Micro GUIDANCE on SME performance. Evidence is somewhat mixed for FROM BUSINESS microenterprises and may be more effective for those CONSULTANTS OR with more experience, formal education, or slightly larger TRAINED MENTORS businesses. SME PEER-TO-PEER Social learning interventions have improved the business LEARNING performance of both SMEs and microenterprises and can be   AND NETWORK cost-effective solutions. FORMATION RELEVANCE Strong relevance Somewhat relevant Not relevant EFFECTIVENESS  emonstrated effectiveness D  ixed or limited evidence M  inimally or not effective M 109 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy 5.1.3 Ensuring gender- mentors, or role models may serve as an inclusive skills-enhancement inspiration for female entrepreneurs or help interventions them relate more with the materials. Logistical arrangements may pose a barrier Even programs that are not gender-specific to women’s participation if the program is not can adopt a “smart design” to ensure that designed with gender in mind. In a review of women are able to equally benefit from them the effectiveness of several entrepreneurship despite the gender-specific constraints interventions for female entrepreneurs, Revenga that they face.38 To ensure skills-enhancing and Dooley (2020) note that many studies programs address the needs and constraints held training during the middle of the day with of female entrepreneurs, policy makers should no childcare options, which was associated consider adaptations to the curriculum and with higher dropout rates and absenteeism for delivery, logistical arrangements, and measures women than men. To address such concerns, to debias service provision. a youth employment program in Benin invited Although traditional business training participants with young children to bring a programs can have an impact on entrepreneurs second person to care for the children during generally,39 “enhanced training programs” that the training and offered transport stipends include mentoring components or different and mid-day meals to both the trainee and types of content are more effective for female to the babysitter.44 In the preparation of a entrepreneurs.40 Adapting content to address training program for microentrepreneurs in gender-specific constraints, such as the Togo, formative qualitative work prior to the Get Ahead for Women in Enterprise training training helped ensure that the logistical program developed by the ILO, has improved the arrangements were appropriate to ensure business performance of female entrepreneurs widespread participation of women. Based on in Kenya.41 Psychological business training, focus group discussions, sessions took place such as personal initiative, has been shown during hours when children would be in school to work for female entrepreneurs with varied and outside of hours when women typically characteristics42 and is also effective for prepare meals for the family. In addition, rather boosting performance of male entrepreneurs.43 than using a centralized location for training, In designing training programs, policy makers hotel conference rooms throughout Lomé should carefully consider not only the needs were rented to ensure that no entrepreneur of the general target population, but also had to travel more than 15 minutes to reach whether the content is adapted for female their training center.45 Bringing training close entrepreneurs. As shown in Section 3 of this to entrepreneurs is particularly important report, female microentrepreneurs have lower for female entrepreneurs, who often face education levels than male entrepreneurs, so more time constraints and may face security it is important also for content and materials issues in transportation. When working in to be adapted for entrepreneurs with lower remote areas, one innovative project brought literacy levels. Although there is limited a mobile team to deliver services, so women rigorous evidence about how the gender of would not need to travel long distances to the trainer impacts women’s experience in participate.46 Beyond the timing, location, and training programs, working with female trainers, need for childcare arrangements, outreach 110 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy and enrollment campaigns should incorporate sexual harassment and abuse. This would gender. For example, if women have smaller include a clear and accessible anti-sexual networks, specific outreach activities targeting harassment code of conduct and policy, training women may be needed to ensure inclusive and awareness raising, grievance redress awareness of the program. Moreover, eligibility mechanisms, and monitoring and evaluation of criteria may inadvertently exclude women, such these systems.47 Beyond ensuring that women as programs requiring a national identity card are not exposed to harassment or gender- in contexts where women are less likely to have based violence during their participation in them. skills programs, it is important to ensure that they are treated equitably. One potential way to It is critical to have the proper protections incentivize trainers to ensure gender-equitable and incentives in place to ensure an unbiased treatment is to offer financial incentives based service provision in addition to considering on achieving specific results, such as the whether content and logistical arrangements number of women who successfully complete are adapted to women’s needs. Skills training the training.48 programs should include mechanisms to prevent, respond to, and report cases of 111 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy 5.2 Interventions to improve access to capital 5.2.1 Rationale and targeting collateral requirements can increase the bank’s risk because less risky borrowers leave the Improving women’s access to finance in credit market and remaining borrowers may Southeast Asia could support potential propose riskier projects. As such, banks do entrepreneurs in opening businesses, close not adjust interest rates until the credit supply gender gaps in microbusiness performance, equals credit demand, and some borrowers and bolster SME performance. As shown in cannot access credit even if they have Section 2.1.3, women who have land that can worthwhile investment proposals.49 Although be used as collateral or are in households that credit rationing affects both male and female have received credit are more likely to engage in borrowers, women may be more exposed if entrepreneurship. Female microentrepreneurs bank officers find men’s funding request pitches in Indonesia are more likely than men to rely to be more convincing50 or if social norms that on savings to open their businesses, and regard men as breadwinners bias decisions of participation in rotating savings and credit how to invest limited funds. Public investment associations seems to enable women to open is needed to address these market failures businesses rather than contribute to family and to ease access to finance for firms with farms or businesses as an unpaid worker. viable projects who are affected by such credit Section 3.3 shows that the relationship between rationing. different sources of capital and business Moreover, supporting female entrepreneurs’ performance is strongly positive for female access to finance is aligned with the value microentrepreneurs, which is not always of equitable development because systemic the case for men. Moreover, the gender gap gender inequalities inhibit women’s access to in value added and the value of production finance. As discussed in previous sections of among Vietnamese microbusinesses is this report, women have more limited collateral completely explained by women’s lower levels and face gender norms that regard husbands of capital, land, and labor, and female-owned as responsible for credit-related transactions microbusinesses in Vietnam face greater or land management. Public investment is financial capital constraints than men and are warranted to address distributional concerns less leveraged. Although neither men-owned or to promote the value of equity. Because men or women-owned SMEs are well incorporated and women do not face equal opportunities into the formal financial system, evidence from to seek financing for their business, policy Vietnam shows that women’s SMEs are more interventions to ease female entrepreneurs’ constrained in land and financial capital. access to credit are justified. Market failures can prevent both men- and A variety of approaches to improving access women-owned MSMEs from accessing finance to capital for MSMEs have been tried globally, and call for policy action. Asymmetries of which can serve as inspiration for policy information make it difficult for bank officers makers in Southeast Asia who would like to to assess the real value of projects and lead address the financing constraints of female to credit rationing. Increasing interest rates or 112 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy entrepreneurs described above. While there (2020) note that microfinance interventions is a vast body of evidence on some types of have a large positive impact on sales but are interventions, others have not been rigorously less effective than basic training in increasing evaluated for effectiveness. The various types profits or encouraging entrepreneurial activity of interventions can be grouped into four large among women. types of interventions, including policies to Although on average, effects of microfinance help improve access to loans, grants, savings interventions have been modest at best, interventions, and equity financing. Although results are heterogeneous. Crépon et al. (2015) equity finance interventions, such as supporting find that in Morocco, although the impacts angel investing, may have promise, they are of microfinance were positive on average, not included in this review due to the lack for some entrepreneurs there were negative of rigorous evidence demonstrating their impacts on profits, suggesting microfinance effectiveness. may not be an appropriate solution for all entrepreneurs. Evidence from India shows 5.2.2 Evidence of promising that the impacts of microfinance on business performance were strongest for those who had interventions to improve existing businesses than for new entrants54 and access to capital for those with more profitable businesses at baseline.55 Improving access to loans Emerging evidence suggests that improving Microfinance interventions promote access the terms of microfinance may enhance its to very small loans that are offered to effectiveness. In most cases, microfinance entrepreneurs or potential entrepreneurs offers loans with a relatively high interest under a variety of conditions. Given the size rate and no grace period before repayment of the loans, they are unlikely to meet the must begin. Field et al. (2013) show that capital needs of SMEs and are targeted at relaxing the immediate repayment obligation microentrepreneurs.51 Often loans are offered for microentrepreneurs in India led to using solidarity group lending, in which groups increased investment in the short-term and of members receive individual loans and the increased profits in the long-term. While members of the group help ensure repayment. microfinance interventions may support female However, some microfinance institutions also entrepreneurs in some contexts and under offer individual loans, typically to less poor certain conditions, the evidence base suggests borrowers.52 that it is not a panacea, so additional policy Reviews of microfinance interventions have interventions are needed to improve access to shown that they do not lead to transformative finance for women’s MSMEs. impacts for female entrepreneurs, although Other programs to help MSMEs gain access most studies do show positive, though to loans, such as psychometric testing, often statistically insignificant, increases in seek to overcome constraints related to lack of business start-up and performance.53 Using collateral, which may be particularly useful for meta-analysis techniques combining results female entrepreneurs who are more likely to from several studies, Revenga and Dooley lack collateral to secure loans. However, while 113 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy there are several types of interventions to ease entrepreneurs’ creditworthiness, credit collateral constraints, there is limited rigorous guarantees are another way to incentivize evidence of how effective these interventions banks to lend to firms without sufficient are. Psychometric testing, which assesses collateral despite that risk. Although there entrepreneurs’ personality characteristics, are no randomized controlled trials testing is one promising alternative to help MSMEs the impact of government-backed credit access credit. Pilot studies have shown that guarantees, there is some suggestive evidence psychometric tests are significant correlates of that these types of programs can also support borrowers’ risk of defaults. In Peru, borrowers MSMEs. Arraíz et al (2014) use propensity who had been rejected by a psychometrically score matching and difference-in-difference enhanced application scorecard had a methodology to show that government-backed probability of defaulting that was up to four credit guarantees in Colombia increased firm times greater than those who were accepted.56 output for the year of the guarantee as well as In Ethiopia, customers who performed better the two subsequent years studied. Although on the psychometric tests were seven times there were no significant impacts on fixed asset more likely to repay their loans than those with investments, as measured by the capital stock lower scores.57 Because psychometric testing per worker, investments in working capital can be available to any potential borrower who may explain the changes in output. A credit accepts to take the test, it has the potential to guarantee scheme in the Republic of Korea also help expand credit to those who lack a credit had positive impacts on SMEs in Korea: Oh et al. history or collateral or who would like to seek (2009) use propensity score matching to show greater loan amounts. Nevertheless, financial that the program increased sales, employment, institutions in developing countries may be wage levels, and survival of firms that received reticent at first to completely change their financing through the guaranteed scheme. lending practices and may be more likely to Loans that use movable collateral may also adopt in an initial stage as a complement to hold promise for supporting female-owned other screening and risk mitigation practices firms with limited collateral. Using non- rather than a substitute.58 experimental methods, Love et al. (2016) show Another new lending methodology to offer that the introduction of collateral registries loans based on cash flow rather than collateral for movable assets increased firm’s access to holds promise for SMEs. A non-experimental finance, in particular for younger and smaller evaluation of MSMEs in Bulgaria, Georgia, firms. Russia, and Ukraine showed access to cash- Another type of policy intervention to improve flow loans had positive effects on firm capital access to loans, which has typically targeted formation, revenue growth, and profit growth. SMEs more than microbusinesses, offers loan The impacts were increasing with firm size, with subsidies, introduces lending quotas for banks, some negative impacts for microbusinesses or blended finance. Three studies that take but positive impacts for small firms and even advantage of natural experiments suggest larger positive impacts for medium firms.59 that such measures may be effective if they While psychometric testing and cash- are well targeted to firms that are truly credit flow lending attempts to demonstrate constrained. Since 1973, Pakistan has offered 114 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy publicly subsidized working capital loans to implementation of the programs are crucial firms that export an eligible set of commodities; to achieving this policy objective. A study of however, in June 2001, cotton yarn became no earmarked credit in Brazil that are distributed by longer eligible, while other types of non-yarn private banks shows that banks are more likely textile firms could continue receiving subsidies. to extend earmarked loans to larger firms and Zia (2008) took advantage of this policy change those with an existing credit relationship, while to examine the effects of losing access to riskier borrowers who do obtain earmarked subsidized loans on SMEs by comparing credit often also face increased prices of free- outcomes before and after the policy change market loans.66 Moreover, subsidized credit for all yarn and non-yarn textile firms. The loss risks introducing market distortions, so the of publicly funded export credits lowered the general equilibrium effects of such a program access to credit in the form of working capital need to be considered and addressed at the loans as well as the overall amount of exports design stage. for smaller, privately owned enterprises. On the other hand, larger, publicly listed and Grants group network firms that presumably have The potential of grant interventions to more access to alternative sources of funding support female entrepreneurs is strongly did not experience declines in their working linked to their design and targeting. Public capital and were mainly affected in terms of transfers to firms in the form of grants may the cost of credit.60 Two studies examined the be warranted if it can generate positive impacts of changes in eligibility for a directed externalities, such as contributions to growth credit policy in India that requires 40 percent or employment creation. In a review of several of net bank credit be reserved for the priority types of entrepreneurship interventions, sector.61 The policy enabled newly eligible Revenga and Dooley (2020) find that grants credit-constrained firms to access credit62 and have a mixed record in terms of impacts for increased their investment, sales, profits,63 female entrepreneurs, with some studies and growth of export earnings.64 Although not even suggesting negative impacts. However, rigorously evaluated, the International Finance the studies with no or even negative Corporation and the World Bank Group’s impacts all involved cash grants given to We-Fi program offer blended finance or other microentrepreneurs and focused on the incentives to boost access to finance for female grantees’ businesses. Female entrepreneurs entrepreneurs, and experience suggests this may have different preferences but also face approach holds promise for supporting female more pressure due to gendered social norms entrepreneurs. These programs offer incentives, and unequal intra-household bargaining power such as subsidies or interest rebates, for that push them to invest cash grants in their banks that hit targets for reaching female households or in another household member’s entrepreneurs.65 A rigorous evaluation of these business. Revisiting studies from India, programs would provide valuable information Ghana, and Sri Lanka that showed only male on their potential. entrepreneurs had a return on investment after Although earmarked and subsidized credit receiving cash grants, Bernhardt et al. (2019) programs have the potential to reach demonstrate that considering household- credit-constrained firms, the design and level returns, cash grants offered to male and 115 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy female entrepreneurs were equally effective. program appeared to reach entrepreneurs that Previous studies had not captured impacts of were truly credit constrained, as it altered firms’ cash grants on women’s businesses because production decisions rather than only boosting the capital they received had been invested in the incomes of the entrepreneurs.71 Although their husbands’ enterprises.67 Aligned with this impacts were sizeable, the YouWiN! competition finding, in-kind grants, which are more difficult in Nigeria was expensive due to the size of the to divert away from the woman’s business, grants and the training and support offered to have been shown to be more effective for entrepreneurs to help develop their business female entrepreneurs in Ghana, in particular for plans. Programs with smaller grant sizes and women with higher initial profits and those in no training have also been shown to support sectors that are not female dominated.68 More firm entry and performance. Fafchamps and evidence, in particular from Southeast Asia, is Quinn (2017) use non-experimental methods to needed to confirm whether in-kind grants can show that a business plan competition offering sustainably boost the performance of female grants of US$1,000 to winners in Ethiopia, microentrepreneurs. Tanzania, and Zambia increased the likelihood of being an entrepreneur by 33 percentage While offering grants to microentrepreneurs points, employment in the firms by two has a mixed record, business plan additional permanent employees, and profits by competitions that result in cash grant prizes approximately US$150 per month six months have been shown to support the creation, after the competition. Cash grant prizes appear growth, and productivity of growth-oriented to be the key to the success of business plan businesses. A randomized controlled trial competitions, as a business plan competition of offering grants averaging US$50,000 to that only offered training and mentoring but not semi-finalists of the YouWiN! business plan funding to winners in Ghana did not have an competition in Nigeria showed that new firm impact on firm growth or employment one or applicant winners were 37 percentage points two years after treatment.72 more likely than the control group to operate a business three years after applying and Savings 23 percentage points more likely to have a business with 10 or more workers. Existing Rather than relying on external sources firm winners were 20 percentage points more of funding from loans or grants, likely to have survived and 21 percentage points savings interventions may help female more likely to have a firm with 10 or more microentrepreneurs overcome negative workers. The program doubled employment income shocks, protect business earnings for new firms and increased employment by 80 from household needs and the demands of percent for existing firms.69 The program closed others, boost productivity, and invest in their gender gaps in firm entry for female winners businesses. Different types of interventions and had similar impacts on employment, have been tested to boost women’s savings, sales, and profits for both men and women; including promoting access to accounts, however, women only represented one-fifth of financial literacy training, commitment savings applicants.70 Firms innovated more, and access products, and behavioral nudges. to capital and hiring of labor seemed to be Aligned with the finding in this report that many pathways for the success of the program. The female entrepreneurs rely on savings for their 116 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy start-up capital, proximal access to safe, formal financial literacy training emphasizing the accounts can help women save and open a non- importance of savings and explaining the farm enterprise. Suri and Jack (2016) showed new product increased female entrepreneurs’ that the mobile money service M-PESA in savings and profits.76 Similarly, Bastian Kenya increased the savings of female-headed et al. (2018) show that mobile money households and shifted their occupational interventions in Tanzania enabled women to choice from agriculture to business. save substantially more, and business training bolstered this effect. Although the mobile Bank accounts can also boost the performance money service and additional savings did of existing female-owned businesses. Dupas not translate into greater investment, sales, and Robinson (2013) offered non-interest- or profit of the primary business, there is bearing savings accounts to market vendors, some evidence that women expanded into a who are predominantly female, and to male profitable secondary business.77 motorbike drivers in Kenya. Despite the large withdrawal fees associated with the accounts, Several other studies have explored 87 percent of women took up the accounts interventions to boost savings without and 41 percent made at least two transactions focusing on entrepreneurs. However, the during the first six months. The accounts lessons learned can serve as inspiration substantially increased women’s savings and for policy makers who would like to investment in their businesses as well as their promote savings as a way to foster female private expenditures. In contrast, men were entrepreneurship. In a study of cash-crop less likely to take up and use the accounts farmers in Malawi, some farmers had the and did not increase their total savings, opportunity to have the proceeds of their cash- suggesting that women may face negative crop harvest deposited directly into new bank private returns to money saved informally.73 accounts in their names. Receiving payments Offering bank accounts to female entrepreneurs in the account led to higher savings in the also showed promise, according to a study months preceding the next planting season, from Malawi. Female entrepreneurs who were raised input usage in that planting season, offered an information session on opening a and led to increases in crop sale proceeds and business bank account as part of a package of household expenditures.78 Future research interventions aiming to support formalization could test whether a similar intervention could increased their use of business bank accounts be effective to help female entrepreneurs who and insurance, saved more, and separated typically sell their products or services to a few household and business money.74 These short- larger firms in the value chain. term impacts translated into large impacts on Commitment devices—defined as “any women’s sales and profits.75 arrangement, entered into by an individual, Training that helps female entrepreneurs with the aim of making it easier to fulfill understand new mobile money or agent his or her own future plans”—used in a banking products can boost women’s savings variety of contexts, including for promoting and has potential to boost their business savings, have been shown to shift individual performance. In Indonesian villages where behavior.79 A commitment savings product in agent banking services were introduced, the Philippines that restricted clients’ access 117 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy to deposits as per their instructions upon these results were likely not driven by peer- opening the account increased the average pressure alone. Feedback text messages that savings balances by 81 percentage points provided recipients with information about after 12 months.80 These commitment-savings their own achievements toward their savings products had longer-term positive impacts goals and the savings of others had similar on women’s decision-making power and impacts.82 While simple informational nudges, purchases of female-oriented durable goods, such as feedback or reminder text messages, particularly for women with lower levels hold promise, less intensive nudges may not of decision-making authority at baseline.81 have a durable impact. In Nigeria, Coville et Deposit collectors who come to households al. (2019) found that an edutainment film or businesses to collect deposits for savings about the importance of savings increased the accounts in banks or microfinance institutions likelihood that individuals opened a savings can simultaneously act as a loose commitment account at the end of the film, but the effects device and reduce transaction costs related dissipated quickly with no detectable impacts to using formal bank accounts. Ashraf et al. four months later. Interestingly, household (2006) show that offering deposit collection savings is relatively inelastic with regard to the services substantially increased savings in the interest rate offered,83 so offering high interest- Philippines, and there is suggestive evidence rate savings accounts may not be an effective that intra-household bargaining issues can way to boost savings. encourage married women to adopt this type of service. Nevertheless, commitment savings Summary of promising interventions products can also have negative repercussions Global evidence suggests several types of if individuals increase their debt levels to offset interventions that policy makers in Southeast their savings commitment, so the products Asia may want to consider to address the may not be adapted to all individuals. Buehren financing gaps that contribute to the gender gap et al. (2018) find that salaried workers in in microenterprise profits and support SME entry Ghana with a worse history of overdrawing and development. Table 3 provides a summary their bank accounts increased their debt levels of the global evidence described above. to offset savings commitments, while those with fewer issues of overdrawing their bank accounts at baseline saved more both during and after the commitment period without increasing their debt. Behavioral nudges also hold promise for increasing savings and helping individuals stick with their savings goals. Microfinance clients in Chile who were invited to savings groups that involved public goal setting, monitoring in group, and non-financial rewards significantly increased their savings—the number of deposits grew 3.7-fold and average savings balances almost doubled. However, 118 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy Table 3 Summary of global evidence of programs to improve entrepreneurs’ access to capital INTERVENTION RELEVANCE SUMMARY OF EVIDENCE EFFECTIVENESS DESCRIPTION MICRO SME Microfinance can generate small, positive changes but MICROFINANCE are unlikely to lead to transformative impacts for female       entrepreneurs. OVERCOMING Although there is a paucity of rigorous evidence, COLLATERAL psychometric testing, cash flow loans, and credit guarantees       CONSTRAINTS hold potential to support female entrepreneurs. SUBSIDIZED LOANS, While these programs can help credit-constrained firms DIRECTED LENDING, access credit, design and implementation are critical to       AND BLENDED ensure proper targeting and avoid distorting the market. FINANCE Cash grants have overall not succeeded in boosting the performance of women’s microbusinesses, often due Cash UNCONDITIONAL to issues of diversion of funds from their businesses.     GRANTS In-kind grants may hold more potential for female microentrepreneurs, but more evidence is needed. In-Kind Business plan competitions offering cash prizes have been proven to increase high growth entrepreneurship BUSINESS PLAN for both men and women; however, the design should       COMPETITIONS include specific targeting to women, such as female-only competitions. Savings interventions have helped increase investment and SAVINGS PROMOTION       business performance of women’s microbusinesses. RELEVANCE Strong relevance Somewhat relevant Not relevant EFFECTIVENESS  emonstrated effectiveness D  ixed or limited evidence M  inimally or not effective M 5.2.3 Ensuring gender- and Sri Lanka did not generate returns to inclusive interventions to their businesses because the money was improve access to capital invested in their husbands’ businesses,84 whereas in-kind cash grants, which are harder Effective capital-related interventions to divert from the intended business, were must take into account gender imbalances more effective than cash grants for boosting in intra-household bargaining power and women’s businesses in Ghana.85 Moreover, normative pressures for women to divert women in several contexts have demonstrated capital flows toward the needs of their an interest in financial products that help households or other members of their family. them protect their individual interests from the Ignoring these constraints can reduce the demands of others, even if these come at a effectiveness of interventions for female cost. Dupas and Robinson (2013) hypothesize entrepreneurs. For example, cash grants to that female market vendors were willing to save female microentrepreneurs in India, Ghana, in formal accounts even though they incurred 119 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy a negative interest rate when considering both unknown, literature from other countries has inflation and withdrawal fees because they shown that women have a lower willingness may face trouble saving on their own and can to compete than men,89 even when they face negative private returns to money saved are equally qualified.90 Lower willingness to informally that is more easily shared under compete may be linked with lower levels of pressure. Similarly, Schaner (2017) shows that self-confidence, which can be perpetuated by ATM cards that reduced transaction costs gender stereotypes. Nevertheless, women were of account use both in terms of money and more likely to enter competition during a lab convenience only increased account usage experiment in Kenya when they were only faced of male-owned and joint accounts of married with female competitors than when faced with couples. Married women were less likely to mixed-gender competitors.91 use the accounts when they became more Rather than avoid competition, program design accessible, especially when they had lower should consider female-only competitions or initial bargaining power. An examination of designing a communications and recruitment the mechanisms suggests that this result is campaign that encourages women to apply. driven by the interpersonal constraints that For example, communications could include married women with low bargaining power female role models to help young women see face in protecting their personal savings from themselves as potential beneficiaries in the external demands.86 Similarly, evidence from program or offer psychological-based training an intervention offering commitment savings aiming to boost self-confidence during capacity devices to men and women in the Philippines building sessions. suggests that financial products that shield income from external demands can boost Social norms and gender stereotypes may savings and increase decision-making power of also affect women’s experience with program women with lower intra-household bargaining implementors or successful receipt of at baseline.87 Capital-related interventions need programs once they decide to apply. Women to account for the risk that support aimed at are often perceived as less competent female entrepreneurs will be diverted to other managers.92 If credit officers or judges in household businesses or needs by enabling a business plan competition hold negative financial information to remain private and gender stereotypes about women’s managerial making capital injections more difficult to abilities, unconscious bias may affect their divert, for example through in-kind provisions or assessments of women’s applications or commitment devices. business plans and potentially restrict women’s access to the programs. Removing the name Social norms and stereotypes may not only and gender of the applicant from the assessed affect whether capital targeted at women’s copies of applications may lower the risk of businesses are indeed invested in those bias. Concealing the gender of the applicant businesses: they may also impact women’s has proven to boost women’s successful propensity to self-select into programs. integration in other domains where gender Women represented only 18 percent of stereotypes persisted.93 Alternatively, policy applicants in the first round of the YouWiN! makers may consider well-designed incentives business plan competition in Nigeria.88 While for decision-makers who have demonstrated the reason women were less likely to apply is 120 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy gender equity in program execution.94 Moreover, apply or participate in the program, safety programs should ensure that anti-harassment concerns, mobility restrictions, and time- and anti-discrimination policies are well constraints may disproportionately exclude developed, clearly communicated about to women from participating. Moreover, eligibility both implementors and potential beneficiaries, requirements should ensure that programs do and that well-functioning grievance redress not inadvertently exclude women. Products mechanisms are in place. that require collateral may exclude female entrepreneurs who have less access to Understanding the local context of gender collateral. In addition, programs that target norms and intra-household decision-making specific sectors of activity may inadvertently is key for targeting interventions to women exclude women due to sectoral segregation, entrepreneurs. Gender norms regarding unless a gendered approach to design is employment, intra-household bargaining, adopted. and domestic labor may constrain the time, capital, and other resources available to Finally, it is critical to consider the types of female entrepreneurs, but may also provide financial products offered to ensure they meet opportunities for intervention. Women are the unique needs of female entrepreneurs. expected to be the primary decision-makers This report showed that female entrepreneurs with regard to household finances in countries may face different types of capital constraints including the Philippines,95 Indonesia,96 and than men—female-owned MSMEs in Vietnam Vietnam,97 which may provide avenues for were more likely than men to be constrained interventions aimed at increasing financial in terms of financial capital (including cash literacy and providing new financial options and receivables). Female-owned SMEs were for female entrepreneurs. In a randomized also more likely than male-owned SMEs experiment in the Philippines, men whose wives to face land-related capital constraints. controlled household financial decision-making However, considering physical capital, female- were more likely to invest money in household owned SMEs face similar and female-owned savings when their financial choices were microbusinesses face fewer constraints than observable, as were women whose husbands men. As such, interventions that target only controlled financial decision-making.98 Women physical capital investments may not address were also more likely to take up a commitment the specific financial constraints that female savings product, which strongly boosted MSMEs face in Vietnam. Products tailored household savings. Interventions targeted at to working capital needs would better meet female entrepreneurs may also have positive the specific needs of women’s MSMEs in impacts on social norms and intra-household Vietnam. In addition, female SME owners would bargaining: a mobile banking and financial skills benefit from tailored products to support land training intervention in Indonesia also increased purchase and lease. During the design stage women’s household decision-making power.99 of an intervention, it is critical to ensure that the product offering is tailored to the needs of As with training, the logistical arrangements female entrepreneurs in that context. Future for capital interventions should also consider interventions could also work with banks gender-specific constraints. If potential to target female entrepreneurs specifically, beneficiaries must travel long distances to including providing additional information on 121 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy the financial needs of female SME owners and the financial products that may be of interest to them. In Cambodia, where the majority of bank loans are made to women, a study found that banks are generally unaware of differences between female- and male-owned SMEs, but that there is interest in understanding their specific needs to become more competitive.100 122 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy 5.3 Interventions to support hiring and managing workers 5.3.1 Rationale and targeting hiring additional workers. This lack of experience and attention to the input prevents Among both microbusinesses and SMEs, entrepreneurs from learning that hiring an women’s businesses are smaller than men’s additional worker could be profitable.106 Given businesses in Southeast Asia, suggesting that women have smaller businesses than men, there are gender-specific constraints to hiring, they are less likely to have had experience hiring managing, and retaining workers. Gender additional workers and may be more likely to differences in the number of workers are linked underestimate how hiring additional workers with the gender profit gap for microenterprises, could boost their business performance. and female microentrepreneurs are operating Labor market frictions can also lead firms to below their optimal size to maximize their operate at a sub-optimal size. Costs related production value in Vietnam. Moreover, to identifying, hiring, and training workers may women are under-represented among SMEs, be prohibitive, especially in contexts with high likely because of gender-specific constraints labor turnover. Imperfect information about to opening or growing into SMEs. Women job candidates’ qualifications can also make it may operate smaller businesses due to time difficult to identify the right worker for the job. constraints related to domestic responsibilities. These costs may be higher for women, who are Alternatively, gender stereotypes that question often more time constrained and have more women’s managerial skills101 can make it narrow networks. harder for women to hire and retain workers.102 Global evidence has revealed some promising If women internalize this notion, they may interventions for addressing informational have lower self-efficacy about their ability to constraints and labor market frictions, manage workers.103 Hesitancy to hire workers which may be relevant for supporting female due to a lack of self-confidence can create a entrepreneurs in Southeast Asia. Nevertheless, negative feedback loop, as women will not learn most studies focus on male-owned firms or do that through experience that they are indeed not report gender-disaggregated results, and capable.104 there are relatively few interventions specifically Economic theory provides a basis for aiming to boost firm size that have been understanding why firm owners may not hire rigorously tested. For example, simultaneously workers, even when it may be profitable to reducing challenges to identifying qualified do so, and gender may interact with several workers and offering subsidies to hire them of these possible constraints. One possibility has proven impactful, but interventions that is that entrepreneurs do not have proper only alleviate search frictions have not been information about the potential returns to rigorously evaluated. The evidence presented in hiring additional workers in their business. this section should be considered as promising They may lack information about their own options, but more research is needed to entrepreneurial ability105 or have little experience understand how effective such interventions 123 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy can be for women’s businesses and to test increased employment in MSMEs in 15 out of additional types of interventions. Moreover, 36 studies included in their review, including policy recommendations can be better tailored four out of six studies of SMEs. to the specific needs of Southeast Asian In addition to formal channels for information women if more research is done on the specific exchange, peer-to-peer learning can also reasons female entrepreneurs in Southeast help MSMEs overcome informational barriers Asia are not hiring more workers. related to HR management. Chatterji et al. (2018) find that firms in India that received 5.3.2 Evidence of promising advice from peers with formal HR management procedures—such as consistently setting goals, labor-related interventions providing feedback, coordinating employees across various tasks—grew 28 percent larger Improving knowledge of labor in terms of the number of employees and were laws and good HR practices 10 percentage points less likely to fail two MSMEs may lack knowledge of how to comply years after the intervention than firms who with labor regulations or how to hire, manage, received advice from peers with an informal and retain workers. The analysis conducted for approach to HR management. Nevertheless, this report shows that generally knowledge of firms may undervalue the importance of laws and regulations is relatively low among training and informational barriers related to HR MSMEs in Vietnam, and these knowledge gaps management. For example, garment factories are larger for female-owned businesses and for in Bangladesh were less likely to demonstrate microbusinesses. If MSMEs are unsure whether interest in a training program and were more they will be able to easily comply with labor sensitive to the program’s price when offered laws and regulations or whether those laws and modules related to HR management, leadership, regulations may negatively impact their ability and social compliance than when offered to adapt their labor needs to changes in their modules on production planning and process or business, they may be hesitant to expand their processes to enhance quality.107 businesses. Wage subsidies Formal interventions to address informational barriers can boost employment in SMEs. In a Wage subsidies have been shown to increase randomized controlled trial of SMEs in South employment during the subsidy period;108 Africa, Bertrand and Crépon (forthcoming) however, these short-term impacts can found that a 21-week membership to a labor dissipate after the subsidy ends.109 Wage law club that shared biweekly newsletters subsidies have the potential to address some covering labor law and HR management informational barriers as well as alleviate some increased employment by between 12 and of the transaction costs associated with hiring 15 percent. Aligned with the possibility of and training new workers. For firms that lack informational issues, some business training information about the potential returns to an programs have increased employment without additional worker or are concerned they may necessarily targeting it specifically. Grimm and not be able to effectively manage workers, Paffhausen (2015) find that business training a wage subsidy can incentivize them to take 124 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy the risk and learn about the potential returns subsidies for consulting services. Outsourcing through experience. significantly increased firm employment, profits, and sales two years after the intervention Nevertheless, the design and targeting of started. Moreover, firms that received insourcing, wage subsidy interventions are critical for outsourcing, or consulting subsidies were more their effectiveness. Although on average De likely to have purchased a business service again Mel et al. (2019) found that wage incentives in the future. Helping firms insource or outsource for microenterprises in Sri Lanka did not workers proved more effective than business have a durable impact on the employment or training and more cost-effective than consulting profitability of male-owned firms on average, for boosting firm growth.110 they found that subsidies had more durable effects for manufacturing firms. Aligned with Summary of promising interventions this finding, Bruhn (2020) finds that wage subsidies for Mexican manufacturing firms Global evidence provides some promising during the 2009 financial crisis increased ideas, summarized in Table 4, for how to employment by 18 percent after the program support female MSMEs in Southeast Asia ended, and this impact increased over time. to hire, manage, and retain more workers. After reviewing several studies, Grimm and However, more research is needed to test Paffhausen (2015) conclude that “business how effective such solutions can be when support services and targeted subsidies can specifically targeting women. Moreover, more contribute to employment generation if they research is needed to better understand are demand-driven, tailor-made and focused.” the underlying reasons for the labor-related Nevertheless, wage subsidy programs can constraints that female entrepreneurs in be very expensive, so additional research is Southeast Asia face and to design and warranted to understand whether they can rigorously test alternative types of interventions effectively address the labor challenges that to address these constraints. female MSMEs face in Southeast Asia and whether they would need to be coupled with additional support. Programs that simultaneously address informational challenges in identifying qualified workers and offer subsidies may be more impactful. A program for MSMEs with between two and 15 workers in Nigeria created a marketplace of service providers and potential employees and offered subsidies for some firms to hire an employee full time (insourcing) and for others to hire someone to work one day per week for the firm on a specific task and the rest of the time for others (outsourcing). These interventions were compared with a traditional business training program and with 125 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy Table 4 Summary of global evidence of programs to support hiring and managing workers INTERVENTION RELEVANCE SUMMARY OF EVIDENCE EFFECTIVENESS DESCRIPTION MICRO SME IMPROVING KNOWLEDGE OF LABOR LAWS AND Limited rigorous evidence is available, but existing evidence GOOD HUMAN seems promising.       RESOURCES PRACTICES A few studies show that wage subsidies may have impacts if they are tailored and targeted to specific types of firms; WAGE SUBSIDIES however, more evidence is needed on how they can impact       female-owned firms. Combining subsidies with support to identify qualified workers may be more impactful. RELEVANCE Strong relevance Somewhat relevant Not relevant EFFECTIVENESS  emonstrated effectiveness D  ixed or limited evidence M  inimally or not effective M 5.3.3 Ensuring gender- favor male entrepreneurs who may hear about inclusive labor-related the program faster or have more time to apply interventions immediately. In addition to ensuring equal access, content Women may be more constrained in terms should be gender-sensitive, and design must of access to information, time, and working mitigate risks of gender-based violence capital, so targeting, application, and selection and sexual harassment. Where possible, processes must enable equal access. informational interventions can address Information campaigns should be designed in a confidence and entrepreneurial self-efficacy, way to target female entrepreneurs, for example for example, by using female role models by working through associations of female or mentors to deliver messages. Capacity entrepreneurs or partnering with banks or building programs and those that support microfinance institutions with a large clientele worker placement should assess the risk of female entrepreneurs. Advertising materials of harassment and gender-based violence, should include female role models when create clear policies and procedures designed possible, so female entrepreneurs can see to mitigate the risk, and ensure appropriate themselves as potential beneficiaries. Selection grievance redress mechanisms to address any criteria should also ensure that it does not violation of the policies and procedures. inadvertently make it more difficult for women to participate when in a situation of limited resources. For example, if women face greater time constraints or informational barriers, a first-come-first-served model may inadvertently 126 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy 5.4 Interventions to alleviate time constraints related to domestic work 5.4.1 Rationale and targeting report their business as their primary activity have higher profits. Domestic responsibilities Policies to help reduce work-family conflict may also affect key strategic business would likely support a broad range of female decisions, such as the location or sector of the entrepreneurs through different channels. business, which can in turn affect business First, they can help address issues of selection performance. Indeed, as discussed in Section into entrepreneurship. Section 2.1.1 shows that 3.1.2, working from home is associated with domestic responsibilities are associated with lower profits of microenterprises in Indonesia, lower labor force participation, so alleviating Vietnam, and Lao PDR, and women are more time constraints related to domestic work could likely to operate their businesses from home. enable some women to enter entrepreneurship Global evidence suggests various policy by enabling their labor market entry. On the tools that can lower women’s domestic other hand, some women who enter the labor workload. Given the ways in which domestic force become entrepreneurs due to the greater work influences women’s engagement in flexibility of hours and location that it can entrepreneurship and ability to dedicate time to afford, which can facilitate combining market their businesses, such policies hold potential and domestic work. This may lead some for enabling female micro, small, and medium women who would rather engage in wage work entrepreneurship in Southeast Asia. to running businesses, which may affect their motivation to grow their businesses. Policies Policy makers have an interest in investing in that reduce the work-family conflict could hence such policies because they promote values support those who choose entrepreneurship of equity and can unleash untapped growth out of necessity to engage in other activities, potential. To the extent that domestic work thereby increasing the relative proportion of constrains women’s choices, it impedes their female entrepreneurs who are opportunity- and ability to exercise complete agency over their growth-oriented. lives, which is an important development outcome as well as a means to development.111 Beyond affecting women’s ability and/or Moreover, as discussed above, time constraints motivation to enter entrepreneurship, policies due to domestic workload orient women’s to lower the time women spend on domestic choices and can lead them to choose labor work may boost their business performance. market activities that are not aligned with their As discussed in Section 3.1.2, female skills or preferences. Existing evidence shows microentrepreneurs in households that likely that such misallocation of talent poses a drag have a greater need for time spent on domestic on economic growth and development.112 work, such as those with small children or who lack time-saving household infrastructure, also have lower profits. Conversely, women who 127 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy 5.4.2 Evidence of promising simultaneously increase parents’ confidence interventions to reduce in using the centers, and hence their uptake, women’s domestic workload as well as improve development outcomes for children. Such regulations and support should ensure that the paid jobs created in the care Access to quality, affordable childcare sector offer decent working conditions, which Access to quality, affordable childcare reduces not only ensures protection of basic worker the amount of time women spend on unpaid rights but also can improve the quality of care work and supports their ability to engage services offered. in productive work. There is a solid evidence Second, governments can help support funding base that childcare services boost women’s for childcare services through a mix of policy labor force participation in low- and middle- tools, such as funding the creation of public income countries.113 Childcare availability childcare centers, support for community- can likewise increase women’s engagement based childcare solutions, or subsidies in entrepreneurship. In China, a 1 percentage offered either directly to parents or to private point increase in access to affordable care providers. Although efforts to support childcare increased women’s engagement in employer-based childcare models can support entrepreneurship by 0.47 percentage points.114 women who work in wage jobs, most female Nevertheless, if women engage in MSME owners would be excluded from these entrepreneurship due to the need for flexible types of programs. Reliable public funding for working arrangements, increased availability childcare providers may have the added benefit of childcare may lower engagement in of giving childcare providers the stability of entrepreneurship from some women who income they need to invest in initiatives to transition to wage work.115 Dang et al. (2019) boost the quality of their services.117 find that childcare in Vietnam increased the Special attention is warranted to support probability that women have a wage job access to high-quality childcare services by 38 percentage points and reduced self- for children under the age of three, as the employment in farm work. Despite evidence availability and use of childcare services for that a lack of childcare is linked with women’s children ages 0-3 is particularly low: 53 out of lower profits,116 there is no rigorous study to 87 countries with available data have gross date testing the impact of access to childcare enrollment rates for children in this age group on the profitability of women’s businesses. of less than 20 percent.118 Governments can play two important roles to increase the availability of accessible, Time-saving infrastructure affordable, high-quality childcare. First, Beyond childcare services, support for the governments can set quality standards and development of infrastructure that reduces the regulations for the childcare sector, support time and labor needed to complete domestic training of childcare workers, and enforce tasks could support female entrepreneurs adherence to standards and regulations. with managing business and household Effective support to ensure that childcare needs. Although a limited number of studies services deliver high-quality services can examine causal impacts, existing evidence 128 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy supports the link between improved household translate into improved business outcomes infrastructure such as drinking water inside for female entrepreneurs. It is possible that the house, time-saving cooking technologies, marginal time savings may not be sufficient or other household appliances and a reduction to close gender gaps in opportunities and in women’s time spent on domestic chores. business performance. Devoto et al. (2012) find that facilitating access to an in-house water connection greatly Shifting norms for a more gender- decreases the time households spend fetching equitable distribution of household work water as well as conflict within the household Interventions to support a reallocation within and with neighbors on water-related issues in the household can be a nice complement urban Morocco. However, the time savings from to care services offered by non-household improved water infrastructure did not boost members and access to time-saving household labor force participation, and households spent infrastructure. The design of parental leave more time on leisure activities.119 On the other policies can encourage men to take a more hand, a program in China to lower the cost active caregiving role in the family and may of certain types of durable goods, including shift social norms about gender and caregiving a refrigerator or a washing machine, did both in the longer term. Paternity leave policies can decrease the time women spent on household signal that fathers should be spending time production by 1.8 hours per week and increase with children and can enable behavioral shifts. women’s engagement in market work. Eligibility A descriptive study from four OECD countries for the durable good rebate program increased shows that fathers who take leave, especially the predicted probability of married women’s of two weeks or more, are more likely to carry labor force participation by 10 to 15 percent, out childcare-related activities when children and eligible women increased their working are young,124 and longitudinal analysis suggests hours by 4.9 hours per week.120 Aligned with this that longer periods of paternity leave are finding, a randomized controlled trial offering associated with more frequent engagement improved cookstoves to pregnant women in developmental tasks and caretaking both in Ghana found that LPG stoves decreased when children are infants and during the first the amount of time women spent collecting few years of children’s lives.125 Beyond the firewood and cooking each meal. During the existence of paternity leave, the characteristics study period, women offered free LPG stoves of parental leave policies can influence the also spent more time on income generating gendered division of labor. Studying parental activities than those in the control group who leave policies in 21 European countries, Castro- continued using traditional three-stone fires García and Pazos-Moran (2015) conclude that for cooking.121 Non-causal evidence also parental leave policies should include equal, supports the findings that improved cookstoves nontransferable, and well-paid leave for each decrease the amount of time women spend on parent. Although parental leave policies provide cooking and related tasks,122 and household one policy instrument to shift norms and appliances more generally are associated with behavior, they are unlikely on their own to be greater female labor force participation.123 sufficient in contexts that have not yet achieved Nevertheless, additional evidence is needed widespread maternity leave protections and to explore whether such time-saving impacts have a sizeable informal sector. 129 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy Other interventions to engage men and communication between couples. Men also boys can improve gender relations within reported a greater increase in household tasks; the household and lead to a more equitable however, women did not report their spouses gendered division of labor. The Bandebereho, being more involved in household tasks.128 or “role model,” program in Rwanda included While interventions to engage men either in 15 facilitated sessions for groups of 12 men’s discussion groups, couples interventions, men that addressed issues surrounding or community-level interventions seems able caregiving, gender and power, communication to shift men’s participation in household tasks, and decision-making, and intimate partner there is currently no evidence on whether or violence. Seven sessions were dedicated how these shifts could improve outcomes for only for men and covered becoming a father, female entrepreneurs. caring for a baby, the impact of their own parents, identifying violence, resolving conflict, Summary of promising interventions alcohol and drug abuse, and reflections on Global evidence suggests several types of how to become a more involved father. Both interventions that policy makers in Southeast the men and their wives were invited to the Asia may want to consider addressing how eight other sessions that covered gender the unequal distribution of domestic work in equality, pregnancy, supporting a pregnant households affects women’s ability to engage partner, childbirth, family planning, gender- and compete in entrepreneurship. Table 5 based violence, raising children, and sharing provides a summary of the global evidence responsibilities at home. The intervention described above. lowered physical and sexual intimate partner violence, improved reproductive health outcomes—including attendance and male accompaniment at antenatal care and use of contraceptives—and lowered men’s dominance in intra-household decision-making. The intervention also increased men’s participation in childcare and household tasks.126 Aligned with this finding, a 16-session curriculum for men’s discussion groups in Côte d’Ivoire significantly increased men’s participation in domestic tasks typically done by women and improved their ability to control their hostility and manage conflict. There is also suggestive evidence that the intervention lowered levels of physical and sexual intimate partner violence.127 Similarly, in Uganda, a community mobilization intervention that aimed to shift harmful social norms and address power imbalances between women and men that perpetuate violence, HIV risk, and inequitable relationships increased joint decision-making and more open 130 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy Table 5 Summary of global evidence of programs to alleviate women’s time constraints related to domestic work INTERVENTION RELEVANCE SUMMARY OF EVIDENCE EFFECTIVENESS DESCRIPTION MICRO SME There is a solid evidence base that childcare can increase women’s labor force participation; however, more IMPROVING ACCESS evidence is needed to understand the impacts on female TO AFFORDABLE,       entrepreneurship. More evidence is also needed on the QUALITY CHILDCARE ideal type of childcare to support female entrepreneurs in different country contexts. Evidence shows that improved household infrastructure, such as in-house drinking water, time-saving cooking SUPPORTING technologies, and other time-saving appliances reduce time ACCESS TO TIME- spent on domestic tasks. However, more evidence is needed SAVING, IMPROVED       to understand the impact of these time savings on female HOUSEHOLD entrepreneurs and whether the amount of time saved is INFRASTRUCTURE sufficient to translate into meaningful impacts for female entrepreneurs. Well-designed parental leave policies that include sufficient, highly paid, nontransferable leave to both parents can boost PARENTAL LEAVE men’s involvement in childcare. However, this may not be a POLICIES policy priority to support female entrepreneurs in contexts       where leave protections are not universal to all workers, such as those in the informal sector. Programs that engage men in discussions related to gender- ENGAGING MEN related issues have shown increases in joint decision- AND SHIFTING making and an increase in men’s involvement in domestic INTRA-HOUSEHOLD       tasks in addition to other development outcomes affecting ALLOCATION OF women. More evidence is needed on how such programs DOMESTIC WORK can support female entrepreneurs more specifically. RELEVANCE Strong relevance Somewhat relevant Not relevant EFFECTIVENESS  emonstrated effectiveness D  ixed or limited evidence M  inimally or not effective M 5.5 The path forward Policy makers have a diverse set of tools perspective or even reported heterogeneous and actions that they can take to support impacts by gender. As discussed throughout female entrepreneurs in Southeast Asia. this section, gender may moderate access to However, knowledge gaps remain in all areas and the resulting effectiveness of different about the relative effectiveness of different programs. interventions, the optimal way to implement A growing body of evidence sheds light them, and how well they translate to different on different policies to enhance the skills contexts. In particular, very few rigorous of MSMEs. Classroom-based training, in studies have been carried out to identify what particular when using more innovative works for MSMEs in Southeast Asia, and many curriculum, holds promise for boosting evaluations have not integrated a gender microenterprise performance, and business 131 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy consulting interventions has the potential to having an SME. In addition, in-kind grants and boost the productivity of SMEs. Social learning directed lending have potential to support interventions that promote peer-to-peer learning microenterprises and SMEs, respectively. and network formation hold promise for various Nevertheless, more evidence is needed to test types of entrepreneurs. capital-related interventions in the Southeast However, more research is needed to Asian context, inform intervention design, understand under which conditions skills- and test more innovative solutions to capital enhancing interventions best meet female constraints. Most evidence on business entrepreneurs’ needs and to test alternative, plan competitions comes from Sub-Saharan more scalable solutions. There is limited Africa, so it would be important to rigorously evidence on how social-learning, consulting, or evaluate this type of intervention elsewhere mentoring interventions may work differently to ensure that it functions well in diverse for male or female entrepreneurs or how contexts. Similarly, more evidence is needed the gender of the consultants, mentors or to demonstrate the effectiveness of in-kind peers may affect the effectiveness of these grants and directed lending in Southeast Asia. interventions for female entrepreneurs. To ensure that directed lending is effective, Moreover, additional evidence is needed it must be well-targeted to meet firms that on how to address market failures and are truly credit constrained, so additional promote sustainable solutions. Relatively research should provide guidance on how to few studies rigorously test why firms do not effectively target such interventions. Although pay for these types of Business Development global evidence suggests that savings Solutions themselves,129 or examine market- promotion can support microentrepreneurs, based solutions such as having firms pay for more evidence is needed about which specific training130 or testing insourcing/outsourcing interventions are the most effective for helping types of models for acquiring skills.131 female entrepreneurs save and invest in their businesses, given that several promising Global evidence sheds light on the types interventions have not yet been specifically of capital-related interventions that hold tested with entrepreneurs. Additional research potential for supporting female entrepreneurs. is also needed to rigorously test the effect of Savings promotion has been shown to boost psychometric testing, cash flow loans, credit investment and business performance of guarantees, and movable collateral registries female microentrepreneurs, and business plan for helping female entrepreneurs without competitions with cash prizes are effective at collateral to access loans, though initial increasing high growth entrepreneurship for studies suggest they hold potential for both both men and women. Female-only competition microenterprises and SMEs. windows could be a particularly promising solution to explore further given the gender Much more evidence is needed on effective gap in participation in SMEs in Southeast Asia. policies to support hiring and managing Notably, business plan competitions for new workers. Knowledge of labor laws and good businesses can help close the gender gap human resource management appears in entry into ownership of growth-oriented promising and combining short-term subsidies businesses and increases the probability of for insourcing or outsourcing labor with the 132 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy development of a marketplace with qualified also test whether less-intensive, more scalable workers holds potential. Nevertheless, interventions may be powerful enough to available studies have not examined gender- generate change. disaggregated impacts, and more evidence Future research should also assess not only is needed on whether interventions that only whether interventions are effective but the reduce informational barriers to identifying extent to which they are cost-effective. In a qualified workers without any subsidies can context with limited resources, information on be effective. Another area for future research cost-effectiveness is critical to choosing the would be to explore psychological interventions package of interventions that will maximize that can boost women’s confidence and help impact for a certain amount of funding. them overcome stereotype threat associated Because it is recommended to adopt policies with negative stereotypes of women’s that address multiple constraints, a deeper managerial abilities. Such interventions may understanding of costs is essential to ensure be stand alone or offered in combination with the budget feasibility of proposed interventions. relaxing other constraints to hiring workers. Although a number of knowledge gaps Global evidence shows promising interventions remain, there is much that policy makers for reducing women’s time spent on domestic can already do to improve MSME policies for work; however, no studies have specifically female entrepreneurs. Interventions discussed studied the impacts of such interventions in Sections 5.1 through 5.4 that have already when targeted to female entrepreneurs. proven successful or promising can be scaled Access to affordable, quality childcare is likely up or introduced in new places. As much very promising, as there is a solid evidence of this research is based on evidence from base that childcare increases women’s labor outside Southeast Asia, program implementers force participation. However, more evidence can conduct formative quantitative or is needed to assess demand for childcare qualitative research to adapt internationally services among female entrepreneurs and to proven interventions to the context and needs identify the form of childcare most adapted to of local entrepreneurs. In parallel, policy the needs of female entrepreneurs in different makers can pilot and rigorously evaluate country contexts. Access to time-saving innovations, including new interventions or household infrastructure and interventions design tweaks to implementation. Policy to shift the intra-household allocation of makers and program implementers can domestic work have been shown to reduce partner with researchers to build a rigorous the amount of time that women spend on research agenda into their work. domestic tasks. However, it remains an open question whether the quantity of time released from such interventions would be sufficient to substantially influence the productivity of women’s businesses. Moreover, programs that engage men on gender-related issues typically require fairly intensive planning and support for implementation, which may make them harder to scale. Future research could 133 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy Endnotes 1  steve-Volart 2004; Cuberes and Teignier 2016; Hsieh et al. E 59  Cassano et al., 2013 2019 60 Zia (2008) 2 Cavalcanti and Tavares 2015; Hsieh et al. 2019 61 Prior to 1998, the priority sector included firms with 3 World Bank 2012a less than Rs. 6.5 million in total investment in plant 4 Sen 1999 and machinery, and in January 1998, the definition was 5 Revenga and Dooley 2020 expanded to include firms with investment in plant and 6 Global Entrepreneurship Monitor 2019-2020 machinery of less than Rs. 30 million. In January 2000, the 7  See McKenzie and Woodruff (2014) for a useful review of eligibility criteria changed again, and the threshold was the literature that includes a discussion of market failures lowered to Rs. 10 million. that can justify public investment in skills development 62 Banerjee and Duflo, 2014 ; Kapoor, Ranjan, and programs for entrepreneurs, which is summarized in this Raychaudhuri, 2012 paragraph. 63 Banerjee and Duflo, 2014 8 World Bank 2012b 64  Kapoor, Ranjan, and Raychaudhuri, 2012 9 McKenzie 2020 65  For more information on these programs, see https://www. 10 McKenzie 2020 ifc.org/wps/wcm/connect/Industry_EXT_Content/IFC_ Anderson et al. 2018; Buvinic et al. 2020; McKenzie and 11  External_Corporate_Site/Financial+Institutions/Priorities/ Puerto 2021 Banking_on_Women/ 12 McKenzie and Puerto 2021 66  Ornelas et al. 2019 13 Beaman et al. 2014 67  Bernhardt et al., 2019 14 Dalton et al. 2021 68  Fafchamps et al., 2014 15 Frese and Fay 2001 69 McKenzie, 2017 16 Campos et al. 2017 70 McKenzie, 2015 17 Campos et al. 2018 71 McKenzie, 2017 18 Alibhai et al. 2019 72  Fafchamps and Woodruff, 2014 19 Shankar et al. 2015 73  Dupas and Robinson 2013 20 McKenzie 2020 74  Campos et al. 2015 21 Revenga and Dooley 2020 75  Campos et al. 2018 Arraíz et al. 2019; Campos et al. 2017; Drexler et al. 2014  22  76  Buvinic et al. 2020 23 Anderson and McKenzie, 2020 77  Bastian et al. (2018) 24 Bruhn et al. 2018 78  Brune et al. 2016 25 Anderson and McKenzie, 2020 79  Bryan et al. 2009 26 Brooks et al. 2018 80  Ashraf et al. 2006 27 Valdivia, 2015 81  Ashraf et al. 2010 28 Drexler et al. 2014 82  Kast, Meier, and Pomeranz 2018 29 McKenzie and Puerto, 2021 83 Kast, Meier, Pomeranz 2018 ; Karlan and Zinman 2018 30 Giné and Mansuri, 2020 84 Bernhardt et al. 2019 31 Anderson et al. 2020 85  Fafchamps et al., 2014 32 Lafortune et al., 2018 86 Schaner 2017 Lafortune et al., 2018 ; Valdivia, 2015 33  87 Ashraf, Karlan and Yin 2006; Ashraf, Karlan, and Yin 2010 34 Cai and Szeidl, 2018 88 McKenzie 2015 35 Lafortune et al. 2018 89 Gneezy, Leonard, and List 2009; Niederle and Vesterlund 36 Field et al. 2016 2007, 2008, 2011; Paryavi, Campos, and Santos 2018  37 Venkatesh et al. 2017 90  Niederle and Vesterlund 2011 38 Buvinic and O’Donnell 2019 91  Paryavi, Campos and Santos 2018 39 McKenzie 2020 92 International Finance Corporation 2008; Macchiavello et al, 40 Revenga and Dooley 2020 2015 41 McKenzie and Puerto 2021 93  Goldin and Rouse 2000 42 Campos et al. 2018 94  Buvinic and O’Donnell, 2019 43 Campos et al. 2017 95  Malapit et al 2020, Bayudan-Dacuycuy 2011 44 Beegle and Rubiano-Matulevich 2020 96  Colfer et al 2015 45 Beegle and Rubiano-Matulevich 2020 97 Ngo 2020 ; Huis et al 2020 46 Beegle and Rubiano-Matulevich 2020 98 Ashraf 2009 47 Beegle and Rubiano-Matulevich 2020 99  Buvinic et al 2020 48 Buvinic and O’Donnell 2019 100 International Finance Corporation 2019, 46 49 Stiglitz and Weiss 1981 101 Macchiavello et al 2015 50 Balachandara et al. 2013 102 International Finance Corporation 2008 51 Alibhai et al., 2018 103 While little evidence is available from Southeast Asia, in 52 See Mehra et al., 2012 for a description of different types of Sub-Saharan Africa, the World Bank (2019) shows that microfinance lending. women generally have lower levels of confidence in their Revenga and Dooley, 2020 ; Meager, 2018 ; Dahal and Fiala, 53  entrepreneurial ability than men, especially when it comes 2018 to domains that are typically associated with men. Banerjee, Brez, Duflo, Kinnan, 2017 54  104 Although not applied to a context of women’s lower Banerjee, Duflo, Glennerster, Kinnan, 2015 55  self-efficacy, Hanna, Mullainathan, and Schwartzstein 56 Klinger et al., 2013 (2014) show that a negative feedback loop can arise when 57 Alibhai et al., 2019 business owners do not experiment with an input and as a 58 Alibhai et al., 2019 consequence do not learn that it matters. 134 Enterprising Women | Toward a targeted, gender-informed entrepreneurship policy 105 Jovanovic 1982 106 Hanna, Mullainathan, and Schwartzstein 2014 107 Machiavello et al. 2015 Betcherman et al. 2010; Kangasharju 2007 108  Card and Hyslop 2005; Groh et al. 2016 109  110 Anderson and McKenzie 2020 111 Sen 1999 112 Esteve-Volart 2004; Cuberes and Teignier 2016 113 Causal evidence of how childcare boosts women’s labor force participation is available for Argentina (Berlinski and Galiani, 2007; Berlinski et. al., 2011), Brazil (Paes de Barros et al., 2011), India (Jain, 2016), Kenya (Clark et al., 2019), Indonesia (Halim et al., forthcoming), and Mexico (Ángeles et al., 2011; Calderon, 2014, Padilla-Romo and Cabrera- Hernández, 2018). Dang et al. (2019) do not find impacts of childcare on women’s labor force participation, likely due to high initial rates of female labor force participation, but they do find an increase in income. 114 Wang, 2015 115 Thébaud, 2015 116 Delecourt and Fitzpatrick, 2020 117 Clark et al. 2017 118 International Labor Organization 2018 119 Devoto et al. 2012 120 Tewari and Wang 2019 121 Prah et al. 2021 ADB 2009; Christianensen and Heltberg 2012; Djedje et al. 122  2009; Habermehl 2007, 2008; Malla et al. 2011 123 De V. Cavalcanti and Tavares 2008 124 Huerta et al. 2014 125 Petts and Knoester 2018 126 Doyle et al. 2018 127 Hossain et al., 2014 128 Kyegombe et al. 2014 129 A notable exception includes Macchiavello et al. (2015) Maffioli et al. 2020 130  131 Anderson and McKenzie 2020 135 Enterprising Women | Appendix & References Appendix A: Details of data sources used in the report Type of business Country Survey name Type of survey Survey years Sample size covered Sample of individuals: Household panel data, 152,493 Indonesia Family Life representative of 83 2000, 2007, Microenterprises Survey (IFLS) percent of the Indonesian 2014 Sample of INDONESIA population businesses: 18,022 World Bank Enterprise Cross-sectional firm survey Sample of SME 2015 Survey (WBES) data businesses: 935 Sample of Panel data of SME non- microbusinesses: state manufacturing firms, Both microenterprises 2005, 2007, 5,845 SME Survey (VSMES) representative of firms in and SME, analyzed 2009, 2011, 10 provinces in 2005 when separately 2013, 2015 Sample of SME: the panel was started 7,396 VIETNAM Sample of Vietnam Access to Household panel data, 2008, 2010, individuals: Resources Household representative for 12 Microenterprises 2012 and 2014 39,020 Survey (VARHS) provinces in rural areas World Bank Enterprise Cross-sectional firm survey Sample of SME 2015 Survey (WBES) data businesses: 749 Sample of individuals: Cambodia Socio- Nationally representative 53,968 Economic Survey cross-sectional household Microenterprises 2014 CAMBODIA (CSES) survey data Sample of businesses: 16,014 World Bank Enterprise Cross-sectional firm survey Sample of SME 2016 Survey (WBES) data businesses: 315 Sample of individuals: Laos Expenditure and Nationally representative 43,641 Consumption Survey cross-sectional household Microenterprises 2012, 2013 LAO PDR (LECS) survey data Sample of businesses: 9,236 World Bank Enterprise Cross-sectional firm survey Sample of SME 2018 Survey (WBES) data businesses: 287 Sample of individuals: Timor-Leste Survey Nationally representative 32,083 of Living Standards cross-sectional household Microenterprises 2014 and 2015 TIMOR-LESTE (TLSLS) survey data Sample of businesses: 797 World Bank Enterprise Cross-sectional firm survey Sample of SME 2015 Survey (WBES) data businesses: 121 136 Enterprising Women | Appendix & References Appendix B: Methods and econometric specifications used in the report Overview of methods used The new analysis conducted for this report uses several different methods. OLS regressions: For studying selection into entrepreneurship, OLS regressions show the correlation between individual and household characteristics and the likelihood of being an entrepreneur. For business performance, OLS regressions show the correlation between individual, household, and business characteristics and the performance of the business. In both cases, interacting the gender of the owner with the characteristics shows whether factors linked with engaging in entrepreneurship or with business performance are different for men and women. It is important to note that these regressions do not suggest causality. Fixed effects regressions: Similar to OLS regressions, fixed effects regressions show the correlations between individual, household, and business characteristics and outcomes of interest, including the likelihood of being an entrepreneur and business performance. However, these models also control for time-invariant characteristics that we cannot observe, such as innate ability or preferences, by detecting how changes in an individual’s or business’s characteristics over time lead to shifts in their propensity to entrepreneurship or in their business performance. Events studies: Event studies show how trends in outcomes of interest (such as engaging in entrepreneurship) shift after the occurrence of an event (such as childbirth). Stepwise regressions: Stepwise regressions are used to measure the raw gender gap in performance and to see how it changes as additional controls for personal, household, business, and community characteristics are added. Oaxaca-Blinder decomposition: This method examines how much of the gender gap in performance is due to differences endowments, how much is due to differences in the relationship between these endowments and business performance, how much is due to the interaction between the two, and how much is left unexplained. Cobb-Douglas production function: The Cobb–Douglas production function explores the relationship between the amounts of inputs, including capital and labor, and the amount of output produced by those inputs. It demonstrates whether there are gender differences in marginal returns to capital and labor. Note: All analysis done using the World Bank Enterprise Surveys data includes sampling weights, due to the design of the survey which oversamples certain types of businesses. Weights are not used for other datasets, as the sampling strategies for the other data sets did not oversample certain types of businesses over others. 137 Enterprising Women | Appendix & References Methods and specifications used in Section 1 Model 1.1: OLS regressions The raw gender gap in performance is calculated using the following OLS regression using robust standard errors: yit = β0 + β1 Fi + θt + εit (Eq. 1.1) Where: yit = Performance indicator of the firm run by individual i in time t Fi = Dummy indicating whether individual i is female θt = Time fixed effect, included in regressions using panel data β1 is the coefficient of interest showing the extent to which there is a gender gap in business performance The variable used for the performance indicator depends on the dataset. Annual profits are used when available. Profits are reported directly in the Indonesia dataset and are calculated by subtracting costs from revenues in datasets from Vietnam, Cambodia and Timor-Leste. Datasets that do not include information on profits or in which there are large numbers of missing values for costs impeding a reliable calculation of profits, revenue is used with controls for sector of activity. Annual revenue is used in all cases except microbusinesses in Laos, for which monthly revenue information was available. Because profit and revenue information is often noisy, the logarithmic transformation is used in most cases, as the number of 0 or negative values was minimal. However, in two cases, more than 10 percent of firms had negative profits, so the Inverse Hyperbolic Sine (IHS) transformation was used instead of the logarithmic transformation. To calculate the percentage difference between men and women’s business performance, the coefficient of interest is transformed according to recommendations in Wooldridge (2012) for logarithmic transformations and Bellemare and Wichman (2019) for IHS transformations, by exponentiating the coefficient, subtracting one, then multiplying by 100. 138 Enterprising Women | Appendix & References The gender gap with all controls is calculated using the following OLS regression: yit = β0 + β1 Fi + β2 Iit + β3 Hit + β4 Bit + β5 Cit + θt + εit (Eq. 1.2) Where: Iit = Individual characteristics of entrepreneur i in time t Hit = Household characteristics of entrepreneur i in time t Bit = Business characteristics of entrepreneur i’s business in time t Cit = Community characteristics of entrepreneur i’s community in time t All other variables are as defined in equation 1.1 Table B.1 shows the list of control variables included in regression 1.2 139 Enterprising Women | Appendix & References Table B.1: Controls included in the regression 1.2 by dataset TYPE OF BUSINESS MICROBUSINESS SME VARIABLE KHM IDN LAO TLS VNM KHM IDN LAO TLS VNM INDIVIDUAL CHARACTERISTICS Marital status X X X X Age X X X X Age squared X X X Years of education X X X X Completed lower secondary X X Completed higher secondary X X Technical worker with or without X X certificate Completed college (technical education) X X Years of experience X X X X Business knowledge and practices X X Primary activity is entrepreneurship X Participation in rotating savings and credit X groups Participation in community groups X Ethnic Kinh X X Communist party member X X HOUSEHOLD CHARACTERISTICS Children 0-2 years old in the household X X X X Children 3-5 years old in the household X X X X Elders over 64 years old in the household X X X X Rural/urban dummy X X X X Province X X X X Drinking water inside house X X X X Uses less time-consuming cooking source X X X X Household members need assistance with X daily activities Someone in household with illness or X X X disability Household received transfers from the X X government Income from Individuals X X Dependency ratio X X BUSINESS CHARACTERISTICS Sector of activity X X X X X X X X X X Business started as a microenterprise X X X X Household members participating in X business activity Dummy for having paid workers X X X X 140 Enterprising Women | Appendix & References TYPE OF BUSINESS MICROBUSINESS SME VARIABLE KHM IDN LAO TLS VNM KHM IDN LAO TLS VNM Number of paid workers X X X X X X Dummy for having unpaid workers or X X X X household members working in business Labor turnover X X Difficulties hiring workers X X Business is outside of the home X Business is on residential property X X Age of business X X X X X Land business is on is owned X X Capital is main constraint to growth X X Capital is main constraint to new projects X X Business applied for formal loan X X Loans received from informal sources X X Debt-equity ratio X X Debt-equity ratio squared X X Business has access to road X X Business has access to rail X X Business received government assistance X X Business has contract with government X X X X Business was formal at start-up X X X X Business is formal X X Keeps accounting books X X Accounting books audited X X Introduced new product groups X X Introduced new technology X X Firm pays social insurance X X Firm pays health insurance X X All people in network X X Network frequency (winsorized) X X Ratio of women in network X X Business is a private, domestic enterprise X X X X Business has a website X X X X COMMUNITY CHARACTERISTICS Community has a bus stop X Community has non-motor vehicle X transportation Market is present in village X Note: Country name abbreviations are as follows: Cambodia (KHM), Indonesia (IDN), Lao PDR (LAO), Timor-Leste (TLS), Vietnam (VNM); Characteristics on the individual, household, and the characteristics of the owner’s network are only available for a sub-sample of firms in Vietnam 141 Enterprising Women | Appendix & References Methods and specifications used in Section 2 Model 2.1: Linear probability model with gender interactions The following equation is used to assess the factors linked with men and women’s labor market choices: lmit= α0 + α1 Fi + α2 Iit + α3 Hit + α4 Cit + α5 Iit * Fi + α6 Hit * Fi + α7 Cit * Fi + θt + εit (Eq. 2.1) Where: lmit shows the labor market outcome of interest of individual i in time t All other variables are as defined in equation 1.1 and 1.2 The regression is run with five specifications: 1. lmit takes the value of 1 if the individual is active in the labor force. 2. lmit takes the value of 1 if the individual is engaged in entrepreneurship, including all working-aged individuals 3. lmit takes the value of 1 if the individual is engaged in entrepreneurship, limited to individuals who participate in the labor force 4. lmit takes the value of 1 if the individual is engaged in wage work, including all working-aged individuals 5. lmit takes the value of 1 if the individual is engaged in wage work, limited to individuals who participate in the labor force The correlation between individual, household and community characteristics and labor market outcomes for men is captured by α2, α3, and α4 respectively. Whether or not these associations are statistically significantly different for men and women is captured by α5, α6, and α7. The relationship between women’s individual, household, and community characteristics and their labor market outcomes is measured by the sum of the coefficient for men and coefficient for the interaction term: (α2 + α5), (α3 + α6), and (α4 + α7), respectively. Table B.2 shows the list of control variables included in regression 2.1 142 Enterprising Women | Appendix & References Table B.2: Controls included in the regression 2.1 by dataset VARIABLE KHM IDN LAO TLS VNM INDIVIDUAL CHARACTERISTICS Marital status X X X X X Age X X X X X Years of education X X X X X Participation in rotating savings and credit groups X Participation in community groups X HOUSEHOLD CHARACTERISTICS Children 0-2 years old in the household X X X X X Children 3-5 years old in the household X X X X X Elders over 64 years old in the household X X X X X Rural/urban dummy X X X X X Province X X X X X Drinking water inside house X X X X Household has a private tap X Household cooks with electricity, gas or kerosene X Uses less time-consuming cooking source X X X X Someone in HH with illness or disability X X X Household members need assistance with daily activities X Spent time in housework in the last 12 months X ACCESS TO CAPITAL CHARACTERISTICS Sector of activity X X X X Household nonbusiness asset index X Incomes from Individuals X X Transfers from government or institutions X X X X COMMUNITY CHARACTERISTICS Community has a bus stop X Community has non-motor vehicle transportation X Market is present in village X Note: Country name abbreviations are as follows: Cambodia (KHM), Indonesia (IDN), Lao PDR (LAO), Timor-Leste (TLS), Vietnam (VNM) 143 Enterprising Women | Appendix & References Model 2.2: Linear probability model with individual fixed effects In Indonesia and Vietnam where panel data is available, individual fixed effects are introduced into the regression 2.1 using the equation: lmit = α0 + α2 Iit + α3 Hit + α4 Cit + α5 Iit * Fi + α6 Hit * Fi + α7 Cit * Fi + θt + δi + εt (Eq. 2.2) Where: δi is the individual fixed effect and all other variables are the same as equation 2.1 The coefficients α2, α3, and α4 show the within-individual changes in labor market outcomes for men associated with changes in covariates over time, controlling for other time-variant covariates as well as time-invariant characteristics and preferences. The coefficients α5, α6, and α7 capture whether these relationships are different for men and women, and the relationship for women is measured by the sum of the coefficients for men and the interaction term: (α2 + α5), (α3 + α6), and (α4 + α7) for individual, household, and community level characteristics respectively. Model 2.3 Events study In Indonesia, panel data captures both information on primary activity annually as well as a detailed pregnancy history enabling the analysis of the impact of having children on women’s labor market choices using an event study framework. The probability of employment before and after first childbirth is estimated using the following equation: (t=-2) (t=10) lmit = δi + ∑ YearToBirthit + ∑ YearToBirthit + interviewt + ϵit (t=-6) (t=0) Where: YearToBirthit is a dummy for year t relative to the incidence of the first pregnancy interviewt is a dummy taking a value of 1 if the respondent was interviewed in year t1 All other variables are as previously defined, and standard errors are clustered at the individual level. 1 Other values are coming from a recall module, which may be more affected by recall bias. 144 Enterprising Women | Appendix & References Methods and specifications used in Section 3 Model 3.1 OLS regressions with gender interaction terms The following equation is used to assess the factors linked with men and women’s business performance: yit = γ0 + γ1 Fi + γ2 Iit + γ3 Hit + γ4 Cit + γ5 Bit + γ6 Iit * Fi + γ7 Hit * Fi + γ8 Cit * Fi + γ9 Bit * Fi + θt + εit (Eq. 3.1) Where: yit = Performance indicator of the firm run by individual i in time t Fi = Dummy indicating whether individual i is female Iit = Individual characteristics of entrepreneur i in time t Hit = Household characteristics of entrepreneur i in time t Bit = Business characteristics of entrepreneur i’s business in time t Cit = Community characteristics of entrepreneur i’s community in time t θt = Time fixed effect, included in regressions using panel data The variable used for the performance indicator depends on the dataset and is calculated as described in Model 1.1. The correlation between individual, household, community, and business characteristics and men’s business performance is captured by γ2, γ3, γ4, and γ5 respectively. Whether or not these associations are statistically significantly different for men and women is captured by γ6, γ7, γ8, and γ9. The relationship between women’s individual, household, community, and business characteristics and their business performance is measured by the sum of the coefficient for men and coefficient for the interaction term: (γ2 + γ6), (γ3 + γ7), (γ4 + γ8), and (γ5 + γ9), respectively. The control variables are the same as those used in equation 1.2 and figure in Table B.1. 145 Enterprising Women | Appendix & References Model 3.2 OLS regressions with individual fixed effects In Indonesia and Vietnam, where panel data is available, we also run the regressions with individual or firm-level fixed effects; however, the precise model and specification used is different in Indonesia and Vietnam due to the different structures of the datasets used. In Indonesia, the panel data is constructed by following the individual across years and the businesses they run at different points in time. Because in Indonesia the data do not enable to determine whether the firm is the same firm or a new firm, the fixed effects are individual-level fixed effects. The following regression is used for Indonesia: yit = γ0 + γ2 Iit + γ3 Hit + γ4 Cit + γ5 Bit + γ6 Iit * Fi + γ7 Hit * Fi + γ8 Cit * Fi + γ9 Bit * Fi + δi + θt + εit (Eq. 3.2.1) Where: δi is the individual fixed effect and all other variables are the same as equation 3.1 In Vietnam, the panel data tracks the firm, not the business owner, so firm-level fixed effects are used. Because current business performance is likely strongly linked with past outcomes, including past business performance, the Arellano-Bond estimator is used to obtain more reliable estimates using the data from Vietnam. The Arellano-Bond estimator essentially involves including the lagged dependent variable from the previous period among the covariates in the regression and instrumenting it using lagged dependent variables from earlier time periods (t-2 and feasible deeper lags). Including the lagged dependent variable among the regressors means that the first year of panel data are not used in the estimation of the relationship between regressors and the dependent variable except through being used as a lagged value for the second time period. Because only three years of panel data are used in Indonesia,2 the Arellano-Bond estimator is not used to avoid loss of the number of observations that can be used. For Vietnam, the following regression is used: yit = γ0 + γ1 Fi + γ2 Iit + γ3 Hit + γ4 Cit + γ5 Bit + γ6 Iit * Fi + γ7 Hit * Fi + γ8 Cit * Fi + γ9 Bit * Fi + β10 yit-1 + φi + θt + εit (Eq. 3.2.2) Where: φi is the firm-level fixed effect yit-1 is the lagged dependent variable All other variables are the same as in equation 3.1 For both equations 3.2.1 and 3.2.2, the coefficients γ2, γ3, γ4, and γ5 show how within-individual or within-firm changes in covariates in Indonesia and Vietnam, respectively, is linked with changes in firm productivity for men. Whether or not these associations are statistically significantly different for men and women is captured by γ6, γ7, γ8, and γ9. How changes in women’s in individual women’s individual, household, community, and business characteristics over time are linked to their business performance is measured by the sum of the coefficient for men and coefficient for the interaction term: (γ2 + γ6), (γ3 + γ7), (γ4 + γ8), and (γ5 + γ9), respectively. 2 Although five waves of IFLS data are publicly available, the way household businesses were asked about in the first two waves was substantially different the in the last three waves, so the analysis is based on analysis of waves 3, 4 and 5 only to avoid biasing the estimates. 146 Enterprising Women | Appendix & References Model 3.3 Stepwise regressions Stepwise regressions start by using the following OLS regression using robust standard errors: yit = β0 + β1 Fi + β6 Sit + θt + εit (Eq. 3.3) Where: yit = Performance indicator of the firm run by individual i in time t, defined as in Model 1.1 Fi = Dummy indicating whether individual i is female Sit = Sector of activity of the firm run by individual i in time t θt = Time fixed effect, included in regressions using panel data Stepwise regressions add progressively one variable or group of variables at a time to regression 3.3 to see how the coefficient of interest β1, which captures the extent to which there is a gender gap in business performance, changes. Model 3.4 Oaxaca-Blinder decompositions The Oaxaca-Blinder decomposition seeks to decompose differences in mean outcomes across two groups, breaking them into three components. The decomposition method relies on the following regression equation of the log of business performance (Y, as defined in Model 1.1) for businesses with a male (M) and a female (F) owner: YG0 = XG’βG + ϵG (Eq 3.4.1) Where G stands for the gender of the owner (M or F), X is a vector containing the predictors (business and owner characteristics) and a constant, and β contains the slope parameters and the intercept, and ϵ is the random error component. The gender gap in business performance, D, can be expressed as the difference in outcomes: D = E(YM) – E(YF) (Eq 3.4.2) Where E(YG) is the expected value of the outcome variable for those with the gender G. It is assumed that E(ϵM) = E(ϵF) = 0 and E(βG) = βG, which means equation 3.4.2 can be rewritten:3 D = (XM)’ βM–(XF)’ βF (Eq. 3.4.3) 3 Given the assumptions, E(YG) = E(X’G βG + ϵG) = E(X’G βG ) + E(ϵG) = E(XG)’ βG 147 Enterprising Women | Appendix & References As discussed in Jann (2008), equation 3.4.3 can be rearranged to identify the contribution of group differences in predictors to the overall outcome difference, as follows:4 D = {E(XM) – E(XF)}’ βF + E(XF)’ (βM–βF) + {E(XM ) – E(XF )}’ (βM–βF ) (Eq. 3.4.4) Component 1: Component 2: Component 3: Endowment Differences in Interaction of effect coefficients endowments and coefficients The first component captures the part of the gender gap in business performance that is due to endowments. In other words, it shows the mean increase in women’s business performance that would arise if women’s businesses had the same characteristics as men’s businesses. Comparing this component to the overall wage gap shows the share of the wage gap that is due to women’s differing levels of endowments. The second component captures the part of the gender gap in business performance that is due to the different relationship between women’s endowments and their business performance. Specifically, it quantifies the change in women’s business performance that would arise if women had the same coefficients as men in a regression showing the correlation between characteristics and business performance. The third component captures the interaction of endowments and their association with business performance, measuring the part of the gender gap in business performance that is simultaneously linked with differences in endowments and their coefficients in a regression linking the characteristics to business performance. The Oaxaca-Blinder decomposition method requires 3 strong assumptions.5 First, it uses a partial equilibrium approach, in which observed outcomes of one group are used to construction different scenarios of counterfactual scenarios for the other group. Second, the estimations are based on correlational analysis and cannot be interpreted as causal, as discussed in Fortin, Lemieux, and Firpo (2011). Third, the analysis assumes linear effects and does not take potentially non-linear relationships into account. Despite these limitations, the Oaxaca-Blinder decomposition is very useful, as it helps quantify the relative importance of different potential explanatory factors in understanding differences in means between two populations. 4 See Winsborough and Dickinson (1971), Jones and Kelley (1984) and Daymont and Andrisani (1984) for more details. 5 See Fortin, Lemieux, and Firpo (2011) for a detailed explanation of the different assumptions and identification restrictions. 148 Enterprising Women | Appendix & References Model 3.5 Cobb-Douglas production function The Cobb-Douglas production function is commonly used to estimate the marginal returns to different factors of production. However, correlations between the observable input levels and the unobservable productivity changes and shocks is a source of bias in OLS estimations of productivity functions. One of the approaches to tackle the issue of unobservable productivity shocks is a control function approach in which investment levels are used as a proxy for productivity. Hence, changes in investment levels would reflect changes in productivity of each firm. The basic assumption of such an approach is that the idiosyncratic shock to productivity at time t does not affect the choice of the level of state variables, which are decided at t-1, but only that of free variables. State variables are mostly understood as physical buildings and machinery, while free variables include labor. Assuming that investment level is a function of productivity and state variables: iit = f(xit,pit), where xit is capital and pit is productivity of firm i in time t, productivity can be expressed as pit = f-1(xit,iit) = h(xit,iit). The estimation of gross output takes the form of a Cobb-Douglas production function:6 yit = α + αwit + γxit +τcit + h(xit-1,iit-1)+ εit where yit is log of gross output, wit is a vector of free variables (male and female labor in our case), xit is a vector of state variables (physical capital), and cit is a vector of control variables (gender, age, etc). Function h(.) is a polynomial (cubic) function level of capital and investment in t-1. While wit-1 serves as an instrument for wit. 6 Estimated using the prodest command in STATA. For more information, please see Mollisi and Rovigatti (2017). 149 Enterprising Women | Appendix & References Model 3.6 Analysis of female entrepreneurs operating in male-dominated sectors Analysis on female entrepreneurs operating in male-dominated sectors is conducted using data on microbusinesses coming from Cambodia (CSES), Indonesia (IFLS), Lao PDR (LECS), and Vietnam (VARHS).7 Sectors are defined as male-dominated if at least 70 percent of businesses in that sector are run by men, which is similar though slightly lower than the 75 percent threshold used in de Mel et al. (2009). The threshold of 70 percent was chosen as it seemed to capture sectors that are truly male-dominated but still had a large enough sample of women in the sector to allow for the analysis of women operating in these sectors. Robustness checks were carried out using thresholds of 65 percent and 75 percent, and the main conclusions hold across specifications, although coefficient magnitude and significance change somewhat across specifications. To identify how the performance of female entrepreneurs operating in male-dominated sectors compare to those in female-concentrated sectors, we estimate: yit = ω0 + ω1F_MDSit + ω2M_FCSit + ω3M_MDSit + θt + εit Where: yit is the business performance of the firm run by individual i in year t, as defined in Model 1.1. F_MDSit is a dummy value taking the value of 1 if individual i is a woman operating in a male-  dominated sector in year t M_FCSit is a dummy value taking the value of 1 if individual i is a man operating in a female- concentrated sector in year t  _MDSit is a dummy value taking the value of 1 if individual i is a man operating in a male- M dominated sector in year t θt is a time fixed effect, included in regressions using panel data  The omitted category is women operating in female-concentrated sectors, so the coefficient denotes the difference in business performance between women in male-dominated sectors when compared with women in female-concentrated sectors. As in Model 1.1, the coefficient ω1 is transformed according to recommendations in Wooldridge (2012) by exponentiating the coefficient, subtracting one, then multiplying by 100 to calculate the percentage difference in performance. To understand whether men in male-dominated sectors outperform men in female-concentrated sectors, the coefficients ω2 and ω3 are compared and the difference between them is tested for statistical significance. 7 Timor-Leste is not included in this analysis because the sample size of women entrepreneurs in male-dominated sectors is too small to produce reliable estimates. 150 Enterprising Women | Appendix & References Methods and specifications used in Section 4 Model 4.1 OLS regressions with gender interaction terms The regressions run are the same as in Model 3.1 but run on the sample of SME. Model 4.2 Cobb-Douglas production function The Model is the same as Model 3.5 but run on the sample of SME. 151 Enterprising Women | Appendix & References Appendix C: Distribution of business performance variables for male and female microenterprises and SME Section 1 shows that on average female microentrepreneurs have lower profits than male microentrepreneurs. On average, there are not statistically significant gender gaps in the performance of SMEs, with the exception of Indonesia. Understanding the average performance differentials is a useful way to identify potential gender-related obstacles to business performance because it shows trends of large groups of businesses. However, average performance differences do not tell the whole story. As this report argues, female and male entrepreneurs are not homogenous groups. Not all female microentrepreneurs perform worse than male microentrepreneurs, and some female-owned SME may outperform or underperform some male- owned SME. As such, exploring the distribution of profits can be useful to gain a more nuanced understanding of gender differences in performance. Figure C.1 shows the distribution of business performance variables for microbusinesses and SME in different countries studied. Figure C1: Distribution microbusiness and SME performance, by gender Business performance of Microbusinesses in Cambodia Business performance of SME in Cambodia Distribution of profits by gender Distribution of revenues by gender .4 Female .4 Female Male Male .3 .3 Density Density .2 .2 .1 .1 0 0 10 12 14 16 18 20 16 18 20 22 24 26 Log of Profit Log of Revenue Business performance of Microbusinesses in Indonesia Business performance of SME in Indonesia Distribution of profits by gender Distribution of revenues by gender .4 Female Female Male .4 Male .3 .3 Density Density .2 .2 .1 .1 0 0 −5 0 5 10 15 20 25 30 35 Log of Profit Log of Revenue 152 Enterprising Women | Appendix & References Business performance of Microbusinesses in Lao PDR Business performance of SME in Lao PDR Distribution of revenues by gender Distribution of revenues by gender .3 .3 Female Female Male Male .2 .2 Density Density .1 .1 0 0 12 14 16 18 15 20 25 30 Log of Revenue Log of Revenue Business performance of Microbusinesses in Timor−Leste Business performance of SME in Timor−Leste Distribution of revenues by gender Distribution of revenues by gender .4 .4 Female Female Male Male .3 .3 Density Density .2 .2 .1 .1 0 0 0 2 4 6 8 8 10 12 14 16 Log of Profit Log of Revenue Business performance of Microbusinesses in Vietnam Business performance of SME in Vietnam Distribution of profits by gender Distribution of profits by gender .6 .4 Female Female Male Male .3 .4 Density Density .2 .2 .1 0 0 −10 −5 0 −6 −4 −2 0 2 Log of Profit Log of Profit 153 Enterprising Women | Appendix & References References Ács, Z. J., Szerb, L., Autio, E., & Lloyd, A. (2017). Alibhai, S., Buehren, N., Frese, M., Goldstein, M., Global Entrepreneurship Index 2017. Washington, Papineni, S., & Wolf, K. (2019). Full Esteem Ahead? D.C.: The Global Entrepreneurship and Development Mindset-Oriented Business Training in Ethiopia. Institute. Retrieved from http://thegedi.org/ World Bank Policy Research Working Paper, product/2017-global-entrepreneurship-index/ 8892. https://openknowledge.worldbank.org/ Agénor, P.-R., & Canuto, O. (2012). Access to handle/10986/31905 Infrastructure and Women’s Time Allocation: Alibhai, S., Buehren, N., & Papineni, S. (2018). Better Evidence and a Framework for Policy Analysis. Loans or Better Borrowers?: Impact of Meso- Retrieved from http://www.ferdi.fr/sites/www. Credit on Female-Owned Enterprises in Ethiopia. ferdi.fr/files/publication/fichiers/wp45_web_ World Bank Policy Research Working Paper, janvier2014_0.pdf 8511. https://openknowledge.worldbank.org/ Ajzen, I. (1985). From Intentions to Actions: A Theory handle/10986/29984 of Planned Behavior. In J. Kuhl & J. Beckmann (Eds.), Allen, E., Langowitz, N., & Minniti, M. (2007). The Action Control: From Cognition to Behavior (pp. 11– 2006 Global Entrepreneurship Monitor: Special Topic 39). Berlin, Heidelberg: Springer Berlin Heidelberg. Report: Women in Entrepreneurship. Babson Park, https://doi.org/10.1007/978-3-642-69746-3_2 MA: Center for Women Leadership, Babson College. Ajzen, I. (1991). The Theory of Planned Retrieved from https://www.gemconsortium.org/ Behavior. Organizational Behavior and Human report/47121 Decision Processes, 50, 179–191. https://doi. Álvarez, I., Marin, R., & Fonfría, A. (2009). The role org/10.1016/0749-5978(91)90020-T of networking in the competitiveness of firms. Alaref, J., Brodmann, S., & Premand, P. (2019). The Technological Forecasting and Social Change, 76 (3), Medium-Term Impact of Entrepreneurship Education 410-421. on Labor Market Outcomes: Experimental Evidence Anderson, S. J., Chandy, R., & Zia, B. (2018). from University Graduates in Tunisia. Washington, Pathways to Profits: The Impact of Marketing D.C.: The World Bank. https://doi.org/10.1596/1813- vs. Finance Skills on Business Performance. 9450-8701 Management Science, 64(12), 5559-5583. https://doi. Alatas, V., & Cameron, L. A. (2008). The Impact of org/10.1287/mnsc.2017.2920 Minimum Wages on Employment in a Low-Income Anderson, S. J., Chintagunta, P. K., Germann, F., Country: A Quasi-Natural Experiment in Indonesia. & Vilcassim, N. (2020). The Impact of Volunteer Industrial and Labor Relations Review, 61(2), 201– Marketers on Product Differentiation and Firm 223. https://doi.org/10.1177/001979390806100204 Growth: A Field Experiment with Ugandan Aldrich, H. E. (1999). Organizations Evolving. London: Entrepreneurs. Chicago Booth Research Paper, Sage Publications. 20(14). http://dx.doi.org/10.2139/ssrn.3613139 Aldrich, H. E., & Zimmer, C. (1986). Entrepreneurship Anderson, S. J., & McKenzie, D. (2020). Improving Through Social Networks. In D. L. Sexton & Business Practices and the Boundary of the R. W. Smilor (Eds.), The Art and Science of Entrepreneur: A Randomized Experiment Comparing Entrepreneurship (pp. 3–24). Cambridge, MA: Training, Consulting, Insourcing and Outsourcing. Ballinger Publishing Company. Policy Research Working Paper, 9502. Retrieved from https://openknowledge.worldbank.org/ handle/10986/34979 154 Enterprising Women | Appendix & References Ángeles, G., Gadsden, P., Galiani, S., Gertler, P., ASEAN. (2018). 2018 ASEAN Statistical Year Book. Herrera, A., Keriger, P. & Seira, E. (2011). Evaluación Retrieved from https://www.aseanstats.org/wp- de Impacto del Programa Estancias Infantiles para content/uploads/2019/01/asyb-2018.pdf Apoyar a Madres Trabajadoras. Report, Centro de ASEAN Secretariat. (2016). Projected Gender Impact Investigación en Evaluación y Encuestas. of the ASEAN Economic Community. Jakarta. Angelucci, M., Karlan, D., & Zinman, J. (2015). Retrieved from https://library.fes.de/pdf-files/ Microcredit Impacts: Evidence from a Randomized bueros/singapur/12676.pdf Microcredit Program Placement Experiment by Ashraf, N., Karlan, D., & Wesley, Y. (2006). Tying Compartamos Banco. American Economic Journal: Odysseus to the Mast: Evidence from a Commitment Applied Economics, 7(1), 151–182. https://doi. Savings Product in the Philippines. The Quarterly org/10.1257/app.20130537 Journal of Economics, 121(2), 635–672. Retrieved Anna, A., Chandler, G., Jansen, E., & Mero, N. (1999). from http://www.jstor.org/stable/25098802 Women business owners in traditional and non– Ashraf, N., Karlan, D., & Yin, W. (2010). Female traditional industries. Journal of Business Venturing, Empowerment: Impact of a Commitment Savings 15, 279–303. Product in the Philippines. World Development, Apedo-Amah, et al. (2020). Unmasking the 38(3), 333–344. https://doi.org/10.1016/j. Impact of COVID-19 on Businesses: Firm Level worlddev.2009.05.010 Evidence from Across the World. World Bank Asia Foundation. (2013a). Access to Trade and Policy Research Working Paper. 9434. Retrieved Growth of Women’s SMEs in APEC Developing from https://openknowledge.worldbank.org/ Economies: Evaluating Business Environments in handle/10986/34626 Malaysia - Philippines - Thailand. San Francisco. Armentos, M., Artellini N., Hoppe A., Urizar, M., & Retrieved from https://www.mimp.gob.pe/ Yaros, B. (2015): “Rethinking FOGAPE. Evaluating weapec2016/files/reportsaccess-to-trade-and- Chile’s Partial Credit Guarantee Program. Mimeo. growth-of-women-smes.pdf Retrieved from https://en.ufm.edu/wp-content/ Asia Foundation. (2013b). Access to Trade and uploads/2017/04/11.-Marco-Urizar.pdf Growth of Women’s SMEs in APEC Developing Arraíz, I. et al. (2019). Less is More: Experimental Economies: Evaluating the Business Environment Evidence on Heuristic-Based Business Training in Indonesia. San Francisco. Retrieved from https:// tin Ecuador. Inter-American Development Bank. asiafoundation.org/resources/pdfs/EvaluatingtheBu Retrieved from https://www.idbinvest.org/en/ sinessEnvironmentinIndonesia.pdf publications/report-less-more-experimental- Asia Foundation and Asian Development evidence-heuristics-based-business-training-ecuador Bank. (2018). Emerging Lessons on Women’s Arráiz, I., Meléndez, M., & Stucchi, R. (2014). Partial Entrepreneurship in Asia and the Pacific: Case Credit Guarantees and Firm Performance: Evidence Studies from the Asian Development Bank and from Colombia. Small Business Economics, 43(3), The Asia Foundation. http://dx.doi.org/10.22617/ 711–724. http://dx.doi.org/10.1007/s11187-014- TCS189585-2 9558-4 Asia Pacific Economic Cooperation. (2021). Small Arugay, A., Jarvis, F., Johnson, H., Liaqat, S., & and Medium Enterprises. Retrieved from https:// Valenciano, A.K. Forthcoming. Understanding gender www.apec.org/Groups/SOM-Steering-Committee- dynamics of decision making in agrarian households on-Economic-and-Technical-Cooperation/Working- of rural Philippines. Washington D.C.: World Bank Groups/Small-and-Medium-Enterprises.aspx ASEAN (2016). ASEAN Strategic Action Plan for SME Asian Development Bank. (2009). Powering the Development 2016-2025. Retrieved from http://www. Poor: Projects to increase access to clean energy smecorp.gov.my/images/pdf/SAP-SMED-Final.pdf for all. Retrieved from https://www.adb.org/sites/ default/files/publication/29138/powering-poor.pdf 155 Enterprising Women | Appendix & References Asian Development Bank. (2013). Gender-Inclusive Balachandra, L., A.R. Briggs, K.Eddleston, and Approaches in Urban Development. Retrieved from C. Brush. (2013). “Pitch Like A Man: Gender https://www.adb.org/sites/default/files/institutional- Stereotypes And Entrepreneur Pitch Success,” document/34136/files/tip-sheet-gender-inclusive- Frontiers of Entrepreneurship Research: Vol. 33: Iss. urban-development.pdf 8, Article 2. Available at: http://digitalknowledge. Asian Development Bank. (2015). Promoting babson.edu/fer/vol33/iss8/2 Women’s Economic Empowerment in Cambodia. Bandiera, O., Buehren, N., Burgess, R., Goldstein, Mandaluyong City, Philippines: ADB. Retrieved M., Gulesci, S., Rasul, I., & Sulaiman, M. (2018). from https://www.adb.org/sites/default/files/ Women’s Empowerment in Action: Evidence from publication/156499/promoting-womens-economic- a Randomized Control Trial in Africa. Washington, empowerment.pdf D.C.: The World Bank. https://doi.org/10.1596/28282 Asian Development Bank. (2017). Accelerating Bandura, A. (1991). Social Cognitive Theory of Financial Inclusion in South-East Asia with Digital Self-regulation. Organizational Behavior and Human Finance. ADB. https://doi.org/10.22617/RPT178622- Decision Processes, 50(2), 248–287. https://doi. 2 org/10.1016/0749-5978(91)90022-L Asian Development Bank. (2018). Emerging Lessons Banerjee, A., Breza, E., Duflo, E. & Kinnan, C. (2017). on Women’s Entrepreneurship in Asia and the Do Credit Constraints Limit Entrepreneurship? Pacific: Case Studies from the Asian Development Heterogeneity in the Returns to Microfinance. Global Bank and The Asia Foundation. Manila, Philippines: Poverty Research Lab Working Paper. 1(104). http:// Asian Development Bank. https://doi.org/10.22617/ dx.doi.org/10.2139/ssrn.3126359 TCS189585-2 Banerjee, A., & Duflo, E. (2014). Do Firms Want to Attanasio, O., & Vera-Hernández, M. (2004). Medium- Borrow More? Testing Credit Constraints Using a and Long Run Effects of Nutrition and Child Care: Directed Lending Program. Review of Economic Evaluation of a Community Nursery Programme Studies, 81, 572–607. doi: 10.1093/restud/rdt046 in Rural Colombia. Retrieved from The Institute Banerjee, A., Duflo, E., Glennerster, R., & Kinnan, for Fiscal Studies website: http://discovery.ucl. C. (2015). The Miracle of Microfinance? Evidence ac.uk/14749/1/14749.pdf from a Randomized Evaluation. American Economic Azmat, G., & Petrongolo, B. (2014). Gender and the Journal: Applied Economics 2015, 7(1), 22–53. http:// labor market: What have we learned from field and dx.doi.org/10.1257/app.20130533 lab experiments? Labour Economics, 30(C), 32-40. Banerjee, A., Karlan, D., & Zinman, J. (2015). Six Babbitt, L. G., Brown, D., & Mazaheri, N. (2015). Randomized Evaluations of Microcredit: Introduction Gender, Entrepreneurship, and the Formal– and Further Steps. American Economic Journal: Informal Dilemma: Evidence from Indonesia. Applied Economics, 7(1), 1–21. https://doi. World Development, 72, 163–174. https://doi. org/10.1257/app.20140287 org/10.1016/j.worlddev.2015.02.019 Bardasi, E., Sabarwal, S., & Terrell, K. (2011). How Bahramitash, R., & Kazemipour, S. (2011). Veiled Do Female Entrepreneurs Perform? Evidence Economy: Gender and the Informal Sector. In from Three Developing Regions. Small Business R. Bahramitash & H. S. Esfahani (Eds.), Veiled Economics, 37(4), 417–441. https://doi.org/10.1007/ Employment: Islamism and the Political Economy s11187-011-9374-z of Women’s Employment in Iran (pp. 226–255). Baron, R. A. (2004). The Cognitive Perspective: A Syracuse, New York: Syracuse University Press. Valuable Tool for Answering Entrepreneurship’s Basic “Why” Questions. Journal of Business Venturing, 19(2), 221–239. https://doi.org/10.1016/ S0883-9026(03)00008-9 156 Enterprising Women | Appendix & References Bastian, G., Bianchi, l., Goldstein, M. & Montalvao, Blackden, C. M., & Wodon, Q. (Eds.). (2006). J. (2018). Short-Term Impacts of Improved Access Gender, Time Use, and Poverty in Sub-Saharan to Mobile Savings, with and without Business Africa. Washington, D.C: World Bank. Retrieved Training: Experimental Evidence from Tanzania. from http://siteresources.worldbank.org/ Retrieved from https://pubdocs.worldbank.org/ INTAFRREGTOPGENDER/Resources/gender_time_ en/529051527991880514/pdf/122-BWC-20- use_pov.pdf Feb-2018.pdf Bloom, N., & Van Reenen, J. (2007). Measuring and Bayudan-Dacuycuy, C. (2019). Examining the Explaining Management Practices Across Firms Women’s Low Labor Market Participation Rate in and Countries. The Quarterly Journal of Economics, the Philippines: Is Housework the Missing Link? 122(4), 1351–1408. https://doi.org/10.1162/ Discussion Paper Series, No. 2019-05. Philippine qjec.2007.122.4.1351 Institute for Development Studies. Boden, R. J. (1996). Gender and Self-Employment Beaman, L., Magruder, J., & Robinson, J. (2014). Selection: An Empirical Assessment. The Journal Minding small change among small firms in Kenya. of Socio-Economics, 25(6), 671–682. https://doi. Journal of Development Economics, 108(C), 69-86. org/10.1016/S1053-5357(96)90046-3 doi: 10.1016/j.jdeveco.2013.12.010 Bollinger, K., & Merchant-Vega, N. (2013, February Beegle, K., E. Rubiano-Matulevich. 2020. Adapting 28). Networking Essential to Women Entrepreneurs Skills Training to Address Constraints to Women’s in Asia. The Asia Foundation. Retrieved from https:// Participation. Jobs Note: No 7. World Bank, asiafoundation.org/2013/02/27/networking- Washington, DC https://openknowledge.worldbank. essential-to-women-entrepreneurs-in-asia/ org/handle/10986/33694 License: CC BY 3.0 IGO. Bosma, N., & Kelley, D. (2018). Global Bellemare, M. & Wichman, C. (2019). Elasticities and Entrepreneurship Monitor2018/2019 Global Report. the Inverse Hyperbolic Sine Transformation. Oxford Retrieved from https://www.gemconsortium.org/ Bulletin of Economics and Statistics, 82(1), 50-61. report/gem-2018-2019-global-report https://doi.org/10.1111/obes.12325 Brody, C., De Hoop, T., Vojtkova, M., Warnock, Berlinski, S., & Galiani, S. (2007). The effect of a R., Dunbar, M., Murthy, P., & Dworkin, S. (2015). large expansion of pre-primary school facilities on Economic Self-help Group Programs for preschool attendance and maternal employment. Improving Women’s Empowerment. The Campbell Labour Economics, 14(3), 665-680. Collaboration. https://doi.org/10.4073/csr.2015.19 Berlinski, S., Galiani, S. & McEwan, P. J. (2011). Brooks, W., Donovan, K., & Johnson, T.R. (2018). Preschool and maternal labor market outcomes: Mentors or Teachers? Microenterprise Training Evidence from a regression discontinuity design. in Kenya. American Economic Journal: Applied Economic Development and Cultural Change, 59(2), Economics, 10(4), 196–221. https://doi.org/10.1257/ 313-344. app.20170042 Bernhardt, A., Field, E., Pande, R., & Rigol. N., (2019). Bruhn, M., Karlan, D., & Schoar, A. (2018). The Household Matters: Revisiting the Returns to Capital Impact of Consulting Services on Small and Medium among Female Microentrepreneurs. American Enterprises: Evidence from a Randomized Trial in Economic Review: Insights, 1(2), 141-60. doi: Mexico. Journal of Political Economy, 126(2). https:// 10.1257/aeri.20180444 doi.org/10.1086/696154 Betcherman, G., Meltem Daysal, N., & Pagés, C. Bruhn, M., & Zia, B. (2011). Stimulating Managerial (2010). Do employment subsidies work? Evidence Capital in Emerging Markets: The Impact of from regionally targeted subsidies in Turkey. Labor Business and Financial Literacy for Young Econ, 17, 710–722. Entrepreneurs. Washington, D.C.: The World Bank. https://doi.org/10.1596/1813-9450-5642 157 Enterprising Women | Appendix & References Brune, L., Giné, X., Goldberg, J., & Yang, D. (2011). Buvinić, M., Furst-Nichols, R., & Courey Pryor, Commitments to Save: A Field Experiment in Rural E. (2013). A Roadmap for Promoting Women’s Malawi. Washington, D.C.: The World Bank. https:// Economic Empowerment. Washington, D.C.: doi.org/10.1596/1813-9450-5748 The United Nations Foundation and ExxonMobil. Brune, L., Gin, X., Goldberg, J., & Yang, D. (2016) Retrieved from http://www.womeneconroadmap. Facilitating Savings for Agriculture: Field org/sites/default/files/WEE_Roadmap_Report_ Experimental Evidence from Malawi. Economic Final_1.pdf Development and Cultural Change, University of Buvinić, M., Johnson, H.C., Perova, E., & Witoelar, F. Chicago Press, 64(2), 187 – 220. (2020). Can Boosting Savings and Skills Support Brush, C. G. (1992). Research on Women Female Business Owners in Indonesia?: Evidence Business Owners: Past Trends, a New Perspective From a Randomized Controlled Trial. The World and Future Directions. Entrepreneurship Bank. Retrieved from http://documents.worldbank. Theory and Practice, 16(4), 5–30. https://doi. org/curated/en/837561585728764714/Can- org/10.1177/104225879201600401 Boosting-Savings-and-Skills-Support-Female- Business-Owners-in-Indonesia-Evidence-From-a- Brush, C. G., Carter, N., Gatewood, E. J., Greene, P., Randomized-Controlled-Trial & Hart, M. (Eds.). (2006). Growth-oriented Women Entrepreneurs and Their Businesses: A Global Buvinić, M., & O’Donnell, M. (2016). Revisiting What Research Perspective. Cheltenham, Northampton: Works: Women, Economic Empowerment and Edward Elgar Publishing. Smart Design. Washington, D.C.: Center for Global Development. Retrieved from https://www.cgdev. Brush, C. G., & Cooper, S. Y. (2012). Female org/sites/default/files/CGD-Roadmap-Update-2016. Entrepreneurship and Economic Development: pdf An International Perspective. Entrepreneurship & Regional Development, 24(1–2), 1–6. https://doi.org/ Buvinić, M., & O’Donnell, M. (2019). Gender 10.1080/08985626.2012.637340 Matters in Economic Empowerment Interventions: A Research Review. The World Bank Research Brush, C. G., De Bruin, A., & Welter, F. (2009). Observer, 34(2). A Gender-Aware Framework for Women’s Entrepreneurship. International Journal of Gender Buttner, E. H., & Moore, D. P. (1997). Women’s and Entrepreneurship, 1(1), 8–24. http://dx.doi. Organizational Exodus to Entrepreneurship: Self- org/10.1108/17566260910942318 Reported Motivations and Correlates with Success. Journal of Small Business Management, 34–46. Brush, C. G., & Hisrich, R. D. (1991). Antecedent Retrieved from https://www.researchgate.net/ Influences on Women-owned Businesses. Journal publication/279548467_Women’s_organizational_ of Managerial Psychology, 6(2), 9–16. https://doi. exodus_to_entrepreneurship_Self-reported_ org/10.1108/02683949110144846 motivations_and_correlates_with_success Bryan, G., Karlan, D. & Scott, N. (2009). Commitment Cai, J., & Szeidl, A. (2018). Interfirm Relationships Contracts. Working Papers, 73. and Business Performance. The Quarterly Journal Budig, M. J. (2006). Gender, Self-Employment, and of Economics, 133(3), 1229–1282. https://doi. Earnings: The Interlocking Structures of Family and org/10.1093/qje/qjx049 Professional Status. Gender & Society, 20(6), 725– Calderón, G. (2014). The Effects of Child Care 753. https://doi.org/10.1177/0891243206293232 Provision in Mexico. Banco de México Working Bulte, E., Lensink, R., & Vu, N. (2016). Gender Training Papers. 07. and Female Empowerment: Experimental Evidence Cambodian Center for Human Rights. (2016). from Vietnam. Economic Letters, 145, 117–119. Cambodia’s Women in Land Conflict. Phnom Penh. https://doi.org/10.1016/j.econlet.2016.06.003 Retrieved from https://cchrcambodia.org/admin/ media/report/report/english/2016_09_27_cchr_ report_Cam_Women_in_Land_Conflict_ENG.pdf 158 Enterprising Women | Appendix & References Cameron, L., Olivia, S., & Shah, M. (2019). Scaling Caruso, B. A., Sevilimedu, V., Fung, I. C.-H., Patkar, up Sanitation: Evidence from an RCT in Indonesia. A., & Bake, K. K. (2015). Gender Disparities in Journal of Development Economics, 138, 1–16. Water, Sanitation, and Global Health. The Lancet, https://doi.org/10.1016/j.jdeveco.2018.12.001 386(9994), 650–651. https://doi.org/10.1016/ Campaña, J.C., Giménez-Nadal, J.I., & Molina, J.A. S0140-6736(15)61497-0 (2020). Self-employed and Employed Mothers in Casella, A., Rauch, J. (2002). Anonymous market Latin American Families: Are There Differences in and group ties in international trade. Journal of Paid Work, Unpaid Work, and Child Care?. Journal of International Economics, 58(1), 19-47. Family and Economic Issues, Springer, 41(1), 52-69. Cassano, F., Jõeveer, K., & Svejnar, J. (2013). Cash Campos, F., Frese, M., Goldstein, M., Iacovone, L., flow vs. collateral‐based credit Performance of Johnson, H. C., McKenzie, D., & Mensmann, M. micro, small and medium‐sized firms in transition (2017). Teaching Personal Initiative Beats Traditional economies. Economics of Transition, 21(2), 269-300. Training in Boosting Small Business in West Africa. https://doi.org/10.1111/ecot.12016 Science, 357(6357), 1287-1290. doi: 10.1126/ Casson, M. (1982). The Entrepreneur: An Economic science.aan5329 Theory. Rowman & Littlefield. Campos, F., Frese, M., Goldstein, M., Iacovone, L., Cetindamar, D., Gupta, V. K., Karadeniz, E. E., & Johnson, H. C., McKenzie, D., & Mensmann, M. Egrican, N. (2012). What the Numbers Tell: The (2018). Is Personal Initiative Training a Substitute Impact of Human, Family and Financial Capital on or Complement to the Existing Human Capital of Women and Men’s Entry into Entrepreneurship in Women? Results from a Randomized Trial in Togo. Turkey. Entrepreneurship & Regional Development, AEA Papers and Proceedings, 108, 256–261. https:// 24(1–2), 29–51. https://doi.org/10.1080/08985626. doi.org/10.1257/pandp.20181026 2012.637348 Campos, F., Goldstein, M., & McKenzie, D. (2015). Cavalcanti, T. and J. Tavares. 2015. “The Output Short-Term Impacts of Formalization Assistance Cost of Gender Discrimination: A Model-Based and a Bank Information Session on Business Macroeconomics Estimate.” The Economic Journal Registration and Access to Finance in Malawi. World 126(February): 109-134 Doi: 10.1111/ecoj.12303 Bank Policy Research Working Paper. 7183. Chhair, S., and L. Ung. 2016. “Economic History of Campos, F., Goldstein, M., McGorman, L., Munoz Industrialization in Cambodia.” Learning to Compete Boudet, A. M., & Pimhidzai, O. (2015). Breaking the Working Paper No. 7 Metal Ceiling : Female Entrepreneurs Who Succeed in Male-Dominated Sectors. World Bank Policy Chatterji, A., Delecourt, S., Hasan, S., & Koning, Research Working Paper, 7503. R. (2019). When does advice impact startup performance?. Strategic Management Journal, 40(3), Card, D. & Hyslop, D. R. (2005). Estimating the 331-356. Effects of a Time‐Limited Earnings Subsidy for Welfare‐Leavers. Econometrica Journal of the Chen, M. A. (2001). Women and Informality: A Global econometric society, 73(6), 1723-1770. https://doi. Picture, the Global Movement. SAIS Review, 21(1), 71–82. https://doi.org/10.1353/sais.2001.0007 org/10.1111/j.1468-0262.2005.00637.x Chen, C. C., Greene, P. G., & Crick, A. (1998). Carr, D. (1996). Two Paths to Self-Employment?: Does Entrepreneurial Self-efficacy Distinguish Women’s and Men’s Self-Employment in the United Entrepreneurs from Managers? Journal of Business States, 1980. Work and Occupations, 23(1), 26–53. https://doi.org/10.1177/0730888496023001003 Venturing, 13(4), 295–316. https://doi.org/10.1016/ S0883-9026(97)00029-3 Carter, N. M., Williams, M., & Reynolds, P. D. (1997). Discontinuance among New Firms in Retail: The Chhay, D. (2011). Women’s Economic Empowerment Influence of Initial Resources, Strategy, and Gender. through Microfinance in Cambodia. Development in Practice, 21(8), 1122–1137. https://doi.org/10.1080/ Journal of Business Venturing, 12(2), 125–145. 09614524.2011.606891 159 Enterprising Women | Appendix & References Chliova, M., Brinckmann, J., & Rosenbusch, N. Cole, S., Sampson, T., & Zia, B. (2011). Prices or (2015). Is Microcredit a Blessing for the Poor? A Knowledge? What Drives Demand for Financial Meta-analysis Examining Development Outcomes Services in Emerging Markets? The Journal of and Contextual Considerations. Journal of Business Finance, 66(6), 1933–1967. https://doi.org/10.1111/ Venturing, 30(3), 467–487. https://doi.org/10.1016/j. j.1540-6261.2011.01696.x jbusvent.2014.10.003 Coleman, B. E. (1999). The Impact of Group Lending Cho, Y., & Honorati, M. (2014). Entrepreneurship in Northeast Thailand. Journal of Development Programs in Developing Countries: A Meta Economics, 60(1), 105–141. https://doi.org/10.1016/ Regression Analysis. Labour Economics, 28, 110– S0304-3878(99)00038-3 130. https://doi.org/10.1016/j.labeco.2014.03.011 Coleman, B. E. (2006). Microfinance in Northeast Chowdhury, I., Johnson, H. C., Mannava, A., & Thailand: Who Benefits and How Much? World Perova, E. (2019). Gender Gap in Earnings in Development, 34(9), 1612–1638. https://doi. Vietnam: Why Do Vietnamese Women Work in Lower org/10.1016/j.worlddev.2006.01.006 Paid Occupations?. Journal of Southeast Asian Conley, T., & Udry, C. (2001). Social Learning through Economies, 36(3), 400-23. doi: 10.1355/ae36-3g Networks: The Adoption of New Agricultural Christiaensen, L., & Heltberg, R. (2012). Greening Technologies in Ghana. American Journal of China’s Rural Energy: New Insights on the Potential Agricultural Economics, 83(3), 668–673. of Smallholder Biogas. World Bank. Conley, T., & Udry, C. (2010). Learning About a New Cirera, X., & Qasim, Q. (2014). Supporting Growth- Technology: Pineapple in Ghana. The American Oriented Women Entrepreneurs: A Review of the Economic Review, 35–69. Evidence and Key Challenges. Washington, D.C.: The Coville, A., Di Maro, V., Dunsch, F. & Zottel, S. (2019). World Bank. The Nollywood Nudge: An Entertaining Approach to Cirera, X., Mason, A.D., de Nicola, F., Kuriakose, Saving. World Bank Policy Research Working Paper. S., Mare, D.S., Tran, T.T. (2021). The Innovation 8920. Imperative for Developing East Asia. Washington Cramer, J. S., Hartog, J., Jonker, N., & Praag, C. M. V. D.C.: The World Bank (2002). Low Risk Aversion Encourages the Choice Cislaghi, B., and L. Heise. 2020. “Gender Norms for Entrepreneurship: An Empirical Test of a Truism. and Social Norms: Differences, Similarities, and Journal of Economic Behavior & Organization, 48, Why They Matter in Prevention Science.” Sociology 29–36. of Health and Illness: 42(2) p. 407-422 https://doi. Crépon, B., Devoto, F., Duflo, E., & Parienté, W. (2015). org/10.1111/1467-9566.13008 Estimating the Impact of Microcredit on Those Who Clark, S., Kabiru, C.W., Laszlo, S. et al. (2019). Take It Up: Evidence from a Randomized Experiment The Impact of Childcare on Poor Urban Women’s in Morocco. American Economic Journal: Applied Economic Empowerment in Africa. Demography, 56, Economics, 7(1), 123-50. 1247–1272. https://doi.org/10.1007/s13524-019- Cuberes, D. and M. Teignier. 2016. “Aggregate 00793-3 Effects of Gender Gaps in the Labor Market: A Clark, S., Laszlo, S., Kabiru, C., & Muthu, S. (2017). Quantitative Estimate.” Journal of Human Capital Can Subsidized Early Child Care Promote Women’s 10(1) Employment?: Evidence from a Slum Settlement Cucagna, E., Iacovone, L., & Rubiano-Matulevich, E. in Africa. Institute for the Study of International (2020). Women Entrepreneurs in Mexico: Breaking Development. Sectoral Segmentation and Increasing Profits. World Cliff, J. (1998). Does one size fit all? Exploring the Bank. relationship between attitudes towards growth, gender, and business size. Journal of Business Venturing, 13, 523–542. 160 Enterprising Women | Appendix & References Dahal, M., Fiala, N. (2018). What do we know about De Mel, S., McKenzie, D., & Woodruff, C. (2019). the impact of microfinance? The problems of power Labor Drops: Experimental Evidence on the Return and precision. Ruhr Economic Papers, 756. http:// to Additional Labor in Microenterprises. American dx.doi.org/10.4419/86788880 Economic Journal: Applied Economics, 11(1), 202-35. Dalton, P., Rüschenpöhler, J., Uras, B. & Zia, B. (2019). De Soto, H. (2002). The Other Path: The Economic Learning to Grow from Peers: Experimental Evidence Answer to Terrorism. New York: Basic Books. from Small Retailers in Indonesia. Policy Research De V. Cavalcanti, T., & Tavares, J. (2008). Assessing Working Paper. 8933. the “Engines of Liberation”: Home Appliances and Dalton, P., Rüschenpöhler, J., Uras, B. & Zia, B. (2021). Female Labor Force Participation. The Review of Curating Local Knowledge: Experimental Evidence Economics and Statistics, 90(1), 81-88. from Small Retailers in Indonesia. Journal of the Debroux, P. (2010). Female Entrepreneurship in East European Economic Association. doi: 10.1093/jeea/ and South-East Asia: Opportunities and Challenges. jvab007 Oxford: Chandos Publishing. Dang, H., Hiraga, M. & Nguyen, C. V. (2019). Childcare Delecourt, S., & Fitzpatrick, A. (2020). Childcare and Maternal Employment: Evidence from Vietnam. Matters: Female Business Owners and the Baby- Policy Research Working Paper Series, 8856. Profit Gap. Retrieved from https://www.aeaweb.org/ Dasso, R., & Fernandez, F. (2015). The Effects of conference/2020/preliminary/paper/ZiE9syz8 Electrification on Employment in Rural Peru. IZA DeMartino, R., & Barbato, R. (2003). Differences Journal of Labor & Development, 4(1). https://doi. between Women and Men MBA Entrepreneurs: org/10.1186/s40175-015-0028-4 Exploring Family Flexibility and Wealth Creation as Daymont, T. N., & Andrisani, P. J. (1984). Job Career Motivators. Journal of Business Venturing, Preferences, College Major, and the Gender Gap in 18(6), 815–832. https://doi.org/10.1016/S0883- Earnings. Journal of Human Resources, 19(3), 408- 9026(03)00003-X 428. Demirgüç-Kunt, A., Beck, T., & Honohan, P. (2008). De Blasio, G., De Mitri, S., D’Ignazio, A., Finaldi Russo, Finance for All? Policies and Pitfalls in Expanding P., & Stoppani, L. (2018). Public Guarantees to SME Access. Washington, D.C.: The World Bank. Retrieved Borrowing: A RDD Evaluation. Journal of Banking from http://documents.worldbank.org/curated/ & Finance, 96, 73-86. https://doi.org/10.1016/j. en/932501468136179348/pdf/417920PAPER0Fi180 jbankfin.2018.08.003. 82137291301PUBLIC1.pdf De Mel, S., McKenzie, D., & Woodruff, C. (2007). Demirgüç-Kunt, A., Klapper, L. F., Singer, D., Ansar, Returns to Capital in Microenterprises: Evidence S., & Hess, J. (2018). The Global Findex Database from a Field Experiment. Washington, D.C.: World 2017: Measuring Financial Inclusion and the Fintech Bank. Retrieved from https://openknowledge. Revolution. Washington, D.C.: The World Bank. worldbank.org/handle/10986/7124 Retrieved from https://globalfindex.worldbank. De Mel, S., McKenzie, D., & Woodruff, C. (2012). org/sites/globalfindex/files/2018-04/2017 One-Time Transfers of Cash or Capital Have Long- percent20Findex percent20full percent20report_0. Lasting Effects on Microenterprises in Sri Lanka. pdf Science, 335, 962–966. https://doi.org/10.1126/ Devine, T. J. (1994). Changes in Wage-and-Salary science.1212973 Returns to Skill and the Recent Rise in Female Self- De Mel, S., McKenzie, D., & Woodruff, C. (2014). What Employment. American Economic Review, 84(2), Generates Growth in Microenterprises? Experimental 108–113. Retrieved from https://www.jstor.org/ Evidence on Capital, Labor and Training. CAGE stable/2117812 Online Working Paper Series 212, Competitive Advantage in the Global Economy (CAGE). Retrieved from http://economics.mit.edu/files/8666 161 Enterprising Women | Appendix & References Devoto, F., Duflo, E., Dupas, P., Parienté, W., & Eagly, A. H., Wood, W., & Diekman, A. B. (2000). Pons, V. (2012). Happiness on Tap: Piped Water Social Role Theory of Sex Differences and Adoption in Urban Morocco. American Economic Similarities: A Current Appraisal. In T. Eckes & Journal: Economic Policy, 4(4), 68–99. https://doi. H. M. Trautner (Eds.), The Developmental Social org/10.1257/pol.4.4.68 Psychology of Gender (pp. 123–174). Mahwah, New Dinkelman, T. (2011). The Effects of Rural Jersey: Erlbaum. Electrification on Employment: New Evidence from Eckardt, S., G. Demombynes, D. Chandrasekharan South Africa. American Economic Review, 101(7), Behr. (2016). Vietnam – Systematic Country 3078–3108. https://doi.org/10.1257/aer.101.7.3078 Diagnostic. Washington, DC: World Bank Group Djankov, S., Porta, R. L., Lopez-de-Silanes, F., & http://documents.worldbank.org/curated/ Shleifer, A. (2002). The Regulation of Entry. The en/334491474293198764/Vietnam-Systematic- Quarterly Journal of Economics, 117(1), 1–37. Country-Diagnostic https://doi.org/10.1162/003355302753399436 Eddleston, K. A., & Powell, G. N. (2012). Nurturing Djédjé, M., Ingwe, A., Wanyohi, P., Brinkmann, V., & Entrepreneurs’ Work-Family Balance: A Gendered Kithinji, J. (2009). Survey on Impacts of the Stove Perspective. Entrepreneurship Theory and Practice, Project in Transmara, Western and Central Cluster of 36(3), 513–541. https://doi.org/10.1111/j.1540- Kenya: Final Report. 6520.2012.00506.x Dong, S. X. (2018). Does economic crisis have Elam, A., & Terjesen, S. A. (2007). Institutional different impact on husbands and wives? Evidence Logics: Gender and Business Creation Across Gem from the Asian Financial Crisis in Indonesia. Rev Countries (SSRN Scholarly Paper No. ID 1060201). Dev Econ, 22, 1489–1512. https://doi.org/10.1111/ Rochester, NY: Social Science Research Network. rode.12521 Ellis, A., Kirkwood, D., & Malhotra, D. (2010). Doyle, K., Levtov, R.G., Barker, G., Bastian, G.G., Economic Opportunities for Women in the East Asia Bingenheimer, J.B., Kazimbaya, S., et al. (2018). and Pacific Region. Washington, D.C.: The World Gender-transformative Bandebereho couples’ Bank. Retrieved from http://elibrary.worldbank.org/ intervention to promote male engagement in doi/book/10.1596/978-0-8213-8300-1 reproductive and maternal health and violence Esteve-Volart, B. 2009. “Gender Discrimination prevention in Rwanda: Findings from a randomized and Growth: Theory and Evidence from India.” controlled trial. PLoS ONE, 13(4). https://doi. Manuscript. org/10.1371/journal.pone.0192756 Eswaran, M., & Kotwal, A. (1986). Access to Drexler, A., Fischer, G., & Schoar, A. (2014). Keeping Capital and Agrarian Production Organisation. It Simple: Financial Literacy and Rules of Thumb. The Economic Journal, 96(382), 482. https://doi. American Economic Journal: Applied Economics, org/10.2307/2233128 6(2), 1–31. http://dx.doi.org/10.1257/app.6.2.1 Fafchamps, M., McKenzie, D., Quinn, S., & Dupas, P., & Robinson, J. (2013). Savings Constraints Woodruff, C. (2014). Microenterprise Growth and and Microenterprise Development: Evidence from the Flypaper Effect: Evidence from a Randomized a Field Experiment in Kenya. American Economic Experiment in Ghana. Journal of Development Journal: Applied Economics, 5(1), 163–192. Economics, 106, 211–226. https://doi.org/10.1016/j. Eagly, A. H. (1987). Sex Differences in Social jdeveco.2013.09.010 Behavior: A Social-Role Interpretation (1 edition). Fafchamps, M., & Quinn, S. (2017). Aspire. The Hillsdale, N.J: Lawrence Erlbaum. Journal of Development Studies, 53(10), 1615-1633. Eagly, A. H., & Wood, W. (2016). Social Role doi: 10.1080/00220388.2016.1251584 Theory of Sex Differences. In The Wiley Blackwell Encyclopedia of Gender and Sexuality Studies (pp. 1–3). American Cancer Society. https://doi. org/10.1002/9781118663219.wbegss183 162 Enterprising Women | Appendix & References Fafchamps, M., & Quinn, S. (2018). Networks and Fossen, F. M. (2012). Gender Differences in Manufacturing Firms in Africa: Results from a Entrepreneurial Choice and Risk Aversion – A Randomized Field Experiment. The World Bank Decomposition Based on a Microeconometric Economic Review, 32(3), 656–675. doi: 10.1093/ Model. Applied Economics, 44(14), 1795–1812. wber/lhw057 https://doi.org/10.1080/00036846.2011.554377 Fairlie, R. W., & Robb, A. M. (2009). Gender Freeman, N. (2015). Financing Small and Medium Differences in Business Performance: Evidence Sized Enterprises for Sustainable Development: A from the Characteristics of Business Owners Survey. View from the Asia-Pacific Region. Jakarta: United Small Business Economics, 33, 375–395. https://doi. Nations Economic and Social Commission for org/10.1007/s11187-009-9207-5 Asia and the Pacific. Retrieved from https://www. Fay, D., & Frese, M. (2001). The concept of unescap.org/sites/default/files/2-ESCAP_FfD_SME personal initiative: An overview of validity studies. percent20finance_24April2015.pdf Human Performance, 14(1), 97–124. https://doi. Gayle, A., Candace, A. M., & Christopher, W. org/10.1207/S15327043HUP1401_06 (2012). Political instability, pro-business market Fayolle, A., & Kyrö, P. (2008). The Dynamics between reforms and their impacts on national systems of Entrepreneurship, Environment and Education. innovation. Research Policy, 41, 638–651. Cheltenham, UK: Edward Elgar Publishing. Gielnik, M. M., Frese, M., Kahara-Kawuki, A., Wasswa Fernandes, D., Lynch, J. G., & Netemeyer, R. G. Katono, I., Kyejjusa, S., Ngoma, M., Munene, J., (2014). Financial Literacy, Financial Education and Namatovu-Dawa, R., Nansubuga, F., Orobia, L., Oyugi, Downstream Financial Behaviors (SSRN Scholarly J., Sejjaaka, S., Sserwanga, A., Walter, T., Bischoff, Paper No. ID 2333898). Rochester, NY: Social K. M., & Dlugosch, T. J. (2015). Action and Action- Science Research Network. Regulation in Entrepreneurship: Evaluating a Student Training for Promoting Entrepreneurship. Academy Field, E., Jayachandran, S., Pande, R., & Rigol, of Management Learning & Education, 14(1), 69–94. N. (2016). Friendship at Work: Can Peer Effects https://doi.org/10.5465/amle.2012.0107 Catalyze Female Entrepreneurship?. American Economic Journal: Economic Policy 2016, 8(2), Giménez-Nadal, J.I., Molina, J.A., & Ortega, R. (2012). 125–153, http://dx.doi.org/10.1257/pol.20140215 Self-employed mothers and the work-family conflict, Applied Economics, Taylor & Francis Journals, Field, F., Pande, R., Papp, F., & Rigol, N. (2013). 44(17), 2133-2147. Does the Classic Microfinance Model Discourage Entrepreneurship among the Poor? Experimental Giné, X., & Karlan, D. S. (2014). Group versus Evidence from India. American Economic Review, Individual Liability: Short- and Long-Term Evidence 103(6), 2196-2226. from Philippine Microcredit Lending Groups. Journal of Development Economics, 107, 65–83. https://doi. Firpo, S., Fortin, N. M. & Lemieux, T. (2011). org/10.1016/j.jdeveco.2013.11.003 Decomposition Methods in Economics. Handbook of Labor Economics, 1-102. Giné, X., & Mansuri, G. (2020). Money or Management? A Field Experiment on Constraints Fisher, J. (2006). For her it’s the Big Issue: Putting to Entrepreneurship in Rural Pakistan. Economic Women at the Centre of Water Supply, Sanitation and Development and Cultural Change. forthcoming. Hygiene. Water, Sanitation and Hygiene: Evidence Report. Water Supply and Sanitation Collaborative Glaub, M., & Frese, M. (2011). A Critical Review of the Council. Retrieved from https://dspace.lboro.ac.uk/ Effects of Entrepreneurship Training in Developing dspace-jspui/bitstream/2134/9970/20/wsscc_for_ Countries. Enterprise Development and Microfinance, her_its_the_big_issue_evidence_report_2006_en.pdf 22(4), 335–353. https://doi.org/DOI: 10.3362/1755- 1986.2011.035 163 Enterprising Women | Appendix & References Glaub, M. E., Frese, M., Fischer, S., & Hoppe, Goyal, P., & Yadav, V. (2014). To Be or Not to Be M. (2014). Increasing Personal Initiative in a Woman Entrepreneur in a Developing Country? Small Business Managers or Owners Leads to Psychosociological Issues in Human Resource Entrepreneurial Success: A Theory-Based Controlled Management, 2(2), 68–78. Retrieved from https:// Randomized Field Intervention for Evidence-Based www.researchgate.net/publication/271520855_TO_ Management. Academy of Management Learning BE_OR_NOT_TO_BE_A_WOMAN_ENTREPRENEUR_ & Education, 13(3). https://doi.org/10.5465/ IN_A_DEVELOPING_COUNTRY amle.2013.0234 Grady, C. A., Van Nguyen, K., Van Nguyen, T., & Global Entrepreneurship Research Association. Blatchley, E. R. (2018). Who is Being Left Behind? (2017). The 2016/2017 Global Entrepreneurship An Analysis of Improved Drinking Water and Basic Monitor. Retrieved from http://www.gemconsortium. Sanitation Access in the Vietnamese Mekong org/report Delta. Journal of Water Sanitation and Hygiene Global Entrepreneurship Research Association. for Development, washdev2018005. https://doi. (2020). The 2019/2020 Global Entrepreneurship org/10.2166/washdev.2018.005 Monitor. Retrieved from https://www. Green, E., & Cohen, L. (1995). Women’s business: gemconsortium.org/report/gem-2019-2020-global- Are women entrepreneurs breaking new ground or report simply balancing the demands of women’s work in a Gneezy, U., Leonard, K. L., & List, J. A. (2009). new way?. Journal of Gender Studies, 4, 297–314. Gender Differences in Competition: Evidence from a Greene, P., Hart, M., Gatewood, E. J., Brush, C., matrilineal and a patriarchal society. Econometrica, & Carter, N. (2003). Moving Front and Center: 77(5), 1637–1664. An Overview of Research and Theory. Retrieved Goheer, N. (2003). Women Entrepreneurs in from www.usasbc.org/knowledge/whitepapers/ Pakistan: How to Improve their Bargaining Power. greene.2003.pd Geneva: ILO. Retrieved from http://ilo.org/wcmsp5/ Greif, A. (1993). Contract Enforceability and groups/public/@ed_emp/@emp_ent/documents/ Economic Institutions in Early Trade: The Maghribi publication/wcms_094011.pdf Traders’ Coalition. The American Economic Review, Goldin, C., & Rouse, C. (2000). Orchestrating 83(3), 525-548. Impartiality: The Impact of “Blind” Auditions on Grimm, M., & Paffhausen, A. L. (2015). Do Female Musicians. American Economic Review, Interventions Targeted at Micro-entrepreneurs 90(4), 715-741. and Small and Medium-sized Firms Create Goldman Sachs Foundation. (2019). Goldman Sachs Jobs? A Systematic Review of the Evidence 10,000 Women Initiative – About the Program. for Low- and Middle-Income Countries. Labour Retrieved from http://www.goldmansachs.com/ Economics, 32, 67–85. https://doi.org/10.1016/j. citizenship/10000women/about-the-program/about- labeco.2015.01.003 the-program-main-page.html Grogan, L., & Sadanand, A. (2013). Rural Goldstein, M., Martinez, P. G., & Papineni, S. Electrification and Employment in Poor Countries: (2019). Tackling the Global Profitarchy: Gender Evidence from Nicaragua. World Development, and the Choice of Business Sector. World Bank 43, 252–265. https://doi.org/10.1016/j. Policy Research Working Paper. https://doi. worlddev.2012.09.002 org/10.1596/1813-9450-8865. Groh, M., Krishnan, N., McKenzie, D. et al. (2016). The impact of soft skills training on female youth employment: evidence from a randomized experiment in Jordan. IZA J Labor Develop, 5(9). https://doi.org/10.1186/s40175-016-0055-9 164 Enterprising Women | Appendix & References Gupta, D. N., Poulsen, A., & Villeval, M. C. (2005). Hani, U., Rachmania, I. N., Setyaningsih, S., & Male and Female Competitive Behavior – Putri, R. C. (2012). Patterns of Indonesian Women Experimental Evidence (SSRN Scholarly Paper No. Entrepreneurship. Procedia Economics and Finance, ID 906766). Rochester, NY: Social Science Research 4, 274–285. https://doi.org/10.1016/S2212- Network. 5671(12)00342-5 Ha, Ngo Ngan. 2020. Household Budget Hanna, R., Mullainathan, S., Schwartzstein, J. (2014). Management and Decision-Making in the Family in Learning through Noticing: Theory and Evidence the Red River Delta, Vietnam. Working Paper. Vietnam from a Field Experiment. Quarterly Journal of Academy of Social Sciences. http://en.vass.gov.vn/ Economics, 129(3), 1311–53. noidung/tintuc/Lists/TinTucHoiNghiHoiThao/View_ Hardy, M., & Kagy, G. (2020). It’s Getting Crowded in Detail.aspx?ItemID=860 Here: Experimental Evidence of Demand Constraints Habermehl, H. (2007). Economic Evaluation of the in the Gender Profit Gap. The Economic Journal, Improved Household Cooking Stove Dissemination 130(631), 2272–2290. https://doi.org/10.1093/ej/ Programme in Uganda. Retrieved from https://www. ueaa040 un.org/esa/sustdev/csd/csd15/lc/GTZ_Uganda.pdf Heilman, M. E., & Chen, J. J. (2003). Habermehl, H. (2008). Costs and Benefits of Entrepreneurship as a Solution: The Allure of Self- Efficient Institutional Cookstoves in Malawi. employment for Women and Minorities. Human Retrieved from http://digicollection.org/eebea/ Resource Management Review, 13(2), 347–364. documents/s20440en/s20440en.pdf https://doi.org/10.1016/S1053-4822(03)00021-4 Hakim, C. (1989). Identifying Fast Growth Small Hossain, M., Zimmerman, C., Kiss, L., et al. (2014). Firms. Employment Gazette, 97, 29–41. Retrieved Working with men to prevent intimate partner from http://www.statistics.gov.uk/statbase/Product. violence in a conflict-affected setting: a pilot asp?vlnk=550&More=N cluster randomized controlled trial in rural Côte Halim, D., Johnson, H. C. & Perova, E. d’Ivoire. BMC Public Health, 14, 339. https://doi. (2021). Preschool Availability and Women’s org/10.1186/1471-2458-14-339 Employment: Evidence from Indonesia. Economic Hsieh, C., E. Hurst, C.I. Jones, P.J. Klenow. 2019. Development and Cultural Change. Retrieved “The Allocation of Talent and U.S. Economic from https://openknowledge.worldbank.org/ Growth.” Econometrica 87(5): 1439-74 handle/10986/35394 Huerta, M. C., Adema, W., Baxter, J., Han, W. Halim, M. L. D., Ruble, D. N., Tamis-LeMonda, J., Lausten, M., Lee, R., & Waldfogel, J. (2014). C. S., Shrout, P. E., & Amodio, D. M. (2017). Fathers’ Leave and Fathers’ Involvement: Evidence Gender Attitudes in Early Childhood: Behavioral from Four OECD Countries. European journal Consequences and Cognitive Antecedents. of social security, 16(4), 308–346. https://doi. Child Development, 88(3), 882–899. https://doi. org/10.1177/138826271401600403 org/10.1111/cdev.12642 Hughes, K. D. (2003). Pushed or pulled? Women’s Hallward-Driemeier, M. (2013). Enterprising Women: entry into self-employment and small business Expanding Economic Opportunities in Africa. ownership. Gender, Work and Organization, 10, Washington, D.C.: The World Bank. Retrieved 433–454. from http://documents.worldbank.org/curated/ Hundley, G. (2000). Male/Female Earnings en/509981468193131963/pdf/782990PUB0EPI00C Differences in Self-Employment: The Effects of 00PUBDATE0604020130.pdf Marriage, Children, and the Household Division of Handa, S., Seidenfeld, D., Davis, B., & Tembo, G. Labor. Industrial and Labor Relations Review, 54(1), (2016). The Social and Productive Impacts of 95–115. Retrieved from http://citeseerx.ist.psu.edu/ Zambia’s Child Grant. Journal of Policy Analysis viewdoc/download?doi=10.1.1.897.4724&rep=rep1 and Management, 35(2), 357–387. https://doi. &type=pdf org/10.1002/pam.21892 165 Enterprising Women | Appendix & References Hundley, G. (2001). Why women earn less than men International Finance Corporation. (2017). Women- in self-employment. Journal of Labor Research, owned Enterprises in Vietnam: Perceptions and 22(4), 817–29. https://doi.org/10.1007/s12122-001- Potential. Washington, D.C.: IFC. Retrieved from 1054-3 https://www.ifc.org/wps/wcm/connect/45c8b501- IBRD. (2011). Vietnam State and People, Central and 6239-45b2-bf4c-dae2b6d7647b/Market-study- Local, Working Together: The Rural Electrification on-Women-owned-enterprises-in-Vietnam_Eng_ Experience. Washington, D.C.: IBRD. Retrieved v1.pdf?MOD=AJPERES from http://www.sddirect.org.uk/media/1353/ International Finance Corporation. (2018). MSME infrastructure-womens-economic-empowerment- Finance Gap database. Retrieved from https://www. scoping-report.pdf smefinanceforum.org/data-sites/msme-finance-gap IFAD, FAO, & ILO. (2009). Gaps, Trends and Current International Finance Corporation. 2019. Research in Gender Dimensions of Agricultural Exploring the Opportunities for Women-owned and Rural Employment: Differentiated Pathways SMEs in Cambodia. Retrieved from https:// out of Poverty. Retrieved from http://www.ipcinfo. www.ifc.org/wps/wcm/connect/9e469291- org/fileadmin/user_upload/fao_ilo/pdf/Reports/ d3f5-43a5-bea2-2558313995ab/ Workshop_report/Workshop_report_Final_ Market+Research+Report+on+Women_ version_30_Sept_2009_to_be_on_line.pdf owned+SMEs+in+Cambodia. Ilahi, N., & Grimard, F. (2000). Public Infrastructure pdf?MOD=AJPERES&CVID=mOU6fpx and Private Costs: Water Supply and Time Allocation International Finance Corporation and USAID. of Women in Rural Pakistan. Economic Development (2016). Women-owned SMEs in Indonesia: A and Cultural Change, 49(1), 45–76. https://doi. Golden Opportunity for Local Financial Institutions. org/10.1086/452490 Washington, D.C. https://doi.org/10.1596/25403 Immyxai, S., & Takahashi, Y. (2009). Further Evidence International Labour Organisation. (2009). Guide of the Performance Contrast between Male and to the New Millennium Development Goals: Female Firms in MSMEs in Lao PDR. Journal of Asia Employment Indicators. Geneva: ILO. Retrieved from Entrepreneurship, V(3), 64–85. https://www.ilo.org/public/english/employment/ Inmyxai, S., & Takahashi, Y. (2010). Performance download/mdg_en.pdf Contrast and its Determinants between Male and International Labour Organisation. (2014). Women’s Female Headed Firms in Lao MSMEs. International Entrepreneurship Development. Geneva: ILO. Journal of Business and Management, 5(4). https:// Retrieved from https://www.ilo.org/wcmsp5/groups/ doi.org/10.5539/ijbm.v5n4p37 public/---ed_emp/---emp_ent/---ifp_seed/documents/ International Finance Corporation. (2008). Voices of publication/wcms_175471.pdf Cambodian Women Entrepreneurs. Washington, D.C. International Labour Organisation. (2018a). Care Retrieved from http://documents.worldbank.org/ Work and Care Jobs for the Future of Decent Work. curated/en/342171468016854477/pdf/465600WP0 Geneva: ILO. Retrieved from https://www.ilo.org/ Box331n1Women1Entreprenure.pdf wcmsp5/groups/public/---dgreports/---dcomm/--- International Finance Corporation. (2010). publ/documents/publication/wcms_633135.pdf Timor-Leste: Gender and Investment Climate International Labour Organization. 2018b. Global Reform Assessment. Washington, D.C. Retrieved Wage Report 2018/19: What lies behind gender pay from https://www.ifc.org/wps/wcm/connec gaps. Retrieved from https://www.ilo.org/global/ t/0d6860804d7381029697b748b49f4568/ publications/books/WCMS_650553/lang--en/index. IFC_GenderICReformAssessments_Timor. htm pdf?MOD=AJPERES 166 Enterprising Women | Appendix & References International Labour Organization (2019). Jovanovic, B. (1982). Selection and the Evolution ILOSTAT database. Retrieved from https://www. of Industry. Econometrica, 50(3), 649-670. ilo.org/ilostat/faces/wcnav_defaultSelection?_ doi:10.2307/1912606 afrLoop=299799283218449&_afrWindowMode=0&_ Kaboski, J. P., & Townsend, R. M. (2011). A Structural afrWindowId=null#! percent40 percent40 Evaluation of a Large-Scale Quasi-Experimental percent3F_afrWindowId percent3Dnull percent26_ Microfinance Initiative. Econometrica, 79(5), 1357– afrLoop percent3D299799283218449 percent26_ 1406. https://doi.org/10.3982/ECTA7079 afrWindowMode percent3D0 percent26_adf.ctrl- Kangasharju, A. (2007). Do Wage Subsidies Increase state percent3D1619aebs49_91 Employment in Subsidized Firms?. Economica, Iqbal, S., Islam, A., Ramalho, R., & Sakhonchik 74(293), 51-67. https://doi.org/10.1111/j.1468- A. (2016). “Unequal before the Law: Measuring 0335.2006.00525.x Legal Gender Disparities across the World.” Policy Kapoor, M., Ranjan, P., & Raychaudhuri, J. (2012). Research Working Paper 7803. Washington, D.C.: The impact of credit constraints on exporting The World Bank. Retrieved from http://documents. firms: Evidence from the provision and subsequent worldbank.org/curated/en/320521471975957942/ Unequal-before-the-law-measuring-legal-gender- removal of subsidised credit. The World Economy, disparities-across-the-world 40(12), 2854-2874. https://doi.org/10.1111/ twec.12498 Isenberg, D. J. (2010). How to Start an Entrepreneurial Revolution. Harvard Business Karlan, D., McConnell, M., Mullainathan, S., & Zinman, J. (2010). Getting to the Top of Mind: How Review, June. Retrieved from https://institute.coop/ sites/default/files/resources/Isenberg percent20- Reminders Increase Saving. NBER Working Paper percent20How percent20to percent20Start No. 16205. National Bureau of Economic Research. percent20an percent20Entrepreneurial Retrieved from https://www.nber.org/papers/ percent20Revolution.pdf w16205.pdf Iversen, J., Malchow-Møller, N., & Sørensen, Karlan, D., & Zinman, J. (2011). Microcredit in Theory A. (2016). Success in Entrepreneurship: A and Practice: Using Randomized Credit Scoring for Complementarity Between Schooling and Wage- Impact Evaluation. Science, 332(6035), 1278–1284. Work Experience. Small Business Economics, 47(2), https://doi.org/10.1126/science.1200138 437–460. https://doi.org/10.1007/s11187-016- Karlan, D., & Zinman, J. (2018). Price and control 9732-y elasticities of demand for savings. Journal of Jain, M. (2016). Public pre-schooling and maternal Development Economics, 130, 145-159. https://doi. org/10.1016/j.jdeveco.2017.10.004 labour force participation in rural India. Oxford Development Studies, 44(2), 246-263. Kast, F., Meier, S., & Pomeranz, D. (2018). Saving more in groups: Field experimental evidence from Jann, B. (2008). The Blinder–Oaxaca decomposition for linear regression models. The Stata Journal, 8(4), Chile. Journal of Development Economics, 133, 275- 453–479. 294. https://doi.org/10.1016/j.jdeveco.2018.01.006 Jennings, J. E., & Brush, C. G. (2013). Research on Kelley, D. J., Baumer, B. S., Cole, M., Dean, M., & Women Entrepreneurs: Challenges to (and from) the Heavlow, R. (2017). Women’s Entrepreneurship Broader Entrepreneurship Literature?. The Academy 2016/2017 Report. Global Entrepreneurship Research Association. Retrieved from http://www. of Management Annals, 7(1), 661–713. http://dx.doi. gemconsortium.org/report org/10.1080/19416520.2013.782190 Khandker, S. R., Bakht, Z., & Koolwal, G. B. (2006). Jones, F. L., & Kelley, J. (1984). Decomposing The Poverty Impact of Rural Roads: Evidence Differences between Groups: A cautionary note on from Bangladesh. Washington, D.C.: The World measuring discrimination. Sociological Methods & Research, 12(3), 323-343. Bank. Retrieved from https://openknowledge. worldbank.org/bitstream/handle/10986/8333/ wps38750rev0pdf.pdf?sequence=1&isAllowed=y 167 Enterprising Women | Appendix & References Khandker, S. R., Barnes, D. F., & Samad, H. A. (2013). Kyegombe, N., Abramsky, T., Devries, K. M., Welfare Impacts of Rural Electrification: A Panel Starmann, E., Michau, L., Nakuti, J., Musuya, T., Data Analysis from Vietnam. Economic Development Heise, L., & Watts, C. (2014). The impact of SASA!, and Cultural Change, 61(3), 659–692. https://doi. a community mobilization intervention, on reported org/10.1086/669262 HIV-related risk behaviours and relationship Kirkwood, J. (2009). Is a Lack of Self‐confidence dynamics in Kampala, Uganda. Journal of the Hindering Women Entrepreneurs? International International AIDS Society, 17(1), 19232. https://doi. Journal of Gender and Entrepreneurship, 1(2), 118– org/10.7448/IAS.17.1.19232 133. https://doi.org/10.1108/17566260910969670 Lacovone, L., Maloney, W., & McKenzie, D. Kite, M. E., Deaux, K., & Haines, E. L. (2008). Gender (2019). Improving Management with Individual Stereotypes. In F. L. Denmark & M. A. Paludi (Eds.), and Group-Based Consulting: Results from a Psychology of Women: A Handbook of Issues and Randomized Experiment in Colombia. World Bank Theories (2nd ed., pp. 205–236). Westport, CT: Policy Research Working Paper. 8854. Retrieved Greenwood. from https://openknowledge.worldbank.org/ handle/10986/31712 Klapper, L. F., Lewin, A., & Quesada Delgado, J. M. (2009). The Impact of the Business Environment Lafortune, J., Riutort, J., & Tessada, J. (2018). Role on the Business Creation Process, Policy Research Models or Individual Consulting: The Impact of Working Paper 4937, Washington, D.C.: The World Personalizing Micro-entrepreneurship Training. Bank. Retrieved from http://documents.worldbank. American Economic Journal: Applied Economics, org/curated/en/537121468149672866/The-impact- 10(4), 222–245. https://doi.org/10.1257/ of-the-business-environment-on-the-business- app.20170077 creation-process La Porta, R., & Shleifer, A. (2014). Informality and Klapper, L. F., & Parker, S. C. (2011). Gender and the Development. Journal of Economic Perspectives, Business Environment for New Firm Creation. The 28(3), 109–126. https://doi.org/10.1257/ World Bank Research Observer, 26(2), 237–257. jep.28.3.109 https://doi.org/10.1093/wbro/lkp032 Lakdawala, L. (2018). From Loans to Labor: Access Klinger, B., Khwaja, A., & LaMonte, J. (2013). to Credit, Entrepreneurship, and Child Labor. Improving Credit Risk Analysis with Psychometrics SSRN Electronic Journal. https://doi.org/10.2139/ in Peru. Inter-American Development Bank. ssrn.3211376 Kobeissi, N. (2010). Gender Factors and Female Langowitz, N., & Minniti, M. (2007). The Entrepreneurship: International Evidence and Entrepreneurial Propensity of Women. Policy Implications. Journal of International Entrepreneurship Theory and Practice, 31(3), Entrepreneurship, 8(1), 1–35. https://doi. 341–364. https://doi.org/10.1111/j.1540- org/10.1007/s10843-010-0045-y 6520.2007.00177.x Kondo, T., Orbeta, A., Dingcong, C., & Infantado, C. Leahy, C., Lunel, J., Grant, M., & Willetts, J. (2017). (2008). Impact of Microfinance on Rural Households Women in WASH. Enterprises: Learning from in the Philippines. IDS Bulletin, 39(1), 51–70. https:// Female Entrepreneurship in Cambodia, Indonesia doi.org/10.1111/j.1759-5436.2008.tb00432.x and Lao PDR. The Institute for Sustainable Futures (ISF-UTS) at the University of Technology Sydney. Koolwal, G., & van de Walle, D. (2013). Access Retrieved from https://opus.lib.uts.edu.au/ to Water, Women’s Work, and Child Outcomes. bitstream/10453/122061/1/Working-Paper-6_ Economic Development and Cultural Change, 61(2), Women-in-WASH-Enterprises.pdf 369–405. https://doi.org/10.1086/668280 Knowles, J. C. (2013). Increasing the Productivity and Earnings of Rural Women: What Works? Washington, D.C.: UN Foundation. 168 Enterprising Women | Appendix & References Lechmann, D., & Schnabel, C. (2012). Why is Love, I., Martínez Pería, M. S., & Singh, S. (2013). there a gender earning gap in self-employment? Collateral Registries for Movable Assets: Does Their a decomposition analysis with German data. IZA Introduction Spur Firms’ Access to Bank Finance?. Journal of European Labor Studies, 1(6), 1–25. Policy Research Working Paper. 6477. Retrieved https://doi.org/10.1186/2193-9012-1-6 from https://openknowledge.worldbank.org/ Lelarge, C., Sraer, D., & Thesmar, D. (2010). handle/10986/15839 Entrepreneurship and Credit Constraints Evidence Loza, E. (2011). Female Entrepreneurship Theory: from a French Loan Guarantee Program. In J. A Multidisciplinary Review of Resources. Journal Lerner & A. Schoar (Eds.), International Differences of Women’s Entrepreneurship and Education, (1–2), in Entrepreneurship (pp. 243–273). https://doi. 26–64. Retrieved from https://papers.ssrn.com/ org/10.7208/chicago/9780226473109.001.0001 sol3/papers.cfm?abstract_id=1833385 Lemmon, G. T. (2012). Entrepreneurship in Macchiavello, R., Menzel, A., Rabbani, A., & Woodruff, Postconflict Zones. Washington, D.C.: Council on C. (2015). Challenges of Change: An Experiment Foreign Relations. Retrieved from https://www.cfr. Training Women to Manage in the Bangladeshi org/content/publications/.../CFR_WorkingPaper12_ Garment Sector. Warwick Economic Research Lemmon.pdf Papers. doi: 10.22004/ag.econ.269725 Lim, K. (2019). Do American mothers use self- Macchiavello, R. ,Rabbani, A., & Woodruff, C. employment as a flexible work alternative?. Rev Econ (2015). The market for training services: a demand Household, 17, 805–842. https://doi.org/10.1007/ experiment with Bangladeshi garment factories. s11150-018-9426-0 American Economic Review. 105(5), 300-304. Lock, R., & Smith, H. (2016). The Impact of Female Macours, K., & Vakis, R. (2014). Changing Entrepreneurship on Economic Growth in Kenya. Households’ Investment Behaviour through International Journal of Gender and Entrepreneurship, Social Interactions with Local Leaders: Evidence 8(1), 90–96. https://doi.org/10.1108/IJGE-11-2015- from a Randomised Transfer Programme. The 0040 Economic Journal, 124(576), 607–633. https://doi. Lokshin, M., & Yemtsov, R. (2005). Has Rural org/10.1111/ecoj.12145 Infrastructure Rehabilitation in Georgia Helped the Maffoli, A., McKenzie, D. & Ubfal, D. (2020). Poor? The World Bank Economic Review, 19(2), Estimating the demand for business training: 311–333. https://doi.org/10.1093/wber/lhi007 Evidence from Jamaica. World Bank Policy Research Looney, R. (2006). Economic Consequences of Working Paper. 9415. Conflict: The Rise of Iraq’s Informal Economy. Malaya, M. (2006). A Gender-Based Analysis of Journal of Economic Issues, 40(4), 991–1007. Performance of Small and Medium Printing Firms https://doi.org/10.1080/00213624.2006.11506971 in Metro Manila. Journal of International Women’s Loscocco, K., & Bird, S. R. (2012). Gendered Paths: Studies, 8(1), 83–95. Retrieved from https:// Why Women Lag Behind Men in Small Business vc.bridgew.edu/jiws/vol8/iss1/6/ Success. Work and Occupations, 39(2), 183-219. Malla, M. B., Bruce, N., Bates, E., & Rehfuess, E. doi:10.1177/0730888412444282. (2011). Applying global cost-benefit analysis Loscocco, K. A., Robinson, J., Hall, R. H., & Allen, J. methods to indoor air pollution mitigation K. (1991). Gender and Small Business Success: An interventions in Nepal, Kenya and Sudan: Insights Inquiry into Women’s Relative Disadvantage. Social and challenges. Elsevier Energy Policy, 39, 7518– Forces, 70(1), 65–85. https://doi.org/10.1093/ 7529. doi:10.1016/j.enpol.2011.06.031 sf/70.1.65 Manolova, T. S., Brush, C. G., Edelman, L. F., Robb, Loscocco, K., & Smith-Hunter, A. (2004). Women A., & Welter, F. (Eds.). (2017). Entrepreneurial Home-Based Business Owners: Insights from Ecosystems and Growth of Women’s Comparative Analyses. Women in Management Entrepreneurship: A Comparative Analysis. Review. 19, 164–173. Cheltenham, UK: Edward Elgar Publishing. 169 Enterprising Women | Appendix & References Mannava, A., Perova, E., & Tran, P. (2020). Who McKenzie, D. (2020). Small Business Training to Benefits from Better Roads and Why? Mixed Improve Management Practices in Developing Methods Analysis of the Gender-Disaggregated Countries Reassessing the Evidence for “Training Impacts of a Rural Roads Project in Vietnam. Doesn’t Work”. World Bank Policy Research Working Policy Research Working Paper. 9216. Retrieved Paper. 9408. from https://openknowledge.worldbank.org/ McKenzie, D. and C. Woodruff. 2014. “What handle/10986/33637 Are We Learning from Business Training and Manning, C. (1998). Indonesian Labour in Transition: Entrepreneurship Evaluations around the Developing An East Asian Success Story? Cambridge, MA: World?” World Bank Research Observer 9(1): 48-82: Cambridge University Press. doi:10.1093/wbro/lkt007 Martin, B. C., McNally, J. J., & Kay, M. J. (2013). McKenzie, D., & Puerto, S. (2021). Growing Markets Examining the Formation of Human Capital through Business Training for Female Entrepreneurs: in Entrepreneurship: A Meta-Analysis of A Market-Level Randomized Experiment in Kenya. Entrepreneurship Education Outcomes. Journal of American Economic Journal: Applied Economics, Business Venturing, 28(2), 211–224. https://doi. 13(2), 297-332. org/10.1016/j.jbusvent.2012.03.002 McKenzie, D., & Woodruff, C. (2014). What Martinez, S., Naudeau, S., & Pereira, V. (2012). The Are We Learning from Business Training and Promise of Preschool in Africa: A Randomized Entrepreneurship Evaluations around the Developing Impact Evaluation of Early Childhood Development World? The World Bank Research Observer, 29(1), in Rural Mozambique. Washington, D.C.: The 48–82. https://doi.org/10.1093/wbro/lkt007 World Bank and Save the Children. Retrieved from McKenzie, D., & Woodruff, C. (2016). Business http://siteresources.worldbank.org/INTAFRICA/ Practices in Small Firms in Developing Countries. Resources/The_Promise_of_Preschool_in_Africa_ Management Science. 63(9). ECD_REPORT.pdf McManus, L. F. (2011). The Anatomy of a Helping Martínez, C., & Perticará, M. (2017). Childcare Hand: Women-Owned Small Businesses and Federal Effects on Maternal Employment: Evidence from Contract Procurement. William and Mary Journal Chile. Journal of Development Economics, 126, 127– of Women and the Law, 18(3), 625–650. Retrieved 137. https://doi.org/10.1016/j.jdeveco.2017.01.001 from https://scholarship.law.wm.edu/wmjowl/vol18/ Mattis, M. C. (2004). Women entrepreneurs: Out iss3/7 from under the glass ceiling. Women in Management McMillan, J., Woodruff, C. (1999). Interfirm Review, 19, 154–163. Relationships and Informal Credit in Vietnam. The McKenzie, D. (2015). Identifying and Spurring High- Quarterly Journal of Economics, 114(4), 1285-1320. Growth Entrepreneurship: Experimental Evidence Meager, R. (2019). Understanding the Average from a Business Plan Competition. World Bank Impact of Microcredit Expansions: A Bayesian Policy Research Working Paper. 7991. Hierarchical Analysis of Seven Randomized McKenzie, D. (2017). Identifying and Spurring High- Experiments. American Economic Journal: Applied Growth Entrepreneurship: Experimental Evidence Economics, 11(1), 57-91. doi: 10.1257/app.20170299 from a Business Plan Competition. American Meeks, R. C. (2017). Water Works: The Economic Economic Review, 107(8), 2278–2307. https://doi. Impact of Water Infrastructure. Journal of Human org/10.1257/aer.20151404 Resources, 52(4), 1119–1153. https://doi. McKenzie, D. (2017). Growing Markets through org/10.3368/jhr.52.4.0915-7408R1 Business Training for Female Entrepreneurs: A Meeks, R. (2018). Property Rights and Water Market-Level Randomized Experiment in Kenya. Access: Evidence from Land Titling in Rural Peru. World Bank Policy Research Working Paper. 7993. World Development, 102, 345–357. https://doi. org/10.1016/j.worlddev.2017.07.011 170 Enterprising Women | Appendix & References Mehra, R., Patel, P., Shetty, A., Golla, A. (2012). Muñoz Boudet, A. M., Petesch, P., Turk, C., & Financial Services for Low-Income Women: Thumala, A. (2013). On Norms and Agency: Opportunities for Economic Empowerment?. ICRW. Conversations about Gender Equality with Retrieved from https://www.icrw.org/wp-content/ Women and Men in 20 Countries. Directions in uploads/2016/10/ICRW_Financial-Services-and- development: human development. Retrieved WEE.pdf from https://openknowledge.worldbank.org/ Meier, R., & Masters, R. (1988). Sex Differences and handle/10986/13818 Risk-Taking Propensity of Entrepreneurs. Journal of Nghiem, H. S., Coelli, T., & Rao, P. (2012). The Welfare Small Business Management, 26(1), 31. Retrieved Effects of Microfinance in Vietnam: Empirical from https://www.questia.com/read/1G1-6515339/ Results from a Quasi-experiment Survey. Journal of sex-differences-and-risk-taking-propensity-of- Development Studies, 48(5), 619–632. https://doi.org entrepreneurs /10.1080/00220388.2011.638051 Miller, C., Peck, J., & Seflek, M. (2019). Integration Nguyen, C., Frederick, H., & Nguyen, H. (2014). Costs and Missing Women in Firms. NBER Working Female Entrepreneurship in Rural Vietnam: Paper. w26271. Retrieved from https://ssrn.com/ An Exploratory Study. In K. V. Lewis, C. Henry, abstract=3454372 E. J. Gatewood, & J. Watson (Eds.), Women’s Minniti, M. (2009). Gender Issues in Entrepreneurship in the 21st Century: An International Entrepreneurship (SSRN Scholarly Paper No. ID Multi-Level Research Analysis. Cheltenham, UK: 1543464). Rochester, NY: Social Science Research Edward Elgar Publishing. Network. Retrieved from https://papers.ssrn.com/ Niederle, M., & Vesterlund, L. (2007). Do Women abstract=1543464 Shy Away From Competition? Do Men Compete Minniti, M. (2010). Female Entrepreneurship Too Much? The Quarterly Journal of Economics, and Economic Activity. The European Journal of 122(3), 1067–1101. https://doi.org/10.1162/ Development Research, 22(3), 294–312. https://doi. qjec.122.3.1067 org/10.1057/ejdr.2010.18 Niederle, M., & Vesterlund, L. (2008). Gender Minniti, M., & Naudé, W. (2010). What Do We Know Differences in Competition. Negotiation Journal, About the Patterns and Determinants of Female 24(4), 447-463. https://doi.org/10.1111/j.1571- Entrepreneurship Across Countries? The European 9979.2008.00197.x Journal of Development Research, 22(3), 277–293. Niederle, M., & Vesterlund, L. (2011). Gender and https://doi.org/10.1057/ejdr.2010.17 Competition. Annual Review of Economics, 3(1), 601- Moss-Racusin, C. A., Dovidio, J. F., Brescoll, V. L., 630. Graham, M. J., & Handelsman, J. (2012). Science Niethammer, C. (2013). Women, Entrepreneurship Faculty’s Subtle Gender Biases Favor Male Students. and the Opportunity to Promote Development and Proceedings of the National Academyu of Sciences, Business. Washington, D.C.: The Brookings Institute. 109, 16474-16479. Retrieved from https://www.brookings.edu/wp- Morris, M. H., Miyasaki, N. N., Watters, C. E., & content/uploads/2016/06/2013-BBR-Women- Coombes, S. M. (2006). The dilemma of growth: Entrepreneurship.pdf Understanding venture size choices of women Nordman, C. J., & Vaillant, J. (2014). Inputs, Gender entrepreneurs. Journal of Small Business Roles or Sharing Norms? Assessing the Gender Management. 44, 221–244. Performance Gap among Informal Entrepreneurs Mu, R., & Walle, D. van de. (2011). Rural Roads and in Madagascar (IZA Discussion Papers No. 8046). Local Market Development in Vietnam. The Journal Institute for the Study of Labor (IZA). Retrieved from of Development Studies, 47(5), 709–734. https://doi. https://ideas.repec.org/p/iza/izadps/dp8046.html org/10.1080/00220381003599436 Noseleit, F. (2014). Female Self-employment and Children. Small Business Economics, 43(3), 549– 569. https://doi.org/10.1007/s11187-014-9570-8 171 Enterprising Women | Appendix & References OECD. (2003). Own-account workers – ILO. Retrieved Patnam, M. (2013). Corporate Networks and Peer from https://stats.oecd.org/glossary/detail. Effects in Firm Policies. Working Paper. asp?ID=1986 Patrick, C., Stephens, H., & Weinstein, A. (2016). OECD. (2017). Strengthening Women’s Where are All the Self-employed Women? Push Entrepreneurship in ASEAN. OECD. Retrieved and Pull Factors Influencing Female Labor Market from https://www.oecd.org/southeast-asia/ Decisions. Small Business Economics, 46(3), 365– regional-programme/Strengthening_Womens_ 390. https://doi.org/10.1007/s11187-015-9697-2 Entrepreneurship_ASEAN.pdf Petts R. J., & Knoester C. (2018). Paternity Leave- OECD. (2018). SME and Entrepreneurship Policy Taking and Father Engagement. Journal of Marriage in Indonesia 2018. Paris: OECD Publishing. and Family, 80(5), 1144-1162. doi: 10.1111/ Retrieved from https://www.oecd-ilibrary.org/ jomf.12494 docserver/9789264306264-en.pdf?expires=1550597 Prah, R. K. D., Carrion, D., Oppong, F. B., Tawiah, T., 550&id=id&accname=ocid195787&checksum=7473 Mujtaba, M. N., Gyaase, S., Kwarteng, A., Ae-Ngibise, 171CFD2BDADB1AACE5A2784080F7 K. A., Agyei, O., Twumasi, M., et al. (2021). Time Use Oh, I., Lee, JD., Heshmati, A., et al. (2009). Evaluation Implication of Clean Cookstoves in Rural Settings of credit guarantee policy using propensity score in Ghana: A Time Use Study. Int. J. Environ. Res. matching. Small Bus Econ, 33, 335–351. https://doi. Public Health, 18, 166. https://doi.org/10.3390/ org/10.1007/s11187-008-9102-5 ijerph18010166 Orhan, M., & Scott, D. (2001). Why Women Enter into Premand, P., Brodmann, S., Almeida, R., Grun, R., & Entrepreneurship: An Explanatory Model. Women in Barouni, M. (2012). Entrepreneurship Training and Management Review, 16(5), 232–247. https://doi. Self-Employment among University Graduates: org/10.1108/09649420110395719 Evidence from a Randomized Trial in Tunisia. Padilla‐Romo, M., & Cabrera‐Hernández, F. (2018). Washington, D.C.: The World Bank. https://doi. Easing the constraints of motherhood: the effects of org/10.1596/1813-9450-6285 all‐day schools on mothers’ labor supply. Economic Poggesi, S., Mari, M., & De Vita, L. (2016). What’s Inquiry, 57(2), 890-909. https://doi.org/10.1111/ New in Female Entrepreneurship Research? Answers ecin.12740 from the Literature. International Entrepreneurship Paes de Barros, R., Olinto, P., Lunde, T., & Carvalho, and Management Journal, 12(3), 735–764. https:// M. (2011). “The Impact of Free Childcare on doi.org/10.1007/s11365-015-0364-5 Women’s Labor Force Participation: Evidence from Porter, G. (1995). The Impact of Road Construction Low-Income Neighborhoods of Rio de Janeiro.” on Women’s Trade in Rural Nigeria. Journal of Conference Paper, World Bank Economists’ Forum. Transport Geography, 3(1), 3–14. https://doi. Parente, S., Prescott, E. (1994). Barriers to org/10.1016/0966-6923(94)00011-D Technology Adoption and Development. Journal of Powell, G. N., & Greenhaus, J. H. (2010). Sex, Gender, Political Economy, 102(2), 298-321. and Decisions at the Family-Work Interface. Journal Paryavi, M., Campos, F., & Santos, I. (2018). Gender of Management, 36(4), 1011–1039. https://doi. and willingness to compete: lab experiments in org/10.1177/0149206309350774 Kenya. Draft paper. Powell, N., Swartling, Å. G., & Mihn Ha, H. (2011). Patel, P. (2014). Effectiveness of Entrepreneurship The National Policy Landscape: the legacy and Development Interventions for Women the present Situation. In N. Powell, Å. G. Swartling, Entrepreneurs. Geneva: ILO. Retrieved from https:// & H. Mihn Ha (Eds.), Stakeholder Agency in Rural www.ilo.org/wcmsp5/groups/public/---ed_emp/- Development Policy: Articulating Co-governance in --emp_ent/---ifp_seed/documents/publication/ Vietnam. Hanoi: World Agroforestry Center ICRAF. wcms_329356.pdf 172 Enterprising Women | Appendix & References Ray, I. (2007). Women, Water, and Development. Schaner, S. (2017). The Cost of Convenience? Annual Review of Environment and Resources, Transaction Costs, Bargaining Power, and Savings 32(1), 421–449. https://doi.org/10.1146/annurev. Account Use in Kenya. The Journal of Human energy.32.041806.143704 Resources. doi:10.3368/jhr.52.4.0815-7350R1 Reuben, E., Sapienza, P., & Zingales, L. (2014). How Schjoedt, L., & Shaver, K. G. G. (2007). Deciding Stereotypes Impair Women’s Careers in Science. on an Entrepreneurial Career: A Test of the Pull Proceedings of the National Academy of Sciences and Push Hypotheses Using the Panel Study of USA, 111, 4403-4408. Entrepreneurial Dynamics Data. Entrepreneurship Revenga, A., & Dooley, M. (2020). What works Theory and Practice, 31(5), 733–752. http://dx.doi. for women microentrepreneurs?: A meta- org/10.1111/j.1540-6520.2007.00197.x analysis of recent evaluations to support female Schutjens, V. A. J. M., & Wever, E. (2000). entrepreneurship. Global Economy and Development Determinants of New Firm Success. Papers in Brookings Institution Working paper, 142. Regional Science, 79(2), 135–159. https://doi. Reynolds, J., & Renzulli, L. A. (2005). Economic org/10.1111/j.1435-5597.2000.tb00765.x Freedom or Self-Imposed Strife: Work-Life Conflict, Sen, A. 1999. Development as Freedom.New York: Gender, and Self-Employment. L. A. Keister (ed.). Alfred Knopf Research in the Sociology of Work, 15, 33–60. Sewell, S. J., & Desai, S. (2016). The Impacts of Reynolds, P. D., Camp, M. S., Bygrave, W. D., Erkko, A., Undeveloped Roads on the Livelihoods of Rural & Hay, M. (2001). The 2001 Executive Report. Global Women. Review of Social Sciences, 1(8), 15–29. Entrepreneurship Research Association. Retrieved https://doi.org/10.18533/rss.v1i8.40 from http://www.gemconsortium.org/report Shankar, S. (2016). Bridging the “Missing Middle” Rijkers, B., & Costa, R. (2012). Gender and Rural between Microfinance and Small and Medium-Sized Non-Farm Entrepreneurship. World Development, Enterprise Finance in South Asia. SSRN Electronic 40(12), 2411-2426. https://doi.org/10.1016/j. Journal. https://doi.org/10.2139/ssrn.2838083 worlddev.2012.05.017 Shankar, A. V., Onyura, M. A., & Alderman, J. (2015). Sabarwal, S., & Terrell, K. (2008). Does Gender Agency-Based Empowerment Training Enhances Matter for Firm Performance? Evidence from Eastern Sales Capacity of Female Energy Entrepreneurs Europe and Central Asia (SSRN Scholarly Paper in Kenya, Journal of Health Communication: No. ID 1262646). Rochester, NY: Social Science International Perspectives, 20(1), 67-75. doi: Research Network. 10.1080/10810730.2014.1002959 Sapienza, P., Zingales, L., & Maestripieri, D. (2009). Sharif, M. Y. (2015). Glass Ceiling, the Prime Gender Differences in Financial Risk Aversion and Driver of Women Entrepreneurship in Malaysia: Career Choices are Affected by Testosterone. A Phenomenological Study of Women Lawyers. Proceedings of the National Academy of Sciences, Procedia – Social and Behavioral Sciences, 169, 329– 106(36), 15268–15273. https://doi.org/10.1073/ 336. https://doi.org/10.1016/j.sbspro.2015.01.317 pnas.0907352106 Shinnar, R. S., Hsu, D. K., & Powell, B. C. (2014). Sasakawa Peace Foundation and Dalberg Global Self-efficacy, Entrepreneurial Intentions, and Development Advisors. (2017). Advancing Women’s Gender: Assessing the Impact of Entrepreneurship Empowerment: Growing Women’s Entrepreneurship Education Longitudinally. The International Journal through ICT on Southeast Asia. Retrieved of Management Education, 12(3), 561–570. https:// from https://www.spf.org/awif/wp-content/ doi.org/10.1016/j.ijme.2014.09.005 uploads/2018/05/Womens-Entrepreneurship-and- Singh, S. P., Reynolds, R. G., & Muhammad, S. (2001). ICT-SE-Asia_2017_en-2.pdf A Gender-Based Performance Analysis of Micro and Small Enterprises in Java, Indonesia. Journal of Small Business Management, 39(2), 174–182. https://doi.org/10.1111/1540-627X.00016 173 Enterprising Women | Appendix & References Sonobe, T., Suzuki, A., Otsuka, K., & Nam, V. H. Teoh, W. M.-Y., & Chong, S.-C. (2007). Theorising (2011). KAIZEN for Managerial Skills Improvement a Framework of factors Influencing Performance in Small and Medium Enterprises: An Impact of Women Entrepreneurs in Malaysia. Journal Evaluation Study in a Knitwear Cluster in of Asia Entrepreneurship and Sustainability, Vietnam. Working Paper Series No. 2012/ 29. III(2). Retrieved from https://www.researchgate. DEPOCEN. Retrieved from http://depocenwp. net/publication/268285452_THEORISING_A_ org/upload/pubs/29- percent20KAIZEN FRAMEWORK_OF_FACTORS_INFLUENCING_ percent20for percent20Managerial percent20Skills PERFORMANCE_OF_WOMEN_ENTREPRENEURS_IN_ percent20Improvement.pdfhttp://depocenwp. MALAYSIA org/upload/pubs/29- percent20KAIZEN Terjesen, S. A., & Lloyd, A. (2015). The 2015 Female percent20for percent20Managerial percent20Skills Entrepreneurship Index. Retrieved from https:// percent20Improvement.pdf papers.ssrn.com/sol3/papers.cfm?abstract_ Stiglitz, J. and A. Weiss. 1981. “Credit Rationing in id=2625254 Markets with Imperfect Information.” The American Tewari, I., & Wang, Y. (2019). Durable Ownership, Economic Review 71(3): 393-410 Time Allocation and Female Labor Force Suri, T. Jack, W. (2016). The long-run poverty Participation: Evidence from China’s “Home and gender impacts of mobile money. Science, Appliances to the Countryside” Rebate. Economic 354(6317), 1288-1292. doi: 10.1126/science. Development and Cultural Change. https://doi. aah5309. org/10.1086/706824 Szarka, J. (1990). Networking and Small Firms. Thébaud, S. (2015). Business as Plan B: International Small Business Journal, 8(2), 10-22. Institutional Foundations of Gender Inequality in Tambunan, T. (2007). Development of SME and Entrepreneurship across 24 Industrialized Countries. Women Entrepreneurs in a Developing Country: The Administrative Science Quarterly, 60(4), 671–711. Indonesian Story. Small Enterprise Research, 15(2), doi: 10.1177/0001839215591627 31–51. https://doi.org/10.5172/ser.15.2.31 Tigges, L. M., & Green, G. P. (1994). Small Business Tambunan, T. (2009). Women Entrepreneurship in Success Among Men- and Women-Owned Firms in Asian Developing Countries: Their Development Rural Areas. Rural Sociology, 59(2), 289–310. https:// and Main Constraints. Journal of Development and doi.org/10.1111/j.1549-0831.1994.tb00534.x Agricultural Economics, 1(2), 27–40. Retrieved from Tilley, E., Bieri, S., & Kohler, P. (2013). Sanitation in http://www.academicjournals.org/app/webroot/ Developing Countries: A Review Through a Gender article/article1379600429_Tambunan.pdf Lens. Journal of Water, Sanitation and Hygiene Tambunan, T. (2015). Development of Women for Development, 3(3), 298–314. https://doi. Entrepreneurs in Indonesia: Are They Being org/10.2166/washdev.2013.090 “Pushed” or “Pulled”? Journal of Social Economics, UNICEF. (2017). UNICEF Cambodia: Girls not brides 2(3), 131–149. Retrieved from http://rassweb. – ending child marriage in Cambodia. Retrieved from org/admin/pages/ResearchPapers/Paper http://unicefcambodia.blogspot.com/2017/03/by- percent206_1495886589.pdf chan-kanha-and-ream-rin-romas-is.html Tambunan, T. (2017). Women Entrepreneurs in UNICEF & World Health Organization. (2017). MSEs in Indonesia: Their Motivations and Main Progress on Drinking Water Sanitation and Hygiene: Constraints. International Journal of Gender 2017 Update and SDG Baselines. Retrieved from and Women’s Studies, 5(1), 88–100. https://doi. https://www.unicef.org/publications/files/Progress_ org/10.15640/ijgws.v5n1a9 on_Drinking_Water_Sanitation_and_Hygiene_2017. Teignier, M., & Cuberes, D. (2014). Aggregate Costs pdf of Gender Gaps in the Labor Market: A Quantitative Estimate. SSRN Journal. https://doi.org/10.2139/ ssrn.2405006 174 Enterprising Women | Appendix & References United Nations Economic and Social Commission Valdivia, M. (2015). Business training plus for for Asia and the Pacific. (2013). Enabling female entrepreneurship? Short andmedium- Entrepreneurship for Women’s Economic term experimental evidence from Peru. Journal of Empowerment in Asia and the Pacific. UNESCAP. Development Economics, 113, 33-51. http://dx.doi. Retrieved from https://www.unescap.org/sites/ org/10.1016/j.jdeveco.2014.10.005 default/files/Enabling percent20women percent27s Valerio, A., Parton, B., & Robb, A. (2014). percent20entrepreneurship.pdf Entrepreneurship Education and Training United Nations Economic and Social Commission Programs around the World. Washington, for Asia and the Pacific. (2017). Fostering Women’s D.C.: The World Bank. Retrieved from https:// Entrepreneurship in ASEAN: Transforming Prospects, openknowledge.worldbank.org/bitstream/hand Transforming Societies. UNESCAP. Retrieved from le/10986/18031/9781464802027.pdf https://www.unescap.org/sites/default/files/ESCAP- Van Der Sluis, J., Van Praag, M., & Vijverberg, FWE-ASEAN-full_0.pdf W. (2008). Education and Entrepreneurship United Nations Economic and Social Commission Selection and Performance: A Review of the for Asia and the Pacific. (2019). Unpaid Work in Empirical Literature. Journal of Economic Surveys, Asia and the Pacific. UNESCAP Social Development 22(5), 795–841. https://doi.org/10.1111/j.1467- Policy Paper. Retrieved from https://www.unescap. 6419.2008.00550.x org/sites/default/files/PP_2019-02_Unpaid Vandegrift, D., & Yavas, A. (2009). Men, Women, percent20Work.pdf and Competition: An Experimental Test of Behavior. United Nations Secretary-General’s High-Level Panel Journal of Economic Behavior & Organization, on Women’s Economic Empowerment. (2016). Leave 72(1), 554–570. https://doi.org/10.1016/j. No One Behind: Taking Action for Transformational jebo.2009.06.003 Change on Women’s Economic Empowerment. VanderBrug, J. (2013). The Global Rise of Female Retrieved from http://hlp-wee.unwomen.org/-/ Entrepreneurs. Harvard Business Review. Retrieved media/hlp percent20wee/attachments/reports- from https://hbr.org/2013/09/global-rise-of-female- toolkits/hlp-wee-report-2016-09-call-to-action-en. entrepreneurs pdf?la=en. Vanek, J. (2013). Entrepreneurship in Informal UN Women. (2020a). Unlocking the lockdown: Employment: Gender Perspective. WIEGO. Retrieved The gendered effects of COVID-19 on achieving from https://unstats.un.org/unsd/gender/ the SDGS in Asia and the Pacific. Retrieved from Events/5.../Session percent203_Vanek.ppt https://data.unwomen.org/publications/unlocking- Varsakelis, N. (2006). Education, political institutions lockdown-gendered-effects-covid-19-achieving-sdgs- and innovative activity: a crosscountry empirical asia-and-pacific investigation. Research Policy, 35, 1083–1090. UN Women. (2020b). Whose time to care: Unpaid Venkatesh, V., Shaw, J., Sykes, T., Wamba, S. & care and domestic work during COVID-19. Retrieved Macharia, M. (2017). Networks, Technology, and from https://data.unwomen.org/sites/default/files/ Entrepreneurship: A Field Quasi-Experiment among inline-files/Whose-time-to-care-brief_0.pdf Women in Rural India. Academy of Management UN Women. (2020c). Leveraging Digitalization to Journal, 60(5), 1709-1740. Cope with COVID-19: An Indonesia Case Study Verheul, I., Thurik, R., Grilo, I., & van der Zwan, on Women-owned Micro and Small Businesses. P. (2012). Explaining Preferences and Actual Retrieved from https://data.unwomen.org/sites/ Involvement in Self-employment: Gender and default/files/documents/Publications/Indonesia- the Entrepreneurial Personality. Journal of Digitalization-Women-MSBs-COVID19-EN.pdf Economic Psychology, 33(2), 325–341. https://doi. org/10.1016/j.joep.2011.02.009 175 Enterprising Women | Appendix & References Vial, V. (2011). Micro-Entrepreneurship in a Hostile Woetzel, J., Madgavkar, A., Ellingrud, K., Labaye, Environment: Evidence from Indonesia. Bulletin E., Devillard, S., Kutcher, E., Manyika J., Dobbs R., of Indonesian Economic Studies, 47(2), 233–262. & Krishnan, M. (2015). The Power of Parity: How https://doi.org/10.1080/00074918.2011.585952 Advancing Women’s Equality Can Add $12 Trillion Vietnam Women Entrepreneurs Council. (2019). to Global Growth. Washington, D.C.: McKinsey About us. Retrieved from http://www.we-apec.com/ Global Institute. Retrieved from https://www. directory/vietnam-women-entrepreneurs-council mckinsey.com/~/media/McKinsey/Featured percent20Insights/Employment percent20and Wang, Q. (2015). Child Care, Work-Family Policy and percent20Growth/How percent20advancing Female Entrepreneurship (SSRN Scholarly Paper percent20womens percent20equality percent20can No. ID 2669282). Rochester, NY: Social Science percent20add percent2012 percent20trillion Research Network. percent20to percent20global percent20growth/MGI Wieland, A. M., Kemmelmeier, M., Gupta, V. K., percent20Power percent20of percent20parity_Full & McKelvey, W. (2019). Gendered Cognitions: A percent20report_September percent202015.ashx Socio-Cognitive Model of How Gender Affects Women’s World Banking. (2018). Access to Finance Entrepreneurial Preferences. Entrepreneurship & of Women-Owned SMEs in Southeast Asia: An Regional Development, 31(3–4), 178–197. https:// Assessment of Five Countries. Retrieved from doi.org/10.1080/08985626.2018.1551787 https://www.womensworldbanking.org/insights-and- Willetts, J., Gero, A., Susamto, A. A., Sanjaya, R., impact/access-to-finance-of-women-owned-smes- Trieu, T. D., Murta, J., & Carrard, N. (2017). Sanitation in-southeast-asia-an-assessment-of-five-countries/ Value Chains in Low Density Settings in Indonesia Wooldridge, J. M. (2012). Introductory econometrics: and Vietnam: Impetus for a Rethink to Achieve a modern approach. Mason, Ohio: South-Western Pro-poor Outcomes. Journal of Water Sanitation and Cengage Learning. Hygiene for Development, 7(3), 445–454. https://doi. org/10.2166/washdev.2017.141 World Bank (2009). Increasing Formality and Productivity of Bolivian Firms. Washington, DC: The Wilson, F., Kickul, J., & Marlino, D. (2007). Gender, World Bank. Retrieved from https://openknowledge. Entrepreneurial Self-Efficacy, and Entrepreneurial worldbank.org/bitstream/handle/10986/2675 Career Intentions: Implications for Entrepreneurship /503580PUB0Box3101OFFICIAL0USE0ONLY1. Education. Entrepreneurship Theory and Practice, pdf?sequence=1&isAllowed=y 31(3), 387–406. https://doi.org/10.1111/j.1540- 6520.2007.00179.x World Bank. 2012a. World Development Report 2012 : Gender Equality and Development. World Bank. © Winsborough, H. H., & Dickenson, P. (1971). World Bank. https://openknowledge.worldbank.org/ Components of negro–white income differences. handle/10986/4391 License: CC BY 3.0 IGO. Proceedings of the Social Statistics Section. 6–8. World Bank. 2012b. Toward Gender Equality in East Asia and the Pacific : A Companion to the World Development Report. World Bank East Asia and Pacific Regional Report;. Washington, DC. © World Bank. https://openknowledge.worldbank.org/ handle/10986/12598 License: CC BY 3.0 IGO. World Bank. (2013). Promoting Youth Entrepreneurship and Employment: The Case of Lao PDR (No. 81674). Washington, D.C.: The World Bank. Retrieved from http://documents.worldbank. org/curated/en/406571468263930377/Promoting- youth-entrepreneurship-and-employment-the-case- of-Lao-PDR 176 Enterprising Women | Appendix & References World Bank. (2016a). Gender Dimensions of World Bank and International Finance Corporation. Small-Scale Cross-Border Trade in Cambodia (2019). Unlocking the Dynamism of the Indonesian and the Lao People’s Democratic Republic. World Private Sector: A Country Private Sector Diagnostic. Bank. https://openknowledge.worldbank.org/ World Bank Group, Washington, DC. https://www.ifc. handle/10986/25759 org/wps/wcm/connect/publications_ext_content/ World Bank. (2016b). Women Entrepreneurs ifc_external_publication_site/publications_listing_ in Indonesia: A Pathway to Increasing Shared page/cpsd-indonesia Prosperity. Jakarta: World Bank. Retrieved from Xavier, S. R., Sidin, S. M., Guelich, U., & Nawangpalupi, http://documents.worldbank.org/curated/ C. (2017). ASEAN Regional Entrepreneurship Report en/738881467782741648/pdf/AUS5568-P147245- 2015/2016: The Context, Impact and Opportunities PUBLIC-WomenEntrepreneursinIndonesia-1.pdf for Women Entrepreneurs and Startups; Key Pivots World Bank. (2016c). Cambodia Economic for Growth and Sustainability. Kuala Lumpur: Update: Enhancing Export Competitiveness The Universiti Tun Abdul Razak. Retrieved from http:// Key to Cambodia’s Future Economic Success. www.gemconsortium.org/report Washington, DC: The World Bank. Retrieved from: Yadav, V., & Unni, J. (2016). Women https://documents1.worldbank.org/curated/ Entrepreneurship: Research Review and Future en/575221480949830789/pdf/108982-WP- Directions. Journal of Global Entrepreneurship ENGLISH-P148100-PUBLIC-FinalCEUOctoberEnglish. Research, 6(1), 1–18. https://doi.org/10.1186/ pdf s40497-016-0055-x World Bank. (2017). Lao PDR – Systematic Country Yishay, A. B., Fraker, A., Guiteras, R., Palloni, G., Shah, Diagnostic: Priorities for Ending Poverty and Boosting N. B., Shirrell, S., & Wang, P. (2017). Microcredit Shared Prosperity. Washington, DC: World Bank and Willingness to Pay for Environmental Quality: Group. http://documents.worldbank.org/curated/ Evidence from a Randomized-controlled Trial of en/983001490107755004/Lao-PDR-Systematic- Finance for Sanitation in Rural Cambodia. Journal Country-Diagnostic-Priorities-for-Ending-Poverty- of Environmental Economics and Management, and-Boosting-Shared-Prosperity 86, 121–140. https://doi.org/10.1016/j. World Bank. (2018). Timor-Leste – Systematic jeem.2016.11.004 Country Diagnostic: Pathways for a New Economy Zhao, H., Seibert, S. E., & Hills, G. E. (2005). The and Sustainable Livelihoods. Washington, DC: World Mediating Role of Self-Efficacy in the Development Bank Group http://documents.worldbank.org/ of Entrepreneurial Intentions. Journal of Applied curated/en/524131528837983427/Timor-Leste- Psychology, 90(6), 1265–1272. https://doi. Systematic-Country-Diagnostic-Pathways-for-a-New- org/10.1037/0021-9010.90.6.1265 Economy-and-Sustainable-Livelihoods Zia, B. H. (2008). Export incentives, financial World Bank. (2019). World Development Indicators constraints, and the (mis)allocation of credit: Micro- database. Retrieved from https://data.worldbank. level evidence from subsidized export loans. Journal org/indicator/SL.TLF.ACTI.MA.ZS of Financial Economics, 87(2), 498-527. https://doi. World Bank. (2020). Project list database. Retrieved org/10.1016/j.jfineco.2006.12.006 from https://projects.worldbank.org/en/projects- operations/projects-list World Bank. (2020a). Women, Business and the Law 2020. Women, Business and the Law. World Bank. doi:10.1596/978-1-4648-1532-4 World Bank. (2020b). Indonesia Systematic Country Diagnostic Update. World Bank, Washington, DC. https://openknowledge.worldbank.org/ handle/10986/34107 License: CC BY 3.0 IGO 177