Report No. 102015-TR PERFORMANCE OF FEMALE EMPLOYERS IN TURKEY December 2015 Performance of Female Employers in Turkey Tolga Cebeci1 1 The World Bank, tcebeci@worldbank.org. I thank very much to Mediha Agar, Ana Maria Munoz Boudet, Carolina Sanchez-Paramo and Cagla Okten for providing very useful comments on the draft of this paper. All remaining errors are mine. The findings expressed in this paper are those of the author and do not necessarily represent the views of the World Bank or the Ministry of Family and Social Policies. 1 Performance of Female Employers in Turkey © 2015 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org Standard Disclaimer: This volume is a product of the staff of the International Bank for Reconstruction and Development/The World Bank, and it has been produced as part of the “Increasing Women’s Acccess to Economic Opportunities in Turkey” Project (2012-2017) which is funded by the Swedish International Development Cooperation Agency (SIDA) and implemented by the Ministry of Family and Social Policies (MoFSP) of the Republic of Turkey and the World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of MoFSP, SIDA, the Executive Directors of The World Bank 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. Copyright Statement: The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development/ The World Bank and the Ministry of Family and Social Policies (MoFSP) of the Republic of Turkey encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. 2 Performance of Female Employers in Turkey TABLE OF CONTENTS 1. INTRODUCTION 4 2. CHARACTERISTICS OF FEMALE EMPLOYERS 5 2.1 INCIDENCE OF BEING A FEMALE EMPLOYER 5 2.2 BASIC CHARACTERISTICS OF FEMALE EMPLOYERS 6 2.3 SURVIVAL ANALYSIS 8 2.4 THE PATH TO BECOME AN EMPLOYER 9 3. INCOME COMPARISON OF MALE AND FEMALE EMPLOYERS 10 4. DISTINGUISHING FEATURES OF SUCCESSFUL FEMALE EMPLOYERS 14 5. CONCLUSION AND POLICY RECOMMENDATIONS 15 REFERENCES 17 APPENDIX 19 A - INCOME COMPARISON OF MALE AND FEMALE EMPLOYERS 19 B- DISTINGUISHING FEATURES OF SUCCESSFUL FEMALE EMPLOYERS 20 3 Performance of Female Employers in Turkey 1. INTRODUCTION 1. It is well documented that economic participation of women is very low in Turkey compared to that of men in the country2. This gender disparity in economic participation is valid not only for participation in the labor market as a wage employee but also for entrepreneurship. Although women lag behind men in entrepreneurship in most countries3, gender disparity in entrepreneurship is particularly large in Turkey. 2. To illustrate, in none of the 44 developed and developing countries (excluding Turkey) analyzed by Estrin and Mickiewicz (2011) using Global Enterprise Monitor survey, female/male start-up ratio are lower than 25%. In 14 of them the ratio is between 25% and 40% and in the remaining group it is above 40%. In contrast, Okten (2013) document that only 7.4% of employers and 17.2% of own-account workers (OAW) were women in 2012 in Turkey. 3. Using European Union Statistics on Income and Living Conditions (SILC) dataset, this paper attempts to provide new insights on the high gender disparity in entrepreneurship in Turkey with giving special emphasis to distinguishing characteristics of successful female employers. SILC dataset covers the 2007-2010 period. Agriculture sector is excluded in the study to identify entrepreneurship patterns in non-agriculture more precisely as patterns in agriculture is usually very different from other sectors. The final dataset includes 49,035 observations of 17,202 individuals. A unique feature of the dataset is that it is a panel dataset. As an individual can be followed over 2-4 consecutive years, it is possible to monitor transformations of individuals across activities. Although entrepreneurs refer to both employers and OAW, this paper focuses on employers due to its importance over OAW in creating jobs and increasing the impact of women in the economy4. All tables and figure in this paper make use of the SILC dataset. 4. The paper documents that female employers are in minority in Turkey not only among all working women but also among all employers in the economy. Moreover, female employers earn less than their male counterparts and their firms are generally smaller. As for their background, there is a low churning among female employers and majority of new female employers are transformed from inactivity and very few from wage employment. 5. The paper confirms the importance of university education in closing gender gap. Proportional income gap between male and female employers is lowest among university graduates and highest among primary school and vocational high school graduates. It is also noteworthy that female employers are less educated than female full-time employers but male employers are more educated than full-time male employees. 2 See for example Hausmann et al. (2012). 3 See for example Estrin and Mickiewicz (2009), Georgellis and Wall (2005), Kim (2007), Minniti and Naude (2010) , De Bruin et al. (2006), Minniti et al. (2005), Reynolds et al. (2002), Verheul et al. (2006). 4 Okten (2013) show that employers and OAW have important differences. 90% of female and 60% of male OAW work informally whereas only 21% of female employers and 20% of male employers are in the informal sector. 4 Performance of Female Employers in Turkey 6. Besides, average tenure of female employers is significantly less than that of their male counterparts and this has a significant role in female employers’ earnings being lower than earnings of male employers. Finally, it is found that female employers are mainly concentrated in less profitable retail and wholesale trade, manufacturing industry and other services sectors including washing and cleaning of clothes as well as beauty services. 7. The remainder of the paper is organized as follows: Section II first attempts to identify basic characteristics of female employers in comparison to male employers and other female work groups. The section concludes with survival and transformation analysis. Based on the insight obtained in Section II, Section III evaluates the success of female employers in terms of income they generate. As the second step, income of female employers belonging to different sector, educational attainment and background categories are compared to each other. Finally, Section IV discusses results and concludes. 2. CHARACTERISTICS OF FEMALE EMPLOYERS 8. This section first investigates the prevalence of being a female employer by providing the share of female employers among all working women and among all employers. Second, basic characteristics of female employers are given in comparison with those of male employers and other working female groups. Third, survival rates of female employers are analyzed. And finally, previous positions of female employers are studied to shed light on the path leading to be an employer. 2.1 INCIDENCE OF BEING A FEMALE EMPLOYER 9. Female employers are in minority in Turkey not only among all working women but also among all employers in the economy. In contrast to a moderate 7% employer rate among working men, only 2% of all working women were employers in 2010 (Figure 1). On the other hand, 16% of working men and 11% of working women were Own Account Workers (OAW). In addition to that full-time wage employment is more popular among women than among men. Around 3/4th of all working women are full-time wage employees in contrast to 2/3rd of all working men. 5 Performance of Female Employers in Turkey 10. Over the 2007-2010 period, only 7-8% of all employers were women, although female share in all working people in the economy was 23-24% (Figure 2). As for other work types, share of women in total OAW in the economy showed an increase from 12% to 18% over the period. Besides, around 25- 27% of all full-time wage employees and 19-23% of all part- time wage employees in the economy were women5. 2.2 BASIC CHARACTERISTICS OF FEMALE EMPLOYERS 11. Educational attainment of employers, full-time wage employees and OAW by gender are given in Figure 3 for the 2007-2010 period. Share of those with a higher education (university) degree among female employers is higher than that among own- account female workers but lower than that among full-time female employees. In contrast, share of female that have at highest a primary school degree6 among all female employers is less than that among own-account female workers but higher than that among full-time wage employees. 25% of female employers have higher education degrees and 38% have primary school degrees at highest. 12. On the other hand, female employers are slightly more educated than their male counterparts. Share of higher education is around 10% lower whereas share of primary school is around 9% higher among male employers. It is noteworthy that female employers are less educated than female full-time employers but male employers are more educated than full-time male employees. This finding for Turkey does not confirm Brush (2006), which states that both male and female entrepreneurs are more educated than non-entrepreneurs. 5 The share of female among all employers found in this paper is consistent with the one (7.4%) documented in Okten (2013). On the other hand, the share of female among all OAW in this paper is slightly less than the one (17.2%) documented in Okten (2013). Exclusion of agriculture from this study accounts for this difference. 6 Primary school category includes: a) those completed 5-year primary school b) those who attended but did not complete primary school c) those who did not attend school at all. 6 Performance of Female Employers in Turkey 13. We next compare tenure7 and age of men and women by work type. Average tenure of female employers is significantly less than that of their male counterparts. This finding is consistent with those documented in Brush (1992), Carter et al. (1997), Boden and Nucci (2000), Kepler and Shane (2007), Fairlie and Robb (2009), Aronson (1991), Lee and Rendall (2001). The tenure gap between male and female in favor male is valid for OAW and full-time wage employees as well. Female employers and OAW slightly more tenured than female full-time wage employees. Table 1: Tenure and Age (years) by Work Type and Gender Wage Employee (ft) Employer OAW Men Women Men Women Men Women Tenure 16.0 9.2 21.7 12.3 22.4 11.7 Age 36.4 32.2 41.8 36.9 42.2 39.1 14. In parallel to the pattern observed for tenure, female, on average, are younger than the male within all 3 work types. On the other hand, difference between male and female average age is less than the difference in average tenure of male an female within all work types. Difference in relative tenure and age is particularly stark among OAW. Although female on average are 3.1 years younger than the male in this group, they have 10.7 years less tenure. This hints that either female start to work at later ages and/or give a break at some point in their career. Regarding marital status, share of never-married among female employers is higher than that among female OAW but lower than that among full-time female wage employees. The opposite ranking is valid for the share of married. Compared to men, share of married among women is lower for all three groups of work type. 15. As for sector composition, most of female employers operate in one of 3 sectors: 29% in retail and wholesale trade, 28% in manufacturing and 25% in other services sectors (Figure 5). Other services sectors mainly consist of washing and cleaning of clothes and beauty treatment such as hairdressing. Compared to their male counterparts, female employers are more concentrated in washing, cleaning and beauty activities, slightly more in manufacturing, slightly less in trade (sales) services and less in food and accommodation services, professional business services as well as construction and mining. This pattern in Turkey is somewhat different than the findings in CWBR (2008), which documents that female entrepreneurs at the global level are over- represented in sales, retail, and services. The observed concentration of female employers in 7 Tenure refers to total number years of current and previous work experience a person has. 7 Performance of Female Employers in Turkey Turkey in manufacturing sector is believed to be due to high share of textile and clothing sector in overall manufacturing industry in Turkey8. 16. Compared to sector composition of full-time female wage employees, female employers are equally concentrated in manufacturing, more concentrated in trade, cleaning, washing and beauty services. In contrast, they are less concentrated in health, education and other business services. Finally, compared to female OAW, female employers operate less in manufacturing but slightly more in education, food and accommodation and other services (washing, cleaning, hairdressing etc.) 2.3 SURVIVAL ANALYSIS 17. Previous Section clearly documented that share of female among all employers is very low in Turkey. One question that would be useful to investigate for the purpose of policy is that whether low share of female among employers is the result of low entry or rather high exit of female? A low entry-low exit combination may be indicative of high barriers whereas a high entry-high exit combination indicates high rate of failure. 18. Table 2 compares entry and exit rates of male and female employers over the 2007-2010 period. Entry rate for female employers is defined as the number of women that become employer in the current year divided by the total number of female employers in that year. Accordingly, exit rate is defined as the number of female employers that stop being employer in the current year divided by the total number of female employers in the year before. According to the Table, both entry and exit rates of female employers is lower than those of male employers. This finding supports Kalleberg and Leicht (1991) and Perry (2002) in that women’s failure rates are not much higher than those of men, contrary to the popular opposite belief. This finding implies an entry barrier problem in Turkey for female employers rather than higher failure of them. Entry rate for both male and female employers is lower than their corresponding exit rates, possibly reflecting the tough economic conditions during the period. 8 Unfortunately, sector information in SILC is not detailed enough to compute sector statistics at finer levels. 8 Performance of Female Employers in Turkey Table 2: Entry and Exit Rates (%) of Employers Gender Entry Rate Exit Rate Men 13 16 Women 8 12 19. Average (mean) tenure of employers are compared in Table 3 by gender and employer status. It is striking that tenure of female becoming employer is much lower than that of male becoming employer (7.3 vs. 20.7 years). That is, women become employer much earlier than men in their career. Second, duration of survival (average tenure of exitters – average tenure of entrants) is much higher for female (6.5 vs. 2.3 years). Table 3: Average Tenure of Entrant and Exitter Employers Men Women All Become Exit Being All Become Exit Being Employer Employer Employer Employer Employer Employer 21.7 20.7 23.0 12.3 7.3 13.9 2.4 THE PATH TO BECOME AN EMPLOYER 20. The panel structure of the dataset allows us to analyze what employers were doing before they become employers and where they end up in the next year. Table 4a demonstrates transition of women among work categories by backward looking. That is, it gives origins of female in a specific category in the current year. It tells what were 100 female in a specific category in the current year were doing the year before. According to the Table, there is a very high correlation between activities of individuals in a year and the previous year. Out of 100 female employers in a year; 90 of them were employers, 1.9 were wage employee, 2.4 were OAW and 5.7 were inactive/unemployed in the previous year. As for 100 OAW in the current year, 62.8 of them were also OAW workers in the previous year, 28.4 of them were inactive/unemployed, 7.7 of them were wage employees and 1 of them was employer. The correlation between current and previous year activitiy is stronger for employers than wage employees and OAW. In other words, churning among female employers is less than the churning among other female work categories. Table 4a: Transition Matrix: Origins t-1 Wage Emp. Employer OAW Other Total Wage Emp. 77.1 0.1 0.4 22.4 100.0 Employer 1.9 90.0 2.4 5.7 100.0 t OAW 7.7 1.0 62.8 28.4 100.0 Other 2.7 0.0 0.5 96.9 100.0 9 Performance of Female Employers in Turkey 21. Table 4b demonstrates transition of women among work categories by forward looking. It responds to the question: What 100 female in a specific category in the previous year are doing now? Out of 100 female employers in the previous year; 86 continue to be employer, 4.2 switch to wage employee, 4.5 switch to OAW and 5.3 become inactive/unemployed in the current year. It is striking that only 0.6% of OAW and 0.1% of wage employees in a year becomes employer in the next year. This supports previous findings in the literature that incentives for male and female entrepreneurs are different (Berglann et al., 2009; Estrin and Mickiewicz, 2011). Female become entrepreneur due to necessity rather than to earn more. Female who has a secure job does not show interest in becoming employer, unlike the case for men. Table 4b: Transition Matrix: Future t-1 Wage Employee Employer OAW Other Wage Employee 81.7 4.2 3.4 3.6 Employer 0.1 86.0 0.6 0.0 t OAW 1.0 4.5 69.6 0.5 Other 17.3 5.3 26.4 95.8 Total 100.0 100.0 100.0 100.0 3. INCOME COMPARISON OF MALE AND FEMALE EMPLOYERS 22. This section compares income of male and female employers. It should be noted that income analysis in this paper are far from being ideal tools to compare success of male and female employers due to absence of information on many variables (e.g. access to finance) that have a potential effect on employer success. 23. As an introduction to this section, size of companies that men and female working as either full-time wage employee and employer are given in Figure 6. Companies owned by women are usually smaller than those owned by men. This is consistent with findings of DuReitz and Henrekson (2000), Coleman (2007) and Okten (2013) for Turkey. 97% of all female-owned companies employ less than 11 employees. Ratio of small size companies among all male-owned companies is 88%. On the other hand, share of women working as full-time wage employee in small companies is slightly less than the share of men working in those companies. 10 Performance of Female Employers in Turkey 24. As for income, in all categories of comparison, women, on average, earn less than men (Table 6). The proportional difference between male and female earnings is highest among OAW, followed by employers and lowest among full-time wage employees. While mean income of full- time wage employee women is 83% of that of men, female employers on average earn 62% of that of male employers and female OAW on average earn 45% of male OAW. Turkey is no exception to the consensus in the literature that women entrepreneurs earn less than male entrepreneurs (Klapper and Parker, 2010; Robb and Wolken, 2002; Chaganti and Parasuraman, 1996; Minniti, 2009). 25. Among working female, employers earn 60% more than full-time wage employees and more than twice as much as OAW. Income gap between employers and OAW is smaller for men because male OAW earn much higher than female OAW. Male OAW earn almost as much as male full-time wage employees. Table 6: Average Income (Annual, 2007 TL) by Employment Type Work Type Gender Mean Income Median Income Wage Employee Men 11,309 9,212 (full-time) Women 9,327 7,513 Men 24,988 15,354 Employer Women 15,367 9,700 Men 10,879 8,530 OAW Women 4,845 2,559 26. Figure 7 depicts income distribution of entrepreneurs for a more complete income comparison between male and female entrepreneurs. Male employers earn the highest, followed by female employers and then male OAW. Female OAW ranks the last by a large margin. 11 Performance of Female Employers in Turkey 27. An important question to address the role of gender in employer income is whether differences in educational attainment of male and female employers play a role in difference in average income of these two groups. Table 7 shows average income of employers by gender and educational attainment. Among male employers two educational groups emerge: Being the more earning group, higher education, regular high school and vocational high school graduates earn around 50% more than the less earning group: middle and primary school graduates. 28. As for women, the picture is quite different. Although female employers with higher education degrees earn the highest, female employers with a vocational school degree earn the lowest, unlike the case for male employers. Besides, female employers with primary school degrees earn considerably less than those with middle school degrees. As a result, income gap between male and female employers is lowest among university graduates and highest among primary school and particularly vocational high school graduates, all in favor of men. Table 7: Average Income (Annual, 2007 TL) of Employers by Education Primary Middle High Vocational Higher Measure Gender Overall School9 School School H. S. Educ. Men 20,144 20,667 31,454 28,670 32,639 24,988 Mean Income Women 7,113 12,609 21,167 5,075 26,636 15,367 Median Men 13,832 13,900 20,000 15,238 24,060 15,354 Income Women 7,000 9,143 14,074 4,180 22,517 9,700 29. In order for a formal treatment of the income gap between male and female employers, employer income is run on female dummy in 5 specifications of the form given in Appendix A. Results of specification (A1) confirm the extent of the previously observed income gap between male and female employers. Average female employer earns around 61% of average male employer. While interpreting this result, it should be remembered that average female employer has much lower tenure than the average male employer. Specification (A2) is run by adding tenure and interacting tenure with female dummy. According to results of this Specification, average income gap between a male and a female employer without any tenure is 5,069 TL. Each year of tenure contributes 269 TL to the income of a male employer but only 98 TL to the income of a female employer. Since tenure is observed to be an important factor in explaining the income gap between male and female employers, it is kept in the following specifications. 30. Specification (A3) includes same variables as in Specification (A2), but it is run within educational attainment groups. Therefore, it compares the income gap between male and female employers that have the same education. In the same manner, Specification (A4) measures the income gap between male and female employers operating in the same sector and finally Specification (A5) compares the income gap between male and female employers that have the same education and operating in the same sector. 9 Primary school category includes those with primary school degrees, those did not complete primary school and illeterate. Those without a primary school degree have a 2% share among male employers and less than 1% among female employers. 12 Performance of Female Employers in Turkey 31. Figure 8 is depicted to illustrate the results of Specifications (A2)-(A5). Due to unequal contribution of additional year of tenure to the income of male and female employers, the income gap depends on the level of tenure of compared employers. Specification (A1) indicates that income of average female employer is 61% of that of average male employer. Specification (A2) compares income of male and female employers with same tenure. At 0 years of tenure, female employers earn 74% of male employers. This proportion declines to 59% at 45 years of tenure. Since proportions generated by Specification (A2) are higher than those generated by Specification (A1) until around 35 years of tenure, it can be concluded that female having less tenure than male counterparts have a significant role in average female employer’s earnings being less than earnings of average male employer. 32. Results of Specification (A3) is not much different than those of Specification (A2). This is because educational composition of male and female employers are not very different. On the one hand, share of university graduates is 10% higher among female employers. On the other hand, share of vocational high school graduates is 5% higher among male employers. Given that male employer who have vocational high school graduates, on average, earn multiple times of female employers of same education and even more than university graduate female employers, these two counter effects cancel out each other. 33. Proportions generated by Specification (A4) follows almost a stagnant path and are increasingly higher than those generated by Specification (A2). Since, Specification (A4) compares income of male and female employers operating in the same industry, it can be inferred that female employers being concentrated in less profitable sectors contributes significantly to the income gap between male and female employers. Finally, Specification (A5) generates the highest proportions for low levels of tenure but proportions decrease rapidly as tenure increases. Since the decrease rate is similar to the rates of decline in Specifications (A2) and (A3), proportions generated by Specification (A5) stays well above those generated by Specifications (A2) and (A3). Specification (A5) also indicates that female employers are concentrated in less profitable sectors compared to their male counterparts. This supports Klapper and Parker (2010) in that sectors women concentrated are characterized by lower average returns. 13 Performance of Female Employers in Turkey 4. DISTINGUISHING FEATURES OF SUCCESSFUL FEMALE EMPLOYERS 34. In this section, income of female employers of different characteristics are compared to understand the distinguishing features of successful female employers. In order to reveal the role of tenure, education, sector operated and previous work type in explaining the income differential among female employers, a specification of the form given in Appendix B is run on the dataset including only female employers. 35. Specification (B1) investigates the role of educational attainment on female employer income. Results of this specification imply that female employers with university degrees earn significantly more than other female employers of same tenure. Higher education is followed by high school. Graduates of middle school earn significantly less than high school graduates. Middle school graduates are followed by primary school graduates and vocational high school graduates earn the lowest. In fact, university graduate female employers earn more than 5 times than those with primary school or vocational high school degrees. 36. Specification (B2) attempts to reply the same question for sectors. Variation of female employer income is very high across sectors. Female employers in health services earn the highest. This group is followed by other business services (e.g. real estate and technical consultancy services), trade services and communication services. Female employers operating in other services (cleaning and washing of clothes and beauty services), accommodation and food services and manufacturing earn the lowest. Average income of female employers in manufacturing is around 10% of those in heath services. 37. In Specification (B3) both education and sectors are controlled. This results in a decrease in all sector coefficients, which are measuring income premium of respective sectors over manufacturing income. This implies that graduates of the lowest earning educational groups (i.e. vocational h. s. and primary school) are concentrated in manufacturing (and to a lesser extent in accommodation and food services since its coefficient does not decrease much). On the other hand, positive premium of communication services and education services observed in Specification (B2) almost disappear in Specification (B3). Therefore, it can be concluded that these two sectors are not associated with higher incomes than that in manufacturing if female employers of same education and tenure are compared. 38. Finally, Specification (B4) is run to measure the role of previous year work type on the current year employer income of female. Results of this specification document that female employers who were also employers in the previous year earn more than female employers who were OAW or who were not working at all in the previous year. 14 Performance of Female Employers in Turkey 5. CONCLUSION AND POLICY RECOMMENDATIONS 39. Using a novel dataset that allows to follow the same individual over respective years, this paper attempted to understand the differences in basic characteristics of male and female employers, differences in income of male and female employers and distinguishing features of successful female employers in Turkey. Key findings and associated policy implications are as follows: 40. Female employers are in minority in Turkey not only among all working women but also among all employers in the economy. This is believed to be due to high entry barriers specifically targeting women rather than higher failure rate of female employers than their male counterparts. Although female employers earn less than their male counterparts and their firms are generally smaller, proportional income difference between male and female employers is lower than the one between male and female OAW. This implies that focusing on policies that will help women become employers beyond OAW may yield better results in achieving gender equity in economic participation. 41. Unlike the case for men, female do not become employers from wage employment. This can be considered bad news as it implies that female become employer due to necessity instead of an offensive entrepreneur mindset. Most of new female employers were previously OAW and some of them were house workers. Moreover, a significant part of OAW are previously non- working people. These support previous findings that women engage in entrepreneurial activity mainly due to necessity. 42. Education matters again: Proportional income gap between male and female employers is lowest among university graduates and highest among primary school and vocational high school graduates. Female employers with university degrees earn more than 5 times than those with primary school or vocational high school degrees do. It is also noteworthy that female employers are less educated than female full-time employers but male employers are more educated than full-time male employees. These findings also point the necessity motivated entrepreneurship among women. Therefore, entrepreneurship policies should more target more educated and wage employee women to create more successful female employers. 43. Average tenure of female employers is significantly less than that of their male counterparts. Female having less tenure has a significant role in female employers’ earnings being lower than earnings of male employers. This pose a problem for entrepreneurship because female with less tenure can be expected to be lacking both entrepreneurship related skills and also financial power required for a strong start-up. So, there should be significant benefits associated with the increase in the availability of external financial sources for the use of female entrepreneurs10. Financial supports and credits to women can be conditioned on participating training programs for entrepreneurship to close the skill gap between male and female employers. 10 See Klapper et al. (2006), Evans and Leighton (1989) and Hurst and Lusardi (2004) for the relation between entrepreneurial behavior and financial capability of individuals. 15 Performance of Female Employers in Turkey 44. Regression results reveal that, sectors in which female employers concentrated (retail and wholesale trade, manufacturing, washing and cleaning of clothes and beauty services such as hairdressing) are less profitable than others. In addition, female employers operating in health services, real estate and technical consultancy services earn higher than female employers that have the same education and tenure but operating in other sectors. Hence, providing support to female start-ups in more profitable sectors over start-ups in other sectors can yield better outcomes. 16 Performance of Female Employers in Turkey REFERENCES Aronson, R. (1991). “Self-employment: A Labour Market Perspective,” Ithaca, NY: ILR Press. Berglann, H., E. Moen, K. Roed and J. Skogstrom (2009). “Entrepreneurship: Origins and Returns, ” IZA Discussion Paper 4250. Boden, R., and A. Nucci (2000). “On the Survival Prospects of Men’s and Women’s New Business Ventures,” Journal of Business Venturing, 15 (4): 347–62. Brush, C. (1990). “Women and enterprise creation: Barriers and opportunities,” Paris: Organization for Economic Cooperation and Development. Brush, C. (2006). “Women Entrepreneurs: A Research Overview,” In: M. Casson, B. Yeung, A. Basu, and N. Wadeson, eds., Oxford Handbook of Entrepreneurship. Oxford UK: Oxford University Press. Carter, N., M. Williams, and P. Reynolds (1997). “Discontinuance among New Firms in Retail: The Influence of Initial Resources, Strategy, and Gender,” Journal of Business Venturing, 12 (2): 125 - 45. Chaganti, R., and S. Parasuraman (1996). “A Study of the Impacts of Gender on Business Performance and Management Patterns in Small Businesses,” Entrepreneurship Theory and Practice, 21 (2):73 – 5. Coleman, S. (2007). “The role of human and financial capital in the profitability and growth of women- owned small firms,” Journal of Small Business Management, 45 (3): 303-319. CWBR (2008). “Key Facts About Women-Owned Businesses,” Washington DC: Center for Women’s Business Research. De Bruin, A., C. G. Brush, and F. Welter (2006). “Introduction to the special Issue: Towards building cumulative knowledge on women’s entrepreneurship,” Entrepreneurship Theory and Practice, 30 (5): 585-593. DuReitz, A. and M. Henrekson (2000). “Testing the female underperformance hypothesis,” Small Business Economics, 14: 1-10. Estrin, S., and T. Mickiewicz (2011). “Do Institutions Have a Greater Effect on Female Entrepreneurs?,” Small Business Economics, 37 (4): 397-415. Evans, D., and L. Leighton (1989). “Some Empirical Aspects of Entrepreneurship,” American Economic Review, 79 (3): 519 - 535. Fairlie, R., and A. Robb (2009). “Entrepreneurship, Self-employment and Business Data: An Introduction to Several Large, Nationally-representative Datasets,” IZA Discussion Paper 4052. Georgellis Y., and H. Wall (2005). “Gender differences in self-employment,” International Review of Applied Economics, 19 (3): 321-342. 17 Performance of Female Employers in Turkey Hausmann, R., L. D. Tyson, and S. Zahidi (2012). “The Global Gender Gap Report,” World Economic Forum. Geneva, Switzerland. Hurst, E., and A. Lusardi (2004). “Liquidity Constraints, Household Wealth and Entrepreneurship,” Journal of Political Economy, 112 (2): 319-347. Kalleberg, A., and K. Leicht (1991). “Gender and Organizational Performance: Determinants of Small Business Survival and Success,” Academy of Management Journal, 34 (1): 136–61. Kim, G. (2007) “The analysis of self-employment levels over the life-cycle”, Quarterly Review of Economics and Finance 47(3): 397. Klapper, L., L. Laeven, and R. Rajan (2006). “Entry Regulation as a Barrier to Entrepreneurship,” Journal of Financial Economics 82 (3): 591–629. Klapper, L., and S. Parker (2010). “Gender and business environment for new firm creation” World Bank Research Observer, 26 (2): 237-257. Lee, M. A., and M. S. Rendall (2001). “Self-employment disadvantage in the working lives of blacks and females,” Population Research and Policy Review, 20: 291–320. Minniti, M., P. Arenius, and N. Langowitz (2005). “The 2004 Global Entrepreneurship Monitor Special Topic Report: Women in Entrepreneurship,” Center for Women Leadership, Babson College, MA. Minniti, M. (2009). “Gender Issues in Entrepreneurship,” Foundations and Trends in Entrepreneurship, 5 (7-8): 497–621. Minniti, M., and W. Naude (2010). “What do we know about the patterns and determinants of female entrepreneurship across countries?,” European Journal of Development Research, 22 (3): 277-93. Okten C. (2013). “Female Entrepreneurship in Turkey: Patterns, Characteristics and Trends”. Perry, S. (2002). “A comparison of failed and non-failed small businesses in the United States,” Journal of Developmental Entrepreneurship, 7: 415–428. Reynolds, P., W. Bygrave, E. Autio, L. Cox, and M. Hay (2002). Global Entrepreneurship Monitor: 2002 Executive Report. Robb, A., and J. Wolken (2002). “Firm, Owner, and Financing Characteristics: Differences between Female and Male-owned Small Businesses,” FRBG Working paper. Verheul, I., A. V. Stel, and R. Thurik (2006). “Explaining female and male entrepreneurship across 29 countries,” Entrepreneurship and Regional Development, 18 (2): 151-183. 18 Performance of Female Employers in Turkey APPENDIX A - INCOME COMPARISON OF MALE AND FEMALE EMPLOYERS Employer income is run on female dummy in 5 specifications of the form: E m t i  p0 I l 1F n eti  cm2T o e a  i m t n 3Fl e uee rT i * t m e a i n et  iEul id e t  r e uiS i  ce t t c (A i t) o r i i where i denotes individuals and t year. Female takes 1 if the individual is a female and 0 if   otherwise. Educ and Sector are Education and Sector dummies. Table A: Regression Results All coefficients significant at 0.1%. 19 Performance of Female Employers in Turkey B- DISTINGUISHING FEATURES OF SUCCESSFUL FEMALE EMPLOYERS In order to reveal the role of tenure, education, sector operated and previous work type in explaining the income differential among female employers, a specification of the form given below is run on the dataset including only female employers. E m t i p  I 0 l 1Tn e ci  t  noiE u mr di  t u e iS c et e it i c P i t r o e i r v (B) i i i Table B: Regression Results   **: Significant at 5% only; All others significant at 0.1%. 20