77508 MONGOLIA:GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS This policy note was prepared by a team consisting of Tehmina Khan, Economist EASPR (TTL) and Rogier Van Den Brink, Lead Economist EASPR (co-TTL); Monazza Aslam (Consultant) Oyunbileg Baasanjav, Operations Officer, EACMF; Munkhnasaa Narmandakh, Economist EASPR; Andy Mason, Lead Economist EASPR; Trang Van Nguyen, Economist EASPR; and Altantsetseg Shiilegmaa, Economist, EASPR. Tehmina Khan and Monazza Aslam were the lead authors of this policy note. The note was prepared under the guidance of Ivailo Izvorsky, Sector Manager (EASPR), Rogier Van Den Brink, Lead Economist (EASPR) and Coralie Gevers, Country Manager, Mongolia. Useful comments and inputs were provided by Sudhir Shetty, Coralie Gevers, Laura Chioda and Erdene Badarch. 2 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS CONTENTS Introduction........................................................................................................................................ 5 Labor market outcomes by gender: How Mongolia compares with other countries.........................9 Labor market outcomes in Mongolia: key findings from the 2008/09 Labor Force Survey................11 Mongolian labor markets are highly occupationally segmented by gender with limited opportunities for self-employment..........................................................................11 Early retirement impacts participation rates among older women and has potentially negative implications for career progression, pensions and poverty.......................................14 Women’s household and care duties negatively impact labor market outcomes.................15 Labor market outcomes are weaker for both men and women with basic/secondary levels of education, but the effects are more acute for women..........................................16 There are large gaps in earnings that cannot be explained by differences in men’s and women’s endowments such as education and experience...............................................17 Key policy recommendationsty.........................................................................................20 Implement equal pay for equal work policies and/or introduce affirmative action laws....20 Increase the funding and visibility of the National Committee on Gender Equality (NCGE) and other programs/policies on gender equality.................................................21 Review and revise maternal/paternal leave and child care policies................................21 Promote gender equality in the private sector.................................................................23 Improve women’s opportunities in entrepreneurial work...............................................25 Overall Summary and Conclusion...................................................................................................26 References...........................................................................................................................................27 Appendix A: Regulations restricting women’s occupational choices, 1999-2008.............................29 Appendix B: Summary statistics, male and female (aged 15-65), MLFS 2009....................................32 Appendix C: Summary statistics, distribution of the Labor Force in formal and informal jobs, by gender, region and age (ages 15-65)...............................................................................................34 Appendix D Selected partial effects on the likelihood of occupational outcome, by gender.................37 Appendix E: Access to Finance........................................................................................................40 Appendix F: Factors that explain participation rates........................................................................41 Appendix G: Probability and duration of unemployment.................................................................44 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 3 Appendix H: Earnings Gaps.............................................................................................................49 Appendix L: Oaxaca’s Decomposition...........................................................................................51 Figures Figure 1 Ratio of female to male enrollment (2007).....................................................................10 Figure 2 Gender Empowerment Measure, selected economy rankings.......................................10 Figure 3 Male and female labor force participation rates.............................................................10 Figure 5 Labor force participation by gender, age and location...................................................12 Figure 6 Distribution of employment by gender, location and age.................................................12 Figure 7 Occupational composition in the top five male and female dominated occupations in Mongolia.......................................................................................12 Figure 8 Reasons for unemployment among those aged 50-65, by gender.............................15 Figure 9 Labor force participation gap between men and women by location, household demographics and age..........................................................................................................15 Figure 10. Senior public officials and managers in government and non-government organizations..........................................................................................15 Figure 11 Distribution of male and female dependents in MLFS, by age...................................15 Figure 12 Hours per week spent on household chores by gender............................................16 Figure 13 Hours per week spent on household chores by women in wage employment..............16 Figure 14 Labor force participation rates by gender and education.........................................17 Figure 15 Gender gap in monthly earnings in selected industries....................................18 Figure 16 Employment by sector and gender........................................................................18 Figure 17 Decomposing the gender wage gap.......................................................................18 Figure 18 Maternity leave in East Asia....................................................................................22 Figure 19 Early childcare: no of students per kindergarten......................................................22 Boxes Box 2 Early Child Care Policies in Mongolia...........................................................................23 Box 3 Improving Gender Equality in the Workplace – the Gender Equity Certification Model in Mexico..................................................................................................24 4 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS INTRODUCTION Mongolia has made strong progress on key treatment in labor markets and labor laws and gender-related Millennium Development Goals regulations such as restrictions on women’s (MDGs) in recent years. Two vital targets for occupational choices that were only annulled in maternal health and child mortality were met 2008, difficulties in accessing child care and also to in 2008. Gender indicators in education and cultural norms, which mean that working women health are also better in many respects than in shoulder a “double burden� of work, outside the comparator countries in the East Asia and Pacific home and household duties within. They impact region. women’s participation rates overall as well as in high-growth sectors in the economy, and have Nevertheless, benchmarking Mongolia against implications for career progression, pensions and other countries reveals considerable inequalities poverty, particularly in female headed households with respect to economic and political power and where levels of dependency are high. decision-making. In the context of labor markets, gender disparities are especially prominent in Accordingly, policy to improve labor market the type of work women do – mostly unpaid outcomes for women needs to address each of with limited engagement in self-employment/ these factors. In 2011 Mongolia passed a Law on entrepreneurial activities and with high levels Promotion of Gender Equality. This is an important of occupational segregation – and the wages first step in a country which previously lacked that they are paid. In addition, women also have any specific national law on gender equality, and a limited presence in higher level managerial particularly because it spells out responsibilities positions. of specific public agencies to ensure gender equality and specifies that at least one quarter Gender inequality in economic participation of all representatives in central and local elected and labor markets has widespread ramifications bodies be women. In addition, the government not only for development but also for growth1 can consider a range of policies to reduce gender as documented in the World Bank’s World disparities in labor markets. In the context of Development Report (2012) on gender. For public sector practices, these could include instance, the Food and Agricultural Organization reviewing where the female workforce is located, estimates that if women and men had equal what impedes women’s career progression and access to productive resources, agricultural output what are their main concerns, and coming up with in developing countries could increase by 2.5-4 strategies to address these. In addition, there is percent. Outside agriculture, studies suggest that some evidence that retirement laws that allow eliminating discrimination against women workers women to retire earlier than men have been and managers could yield increases in productivity misused with potentially adverse impacts on per worker of 25-40 percent and 7-18 percent for women’s careers and pensions. This suggests the East Asia (Cuberes and Teignier-Baque, 2011)2 need for policies that limit the scope for abuse, These figures only represent an estimate while at the same time ensuring that employers of economic costs: limiting women’s economic retain the flexibility to move workers to their opportunities also has pervasive intergenerational most productive uses. The government may also social costs that remain unmeasured. In Mongolia, consider introducing affirmative action policies these inequalities are linked to differential in sectors where women are acutely under- 1 World Bank World Development Report (2012a), “Gender Equality represented such as mining and transport, storage and Development� and communication. This may help to offset any 2 World Bank (2012b), “Toward Gender Equality in East Asia and the legacy of occupational segmentation from labor Pacific: A Companion to the World Development Report� MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 5 regulations referred to above as well as any be increasing. Mongolia is currently ranked 86th cultural/social biases that prevent women from out of 171 economies3 on the overall ease of doing working in traditionally male-dominated sectors. business and women who lack access to business networks or finance may be doubly discouraged Furthermore, policies that help to balance from setting up their own businesses. This suggests work and family life need to be introduced, such a greater role for policies that make it easier to as paternity leave for fathers and improving start and operate businesses, including in high- childcare facilities that currently suffer from a growth sectors. Other policies that may be helpful lack of funding and overcrowding, particularly in include promoting awareness of and encouraging Ulaanbaatar which experiences large inflows of the development of (appropriately regulated and migrants from rural areas. Policies in the private supervised) micro-lending institutions. Broad sector that support a more equitable participation policy actions for the government are set out in of women also need to be encouraged. Table 1. Finally, women’s participation in entrepreneurial work is lower than that of men. 3 World Bank 2012 Doing Business Indicators Report, ranking for Moreover, overall levels of informality appear to 2011 6 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS Table 1: Key Policy Recommendations and Implementing Agencies and Partners Key Policy Recommendations Implementing Agencies/Partners • Implement the Law on Gender Equality. The Law sets out the • The main implementing/ responsibilities of public agencies with regard to gender discrimination and monitoring agency for the Law on ensuring adequate representation of women at all levels of government and Gender Equality is the National also mandates the establishment of a complaint mechanism through the Committee on Gender Equality National Human Rights Commission and employment dispute commissions. (NCGE), supported by all organs Improving gender equality in the public sector will require auditing where of government. Partners: ADB, women are located, what impedes their progress and setting benchmarks World Bank, other international for senior management. donors and national and international agencies. • Increase the funding of the National Committee on Gender Equality, the • Ministry of Finance, the national main agency in charge of monitoring the Law on Gender Equality. Cabinet and Parliament • Introduce affirmative action regulations in sectors where women • The Ministry of Social Welfare are acutely under-represented such as mining. Aside from cultural/ and Labor/ Ministry of Mineral social norms, low levels of participation are also likely the result of labor Resources and Energy regulations that restricted the degree to which women could participate in these sectors and which were only annulled in 2008. The government could encourage greater participation through affirmative action regulations that help to raise female participation e.g as has happened in South Africa where the government mandated that 10 percent of the mining workforce be female by 2010 • Review retirement laws to ensure that women are not induced to retire • The Ministry of Social Welfare earlier than they want and Labor, supported by NGOs. These allow women to voluntarily retire earlier than men, but there is some evidence to suggest that they have been misused with adverse impacts on women’s careers and pensions. Accordingly policies that limit any potential scope for abuse, for example by setting up appropriate complaint mechanisms, while at the same time ensuring that employers retain the flexibility to move workers to their most productive uses, need to be considered. Review childcare and maternal/paternal leave policies. Mongolia currently • The Ministries of Education, does not provide paternity leave, and child care by the state suffers from Culture and Science and of Social under-funding, overcrowding (particularly in urban areas) and a lack of Welfare and Labor, supported by teacher training, with national policies emphasizing family-based care of NGOs, international donors pre-schoolers. • Promote gender equality olicies in the private sector. Introducing/funding • The National Committee on workshops on gender sensitive policies, introducing national awards/ Gender Equality (NCGE), the rankings of companies that follow the most gender-friendly policies. The Ministry of Social Welfare and government could also set out best practice guidelines for the private sector Labor, supported by NGOs. and ask firms to document to what extent they comply with these at the time of the publication of their annual reports or why they do not. • Strengthening women’s opportunities as entrepreneurs – and in the • Ministry of Finance supported by private sector more broadly. Far fewer women are self-employed compared NGOs and international donors to men, particularly in urban areas. In addition, the costs of doing business such as the World Bank and are high, which may doubly disadvantage women who may lack access to Asian Development Bank social and financial capital. Mongolia is currently ranked 86th out of a 171 economies on the overall ease of doing business and women who lack access to business networks or finance may be doubly discouraged. Aside from policies that reduce the costs of business regulations, other policies that may be helpful include promoting awareness of and encouraging the development of (appropriately regulated and supervised) micro-lending institutions. These would benefit both men and women. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 7 This policy note is structured as follows. It within the country. It then concludes with a set of starts by comparing gender outcomes in Mongolia strategies, policies and practices that could help with other comparator countries, before looking improve economic participation and labor market deeper into gender differences in labor markets outcomes for women. 8 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS LABOR MARKET OUTCOMES BY GENDER: HOW MONGOLIA COMPARES WITH OTHER COUNTRIES Gender inequalities vis-à-vis the level of human compared to 10.5 percent in 1990. The proportion development are fairly low in Mongolia, but highly has since increased to 12 percent in the latest 2012 prominent in economic and political power and parliamentary elections, but this is lower than the participation in decision making. For instance, global average of 19.7 percent and is also lower consider the Gender Development Index (GDI), compared to several other developing countries is simply the Human Development3 index which in the region. sMoreover, until 2011, Mongolia measures inequalities in three dimensions – a long lacked any specific national law on gender equality and healthy life, education, and a decent standard although it was a signatory to major international of living – adjusted for gender imbalances. The conventions on the rights of women and children ratio or the difference between the GDI and HDI (ADB, 2010). The new law on gender equality measures the gender gap in human development. passed by Parliament in February 2011 spells out However, at 100%, the ratio places Mongolia the responsibilities of specific public agencies to as the best among 155 countries in 2009.4 In ensure gender equality in various sectors such as education, if anything, the country is faced with employment, education, health, and across the reverse (or pro-female) gender gaps in enrolment public sector in general. Furthermore, under the rates. There is no major gender difference in net labor regulations introduced in 1999 (Appendix A) enrolment rates at the primary level while at which are a leftover from the country’s communist the secondary and tertiary level gender gaps in past,6 for many years women were prohibited enrolment strongly favor female students (Figure from an extensive range of activities (e.g. driving 1). vehicles which carry more than 25 passengers, to work as machinists, to butcher cattle). These However, a comparison of economic and regulations were only annulled in 2008, and it political power shows Mongolia in a much less is likely that they were a key contributing factor flattering light. The Gender Empowerment in reinforcing occupational segmentation in Measure (GEM) indicates whether women take an Mongolia alongside gender stereotypes. active part in economic and political life. It tracks women’s share of legislators, senior officials and managers; of professional and technical workers; seats in parliament held by women; and the gender disparity in earned income. Here Mongolia scores a lowly 94th out of a 109 countries (Figure 2) much worse than in other East Asian economies. That participation in political decision making is poor can be gauged from the fact that in 2009 only 45 percent of Mongolian MPs were women, 4 The greater the gender disparity in basic human development, the lower is a country’s GDI relative to its HDI. Mongolia’s GDI value was 0.727 in 2009 compared to its HDI value of 0.727 5 UNDP (2010) Asia Pacific Human Development Report 6 2010 Women, Business and Law Report by the World Bank MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 9 In addition although women in Mongolia from the regional average7. Nevertheless, a third participate only slightly less than men in the of women are employed as unpaid family workers labor market, when they do, many more are (Figure 4). This proportion is large compared to concentrated in the non-remunerated part of the the East Asia average and suggests that gender labor market. At 70 percent (Figure 3), women’s inequalities in access to economic opportunities incorporation into the labor force in Mongolia are considerable compared to other countries in is much higher than in a number of East Asian the region. and Central Asian economies, while official unemployment rates are also not very different 7 The official unemployment rate in Mongolia for women in 2008 was 3.3% compared with a regional average of 3.1% in East Asia. However official rates often understate actual unemployment rates. 10 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS LABOR MARKET OUTCOMES IN MONGOLIA: KEY FINDINGS FROM THE 2008/09 LABOR FORCE SURVEY Gender asymmetry in Mongolia’s labor or the informal sector, are concentrated in the market and economic participation is especially tertiary sector while men are located in the remarkable given the overall degree of equality in secondary sector (Appendix C). Interestingly, the educational outcomes and the pro-female gaps at declining trend in informality rates that was visible higher levels of edutcation. 8 This section identifies in first half of the 2000s due to the decreasing key gender differences in labor market outcomes share of agriculture in total employment appears using the Mongolian Labor Force Survey (MLFS to have reversed11. This appears to be driven by 2009) and the factors driving these outcomes. the increase in informality among salaried or wage workers, and likely reflects the effects of the Generally, gender outcomes differ by age-cohort severe economic downturn in 2009 when GDP and by rural versus urban location of residence. growth fell to minus 1.3 percent from 8.9 percent These aspects matter because Mongolia has a the year before. relatively young population. Almost 44 percent of the female and 46 percent of the male working Overall the main findings are: age population (15-65 year olds) are younger than MONGOLIAN LABOR MARKETS 30 (Appendix B). In addition, social and economic ARE HIGHLY OCCUPATIONALLY transformations initiated in the 1990s have led SEGMENTED BY GENDER WITH to large-scale migration to urban areas and the LIMITED OPPORTUNITIES FOR SELF- capital city (Ulaanbaatar), with currently some EMPLOYMENT 40 percent of the working population residing in Overall, women in Mongolia participate less urban areas in the MLFS9. than men in the labor force and when they do, a It is also worth noting that the majority large portion work as unpaid family workers notably (66 percent) of working age individuals – men in rural areas (Figure 5 and Figure 6). Although and women – work in the informal sector10, education plays an important role in bringing mostly agriculture Outside agriculture, a higher women out of unpaid work, the only alternative proportion of women, whether in the formal occupation open to them is wage work. Men for 8 One possibility is that women’s quality of schooling is worse than example are twice as likely to take advantage of that of men. But while the overall quality of general and vocational self-employment (outside the agriculture sector) schooling in Mongolia is poor (ADB 2008 and UNICEF 2009), there compared to women at all levels of education is no evidence of large gender differences in schooling quality or that it impacts men and women differently. Trends in International (Figures d1-d4, Appendix D). In addition, women Mathematics and Science Study (TIMSS) data from 2007 indicate are also concentrated in a relatively narrow set of that boys outperform girls in 8th grade mathematics. Whether this occupations such as teaching, catering and retail disparity arises due to differences in the quality of schooling, home environment or inherent ability is unclear. And, the quality of data support services (Figure 7). The top 5 ‘male’ and was doubtful and these results are relegated to the appendices of ‘female’ occupations engage about 57 percent of the reports (TIMSS 2008). 9 Overall, 57 percent of the population lives in urban areas (World all men and 55 percent of women. Although some Development Indicators, 2010) professions overlap (commercial livestock work 10 Informal’ work is defined to include all unpaid workers (regardless and street vending), women are noticeably absent of whether they work in formal/informal enterprises), own- account workers in informal enterprises (employing less than or from transportation and construction but heavily equal to 5 persons), wage workers in informal enterprises (workers concentrated in support positions in retail and do not receive pensions) and ‘others’ to include employers and catering and in teaching. members of cooperatives employing less than or equal to 5 persons. 11 See Appendix C . MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 11 These findings reflect a combination of factors capital suitable for self-employment can including limited business and entrepreneurial also be possible explanations, although it is capital appropriate for participating in the not really possible to explore these issues high growth sectors of the economy and labor analytically in the Labor Force Survey. regulations12 that existed until 2008 combined First, to the extent that the labour market with cultural norms that discouraged women from experiences of women and men differ, fully participating in all sectors of the economy. male and female entrepreneurs may also a) Limited participation in self-employment choose to locate in sectors that they are may reflect smaller amounts of human more familiar with and may also affect capital appropriate for setting up businesses, the likelihood that they may even start a or limited economic opportunities, business. 13 For example, if work experience including access to external capital or is restricted to clerical and administrative business networks. That said, self-selection support positions, and fewer women have by women into relatively stable wage managerial experience as is clearly seen in work, or a smaller accumulation of human the Labour Force surveys (see discussion 12 See Appendix A. 13 See Klapper and Parker (2010) for an excellent review. 12 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS later), then the work skills accumulated by firms in Mongolia indicates that women Mongolian women may make it difficult to do not have necessarily more difficulty in start a business in the first place. accessing finance than men16. A roughly similar proportion of male and female In addition, disparities in asset ownership owned firms borrow from banks (Appendix by gender in Mongolia may also impact E), and although amounts borrowed by women’s access to finance. Unequal female owned firms are smaller (reflecting distribution of state assets in the shift their size and industry location), both male- towards privatization in the 1990s when and female-owned firms also post roughly lands and livestock were registered in the the same collateral (about 40 percent of the names of heads of households, mostly men, loan’s value). However, the fact that there left women without legal title or control. are no differences between existing male As a result, women require the consent of and female entrepreneurs is a problem of heads of households (usually husbands) to sample selection: i.e. there exists data only use assets as collateral for loan or credit14. on those that have managed to clear all For example, a survey in 2008 found that hurdles associated with starting a business for 88 percent of the women interviewed, in the first place. It is quite possible that property and assets were registered solely access to finance remains an important in the name of their spouses and only 12 issue for women attempting to start a percent had joint ownership.15 business in the first place, compared to While it is not possible to comment on male counterparts. access to finance for would-be female b) Extensive labor regulations dating back entrepreneurs compared to male to 1999 (Appendix A) and that were only counterparts, there does not appear to be annulled recently in 2008 contained broad a difference in access to finance by gender prohibitions that likely prevented Mongolian among existing entrepreneurs. Admittedly women from fully participating in key high- the proportion of entrepreneurs/self- growth occupational areas including mining, employed able to take out a loan in the construction and transport. Mongolia’s GDP Labor Force survey – these are mostly small growth is expected to pick up from around informal sector firms – is low, but there 6 percent in 2010 to more than 25 percent are no significant differences according to by around 2013 as large mining projects gender (Table 2). come on-stream. This is expected to lead to significant knock-on effects on related sectors including infrastructure, utilities, Table 2. Access to finance among self-employed, transportation and mining, i.e. the very by gender sectors that were ‘closed’ off for a very % of total long time to women by these regulations Able to take out loan in the male female name of the business 25.2 22.3 The ostensible reason for these restrictions was Not able 67.9 69.6 health and safety concerns, but they were quite broad, preventing women from undertaking Dont know 6.9 8.2 jobs that are deemed “safe� in other countries. Source: MLFS (2009) In addition, health and safety concerns could arguably have been more directly addressed, to Moreover, 2008/09 survey data from a the benefit of both men and women, rather than sample of larger, formal, private sector through exclusionary labor regulations. Indeed 14 Robinson and Solongo (2000) the list was so extensive that it covered industries 15 excluding single mothers who had property registered in their names. Survey from Khas Bank, Measuring the Impact 16 The World Bank’s 2008/09 Business Environment and Enterprise of Microfinance on the Poor Rural Women in Mongolia, Draft Performance Survey (BEEPS) dataset comprises of 362 mostly Baseline Report, July 2008. large, formal sector firms. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 13 such construction, transportation (rail, road and What is evident though is that relatively fewer air), meatpacking, textile, tailoring and publishing, women reach higher level managerial positions. where such health and safety concerns would For example, women tend to account for one- have been quite small. In addition, these sectors fifth of director and executive director positions in are already male-dominated with limited support the 2009 Labor Force Survey (Figure 10) and are facilities for women17. As a result, in combination overwhelmingly concentrated in mid-low level with cultural factors, these labor regulations have managerial and support staff positions both in the likely contributed to limited female participation public and private sector. While other factors may in these sectors, the legacy of which is still visible also important, such as gender discrimination in today. the workplace, early retirement may also be one contributory factor. EARLY RETIREMENT IMPACTS In addition early retirement impacts women’s PARTICIPATION RATES AMONG OLDER earnings (which, in the worst case, fall to zero WOMEN AND HAS POTENTIALLY if the woman becomes unemployed) and also NEGATIVE IMPLICATIONS FOR CAREER pension payout levels to the extent that these PROGRESSION, PENSIONS AND depend on the length of job tenure. It therefore POVERTY. increases the risk of poverty for low-income The retirement age in Mongolia for women is households particularly those headed by females 55 years, 5 years earlier than for men and rising – there are twice as many elderly female headed to 10 years earlier if they have four or more households (around 15,000 according to NSO children. Although legally retirement occurs “at data) in Mongolia as there are male – and as one’s own request,� it is unclear to what extent noted before, there is some evidence to suggest women self-select into early retirement and that women have fallen into poverty as a result of to what extent they are required or otherwise “involuntary� retirement.19 . A study by the UNDP induced to leave. There is some evidence to and the Ministry of Finance on pension issues in suggest that the law has been used as a pretext for 2004 found that only 25 percent of retired men unfairly dismissing women who have then slipped received the lowest pension20 compared to 84.2 into poverty.18 Furthermore, retirement is the percent of women. most commonly cited cause for unemployment It is also worth noting that levels of male among women older than 50 years of age in the dependency are much higher in female headed 2009 Mongolian Labor Force Survey (MLFS), and households (Figure 11). For example, working age women’s participation rates drop quite sharply males account for about a third of dependents thereafter (Figure 8 and Figure 9). In the MLFS, in female headed households (FHH), twice the three-quarters of 50-65 year old women cited amount in male headed households) with a retirement as the main reason for unemployment, significant portion of these over the age of 30. The compared to 55 percent of men. Moreover, of probability of labor force participation for these these women roughly 30 percent were heads of males is much lower, particularly for older males households suggesting self-selection is not likely – it falls by 8 percentage points if the male is over to have played much of a role with respect to early the age of 30 (Table b, Appendix F) which suggests retirement. that female headed households are much more likely to be under financial stress than male- headed households. 17 See companion policy paper “Raising female participation in the large-scale mining sector in Mongolia,� World Bank (2011) 19 CEDAW Watch, Reports In Response To Request For Information, forthcoming. January 24 –February 15, 2000. http://www.un.org/womenwatch/ 18 CEDAW Watch, Reports In Response To Request For Information, daw/cedaw/cedaw24/cedawcmng34.pdf January 24 –February 15, 2000. http://www.un.org/womenwatch/ 20 Ministry of Finance and UNDP, Gender Analysis of Public Spending daw/cedaw/cedaw24/cedawcmng34.pdf in the Field of Social Security (Ulaanbaatar, 2004), 17-18. 14 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 21 WOMEN’S HOUSEHOLD AND CARE younger in a household reduces the probability of DUTIES NEGATIVELY IMPACT LABOR participation, especially among older women for MARKET OUTCOMES whom the probability of participation drops by Women spend roughly twice the amount nearly 4 percentage points (Appendix F). of time as men on household and care duties, However an additional factor compounding and this does not decline even when they are these effects may be the relatively generous cash engaged in paid productive work in the labor transfers to women of child bearing age that market (Figure 12 and Figure 13). For working proliferated from the mid-2000 onwards. These women this represents a “double burden� and the included monthly cash allowances of MNT3,000 impacts are evident in labor market outcomes. (the Child Money Program) for each child, For example, the presence of children aged 15 or additional quarterly allowances of MNT25,000 to 21 Technically these are participants in the MLFS that identified each child, and cash allowances of MNT100,000 themselves as belonging to a male headed or a female headed households, but were not household heads themselves MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 15 to every newborn child, to name but a few22 with children to opt out of the labor market. Data which may have encouraged younger women shows that birth rates in Mongolia, which had been falling rapidly since the early 1990s, started 22 The Child Money Program, which was the most important of these transfers, disbursed nearly 5.6 percent of total fiscal spending to inch up again since 2005 suggesting that the in 2008. These social transfers were suspended in the 2010 budget and replaced with universal, consolidated and generous benefits provided by the government encouraged cash transfers in the 2011 budget for every Mongolian citizen. women to have more children. Following national elections in 2012, the Child Money Program was reinstated in October, paying out MNT 20,000 per month per child under the age of 18 and will replace the universal cash transfers. LABOR MARKET OUTCOMES ARE than those with higher levels of education or those WEAKER FOR BOTH MEN AND WOMEN with no education at all. This somewhat puzzling WITH BASIC/SECONDARY LEVELS OF finding could be because workers who have these EDUCATION, BUT THE EFFECTS ARE educational qualifications are bringing inadequate MORE ACUTE FOR WOMEN skills to the labor market – there is no shortage Labor market outcomes are weaker among both of educated men and women in Mongolia but a men and women who have completed basic and shortage of individuals with the right skills. secondary education compared with those with Nonetheless, negative outcomes are more much higher levels of education or no education pronounced for women with basic/secondary at all. This affects a majority of the productive education than for men with similar levels of workforce in Mongolia (a total of 56 percent education. There is, for example, a 9 percentage women and 60 percent men have completed basic point difference in participation rates between and/or secondary education levels. Less educated men with a basic education compared to women men and women are less likely to participate in the with a basic education and the corresponding gap labor market23 and more likely to be unemployed24 for men and women with secondary education 23 See table c in Appendix F is 13 percentage points (Figure 14). Similarly, the 24 Table g2 in Appendix G. Overall though, gender differences in unemployment rates are not very large and gender does not play probability of unemployment for a woman with a a significant role in the probability of being unemployed or in secondary education, compared to a woman with the duration of unemployment. Generally, both young men and a university education is 7.4 percentage points women have a higher probability of being unemployed compared to older men and women. 16 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS higher, compared with 3.5 percentage points for THERE ARE LARGE GAPS IN EARNINGS men (Table g2, Appendix G). THAT CANNOT BE EXPLAINED BY DIFFERENCES IN MEN’S AND WOMEN’S ENDOWMENTS SUCH AS EDUCATION AND EXPERIENCE. Women may earn less than men due to several reasons. It is sometimes suggested that women’s education and subject-choices are consistent with their lower career aspirations and they ‘self select’ into low wage occupations and jobs (ILO 2004). They may also accumulate a wage gap over their careers due to having less work experience because of career breaks with child birth. Thus, women may get paid less than men because they bring lower productivity characteristics (education or experience among others) to the labor market. Their earnings may also be lower because the labor market rewards women differently than men for possessing the same set of characteristics. Put simply, women may be treated differently than men. A vast literature in both developing A jobs/skills mismatch argument cannot explain and developed countries notes that a sizeable why the effects are more acute for women than portion of the earnings gap cannot be explained men with basic/secondary education. Instead, the by differences in observed attributes of men and answer may be more prosaic, namely more limited women. work opportunities or perhaps discrimination. Both are plausible. Job advertisements that set out In Mongolia, large and increasing raw gender requirements regarding the physical appearance earnings25 gaps exist across almost all industry (or age) of women are documented in both the sectors that women are concentrated in, the main ADB 2010 gender report and also in World Bank- exceptions being wholesale and retail trade and sponsored qualitative research on the impacts of public administration (Figure 15 and Figure 16). the economic crisis (Reva et al, 2010). Meanwhile, Overall men earn about 10 percent more than as discussed earlier, women appear to have fewer women and the gap is also obvious across all work opportunities in self-employment compared education levels and is particularly high in urban to men: only 27 percent of women with basic and areas and in the formal sector (Table h1, Appendix secondary education levels for example are self- H). employed compared to 40 percent of men. 25 Appendix H depicts mean earnings of wage earners in the MFLS (2009) data. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 17 The raw gap potentially hides differential treatment (or even discrimination) against women if the average productivity characteristics of women are better than those of men and if the returns to women’s characteristics (such as levels of education) are the same or better than men’s, as indeed they are (Table h2 Appendix H). In the 2009 labor force survey, wage earning women26 have on average better mean characteristics than men; they have completed slightly more years of schooling than men (13 years versus men’s 12), have only marginally less experience (17 years versus 18 for men) and 42 percent of women have 10 years or more experience in their current job (tenure) compared to 37 percent of men. 26 Earnings data are available only for wage workers in our sample. However a majority of women are actually unpaid workers and some are engaged in self employment activities. But earnings in agriculture and non-agriculture self employment are not available and gender gaps in earnings may be substantially larger in these occupations partly because they reflect women’s weaker access to capital, land and social networks that are known to have implications for earnings in self employment. 18 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS A decomposition of the wage gap indicates The overall gender wage gap in log wages is substantial differences in treatment of women 0.18 which implies that male wages are 1.2 times and men by employers in Mongolia’s labor market. higher than female wages (Figure 17). Controlling Wage gaps can be disaggregated into “explained� for industry and occupation characteristics, the and “unexplained� portions.27 The first is that gap rises further, suggesting that the distribution part of the wage gap that arises from differences of men and women across occupations and in the average characteristics of men and women industries is not a driving force behind the (e.g. in productivity, schooling or tenure). The explained component of the gender wage ‘unexplained’ portion measures the effect of differential. Among younger workers, the gap is unobserved characteristics between the sexes even larger, indicating that young men are paid (e.g. motivation or ability) but is sometimes taken 1.35 times what young women are paid. in the literature as evidence of discrimination, Decomposing the gap however suggests that the namely that the market rewards women and overwhelming bulk of it is due to the fact that the men differently given observably identical market values men and women’s work differently characteristics. rather than due to differences in observed 27 Using the Oaxaca technique See Appendix H for details. characteristics. In fact, if they were paid in the This decomposes labor market endowments and returns to same way as men, based on the characteristics endowments by gender. The gap between wages that are explained that are observed, women would earn more than by observ able and non-observable characteristics is usually taken to provide an indication of the degree of discrimination. men because they have better characteristics than However, itcould reflect a number of things, including the effects wage working men. Only among older workers do of measurement error (e.g., poor proxies for years of labor market differences in observed characteristics help to experience), omitted variables (on-the-job training, intermittency in the labor force), and other unobservables (e.g., effort, “innate explain a small portion of the gender wage gap. ability�). It is therefore, at best, an indirect and imperfect measure of discrimination. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 19 KEY POLICY RECOMMENDATIONS The above findings suggest a number of policy objectives. However improving gender equality actions for improving outcomes for women in the in the public sector will entail, at the very least, labor market and increasing gender equality. The reviewing where the female workforce is located, main recommendations are what impedes women’s career progression and what are their main concerns and addressing these. It might also involve introducing clear IMPLEMENT EQUAL PAY FOR EQUAL benchmarks or affirmative action policies for WORK POLICIES AND/OR INTRODUCE senior management on how to incorporate gender AFFIRMATIVE ACTION LAWS sensitive policies, conducting gender sensitization Such policies are needed to address: a) the workshops and so on. Currently the Law on Gender widespread differential treatment against women Equality mandates that the National Committee with regard to pay that cannot be explained by on Gender Equality monitor the implementation factors such as education or job experience; b) of the Law. Affirmative action policies may also women’s low rates of participation in high growth be useful in both increasing women’s presence at sectors of the economy and c) the inadequate higher levels of decision making and also helping representation of women at higher levels of to create mentors and role-models for more junior decision-making, in effect the existence of ‘glass female staff. ceiling’ for women in management positions. Specific policy issues may also merit close Mongolia is a signatory to most international attention. For example under labor regulations conventions on the rights of women that mandate that existed until 2008 women were prohibited equal treatment of men and women, and in from engaging in a broad list of activities February 2011 passed a national law on gender (Appendix A), which in combination with gender equality as well. The law is important because it stereotyping and cultural norms about which addresses discrimination against women in the jobs for women are “appropriate,� resulted in political, economic, family and educational arenas high levels of occupational segmentation in high and also sets out sexual harassment as a form of growth sectors of the economy such as mining. discrimination. It also indicates the responsibilities To reverse the effects of such regulations and to of public agencies, including the Prime raise female participation in such sectors, the Minister’s office, with regard to guarding against government may consider introducing affirmative discrimination generally, proposes affirmative action regulations e.g. as has been done in South action policies in public sector, and establishes Africa where the government mandated that a complaint mechanism through the National 10 percent of the mining labor force comprise Human Rights Commission and employment women. dispute commissions. Early retirement laws are of concern since Ultimately however enforcing the law will they can be used to induce women to leave the prove to be the litmus test for how gender equality labor market earlier than they desire of the labor is prioritized by policy makers. As it is, the version force with corresponding impacts on pensions and of the Law on Gender Equality that was finally vulnerability to poverty. A study by the UNDP and approved by Parliament discarded a provision the Ministry of Finance on pension issues in 2004 in earlier drafts that set out the government’s found that only 25 percent of retired men received responsibility for undertaking preliminary gender- the lowest pension28 compared to 84.2 percent of based analysis of laws, policies and programs with respect to compliance with gender equality 28 Ministry of Finance and UNDP, Gender Analysis of Public Spending in the Field of Social Security (Ulaanbaatar, 2004), 17-18. 20 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS women. Household surveys show that roughly 31 and support their return into the labor force after percent of urban FHH are poor compared to 26 childbirth. percent of regular two-breadwinner male headed Maternity/ paternity leave policies and child households. Meanwhile, the 2009 LFS survey benefits: A revision of current maternity and shows that three-quarters of women aged 50-65 paternity leave policies may be a step in the right who are unemployed and looking for work, cite direction. Maternity leave is currently 120 days retirement as the reason for being unemployed but with only 70 percent of wages paid (Figure in the first place (Figure 8). Accordingly, it may be 18). Moreover, maternity leave is unlikely to be appropriate to reconsider these laws and to set up offered at all in the informal sector, where almost appropriate complaint mechanisms where unfair 64 percent of the female labour force is located. dismissal issues can be appropriately addressed. Currently, the government does not mandate paid or unpaid paternity leave for men. However, INCREASE THE FUNDING AND until 2009 the government provided generous VISIBILITY OF THE NATIONAL child benefits (for existing children and newborn COMMITTEE ON GENDER EQUALITY children) which may also have encouraged young (NCGE) AND OTHER PROGRAMS/ mothers to opt out of the labour market. POLICIES ON GENDER EQUALITY Child care and flexible working hour policies. The NCGE was set up in 2005 and is chaired by Access to affordable and accessible child care, the prime minister and comprises more than 30 which helps to balance women’s caregiving and members including key ministers and prominent market roled is regarded as critical in strengthening private sector and civil society members. The women’s access to economic opportunities29. The NCGE is an important body –the recently passed analysis in this policy note indicates, the presence law on gender equality was prepared by it of children has significant impacts on labor force and it is the main body in charge of ensuring participation rates particularly among young compliance with the Law – and it operates to women (mothers) and older women (possibly ensure consultation, coordination and monitoring grandmothers) in urban areas. Flexible work of the National Program on Gender Equality set policies (shorter shifts to accommodate school up in 2002. However, it lacks visibility and as hours, home based working) and accessibility to the Asian Development Bank’s Country Gender affordable good quality child care are needed but Assessment (2010) identified, and is also lacking they are not widespread, and as with maternity in financial resources and manpower. The already leave, also not likely to be available in the informal small budget for the NCGE has been cut sharply in sector. Meanwhile, although national child recent years, although the 2011 budget provided care policies have been reviewed and revised an increase of 6 percent on the 2010 allocation, in recent years, their implementation needs to to MNT 46mn. The NCGE should be provided the be strengthened, including through increased financial and physical manpower to be better able funding, teacher training and certification (see to argue and pursue its objectives of promoting Box 2). For example, the national average ratio of gender equality and monitoring progress on the children to one teacher is 29 in kindergartens but implementation of the new gender equality law. is close to 40 children to a teacher in Ulaanbaatar30 where 60 percent of the population is located. Also as Figure 19 shows, in UB the average number of REVIEW AND REVISE MATERNAL/ children per kindergarten is much higher than the PATERNAL LEAVE AND CHILD CARE POLICIES national average Policies are needed that encourage women’s However, any revision of leave and child care entry into the labor force (and particularly policies should carefully consider design features remunerative work), that promote their 29 See World Bank (2012b), “Toward Gender Equality in East Asia and the Pacific: A Companion to the World Development Report� continuous presence during their ‘fertile’ years 30 ADB (2010b) MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 21 such as whether paternity leave is transferable, and institutional factors are also important, with whether they encourage women to exit too long fathers often reluctant to undertake care duties, from the labour force – the length of maternity leave and this needs to be addressed as well through is higher in Mongolia compared to neighbouring greater social awareness campaigns, making countries, but benefits are lower (Figure 18) – paternity leave mandatory etc. financing or funding features and so on. Cultural 22 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS Box 2. Early Child Care Policies in Mongolia In Mongolia, state-funded early childcare and education was rolled back in the 1990s as the economy transitioned to a capitalist system and this has added to the “double burden� of women, with national policies emphasizing family-based care of pre-schoolers (under-3 year olds) and (mostly) state funded kindergarten care for older children.2 With low enrolment rates, and growing concerns about marginalized young children in rural areas, there has been considerable national debate on early child care policies leading to policy reform For instance, in 2008 Parliament ratified a Law on Pre-school Education in 2008 which legalized alternative forms of education (to address the education needs of the most marginalized children in rural areas e.g through ger summer kindergartens) and also required every child to be engaged in early childhood education (ECE) activities. In addition, the law indicated that the government would fully cover the cost of school meals in kindergartens versus half as had been the case before. The law is complemented by the Education Sector Master Plan-2 for 2006–2015 (ESMP-2) adopted in 2006 that aims to enroll 35 percent of rural children aged 2–6 in kindergartens and 64 percent in alternate forms of preschool education. In addition, beginning in 2008-09, Mongolia shifted from an 11 year to 12 year school cycle lowering the age of enrollment from 7 to 6 that lead to a 9 percent increase in total enrollments by 6-year-old children How well are these policies doing? According to a recent ADB review: “For the 2007/2008 academic year, there were 768 kindergartens (87% of them are public) with 130,758 children nationwide. Only 57.1 percent of children aged 2–6 are enrolled in kindergartens. Of these, 43.8 percent are enrolled in formal kindergartens and another 13.3 percent in alternate forms of preschool education. According to Ministry of Education, Culture and Science, of the 43 percent of children not enrolled in ECE, around 30 percent are children of migrant families and 13 percent are nomadic and rural children�. The report also identifies a number of problems that include: --Short duration and quality of alternative ECE programs in rural areas such as mobile kindergartens --Lack of teacher certification and service standards, registration standards, and curriculum in the alternative ECE programs with no official guidelines on how to benchmark these appropriately --Inadequate funding in rural areas and overcrowding in urban areas as large rural-urban migration flows have increased demand for social services in underfunded peri-urban ger areas Source: ADB (2010b) Description of the Pre-school System http://www.adb.org/documents/ supplementary-appendixes/43127/43127-mon-sa-c.pdf PROMOTE GENDER EQUALITY IN THE An alternative approach would be for the PRIVATE SECTOR government to set out best practice guidelines for the private sector and ask firms to document to what extent The government can also promote best practice they comply with these at the time of the publication in the private sector e.g. by introducing/ funding of their annual reports or why they do not. This could workshops on gender sensitive policies and the increase peer pressure (notably among larger firms) advantages of these to firms (a more diversified and to report on the degree to which they follow gender productive pool of labor) or by introducing national friendly employment policies, and is in their own awards/rankings of companies that follow the most advantage given that a diversified labor pool will help gender-friendly policies. An example of a successful, their own bottom-lines. low cost gender awareness and reward campaign that was followed in Mexico is given in Box 3. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 23 Box 3. Improving Gender Equality in the Workplace – the Gender Equity Certification Model in Mexico The Gender Equity Model (MEG), a process towards certification in Mexico, is a public-private partnership between the Mexican government and private companies interested in promoting gender equality. The program offers lessons on creative, innovate approaches to address gender discrimination in the private sector workplace and improve the business enabling environment for the benefit of all. It was implemented as part of the Gender Awareness Component of the Mexico Gender Equity Learning and Innovation Loan (LIL) Project implemented between 2001 and 2005 and which sought to address gender inequalities by helping build public capacity, pilot community based initiatives on gender, and increase gender awareness. The World Bank financed the project through a $3.03 million Learning and Innovation Loan. A particularly innovate aspectwas its awareness raising component, which set up the MEG to recognize progress on gender equity in private firms, public entities, and non-governmental organizations (NGOs). Prepared in cooperation with the Mexican government’s National Women’s Institute and through consultation with private and public sector leaders, the academia and NGOs, MEG covered four areas: recruitment, career advancement, training, and sexual harassment. The aim was for the participating firms to foster equal opportunity practices for the benefit of both women and men. The program was voluntary and dynamic. It did not require a participation fee from the companies and had no strategy to target particular firms or industries. In the first round there was a public call for participation and the project’s team approached businesses. Since the project did not have much funding for dissemination, conducting outreach and awareness activities was a challenge. Yet there was considerable interest among firms and demand quickly outpaced supply. The process began with a self-assessment and identification of strengths and weaknesses, followed by a training program and support to overcome limitations and institutionalize best practices. Four independent certification firms were hired to assess and evaluate the companies. Once a firm completed the process, it received the gender equity seal. The monitoring continues on a yearly basis, for a period of three years to ensure sustainability. Since the firms could define their own goals and steps on incorporating gender equality measures, participation was not costly for them. The project provided training and paid the certification firms around US$500 for each participating firm. Thirty percent of participating firms were international organizations such as Wal-Mart, Manpower, Kraft, or Motorola, which tended to already have the capacity to incorporate the needed measures. For firms that are careful to maintain a brand reputation, MEG served to strengthen the corporate image. MEG firms have placed the seal on their published materials and are using the certification to show their commitment to diversity. For example, Wal-Mart which has been the subject of the largest gender discrimination class action lawsuit in the United States, has used its participation in the Certification process to demonstrate its commitment to gender equality By 2010, 42 firms with around 170,000 employees have completed the process and obtained the seal, and 550 persons have received training on how to implement gender equity action. Gender committees and women’s networks have been launched in & between certified firms. Firms report a better labor atmosphere including better communication between management and labor and a more motivated workforce. Most firms have incorporated gender equity into recruitment. There are verbal reports from participating firms that some women have been promoted and are in management positions. The success of the program and continued interest among the private sector and Mexican women’s groups are the basis for turning the MEG into a regular government program. Source: IFC (2010) Gender Entrepreneurship and Markets (GEM). http://www.ifc.org/ifcext/enviro.nsf/attachmentsbytitle/art_gemquicknote_mexico/$FILE/ gem+flyer_mexico.pdf 24 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS IMPROVE WOMEN’S OPPORTUNITIES With regard to the joint registration of property IN ENTREPRENEURIAL WORK by men and women who are married, public policy A significant portion of men and women in will need to address ‘demand side’ side issues – Mongolia work in the informal sector, but women e.g. through public advertising campaigns which are mostly unremunerated or unpaid workers, highlight the importance to women of jointly while more men are likely to be self-employed. held assets and their rights in this regard – and There are two main issues with regard to informal “supply side,� issues through measures such as sector work for women. First by definition, the training government employees or providing clear informal sector does not adhere to labor, taxation guidelines on ensuring that all family members or other regulations (usually because of associated are taken into account in the first instance that costs), and this means that maternity, pension and property is registered. gender discrimination laws and regulations that Strengthening women’s opportunities as help women do not get implemented. entrepreneurs – and in the private sector more On the flip side, the informal sector is also a broadly – will require policies that improve the source of high earnings for micro-entrepreneurs, ease with which businesses can be established e.g. which women are not being able to tap to the through policies that reduce the costs of business same degree as men. Limited opportunities for regulations. Mongolia is currently ranked 86th out self-employment, particularly among women of a 171 economies on the overall ease of doing could reflect one of several things. Firstly, it could business, 97th with respect to starting a business, point towards a lack of access to and ownership 119th on dealing with construction permits and of capital and assets that are needed to enter 67th on the ease of obtaining credit. These are into other productive opportunities such as self- fairly poor compared to other countries in the employment. As discussed earlier, in the transition region and also increase the incentives for firms to privatization, property was overwhelmingly to remain in the informal sector. Given how hard registered in the name of male heads of it is to do business in Mongolia, women who lack households reducing the ability of women to access to business networks or finance may be provide collateral for loans. Recent surveys doubly discouraged. Evidence from other parts of indicate that property and asset ownership is still the world suggests that carefully designed policies very unequally distributed.31 Although it is not (such as making business registry cheaper and possible to determine to what extent potential easier, making tax rates proportional to firm size female entrepreneurs suffer from a lack of access etc.) can increase the size of the formal economy to finance, among existing small-scale male and and accelerate growth (Gutierrez-Romero 2010). female entrepreneurs in the informal sector, both Other policies that may be helpful include find it hard to obtain a loan (refer back to Table promoting awareness of and encouraging the 2 ). In addition, women may lack the requisite development of (appropriately regulated and experience, and also access to business networks supervised) micro-lending institutions. Micro- that could help them set up businesses. finance can play an important role in providing credit to those overlooked by large financial institutions, the most famous example of which is the Grameen Bank in Bangladesh. The government can also help to facilitate business networks in cooperation with civil society organizations. Business networks help foster connections and can generate cross-selling activities as well as provide sources of finance, market information and generally serve as a support mechanism for 31 Khas Bank, Measuring the Impact of Microfinance on the Poor Rural Women in Mongolia, Draft Baseline Report, July 2008. individual entrepreneurs who might otherwise feel isolated. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 25 OVERALL SUMMARY AND CONCLUSION The passage of the Gender Equality Law in 2011 with wider benefits for the economy and within marked a substantial achievement for country the household.32 that previously lacked any such law and adds to Accordingly, policy action is needed to improve Mongolia’s previous successes in meeting key labor market outcomes for women and to ensure gender-related Millennium Development Goals that, both men and women, are able to benefit (MDGs) on maternal health and child mortality. from the economic transformation that Mongolia is Mongolian women are also much better educated currently undergoing, including new employment than their peers in the region, and compared to opportunities that are opening up in mining and Mongolian men. However, as this policy note services. A range of potential policy actions can documents, despite these achievements there be considered, including improving employment are considerable differences between men and outcomes (wages, career progression) for women women in labor markets, in terms of participation, in the public sector, introducing more friendly wages and occupational roles. Women also have parental leave policies that cover both fathers a limited presence in higher level managerial and mothers, improving child care services and positions and in entrepreneurial work, and introducing affirmative action policies in sectors working women also have to shoulder most of where women are acsutely under-represented the household and care duties compared to such as mining. In addition, business regulations men. These inequalities can have large impacts could be streamlined to make it easier to start and on development, growth and productivity as well operate businesses for both men and women. as pervasive intergenerational social costs. By the Other policies that may be helpful include same token, removing impediments to full and promoting awareness of and encouraging the equal participation for women in the economy, development of (appropriately regulated and providing equal access to economic resources and supervised) micro-lending institutions opportunities and eliminating discrimination can boost productivity and competitiveness for firms, 32 See the World Bank World Development Report (2012a)�Gender Equality and Development� for a review 26 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS REFERENCES ADB (2010) Country Gender Assessment Review of Development Economics, 8(2), (forthcoming). 198-222. ADB (2010b) Description of the Pre-school Liu, M. and Y. Yin (2010), ‘Human Development in System http://www.adb.org/documents/ East and Southeast Asian Economies: 1990s- supplementary-appendixes/43127/43127- 2010’, United Nations Human Development mon-sa-c.pdf Program, Human Development Reports Research Paper 2010/17. ADB (2008), Mongolia: Education Sector ADB Evaluation Study (2008), Asian Development Nickell, S. (1980), ‘A picture of male unemployment Bank. in Britain’, Economic Journal, 90 (363), 776- 794, December. Altonji and Blank (1999),’Race and Gender in the Labor Market’, in Handbook of Labor NSO (2008), National Statistical Office Yearbook, Economics (eds. Ashenfelter, O. and D. E. Mongolia Statistical Office. Card), Chapter 48, Vol. 3, Part 3, 3123-3259. Oaxaca, R. (1973), ‘Male-female differentials Aslam, M., Kingdon, G. and M. Söderbom (2008), in urban labour markets’, International ‘Is Education a Path to Gender Equality in Economics Review, 3: 603-709. the Labour Market? Evidence from Pakistan’, in Tembon, M. and L. Fort (eds.) Educating Pastore, F. (2008), ‘School-to-work Transitions Girls for the 21st Century: Gender Equality, in Mongolia’, ILO Employment Sector, Empowerment and Economic Growth, 2008. Employment Sector working paper No. 14. Washington D.C: The World Bank. Pastore, F. (2009), ‘The Gender Gap in Early Career CIA World Factbook, https://www.cia.gov/library/ in Mongolia’, The Institute for the Study publications/the-world-factbook/ of Labor (IZA), Discussion paper series No. 4480, October 2009. Emmett, B. (2009), ‘Paying the Price for the Economic Crisis’, Oxfam International Reva, A. Heltberg, R., Sodnomtseren Altantsetseg Discussion Paper, Oxfam GB. and J. Sarantuya (2010) The Impacts of Economic Crisis in Mongolia: Findings from Flintan, F. (2008), ‘Women’s Empowerment in Focus Group Discussions, Discussion Draft Pastoral Societies’, UNDP 2008. Robinson, B. and Solongo, A. (2000) The Gender Gutierrez-Romero, R. (2010), ‘The dynamics of Dimension of Economic Transition in the informal economy’, CSAE WPS/2010- Mongolia, in The Mongolian Economy: A 07, Department of Economics, University of Manual of Applied Economics for a Country Oxford. in Transition. Editors: F. Nixson, B. Suvd, P. Luvsandorj, B. Walters . HDR (2009), Human Development Report, United Nations, available online http://hdr.undp. Steiner-Khamsi, G. and Gerelmaa, A. (2008), org/en/reports/ ‘Quality and Equity in the Mongolian Education Sector’, Prospects: Quarterly IFC (2010) Gender Entrepreneurship and Markets Review of Comparative Education, 38 (3), (GEM) 409-414 Sep 2008. h t t p : / / w w w. i f c . o r g / i f c e x t / e n v i r o . n s f / TIMSS (2008), TIMSS 2007 International attachmentsbytitle/art_gemquicknote_ Mathematics Report, TIMSS and PIRLS mexico/$FILE/gem+flyer_mexico.pdf International Study Centre, Lynch School of Education, Boston College. ILO (2004), Breaking Through the Glass Ceiling, Update 2004, International Labor Office, TIMSS (2008), TIMSS 2007 International Science Geneva. Report, TIMSS and PIRLS International Study Centre, Lynch School of Education, Boston Kingdon, G. and John. Knight (2004), ‘Race and the College. incidence of unemployment in South Africa’, MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 27 UNDP Asia Pacific Human Development Report World Bank (2012a), World Development Report: (2010), ‘Power, Voice and Rights: A turning Gender Equality and Development. point for gender equality in Asia and the Pacific’, Macmillan Publishers India Ltd. World Bank (2012b), Toward Gender Equality in UNESCAP (2007), ‘Gender Equality and East Asia and the Pacific: A Companion to the Empowerment: A Statistical Profile of the World Development Report ESCAP region’, United Nations Economic and Social Commission for Asia and the Pacific. UNICEF (2009), Situation Analysis of Women and Children in Mongolia, United Nations Children’s Fund, Mongolia. 28 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS APPENDIX A: REGULATIONS RESTRICTING WOMEN’S OCCUPATIONAL CHOICES, 1999-200833 ORDER OF THE MINISTER OF HEALTH AND SOCIAL PROTECTION 13th August 1999 Number A/204 Ulaanbaatar city TO ORDER on the basis of the article 101 and the section 5 of the article 109 of the Law on Labor of Mongolia: 1. To reapprove the “Listing of works prohibited to be performed by women� as an annex 1 and the “Listing of workplace prohibited to be performed by minors� as an annex 2. 2. To oblige senior authorities of organizations and entities to work in compliance with the listing. 3. MINISTER’S ROLE EXECUTIVE S. SONIN Mining • All kinds of underground work, Air Transport • Airplane mechanic Extractive oil exploitation and refining, • Airplane connector Industry use of flammable gas • To perform a technical service for airplane and its engine Leather • All kinds of work using chalk Wood • All works transmitted through Industry (bleaching powder) in its preparation wood generation and water float lifecycle such as tanning, and • To load and unload timber by preparing extract from sodium generation hand and chrome, to carry and industry • To cut, sort and collect wood in gather a tanned leather the underground storehouse • To compose chlorine of a paper mill • To crush wood 33 Annulled in 2008 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 29 Glass • Blow a glass by mouth Construction • To assemble a stovepipe and industry • To crush a carbide by hand assembly canal • To dig a well • To pour, broke and crush a platen by hand • To crush a rock • Autoclave men Textile • To prepare aniline salt and flux Social • Clean the drainage filter industry by hand Services • Plumber of drainage canal • To prepare hydrochloride by hand Publishing • Printing machine Common • Antenna work industry • To develop and poison a work and • To pot bitumen and asphalt picture, zinc and an offset profession • Mountain rescue work printing • Parachutist and fireman • To mould a stereotype • To work in contact with mercury • To melt lead • To blend and dye paint with • Letter roll mercury • Machinist of an one-sided chop • Stoker of all kind of heat boilers machine Tailor • Machinist of a minute and • To clean, dye, repair and seal Industry separate chopping machine tank and cistern going inside of used for incise transmission it, which stores flammable and • To operate a special function greasing materials iron press • To mix ethyl with fuel Meat • To cause numbness of cattle, • To clean heat boilers, stovepipe, industry pig and bird, and to execute canal and camera them • To extract and carry coal, lava and • To butcher cattle ash • To work in a storehouse with • To fire heat boilers, repair and amikan refrigerator clean boiler house, carry fuel and ash • To melt, cast, discharge and pack cast iron and metal 30 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS Railway • Train connector and composer • To melt, pour, stretch, cast, mould transport • Install twin wheel band and rim and scroll pot lead, to make and seal an accumulator, to plumb • Bridge locksmith (plumber) cable, to generate a pot lead • To tan, lift and unload bearer made things by hands (tie) • To melt a base metal such as pot • To clean, bloat and blow lead, copper, mercury, gold, zinc stovepipe and silver out of ore • Padding men • To seal channel and canal • To go under the train for service • Striker and control check • Auto machine operator and • Maintenance work of diesel- presser that prints metal in a cold locomotive, supply equipment, condition with a power higher fuel and electric motor than 25 tones • To block in a road by an electric • Base machinist of drum chipping blocker • Smasher • Driver of heavy vehicles for • Steam auto machine’ wheel road maintenance • Stove/boiler repairmen Water • Diesel engineer of a ship • To seal and repair high pressure transport • Steersman/pilot, sailor-man canal • To dive • To pour an electric filter • Ship mechanic • Machine operator and repairmen of a stationary diesel electric • To put a seamark station • Compressor man Auto • Vehicle with more than 2.5 ton • To polish and cover metal by a Transport of carrying capacity, driver of a chemical method vehicle with more than 25 seats • Flammable and greasing material, storing reservoir, to repair internal part of cistern, to seal Source: Government of Mongolia and World Bank MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 31 APPENDIX B: SUMMARY STATISTICS, MALE AND FEMALE (AGED 15-65), MLFS 2009 Profile of women aged 15-65 in All Rural Urban Old Mongolia Age 34.5 34.4 34.5 22.1 44.0 Proportion young (%) 43.6 43.0 44.0 - - Years of Education 10.2 8.4 11.4 9.8 10.5 No education (%) 2.4 5.2 0.7 3.1 1.9 Primary (%) 9.6 19.3 3.5 10.2 9.2 Basic (%) 21.2 32.0 15.6 25.0 20.0 Secondary (%) 35.0 26.7 40.4 40.0 31.2 Initial Technical (%) 5.4 4.2 6.1 2.2 7.8 Advanced Technical (%) 11.7 7.6 14.2 4.6 17.1 University or more (%) 14.2 5.6 20.0 15.2 13.4 Married (%) 55.8 60.9 52.7 33.1 73.4 Proportion living in rural areas (%) 38.7 - - 38.1 39.1 Living in hh with children aged 15 60.5 65.4 57.4 62.7 58.7 or less (%) Hours worked/week 43.5 39.6 47.4 41.6 44.4 Labor Force Participation rate 59.2 72.6 50.8 44.1 70.9 Among wage earners: Tenure less than 1 year (%) 7.0 8.0 7.0 15.0 4.0 Tenure 1-2 years (%) 13.0 15.0 12.0 28.0 7.0 Tenure 3-4 years (%) 15.0 12.0 16.0 28.0 10.0 Tenure 5-9 years (%) 21.0 17.0 22.0 22.0 20.0 32 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS Tenure 10 year or more (%) 43.0 46.0 42.0 4.5 58.0 Profile of men aged 15-65 in All Rural Urban Young Old Mongolia Age 33.8 33.6 33.8 21.9 44.0 Proportion young (%) 46.1 46.0 46.4 - - Years of Education 9.6 7.8 10.8 8.9 10.2 No education (%) 3.0 5.5 1.2 4.1 2.0 Primary (%) 11.4 22.1 3.8 14.7 8.7 Basic (%) 26.5 37.0 19.1 30.0 23.7 Secondary (%) 34.4 22.8 42.6 36.6 32.6 Initial Technical (%) 6.5 4.9 7.6 2.8 9.7 Advanced Technical (%) 7.5 4.2 9.7 2.6 11.6 University or more (%) 10.7 3.4 15.9 9.5 11.8 Married (%) 57.4 59.8 55.8 25.0 85.1 Proportion living in rural areas (%) 41.4 - - 40.9 41.8 Living in hh with children aged 15 57.3 61.3 54.4 53.8 60.3 or less (%) Hours worked/week 49.9 49.4 50.4 49.1 50.3 Labor Force Participation rate 69.1 80.8 60.9 53.5 82.5 Among wage earners: Tenure less than 1 year (%) 8.0 12.0 7.0 13.0 5.0 Tenure 1-2 years (%) 15.0 15.0 16.0 31.0 8.0 Tenure 3-4 years (%) 17.0 16.0 18.0 27.0 13.0 Tenure 5-9 years (%) 21.0 22.0 21.0 20.0 21.0 Tenure 10 year or more (%) 37.0 32.0 38.0 6.0 51.0 Source: Calculations based on MLFS (2009) MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 33 APPENDIX C: SUMMARY STATISTICS, DISTRIBUTION OF THE LABOR FORCE IN FORMAL AND INFORMAL JOBS, BY GENDER, REGION AND AGE (AGES 15-65) All All Male Female Formal (as % of LF) 24.0 21.9 25.7 Informal (as % of LF) 66.2 68.3 63.9 Of which: Own account 47.7 65.8 27.0 Unpaid 29.8 13.2 48.8 Wage 21.9 20.3 23.7 Other 0.6 0.7 0.5 Rural Formal 10.5 9.2 12.0 Informal 83.4 84.5 82.2 Of which: Own account 46.7 71.4 18.8 Unpaid 44.1 19.9 71.5 Wage 9.0 8.5 9.5 Other 0.2 0.2 0.2 Urban Formal 35.9 33.8 38.1 Informal 50.3 53.1 47.3 Of which: 34 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS Own account 49.3 57.3 39.9 Unpaid 7.9 3.3 13.2 Wage 41.6 37.8 46.0 Other 0.2 1.6 0.9 Young (aged <=30) Formal 18.0 17.0 19.0 Informal 68.7 70.4 66.6 Of which: Own account 35.4 50.8 16.0 Unpaid 42.5 29.2 59.0 Wage 21.8 19.5 24.7 Other 0.3 0.5 0.3 Old (>30 & <=65) Formal 26.8 24.7 29.0 Informal 64.9 67.2 62.5 Of which: Own account 54.5 74.4 32.6 Unpaid 22.8 3.9 43.6 Wage 21.9 20.7 23.2 Other 0.8 1.0 0.6 Note: Informal + formal in any column for a category doesn’t add up to 100% because the labor force includes the unemployed (the remaining difference from the 100% sum is the corresponding unemployment rate); ‘Informal’ work is defined to include all unpaid workers (regardless of whether they work in formal/informal enterprises), own-account workers in informal enterprises (employing less than or equal to 5 persons), wage workers in informal enterprises (workers do not receive pensions) and ‘others’ to include employers and members of cooperatives employing less than or equal to 5 persons. Persons reporting primary and secondary ‘informal’ jobs are reported. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 35 36 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS APPENDIX D. SELECTED PARTIAL EFFECTS ON THE LIKELIHOOD OF OCCUPATIONAL OUTCOME, by gender Males Females 1. Out of the labor force Years of schooling -0.027 -0.042 (-10.94)*** (-19.06)*** Household size -0.004 -0.004 (-0.67) (-0.78) Kids under 15 -0.012 0.023 (-1.12) (2.52)** Married -0.118 0.008 (-5.13)*** (0.40) 2. Unemployed Years of schooling 0.0003 0.0004 (0.30) (0.42) Household size 0.004 0.006 (1.27) (2.67)*** Kids under 15 0.004 0.003 (0.75) (0.79) Married -0.031 -0.020 (-2.53)** (-2.21)** 3. Unpaid worker Years of schooling -0.0002 -0.001 (-8.10)*** (-13.71)*** Household size 0.0001 -0.0004 (2.30)** (-1.89)* Kids under 15 -0.0001 0.0004 (-1.57) (1.64) Married -0.0002 0.007 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 37 (-0.85) (7.37)*** 4. Agricultural worker Years of schooling -0.022 -0.004 (-14.69)*** (-7.75)*** Household size -0.012 -0.001 (-3.86)*** (-0.73) Kids under 15 0.016 -0.000 (3.87)*** (-0.23) Married 0.021 -0.004371 (2.06)** (-1.09) 5. Non-agriculture self employed Years of schooling 0.005 0.003 (3.23)*** (2.40)** Household size 0.005 0.002 (1.22) (0.77) Kids under 15 -0.001 0.002 (-0.17) (-3.00)*** Married 0.045 0.026 (2.90)*** (2.58)*** 6.Wage worker Years of schooling 0.044 0.044 (22.38)*** (27.42)*** Household size 0.006 -0.003 (1.12) (-0.69) Kids under 15 -0.007 -0.012 (-0.77) (-1.79)* Married 0.084 -0.017 (4.50)*** (-1.06) Note: These results are based on the multinomial logits not reported here; Robust z-statistics in parentheses. * denotes significant at 10% level; ** significant at 5% level; *** significant at 1% level; Explanatory variables include age, age squared, years of schooling, household size, number of children aged 15 or less, whether person is married or not, whether person is household head or not, whether person belongs to a female headed-household or not and aimag-level fixed effects. 38 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS The charts below are based on multinomial logit models that estimate predicted probabilities for being in either of the following outcomes: out of the labor force, unemployed, unpaid family worker, self employed in agriculture, self employed otherwise and wage worker. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 39 APPENDIX E: ACCESS TO FINANCE male female All males At least Total (single) (single) 1 female (multiple (multiple owner firms ) owner firms) % of firms with Bank Account 61% 69% 58% 52% 61% % Investment financed 82% 68% 70% 67% 73% internally % of firms with Bank loan 58% 58% 49% 56% 57% Median Loan Value (millions of 46.5 40 200 45 49 MNT) Median Collateral Val as % of 40% 40% 45% 35% 40% Loan % of firms that applied for loan 55% 60% 51% 65% 58% in 2007 % loan rejected 26% 13% 13% 33% 22% % firms with results audited 82% 76% 89% 79% 80% Source: Business Environment and Enterprise Performance Survey (BEEPS) conducted in 2008/2009, World Bank. The survey consisted of 362 firms that were present on the Mongolian business register and are thus are formal sector firms. 40 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS APPENDIX F: FACTORS THAT EXPLAIN PARTICIPATION RATES Table (a) Marginal effects of female participation in the labor market All By location By age category Labor Force (1) (2) (3) Participation: Rural Urban Young Old (15-30 years) (31-65 years) Years of 0.022 0.003 0.037 0.014 0.022 schooling (11.94)*** (1.20) (14.41)*** (5.03)*** (10.95)*** Age 0.126 0.085 0.148 0.270 0 . 0 7 4 (9.06)*** (28.45)*** (15.91)*** (22.51)*** (10.78)*** Age squared -0.002 -0.001 -0.002 -0.005 -0.001 (-30.27)*** (-16.90)*** (-24.12)*** (-8.30)*** (-11.23)*** Kids under 15 in -0.013 -0.006 -0.019 0.021 -0.036 household (-1.93)* (-0.72) (-2.04)** (2.03)** (-4.78)*** Married 0.044 0.153 -0.030 -0.033 0.061 (2.63)*** (6.18)*** (-1.38) (-1.40) (2.48)** Female headed 0.014 0.028 0.003 0.040 0.004 household (0.83) (1.16) (0.14) (1.62) (0.17) Aimag fixed yes yes yes yes yes effects Observations 8923 3450 5473 3887 5036 Pseudo-R2 0.287 0.259 0.295 0.313 0.214 Mean of dependent 0.592 0.726 0.508 0.441 0.709 variable Note: * denotes significant at 10% level; ** significant at 5% level; *** significant at 1% level; estimation method is probit, and robust z-values reported; dependent variable = 1 if person participates in labor market, 0 otherwise; Model (1) also controls for ‘young’ and ‘rural’ dummy, (2) controls for ‘young’ and (3) controls for ‘rural’ dummy. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 41 Table (b) Marginal effects of male participation in the labor market All By location By age category Labor Force (1) (2) (3) Participation: Rural Urban Young Old (15-30 years) (31-65 years) Years of 0.013 -0.003 0.032 0.009 0.011 schooling (7.66)*** (-1.42) (10.99)*** (2.72)*** (6.90)*** Age 0.119 0.070 0.165 0.330 0.046 (33.13)*** (18.06)*** (27.25)*** (11.48)*** (8.28)*** Age squared -0.001 -0.001 -0.002 -0.006 -0.001 (35.40)*** (-18.62)*** (29.41)*** (-8.96)*** (-10.26)*** Kids under 15 in 0.019 0.010 0.026 0.044 0.004 household (2.93)*** (1.60) (2.56)*** (3.69)*** (0.66) Married 0.139 0.053 0.182 0.221 0.082 (8.75)*** (2.78)*** (7.91)*** (7.72)*** (4.72)*** Female headed -0.034 -0.029 -0.039 -0.022 -0.080 household (-1.84)* (-1.17) (-1.44) (-0.84) (2.48)** Aimag fixed Yes yes yes yes yes effects Observations 8227 3406 4821 3789 4438 Pseudo-R2 0.371 0.317 0.390 0.413 0.232 Mean of dependent 0.691 0.808 0.609 0.535 0.825 variable Note: * denotes significant at 10% level; ** significant at 5% level; *** significant at 1% level; estimation method is probit, and robust z-values reported; dependent variable = 1 if person participates in labor market, 0 otherwise; Model (1) also controls for ‘young’ and ‘rural’ dummy, (2) controls for ‘young’ and (3) controls for ‘rural’ dummy. 42 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS Table (c) – Marginal effects for male and female participation – the effect of education Education Level Male Female Difference is Completed: significant? (Base: no education or primary) Basic -0.064 *** -0.027 *** (-3.06) (-1.13) Secondary -0.050 ** -0.042 * (2.35) (-1.74) Initial technical 0.051 * 0.074 ** (1.81) (2.30) Ad. Technical 0.030 0.096 *** *** (1.09) (3.61) University/ more 0.135 *** 0.219 *** *** (7.01) (9.81) Aimag fixed effects Yes Yes Observations 8227 8923 Pseudo-R2 0.378 0.295 Note: * denotes significant at 10% level; ** significant at 5% level; *** significant at 1% level; estimation method is probit, and robust z-values reported; dependent variable = 1 if person participates in labor market, 0 otherwise; Controls include: age, age squared, kidsunder15, married, female-headed-household. Overall notes Levels of schooling are defined as: no education (a dummy variable equalling 1 if person has no education, 0 otherwise), primary (a dummy variable equalling 1 if person has completed 4 years of schooling, 0 otherwise), basic (a dummy variable equalling 1 if person has completed 8 years of schooling, 0 otherwise), secondary (a dummy variable equalling 1 if person has completed 10 years of schooling, 0 otherwise), initial technical (a dummy variable equalling 1 if person has completed 12 years of schooling, 0 otherwise), advanced technical (a dummy variable equalling 1 if person has completed 13 years of schooling, 0 otherwise) and university or more (a dummy variable equalling 1 if person has completed 17 or more years of schooling, 0 otherwise). The variable ‘kidsunder15’ captures the number of persons aged 15 or less in the household. Ideally one needs the number of own children of the woman in a household to identify the true effect of presence of children on labor market outcomes. The MLFS (2009) dataset does not allow us to match children to parents and this is the best variable capturing household dynamics available to us. As a robustness check, we restricted the sample of women aged 15-65 to only those who are household heads or wives of heads with the view that this allows a better matching of biological children to parents. As expected, this causes the effect of presence of children on the probability of participation to become even larger and more precise. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 43 APPENDIX G: PROBABILITY AND DURATION OF UNEMPLOYMENT Table g1 Narrow and broad unemployment Broad Definition Narrow Definition Age Male Female Male Female 15-65 9.8 10.4 7.8 8.6 Young (15-30) 12.7 14.5 9.7 11.3 Old (31-65) 8.2 8.4 7.0 7.4 Education None/Primary 5.7 4.5 3.5 2.6 Basic/Secondary 11.5 13.5 9.3 10.6 More than Secondary 8.7 8.9 7.6 8.0 Location Rural 6.3 5.8 4.5 4.1 Urban 13.1 14.6 11.2 12.8 Region West 8.5 8.2 8.2 6.8 Khangai 8.5 8.9 8.3 7.2 Central 9.5 9.1 7.9 7.6 East 10.9 9.7 8.0 6.7 Ulaanbaatar 11.4 13.7 7.2 11.9 Source MLFS (2009). The ‘narrow’ definition includes individuals who are not currently employed but who actively looked for work in a given time period. The ‘broad’ definition includes the narrow unemployed plus those who say they want work but did not look for work in a given time period. Also see Kingdon and Knight (2004). 44 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS Table g2: Probability of unemployment (among all labor force participants) All Male Female Probability of being M a r g i n a l Robust M a r g i n a l Robust M a r g i n a l Robust unemployed: effect z-value effect z-value effect z-value Personal characteristics Age*100 0.204 (0.95) 0.310 (1.07) 0.340 (1.05) Age2*100 -0.003 (-1.13) -0.003 (-0.94) -0.005 (-1.36) Male -0.004 (-0.73) - - - - Young 0.035 (3.02)*** 0.028 (1.76)* 0.044 (2.56)** Education (Base: university or more) No education 0.039 (1.69)* 0.019 (0.67) 0.062 (1.62) Primary -0.029 (-2.66)*** -0.035 (-2.51)** -0.026 (-1.47) Basic 0.030 (2.74)*** 0.012 (0.85) 0.049 (2.92)*** Secondary 0.054 (5.76)*** 0.035 (2.72)*** 0.074 (5.49)*** Initial Technical 0.023 (1.64) -0.004 (-0.24) 0.056 (2.37)** Ad. Technical -0.024 (-2.49)** -0.024 (-1.70)* -0.017 (-1.34) Other variables Kids under 15 0.011 (3.88)*** 0.008 (2.01)** 0.014 (3.72)*** Female headed 0.010 (1.17) 0.069 (3.78)*** -0.011 (-0.97) Married -0.058 (-6.78)*** -0.072 (-5.78)*** -0.056 (-4.36)*** Rural -0.095 (-12.53)*** -0.087 (-8.38)*** -0.104 (-9.66)*** Region yes yes yes Observations 10967 5686 5281 Pseudo-R2 0.072 0.068 0.088 Mean of dependent variable 0.101 0.098 0.104 Note: * denotes significant at 10% level; ** significant at 5% level; *** significant at 1% level; estimation method is probit, and robust z-values reported; dependent variable = 1 if person is unemployed (broadly defined) i.e. not working and may or may not be looking for work, 0 otherwise; controls include ‘young’ dummy. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 45 Table g3: Duration of unemployment by age, education, location and region Duration of Unemployment % duration of unemployment (average months) <1 1-2 3-6 7-11 1-2 3 or > month months months months years years Males Age 15-30 28.6 2.7 3.9 12.4 8.5 25.2 47.3 31-65 33.5 1.3 3.7 7.7 7.7 19.4 60.2 Education None or primary 39.5 0.0 0.0 6.3 2.1 16.7 75.0 Basic or secondary 32.4 1.7 4.1 9.1 8.0 19.6 57.6 More than 25.8 3.4 4.1 13.0 10.3 30.1 39.0 secondary Location Rural 35.2 2.9 2.3 8.7 5.2 15.1 65.7 Urban 29.5 1.6 4.4 10.4 9.4 25.2 49.1 Region West 35.3 1.1 3.2 4.3 4.3 23.7 63.4 Khangai 34.7 0.9 4.4 6.1 6.1 19.3 63.2 Central 31.8 5.3 3.2 16.0 2.1 14.9 58.5 East 34.0 2.3 9.1 9.1 6.8 6.8 65.9 Ulaanbaatar 26.8 1.4 2.8 11.8 13.7 29.3 41.0 Type of U Searching 31.1 1.6 2.5 10.4 9.0 23.3 53.3 Non-search 31.8 3.5 8.8 7.9 4.4 17.5 57.9 Females Age 15-30 26.0 3.6 4.4 13.3 10.4 28.1 40.2 46 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 31-65 28.6 1.7 3.3 7.0 4.0 23.6 60.5 Education None or primary 42.4 0.0 0.0 7.1 3.6 3.6 85.7 Basic or secondary 32.0 1.2 2.9 7.9 5.3 28.5 54.3 More than 25.3 5.5 6.1 13.8 10.5 23.8 40.3 secondary Location Rural 35.1 1.4 2.8 6.9 5.5 19.3 64.1 Urban 28.7 3.0 4.2 10.9 7.4 27.9 46.7 Region West 32.6 0.0 2.4 13.1 4.8 22.6 57.1 Khangai 29.8 1.0 5.3 12.4 3.5 28.3 49.6 Central 34.0 1.2 3.6 11.9 2.4 19.1 61.9 East 33.6 5.6 8.3 5.6 5.6 11.1 63.9 Ulaanbaatar 28.0 4.3 3.0 7.3 11.2 30.0 44.2 Type of U Searching 29.9 2.5 3.8 9.8 7.6 26.6 49.8 Non-search 32.5 2.9 3.9 9.8 3.9 21.6 57.8 Note: Duration of unemployment in the MLFS (2009) is a categorical variable (less than 1 month, 1-2 months etc.). To arrive at the mean monthly duration variable, we attribute the mid-points of the category to create a continuous variable. Table g4: Determinants of unemployment duration Determinants of All Male Female unemployment Robust Robust Robust duration: Coefficient Coefficient Coefficient t-value t-value t-value Personal characteristics Age 2.066 (4.14)*** 1.478 (2.07)** 2.697 (3.61)*** Age2 -0.022 (-3.57)*** -0.016 (-1.81)* -0.029 (-3.10)*** MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 47 Male 0.306 (0.28) - - - - Young 2.594 (1.11) 2.068 (0.59) 3.215 (0.99) Education No education 15.139 (4.76)*** 14.461 (3.46)*** 14.210 (2.79)*** Primary 15.995 (6.19)*** 14.473 (4.11)*** 17.653 (4.65)*** Basic 7.927 (4.03)*** 8.641 (3.08)*** 6.871 (2.48)** Secondary 7.620 (4.65)*** 7.706 (3.21)*** 7.371 (3.23)*** Initial Technical 8.225 (3.36)*** 10.711 (3.04)*** 5.525 (1.61) Advanced 0.753 (0.28) -2.125 (-0.51) 2.247 (0.63) technical Household demographic variables Married -3.582 (-2.56)** -3.166 (-1.62) -3.606 (-1.77)* Kids under 15 0.999 (2.15)** 1.032 (1.44) 0.897 (1.43) Female-headed 0.667 (0.45) -1.579 (-0.67) 1.655 (0.84) HH Location Rural 2.482 (1.75)* 1.409 (0.73) 3.467 (1.63) Region West 2.482 (1.98)** 4.804 (1.92)* 2.326 (0.92) Khangai 2.458 (1.60) 5.567 (2.51)** -0.231 (-0.11) Central 2.017 (1.08) 2.470 (0.93) 1.734 (0.64) East 0.695 (0.27) 1.409 (0.40) 0.265 (0.07) Observations 1107 557 550 R2 0.119 0.112 0.144 Mean of dependent 30.801 31.238 30.359 variable Note: * denotes significant at 10% level; ** significant at 5% level; *** significant at 1% level; estimation method is ordinary least squares (OLS), and robust t-values reported; dependent variable = unemployment duration (months), see definition in Note under Table A5; controls include ‘young’ dummy. 48 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS APPENDIX H: EARNINGS GAPS Table h1: Raw mean wages (thousands MNT/month) for wage workers (aged 15-65) Male Female Gap (M-F) Overall 235.1 213.8 21.37*** By education level No education 160.8 51.4 109.4 Primary 131.3 118.3 13.0 Basic 178.6 150.4 28.3*** Secondary 213.6 176.0 37.6*** Initial Technical 219.1 186.7 32.4*** Advanced Technical 257.7 223.1 34.7*** University or more 290.7 257.5 33.2*** By Location Urban 245.6 219.5 26.1*** Rural 198.9 194.7 4.3 By Age group Young 229.7 208.3 21.4*** Old 237.6 215.9 21.7*** By Region West 191.5 205.1 -13.6* Khangai 230.1 208.8 21.3** Central 218.9 201.4 17.5** East 212.1 193.6 18.5 Ulaanbaatar 254.1 223.7 30.4*** By Industry sector Primary 253.4 222.4 31.0* Secondary 236.8 210.0 26.8*** Tertiary 231.6 214.1 17.5*** By Occupation group Legislators 287.1 265.4 21.7** Specialists 286.6 246.4 40.2*** Other specialists 248.6 227.4 21.2*** Clerks 197.1 205.6 - 8.5 Sales 195.2 158.7 36.5*** Agriculture 133.5 113.3 20.2 Construction 229.8 199.8 30.0*** Operators 220.6 223.3 - 2.7 Elementary 167.0 145.6 21.4*** By formal/informal Informal sector wage job 202.4 185.2 17.2*** Formal sector wage job 257.3 231.2 26.3*** Source: MLFS (2009) data. MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 49 50 Table h2: Wage earnings and years of schooling by location, age and gender (ages 15-65) Log Monthly All Mongolia By Location By Age Earnings: (1) (2) (3) Urban Rural Young Old All Male Female All Male Female All Male Female All Male Female All Male Female Years of 0.054 0.049 0.058 0.052 0.047 0.057 0.059 0.056 0.064 0.048 0.035 0.064 0.057 0.058 0.056 schooling (28.13)*** (17.74)*** (22.44)*** (25.66)*** (15.82)*** (20.77)*** (12.71)*** (8.67)*** (9.08)*** (12.81)*** (7.02)*** (12.86)*** (24.73)*** (17.33)*** (17.51)*** Age 0.005 -0.007 0.019 0.013 0.004 0.024 -0.027 -0.062 0.005 -0.003 0.026 -0.048 0.002 -0.009 0.014 (0.79) (-0.80) (1.89)* (1.87)* (0.49) (2.12)** (-1.64) (-2.61)*** (0.27) (-0.45) (0.36) (-0.59) (0.24) (-0.63) (0.78) Age2* -0.007 0.005 -0.021 -0.016 -0.007 -0.028 0.003 0.066 0.000 0.059 -0.052 0.116 -0.005 0.007 -0.016 100 (-0.97) (0.50) (-1.75)* (-2.03)** (-0.68) (-2.08) ** (1.67)* (2.43)** (0.03) (0.53) (-0.36) (0.72) (-0.37) (0.43) (0.76) Tenure 1-2 0.131 0.102 0.146 0.135 0.114 0.140 0.105 0.080 0.005 0.099 0.082 0.095 0.180 0.139 -0.000 years (4.11)*** (2.29)** (3.28)*** (3.71)*** (2.40)** (2.62)*** (1.64) (0.82) (1.79)* (2.62)*** (1.56) (1.81)* (3.23)*** (1.79)* (2.82)*** Tenure 3-4 0.168 0.081 0.245 0.187 0.123 0.238 0.088 0.080 0.269 0.122 0.064 0.163 0.208 0.0912 0.331 years (5.26)*** (1.79)* (5.58)*** (5.28)*** (2.69)*** (4.58)*** (1.17) (-0.61) (3.07)*** (2.86)*** (1.06) (2.90)*** (4.15)*** (1.33) (4.64)*** Tenure 5-9 0.268 0.214 0.308 0.269 0.222 0.238 0.262 0.203 0.319 0.207 0.174 0.224 0.313 0.231 0.389 years (8.60)*** (4.86)*** (7.00)*** (7.61)*** (4.76)*** (5.71)*** (4.06)*** (2.16)** (3.68)*** (5.06)*** (2.90)*** (4.10)*** (6.45)*** (3.49)*** (5.60)*** Tenure >=10 0.354 0.305 0.381 0.345 0.313 0.353 0.374 0.289 0.447 0.274 0.279 0.218 0.393 0.316 0.461 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS years (11.07)*** (6.88)*** (8.46)*** (9.57)*** (6.76)*** (6.58)*** (5.92)*** (3.08)*** (5.25)*** (5.30)*** (3.65)*** (3.22)*** (8.43)*** (4.96)*** (6.88)*** Married 0.014 0.065 -0.017 0.015 0.047 -0.005 0.012 0.113 -0.049 0.048 0.068 0.022 -0.009 0.045 -0.035 (0.99) (2.79)*** (-0.97) (1.00) (2.09)** (-0.24) (0.33) (1.55) (-1.13) (2.03)** (2.00)** (0.68) (-0.50) (1.38) (-1.66)* Male 0.152 - - 0.162 - - 0.123 - - 0.167 - - 0.148 - - (12.77)*** (12.77)*** (4.03)*** (7.64)*** (10.26)*** N 4114 1973 2141 3173 1530 1643 941 443 498 1219 617 602 2895 1356 1539 Rsq 0.313 0.301 0.328 0.308 0.278 0.326 0.266 0.266 0.307 0.319 0.269 0.402 0.317 0.330 0.309 Mean dep var 5.31 5.40 5.30 5.37 5.43 5.32 5.18 5.18 5.18 5.31 5.36 5.26 5.33 5.38 5.29 Note: * denotes significant at 10% level; ** significant at 5% level; *** significant at 1% level; estimation method is ordinary least squares (OLS), and robust t-values reported; Tenure less than 1 year is excluded category; Controls include regional fixed effects and ‘young’ dummy (except in model 3), rural dummy (except in model 2). The coefficient on the gender variable i.e “Male� in the table indicates the percentage by which men are paid more than women. For example in the first column, the coefficient estimate is 0.152, indicating that after taking into account all factors such as years of schooling, age, work experience and demographic characteristics, if the worker is male then he will be paid 15 percent more than a woman with similar characteristics. APPENDIX L: OAXACA’S DECOMPOSITION Oaxaca’s (1973) technique entails decomposing the wage gap into two components: 1) that explained by differences in individual characteristics and 2) the residual, unexplained portion, reflecting differences in earnings structure. Assume that the mean earnings of females (f) are Yf and those of males (m) are Ym. Mean earnings will be determined by: Yi = biXi where i=m,f (1) where X is the vector of average characteristics of i and bi is the vector of estimated parameters for i. Standardizing by male means, the total wage gap in mean earnings can be divided into the explained (E) component and the unexplained (D) component as follows: Wage Gap = Ym – Yf Wage Gap = bmXm – bfXf Wage Gap = {Xm (bm– bf)} + {bf (Xm – Xf)} Wage Gap = Unexplained Component+ Explained Component (2) The term {bf (Xm – Xf)} represents the explained component. In other words it is the part of the wage gap due to differences in the average characteristics of men (Xm) and women (Xf). If average characteristics were the same, then this term would be zero. If men have better characteristics than women, then a positive share of the gap is explained. The term {Xm (bm– bf)} represents the unexplained component, i.e. they are due to differences in sex or how the labor market values differences in sex for a given set of characteristics (e.g male). Table (a) Oaxaca Decomposition of log monthly wages Standardizing by male means Standardizing by female means Unexplained Explained Total gender Unexplained Explained Total gender (D) (E) wage gap (D) (E) wage (log monthly gap (log earnings) monthly earnings) Basic 0.198 -0.015 0.184 0.202 -0.019 0.184 Earnings Function With 0.236 0.003 0.239 0.252 -0.013 0.239 Occupation and Industry controls MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS 51 Young (with 0.338 -0.042 0.296 0.337 -0.041 0.296 occupation/ industry controls) Old (with 0.148 0.042 0.190 0.176 0.014 0.190 occupation/ industry controls) Note: Estimates (a) based on results reported in model (1) in Table (a) in Appendix K; estimates for columns (b), (c) and (d) based on results not reported here. 52 MONGOLIA: GENDER DISPARITIES IN LABOR MARKETS AND POLICY SUGGESTIONS