The Gender Gap in Entrepreneurship in Romania Background Study for the Romania Gender Assessment Monica Robayo-Abril and Britta Rude The World Bank © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.orgpubrights@worldbank.org. Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank, therefore, does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. 2 Summary Although female entrepreneurship is crucial to generating sustainable and equitable growth patterns, international evidence shows that women tend to be underrepresented in entrepreneurship, and this gender gap has exhibited remarkable persistence. In this study, we first measure the gender gap in entrepreneurship in Romania by using various data sources. We observe significant gender gaps, with the average gender gap in self-employment rates being 4.2 percentage points when abstracting from observable characteristics. Even when controlling for observable characteristics, the gender gap is persistent (3.7 percentage points). Other measures, such as the share of firms with female owners and top managers, indicate that the gap could be even larger. Moreover, we observe that the entrepreneurial gender gap varies across income quintiles and between rural and urban areas. In the second step, we analyze the potential drivers of women’s engaging less in entrepreneurship by following the model of the “5 M’s” developed by Brush, De Bruin, and Welter (2009). We find that the following drivers play a role in the entrepreneurial gender gap in Romania: gender gaps in financial inclusion and access to assets, harmful gender norms, motherhood, lack of childcare, and eldercare. Our findings suggest the need for a nuanced approach toward female entrepreneurship that factors in the distinct challenges of different groups of women and consists of a menu of policy interventions. Policies should range from improving women’s access to relevant assets, human capital, and networks to addressing harmful gender norms and sparking an entrepreneurial culture in Romania more generally. Lastly, our evidence indicates that women are more interested in “impact” entrepreneurship. As women entrepreneurs in Romania mainly operate in the primary sector, giving them a leading role in the green transition has great potential for more sustainable and equitable growth patterns. 3 Table of Contents Summary ................................................................................................................................................. 3 Table of Contents .................................................................................................................................... 4 Figures ..................................................................................................................................................... 5 Tables ...................................................................................................................................................... 5 1. Introduction ..................................................................................................................................... 6 2. Measuring the Gender Gap in Entrepreneurship and Characterizing Female Entrepreneurs in Romania................................................................................................................................................... 9 3. Potential Drivers ............................................................................................................................ 19 Legal, Institutional, and Policy Frameworks ...................................................................................... 20 Access to finance ............................................................................................................................... 21 Access to assets ................................................................................................................................. 23 Institutional and cultural factors ....................................................................................................... 24 Motherhood ...................................................................................................................................... 25 Eldercare............................................................................................................................................ 31 4. Conclusion and Policy Recommendations..................................................................................... 32 References ............................................................................................................................................. 36 Appendix................................................................................................................................................ 42 4 Figures Figure 1. The gender gap in entrepreneurship in Romania measured as the share of working women and men who are self-employed, 2020) ............................................................................................... 11 Figure 2. Self-employment rates by gender and income quintiles, 2020 ............................................. 11 Figure 3. Self-employment rates by gender and rural and urban areas, 2020 ..................................... 12 Figure 4. The gender gap in entrepreneurship in Romania measured as the proportion of firms with female ownership and top female managers ....................................................................................... 12 Figure 5. Female share of all self-employed by income quintiles, 2020 ............................................... 13 Figure 6. Share in self-employed by gender and age groups, 2020 ...................................................... 13 Figure 7. Self-employment rates by gender and rural and urban areas, 2020 ..................................... 14 Figure 8. Sectoral distribution of self-employed women, 2020 ............................................................ 14 Figure 9. Percentage of female owners by company size, 2018–20 ..................................................... 16 Figure 10. 5 Ms framework of Brush, De Bruin, and Welter (2009)...................................................... 19 Figure 11. Account ownership rate by gender, 2011–21 ...................................................................... 22 Figure 12. Access of capital to start, expand, and operate a business by gender, 2014 vs. 2017 ........ 22 Figure 13. Distribution of responses to question about whether men make better business executives than women by gender, 2017–22 ......................................................................................................... 25 Figure 14. Distribution of responses to question about whether preschool children suffer because the mother is working by gender, 2017–22 ................................................................................................ 25 Figure 15 Self-employment rates for households with or without children below 16 years old by gender, 2020 ......................................................................................................................................... 26 Figure 16. Self-employment rates for working women from households with or without children below 16 years by income quintiles, 2020 ............................................................................................ 26 Figure 17. Self-employment rates for households with and without children below 16 years old by gender and rural and urban areas ......................................................................................................... 27 Figure 18. Forms of childcare used by self-employed people from households with children below six years old by rural and urban areas ........................................................................................................ 27 Figure 19. Access to childcare by the self-employed from households with at least one child below six years old by income quintile, 2020 ....................................................................................................... 28 Tables Table 1. Summary statistics of entrepreneurs by gender, 2020 ........................................................... 15 Table 2. The impact of gender on the probability of being self-employed (marginal effects) ............. 19 Table 3. Probit model on the probability of women in Romania being self-employed, 2020 .............. 30 Table 4. The impact of eldercare on the probability of being self-employed among women, 2020 .... 31 Table A.1. Probability of working women in households with children being self-employed, 2020 .... 42 Table A.2. Probability of working women in households with children being self-employed (with self- employed restricted to employers with employees), 2020 .................................................................. 43 Table A.3. Probability of working mothers in households with children being self-employed, 2020 .. 44 5 1. Introduction Female entrepreneurship is crucial for generating sustainable and equitable growth patterns. Entrepreneurship in general and social entrepreneurship in particular have recently been taken up as topics in research, because they are increasingly important sources of employment for women across many countries (Langowitz and Minniti 2007). Female entrepreneurship is crucial to generating sustainable and equitable growth patterns: first, female entrepreneurs contribute significantly to economic growth by creating jobs and wealth for women, who suffer from an economic gender gap in many countries, and contribute to innovation (Low, Henderson, and Weiler 2005). Second, in some countries women are more interested in social entrepreneurship and “making a difference” than men (GEM 2022b), which can create positive social and environmental externalities. Finally, female entrepreneurs can serve as role models for other women and girls, showing them that they can succeed in business and achieve their goals (Bechthold and Huber 2018). Overall, promoting female entrepreneurship is important for creating a more inclusive, diverse, and prosperous development path. The policy discussion regarding self-employment and its benefits for development involves multiple issues. Self-employment is viewed as a way to stimulate innovation and job creation. Policy makers aim to encourage more people to start their own businesses and become self-employed, under the assumption that they desire more autonomy over their work lives. However, some critics argue that self-employment may not be a voluntary decision, but rather a necessity as a way to escape unemployment and economic inactivity and could lead to economic dependence, resulting in a lack of autonomy and limited social protection. The long-term financial sustainability of self-employment is also questioned, as individuals in this sector tend to be overrepresented in both low- and high-income brackets. To address these concerns, the European Pillar of Social Rights proposed by the European Commission takes up adequate social protection for the self-employed (Vermeylen et al. 2017), and the international community has recognized the need to provide social security to self-employed workers (European Parliament 2017; G20 2017; ILO 2021). Despite the potential benefits, the gender gap in entrepreneurship remains large in most economies, both developed and developing. The gender gap in business ownership persists globally, with less than one-third of new limited liability company owners being women in most analyzed economies (Meunier et al. 2017). Women are particularly underrepresented in low-income economies, where they are less likely to start a new business than men. The OECD and the European Commission (2021) find that 75 percent of “missing"” entrepreneurs who face barriers to business creation are women. Moreover, women only own one in three businesses worldwide (World Bank, 2020). International research has highlighted several barriers to female entrepreneurship that impede the fulfillment of this potential and generate a persistent gender gap in entrepreneurship. Women tend to face significant obstacles in relation to access to capital, networking opportunities, and biases in the business world (Adikaram and Razik 2022). The gender gaps in entrepreneurship might be related to the fact that women face additional barriers to entrepreneurship on top of those men face (Brush, De Bruin, and Welter 2009). These barriers are present throughout the entire entrepreneurial process. Women are less likely to start a business in the first place. Additionally, those women who found businesses and seek capital are 63 percent less likely to receive funding from VCs due to investor bias and gender stereotypes (Guzman and Kacperczyk 2019). But the barriers do not stop here. There are many contextual factors influencing the gender gap in entrepreneurship, such as having less access to networks (Caliendo et al. 2015), but also entrepreneurial human capital (Krieger et al. 2022) and the existence of traditional gender norms and institutional factors (Dheer, Li, and Treviño 2019). More 6 recently, there has been increasing interest in the concept of “mompreneurship,” entrepreneurship by mothers, spurred by evidence that it is difficult for women to combine entrepreneurship with motherhood and their family context (Richomme-Huet, Vial, and d’Andria 2013; Ekinsmyth 2014). By identifying these key barriers and promoting female entrepreneurship, governments can implement policies to reduce these gender inequalities and create a more level playing field, inducing more- equitable and inclusive growth patterns. There is no recent updated quantitative evidence on female entrepreneurship in Romania. Previous research on Romania has noted the persistence of stereotypes of women. However, Herman and Szabo (2015), in their analysis of gender (in)equalities in entrepreneurship in Romania, found that the situation has improved since 1990. The number of women-owned businesses has significantly increased during the period they studied, although the share remained low: not even 10 percent of the Top Forbes 500 firms in Romania were led by women in 2013. This study aims to measure gender gaps in entrepreneurship in Romania and characterize female entrepreneurs in Romania using several data sources, including global databases and recently available household surveys; we also investigate potential drivers of these gaps through the lens of the so-called 5 Ms framework. The absence of granular data that separates information on business creation and ownership by gender creates a major barrier to the measuring and analyzing of female entrepreneurship. The available data are inadequate and nonstandardized and as such cannot be compared across countries. Nevertheless, we start with a descriptive evaluation of gender gaps for the overall population and different population groups in Romania and compare Romania’s performance with regard to female entrepreneurship to that of other countries to the extent possible. For cross- country comparisons, we mainly rely on data from global databases, such as the Mastercard Index of Women Entrepreneurs, the Global Entrepreneurship Monitor (GEM), The World Bank’s Global Financial Inclusion Database (Findex) and Women, Business and the Law database, the OECD Gender Data Portal, and the World Bank Enterprise Survey.1 For microlevel empirical analysis, we use the 2020 European Union Statistics on Income and Living Conditions (EU-SILC) to generate a measure of entrepreneurship using self-employment as a proxy.2 After mapping the most critical gender gaps in entrepreneurship in Romania, we describe the average characteristics of female entrepreneurs in Romania. We then turn our attention to potential drivers, using as a lens the framework developed by Brush, De Bruin, and Welter (2009), the so-called 5 M framework. The 5 Ms refer to five main drivers: the macro and meso environment, motherhood, and the three Ms (Market, Management, Money) from the traditional literature on entrepreneurship. By understanding the strengths and weaknesses of their local ecosystem, policy makers can work to improve it and create an environment that is conducive to entrepreneurship and innovation. 1 The Mastercard Index of Women Entrepreneurs (MIWE) uses 12 indicators and 27 sub-indicators to create 3 “Components”: (a) women’s advancement outcomes, (b) knowledge assets and financial access, and (c) entrepreneurial supporting conditions. The Global Entrepreneurship Monitor (GEM), established in 1999, currently involves over 100 research and academic institutions and has more than 400 entrepreneurship experts participating in its execution. The data collection under the GEM is conducted using two complementary survey tools, the Adult Population Survey (APS) and the National Expert Survey (NES). The APS is a nationally representative survey of at least 2000 respondents used to measure the level and nature of entrepreneurial activity around the world. The National Experts Survey (NES), part of the standard GEM methodology, assesses various conditions that enhance (or hinder) new business creation as well as shedding light on some other topics related to entrepreneurship. It is intended to obtain the views of additional experts on, for example, women entrepreneurship support, high growth business encouragement, and questions related to the special topic included in the current GEM cycle. The Global Financial Inclusion Database (Findex) is a World Bank database that measures and tracks access to financial services around the world. 2 We use the EU-SILC because it provides cross-detailed information on childcare with information on self-employment. Moreover, the EU-SILC contains information on the gross monetary income of paid employees and gross monetary income or losses for self-employed persons during the previous 12-month period. In addition, it enables the evaluation of how entrepreneurship varies along the income distribution. 7 The 5 M framework developed by Brush, De Bruin, and Welter (2012), as noted above, identifies five key drivers that can influence entrepreneurship among women and uses empirical indicators to approximate these drivers, drawing data from several sources, such as the Women, Business, and Law Index, the Global Findex, and the Social Institutions and Gender (SIGI) Index, developed by the OECD.3 The first driver is the macro and meso environment, which includes cultural attitudes toward gender roles, legal and regulatory frameworks, and access to infrastructure and resources. The second driver is motherhood, which refers to women’s caregiving responsibilities and experiences shaping their entrepreneurial choices and strategies. For example, women may prioritize businesses that enable them to work from home or have a flexible schedule so as to balance caregiving responsibilities with entrepreneurial activities. The remaining three drivers are the “three Ms” from the traditional literature on entrepreneurship: Market, Management, and Money. The Market driver refers to factors such as market opportunities, competition, and consumer demand, which can influence the viability and success of women’s entrepreneurial ventures. The Management driver refers to the skills, knowledge, and networks that women entrepreneurs need in order to manage and grow their businesses effectively. Finally, the Money driver refers to access to financial resources, such as credit and investment, including venture capital, angel investors, and crowdfunding, which are critical to the growth and sustainability of women’s entrepreneurial ventures. This framework facilitates the identification of potential drivers of women’s entrepreneurial activity, given a specific context. The drivers in turn highlight the importance of understanding the complex and interrelated factors that can influence women’s entrepreneurial activity and the need for targeted policies and programs that address these factors holistically. We put a special focus on the newly emerging concept of “mompreneurship,” a growing area of research in entrepreneurship and gender studies. “Mompreneurship” refers to the practice of mothers’ starting and running their own businesses while balancing family responsibilities. Richomme- Huet and Vial (2014) and Landour (2020) both explore the mompreneurship movement in France, with the latter highlighting the challenges and expectations faced by mompreneurs. Dhaliwal (2022) examines the antecedents and challenges of mompreneurs, offering practical recommendations for those interested in pursuing this path. Ekinsmyth (2014) explores the debate surrounding the concept of “mompreneurship” and its potential to either empower or reinforce gender role expectations. Overall, this evidence suggests that mompreneurship is a growing phenomenon that offers mothers the opportunity to pursue their entrepreneurial goals while balancing family responsibilities. This new evidence is particularly relevant for Romania, a country where the availability of child- and eldercare has been identified as a constraint to female labor force participation (Robayo-Abril and Rude forthcoming). As the number of women entrepreneurs rises, it is essential to develop a more nuanced understanding of the experiences and realities of women who are balancing the demands of motherhood and entrepreneurship. In particular, we investigate the interaction between motherhood, childcare, and entrepreneurship. The report is organized as follows: chapter 2 presents the gender gaps in entrepreneurship in Romania and characterizes female entrepreneurs; chapter 3 presents a picture of potential drivers using the 5 M framework, with a focus on “mompreneurship”; and chapter 4 concludes and presents policy recommendations. 3 The Social Institutions and Gender Index (SIGI) is a composite measure produced by the Organization for Economic Co- operation and Development (OECD) that assesses the discriminatory practices and social institutions that perpetuate gender inequalities in countries around the world. The SIGI covers five dimensions of discrimination against women: discriminatory family code, restricted physical integrity, restricted access to productive and financial resources, limited civil liberties, and son preference. It measures these dimensions using a set of 14 indicators, the data for are derived from a range of sources, including national surveys, censuses, and administrative data sets. 8 2. Measuring the Gender Gap in Entrepreneurship and Characterizing Female Entrepreneurs in Romania Romania does not have a long-lasting culture of entrepreneurship. Romania generally has an entrepreneurial ecosystem consisting of solid early-stage activity rates but a limited rate of achieving established business ownership (EBO) (GEM 2022a). The low EBO rate could be due to the low success rates of early businesses or lower entrepreneurial activity rates in the past (GEM 2022a). The low EBO rate comes along with a culture of people who are not used to entrepreneurial activities. For example, only 37.7 percent of Romanians know somebody who started a business (the 43rd-lowest value out of 47 countries for which data is reported on this indicator as part of the GEM 2022a), and only half of the population feels equipped to start a business (the 33rd-lowest value out of 47) (GEM 2022a). In Romania, the conditions are less conducive to entrepreneurship compared to other European countries. The OECD rates the conditions for entrepreneurship as less favorable than the EU average (OECD and the European Union 2020). In line with this assessment, Romania is categorized as an “Emerging Inventor” on the European Innovation Scoreboard 2022. Its innovation performance is 32.6 percent below the EU average and the average of other Emerging Inventors in the EU. The gap between Romania’s score and the EU average has increased over the last years, because other countries have improved more than Romania (European Commission 2022). Many factors have been identified that impede a flourishing entrepreneurial culture in Romania. Although Romania has many skilled, tech-affine young graduates, there is a general reluctance to take risks and a preference to work as salary workers (European Commission 2017). High levels of corruption and bureaucracy in the public sector are other important barriers that impede entrepreneurial activities, along with high taxes and few subventions, grants, and other forms of government support (European Commission 2017). An underdeveloped angel investor and venture capital (VC) market present further challenges (European Commission 2017). In fact, Romania’s private equity investments as a percentage of GDP are one of the lowest in Europe and significantly below the EU and Central and Eastern European average (Invest Europe 2022). Lastly, the Romanian diaspora and brain drain also play a role (European Commission 2020). Nevertheless, an increasing number of venture capital firms, angel investor groups, and crowdfunding platforms operate in Romania. Given the growing start-up ecosystem in Romania, funding opportunities from both abroad and within Romania are increasing (Fortech Investments 2022). On the latest available “Ease of Doing Business” Indicator from 2019, Romania ranked 55th out of 190 countries and faced a negative trend over time (World Bank 2020). More recently, there have been efforts to shine some light on women in the start-up sector in Romania (Groszkowska 2023). According to global databases, Romania ranks low compared to other countries of similar income levels, with no recent improvements. The country ranked low in the lower half of the Mastercard Index of Women Entrepreneurs (MIWE) 2019 (Mastercard 2023) and had lost ground compared to 2018. The MIWE has measured the progress of women in business across 58 countries since 2017 and in so doing has shed light on the many drivers behind female entrepreneurship, such as access to education, finance, and cultural elements. The 2019 index positioned Romania below countries like Argentina, South Africa, or Peru but above Italy (Mastercard 2023). Within the group of upper-middle- 9 income countries included in the index, Romania ranked among the lowest. Overall, its index score decreased by 3.3 percent when compared to 2018: where it had taken 35th place in 2018, in 2019 it had fallen to 39th place. Romania declined on all subdimensions of the indicator. First, its ranking decreased by four places on the Women’s Advancement Outcomes Score.4 Second, Romania lost ground on the Knowledge Assets & Financial Access Score, dropping eight places from where it had ranked in 2018.5 Lastly, its performance also decreased on the Supporting Entrepreneurial Conditions Score,6 falling six places between 2019 and 2018. Although Romania fares well when compared globally, women still occupy a limited number of leadership positions in the private sector. Only 17 percent of firms had a top female manager in 2019.7 While this is comparable to the EU average (18 percent), it represents a deterioration compared to 2009, when one-quarter of firms had a top female manager. Similarly, only one-third of the senior and middle management were female in 2020.8 While there have been some fluctuations over the years in this indicator, the share has remained stable over time and these numbers show that there is room for improvement. Still, Romania ranked the highest in terms of the share of female business owners (of all business owners) in 2018, the latest available data point across countries:9 in that year 4 out of 10 business owners were women. Romania also ranked high on the share of female directors (of all directors) and took fifth place on this indicator in 2018.10 To have a better understanding of the size of the gender gap and potential driving factors, we use household-level data to measure the entrepreneurship gender gap using self-employment as a proxy and characterize female entrepreneurs in Romania. To investigate the gender gap in entrepreneurship and the current state of female entrepreneurs in Romania, we mainly rely on data from the EU-SILC from 2020. We use self-employment as a proxy for entrepreneurship, given that no explicit question in the EU-SILC enables for a more granular distinction between entrepreneurs and the self-employed. While some argue that there are systematic differences between self-employment and entrepreneurship and that self-employment as a measure has important shortfalls (Gartner and Shane 1995),11 an overwhelming share of the literature has relied on self-employment to analyze 4 This subdimension of the MIWE consists of four indicators: women’s participation i n business leadership, their share in professionals and technical workers, the women entrepreneurial activity rate, and women labor force participation. 5 This subdimension of the MIWE measures women’s access to knowledge and finance via 4 indicators (10 subindicators). Examples are the gender ratio of women borrowing or saving for business, female financial inclusion, support for small and medium enterprises, and the gender ratio of tertiary enrollment. 6 This subdimension of the MIWE analyzes contextual factors that could enable or constraint female entrepreneurship via 4 indicators (15 sub-indicators), such as the general ease of doing business, the cultural perception of female entrepreneurs, the quality of governance, and entrepreneurial supporting factors. 7 World Bank, World Bank Data, Female share of employment in senior and middle management (%) - Romania | Data (worldbank.org). 8 World Bank, World Bank Data, Female share of employment in senior and middle management (%) - Romania | Data (worldbank.org). 9 World Bank, Gender Data Portal, Indicators - World Bank Gender Data Portal. 10 World Bank, Gender Data Portal, Indicators - World Bank Gender Data Portal. 11 Self-employment, often used as a proxy for measuring entrepreneurship, is not a perfect measure. While self- employment can be an indicator of entrepreneurship, it can also include individuals who are self-employed out of necessity rather than choice or who are engaged in activities that are not entrepreneurial in nature. For example, some individuals may be self-employed because they cannot find regular employment, rather than because they have identified a new business opportunity. Others may be engaged in low-growth or low- innovation activities, such as selling goods in a market or providing basic services like cleaning or cooking. Other potential measures of entrepreneurship are, for example, business creation, small businesses ownership, venture capital, R&D spending, and patents. 10 entrepreneurship, given the similar data limitations associated with most surveys (for example Hamilton 2000; Dawson, Henley, and Latreille 2009; Verheul et al. 2012). We start by analyzing the rate of self-employment by gender and then cross it with income quintiles and regions. In Romania, the raw gender gap in entrepreneurship using self-employment as a proxy is 4.2 percentage points, with fewer women declaring themselves self-employed than men. The gap varies by income quintiles, being largest in the lowest quintile (15.3 percentage points) and smallest in the upper three quintiles. It is also slightly higher in rural areas. In Romania, we estimate 4 out of 10 self- employed people were women in 2020.12 At the same time, a lower share of working women than men reported being self-employed, with or without employees (14.9 versus 19.1 percent, respectively), representing a raw gender gap in entrepreneurship of 4.2 percentage points (figure 1). The gender gap varied significantly across income quintiles (figure 2): the gap in the lowest income quintile was 15.3 percentage points in 2020, whereas it was negligible in the upper three income quintiles (ranging from 1.5 to 2.3 percentage points) and it was negative for the second income quintile (-6.1 percentage points). Gender gaps in entrepreneurship were similar for urban and rural regions, although slightly higher in rural areas (3.7 versus 2.4 percentage points in urban areas) (figure 3). Alternative measures of the entrepreneurial gender gap using firm ownership and female manager positions confirm that women play a less significant role in entrepreneurship. Data from the World Bank Enterprise Survey (2018–2020) show that one-third of companies have at least one female owner.13 The share of companies where all owners are female is even lower, though just slightly. In addition, only 17.2 percent of companies have a female top manager (figure 4). Figure 1. The gender gap in entrepreneurship in Romania Figure 2. Self-employment rates by gender and income measured as the share of working women and men who quintiles, 2020 are self-employed, 2020) 0.7 0.6 female 0.149 0.5 0.4 0.3 0.2 male 0.191 0.1 0 male male male male female male female female female female male female 1 2 3 4 5 Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu) Note: Self-employment is self-declared and includes those with and without employees. Income quintiles are based on the per capita disposable income; the lowest quintile is 1. 12 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 13 World Bank Enterprise Survey, https://login.enterprisesurveys.org/content/sites/financeandprivatesector/en/library.html. 11 Figure 3. Self-employment rates by gender and rural and Figure 4. The gender gap in entrepreneurship in Romania urban areas, 2020 measured as the proportion of firms with female ownership and top female managers 0.35 0.350 0.324 0.317 0.3 0.300 0.25 0.250 0.2 0.200 0.172 0.15 0.150 0.1 0.100 0.05 0.050 0 Female Male Female Male 0.000 Has female All-female Female top Rural Urban owner owners manager Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu) (figure 3) and an elaboration based on World Bank Enterprise Survey data (2018–2020), https://login.enterprisesurveys.org/content/sites/financeandprivatesector/en/library.html (figure 4). Note: Self-employment is self-declared and includes those with and without employees. Figure 4 represents the entrepreneurship gap using alternative measures. We use median survey weights. We do not show estimates for self- employed with employees due to the small sample size (N=126). When restricting the analysis to new businesses to capture start-up activity, there are no gender gaps, but Romania lags behind most countries in the activity rates of both men and women. Based on recent data from GEM (2022a), the proportion of Romanian adults actively starting or running a new business is lower than in most countries. It puts Romania in 35th place out of 47 countries, but the female-to-male ratio is 0.98, which is close to parity. Furthermore, women’s share of early-stage entrepreneurs is higher than the EU average (OECD and the European Union 2020). While in the lowest income quintile, less than one-third of the self-employed population is female, the female share is above 40 percent in the rest of the income quintiles. Data from the EU-SILC 2020 reveal that the share of female self-employed in all self-employed was especially high in the second- lowest income quintile, with more than half of the self-employed population being female (figure 5). In comparison, only one-third of the self-employed in the lowest income quintile were women, while the share fluctuated around 40 percent for the upper three income quintiles.14 This could indicate that women in the lowest income quintile have fewer opportunities to start and grow their businesses due to various barriers, such as limited access to capital, markets, education, and networks. In addition, it may be that women in the lowest income quintile may face greater economic vulnerability and have to rely more on other forms of work, such as informal or precarious employment, to make ends meet. Addressing the gender gap in entrepreneurship among the poor could therefore require policies and programs that address the underlying barriers women face in accessing entrepreneurship opportunities, such as providing access to finance, training, and mentoring, as well as addressing social norms and discrimination that limit women’s economic empowerment. Less than half of self-employed people are female, independent of age group. We investigate whether there are any age patterns in self-employment in Romania. This analysis can help to identify potential barriers to self-employment. If older women are more likely to be self-employed, for example, there might be barriers for women when they are younger or of child-bearing age. In 14 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 12 addition, this analysis enables us to determine whether policies supporting female entrepreneurs should be tailored toward a certain age group. We find that women account for a slightly larger share of the self-employed population among older workers than younger workers (figure 6). Figure 5. Female share of all self-employed by income Figure 6. Share in self-employed by gender and age groups, quintiles, 2020 2020 0.60 0.559 Female share in self- 0.50 0.463 employed (seniors) 0.430 0.395 0.40 Female share in self- 0.291 employed (adults) 0.30 0.20 Female share in self- employed (youth) 0.10 0% 20% 40% 60% 80% 100% 0.00 1 2 3 4 5 male female Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: Self-employment is self-declared and includes those with and without employees. Income quintiles are based on the per capita disposable income; the lowest income quintile is 1. Self-employment plays a significant role in the employment of the poorest women. While only 5.9 percent of working women in the upper-income quintile declared themselves to be self-employed in 2020, this was true of 43.1 percent of working women in the lowest income quintiles (figure 1).15 Importantly, differences are even more marked for working men.16 Self-employment ratios are higher in rural areas for both sexes. At the same time, self-employment ratios (the ratio of self-employed workers to all workers) were higher in rural than urban areas for both genders (figure 7). Approximately 3 out of 10 working men and women in rural areas reported being self-employed in 2020, compared to fewer than 1 out of 10 working men and women in urban areas.17 The ratio was slightly lower for working women than for working men in rural and urban areas. 15 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 16 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 17 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 13 Figure 7. Self-employment rates by gender and rural and Figure 8. Sectoral distribution of self-employed women, urban areas, 2020 2020 0.35 100% NACE R-U sectors 90% Human health and social 0.3 Education 80% 0.25 Public admin, defense, 70% social security NACE L-N sectors 0.2 60% Finance and Insurance 0.15 50% ICT 40% Accomodation and food 0.1 Transport and storage 30% 0.05 Wholesail and retail; repair 20% Construction 0 Female Male Female Male 10% NACE B-E sectors Rural Urban 0% Primary sector Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: Self-employment is self-declared and includes those with and without employees. In figure 8, the economic sectors are per NACE Rev. 2. Not surprisingly, given that self-employment plays a more significant role in rural areas, the majority of the self-employed worked in the primary sector. Nearly 8 out of 10 self-employed women (figure 8) and nearly 7 out of 10 self-employed men worked in the primary sector in 2020.18 These sectoral distributions are also reflected in the occupational distribution of the self-employed. Most self- employed workers declared themselves as being market-oriented skilled agricultural workers or agricultural, forestry, and fishery laborers (more than 80 percent in the case of women and close to 70 percent in the case of men).19 Surprisingly, these occupations also played a significant role in urban areas (for more than 60 percent of self-employed women and nearly half of self-employed men). In urban areas, self-employed men also worked as building and related trades workers, excluding electricians (14.4 percent) and drivers and mobile plant operators (5.2 percent), while self-employed women also worked as personal service workers (5.2 percent) and food processing, woodworking, garment, and other craft and related trades workers (4.0 percent).20 In 2020, female entrepreneurs were, on average, older and less skilled than their male counterparts and belonged to a lower share of households with children than did male entrepreneurs. Table 1 presents the characteristics of male and female entrepreneurs as of 2020 in more detail: female entrepreneurs were, on average, 10 years older than males and that only a small share were youth entrepreneurs (between 20 and 30 years old). They were mainly middle-skilled and, on average, less skilled than their male counterparts. Compared to male entrepreneurs, women were less likely to 18 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 19 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 20 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 14 come from households with children. They were slightly more likely to employ a professional for childcare, but the share of those from households in which that would be necessary was still negligible. Table 1. Summary statistics of entrepreneurs by gender, 2020 (1) (2) Female Male mean SD mean SD Age 57.70 17.75 47.42 15.69 No schooling 0.04 0.20 0.03 0.16 Primary education 0.21 0.41 0.10 0.30 Secondary education 0.65 0.48 0.73 0.45 Tertiary education 0.04 0.21 0.05 0.21 Ph.D. 0.00 0.00 0.00 0.00 Youth 0.08 0.27 0.12 0.32 Adults 0.26 0.44 0.44 0.50 Seniors 0.24 0.43 0.29 0.45 Urban 0.27 0.44 0.26 0.44 HH with child (<16) 0.31 0.46 0.42 0.49 HH with young child (<6) 0.09 0.29 0.12 0.33 HH with child (<6) attending 0.03 0.16 0.04 0.20 preschool HH with child (<6) and 0.00 0.00 0.00 0.02 childcare (center based or daycare center) HH with child (<6) and 0.01 0.08 0.00 0.03 childcare (by professional) HH with child (<6) and 0.14 0.35 0.17 0.37 childcare by HH member Employer 0.06 0.23 0.08 0.26 Income quintile 1 (poorest) 0.28 0.45 0.48 0.50 Income quintile 2 0.40 0.49 0.22 0.41 Income quintile 3 0.14 0.35 0.12 0.32 Income quintile 4 0.08 0.28 0.09 0.29 Income quintile 5 (richest) 0.10 0.30 0.10 0.29 Primary sector 0.78 0.42 0.64 0.48 NACE B–E sectors 0.02 0.15 0.03 0.18 Construction 0.01 0.10 0.20 0.40 Wholesale and retail; repair 0.06 0.24 0.06 0.24 Transport and storage 0.00 0.00 0.03 0.16 Accommodation and food 0.02 0.12 0.01 0.08 ICT 0.00 0.04 0.00 0.04 Finance and Insurance 0.00 0.04 0.00 0.05 NACE L–N sectors 0.01 0.12 0.01 0.10 Public admin, defense, social 0.00 0.04 0.00 0.00 security Education 0.01 0.10 0.00 0.00 Human health and social 0.04 0.19 0.00 0.06 work activities NACE R–U sectors 0.05 0.21 0.01 0.11 Employee cash or near cash 270.4 1339.4 470.0 1898.5 income Income from the rental of a 11.2 144.3 12.0 200.5 property or land 15 Interest, dividends, profit 1.4 38.9 13.2 261.3 from capital investments in unincorporated business Cash benefits or losses from 877.3 1852.9 2013.9 2796.9 self-employment Value of goods produced for 1139.0 1043.9 1063.4 949.3 own consumption Observations 951 1205 Source: EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: We approximate entrepreneurship by those who reported being self-employed (with or without employees) in 2020. Households without children are assigned a value of zero in the childcare-related variables. The economic sectors are at the NACE Rev. 2 level. Income quintiles are based on per capita disposable household income. Income variables are reported in the net values of national currencies. In 2020, female entrepreneurs were mainly from the second (poorest) income quintile, were less likely to employ people, and worked primarily in the primary sector. As can be seen in table 1, in 2020 female entrepreneurs in Romania were mainly from the second income quintile, unlike male entrepreneurs, who were most likely to belong to the lowest income quintile. A lower share of female entrepreneurs than male entrepreneurs were employing other people and there were some important differences in the sectoral distribution. However, both self-employed men and women worked to a large extent in the primary sector, which is also reflected in the low share of self-employed who were living in urban areas. In line with this evidence, the share of female owners falls with the number of employees for the period 2018–20. Figure 9 plots the share of female owners by company size for 2018–20. While a higher share of owners in small firms was female, the number drops significantly for medium-sized firms and large firms. This could be a sign that female-owned firms faced performance and growth restrictions (Ubfal 2023). This phenomenon is commonly referred to as the “leaky pipeline” or the “glass ceiling” effect. The leaky pipeline suggests that women are able to enter entrepreneurship at an equal rate as men, but businesses that women entrepreneurs own do not grow beyond small-sized enterprises, likelydue to a multiplicity of barriers. Figure 9. Percentage of female owners by company size, 2018–20 70.0 63.3 60.0 50.0 Percentage 40.2 40.0 34.6 30.0 20.0 10.0 0.0 Small (0-19) Medium (20-99) Large (>=100) Source: World Bank Enterprise Survey 2018–2020, https://login.enterprisesurveys.org/content/sites/financeandprivatesector/en/library.html. Note: We weight observations by median survey weights. 16 There are important gender gaps in several income components between male and female entrepreneurs. Despite the data challenges,21 we use self-employment income data from the EU-SILC 2020, which includes information on gross cash profits or losses from self-employment and the value of goods produced for own consumption.22 Our estimates show that female entrepreneurs only earned 43.6 percent of the earnings of male entrepreneurs when relying on data gathered on cash benefits or losses from self-employment. On the other hand, the value of goods produced for own consumption is larger for women (lei 1,139.0 versus lei 1,063.4 for men). Taking both income streams together, female entrepreneurs earn 65.5 percent of the income generated by male entrepreneurs. Female entrepreneurs’ income from the rental of a property or land is also slightly lower than that of male entrepreneurs. There are marked gender gaps in the income generated from capital investments in unincorporated businesses, with women’s average profits, interests, and dividends being only 10 percent of men’s. This could be due to a range of factors, such as gender discrimination, differences in access to financial and human capital, and differences in business strategies and markets. As of 2014, female entrepreneurs in Romania reported being, on average, happy with their decision to become an entrepreneur, though satisfaction was related to income, education level, and age. Pocol and Moldovan-Teselios (2014) interviewed 602 women entrepreneurs in Romania to study their satisfaction with the decision to become an entrepreneur. They find that over 70 percent of the study participants are satisfied with their decision to become an entrepreneur and feel respected based on their status. There are some important differences in income level, with those with higher income reporting higher satisfaction rates. Furthermore, according to the study, young women entrepreneurs are more likely than older ones to repeat their decision to become entrepreneurs. Moreover, those with secondary education are less likely to repeat this decision. Although women entrepreneurs are mainly satisfied with their status, many also report being stressed, not having enough time to themselves, or feeling that they do not spend enough time with their children. Incentives for female entrepreneurship center around financial and economic reasons; at the same time, women are more interested in “impact entrepreneurship.” Another finding of Pocol and Moldovan-Teselios (2014) is that women in Romania mainly become entrepreneurs for financial reasons (27 percent), to gain independence (17 percent), because they are not employed (10 percent), or to take advantage of an emerging opportunity (8 percent). Only 11 percent of the interviewed women entrepreneurs took over a family business. Beyond these self-focused reasons for entrepreneurship, evidence from GEM (2022b) shows that women are more interested in “impact entrepreneurship,” which is values driven, socially oriented, and locally focused. The female-to-male ratio in the motivation to start a business to make a difference is 7 to 6: 71.6 percent of women report 21 These estimates must be taken with caution, because income from self-employment is one of the most difficult components to measure in income surveys, which poses a major challenge in terms of data quality and comparability. Among the problmes survey research runs into when attempting to obtain accurate information about the incomes of the self-employed are the self-employed’ s accounting practices, which may not separate their business and personal finances; and the difference between their financial and accounting frameworks and that used by statisticians in constructing national accounts. Moreover, the self-employed are less likely than employed individuals to respond to income surveys and their income variables are subject to higher levels of item nonresponse. Additionally, those self-employed individuals who do respond to surveys are more likely to underreport their income. Finally, the growth of self-employment as a secondary activity for employees presents additional problems, because unless such secondary activities are appropriately covered in an income survey with detailed questions, it too will result in underreporting (Ciampalini, Bartoletti, and Verma 2007). Ciampalini, Betti, and Verma (2009) examine data on self-employment income as obtained via the EU-SILC from a comparative perspective and find substantial differences across countries, potentially due to methodological differences among the surveys. Therefore, these results cannot be compared against other countries. 22 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 17 that they started a business to make a difference, compared to 60.3 percent of men (GEM 2022b). The greater attraction of women to social entrepreneurship has great potential to spark more-sustainable business models. To validate that there is a female entrepreneurship gap in Romania, we estimate a probit model. The measures of the gender gap we reported in previous figures do not control for workers’ demographic characteristics (so they were unadjusted measures). The raw gender gap in entrepreneurship refers to the difference in the probability of being self-employed between men and women, without considering any other factors that might influence this outcome. To analyze this gap, we run a probit model to investigate whether the probability of being self-employed differs by gender (see Equation 1 below). Y is a dummy variable that is equal to one if a working person is self-employed and zero otherwise. In other words, it denotes the probability of a working person being self- employed. F is a dummy variable that takes the value of one if the working person is a woman and zero otherwise. In the second step, we control for several factors that might influence the outcome to isolate their effect on gender, such as age, education level, and area women or men live in. We also include the economic sectors and income quintiles as control variables. These factors are denoted by X. For this purpose, we estimate the following empirical model: Equation 1 = + + , ℎ ~(0, 2 ) 0 ≤ 0 = { 1 ≥ 0 Standard errors are robust to account for potential heteroskedasticity. In what follows, we report the marginal effects of the underlying probit model to get a sense of the size of the gender gap. We report the findings in columns 1 and 2 of table 2. We run a logit model to confirm the robustness of our findings and present the results in column 3 of the table. The probit regression confirms our findings presented in the previous figures: there is an entrepreneurial gender gap in Romania. The coefficient in row 1 in table 2 is negative across model specifications. The coefficient in column 1 presents results from a simple probit regression without controlling for observable characteristics and is -0.0422. This means that the probability of being self- employed decreases by 4.22 percentage points for females when compared to males (without controlling for observable characteristics such as age). This result is in line with the results presented in figure 1. We find that even if you compare men and women who have similar characteristics, men are still more likely to be self-employed than women. We find that the negative effect of gender on the probability of being self-employed shrinks from -0.0422 to -0.0373 after controlling for observed differences between males and females. The coefficient in column 2 in table 2 indicates that the probability of being self-employed decreases by 3.73 percentage points when a working person is female compared to when the working person is male, controlling for the rest of the variables. The differences in the coefficients presented in columns 1 and 2 indicate that while the gender gap is to some extent explained by systematic differences between men and women, for the most part the gap is explained by unobservable characteristics, such as discriminatory structures against women entrepreneurs and those who aspire to become such. The coefficient from a logit estimation reported in column 3 in table 2 is very similar in magnitude and confirms our results. These findings generate evidence of an important “adjusted” gender gap in entrepreneurship in Romania, even when accounting for different observable characteristics. Many factors could explain this, ranging from discrimination to differences in unobservable characteristics. We next analyze the potential drivers of this gap in more detail. 18 Table 2. The impact of gender on the probability of being self-employed (marginal effects) Probit Probit Logit Female -0.0422 -0.0373 -0.0361 (0.000207) (0.000228) (0.000232) Mean (Dep. Var) .17 .17 .17 SD (Dep. Var.) .38 .38 .38 Controls No Yes Yes N 13,137 13,137 13,137 Source: EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: The outcome variable is an indicator variable that is equal to one for the self-employed and zero otherwise. We restrict the sample to the working population and control for covariates in columns 2 and 3. 3. Potential Drivers To investigate the underlying drivers behind the entrepreneurial gender gap in Romania and to identify potential barriers to female entrepreneurship, we follow the framework developed by Brush, De Bruin, and Welter (2009). The gender-aware framework of female entrepreneurship formulated by Brush, De Bruin, and Welter (2009) draws from the traditional entrepreneurship framework (Bates, Jackson, and Johnson 2007), often denoted as the “3 Ms,” and combines it with institutional theory to derive a so-called 5 Ms framework. As mentioned before, while the 3 Ms approach only considers three factors that enable entrepreneurship and venture growth—markets, money, and management—the 5 Ms add two additional factors important to female entrepreneurship: motherhood and the meso- and macroenvironment (figure 10). Motherhood proxies the household and family context of female entrepreneurs and the meso- and macroenvironment captures societal and cultural norms as well as institutional structures. Brush, De Bruin, and Welter (2009) stress that individual and societal factors interact with each other, ultimately generating the need for integrated approaches to facilitate female entrepreneurship. Figure 10. 5 Ms framework of Brush, De Bruin, and Welter (2009) Source: Brush, De Bruin, and Welter 2009. To shed light on these mechanisms in the case of Romania, we explore all these different aspects by exploratory data analysis. We start by investigating the role of the legal framework by using data from the Women, Business, and the Law Index (WBL) of the World Bank. Next, we analyze whether financial inclusion could affect the observed gaps. To this end, we conduct a descriptive analysis using data from the Global Findex Database and the “Ease of Doing Business” indicator. In addition, we explore 19 potential drivers around access to assets by analyzing data on educational attainment from the EU- SILC, on soil ownership from the World Bank Data Gender Portal, and on access to entrepreneurial training from GEM. Furthermore, we investigate the influence of social and cultural factors by using data from the World Values Survey and the SIGI Index.23 Lastly, we take a deep dive into the influence of motherhood and childcare by presenting new empirical evidence on their interaction with the probability of being self-employed. Legal, Institutional, and Policy Frameworks The legal framework for female entrepreneurship in Romania seems to be sound. According to the latest Women, Business, and the Law Index (WBL) of the World Bank, there are no legal barriers with respect to female entrepreneurship in Romania.24 The index measures four different dimensions of the legal framework: discrimination in access to credits, the possibilities of signing a contract as well as registering a business, and the right to open a bank account. All these potential gender gaps in the legal system are closed in Romania. Consequently, legal factors might not be the leading cause of barriers to female entrepreneurship. At the same time, consultations with the Romanian Ministry of Entrepreneurship and Tourism revealed that there is no gender-specific legal framework in Romania. Consultations conducted by a World Bank team in April 2023 with the ministry revealed that strategies around female entrepreneurship do not follow specific laws, instead referring to EU directives on the private sector. Based on this exchange, the public sector in Romania seems to lack laws and policies guiding the program around female entrepreneurship. During 2014–20, Romania had an entrepreneurship strategy that included some targeted schemes. Until recently, the entrepreneurship strategy in Romania was guided by the SME and Entrepreneurship Strategy 2014–2020, with additional support for underrepresented groups (OECD and the European Union 2020). Some dedicated schemes, such as business training and advice, mainly targeted the youth. One program tailored explicitly toward women was the Programme for the Development of Entrepreneurial Culture among Female Managers in the SME Sector 2020, which had a strong focus on building entrepreneurial skills. It also offered grants and loan guarantees, but was small in nature, only targeting around 160 beneficiaries (OECD and the European Union 2020). Overall, the entrepreneurial programming focused on grants to new entrepreneurs, mainly financed by EU Structural Funds and the Start-Up Nation program (OECD and the European Union 2020). In 2022, the Ministry of Entrepreneurship and Tourism introduced the Woman Entrepreneur Program. The program aims to stimulate and support the establishment and development of private economic structures established and managed by women, improve their economic performance, to achieving sustainable and inclusive economic growth, with the main goal of closing the entrepreneurial gender gap in Romania (Van Kline 2022). Its budget is approximately 200 million euros for the years 2022–27 and targets 5,000 beneficiaries (Van Kline 2022). Participation in the program is restricted to companies that have at least one female partner who holds at least 50 percent of shares. Another novelty of the program is its digital design. Applications, as well as the implementation of the support measure, take place entirely online on a dedicated platform. In recent years, banks have introduced financial products that target smaller businesses and women entrepreneurs. Examples are lending initiatives to SMEs by Patria Bank, backed up by a loan from the IFC, of which 50 percent are earmarked for women-led SMEs (IFC 2022). The IFC’s Banking on Women 23 World Values Survey, Online Data Analysis, WVS Database (worldvaluessurvey.org). 24 World Bank, Women, Business and the Law. Romania, https://wbl.worldbank.org/en/data/exploreeconomies/romania/2022. 20 Program is another example. A study evaluating its impacts shows positive effects on the performance of women-led businesses (IFC 2019). Another important initiative targeting entrepreneurial activity in Romania is the Start-Up Nation program. In 2023, the government launched the third edition of this program, which aims to stimulate the establishment and development of SMEs. More broadly speaking, its goal is to generate sustainable development, a smart economic recovery, sustainable and inclusive digitization, innovation and entrepreneurial training, and new jobs (Barac 2022). Based on information gathered during consultations with the World Bank team in April 2023, start-ups applying to the program receive more points during the application procedure if women are employed by that start-up or if it is a woman applying, but this initiative does not follow through to monitor how specifically female-led start-ups are doing. According to data provided by the government in September 2023, in 2022, 50.2 percent of companies applying under Pillar 1 and Pillar 2 of the Start-Up Nation program, have women as shareholders. These numbers indicate that the program might be an important mechanism to support female entrepreneurship. The Ministry of Economy is currently coordinating at the national level a National multi-annual program – The Woman Entrepreneur 2022. The National Confederation for Female Entrepreneurship (CONAF) has been an important advocate for the rights and interests of businesswomen in Romania. The CONAF, a federation of female entrepreneur, was established in 2018 (Wegate 2023). Its mission is to promote and protect the rights of businesswomen in Romania, connect them to each other, and broaden their horizons through missions to other countries (Wegate 2023). The confederation organizes a yearly “Women in Economy” Gala, during which women entrepreneurs connect with officials with ministries, journalists, banks, and brokers (CONAF 2023). Access to finance Although the European venture capital market is booming, with record-breaking growth in 2021, a strong female funding gap in the EU persists. In 2021, the European venture capital (VC) market invested over 100 billion euros in European start-ups (IDC 2022). Although these investments are highly beneficial, resulting in almost 100 unicorns across Europe, investments have not gained any ground with respect to gender diversity. Nevertheless, there are important initiatives in place and specific criteria in European financed programs aiming to address this lack of diversity. Most importantly, all-women teams have only acquired 2.7 percent of the overall start-up investment per year in the EU since 2018 (IDC 2022). When adding mixed-gender-led start-up teams, the number remains low—11 percent. Existing research consistently affirms the existence of gender disparity within the European venture capital ecosystem, indicating that this is not a specific issue affecting Romania (Pavlova and Gvetadze, 2023). Disaggregated data at the country level on these gaps in Romania are not available but given the low relative performance on gender equality in Romania, the situation is unlikely to be much better. Generating this data for Romania would be useful to get a better understanding of the magnitude of the problem. The European start-up scene is significantly male dominated, which might create barriers to women’s accessing of start-up funding. Only 8 percent of start-ups in Europe are founded by all- women teams, and three-quarters by all-men teams (IDC 2022). In addition, only 15 percent of VC general partners in the EU are female, and those women who are general partners have less investment power (IDC 2022). Similarly, four out of five members of VC investment committees in Europe are male (IDC 2022). The angel investment world is similarly male dominated, with only 1 in 10 angel investors being women (Jormalainen 2023). Research shows that women face many challenges in these types of male-dominated environments, ranging from pervasive stereotypes to higher levels 21 of stress and anxiety (García Johnson and Otto 2019). Evidence shows that women and men get asked different questions by VCs, which significantly influences funding (Kanze et al. 2017). The female funding gap is problematic for the business environment in Romania, given that most women lead their businesses on their own or as part of all-women teams. According to data collected by Pocol and Moldovan-Teselios (2014) through interviews, half of the women entrepreneurs had started their business on their own, and another 8 percent had done so as part of all-women teams. Two out of 10 respondents indicated that they had started their business with both women and men. Therefore, it is problematic that all-women teams are less able to access funding. The female funding gap might result in women’s gaining lesss funding or having a lower probability of scaling up their businesses. Romanian women also report lower levels of financial inclusion. Women’s share of bank accounts has increased significantly over time, from 41 to 66 percent between 2011 and 2021.25 Still, the gap between their share and that of men remains persistent; 73 percent of men had an account in 2021 (figure 11). Analyzing financial literacy indicators related to starting a business reveals contradictory trends (figure 12). On the one hand, the share of women who save to start, operate, or expand a farm or business increased between 2014 and 2017 from 2 to 10 percent and in 2017 exceeded the share of men (7 percent).26 On the other hand, the share of women who borrow to start, operate, or expand a farm or business decreased in that period from 2 to 1 percent and as of 2017 was below the level of men (3 percent).27 Thus per the latest available data, women still seem less financially literate than men on average. Figure 11. Account ownership rate by gender, 2011–21 Figure 12. Access of capital to start, expand, and operate a business by gender, 2014 vs. 2017 80% 70% Borrowed to start, operate, or expand a farm or 60% business, male (% age 15+) 50% Borrowed to start, operate, or expand a farm or 40% business, female (% age 15+) 30% Saved to start, operate, or 20% expand a farm or business, male (% age 15+) 10% Saved to start, operate, or 0% expand a farm or business, 2011 2013 2015 2017 2019 2021 female (% age 15+) Account, female (% age 15+) 0% 5% 10% 15% Account, male (% age 15+) 2017 2014 Source: World Bank, Global Findex Database, The Global Findex Database 2021 (worldbank.org). Women are less financially literate when it comes to digital payment services, and the gaps in this respect between Romania and the euro area are marked. While 66 percent of men made or received 25 World Bank, Global Findex Database, The Global Findex Database 2021 (worldbank.org). 26 World Bank, Global Findex Database, The Global Findex Database 2021 (worldbank.org). 27 World Bank, Global Findex Database, The Global Findex Database 2021 (worldbank.org). 22 a digital payment in 2021, this was true of just 61 percent of women.28 On average, in 2021 both men and women in Romania lagged significantly behind their counterparts in the euro area on this indicator; in the eurozone nearly the full universe of men and women engaged in digital payments in 2021. Access to assets Access to entrepreneurial assets is crucial for successful business ventures and a large body of literature demonstrates their interconnection. Entrepreneurial assets span a range of different factors, such as financial, human, and social types of capital (Envick 2005). Examples of these are land tenure as well as social capital in the form of networks and skills—especially creativity, problem-solving skills, and management skills (Kodithuwakku and Rosa 2002). Moreover, Envick (2005) adds the idea of psychological capital, that is, positive thinking, hope, and confidence to the theoretical literature. Others have studied the impact of experience and knowledge (Barreira 2005). Both male and female entrepreneurs in Romania generally have secondary education, though self- employment rates are highest among those without schooling. Both male and female entrepreneurs generally have secondary education (table 1); however, share of those with primary education is higher among female entrepreneurs (table 1). Self-employment rates are higher among those without schooling and decrease with the level of education. A better understanding of why self-employment plays such a significant role among the low-skilled in Romania requires more evidence and more- detailed analyses. In general, the evidence presented so far in this note indicates that self-employment centers around low- and middle-skilled agricultural employment, both for working men and women. There are important deficits in entrepreneurial education in Romania. Data presented by GEM (2022) reveal that experts rate entrepreneurial education poorly in Romania. Deficits in comparison to other countries persist both during and after school. The low educational capacity in entrepreneurship could partly explain why so few Romanians feel equipped to start a business; Furdui, Lupu-Dima, and Edelhauser (2021) found positive effects of entrepreneurial training during school on attitudes toward entrepreneurship in Romania. In addition, women-led SMEs report that an inadequately educated workforce is one of their biggest challenges (IFC 2019). At the same time, women entrepreneurs are as likely as men to report that they have the necessary skills to start a business but are less likely to agree with the statement that it is easy to start a business (GEM 2022a). Women are less likely to own land, which could be an important driver of entrepreneurial gender gaps, given the importance of the primary sector. The latest available data on land ownership are from 2018 and show that the male-to-female soil ownership ratio is 1.4, indicating a significant gender gap on this indicator.29 This gap could be problematic, given that many self-employed in Romania engage in the primary sector and land ownership might play a crucial role in acquiring loans. A study by the IFC (2019) finds that 80 percent of loans received by interviewees required land as collateral. Gender gaps in digital and technological skills could also affect the entrepreneurial gender gaps. While overall, a larger share of women than men report having internet access for personal use at home (57.8 percent versus 45.6 percent, respectively), the share is lower for women than men among the self-employed (30.5 percent versus 44.2 percent).30 In addition, only one-third of ICT graduates in 2018 were women (EIGE 2020). These numbers translate into a low share of women in the ICT sector in the labor market, where only one-fourth of ICT specialists are women (EIGE 2020). Similarly, only 28 World Bank, Global Findex Database, The Global Findex Database 2021 (worldbank.org). 29 World Bank, Gender Data Portal, Indicators - World Bank Gender Data Portal. 30 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 23 one-fourth of the scientists and engineers in high-technology sectors are women (EIGE 2020). While there are no gender differences in the share of the population using certain ICT tools, such as computers, email, electronic documents, social media, or software, the gaps between the usage rates in Romania and the EU average are significant for both men and women (EIGE 2020). Institutional and cultural factors Overall, both male and female entrepreneurs face a challenging institutional environment in Romania. According to recent data presented by GEM (2022), experts rank Romania among the lowest on the governance indicators assessing the entrepreneurial environment of countries, indicating that the government needs to introduce more policies that incentivize entrepreneurship and help new businesses grow, such as tax incentives and matching grants (GEM 2022). In addition, Romania suffers from unequal cultural support for women entrepreneurs, has unfavorable regulations for women entrepreneurs, and lacks sufficient family support services (GEM 2022). On top of the challenging institutional setup for entrepreneurship in general, women face institutional gender discrimination. While the institutional environment is challenging for all entrepreneurs, women also face institutional discrimination in relation to gender equality and thus a double barrier. The concept of such a double barrier has been investigated by Gohman (2012), for example, who finds that institutions, such as economic freedom, connect with people’s preference for being self-employed. Similarly, Afandi and Kermani (2015) demonstrate that a significant share of the gender entrepreneurship gap in European countries is likely due to gender discrimination. While Romania was ranked relatively high on the 2019 SIGI (OECD 2019), an index that measures discrimination against women in social institutions across 180 countries, there is room for improvement. Romania came in 20th place in the index, but only performed average when the list is limited to countries to Europe. According to the different subdimensions of the index, women are especially confronted with discrimination within their families and face restricted civil liberties. Harmful gender norms in the population could create additional barriers to female entrepreneurship. As can be observed in figure 13, in the World Values Survey of 2017–22 4 out of 10 Romanian men (strongly) agree with the statement that men make better business executives than women. Only 28.6 percent of women do so, but the perception of men could negatively influence them. Gender differences are less marked when Romanians were asked about preschool children’s potential suffering due to working mothers (figure 14): 4 out of 10 both men and women agreed with this statement. The holding of such views by a large share of the population could impede the entry into entrepreneurship of women and mothers, on the one hand, and create stress for those who are already “mompreneurs” on the other. 24 Figure 13. Distribution of responses to question about Figure 14. Distribution of responses to question about whether men make better business executives than whether preschool children suffer because the mother is women by gender, 2017–22 working by gender, 2017–22 45 45 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 0 5 0 Average of Male Average of Female Average of Female Average of Male Source: World Values Survey 2017–2022, Online Data Analysis, WVS Database (worldvaluessurvey.org). In addition, studies have shown that differences in risk aversion play a significant role. Women are, on average, more risk averse than men (Filippin 2022). A study by Caliendo et al. (2015) that uses microdata from Germany shows that these differences in risk aversion explain a large share of the entrepreneurial gender gap. In Romania, Paun (2012) does find gender differences in risk aversion, namely that women are more risk averse than men, but that this gap decreases during crises. Motherhood We next take a deep dive into the interaction between motherhood and entrepreneurship in Romania. Putting special emphasis on this point is interesting, given that childcare has previously been identified as problematic in Romania (Lokshin and Fong 2006). In a study on the subject of “mompreneurship” in Romania, Leovaridis, Bahnă, and Cismaru (2018), the authors relied on a qualitative approach, though they indicated that a quantitative analysis could shed further light on the topic. We close this empirical gap through the analysis in this section. Importantly, we approximate motherhood by women who live in households with children below 16 years old.31 Consequently, our concept goes broader, in that it analyzes women’s potential care responsibilities as associated with living in the same household as a child, given that female grandparents, aunts, stepmothers, and older sisters might also be more adversely affected than their male counterparts (see related work by East, Weisner, and Slonim 2009; Posadas and Vidal-Fernandez 2013; Compton and Pollak 2014; Arpino and Bellani 2022). On average, self-employment rates of women in households with and without children are similar, but there are significant gaps for men. To investigate whether motherhood and the family context are significant constraints to entrepreneurship in Romania, we analyze the share of self-employed for households with and without children (below 16 years old) (figure 15). We find that, in 2020, there 31 Because the EU-SILC was mainly designed to analyze household income and living conditions, children of respondents are only observed when living permanently in their parents’ household. The questionnaire does not ask about the number of children women and men ever had, leaving unobserved those children who live outside the household (Greulich and Dasré 2018). 25 were no significant differences in the case of women, although there were marked gaps for men (figure 15).32 The patterns of results look similar when analyzing households with younger children. Young children might present greater barriers to self-employment, because they are less likely to be in school and might require more care from their parents. For this reason, we also look at households with young children (below six years old). In households with children as a whole, 23.5 percent of working men were self-employed, while this was true of only 16.9 percent of working men in households without children below six years old.33 In the case of households with small children (below six years old), there was again no gap in the self-employment rates of women between households with and without small children. The gender gap in self-employment was more significant for households with children. Notably, the gender gap in self-employment rates was larger for households with children (8.1 percentage points) than for households without children (2.2 percentage points) (figure 15). These gaps could mean that parenthood is a larger barrier to entrepreneurship for women than for men. Figure 15 Self-employment rates for households with or Figure 16. Self-employment rates for working women without children below 16 years old by gender, 2020 from households with or without children below 16 years by income quintiles, 2020 0.25 0.5 0.2 0.4 0.15 0.3 0.1 0.2 0.05 0.1 0 Female Male 0 1 2 3 4 5 HH with child (<16) No child No child HH with child (<16) Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: Income quintiles are based on per capita disposable household income; 1 is the lowest income quintile and 5 the highest. The gender gap in self-employment is especially large within the lowest income quintile, for both households with and without children. According to data from 2020, the self-employment rate was higher for women without children (below 16 years old) for all income quintiles but the highest quintile (figure 16).34 The gender gap in self-employment rates was highest in the lowest income quintile, independent of having children or not, although the gap was slightly higher for those low-income households with children (16 percentage points versus 14.5). This pattern of results could be a first hint that access to childcare might be lower in for households in lower-income quintiles and affect female entrepreneurship. The gender gap in self-employment is also higher for households with children than for households without children in the rest of the income quintiles, other than the 32 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 33 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 34 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 26 highest income quintile, where gender gaps in self-employment were very close, independent of having a child in the household or not.35 In both rural and urban areas, the gender gap in self-employment was larger for households with children than without children. As can be seen in figure 17, the differences in self-employment rates between working men and women were more significant in households with children than in households without children. The gender gap was especially large for households with children in rural areas: it was 12 percentage points, compared to a gap of 2.6 percentage points for households in urban areas.36 At the same time, there was a slightly reversed gender gap between working men and women in rural households without children. Figure 17. Self-employment rates for households with and Figure 18. Forms of childcare used by self-employed without children below 16 years old by gender and rural and people from households with children below six years urban areas old by rural and urban areas 0.4 0.35 0.3 0.25 HH with child (<6) and child-care by hh member 0.2 0.15 HH with child (<6) and 0.1 child-care (by professional) 0.05 HH with child (<6) and child-care (center-based or 0 day-care center) Female Female Female Female Male Male Male Male HH with child (<6) attending pre-school HH with HH without HH with HH without 0 0.2 0.4 0.6 child (<16) child (<16) child (<16) child (<16) rural urban urban rural Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). As of 2020, the self-employed in households with children under six years old mainly relied on childcare provided by other household members. In 2020, of those who were self-employed and lived in households with children under six years old, more than half had childcare provided by another household member.37 Additionally, one-third indicated that their child attended preschool, while only a small share employed a professional for childcare (2.6 percent). A negligible share indicated that their child accessed center-based childcare or attended a daycare center. Urban households seemed to have more access to public childcare than rural households. When distinguishing between rural and urban areas, a lower share of the self-employed in households with children under six years old in urban areas reported that their child attended preschool (26.1 percent versus 39.0 percent) (figure 18). Moreover, the share of self-employed relying on paid professionals for childcare was entirely driven by those in urban areas (figure 18). 35 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 36 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 37 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 27 The type of childcare used significantly differed across income quintiles, especially in the share of households employing professionals for childcare. There were significant differences across income quintiles in the type of childcare used (figure 19). The self-employed in households with children under six years old in the lowest three income quintiles did not employ professionals for childcare, while nearly 4 out of 10 self-employed in the higher income quintiles did so.38 At the same time, the self- employed in the lowest income quintile seemed to rely less on childcare provided by other household members. Only one-third of the households in the highest income quintile reported that another household member cared for their child, while more than half of the households in the lowest income quintile did so.39 Figure 19. Access to childcare by the self-employed from households with at least one child below six years old by income quintile, 2020 1 0.8 0.6 0.4 0.2 0 HH with child (<6) attending HH with child (<6) and HH with child (<6) and HH with child (<6) and pre-school child-care (center-based or child-care (by professional) child-care by hh member day-care center) Q1 Q2 Q3 Q4 Q5 Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). To analyze the influence of motherhood in more detail, we run a probit model to shed light on the interaction between having a child, having access to childcare, and the probability of women being self-employed. We restrict the sample to working women and run regressions at the individual level. Equation 2 below details our empirical model. The outcome variable Y is an indicator variable equal to one if a working woman is self-employed and zero otherwise. The main explanatory variable is a dummy variable, here denoted as C, equal to one for women living in a household with children and zero otherwise. We are also interested in the impact of having access to childcare, here denoted as CC, measured by an indicator variable that accounts for access to childcare, namely a dummy variable that is equal to one for households that report some form of childcare, either from a relative, professional, or daycare center, and zero otherwise. Lastly, we control for several sociodemographic characteristics with the variables X, such as age, education level, and the area women live in. We also include the economic sectors and income quintiles as control variables. Standard errors are robust to account for potential heteroskedasticity. We present the marginal effects of the probit model in table 2. Equation 2 = + + + , ℎ ~(0, 2 ) 0 ≤ 0 = { 1 ≥ 0 38 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 39 EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). 28 There are significant negative correlations between the probability of working women ’s being self- employed and living in a household with at least one child but positive correlations with having access to childcare. As shown in table 2, when controlling for sociodemographic characteristics, income, and employment sectors, working women who live in households with children have a lower probability of being self-employed than those who live in households without children. The probability of being self-employed decreases by 1.7 percentage points when a woman lives in a household with at least one child. On the contrary, having access to some form of childcare (compared to not having childcare or not having children) increases the probability of being self-employed by 0.4 percentage points. Both coefficients are significant at the 1-percent level. These results could imply that, first, motherhood and entrepreneurship are difficult to combine, and second, that childcare makes mompreneurship more feasible. We confirm the negative effect of motherhood on entrepreneurial activities by estimating a regression of motherhood on the probability of being self-employed. Table 3 presents the results. The negative relationship between motherhood and entrepreneurship could be related to the drivers identified by Leovaridis, Bahnă, and Cismaru (2018). The authors find that mothers face a complex social and cultural context and a high level of complexity around them when they attempt to balance childcare with care at home and work. Both community and family perceptions and attitudes toward mompreneurships could play a role. Moreover, Leovaridis, Bahnă, and Cismaru (2018) stress that Romanian society’s perception of mothers as caretakers might present an additional barrier, compounding the difficulty created by existing barriers related to lack of financing, taxation, and bureaucracy. Our findings are in line with evidence from other studies. Dutta and Mallick (2018) find negative effects of the fertility rate on entrepreneurship among women. Moreover, data from GEM (2022b) reveal that women are more likely than men to discontinue their businesses due to family reasons. We conduct several robustness tests in our analysis of the interaction between motherhood and entrepreneurship, starting with restricting the concept of entrepreneurship. First, we restrict the self- employed to those self-employed who have employees in order to address concerns about a valid approximation of entrepreneurship by self-employment. The self-employed without employees might also include bogus self-employment or other forms of vulnerable employment types. Given that we cannot distinguish between these types of employment and real solo entrepreneurship, excluding the self-employed without employees serves as validation. Table A.2 presents the marginal effects and shows that the results still hold. The marginal effect of having children in the household remains negative, standing at -0.00266, and the marginal effect of having access to childcare remains positive, standing at 0.00711. Both effects are significant at the 1-percent level. Our main analysis relies on the impact of the working women’s living in the same household as children; we validate our findings on motherhood by creating an indicator variable that identifies mothers of children in the household. The EU-SILC enables the identification of the mother of children via a unique identifier. In the main analysis, we consider all women, given that children might not live with their birth mothers. In addition, other women in the household, such as older sisters or grandmothers, might also be influenced in their decisions to become entrepreneurs when there are children present in the household. Still, these women might not be affected to the same extent as birth mothers and thus our observed effect might be an underestimation of the true underlying effect. To shed light on this possibility, we create an indicator variable that enables us to identify the mother of children in the household. Table A.3 presents the results. We find that marginal effects are indeed larger for mothers than for women, generally speaking. The marginal effect of being a mother is –5.76 percentage points and larger than the coefficient on women presented in table 3. 29 Table 3. Probit model on the probability of women in Romania being self-employed, 2020 (1) Variables Self-employed Age 0.000487*** (1.35e-05) No schooling 0.0230*** (0.00184) Primary educ. -0.0352*** (0.000810) Secondary educ. 0.0109*** (0.000430) Tertiary educ. -0.0165*** (0.000626) Urban -0.0400*** (0.000287) HH with child (<16) = 1 -0.0171*** (0.000362) Has access to childcare = 1 0.00416*** (0.000431) Income quintile 1 0.0462*** (0.000530) Income quintile 2 0.0623*** (0.000456) Income quintile 3 0.00879*** (0.000460) Income quintile 4 -0.00563*** (0.000497) Primary sector 0.0831*** (0.000557) NACE B–E sectors -0.165*** (0.000657) Construction -0.144*** (0.00157) Wholesale and retail; repair -0.156*** (0.000621) Accommodation and food -0.122*** (0.000944) ICT -0.133*** (0.00157) Finance and Insurance -0.158*** (0.00176) NACE L–N sectors -0.110*** (0.000850) Education -0.171*** (0.00107) Human health and social -0.0952*** (0.000670) Observations 3,017 Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: The table presents the estimates of a regression of having a child below 16 in the household on the probability of being self-employed among employed women. The outcome variable is an indicator variable that is equal to one if a woman is self- employed and zero otherwise. “HH with child (<16) = 1” is the main explanatory variable of interest and is equal to one if a woman lives in a household with a child below 16 years old and zero otherwise. “Has access to childcare = 1” is an indicator variable that is equal to one if a household has access to some form of childcare and zero otherwise. It is also zero for those 30 households without children to avoid listwise deletion. Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. Eldercare Nearly 2 out of 10 people in Romania are above 64 years old, which present a significant challenge for Romanian society. Demographic change increased the old-age dependency ratio from 23 percent to 29 percent between 2010 and 2021,40 which is evidence of the increasing challenges these demographic developments pose for Romania. Women might face an unequal care burden not only in the case of childcare but also in eldercare, which might further disincentivize them from taking up entrepreneurial and self-employment activities. To investigate this possibility, we run a probit model similar to the one on childcare, but instead of analyzing the impact of having children in the household, we investigate the effect of having elderly people (more than 64 years old) in the household. We find a significant but small impact of living together with at least one elderly person on women’s probability of being self-employed. Table 4 shows the marginal effects from a probit regression: although the effect is significant at the common significance levels, the marginal effect is close to zero. Consequently, the impact is negligible. Nevertheless, given demographic change, the effect size might change in the (near) future. In addition, the analysis is subject to important empirical shortfalls, as not all elderly people might live with their children. Consequently, we cannot observe the true eldercare responsibilities and the results shown here might underestimate the true impact. Table 4. The impact of eldercare on the probability of being self-employed among women, 2020 (1) Variables Self-employed HH with elderly -0.00468*** (0.000348) Age 0.00134*** (1.34e-05) No schooling -0.00304 (0.00202) Primary educ. -0.0480*** (0.000871) Secondary educ. 0.000929** (0.000451) Tertiary educ. 0.00900*** (0.000558) Urban -0.0375*** (0.000293) Mother = 1 -0.0265*** (0.000331) Income quintile 1 0.0446*** (0.000534) Income quintile 2 0.0453*** (0.000428) Income quintile 3 -0.00495*** (0.000439) Income quintile 4 -0.0229*** (0.000469) Primary sector 0.0927*** 40 World Bank, World Bank Data, Female share of employment in senior and middle management (%) - Romania | Data (worldbank.org). 31 (0.000606) NACE B–E sectors -0.178*** (0.000712) Construction -0.0927*** (0.00104) Wholesale and retail; repair -0.134*** (0.000625) Accommodation and food -0.0961*** (0.000885) ICT -0.178*** (0.00170) Finance and Insurance -0.203*** (0.00187) NACE L–N sectors -0.131*** (0.000872) Public admin, defense, social security -0.242*** (0.00161) Education -0.168*** (0.000941) Human health and social -0.109*** (0.000714) Observations 3,017 Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: The table presents the estimates of a regression of having an elderly person in the household on the probability of being self-employed among employed women. The outcome variable is an indicator variable that is equal to one if a woman is self-employed and zero otherwise. “HH with elderly” is the main explanatory variable of interest and is equal to one if a woman lives in a household with a child below 16 years of age and is zero otherwise. Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. 4. Conclusion and Policy Recommendations In this background note we analyze gender gaps in entrepreneurship in Romania and investigate potential drivers of these gaps using the 5 Ms framework, with a particular focus on mompreneurs. We use data from various sources, including the EU-SILC, the Mastercard Index of Women Entrepreneurs, and GEM, to measure these gender gaps, describe the average characteristics of female entrepreneurs in Romania, and measure potential drivers. As part of the analysis in this background note, we construct a probit model that focuses on the concept of mompreneurs in order to investigate the interaction between motherhood, childcare, and self-employment. Our results show significant gender gaps in entrepreneurship that disfavor females in Romania, especially those in the poorest income quintile. Our analysis reveals that 4 out of 10 self-employed persons in Romania are female. At the same time, a lower share of working women than men report being self-employed, with or without employees (14.9 versus 19.1 percent, respectively). The gender gap is mainly driven by the lowest income group, where the gap is quite sizable (15.3 percent). Moreover, it is slightly higher in rural than urban areas. We show that the average female entrepreneur is more likely to live in a rural area, be around 57 years old, and live in a household without a child under six. She most likely has completed secondary education, belongs to the second income quintile, and works in the primary sector. Our analysis also reveals a significant earning gap between self- employed men and women in Romania. 32 Our results indicate that all the drivers identified in the 5 Ms framework likely contribute to the observed gender gaps in entrepreneurship. While the legal framework is sound, there are other discriminatory structures in the institutional system in Romania, especially within women’s family contexts, which could negatively affect women’s likelihood of becoming entrepreneurs. Harmful gender norms most likely impose further barriers on female entrepreneurship, as do gaps in access to finance and financial literacy. Gender gaps in access to education also seem to play a role. In addition, our analysis reveals that there is a negative relationship between motherhood and self-employment. Access to childcare, on the other hand, is positively related to the probability of self-employment of women with children. These results indicate that women in Romania face multiple barriers to entrepreneurship. Moreover, given that the environment in Romania is already challenging for entrepreneurs in general (OECD and the European Union 2020), women face a double barrier compared to men. We confirm the validity of our findings by conducting several robustness tests, such as restricting the concept of entrepreneurship to self-employment with employees or only looking at women clearly identified as mothers in the EU-SILC. Our findings support the design and implementation of a nuanced approach toward female entrepreneurship that factors in the distinct challenges of different groups of women and consists of a menu of policy interventions. Our results show that gender gaps in entrepreneurship vary across subgroups of the country’s population. Therefore, there is no one-fits-all solution to increase the number of women entrepreneurs in Romania. Women in the lowest income quintile seem to face more-severe barriers, yet they rely more on entrepreneurship. The same applies to women living in rural areas. Challenges among women with and without children also differ. Overall, given that we detect gender differences in nearly all areas that could potentially hinder female entrepreneurship, the government should introduce a menu of different policy interventions that target all these drivers. Interventions that improve access to childcare, education, and finance, as well as strategies that target harmful gender norms, might be equally important. Eliminating barriers to female entrepreneurship could result in nearly half a million additional women entrepreneurs in Romania and spur inclusive economic growth (OECD and the European Union 2020). Fostering an environment that facilitates female entrepreneurship could have beneficial results for the green transition and inclusive economic models, given that women are more interested in “impact” entrepreneurship. Data from GEM shows that women are 11.3 percentage points more likely than men to say they would start a business in order to make a difference. In fact, three-quarters of women are motivated by intentions to make a difference, which shows that they could play a leading role in creating more-sustainable and inclusive business models. Combined with their concentration in the primary sector, female entrepreneurs could also assume a leading role in the green transition. Based on our findings, we recommend a tailored approach to decrease the gender gaps in entrepreneurship in Romania that factors in the distinct barriers faced by different groups of women. The analysis in this report points out important gender gaps in Romania. In addition, our analysis reveals that gender gaps in entrepreneurship differ for certain population groups. There are significant differences across income groups and between rural and urban regions. Moreover, motherhood and not having access to childcare seem to create additional barriers in some cases. These findings show that there is no one-fits-all solution to the mitigation of the barriers to female entrepreneurship in Romania and that there is a need for more-nuanced approaches. A recent review of potential interventions to support female entrepreneurship by Ubfal (2023) comes to a similar conclusion, indicating that interventions need to be more nuanced and better targeted. Moreover, rolling out a package of interventions targeting several of the identified drivers of these barriers simultaneously is recommended. We show important gender gaps in nearly all the potential 33 drivers of the barriers to female entrepreneurship in Romania. These start with differences in access to financial resources and education and other forms of assets, such as soil ownership. Moreover, we show that—on top of the already challenging environment for entrepreneurs in general—women face a double barrier, given that they are also confronted with discriminatory structures and institutions. Harmful gender norms might discourage or deteriorate women’s engagement in entrepreneurship. We also generate evidence that shows a negative relationship between entrepreneurship and motherhood. All these factors likely play a role in Romania’s gender gap in entrepreneurship and probably interact with each other. Therefore, we recommend introducing a menu of interventions that mitigate these barriers through a comprehensive approach. This recommendation is in line with previous research showing that interventions tackling barriers to female entrepreneurship should by design consider multiple barriers (Ubfal 2023). Increasing the participation of women in entrepreneurship and the business world also entails supporting women-led businesses. Closing the gender gap in entrepreneurship means not only increasing the number of female entrepreneurs and tackling the barriers that keep women from becoming entrepreneurs, it means providing support to women who already operate businesses. We show that the female share of business owners decreases by firm size, which could mean that women are trapped in lower-growth, lower-productivity businesses and do not receive sufficient support to scale their businesses. Based on these three main findings, we recommend the following policy actions: - Improve access to entrepreneurial training and education. Giving women greater access to entrepreneurial training, not just on how to start but also how to scale up and sustain a business, could facilitate female entrepreneurship. Teaching business and entrepreneurial skills during childhood, independent of gender, can have beneficial effects later in life (Jardim, Bártolo, and Pinho 2021). Closing gender gaps in tertiary education (Dutta and Mallick 2018; Gawel 2021) and increasing the share of women in ICT and STEM education fields could also have a positive effect.41 Tailored approaches are more effective in the case of entrepreneurs from vulnerable groups (OECD and the European Commission 2021). - Foster the development of women entrepreneurship networks, mentoring, and tutoring. Studies such as Markussen and Røed (2017) show that peer effects explain half of the gender gap in early career entrepreneurship. Fostering women’s entrepreneurship networks and mentoring and tutoring programs among female (and male) entrepreneurs and younger women could have positive trickle-down effects (Noguera et al. 2015). - Increase the number and the investment power of female GPs and push for stronger VC and angel investor Diversity and Inclusion Strategies. The European VC and angel investor environment is very male dominated, and there is evidence that this might create gender biases that prevent aspiring women entrepreneurs from accessing financial resources (IDC 2022). Creating a more diverse VC environment could decrease investor biases and improve women’s access to financial resources. - Generate sustainable financing schemes for women and better communicate initiatives targeting gender-smart investment. We show that there are still gaps in the access of women to finance, a phenomenon that has been flagged in previous research as well. Potential solutions are designing loans in a more inclusive way (Chowdhury, Yeasmin, and Ahmed 2018) and increasing financial inclusion, more generally speaking (Goel and Madan 2019). Given the gender gap in land ownership, finding alternative forms of collateral for bank loans could also increase women’s access to financing (IFC 2019). 41 For an introduction to the literature, see Poggesi et al. 2020. 34 - Include female entrepreneurs in the green transition. The fact that a significant share of the female self-employed in Romania is concentrated in the primary sector offers possibilities for an inclusive, green transition and for giving women an important role in this transition. In fact, evidence from Germany shows that female entrepreneurs are more active in the green economy and social entrepreneurship because they are more driven by overriding goals than their male counterparts (German Startups Association 2020). This represents an opportunity to foster more- inclusive and sustainable economic models. - Roll out interventions that target harmful gender norms around women in business. We show that women see themselves confronted with social norms that limit their intention to engage in entrepreneurship. In line with this argument, previous studies show that the success of entrepreneurship training for women depends on the degree to which they are subject to traditional norms (Field, Jayachandran, and Pande 2010). We recommend interventions that target these conservative norms and attitudes around gender and entrepreneurship in general, in line with previous research (Yordanova and Tarrazon 2010; OECD and the European Commission 2021). - Improve access to childcare and facilitate a better work-life balance for mothers and aspiring mothers. Our analysis generates new evidence on the negative relationship between motherhood and entrepreneurship. It also shows that having access to childcare improves the probability of being self-employed among women in households with children. These results make the case for investing in childcare infrastructure to facilitate the combination of entrepreneurship and motherhood. - Create a larger number of gender-disaggregated indicators and implement gender-sensitive monitoring and evaluation initiatives. Generating a larger number of gender-disaggregated indicators, both by the government and the banking sector, could help with understanding the entrepreneurial gender gap and its drivers in more detail.42 This type of information could also yield interesting insights that are valuable to the financial sector. For example, gender- disaggregated data by Garanti Bank revealed that women are more desirable customers, given that they demonstrate better payback behavior (McCartney et al. 2016).43 - Foster a healthier entrepreneurial ecosystem. In addition to the barriers specific to women, women also face general barriers to entrepreneurship in Romania. Consequently, it is also important to create a better entrepreneurial ecosystem. To do so, the government should introduce measures that facilitate entrepreneurs’ access to talent, markets, and finance, as well as invest in in an open government data environment and in the elaboration of an entrepreneurial culture (European Commission 2020). 42 Consultations conducted by a World Bank team during April 2023 confirmed the limited availability of gender- disaggregated data around female entrepreneurship and revealed minimal monitoring and evaluation efforts around this topic. 43 These findings are in line with qualitative evidence gathered during consultations conducted by a World Bank team in April 2023, which revealed that there is a general perception of women-led businesses’ doing better because they are more centered and have clear objectives and drive. 35 References Adikaram, A. S., and R. Razik. 2022. “Femininity Penalty: Challenges and Barriers Faced by STEM Woman Entrepreneurs in an Emerging Economy.” Journal of Entrepreneurship in Emerging Economies. Afandi, E., and M. Kermani. 2015. “Bridging the Gender Gap in Entrepreneurship: An Empirical Analysis.” Journal of Developmental Entrepreneurship 20 (02): 1550008. Arpino, B., and D. Bellani. 2022. “Juggling Grandchild Care and Labor Force Participation: The Effect on Psychological Wellbeing of Older Women.” Frontiers in Sociology 6: 252. Barac, Petre. 2022. “Start-Up Nation Program Is Launched, at its Third Edition.” Thediplomat.ro.com, July 19, 2022. https://www.thediplomat.ro/2022/07/19/start-up-nation-program-is-launched-at-its- third-edition/. Barreira, J. C. D. 2005. “The Influence of Business Knowledge and Work Experience as Antecedents to Entrepreneurial Success.” Doctoral dissertation, University of Pretoria. Bates, T., W.E. Jackson III, and J. H. Johnson Jr. 2007. “Introduction to the Special Issue on Advancing Research on Minority Entrepreneurship.” Annals of the American Academy of Political Science and Social Science 613: 10–17. Bechthold, L. A., and L. Rosendahl Huber, L. 2018. “Yes, I Can!–A Field Experiment on Female Role Model Effects in Entrepreneurship. Academy of Management Proceedings 2018 (1): 12081. Briarcliff Manor, NY: Academy of Management. Brush, C. G., De Bruin, A., & Welter, F. (2009). A gender‐aware framework for women's entrepreneurship. International Journal of Gender and Entrepreneurship. Caliendo, M., F. M. Fossen, A. Kritikos, and M. Wetter. 2015. “The Gender Gap in Entrepreneurship: Not Just a Matter of Personality.” CESifo Economic Studies 61 (1): 202–38. Chowdhury, T. Y., A. Yeasmin, and Z. Ahmed. 2018. “Perception of Women Entrepreneurs to Accessing Bank Credit.” Journal of Global Entrepreneurship Research 8: 1–16. Ciampalini, G., S. Bartoletti, and V. Verma. 2007. “Reporting of Self-Employment Income (EU-SILC 2005), Output I.17, Contract 36401.2006.004-2006.413.” EU-SILC: Methodological Studies to Estimate the Impact on Comparability of the National Methods Used, ISR International. Ciampalini, G., G. Betti, G., and V. Verma. 2009. “Comparability in Self-Employment Income.” Working Paper 82/09, Department of Quantitative Methods, University of Siena. Compton, J., and R. A. Pollak. 2014. “Family Proximity, Childcare, and Women’s Labor Force Attachment.” Journal of Urban Economics 79: 72–90. CONAF. 2023. “About CONAF.” https://conaf.ro/despre-conaf/. Dawson, C. J., A. Henley, and P. L. Latreille. 2009. “Why Do Individuals Choose Self-Employment?” IZA Discussion Paper No 3974. Institute of Labor Economics. Dhaliwal, A. 2022. “The Mompreneurship Phenomenon: An Examination of the Antecedents and Challenges of Mothers in Business.” International Journal of Service Science, Management, Engineering, and Technology 13 (1): 1–17. 36 Dheer, R. J., M. Li, and L. J. Treviño. 2019. “An Integrative Approach to the Gender Gap in Entrepreneurship across Nations.” Journal of World Business 54 (6): 101004. Dutta, N., and S. Mallick. 2018. “Enabling Women Entrepreneurs: Exploring Factors that Mitigate the Negative Impact of Fertility Rates on Female Entrepreneurship.” Kyklos 71 (3): 402–32. East, P. L., T. S. Weisner, and A. Slonim, A. 2009. “Youths’ Caretaking of Their Adolescent Sisters’ Children: Results from Two Longitudinal Studies.” Journal of Family Issues 30 (12): 1671–97. EIGE. 2020. “Gender Equality Index.” https://eige.europa.eu/gender-equality-index/thematic- focus/digitalisation/country/RO. Ekinsmyth, C. 2014. “Mothers’ Business, Work/Life and the Politics of ‘Mumpreneurship.’” Gender, Place & Culture 21 (10): 1230–48. Envick, B. R. 2005. “Beyond Human and Social Capital: The Importance of Positive Psychological Capital for Entrepreneurial Success.” The Entrepreneurial Executive 10 (1): 41–52. European Commission. 2017. Business Angels Funding. https://www.eban.org/wp- content/uploads/2017/12/Final-Report_Understanding-the-Nature-and-Impact-of-the-business- angels-in-Funding-Research-and-Innovation_FV_Formatted_Revised13.12.2017.pdf European Commission. 2020. Start-ups, Scale-ups and Entrepreneurship in Romania. https://ec.europa.eu/research-and-innovation/sites/default/files/rio/report/KI-AX-18-008-EN-N.pdf. European Commission. 2022. European Innovation Scoreboard. Romania. ec_rtd_eis-country-profile- ro.pdf (europa.eu). European Commission. 2021. Methodological Guidelines and Description of EU-SILC Target Variables. https://ec.europa.eu/eurostat/documents/203647/203704/DOC65.pdf/434b2180-33b3-0d8c-ed1e- 2da912d6a685?t=1655461990699. European Parliament. 2017. “A European Pillar of Social Rights European Parliament Resolution of 19 January 2017 on a European Pillar of Social Rights (2016/2095(INI)).” Field, E., S. Jayachandran, and R. Pande. 2010. “Do Traditional Institutions Constrain Female Entrepreneurship? A Field Experiment on Business Training in India.” American Economic Review 100 (2): 125–29. Filippin, A. “Gender Differences in Risk Attitudes.” IZA World of Labor 2022: 100 doi: 10.15185/izawol.100.v2. Fortech Investments. 2022. “Romanian Investors, VCs and Other Funding Sources for Your Start-Up.” fortechinvestments.ro, January 21, 2022. https://www.fortechinvestments.ro/blog/investors-in- romania/. Furdui, A., L. Lupu-Dima, and E. Edelhauser. 2021. “Implications of Entrepreneurial Intentions of Romanian Secondary Education Students, over the Romanian Business Market Development.” Processes 9 (4): 665. G20. 2017. “Towards an Inclusive Future: Shaping the World of Work. Ministerial Declaration, G20 Labour and Employment Ministers Meeting, Bad Neuenahr, 18–19 May 2017.” http://www.g20.utoronto.ca/2017/170519-labour.html. García Johnson, C. P., and K. Otto. 2019. “Better Together: A Model for Women and LGBTQ Equality in the Workplace.” Frontiers in Psychology 10: 272. 37 Gartner, W. B., and S. A. Shane. 1995. “Measuring Entrepreneurship over Time.” Journal of Business Venturing 10 (4): 283–301. Gawel, A. 2021. “The Gender-Based Attainment of Education and Female Entrepreneurship: The European Perspective.” Entrepreneurship and Sustainability Issues 8 (4): 403. GEM (Global Entrepreneurship Monitor). 2022a. Global Entrepreneurship Monitor 2021/2022. Global Report: Opportunity Amid Disruption. London: GEM. GEM (Global Entrepreneurship Monitor). 2022b. Global Entrepreneurship Monitor 2021/22 Women’s Entrepreneurship Report: From Crisis to Opportunity. London: GEM. German Startups Association. 2020. Female Founders Monitor. ffm_2020_en.pdf (startupverband.de) Goel, N., and P. Madan. 2019. “Benchmarking Financial Inclusion for Women Entrepreneurship—A Study of Uttarakhand State of India.” Benchmarking: An International Journal 26 (1): 160–75. Gohmann, S. F. 2012. “Institutions, Latent Entrepreneurship, and Self–Employment: An International Comparison.” Entrepreneurship Theory and Practice 36 (2): 295–321. https://doi.org/10.1111/j.1540- 6520.2010.00406.x. Greulich, Angela, and Aurélien Dasré. 2018. “Observing the Number of Children with EU-SILC: A Quantification of Biases.” Population 73 (4): 685–720. https://www.cairn-int.info/journal-population- 2018-4-page-685.htm. Groszkowska, Katarzyna. 2023. “Top Women From VC and Start-up Ecosystem In Romania.” Vestbee.com, January 31, 2023. https://www.vestbee.com/blog/articles/top-women-from-vc-and- startup-ecosystem-in-romania. Guzman, J., and A. O. Kacperczyk. 2019. “Gender Gap in Entrepreneurship.” Research Policy 48 (7): 1666–80. Halim, Daniel. 2020. “Women Entrepreneurs Needed—Stat!” World Bank Blogs, March 5, 2020. Women entrepreneurs needed—stat! (worldbank.org). Hamilton, B. H. 2000. “Does Entrepreneurship Pay? An Empirical Analysis of the Returns to Self- Employment.” Journal of Political Economy 108 (3): 604–31. Herman, Emilia, and Zsuzsanna Katalin Szabo. 2015. “Gender (In)equality in Entrepreneurship: Challenges for Romania.” In Female Entrepreneurship in Transition Economies, edited by Veland Ramadani, Shqipe Gërguri-Rashiti, and Alain Fayolle, 129–58. Palgrave Macmillan. IDC. 2022. European Women in VC. IDC European Woman in Venture Capital (idcinteractive.net) IFC. 2019. Taking Women Entrepreneurs to the Bank in Romania. https://www.ifc.org/wps/wcm/connect/c4e58de9-5c29-435a-825c-66420ce4376d/GarantiBank- Romania.pdf?MOD=AJPERES&CVID=mz2MFVT. IFC. 2022. “IFC Supports Smaller Businesses and Women Entrepreneurs in Romania to Build Economic Resilience.” https://pressroom.ifc.org/all/pages/PressDetail.aspx?ID=27337. ILO. 2021. “Extending Social Security to Self-Employed Workers: Lessons from International Experience.” ILO Policy Paper. Invest Europe. 2022. 2021 Central and Eastern Europe. Private Equity Statistics. 38 Jardim, J., A. Bártolo, and A. Pinho. 2021. “Towards a Global Entrepreneurial Culture: A Systematic Review of the Effectiveness of Entrepreneurship Education Programs.” Education Sciences 11 (8): 398. Jormalainen, Janne. 2023. “Unlocking Profits with Gender Diversity: EBAN’s Commitment to Angel Investing.” www.eban.org, March 30, 2023. https://www.eban.org/unlocking-profits-with-gender- diversity-ebans-commitment-to-angel-investing/. Kanze, D., L. Huang, M. A. Conley, and E. T. Higgins. 2017. “Male and Female Entrepreneurs Get Asked Different Questions by VCs—And It Affects How Much Funding They Get.” Harvard Business Review, June 27, 2017. https://hbr.org/2017/06/male-and-female-entrepreneurs-get-asked-different- questions-by-vcs-and-it-affects-how-much-funding-they-get Kodithuwakku, S. S., and P. Rosa. 2002. “The Entrepreneurial Process and Economic Success in a Constrained Environment.” Journal of Business Venturing 17 (5): 431–65. Krieger, A., J. Block, M. Stuetzer, M. Obschonka, and K. Salmela-Aro. 2022. “Closing the Gender Gap in Entrepreneurship: The Importance of Skill Variety.” Plos One 17 (7): e0270976. Landour, Julie. 2020. « Mompreneurs », Travail et Emploi [En ligne], Hors-série | 2020, mis en ligne le 01 mars 2021, consulté le 28 avril 2023. http://journals.openedition.org/travailemploi/10041. DOI : https://doi.org/10.4000/travailemploi.10041 Langowitz, N., and M. Minniti. 2007. “The Entrepreneurial Propensity of Women.” Entrepreneurship Theory and Practice 31 (3): 341–64. Leovaridis, C., M. Bahnă, and D. M. Cismaru. 2018. “Between Motherhood and Entrepreneurship: Insights on Women Entrepreneurs in the Creative Industries.” Challenging the Status Quo in Management and Economics 1383. Lokshin, M., and M. Fong. 2006. “Women’s Labour Force Participation and Child Care in Romania.” The Journal of Development Studies 42 (1): 90–109. Low, S., J. Henderson, and S. Weiler. 2005. “Gauging a Region’s Entrepreneurial Potential.” Economic Review-Federal Reserve Bank of Kansas City 90 (3): 61. Markussen, S., and K. Røed. 2017. “The Gender Gap in Entrepreneurship–The Role of Peer Effects.” Journal of Economic Behavior & Organization 134: 356–73. Mastercard. 2023. Mastercard Index of Women Entrepreneurs (MIWE) 2019. mastercard-index-of- women-entrepreneurs-2019.pdf. McCartney, Andrew Charles, Chiaki Yamamoto, Manar Ahmed Zaki Mahmoud Zaki Korayem, Olga Y. Ioffe Kasher, Roshin Mathai Joseph, Sammar Essmat, and Ulugbek Yusufdjanovich Tilyayev. 2016. “Garanti Bank Romania: Striving to Lead—A Study on Women in Banking: A Learning Study on Impact of Banking on Women Programs in Europe, Central Asia, Middle East and North Africa.” World Bank, Washington, DC. https://documents.worldbank.org/en/publication/documents- reports/documentdetail/791761478006389684/learning-study-on-impact-of-banking-on-women- programs-in-europe-central-asia-middle-east-and-north-africa. Meunier, F., Y. Krylova, and R. Ramalho. 2017. “Women’s Entrepreneurship: How To Measure the Gap between New Female and Male Entrepreneurs?” World Bank Policy Research Working Paper 8242, World Bank, Washington, DC. 39 Noguera, M., C. Alvarez, J. M. Merigo, and D. Urbano. 2015. “Determinants of Female Entrepreneurship in Spain: An Institutional Approach.” Computational and Mathematical Organization Theory 21: 341–55. OECD. 2019. “Social Institutions and Gender Index (SIGI).” https://www.genderindex.org. OECD and the European Union. 2020. Inclusive Entrepreneurship Policies, Country Assessment Notes. Romania. Romania (oecd.org). OECD and the European Commission. 2021. The Missing Entrepreneurs 2021: Policies for Inclusive Entrepreneurship and Self-Employment. Paris: OECD Publishing. https://doi.org/10.1787/71b7a9bb- en. Pavlova, E. and Gvetadze, S. 2023 “Female access to finance: a survey of the literature”, EIF Research and Market Analysis, Working Paper 202 3/87. Paun, Cristian. 2012. “Risk Tolerance Analysis: Romanian Case Before and During Financial Turmoil.” Economics & Sociology 5 (2a): 11–23. Pocol, C. B., and C. Moldovan-Teselios. 2014. “Women Entrepreneurship in Romania: A Regional Evaluation.” Bulletin of the University of Agricultural Sciences & Veterinary Medicine Cluj-Napoca. Horticulture 71 (2). Poggesi, S., M. Mari, L. De Vita, and L. Foss. 2020. “Women Entrepreneurship in STEM Fields: Literature Review and Future Research Avenues.” International Entrepreneurship and Management Journal 16: 17–41. Posadas, J., and M. Vidal-Fernandez. 2013. “Grandparents’ Childcare and Female Labor Force Participation.” IZA Journal of Labor Policy 2: 1–20. Richomme-Huet, K., V. Vial, and A. d’Andria, A. 2013. “Mumpreneurship: A New Concept for an Old Phenomenon?” International Journal of Entrepreneurship and Small Business 19 (2), 251–75. Richomme‐Huet, K., and V. Vial. 2014. “Business Lessons from a ‘Mompreneurs’ Network.” Global Business and Organizational Excellence 33 (4): 18–27. Robayo-Abril, M. and B. Rude. Forthcoming. Preparatory School Years and Maternal Employment in Romania. Ubfal, Diego Javier. 2023. “What Works in Supporting Women-led Businesses?” World Bank Group Gender Thematic Policy Note, World Bank, Washington, DC. https://openknowledge.worldbank.org/entities/publication/be7df346-4062-5ebc-848b- d80d3a316381. Van Kline, Maia. 2022. “RO Govt’s Call for Projects to Enter the Woman Entrepreneur Program Begins on August 18.” Romania-Insider.com, August 12 2022. https://www.romania-insider.com/call- projects--woman-entrepreneur-program-ro-aug-2022. Verheul, I., R. Thurik, I. Grilo, and P. Van der Zwan. 2012. “Explaining Preferences and Actual Involvement in Self-Employment: Gender and the Entrepreneurial Personality.” Journal of Economic Psychology 33 (2): 325–41. Vermeylen, Greet, Mathijn Wilkens, Isabella Biletta, Andrea Fromm, Agnès Parent-Thirion, and Ricardo Rodriguez Contreras. 2017. Exploring Self-Employment in the European Union. European Foundation for the Improvement of Living and Working Conditions. 40 Wegate (European Gateway for Women’s Entrepreneurship). 2023. “High-Level Cooperation Promotes Romanian Business Excellence.” https://www.wegate.eu/high-level-cooperation- promotes-romanian-business-excellence/. World Bank. 2020. Doing Business 2020. https://www.doingbusiness.org/content/dam/doingBusiness/country/r/romania/ROM.pdf. Yordanova, D. I., and M. A. Tarrazon. 2010. “Gender Differences in Entrepreneurial Intentions: Evidence from Bulgaria.” Journal of Developmental Entrepreneurship 15 (03): 245–61. 41 Appendix Table A.1. Probability of working women in households with children being self-employed, 2020 (1) Variables Self-employed Age -9.26e-05*** (2.67e-05) No schooling 0.149*** (0.00296) Primary educ. -0.0256*** (0.00136) Secondary educ. 0.0173*** (0.000696) Tertiary educ. 0.0126*** (0.000993) Urban -0.0125*** (0.000511) Has access to childcare = 1 0.00117** (0.000459) Income quintile 1 0.0626*** (0.00102) Income quintile 2 0.117*** (0.000952) Income quintile 3 0.0227*** (0.00112) Income quintile 4 0.0304*** (0.00107) Primary sector 0.148*** (0.00121) NACE B–E sectors -0.161*** (0.00145) Construction -0.108*** (0.00216) Wholesale and retail; repair -0.114*** (0.00126) Accommodation and food -0.114*** (0.00205) NACE L–N sectors -0.0941*** (0.00168) Education -0.122*** (0.00174) Human health and social -0.0605*** (0.00134) Observations 676 Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. 42 Table A.2. Probability of working women in households with children being self-employed (with self-employed restricted to employers with employees), 2020 (1) Variables Self-employed (with employees) Age 0.000799*** (7.20e-06) Secondary educ. -0.0159*** (0.000236) Tertiary educ. 0.00574*** (0.000195) Urban -0.00649*** (0.000155) HH with child (<16) = 1 -0.00266*** (0.000140) Has access to child care = 1 0.00711*** (0.000246) Income quintile 1 0.00358*** (0.000327) Income quintile 3 -0.00895*** (0.000226) Income quintile 4 -0.0122*** (0.000186) Primary sector 0.0105*** (0.000375) NACE B–E sectors -0.00782*** (0.000446) Construction 0.0331*** (0.000393) Wholesale and retail; repair 0.0235*** (0.000302) Accommodation and food 0.0320*** (0.000378) NACE L–N sectors 0.00437*** (0.000374) Public admin, defense, social security -0.00350*** (0.000463) Education 0.00651*** (0.000350) Observations 2,455 Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: The table presents the marginal effects of a probit model. The outcome variable is an indicator variable that is equal to one if a woman is self-employed and employs employees and zero otherwise. We restrict the sample to working women. Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. 43 Table A.3. Probability of working mothers in households with children being self-employed, 2020 (1) Variables Self-employed (mothers) Age 0.00152*** (9.74e-06) No schooling 0.134*** (0.00162) Primary educ. 0.0298*** (0.000605) Secondary educ. -0.00339*** (0.000325) Tertiary educ. -0.000383 (0.000417) Urban -0.0445*** (0.000211) Mother = 1 -0.0576*** (0.000342) Income quintile 1 0.108*** (0.000346) Income quintile 2 0.0482*** (0.000322) Income quintile 3 -0.00542*** (0.000331) Income quintile 4 -0.0171*** (0.000333) Primary sector 0.132*** (0.000549) NACE B–E sectors -0.183*** (0.000597) Construction 0.00605*** (0.000559) Wholesale and retail; repair -0.105*** (0.000568) Transport and storage -0.128*** (0.000676) Accommodation and food -0.0859*** (0.000843) ICT -0.191*** (0.00132) Finance and Insurance -0.122*** (0.00127) NACE L–N sectors -0.128*** (0.000733) Public admin, defense, social security -0.311*** (0.00177) Education -0.198*** (0.00103) Human health and social -0.111*** (0.000744) Observations 367 44 Source: Own estimates based on EU-SILC, European Union Statistics on Income and Living Conditions, European Union statistics on Income and living conditions - Microdata - Eurostat (europa.eu). Note: The table presents the marginal effects of a probit model. The outcome variable is an indicator variable that is equal to one if a worker is self-employed and zero otherwise. Standard errors are in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1. 45