Policy Research Working Paper 9451 Measuring the Biases, Burdens, and Barriers Women Entrepreneurs Endure in Myanmar Chiara Dall’aglio Fayavar Hayati David Lee Macroeconomics, Trade and Investment Global Practice October 2020 Policy Research Working Paper 9451 Abstract Entrepreneurs in Myanmar face many challenges to start- their businesses and identifies firm-level characteristics ing and operating a business. As is the experience globally, leading to the use of good business practices. Further, the women often experience these challenges to a greater extent analysis investigates the adoption of gender and fami- and face additional sociocultural barriers, limiting their ly-friendly policies, as an outcome and as a determinant equal participation in the economy. To develop a better of business success. The purpose of the study is to gain a understanding of the dynamics holding back private sector better understanding of the barriers to gender-inclusive pri- development, especially for women, this paper uses data vate sector development in Myanmar and provide tangible from the first-of-a-kind, firm-level data set available in recommendations to private- and government-level actors. Myanmar. The analysis explores the variance of experience Overall, the analysis finds the major constraints for women female-owned micro, small, and medium-size enterprises entrepreneurs are access to finance and sociocultural factors, face compared with their male-owned counterparts. The such as family responsibilities and household work. paper assesses the barriers imposed on entrepreneurs and This paper is a product of the Macroeconomics, Trade and Investment Global Practice. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at fhayati@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Measuring the Biases, Burdens, and Barriers Women Entrepreneurs Endure in Myanmar Written by Chiara Dall'aglio, Fayavar Hayati, David Lee. This paper benefited from detailed comments and feedback from Ashani Alles, Sarah Anne Ebrahimi, Bronwyn Grieve, Amy Luinstra, Ellen Maynes, Deepak Mishra, Clarence Tsimpo Nkengne, Sjamsu Rahardja and Degi Young. Earlier input and research was supported by Kyi Kyi Thin. JEL: J16 J71 L26 015 Keywords: Female-owned firms, Access to Finance, Human Resource Policy, Myanmar Table of Contents Introduction: The Myanmar Context ............................................................................................................ 3 Literature Review .......................................................................................................................................... 7 Data and Summary Statistics ...................................................................................................................... 10 Model and Results ...................................................................................................................................... 20 Outcome: Access to Finance ................................................................................................................... 22 Outcome: Business Development Practices ........................................................................................... 26 Outcome: Barriers to growth .................................................................................................................. 29 Outcome: Adoption of informal HRM policies........................................................................................ 32 Recommendations ...................................................................................................................................... 35 Conclusion ................................................................................................................................................... 36 References .................................................................................................................................................. 37 Appendix A – Descriptive statistics table .................................................................................................... 39 Appendix B – Variable description .............................................................................................................. 39 2 Introduction: The Myanmar Context A well-functioning private sector is crucial to the growth and prosperity of all countries. The private sector is the engine of growth for a country by its contribution toward creating jobs and opportunities for people to move out of poverty and improve the quality of their lives (IFC, 2012). 1 The World Bank Enterprise Survey indicates that small firms (<50 employees) in developing countries have the highest job growth rates; thus, understanding how to improve employee retention, job satisfaction, and business growth, is crucial in this sector. In Myanmar, enterprises of all sizes face a range of challenges and barriers to their development, including lack of access to capital, a burdensome bureaucratic and administrative environment, a low-skilled workforce, infrastructure gaps, as well as conflict. As is the experience globally, women face similar challenges to men but often to a greater extent, thereby limiting their equal participation in the economy as leaders, entrepreneurs, employees, consumers, and community members. Given the importance of the private sector to the economic development of a country, it is fundamental that all members of society have equal access. Micro, small and medium-size enterprises (MSMEs) dominate Myanmar’s private sector. According to the OECD Development Pathway Initial Assessment (2013), it was estimated that approximately 83 percent of businesses in Myanmar were in the informal sector. Of the approximately 17 percent in the formal sector, it was estimated that 99.4 percent were small and medium-size enterprises. As such, enhancing the opportunities for MSMEs to grow is a key policy area of interest for a country in transition such as Myanmar. The ongoing liberalization of the economy provides enormous opportunities for Myanmar to grow in terms of economic and social development. Across the world, women face numerous challenges to participate equally in the economy as leaders, entrepreneurs, employees, consumers, and community members. Laws and regulations continue to prevent women from entering the workforce or starting a business, resulting in lasting effects on women’s economic inclusion, labor force participation, and entrepreneurial aspirations. Moreover, barriers such as the lack of family- friendly workplace policies, care and household responsibilities, lack of training or mentoring programs, access to finance, and a lack of safe transportation, create a situation that prevents women from achieving their full potential. This, in turn, prevents the whole of society from reaching its full potential. According to a World Bank Group study, 2 countries are collectively losing USD160 trillion in wealth because of differences in lifetime earnings between women and men and the situation for women in Myanmar, as this study confirms, is no different to the experience of women globally. Currently, the benefits from growth are not being shared equally in Myanmar. The World Bank Group’s World Development Indicators (2019) shows that the labor force participation rate among those aged 15 and over in Myanmar is 77 percent for men but only 48 percent for women. The Myanmar Annual Labor Force Survey (2017) 1 Private sector matters for job creation, IFC, 2012. 2 Unrealized Potential: The High Cost of Gender Inequality in Earnings, World Bank, 2018. 3 reported that of those women who are working, most are likely to be in informal employment. In terms of business ownership and management, only 35 percent of firms are reported as having female participation in ownership, and 41 percent as having a female top manager, despite women having higher average education levels than men (10.3 years of schooling compared to 9.8 for men). Moreover, a recent World Bank study (2020) 3 highlights how nationally women are widely excluded from paid labor, with men being twice as likely as women to be engaged in paid work. According to the report, women’s exclusion from paid labor did not stem from unwillingness to participate but from time consumed on unpaid labor, lack of job opportunities, limited access to credit and lack of skills. Husbands and family members surveyed as part of this report appeared supportive of women accessing paid labor as long as activities outside the household would not interfere with their full responsibility of housework and child and elderly care. The Equal Measures 2030: 2019 Report, which examines progress against gender equality in the Sustainable Development Goals (SDGs) across 129 countries, placed Myanmar at 98th. The country’s most significant gender gaps were found in the economic participation of women, equal representation in government/powerful positions, and gender-based violence. In the Asia and the Pacific region, Myanmar scores on the lower end of the gender index positioning itself only slightly above Nepal, Lao PDR, Bangladesh, and Pakistan. Table 1 contains some select Equal Measures indicators showing the gender gap in Myanmar. Table 1. Select indicators from Equal Measures 2019 Report Indicator Score out of 100 Proportion of ministerial/senior government positions held by women 10 Proportion of seats held by women in national parliament 20.3 Extent to which the country has laws mandating women’s workplace equality 20.0 Proportion of women who hold a bank account at a financial institution 26.0 Proportion of women who have made or received digital payments in the past year 7.4 Proportion of women with access to internet services 27.6 Level of personal autonomy, individual rights and freedom from discrimination 31.3 Openness of gender statistics 45.0 The participation of women in Myanmar’s labor force is less than that of men, as is their representation in senior or decision-making roles across the public and private sectors. Despite women accounting for 51.8 percent of Myanmar’s population, a United Nations Capital Development Fund Assessment of women and girls’ financial inclusion estimated that women owned only around 25 percent of all MSMEs in Myanmar (28 percent of small enterprises and 20 percent of medium enterprises), predominantly in services, but also in the manufacture of textiles 3 Women’s agency in Mon and Kayin States, World Bank, June 2020. 4 and footwear. Large enterprises were found to be predominantly owned by men, with women owning only around 7 percent of businesses with more than 100 employees. The report also noted that most MSMEs, including those owned by women, remained in the informal sector, which in turn, hampered access to formal finance. 4 An unpublished assessment of the relationship between gender and firm performance from the World Bank Enterprise Survey data found that female-owned firms tend to have lower sales revenues and number of workers. 5 The assessment also found that firm characteristics such as firm size and sector were driving the performance gap between male and female owned and operated firms, rather than just gender. Further, female-owned small firms in the manufacturing sector consistently had lower labor productivity and sales revenues compared to their male- owned counterparts of the same firm size and in the same sector. Myanmar’s social and cultural traits present significant challenges to addressing gender inequality. While the Constitution prohibits the state from discrimination of its citizens based on gender, the same Constitution also states that “…nothing shall prevent appointment of men to the positions that are suitable for men only.” A gender equality situational analysis for Myanmar noted that over the past decade there have been several improvements in the economic and social status indicators for women in Myanmar. 6 However, this improvement has helped nurture a pervasive assumption that men and women enjoy equal status and opportunity, and that this is a unique trait of Myanmar society. While on paper Myanmar is a signatory to multiple international and regional conventions against discrimination, and the Constitution prohibits discrimination based on gender, 7 in reality, social and cultural norms continue to undermine gender equality. Several parts of the legislation regarding family law, including divorce, are discriminatory towards women. For example, adultery alone, if committed by a man, is not grounds for divorce, however, women found guilty of adultery alone may be ordered by the court to relinquish their rights to joint marital property for the benefit of the husband or children. 8 The situational analysis noted that women continue to bear the major responsibility for unpaid care work in addition to their paid jobs, while men are typically the household heads. A World Bank (2020) study reports that there is a lack of women’s agency over the division of labor in the household and that this constitutes a burden to women. In the survey this was also indicated as the main impediment to accessing both paid labor and community participation. Gender segregation can also be seen in the education system. The Admission Standards page of the government operated Myanmar Maritime University states that 320, or 80 percent of students, will be male, and the remaining 20 percent will be female. 9 Traditional views also shape 4 http://www.uncdf.org/download/file/127/6266/uncdf-power-country-assessment---myanmar.pdf 5 Women Entrepreneurs in Myanmar: What does the Enterprise Survey Data Tell Us? Working Document, WBG (Anne Ong Lopez, Sjamsu Rahardja) 2017. 6 Asian Development Bank, United Nations Development Program, United Nations Population Fund, and the United Nations Entity for Gender Equality and the Empowerment of Women, 2016, Gender equality and women’s rights in Myanmar: A situation analysis. 7 At the international level, Myanmar is a signatory to the Beijing Platform for Action and the United Nations Convention on the Elimination of All Forms of Discrimination Against Women (CEDAW), and the 2030 Agenda for Sustainable Development Goals (SDG). At the regional level, Myanmar is an active member of the ASEAN Commission on Protection and Promotion of the Rights of Women and Children (ACWC) and the ASEAN Committee on Women (ACW). The Myanmar Constitution prohibits the state’s discrimination in many forms against its citizens, including sex. The Constitution also guarantees equal pay for similar work and promises equal rights under the law for both genders. 8 Gender Equality Network, “Raising the Curtain: Cultural Norms, Social Practices and Gender Equality in Myanmar,” 48; Crouch, “Promiscuity, Polygyny, and the Power of Revenge: The Past and Future of Burmese Buddhist Law in Myanmar,” 85–104; The Burma Divorce Act, 39 Section X. 9 http://www.mmu.edu.mm/AdmissionStandards.aspx (downloaded 22 October 2019). 5 expectations around education, marriage, employment, and occupation. Moreover, when it comes to actively participating in community-level decision making, women enjoy very limited agency and they are also less likely to hold positions of leadership. This can be mainly connected to the widespread belief that men’s opinions matter more than women’s, the overwhelming weight of household work and family duties, the strength of the status quo and the fear of transgressing gender norms. In addition, there is evidence that elected women officials continue to face discrimination in access to training, information, day-to-day interactions and how they are treated by male colleagues (World Bank, 2020). Ensuring that women entrepreneurs and employees gain from the opportunities presented by economic development must be a priority. Development of gender-related support interventions necessitates an understanding of the differences between male and female entrepreneurs. This should include the sectors in which they operate, location, the experience of social and physical barriers, flexible workplace policies, and whether performance gaps exist between male and female entrepreneurial firms, including the drivers behind these gaps. The aim of the note is to gain a better understanding of the barriers to gender-inclusive Private Sector Development (PSD) in Myanmar, explore the differential experience female-owned enterprises have in comparison to their male- owned counterparts, and finally, to provide recommendations for inclusive PSD based on the findings. To achieve this, the note uses data from the 2019 Myanmar Gender and Entrepreneurship Survey (MGES) to explore the differential experiences of female-owned micro, small and medium sized enterprises in comparison to their male counterparts. Taking the above into account, this note covers and highlights the following issues: • Characteristics of female and male owned enterprises • Barriers faced by women entrepreneurs • Determinants of Human Resource Management (HRM) policies • Impact of HRM policies on business development • Impact of HRM policies on workplace attitude/employee satisfaction and other areas. The survey data were analyzed from a gender perspective to create an overview of the SME environment in Myanmar, assess the barriers and challenges faced by entrepreneurs and their businesses, and identify firm-level characteristics leading to the use of good business practices. The analysis also investigates the adoption of gender and family-friendly policies both as an outcome and as a determinant of business success. This document brings together the analysis of the MGES with other contextual information about the gender dimensions of PSD in Myanmar. 6 Literature Review A significant body of research on gender and entrepreneurship exists globally, however, it is a relatively new area of study for Myanmar. The research listed in the following paragraphs presents the latest findings and thinking which exists globally, regionally, and where available, specifically to Myanmar. It covers the adoption of informal employee benefits, such as family friendly working conditions and the provision of sick leave and maternity leave, as well as the barriers and challenges to business growth relevant to both male and female entrepreneurs. • Barriers to business growth In Myanmar, SMEs face a range of challenges and barriers to their development. Local businesses have been suffering from challenges such as lack of access to capital and burdensome tax and monetary policies. One of the largest barriers to doing business as an SME in Myanmar is the lack of access to finance and high interest rates. 10 Bureaucratic and administrative processes create an added burden on business owners, indeed, access to public services is reduced due to cumbersome and time-consuming procedures (Kapteyn and Wah, 2016). The regulatory environment is complex, which results in different degrees of non-compliance in several areas of enterprise activity. Many different government authorities are responsible for business registration and licensing. Enterprises perceive that the role of government in relieving business difficulties could be in providing easier access to credit, assisting with technical know-how, and easing access to quality raw materials. 11 All of these factors limit the opportunities for entrepreneurs to grow their businesses. An additional limiting factor for the development of the private sector in Myanmar is the ongoing political conflict of the past 60 years. The nexus between fragility and business in the case of Myanmar is unique and is marked by idiosyncratic characteristics of the country’s historical trajectory, ethnic subnational conflict, and other cultural elements, which make it significantly different from other fragile contexts (Egreteau and Mangan, 2018). Conflict affects business development through several different channels; trade through ethnic areas suffers from the legacy and persistence of conflict; transit fees, levied at checkpoints at the porous borders in Myanmar’s mountainous periphery, disincentivize trade, business growth and long-term investment in these areas (Jolliffe, 2015); furthermore, the informal trade that transits through these areas, undermines legitimate businesses, as well as, government efforts to mobilize taxes and build better trade facilitation institutions (Egreteau and Mangan, 2018). In Myanmar, women entrepreneurs face different and more severe challenges compared to their male counterparts. Constraints faced by women entrepreneurs in low income countries mainly arise from gender discrimination, work-family conflict, difficulty raising capital, lack of infrastructure, unstable business, economic and political (BEP) environments, and lack of training and education (Panda, 2018), in addition to heavy household chores and limited access to finance (Tambunan, 2017). Moreover, a study conducted in Indonesia by The Asia Foundation (2013) found that access to finance is a constraint for women because they use personal savings for the initial capital 10 Descriptive report, Myanmar Micro, Small and Medium Enterprise Survey, 2017. 11 Ibid. 7 needed for a business, they see government support as inaccessible and unsupportive for business, and most importantly, household and domestic responsibilities limit their participation in business. The major impediment to the emergence of women entrepreneurs is the cultural, gender-based traditions regarding the role of women in society. 12 These include the burden of household and family responsibilities, general lack of respect for women entrepreneurs, and an unwelcoming environment for women entrepreneurs in male- dominated sectors. While not only specific to women-owned businesses, SMEs in Myanmar face many challenges related to access and use of financial instruments. Lack of financial service sophistication, short loan tenors, and onerous collateral requirements, are all developmental challenges faced by firms in the country. 13 Moreover, research suggests that women entrepreneurs face specific barriers when dealing with bureaucracy, this is quite relevant in the context of Myanmar. These challenges may arise from lower educational attainment, literacy levels, and underrepresentation of women in senior positions. 14 To improve women’s participation in the labor force, especially their access to quality employment, progressive firms have adopted formal family-friendly policies aimed at reducing the gender gap. When talking about HRM policies, the literature refers to maternity and paternity leave, subsidized childcare, paid sick and annual leave, housing, and food and transport subsidies. The adoption of such policies, especially maternity leave and childcare, is conducive to a higher participation of women in the workforce (Ferragina, 2017), and they are generally associated with positive employment outcomes for mothers (Misra, Budig and Boeckmann, 2010). The She Works report conducted across several different countries, including LICs, highlights how policies such as flexible workplace and leave, child and elderly care, and equal pay for equal work, are all important for the recruitment and retention of female talent. 15 Moreover, the report calls attention to practices such as complementing adequate parental leave and flexible working arrangements with proactive career management to help women (and men) staying/getting back on the career track. Overall, these practices create an enabling environment for women to be able to participate more in the private sector and to overcome the barriers they face. In the context of Myanmar, the adoption of HRM policies is mandated by the government in the Social Security Law (2012) and the Leave and Holidays Act (1951/2014). • Business development Strengthening SME capacities to improve their competitiveness in domestic, regional and global markets is crucial. According to the OECD report (2004) on promoting SMEs for development, it is important to promote tools, such as value chain analysis, in order to enable entrepreneurs to understand what problems and challenges they need to address within and beyond their own borders. The report also stresses the importance of investments in physical 12 National Assessment of Women’s Entrepreneurship Development Framework Conditions – Myanmar, ILO and Sasakawa Peace Foundation, Draft Executive Summary. 13 Country report Myanmar, Entrepreneurial Ecosystem Assessment, Dutch Good Growth Fund, April 2019. 14 Ibid. 15 She Works, Recruitment and retention of female talent in the workplace, The IFC, 2016. 8 infrastructure and business services and the use of tools such as professional market research findings, advisory services, marketing assistance and training for SME growth. Business development services have also been identified as the most needed competitiveness improving services by SMEs in developing countries. 16 Given the recognized importance of these practices, in our analysis, we use a combination of business development and financial planning measures to create two indices as proxies for business success for SMEs in our sample. Access to finance has been identified as a key element for SMEs to succeed. Financial resources are crucial for SMEs in their drive to build productive capacity, to compete, to create jobs, and to contribute to poverty alleviation in developing countries. 17 Finance has been identified in many business surveys as the most important factor determining the survival and growth of SMEs in both developing and developed countries. Access to finance allows businesses to undertake investments to expand their operations and improve their competitiveness. In this paper, we proceed to use access to finance as another indicator to measure business development of the SMEs in our study. Adoption of formal HRM policies is linked to improvement of business development outcomes and to the creation of an enabling business environment. It has been shown that the adoption of formal family-friendly policies promotes business development, and improves productivity, job satisfaction, and employee retention (Bin Bae and Yang, 2017 and Born, 2015). A UNICEF report (2019) presents evidence that manufacturing workers in Bangladesh and Vietnam are more likely to be retained in workplaces that implement family-friendly policies and shows results from a study in China which found that HRM policies were an important factor that increased trust in management and overall employee retention—by reducing the cost of staff absenteeism and turnover, they would increase business performance by increasing productivity (IFC, 2019). At a global level, it has been shown that the adoption of HRM policies is positively correlated with business development, organizational performance and firm productivity (Bloom, Kretschmer and Van Reenen, 2011, Jones, Kalmi and Kauhanen, 2006, Arthur and Cook, 2004 and De Kok and Uhlaner, 2001), including for smaller businesses (<100 employees) (Sels, De Winne, Delmotte, 2006). Shafeek (2016), studies the impact of HR policies in SMEs (ranging from micro-firms, <10 employees to large firms with over 250 employees), in the context of Saud Arabia. He finds that the adoption of these practices positively affected SME performance regardless of size. Naz et al., (2016) and Yordanova, (2011), in their studies in Pakistan and Bulgaria, provide further evidence supporting this finding. Overall HRM policies are conducive to business growth, improved productivity, employee retention, and a reduced gender gap due to higher female employee retention, even when looking at small firms. In our study, we are going to evaluate whether the adoption of informal employee practices has a similar impact on different measures of business development and access to finance. Further, given that evidence points to formal family-friendly policies being beneficial in reducing gender barriers, we proceed to test whether a similar story can be told in our context. 16 A survey conducted in Tanzania indicates that micro and small enterprises perceived training in bookkeeping, financial services, marketing and sales promotion, training in costing and pricing and transport services as the most needed competitiveness improving services for their businesses. BDS market survey conducted by Swisscontact 2003. 17 Improving the competitiveness of SMEs in developing countries. The Role of Finance to Enhance Enterprise Development, UNCTAD, 2001. 9 To our knowledge this is the first paper that investigates the relationship between informal policies and measures of business success. • Adoption of informal employee benefits Small and medium enterprises in Myanmar provide their employees with ‘informal’ employee benefits. The literature suggests that there are several different drivers at both the country and firm levels pushing businesses to adopt family-friendly policies. In a study set in India, Budhwar (2000), found evidence that age, size, ownership, type of industry, and union membership, are all positively related to the adoption of HRM policies. Other studies focus on either country level or firm-level characteristics. A country’s regulation and a firm’s compliance (UNICEF, 2019), the size of the firm, and the percentage of female employees (Poelmans, Chincilla and Cardona, 2003 and Ozturkler and Ozutku, 2009), are all positively related factors to the likelihood of a firm having family-friendly policies in place. Overall, a country’s legislation, the dynamics of the business environment, and the overall national culture, play an important role. At the firm level, the sector in which the business operates, the size of the firm, and the ratio of female employees, seem conducive to the adoption of better HRM practices. Evidence suggests that large organizations are better able to provide a broad base of work-family benefits and tend to adopt more policies pertaining to individual support and dependent-care benefits than smaller organizations (Bardoel, Tharenou and Moss, 2013). This is interesting because in our data gathered from micro and SMEs, we find that the majority of these enterprises are offering what we refer to as ‘informal HRM practices’, including provisions for food, housing, sick leave and maternity leave, among other things. We proceed to test whether the main characteristics, recognized by the literature to be conducive to the adoption of these practices, are also significant in the adoption of family- friendly policies in the private sector in Myanmar. Data and Summary Statistics • Research methodology The data used for this section come from the MGES 2019. The survey was conducted to collect data to understand how policy changes, and formal and informal support programs, can achieve greater impact for female entrepreneurs (description of variables used in the analysis in Appendix A). The focus is on how gender can impact specific entrepreneurs in the non-farming agriculture, light-manufacturing, hospitality and tourism, and trade and logistic sectors. Data was collected with the aim to inform the WBG/IFC operations on private sector development and dialogue with government counterparts on promoting more inclusive enterprise development in Myanmar. 18 Particularly the survey aims to shed light on the following: • How cultural norms and decision making in the household interact with decision making within an enterprise; 18 Myanmar Gender Entrepreneurship Survey Report, Myanmar Survey Research, 2019. 10 • How household responsibilities and childcare arrangements interact with entrepreneurial activity; • How men and women may experience barriers to growth differently; • How gender differences manifest in accessing and using business development tools, for example access to finance, and the use of good managerial practices; • How gender differences manifest in sector activities; and • How these issues and challenges may differ in areas that have been, or are affected, by conflict. The survey was carried out between the end of 2018 and the beginning of 2019, and targeted micro firms (1-5 employees), small firms (6-10 employees), and medium sized firms (>10 employees). Data were collected from 513 male and female owned businesses in the following sectors: • Non-farming agriculture – Milling and processing, agriculture wholesaling, beverage and food processing • Light manufacturing – Garments, textiles, leather; handicrafts and furniture; fabric metals, machinery and electronics; publishing, printing, media; rubber and plastics • Tourism and hospitality – Hotels and guesthouses, cafes and restaurants, tour operators • Trade and logistics – Retailing and wholesaling, freight forwarders, customs agents, warehousing, courier services; freight transport operators. The research was conducted in the urban areas of five states/regions of Myanmar. The townships selected for the study were selected using a purposive sampling approach, with priority given to townships where the required sectors could be found. Both conflict and non-conflict affected townships were selected. The conflict townships were selected where no clashes had taken place in the past 6 months, in order to ensure the team’s safety. In Table 2 we can see the townships and the number of firms selected per area. Table 2. Geographic and Sectoral Dimension 19 Conflict/Non- Non-Farming Light Hospitality & Trade & Township conflict Agriculture Manufacturing Tourism Logistics Moegaung Conflict 10 10 10 10 Kachin Myitkyina Non-conflict 10 10 10 13 Myawaddy Conflict 10 10 10 10 Kayin Hpa-an Non-conflict 10 10 10 10 Lashio Conflict 10 10 10 11 Shan Taunggyi Non-conflict 11 11 10 11 Chanmyatharzen 11 12 10 11 Mandalay Non-conflict Maharaungmyay 9 11 11 14 Tamwe 10 14 9 14 Yangon Dagon Myothit Non-conflict 10 12 11 17 South Okkalapa 10 10 10 10 19 Myanmar Gender Entrepreneurship Survey Report, Myanmar Survey Research, 2019. 11 Insein 11 10 10 9 Total 122 130 121 140 Once the townships were selected, we proceeded with the selection of urban blocks in which to carry out the enumeration and the selection of firms. Four starting points were selected per township and at least 10 interviews were conducted per starting point, giving a total of 40 interviews per township. For the selection of respondents, interviewers had to first pick a corner as a starting point. From there the interviewers walked around the block in search of enterprises within the four sectors included in the survey, with a target of ten interviews to complete in each Enumeration Area (EA). When an enterprise was identified, a screening questionnaire was used to determine if that enterprise met the requirements of firms to be included in the study. 20 When an enterprise was deemed eligible, interviews were conducted with owners or managers and the findings are presented based on the owners’ perspective. Enumerator training included a detailed explanation of the background and objectives of the survey, as well as an explanation of the rights of respondents, including privacy. The training also included sampling design, call-back procedures, detailed explanation of the questionnaire, question by question and a comprehensive discussion of probing. Moreover, respondent confidentiality, quality control procedures, and how to adopt a gender sensitive approach when interviewing respondents, were also covered. The sample was purposely selected to have a rough share of 30 percent male-owned firms and 70 percent female- owned businesses. Indeed 379 women were interviewed, of which 317 (84 percent) were owners. 134 men were interviewed, 73 of which (54 percent) were owners. The interviews were conducted face-to-face. • Snapshot of the business environment In this section we provide an overall snapshot of the business environment of the firms in the sample, highlighting some of the most interesting findings stemming from the MGES survey. 21 Women tend to operate smaller, younger, and home-based businesses. Starting with firm-level characteristics, one of the relationships that stands out is between the size of the firm and the gender of the owner: women are more likely to be part of smaller firms than men. Women are also more likely to have their businesses at home, implying that they usually sleep and work in the same location. This could indicate the limiting influence of cultural and gender-based traditions in the opportunities to establish a business, as well as family obligations that may prevent them from growing their business. These include family discouragement towards women becoming entrepreneurs and mobility restrictions constraining women’s access to markets (places where they can go and cannot go, security issues). 22 Moreover, there is also a relationship between the size of the firm and number of years in operation. While 48 percent of micro firms have been operating for 5 years or less, larger firms have been in the business for a longer time. In fact, 37 percent of small, and 40 percent of medium sized firms have been operating for 11 years or more. 20 Myanmar Gender Entrepreneurship Survey Report, Myanmar Survey Research, 2019. 21 This section also relies on the MGES report, Myanmar Survey Research, 2019. 22 ILO and Sasakawa, National Assessment of Women’s Entrepreneurship Development Framework Conditions – Myanmar, Draft executive summary. 12 Thus, the larger the firms, the older they are. This might be related to a number of different factors: the recent establishment of services related to communications and electronic devices, the expansion of small businesses in more recent years due to major economic, political and economic reforms, and more opportunities in general for entrepreneurs. It is also possible that smaller firms are only able to sustain themselves for a few years before closure. Businesses present consistent characteristics within sector of operation. The survey identified a wide range of businesses; indeed, the 513 surveyed firms were classified into 63 different types. Tea shops and printing services were the most common in the sample, but still accounted for no more than 11 percent of the total. The most common business run by women are printing services (11 percent), tea shops (8 percent), non-specialized stores with food and beverage (7 percent), and wholesale and retail of clothing (7 percent). The educational background of the owner is related, to some extent, to the industry the respondents work in. Those in tourism and hospitality have less years of schooling compared to other industries. Another difference across sectors shows that trade and logistics businesses have started operating more recently, possibly related to increased business activity and internal mobility in recent years. The trend indicates political and economic restructuring influencing the development of the latter. In line with this, business related to the more traditional non-farming and light manufacturing tend to be older. Small and medium firms in Myanmar offer a number of HRM benefits. Firms in the sample offer a range of informal benefits, including, maternity/paternity leave, subsidized childcare, annual and sick leave, and subsidized housing. Among all the businesses operating in the different industries, the most commonly offered policy is paid sick leave. Across the difference industries, the most noticeable pattern comes from firms operating in the hospitality sector, which are most likely to provide subsidized housing. On the other hand, maternity leave is low across all firms (42 percent), regardless of the owner’s gender. This represents an extra challenge for mothers wanting to remain in the workforce and could also partly explain why women entrepreneurs are older than men by their entering the workforce once their children have grown. This pattern can be clearly seen in Chart 1. Chart 1. Age distribution of owners by gender 30% 33% 24% 24% 21% 20% 15% 13% 10% 3% 4% 1% 0 1% 18 - 20 21 – 30 31 – 40 41 – 50 51 – 60 61 - 70 70 - 75 Male Female The lack of maternity leave also represents a decrease of income for mothers, which in turn is related to autonomy, social equality, and ability to plan for the future. 23 In addition, subsidized childcare helps women who have, or plan 23 ILO and Sasakawa, National Assessment of Women’s Entrepreneurship Development Framework Conditions – Myanmar, Draft executive summary. 13 to have children, enter and continue in the work force, however, the proportions of firms providing subsidized childcare are particularly low. While female and male owned businesses face similar barriers, female entrepreneurs are more likely to have reduced access to finance. The amount of collateral a woman entrepreneur needs to provide to secure a loan is larger than the amount needed by men business owners, by an overage of over one million Myanmar kyats 24 According to the ILO-Sasakawa report, not only are women entrepreneurs asked to provide higher collateral, but they are disadvantaged in meeting those higher collateral requirements because they are less likely than men to hold title to land and property. 25 Group-lending through microfinance institutions (MFIs) is the most commonly reported lending methodology used by women in Myanmar (amounts of ~USD 400), with individual loans, by law, capped at USD 10,000. This type of borrowing presents a significant financial gap for women business owners interested in taking their microenterprise to the next level of growth because the minimum threshold for bank lending is ~ USD 50,000. Another limiting factor is that women entrepreneurs use their own personal savings and informal sources of external financing to finance their entrepreneurial activities more than men. 26 Women-owned businesses are less likely overall to use financial services. The survey shows a low proportion of women-owners with bank accounts compared to men-owners. This difference is statistically significant, as can be seen in Table 4. Women-owned businesses are less likely to have a bank account, borrow less, and are more often required to provide higher collateral as security, making access to finance one of their main challenges. Family responsibilities and household work are the biggest challenges for growth reported by both genders, but more so for women. This can be explained by the fact that women spend disproportionately more time on household work and looking after family than their male counterparts. On average, women spend almost two more hours per week caring for dependents and six more hours doing household chores, as can be seen in the descriptive statistics table (Table 3a) reported below. More than twice as many women entrepreneurs report gender as a barrier to business growth than male entrepreneurs. Conflict, a key aspect in the context of Myanmar, affects respondents in different and non-uniform ways. As mentioned above, conflict areas in Shan have more access to banking than those in non-conflict areas, while Kachin presents the opposite scenario. Moreover, education attainment is lower in conflict-affected areas; indeed, entrepreneurs operating in these areas report fewer years of schooling. In the conflict areas of Kachin and Kayin, business owners on average have completed less than high school. Shan is once again the exception as respondents in these conflict areas report quite high levels of education completed: on average respondents in Shan conflict areas have completed high school and some university. This is most likely due to Lashio, a conflict area in Shan, having a university and teachers’ college, while Mogaung, in Kachin, has neither. 24 736 USD as of August 2020. 25 ILO and Sasakawa, National Assessment of Women’s Entrepreneurship Development Framework Conditions – Myanmar, Draft executive summary. 26 Ibid. 14 • The highest number of firms with bank accounts were in the conflict areas of Shan State. Indeed, 61 percent of firms in the conflict area in Shan have a bank account, compared to only 10 percent of firms in the conflict areas of Kachin. This finding could be partly related to the large presence of Chinese banks, particularly in Northern Shan. • Descriptive statistics In this section we provide summary statistics for the variables used in our analysis. Table 3a provides summary statistics on key characteristics of the firms in our sample by gender of owner, while Table 3b presents results on the perception of the various barriers to business growth by gender of owner. Table 4 presents balance tests results carried out on the same variables by gender of owner. Panel B of Table 3a provides descriptive statistics on the informal HRM practices adopted. It is important to note that while respondents were surveyed, they were not asked to provide proof of whether a gender equality policy or any of the other HRM policies were in place in their business. 27 In most cases, especially for smaller businesses, we assume that while these practices exist, they are informal. In the context of Myanmar, it is common for Small and Medium Enterprises to provide their employees with food, a place to sleep, and sick and maternity leave, however, these policies are not formalized as they might be in a larger company. The interpretation of having a gender equality policy may vary considerably between respondents, ranging from being open to hiring women as well as men, to paying them equally. 28 Firms in our sample have existed for an average of 10 years and hold a permit to operate. As seen in panel A of Table 3a, male-owned firms are almost 50 percent larger than women-owned businesses, with an average of 9.5 employees versus 6.6 for female-owned enterprises. There is little and non-significant variation between male and female owned firms in the share of employees with a high level of education (completed high school or above), and in the share of firms who have a license or permit to operate, as almost all firms have this (94 percent). However, female-owned businesses employ more women than male-owned businesses, with more than half their employees being women (~57 percent). While there are more female than male-owned businesses working in conflict areas (25 compared to 19 percent, respectively), the difference between the two is not significant, as can be seen in Table 4. There is no clear pattern between female and male owned firms when it comes to implementing family-oriented practices. On average, the share of male and female owned enterprises offering these benefits are similar, and in most cases, the differences are not significant. The only exception is found in the provision of subsidized food, which is provided more by women than male owned enterprises (78 percent vs 68 percent). It must be noted that there is 28 Unfortunately,enumerators were not instructed to collect information on what having a gender equality policy in place meant for the business they were surveying. To adjust for concerns over differing interpretations, in the analysis, respondents were excluded who said they had a gender equality policy but did not use any other informal HRM practices. 15 no significant difference between male and female owners of restaurants/businesses in the tourism sector (which includes teashops and restaurants), as can be seen in Table 4. Women entrepreneurs own significantly fewer bank accounts. While the banking system in Myanmar has improved considerably since the country opened up, a very high proportion of SMEs do not have a bank account, more so amongst women. On average, 20 percent more male-owned businesses have a bank account than female-owned businesses. As having a bank account is a requirement to apply for a bank loan, this also affects women unfavorably when trying to access credit to expand their businesses. Indeed, this highlights a gender-gap in the use and access of financial instruments in Myanmar. Overall, as compared to men, women face structural barriers as they borrow considerably less, are more likely to be asked for higher collaterals (61 vs 50 percent of men who have borrowed), are less likely to have bank accounts and have smaller businesses. There are no major differences in the adoption of ‘good’ business practices between male and female entrepreneurs. When it comes to adopting business practices such as advertising, negotiating with suppliers, record keeping, and preparing a budget and cash flow statements, there is no gender-gap. This is an interesting finding, as in a previous study conducted in 2017, 29 focusing only on the manufacturing sector, it was found that female entrepreneurs are less likely to apply as many beneficial business practices as their male counterparts. Overall, in our sample, the adoption of good business practices is high, with over 50 percent of businesses adopting at least one. Women entrepreneurs spend twice as much time as male business owners doing household chores. In a month, women owners will spend 24 hours more than men doing household chores, again indicating the burden of cultural norms imposed on female entrepreneurs. This is consistent with studies on the gender barriers in the country. One of the most consistently raised issues by women trying to enter the business ecosystem in Myanmar is their excessive burden of domestic and household responsibilities. 30 Despite women business owners facing discrimination, gender is not perceived as a barrier to growth by women entrepreneurs themselves. This can probably be imputed to the fact that gendered roles are so engrained in Myanmar society that the different opportunities one has depending on their gender is accepted as the inviolable norm. Despite women entrepreneurs carrying out twice as much household related work than men, they do not report their gender as a barrier to the growth of their business (Table 3b Appendix A). In our sample, only 18% of women reported their gender as some sort of barrier, with the vast majority reported gender as not being a barrier at all. 29 Myanmar MSME Survey, 2017. 30Ibid. 16 Table 3a. Summary Statistics Female Owner Mean SD N Male Owner Mean SD N Panel A: Firm Characteristics % owners with high education 64.91 0.48 379 57.46 0.50 134 Age of firm 10.50 11.26 379 10.54 11.91 134 Total employees 6.64 8.75 379 9.46 10.61 134 % female employees 56.76 0.40 379 41.49 0.33 134 % employees with high education 26.69 0.38 379 28.88 0.36 134 % of firms operating in a conflict area 25.33 0.44 379 18.66 0.39 134 % of firms with license or permit to operate 93.93 0.42 379 94.03 0.24 134 Panel B: HRM Policies % firms providing: Gender equality policy 74.79 0.44 361 69.17 0.46 133 Subsidized childcare 22.04 0.42 372 21.09 0.41 128 Paid annual leave 31.21 0.46 378 37.59 0.49 133 Sick leave 86.77 0.34 378 81.34 0.39 134 Maternity leave 43.38 0.50 355 46.46 0.50 127 Paternity leave 38.85 0.49 350 45.24 0.50 126 Subzidized housing 70.18 0.46 379 63.43 0.48 134 Subsidized food 78.36 0.41 379 67.91 0.47 134 Subsidized transport 7.65 0.27 379 15.67 0.37 134 Panel C: Access to Finance Firm has a bank account 36.55 0.48 372 55.91 0.50 127 Firm has documents to request a loan 61.43 0.49 363 61.29 0.49 124 Firm has a loan 8.89 0.29 371 8.07 0.27 124 Panel D: Finance and Business Development Practices Business Development Past 3 months % of firms that: Visited a competitor to see what products/services they offer and at what price 46.70 0.50 379 49.25 0.50 134 Asked your existing customers whether there are any other products/services they 66.75 0.47 379 70.15 0.46 134 would like you to offer Used a special offer to attract customers 36.41 0.48 379 41.79 0.50 134 Done any form of advertising 23.22 0.42 379 26.12 0.44 134 Attempted to negotiate with a supplier for lower prices 41.42 0.49 379 46.27 0.50 134 % of firmst that: Have a record-keeping system (written or computerized) which allows you to know 64.29 0.48 378 67.91 0.47 134 the quantity of supplies or inventory you have on hand Keep written or computerized records relating to your enterprise 71.24 0.45 379 74.63 0.44 134 Record every purchase and sale made by the enterprise 75.13 0.43 378 72.93 0.45 133 Regularly use your records to know whether sales of a particular product are 64.98 0.48 377 62.69 0.49 134 increasing or decreasing from one month to another Worked out the cost to you of each main product or service you sell 77.78 0.42 378 78.20 0.41 133 Know which goods or services you make the most profit per unit in selling 85.48 0.35 379 80.60 0.40 134 Have a written budget, which tells you how much you have to pay each month for 66.57 0.47 377 65.91 0.48 132 rent, electricity, and other indirect costs of the enterprise? Financial Practices % of firmst that: Have target sale for next year 17.72 0.38 378 17.91 0.38 134 % of firmst that prepare the following at least annually: Profit and loss statement 60.42 0.49 379 62.69 0.49 134 Statement of cash flow 57.78 0.50 379 58.96 0.49 134 Balance sheet 59.63 0.49 379 64.18 0.48 134 Income/Expenditure sheet 72.29 0.45 379 67.91 0.47 134 Panel E: Other activities Hours spent looking after dependents 16.55 10.90 317 14.88 10.54 73 Hours spent doing household chores 12.85 9.87 317 7.29 9.64 73 Hours spent practicing hobbies and personal pursuits 12.63 7.51 317 14.34 9.12 73 Note: The table shows the mean, standard deviation, and number of observations from MGES 2015. Age of firm is reported in number of years, hours spent looking after dependents, doing household chores and doing hobbies refer to weekly averages. All the variables used in the analysis are describes in more detail in Appendix A1. 17 Table 4. Balance Test - Female vs Male Owned Firms Difference in Mean for Variable Mean for Male Owned Female Owned Mean/sd Mean/se % owners with high education 57.46 0.07 (0.50) (0.05) Age of firm 10.54 -0.04 (11.91) (1.15) Total employees 9.46 -2.81*** (10.61) (0.93) Trade 24.62 0.036 (0.43) (0.045) Manufacturing 27.61 -0.031 (0.45) (0.044) Tourism 23.13 0.006 (0.42) (0.043) Non-farming agriculture 24.63 -0.011 (0.43) (0.043) % female employees 41.49 15,28*** (0.33) (3,85) % employees with high education 28.88 -2,20 (0.36) (3.79) % of firms operating in a conflict area 18.66 0.067 (0.39) (0.04) 94.03 0.00 % of firms with license or permit to operate (0.24) (0.02) % firms providing: Gender equality policy 69.17 0.06 (0.46) (0.05) Subsidized childcare 21.09 0.01 (0.41) (0.04) Paid annual leave 37.59 -0.06 (0.49) (0.05) Sick leave 81.34 0.05 (0.39) (0.04) Maternity leave 46.46 -0.03 (0.50) (0.05) Paternity leave 45.24 -0.06 (0.50) (0.05) Subzidized housing 63.43 0.07 (0.48) (0.05) Subsidized food 67.91 0.11** (0.47) (0.04) 15.67 -0.08*** Subsidized transport (0.37) (0.03) Firm has a bank account 55.91 -0.19*** (0.50) 0.05 Firm has documents to request a loan 61.29 0.00 (0.49) (0,05) 8.07 0.01 Firm has a loan (0.27) (0.03) 18 Business Development Past 3 months % of firms that: Visited a competitor to see what products/services they offer and at what price 49.25 -0.03 (0.50) (0,05) Asked your existing customers whether there are any other products/services they 70.15 -0.03 would like you to offer (0.46) (0.05) Used a special offer to attract customers 41.79 -0.05 (0.50) (0.05) Done any form of advertising 26.12 -0.03 (0.44) (0.04) Attempted to negotiate with a supplier for lower prices 46.27 -0.05 (0.50) (0.05) % of firmst that: Have a record-keeping system (written or computerized) which allows you to know 67.91 -0.04 the quantity of supplies or inventory you have on hand (0.47) 0.05 Keep written or computerized records relating to your enterprise 74.63 -0.03 (0.44) (0,05) Record every purchase and sale made by the enterprise 72.93 0.02 (0.45) (0.04) Regularly use your records to know whether sales of a particular product are 62.69 0.02 increasing or decreasing from one month to another (0.49) (0.05) Worked out the cost to you of each main product or service you sell 78.20 0.00 (0.41) (0.04) Know which goods or services you make the most profit per unit in selling 80.60 0.05 (0.40) (0.04) Have a written budget, which tells you how much you have to pay each month for 65.91 0.01 rent, electricity, and other indirect costs of the enterprise? (0.48) (0.05) Financial Practices % of firmst that: Have target sale for next year 17.91 0.00 (0.38) (0.04) % of firmst that prepare the following at least annually: Profit and loss statement 62.69 -0.02 (0.49) (0.05) Statement of cash flow 58.96 -0.01 (0.49) (0.05) Balance sheet 64.18 -0.05 (0.48) (0.05) 67.91 0.04 Income/Expenditure sheet (0.47) (0.05) Hours spent looking after dependents 14.88 1.68 (10.54) (1.41) Hours spent doing household chores 7.29 5.56*** (9.64) (1.28) Hours spent practicing hobbies and personal pursuits 14.34 -1.72* (9.12) (1.02) note: .01 - ***; .05 - **; .1 - *; 19 Model and Results In this section we proceed to test the relationship between firm and employee-level characteristics and our three main outcomes of interest: measures of business development, barriers to growth, and the adoption of informal HRM practices. The purpose of the analysis is to understand what characterizes firms that have good business development measures have in place, whether there is a relationship between having informal HRM practices and measures of business success, whether these affect barriers to business growth and if they affect women and men owned businesses differently. We then turn our attention to study the firm and employee level characteristics that are conducive to having informal HRM practices in place to determine if what is true for formalized policies holds in our case as well. This discussion is relevant in terms of business-level recommendations that we can provide to firms. The analysis adopted in this paper is of an explorative nature and aimed at better understanding the business sector environment with a special focus on the importance of gender in the private sector. We start our analysis by testing which features characterize firms that have adopted measures of business development; access to financial instruments and good business practices. Access to finance is measured as the probability of having a bank account and having documents to request a loan. To study this relationship, we use a probit specification. To test the relationship between firm and employee-level characteristics and two different indices incorporating several different measures of business development and financial planning, we use OLS. In this case the dependent variable are two different indices. 31 For each outcome of interest, we carry out two sets of regressions. In the second set, we use interaction variables to isolate the role played by the owner’s gender. Equation 1a and 2a represent the reduced form specification for the probit and OLS regressions in the first group presented in Tables 5a and 6a. Equations 1a and 2a: Ij =α+Xj β+ϵj (1a) Yj =α+Xj β+ϵj (2a) For these specifications, Ij is the binary variable representing whether firm j has a bank account or the documents to request a loan, while Yj represents the two different indices constituting good business practices and financial planning for firm j. Xj is a set of firm and employee-level covariates and the error term is clustered at township level. As suggested to be important predictors by the literature, (Poelmans, Chincilla and Cardona, 2003; Bardoel, Tharenou and Moss, 2013), covariates include a series of firm characteristics (gender of owner, age and size of firm, whether it operates in a conflict area, whether the firm has a license to operate as a proxy for firm stability, and 31 Details on the indices’ construction can be found in Appendix B. 20 whether it has documents to request a loan as a proxy for financial stability, and finally what sector it operates in), as well as, employee level characteristics (level of employee education, percentage of female employees and the number of hours spent by the owner taking care of dependents and household chores). In these specifications, Xj also includes covariates on whether the firm has any of the informal HRM practices in place, being: gender equality policy (proxy for whether the firm has any of the practices in place as previously mentioned), paternity/maternity leave, subsidized childcare, paid annual/sick leave and housing subsidy. The aim in this case is to understand whether there is any relationship between having these policies in place and our measures of business success. The literature shows that having formal HRM policies is conducive to improved business performance. We thus assess whether this holds true in our context of informal practices as well. In our case, we define performance in terms of access to finance, business and financial development. Equations 1b and 2b: Ij =α+Xj β+ α+( ) + ϵj (1b) Yj =α+Xj β+ α+( ) + ϵj (2b) In the second group of regressions, the outcome variables remain the same: having a bank account, having documents in place to request a loan, and measures of business and financial development. However, in this instance, we seek to understand the effect of different explanatory variables depending on the gender of the owner. Interaction variables were created between the owner’s gender and the following covariates: documents to request a loan, firm size, license to operate, and sector of operation. This will allow us to determine whether characteristics that have already been evaluated as significant, are conducive to better outcomes for a specific gender. We do this to understand the differential experience between genders in terms of using business development practices, and whether the factors have differing effects on male and female businesses. To have a clear interpretation of each interaction term, we report the margins estimations tables after the regression tables, providing an actual predicted value for each combination. For example, in the case of gender of owner and loan documents, the margins table will present the actual predicted value for both genders with and without loan documents enabling a comparison to be made. The next set of equations, 3a and 3b present the reduced form for the ordered probit regressions in Table 7a and 7b: Barriersj =α+Xj β+ϵj (3a) Barriersj =α+ j β+ α+( ) + ϵj (3b) 21 In these regressions, the focus is on barriers to growth faced by business owners. In this case, given that Barriersj is an ordinal categorical variable, we use ordered probit regression. Barriersj measures, on a scale from 0 to 5, how much gender, household chores, family responsibilities, and conflict, represent a barrier to the owner. Xj, in this case, includes the same set of covariates as equations 1 and 2, and includes informal HRM practices. We ran the specification with family responsibility as the outcome of interest on separate samples to understand how different factors affect genders differently. When looking at gender, instead, we only analyze female-owned businesses to try to make sense of the finding that women do not perceive their gender as a barrier despite clear evidence to the contrary. The last barrier we analyze is conflict, as we did for the measures of business success, we include interaction terms (equation 3b) to understand whether operating in a conflict area affects the perception of conflict as a barrier depending on gender. Once again, the margins table for the interaction term is reported. In this case, given the outcome is an ordinal categorical variable, the table will present the predicted value for each combination of the interaction and each possible value of the outcome variable. Finally, we test which traits characterize firms that have adopted informal HRM practices to evaluate whether the same factors highlighted in the literature also hold in our context. Probit regression analysis is used to estimate the probability of having adopted a certain policy depending on the firm and employee level characteristics. One-person firms are excluded from this estimation as their policies would only apply to themselves. Equation 4 presents the reduced form for the informal HRM probit regressions: HRMj =α+Xj β+ϵj (4) The dependent variable in this case is the binary variable representing the adoption of HRMj practices by firm j, Xj is the same set of firm and employee-level covariates previously used, minus the HRM practices which are now our outcomes of interest. Outcome: Access to Finance We start this section by focusing on access to finance, a crucial indicator for business development and growth. Table 5a presents the results without the interaction terms, while Table 5b includes interaction terms. Tables 5c and 5d present the margins result for the predicted values of the interaction terms. Female ownership significantly reduces the probability of the firm to have a bank account. Indeed, being a female business owner reduces the likelihood of the business having a bank account by 38 percent. This supports our discussion on the gender-gap concerning the use of financial instruments in the descriptive statistics section. While this does not hold true when we look at having documents to obtain a loan, there is still a relevant gender factor affecting this outcome as well. In fact, even though the magnitude is not very large (4 percent), having a larger share of female employees reduces the probability of having these documents in place. These findings are in line with 22 other studies on the financial inclusion of women in Myanmar, such as the UNCDF’s PoWER Women and Girls Financial Inclusion Country Assessment, 32 which shows a gender gap in the access to formal and informal finance. Women business owners in Myanmar remain disproportionately excluded from the formal financial sector. The larger the firm is, the more likely it is to have access to financial instruments. Size of firm plays an important role in increasing the likelihood of the firm having a bank account (63 percent) and documents to request a loan (54 percent). Medium sized firms are also more likely to have an account specifically for their business. While 41 percent of medium sized firms have an enterprise account, only 11 percent of small and 8 percent of micro have one. Having a license to operate also increases the likelihood of having a bank account and documents to request a loan (by 21 and 45 percent). The level of education of employees, although small (3 percent), is also a significant and positive predictor of having loan documents. Larger, more stable firms, and firms with better-educated employees, are more likely to use financial services. Having documents to request a loan is also significant in bank account ownership, increasing the likelihood by 55 percent. Age of the firms, on the other hand, reduces the likelihood of firms having necessary documents by 14 percent, the older the firm the less likely they are to have documents in place. Belonging to the trade industry matters for financial access. Firms in the trade industry display a 40 percent higher likelihood of having a bank account, most likely given it is critical in conducting trading business. Moreover, in the context of Myanmar, those companies that hold an import/export license must open a bank account, 33 potentially accounting for part of this result. Interestingly, the source of loans varies according to industry. Non-farming agriculture, trade, and logistics rely mainly on private commercial banks, while hospitality and light manufacturing instead rely on micro-finance institutions. The regression outcomes that include the interaction terms present little variation from those with no interactions. Before turning to the description of the outcomes, we will explain how to interpret the margins tables. In our scenario, we are interacting the gender of the owner with binary explanatory variables, therefore, each interaction has four possible outcomes: female & characteristic – denoted by 1,1; female & no characteristics – 1,0; male & characteristic – 0,1; and, male & no characteristic – 0,0. These outcomes allow us to compare the effects of each of these interaction on the outcome of interest and determine the different impact of one variable depending on the owner’s gender. We start by looking at differentials in bank account ownership and move on to having loan documents. Loan documents matter more if you are a male entrepreneur. Having loan documents increases the likelihood of having a bank account by almost 20 percent more for a male business owner than for a female entrepreneur. In Table 5a we saw how this variable was significantly related to having a bank account, we can now see how the magnitude of the effect actually depends on the gender of the business owner. Overall, a female entrepreneur with 32 http://www.uncdf.org/download/file/127/6266/uncdf-power-country-assessment---myanmar.pdf 33 https://myanmar.gov.mm/en/business/service/-/asset_publisher/idasset472/content/a-2 23 loan documents is 16 percent more likely to have a bank account than a female owner without loan documents but is still 20 percent less likely to have one compared to a male with these documents in place. The effect of firm size on having a bank account presents the smallest difference between genders. The difference in bank account ownership, between a medium sized female and male owned firm is 8 percent (60 vs 68 percent). Larger firm size is conducive to owning a bank account, regardless of gender. Firm formalization only matters for male owners. Female business owners with and without a license to operate present no difference in bank account ownership, both having a 40 percent likelihood of holding a bank account. The narrative changes when looking at male owners. A male-owned registered business is 25 percent more likely to have a bank account than a male entrepreneur with no permit, and 15 percent more likely than a woman with no license to operate. Industry of operation is important for both genders, but more so for male entrepreneurs. Women entrepreneurs working in the trade industry are 13 percent more likely than women entrepreneurs operating in any other sector to have a bank account. This difference for male business owners is 18 percent. However, the difference in bank account ownership between a male and a female business operating in the trade industry remains large, at 17 percent. Female entrepreneurs operating medium sized firms, having permits and belonging to the trade industry are as likely, if not more, to have loan documents in place. The difference in the likelihood of having documents is almost negligible between genders. Having a permit and belonging to the trade industry increases the likelihood for female entrepreneurs to have loan documents by 5 and 11 percent respectively, compared to male entrepreneurs. Despite the impact of having a permit and operating a larger firm is quite similar between genders, the magnitude of increase from not having to having a permit for a bigger firm is larger for male entrepreneurs, while the opposite is true when looking at sector of operation. Overall, this analysis finds that while certain characteristics are conducive to having a bank account or loan documents, the gender of the owner is almost always significant in determining the actual magnitude of the effect. 24 25 Table 5c - Bank account margins Table 5d - Loan documents margins Margin Std. Err. [95% Conf. Interval] Margin Std. Err. [95% Conf. Interval] Female owner#Loan Feamle owner#Medium documents firm 00 0.37 0.07 0.23 0.51 00 0.54 0.08 0.38 0.71 01 0.62 0.08 0.47 0.78 01 0.77 0.05 0.67 0.86 10 0.29 0.04 0.22 0.37 10 0.63 0.03 0.57 0.68 11 0.45 0.04 0.38 0.53 11 0.78 0.06 0.65 0.9 Feamle owner#Medium firm Female owner#Permit 00 0.5 0.07 0.36 0.64 00 0.24 0.17 -0.09 0.57 01 0.68 0.08 0.52 0.84 01 0.6 0.07 0.46 0.74 10 0.35 0.04 0.28 0.42 10 0.58 0.08 0.42 0.73 11 0.6 0.05 0.5 0.7 11 0.65 0.03 0.59 0.72 Female owner#Permit Female owner#Trade 00 0.3 0.18 -0.06 0.66 00 0.58 0.06 0.45 0.7 01 0.55 0.06 0.42 0.67 01 0.58 0.11 0.37 0.79 10 0.4 0.07 0.26 0.53 10 0.63 0.04 0.56 0.7 11 0.4 0.03 0.33 0.46 11 0.69 0.06 0.58 0.8 Female owner#Trade 00 0.48 0.07 0.34 0.62 01 0.66 0.08 0.51 0.82 10 0.36 0.03 0.3 0.42 11 0.49 0.07 0.36 0.63 Outcome: Business Development Practices We now focus on good management practices (Table 6a). To do so we built two indices measuring business development and financial planning by aggregating different binary variables using a summative approach. The components are described in Appendix B. In this case, coefficients of the covariates in the specifications represent an increase or decrease in the overall value of the index, which translates into more or fewer practices being adopted. The business development index takes values from 0 to 12 and the financial planning index from 0 to 6. Table 6b reports the results of the two regressions, including the interaction terms. Tables 6c and 6d report the predicted values of the interaction terms. Firm size, sector of operation, and level of education are all determinants in the adoption of business development and good management practices. Having a medium sized firm increases the number of adopted practices by 1.04. The high level of education of the entrepreneur is positively and significantly related with the adoption of business development measures (~0.9). A similar trend can be seen for firms belonging to the trade sector, which is significantly related to the number of adopted ‘good-practices’. In this case, a firm in the trade sector will have adopted ~0.8 more good management practices compared to a firm in the non-farming agriculture sector. Tourism is instead negatively related to this index, reducing by 0.61 the number of measures adopted. Providing informal HRM practices increases the adoption of business development and financial practices. While the magnitude of the effect is not very large, ~0.4, there is a significant relationship between the provision of maternity/paternity leave and the number of good management practices adopted. A similar relationship can be 26 seen when looking at the adoption of financial planning instruments. In this case, the provision of subsidized housing increases the number of financial instruments in place by 0.32. The stability of the firm, proxied by its ownership of a permit, having documents to request a loan, and the age of the firm, are significant predictors. What seems to matter most for both outcomes is whether the firm has documents in place to request a loan. In both instances, the magnitude of this covariate is very large, ~2.5. This translates into firms having these documents, on average, having 2–3 more business development and financial tools in place, compared to a business with no documents. While quite a small effect, age is negatively related with having business development (-0.02) and financial planning measures (-0.01). An older firm is probably less likely to be innovative in terms of marketing and other good managerial practices. The stability of the firm, instead, is only a positive predictor for business development, increasing the value of the index by ~1. Next, the interaction terms are analyzed to see whether the significant firm-level characteristics we have seen to be conducive to adopting more business development and financial practices differ in magnitude depending on gender. On adoption of good business practices, there is no significant difference observed by gender of the owner. Being a medium sized firm, having loan documents, relevant permits, and belonging to the trade sector, all result in a similar number of business development practices in place regardless of gender. Overall, the magnitude of the impact of these characteristics is slightly higher for men than for women entrepreneurs due to men starting from a lower level of implementation but reaching the same number of adopted practices. For example, having a permit increases the number of good business practices adopted by 2.7 for male business owners, but only 0.36 for female entrepreneurs. The resulting number of practices adopted is then almost the same as 7.38 for male-owned firms and 7.27 for female-owned businesses. A similar trend can be observed when we consider the number of financial planning measures adopted. The resulting number of financial practices adopted once the firm has loan documents, a permit, is medium sized, and operates in the trade sector, varies little across genders. However, the magnitude of the impact varies between male and female firms. Operating a medium sized firm, or belonging to the trade sector, have a slightly larger impact on financial practices used for female-owned firms, whereas having loan documents and a registered firm have a larger impact on the use of financial instruments for male-owned firms. The difference in the magnitude of impact across genders is small, except when looking at having a registered business where the difference is larger for men than women. 27 Table 6a.Business Management Practices Table 6b.Business Management Practices OLS Regression Analysis OLS Regression Analysis Business Business Financial Financial development development planning index planning index index index coef/se coef/se coef/se coef/se Female owner 0.007 0.117 Female owner 2.540* 2.371*** (0.270) (0.148) (1.387) (0.620) High education 0.899*** 0.253 High education 0.911*** 0.268 (0.254) (0.178) (0.258) (0.179) Small firm (6-10) 0.341 -0.144 Small firm (6-10) 0.272 -0.206 (0.386) (0.192) (0.380) (0.202) Medium firm (>10) 1.037*** 0.193 Medium firm (>10) 1.274*** 0.185 (0.327) (0.294) (0.402) (0.415) % of female employees 0.001 -0.002 Medium firm#Female owner -0.239 0.126 (0.002) (0.003) (0.417) (0.431) % employees high education -0.004** -0.001 % of female employees -0.000 -0.003 (0.002) (0.002) (0.003) (0.003) Loan documents 2.524*** 2.424*** % employees high education -0.004** -0.001 (0.322) (0.149) (0.002) (0.002) Firm has a license/permit 0.955*** 0.322 Loan documents 2.574*** 2.251*** (0.330) (0.341) (0.363) (0.256) Age of firm -0.018* -0.013** Loan documents#Female owner -0.309 -0.034 (0.011) (0.006) (0.521) (0.304) Conflict area -0.224 0.102 Firm has license/permit 2.734** 2.120*** (0.526) (0.253) (1.147) (0.484) Hours dependents 0.001 0.006 Firm has permit#Female owner -2.373* -2.400*** (0.016) (0.006) (1.435) (0.635) Hours hh chores 0.013 -0.004 Age of firm -0.019* -0.014** (0.010) (0.006) (0.010) (0.006) Maternity/paternity leave 0.387** 0.000 Conflict area -0.205 0.121 (0.190) (0.140) (0.545) (0.271) Paid/sick leave 0.480 0.106 Hours dependents 0.003 0.007 (0.551) (0.308) (0.017) (0.007) Subsidized housing 0.051 0.316** Hours hh chores 0.010 -0.006 (0.227) (0.156) (0.011) (0.007) Light manufacturing -0.235 -0.015 Maternity/paternity leave 0.396** 0.017 (0.380) (0.223) (0.202) (0.140) Tourism -0.614* -0.209 Paid/sick leave 0.468 0.090 (0.370) (0.175) (0.543) (0.307) Trade 0.806** 0.284 Subsidized housing 0.024 0.291* (0.338) (0.319) (0.240) (0.160) _cons 3.260*** 0.936 Light manufacturing -0.264 -0.063 (0.741) (0.692) (0.419) (0.238) N 513 513 Tourism -0.661* -0.267 note: .01 - ***; .05 - **; .1 - *; (0.397) (0.194) Note: Clustered standard errors by township for each specification. Trade 1.004 0.289 The variables: female owner, high education, small and medium firm, (0.904) (0.531) whether the firm has a permit to operate, conflict area has documents to Trade#Female owner -0.154 0.072 request loan and light manufacturing, tourism and trade are all dummy (0.870) (0.478) variables. _cons 1.567 -0.552 % employees high education - % of employees who have completed at (1.133) (0.745) least high school. N 513 513 note: .01 - ***; .05 - **; .1 - *; Hours dependents and hh chores - number of hours spent looking after dependents and doing hh chores - given these two variables have more than 10% of values missing, we recoded missing to 0 and included a missingness dummy as an additional covariate in the analysis. 28 Table 6c - Business development margins Table 6d - Financial planning margins Margin Std. Err. [95% Conf. Interval] Margin Std. Err. [95% Conf. Interval] Female owner#Loan Female owner#Loan documents documents 00 5.59 0.24 5.05 6.12 00 1.75 0.21 1.29 2.2 01 8.16 0.27 7.58 8.74 01 4 0.15 3.66 4.33 10 5.81 0.41 4.91 6.71 10 1.9 0.19 1.48 2.33 11 8.08 0.26 7.5 8.65 11 4.12 0.16 3.76 4.48 Feamle owner#Medium Feamle owner#Medium firm firm 00 6.99 0.2 6.55 7.42 00 3.14 0.15 2.81 3.46 01 8.26 0.39 7.4 9.13 01 3.32 0.35 2.55 4.1 10 7.06 0.28 6.45 7.67 10 3.25 0.19 2.83 3.67 11 8.1 0.29 7.46 8.73 11 3.56 0.19 3.14 3.98 Female owner#Permit Female owner#Permit 00 4.65 1.17 2.08 7.21 00 1.18 0.38 0.35 2.01 01 7.38 0.17 7 7.76 01 3.3 0.14 2.98 3.62 10 6.91 0.39 6.04 7.77 10 3.57 0.39 2.71 4.43 11 7.27 0.24 6.73 7.8 11 3.29 0.14 2.99 3.59 Female owner#Trade Female owner#Trade 00 6.94 0.3 6.28 7.6 00 3.09 0.17 2.72 3.47 01 7.95 0.7 6.41 9.48 01 3.38 0.43 2.44 4.32 10 7.01 0.24 6.48 7.55 10 3.21 0.17 2.84 3.58 11 7.87 0.29 7.22 8.51 11 3.57 0.25 3.03 4.11 Outcome: Barriers to Growth We study the most significant barriers to business growth: family responsibilities, gender, and conflict, to understand which firm and employee characteristics play a role in reporting these as barriers. In these specifications, the interpretation of the coefficient is the increase or decrease in the likelihood that the observation falls into a higher or lower category. In Table 7b, conflict is considered as a barrier to growth and the model includes interaction terms for gender and conflict to understand whether the effect of operating in a conflict area differs depending on the gender of the business owner. Table 7c presents the results for the predicted values of this interaction. Women owned firms are significantly less likely to report conflict as being a significant barrier to growth. A female- owned firm is 41 percent less likely to perceive conflict as an important barrier to growth. There are several potential narratives to help explain this phenomenon, including the possibility that female-owned businesses may not be perceived as a threat during conflict and consequently suffer less disruption, or otherwise the perception that female-owned businesses are less likely to be in insecure areas or in a position to make payments to conflict actors or suffer payment insecurity. Firms with loan documents are almost 30 percent more likely to perceive conflict as a greater barrier. This could be the result of more stable firms being capital intensive and having characteristics that could cause a larger loss in case of conflict. The larger the share of highly educated employees, the more conflict is perceived as a barrier (although with a minor impact of 3 percent) potentially because firms with more educated employees could be more complex and developed and have more to lose in case of conflict. Finally, firms operating in a conflict zone increases the perception of conflict as a barrier by 87 percent. 29 Provision of informal HRM practices is determinant in the perception of family responsibilities as a barrier to business growth for women owners. The provision of sick and paid leave reduces the perception of family responsibilities being a challenge for women-owned enterprises, suggesting these types of leave entitlements assist with navigating family responsibilities. The provision of housing increases the perception of family responsibilities as a more significant barrier for women-owned businesses by ~30 percent, suggesting that care of employees may add to existing household responsibilities. Provision of housing ranges from 50 percent in light manufacturing to 87 percent in tourism. Given the sample was not randomly selected, the number of women operating in each sector is the same. Other significant factors in determining perception of gender and family responsibilities as major barriers vary by gender. Hours looking after dependents and carrying out household chores are positively related to reporting family responsibilities as a bigger barrier to enterprise growth. This is especially true for women-owned businesses. In the case of male-owned firms, only hours spent carrying out chores is significant, but in each instance the magnitude of the effect is not very large. For women, each hour spent looking after dependents or carrying out household chores increases the perception of family responsibilities as a barrier by 3 percent, compared with men where each hour increases the perception by 8 percent. For women, each hour spent doing household chores increases the perception of gender as a barrier to their business growth by 2 percent. These findings are consistent with the Myanmar context where family responsibilities largely fall to women. One of the reasons why there is a small positive correlation could be partly explained by the fact that the sample is of women owners who may have found ways to alleviate these barriers. For men, a high level of education increases the likelihood of perceiving family responsibilities as a barrier. When analyzing family related challenges, men and women report different firm and employee-level characteristics affecting their perception. The provision of informal HRM policies only affects women-owners, while the level of education increases this only for male entrepreneurs. While gender is not reported as a barrier by a large share of women, it is clear that women recognize this as a more significant barrier, doing more hours of household chores. We have seen how firms operating in conflict areas perceive conflict as a bigger barrier than firms in non-conflict zones. We proceed to evaluate whether gender of ownership plays a role in this context as well. Given the outcome of interest is a categorical variable, the interpretation of the predicted values is not as straightforward as in the previous cases, for each value of the outcome (0-5), we have four potential interactions. Table 7c provides further evidence and supports the main finding that male entrepreneurs operating in conflict areas systematically report conflict as a more disruptive barrier than women owners. As such, 71 percent of women operating a business in a conflict area report conflict not being a barrier, compared to 56 percent of men. On the other extreme, 3 percent of women in conflict areas report conflict as being a very significant barrier to growth, compared to 7 percent of men. 30 Table 7a.Barriers to Growth Table 7b.Barriers to Growth Ordered Probit Regression Oprobit Barrier to growth: Barrier to growth: Barrier to growth: Barrier to growth: Barrier to growth: Family Family Gender (female Conflict (whole Conflict (whole responsibilities responsibilities owners) sample) sample) (female owners) (male owners) coef/se coef/se coef/se coef/se coef/se Female owner N/A -0.414** Female owner -0.348** (0.196) (0.156) High education 0.127 0.944*** -0.153 0.168 High education 0.161 (0.137) (0.340) (0.146) (0.129) (0.125) Small firm (6-10) -0.426 -0.891 0.013 -0.059 Small firm (6-10) -0.071 (0.362) (0.645) (0.253) (0.160) (0.162) Medium firm (>10) -0.241 -0.460 -0.051 -0.108 Medium firm (>10) -0.104 (0.248) (0.420) (0.157) (0.315) (0.318) % of female employees 0.006*** -0.003 -0.004* 0.002 % of female employees 0.002 (0.002) (0.007) (0.002) (0.002) (0.002) % employees high education 0.002 -0.011 -0.000 0.003* % employees high education 0.003* (0.002) (0.011) (0.002) (0.002) (0.002) Loan documents 0.115 0.408 -0.116 0.285** Firm has a license/permit 0.279* (0.147) (0.376) (0.201) (0.145) (0.143) Firm has a license/permit 0.182 6.143*** 0.182 0.663 Age of firm 0.666 (0.283) (0.997) (0.345) (0.452) (0.445) Age of firm -0.002 -0.049 0.001 0.006 Conflict area 0.006 (0.005) (0.033) (0.005) (0.008) (0.008) Conflict area 0.188 -0.854 0.021 0.869*** Conflict area#Female owner -0.199 (0.124) (0.612) (0.166) (0.178) (0.380) Hours dependents 0.029*** 0.036 0.004 -0.003 Hours dependents -0.003 (0.007) (0.029) (0.006) (0.006) (0.006) Hours hh chores 0.032*** 0.079** 0.023*** 0.012 Hours hh chores 0.012 (0.007) (0.032) (0.006) (0.009) (0.009) Maternity/paternity leave 0.083 -0.416 0.058 0.417* Maternity/paternity leave 0.418* (0.156) (0.525) (0.141) (0.215) (0.216) Paid/sick leave -0.669*** -0.230 -0.034 0.200 Paid/sick leave 0.212 (0.140) (0.559) (0.285) (0.249) (0.261) Subsidized housing 0.296** 0.558 0.088 -0.063 Subsidized housing -0.063 (0.150) (0.516) (0.151) (0.251) (0.251) Light manufacturing 0.179 -0.891 0.242 -0.129 Light manufacturing -0.135 (0.242) (0.679) (0.222) (0.184) (0.184) Tourism 0.287 0.091 -0.087 -0.028 Tourism -0.035 (0.270) (0.235) (0.216) (0.107) (0.105) Trade -0.166 -0.214 0.148 0.269 Trade 0.262 (0.229) (0.490) (0.180) (0.186) (0.188) N 379 134 379 513 N 513 note: .01 - ***; .05 - **; .1 - *; note: .01 - ***; .05 - **; .1 - *; Note: Clustered standard errors by township for each specification.The variables: female owner, high education, small and medium firm, whether the firm has a permit to operate, conflict area has documents to request loan and light manufacturing, tourism and trade are all dummy variables. % employees high education - % of employees who have completed at least high school Hours dependents and hh chores - number of hours spent looking after dependents and doing hh chores - given these two variables have more than 10% of values missing, we recoded missing to 0 and included a missingness dummy as an additional covariate in the analysis. 31 Table 7c - Conflict margins Margin Std. Err. [95% Conf. Interval] Female owner#Conflict area 100 0.83 0.03 0.76 0.9 101 0.56 0.1 0.37 0.76 110 0.89 0.03 0.84 0.94 111 0.71 0.02 0.67 0.76 200 0.08 0.01 0.06 0.1 201 0.13 0.02 0.09 0.17 210 0.06 0.01 0.04 0.08 211 0.11 0.02 0.08 0.15 300 0.03 0.01 0.01 0.06 301 0.08 0.02 0.03 0.12 310 0.02 0.01 0.01 0.04 311 0.06 0.01 0.04 0.08 400 0.02 0.01 0 0.03 401 0.05 0.01 0.03 0.07 410 0.01 0.01 0 0.02 411 0.03 0.01 0.01 0.05 500 0.03 0.01 0.01 0.04 501 0.11 0.04 0.02 0.19 510 0.01 0.01 0 0.02 511 0.06 0.01 0.03 0.08 600 0.01 0.01 0 0.02 601 0.07 0.04 -0.01 0.16 610 0 0 0 0.01 611 0.03 0.01 0 0.05 Note: In this case we are calculating the margins after running an ordered probit regression, therefore we are getting the margins' coefficients for each possible outcome of the dependent variable as well. In this case the dependent variable measures from 0- 5 (6 categories) how big of a barrier is conflict for the respondent. Outcome: Adoption of Informal HRM Policies We test the impact of firm and employee level characteristics on the adoption of five informal HRM practices: gender equality policy, maternity/paternity leave, subsidized childcare, paid annual/sick leave, and housing. The results are in Table 8. While some characteristics significantly affect the adoption of small businesses to say that they had more than one policy, the probability of having each policy in place is mostly affected by different factors. It is important to note that maternity and paternity leaves are legislated, while other informal HRM practices are not. Operating in the tourism sector positively affects the outcomes of interest, increasing the probability of a business responding positively to having any HRM practice in place, subsidized childcare, and housing by 72, 84, and 62 percent respectively. Compared to other sectors, tourism in Myanmar is well known for having ‘more published government policies relating to responsible business and sustainability than any other sector’ (Government Legal and Regulatory Framework). In addition, operators may come more under scrutiny from customers and have more exposure to international standards than other sectors in this paper. On the other hand, there is a reduced probability of a firm having maternity/paternity leave, by almost 50 percent and housing subsidy by 75 percent, for those firms operating in light manufacturing and trade. Indeed, housing is directly related to industry, offered by 70 32 percent of non-farming agriculture firms, 86 percent in hospitality and tourism, 46 percent in light manufacturing and 56 percent in trade and logistics. Size of firm plays an important role in determining the adoption of informal HRM practices. This is consistent with the literature (Ozturkler and Ozutku, 2009; Budhwar, 2000), which highlights how size is positively related to the adoption of HRM practices. It seems that what is overall true for formal HRM practices also hold true in our context on informality as well; the larger the firm, the more likely it is to have these practices in place. A medium-sized firm (>10 employees) increases the probability of the firm providing housing subsidy by 72 percent, while at the same time, a small-sized firm (6-10 employees), reduces the likelihood of the firm offering maternity/paternity leave and paid annual/sick leave by almost 30 percent in both cases. Having documents to apply for a loan and having a permit/license to operate increase the chance of having a gender equality policy and subsidized childcare. This finding is also supported by the literature which identifies the formalization of business as being conducive to the adoption of HRM practices. Having permits to operate increases the probability of the firm offering childcare by 58 percent, while having documents to request a loan increases the probability of a firm having a gender equality policy by roughly 30 percent. As these measures can be seen as proxies for financial and legal stability, this indicates more stable firms are more likely to report having family-oriented policies. A positive relationship between education and the provision of informal HRM practices can also be seen. An owner who has completed high school or more is 30 percent more likely to have a gender equality policy in place. When looking at gender related explanatory variables, such as female ownership and share of female employees, results are mixed. Female ownership reduces the probability of a firm providing subsidized childcare by 23 percent, while the higher the share of female employees, and the larger the number of hours spent carrying out household chores, both increase the probability of having childcare by 0.4 percent, for each percent increase in the number of female employees, and 1.8 percent per hour spent doing chores. The finding relating the higher share of female employees with policy adoption is supported by Poelmans, Chincilla and Cardona (2003) who show that a higher share of women employees is positively associated with the implementation of a work/family program. Counterintuitively, some factors conducive to the adoption of formal HRM practices also increase the adoption likelihood of informal HRM practices. Size and formality – being two of the main drivers of the adoption of family- friendly policies – also influence the presence of informal policies. In our case, sector of operation also matters, but this could potentially be another channel through which formality effects the existence of these policies. Factors that are, instead, determinant for gender and family related policies, such as owner’s gender and share of female employees, seem to be less relevant in our context as they present mixed results. This could be explained by the fact that women entrepreneurs do not perceive their gender as a significant barrier and as a result, may be less likely to adopt policies to benefit the business and alleviate the burdens related to household responsibilities. 33 Table 8. Adoption of HRM policies Probit regression analysis Maternity Paid annual Gender equality Subsidized and/or paternity and/or sick Housing subsidy policy childcare leave leave coef/se coef/se coef/se coef/se coef/se Female owner 0.114 -0.138 -0.231** 0.001 0.257 (0.237) (0.163) (0.103) (0.160) (0.183) High education 0.303** 0.080 0.076 0.280 0.059 (0.149) (0.117) (0.161) (0.196) (0.169) Small firm (6-10) 0.052 -0.290* -0.012 -0.273** -0.032 (0.165) (0.162) (0.208) (0.118) (0.181) Medium firm (>10) 0.036 0.122 -0.249 0.044 0.723*** (0.215) (0.188) (0.247) (0.194) (0.150) % of female employees 0.003 -0.000 0.004* -0.003 -0.003 (0.002) (0.001) (0.002) (0.003) (0.003) % employees high education 0.001 -0.001 -0.004 0.001 -0.005** (0.002) (0.002) (0.003) (0.002) (0.002) Firm has documents to request loan 0.293* 0.265 0.014 0.044 0.168 (0.150) (0.199) (0.178) (0.108) (0.111) Firm has a license/permit -0.000 -0.285 0.581* -0.331 0.219 (0.356) (0.276) (0.309) (0.372) (0.294) Age of firm 0.001 -0.001 -0.009 0.004 -0.011** (0.004) (0.006) (0.007) (0.007) (0.005) Conflict area -0.141 0.089 -0.413 0.375 0.044 (0.218) (0.276) (0.458) (0.273) (0.125) Hours dependents 0.009 0.001 0.004 0.013 -0.003 (0.008) (0.009) (0.010) (0.009) (0.008) Hours hh chores -0.005 0.003 0.018*** -0.013 0.001 (0.011) (0.006) (0.007) (0.009) (0.011) Light manufacturing -0.008 -0.468*** 0.111 -0.264 -0.750*** (0.187) (0.141) (0.140) (0.161) (0.170) Tourism 0.724*** -0.090 0.210 0.839*** 0.624** (0.221) (0.195) (0.184) (0.274) (0.251) Trade 0.224 -0.330* -0.098 0.285 -0.280 (0.220) (0.168) (0.155) (0.259) (0.220) _cons -0.266 0.402 -1.352*** 1.225*** 0.566 (0.312) (0.315) (0.494) (0.388) (0.394) N 441 441 441 441 441 note: .01 - ***; .05 - **; .1 - *; Note: Clustered standard errors by township for each specification.The sample size was reduced to 441 firms as those with only 1 employee were not considered for this analysis. The variables: female owner, high education, small and medium firm, whether the firm has a permit to operate, conflict area has documents to request loan and light manufacturing, tourisma and trade are all dummy variables. % employees high education - % of employees who have completed at least high school Hours dependents and hh chores - number of hours spent looking after dependents and doing hh chores - given these two variables have more than 10% of values missing, we recoded missing to 0 and included a missingness dummy as an additional covariate in the analysis. Gender equality index: gender equality, maternity and paternity leave and subsidized childcare. Social policies index: all the gender equality policies plus sick and annual leave. Subsidy policies index: house, food and transport subsidy. Gender equality - 308 yes Maternity and/or paternity leave - 195 yes Subsidized childcare - 92 yes Paid annual and/or sick leave - 392 yes Subsidized housing - 315 yes 34 Recommendations This paper revealed a number of barriers and challenges that relate to women entrepreneurs, as well as issues that affect both genders when operating a firm. The most significant challenges for female business owners arise from their traditional household and family responsibility role in Myanmar society, as well as their disadvantaged position when seeking access to finance. The recommendations below are aimed at alleviating these challenges. • Improving access to finance Improve access to finance for women by sharing findings of this paper with banks, MFIs and MMFA to raise awareness of gender gaps and provide support to create new gender sensitive financial products for MSMEs. Despite MFIs being the most commonly reported lending institution used by women in Myanmar, these organizations lack financial products tailored specifically to women-owned enterprises. It is thus crucial that Myanmar’s financial institutions are supported and guided, by creating financial products tailored to the needs of women. Further, these products would help to reduce the reliance on women using personal savings or informal lending sources to support their business, a significantly more pressing issue for women than men. • Raising awareness Increasing understanding of the impact that gender gaps have on private sector development is crucial. As reported, women entrepreneurs do not recognize their own gender as being a barrier to business growth despite evidence to the contrary, which highlights underlying issues such as limitations in access to finance and a far greater share of household responsibilities as gender-specific barriers. To increase awareness of these gender-barriers, culturally-sensitive messaging needs to be delivered both at the institutional and societal levels, which this paper proposes to do through the following proposed avenues: Editorial – As a qualitative exercise, collecting case studies from women entrepreneurs who have been successful in running a business and discussing the challenges they have faced along the way. An editorial would provide an avenue for awareness-raising on the various barriers faced by women entrepreneurs in a more accessible and tangible way, and potentially generate public discussion on the problem and potential solutions. Share findings with business associations, organizations, and the government – Engage with chambers of commerce, entrepreneur associations, government stakeholders, and other relevant organizations, in a dialogue on the challenges and potential solutions to the barriers faced by women entrepreneurs. Moreover, at the government level, reach out to the gender working group (part of the Department of Social Welfare), to share findings, and discuss potential areas for collaboration. 35 Conclusion Many factors contribute to the pervasive structural problems in Myanmar, with social and cultural norms being deeply embedded in institutional and workplace arrangements. Given this context, the policy challenge must center on how to bring about societal change, thereby expanding economic opportunities in which women can engage, particularly as leaders, entrepreneurs, and employees. While the constraints faced by women entrepreneurs may not be immediately apparent, certain gender differences come into play when the societal role of women is considered. Women tend to face more barriers and challenges than men in several different aspects, as seen from the analysis conducted. The major barrier to the emergence and growth of women entrepreneurs in Myanmar is the cultural, gender-biased traditions regarding the role of women in society. These include the time burden of household and family responsibilities and the perception of their gender being a challenge for business development. Household chores and family responsibilities disproportionately fall upon women’s shoulders, and, as such, they are perceived as a much bigger barrier to business growth than for men. When examining barriers to growth, there is a clear relationship between the number of hours spent engaged in household chores and caring for family members, and the perception of these activities as significant obstacles to business development. The societal role of women also translates into a general absence of respect for women as entrepreneurs and business leaders, and a lack of a welcoming environment for women entrepreneurs in male-dominated sectors. One example of the latter that stands out from the analysis is the reduced access to finance faced by women business owners. Banks in general are not reaching out to the women-owned enterprise market and women are disadvantaged in being able to meet collateral requests. Overall, women tend to face more barriers and challenges in terms of their business size, which constrains them from accessing finance—a key prerequisite for business growth—thus perpetuating the problem. This study highlights the importance of targeted interventions for women owned MSMEs to move beyond micro businesses and to increase the overall participation of women in the economy. This paper underlines the clear need for legal and regulatory reform, but more emphatically, for awareness-raising and effective leadership to ensure gender equality is understood and appreciated, and for incentives to encourage non-discriminatory practices to enter the mainstream. An example of a key awareness-building outcome could be enhancing an understanding of the wider economic benefit that increased gender participation has on business and trade generally, potentially benefiting male-owned businesses. Separately, an improved and more family-friendly business environment could be partly achieved by a wider adoption and use of HRM policies. For instance, we observed how the ability to take paid sick and annual leave reduces the perception of family responsibilities as a barrier to business growth. Size seems to be the most conducive firm characteristic to the adoption of the policies and overall good business practices. This highlights the importance of providing women entrepreneurs with the appropriate instruments, access to formal finance being one of the most important ones, to be able to expand their businesses. 36 Overall, in Myanmar, women tend to face disproportionately more barriers to operating and growing their business (with the exception of conflict, which affects men entrepreneurs more), given the small size of their firms, additional household responsibilities, reduced access to finance and lack of family-friendly work environments. To overcome these challenges and foster business development, it is thus important that awareness is raised and that policies, products and services that encourage better business environments and development are put in place. References Asian Development Bank. 2016. Gender equality and women’s rights in Myanmar: a situation analysis. ADB. Bardoel, E. Anne et al. 2013. Organizational predictors of work-family practices. Asia Pacific Journal of Human Resources (36): 31-49. Berkel, H. et al. 2018. Myanmar Micro, Small, and Medium Enterprise Survey 2017: Descriptive Report. Helsinki: UNU-WIDER, University of Copenhagen, Central Statistical Organization. Bin Bae, Kwang and Yang, Gigeun. 2013. The effects of family-friendly policies on job satisfaction and organizational commitment: a panel study conducted on South Korea’s public institutions. Public Personnel Management (46): 25-40. Bloom, Nick et al. 2011. Are family‐friendly workplace practices a valuable firm resource? Strategic Management Journal (32): 343-367. Boeckmann, Irene et al. 2010. Work–Family policy trade-offs for mothers? Unpacking the cross-national variation in motherhood earnings penalties. Work and Occupations (43): 119-177. Bollinger, Kate and Seaman, Laura. 2013. Women biz owners in Indonesia defy some long-entrenched barriers. The Asia Foundation. Born, Nicholas and Kang, Seoung Kil. 2015. What are best practices in the space of employer branding that enable organizations to attract and retain the best talent? Cornell University ILR Review. Budwhar, Pawan S. 2000. Determinants of HRM policies and practices in India: an empirical study. Global Business Review (1): 229-247. Crouch, Melissa. 2016. Promiscuity, polygyny and the power of revenge: the past and future of Burmese Buddhist law in Myanmar. Asian Journal of Law and Society (3): 85-104. De Kok, Jan and Uhlaner, Lorraine. 2001. Organization context and human resource management in small firm. Small Business Economics (17): 273-291. De la Briere, Benedicte and Wodon, Quentin. 2018. Unrealized potential: the high cost of gender inequality in earnings. World Bank. Dutch Goof Growth Fund. 2019. Country report Myanmar, entrepreneurial ecosystem assessment. DGGF. Egreteau, Renaud and Mangan, Cormac. 2018. State fragility in Myanmar: Fostering development in the face of protracted conflict. IGC. Equal Measures 2030. 2019. Harnessing the power of data for gender equality. Equal Measures 2030. Ferragina, Emanuele. 2017. Does family policy influence women’s employment? Reviewing the evidence in the field. Political Studies Review (17): 65-80. Gender Equality Network. 2015. Raising the curtain: cultural norms, social practices and gender equality in Myanmar. LIFT. 37 International Labor Organization and Sasakawa Peace Foundation. 2020. National assessment of women’s entrepreneurship development framework conditions – Myanmar. ILO. International Finance Corporation. 2016. She works, recruitment and retention of female talent in the workplace. IFC. International Finance Corporation. 2019. Tackling childcare: the business case for employer-supported childcare in Vietnam. IFC. International Finance Corporation and Myanmar Survey Research. 2019. Myanmar Gender Enterprise Survey Report. MSR. Jayasekaran, Subajini et al. 2019. Business and family friendly policies. An evidence brief. UNICEF. Jolliffe, Kim. 2015. Ethnic armed conflict and territorial administration in Myanmar. The Asia Foundation. Jones, Derek C. et al. 2006. How does employee involvement stack up? The effects of human resource management policies on performance in a retail firm. Cornell University. Kapteyn, Arie and Wah, Saw Htay. 2016. Challenges to small and medium-size businesses in Myanmar: What are they and how do we know? Journal of Asian Economics (47): 1-22. Lopez, O. Anne and Rahardja, Sjamsu. 2017. Women entrepreneurs in Myanmar: what does the Enterprise Survey Data tell us? Working Paper, World Bank. Michel, M. Arthur et al. 2004. Taking stock of work-family Initiatives: How announcements of “family-friendly” human resource decisions affect shareholder value. Cornell University ILR Review (57): 599-613. Michelitsch, Roland. 2012. Private sector matters for job creation. International Finance Corporation. IFC Jobs Study Newsletter. Myanmar Annual Labour Force Survey Report. 2017. ILO. Naz, Farah et al. 2016. Financial performance of firms: Evidence from Pakistan cement industry. Journal of Teaching and Education (5): 81-94. Organization for Economic Cooperation and Development. 2004. Promoting SMEs for Development: The Enabling Environment and Trade and Investment Capacity Building. OECD. Organization for Economic Cooperation and Development. 2013. Multi-dimensional review of Myanmar. Initial assessment, OECD Development Pathways. OECD publishing. Ozutku, Hatice and Ozturkler, Harun. 2009. The determinants of human resource practices: an empirical investigation in the turkish manufacturing industry. Ege Academic Review (9): 73-93. Panda, Swati. 2018. Constraints faced by women entrepreneurs in developing countries: review and ranking. Gender in Management: An International Journal. Poelmans, Steven et al. 2003. The adoption of family-friendly HRM policies: competing for scarce resources in the labour market. International Journal of Manpower (24): 128-147. Sels, Luc et al. 2006. Unravelling the HRM–performance link: value‐creating and cost‐increasing effects of small business HRM. Journal of Management Studies (43): 319-342. Shafeek, Hani. 2016. The impact of human resources management practices in SMEs. Annals of Faculty Engineering Hunedoara – International Journal of Engineering (14): 91-102. Tambunan, Tulus T.H. 2017. The performance of Indonesia’s public credit guarantee scheme for MSMEs. Journal of Southeast Asian Economies (35): 319-332. The World Bank Group’s World Development Indicators. 2019. World Bank. 38 United Nations Capital Development Fund. 2017. PoWER Women and girls’ financial inclusion country assessment, Myanmar. Dalberg Global Development Advisors. United Nations Conference on Trade and Development. 2001. Improving the competitiveness of SMEs in developing countries. The role of finance to enhance enterprise development. United Nations. World Bank. 2020. Women’s agency in Mon and Kayin States. World Bank. Yordanova, Desislava Ivanova. 2011. The effects of gender on entrepreneurship in Bulgaria: an empirical study. International Journal of Management (28): 289-305. Appendix A – Descriptive statistics table Table 3b. Weighted average of barrier perception (whole sample) Family responsibility Gender Conflict Male weighted average 0.19 0.08 0.31 Female weighted average 0.44 0.29 0.38 Note:The question asked to report out 5 (0 being no barrier at all, 5 being biggest barrier) how m.uch these affected their business growth. In the sample: 24.5% of women and 10.5% of men reported family responsibilities as a barrier 18.2% of women and 4.5% of men reported gender as a barrier 10% of women and 13.5% of men reported conflict as a barrier Appendix B – Variable description HRM policies HRM policies: gender equality policy, maternity and/or paternity leave, subsidized childcare, paid annual and/or sick leave, and housing subsidy are all binary variables that take value 1 if the firm provides each or the combination of the policies, as in the case of maternity and/or paternity leave. Gender equality policies index: this index is out of 4 points, it was created with the summative approach (as all indices in this paper) and it is composed by gender equality, maternity and paternity leave, and subsidized childcare. Social policies index: this index is out of 6 points and it is composed by all variables comprising the gender equality index, plus whether the firm provides sick and/or annual leave. Subsidy policy index: this index is out of 3 points and it is composed by the variables asking whether the firm provides housing, food and transport subsidy. Access to finance Firm has bank account and documents to request loan: binary variables taking value 1 if the firm has a bank account and if they have the necessary documents in place to request a loan. 39 Good business practices Business development index: this index is out of 12 possible points and it comprises the following business development/management practices: whether in the past 3 months the owner has visited competitors to see what product they were offering and at what price, asked customers for product suggestion, had a special offer to attract customers, did any form of advertising, negotiate with supplier for lower price, and whether they have written/computerized record-keeping system to check inventories, record purchases and sales made by enterprise, have records to keep track of sales, know cost of each product, know which product makes the most profit, and have a written budget for enterprise expenses. Financial planning index: this is out of 6 points and it is composed of the following binary variables: whether the business has documents to apply for a bank loan, whether they have a sale target, and whether they prepare at least annually: profit and loss statement, statement of cash flow, balance sheet, income and expenditure sheet. Barriers to growth Family responsibilities, household work, gender and conflict: These are all ordered categorical variables ranging from 0 to 5, where 0 is no barrier and 5 is biggest barrier. Businesses were asked to rate how each of these affected their business growth. Covariates Female owner – Binary variable taking value 1 if the owner of the firm is a woman. High education – Binary variable taking value 1 if the owner has completed high school education or higher. Small and medium firm – Binary variables taking value 1 if the firm has between 6-10 employees (small) and more than 10 employees (medium). Percent of female employees – Share of female employees in the firm. Percent of employees’ high education – Share of employees who have completed at least high school or more. Firm has a license/permit to operate – Binary variable taking value 1 if the business has a license. This is used as a proxy in the analysis for business stability/formality. Age of firm – Number of years the firm has existed. Conflict area – Binary variable taking value 1 if the firm operates in a conflict township. Hours dependents and household chores – Reports the average number of hours spent in a week looking after dependents: children, elderly, spouse etc. and the average number of hours spent doing household work. Has documents to request loan – Binary variable taking value 1 if the business has the appropriate documents in place to request a loan. This is used as a proxy in the analysis for financial stability of the firm. Light manufacturing, tourism and trade – Binary variables reporting on the sector of operation of the firm. In the access to finance, business development and barriers to growth tables we use HRM policies as covariates as well. 40