GENDER RELATIONS IN EUROPE AND CENTRAL ASIA RESULTS FROM THE LIFE IN TRANSITION SURVEY III Nazli Aktakke (Development Analytics), Meltem Aran (Development Analytics), Ana Maria Munoz Boudet (World Bank, Poverty and Equity Global Practice) Summary This report presents a detailed gender analysis of the Life in Transition Survey III (LITS III), conducted in 2015-2016 in 34 countries of the Europe and Central Asia region. LITS is a unique dataset as it covers for the first time issues related to asset ownership, care need in the household and gender norms. In addition, in a subset of the households that participated information was collected for both men and women allowing an analysis of gender intra-household dynamics. In total, over 51,000 respondents were covered in the survey. The results indicate that while gender relations, views, and women’s access and use of some opportunities have transitioned to more egalitarian ones, women still face challenges to fully participate in economic activity and have an equal say in household decision-making. Across all countries, women’s employment appears to be a main driver of greater quality across the domains analyzed in the report. 2 Table of Contents Summary............................................................................................................................................................................... 2 Women’s economic participation ..................................................................................................................................... 8 A. Employment .......................................................................................................................................................... 8 B. Entrepreneurial activity ...................................................................................................................................... 13 Care: Needs and Provision .............................................................................................................................................. 15 Spotlight: Female Headed Households .......................................................................................................................... 22 Asset Ownership ............................................................................................................................................................... 22 A. Land and Dwelling Ownership ......................................................................................................................... 23 B. Bank Account Ownership ..................................................................................................................................... 29 Norms and Voice in Household Decisions .................................................................................................................. 31 A. Voice in household decisions ............................................................................................................................ 37 Conclusions........................................................................................................................................................................ 40 Annexes .............................................................................................................................................................................. 42 Annex 1 Data Cross Checks Educational Attainment, Employment Rates in LITS III Data ........................ 42 Annex 2. Construction of the Empowerment Indices ........................................................................................... 44 Annex 3. Laws on Ownership of property .............................................................................................................. 45 Annex 4. Cross tabulations ......................................................................................................................................... 46 Annex 5 Regression Results ....................................................................................................................................... 58 Regression 1: How are men’s characteristics like education or employment associated with women’s employment?............................................................................................................................................................. 58 Regression 2: How is entrepreneurship associated with education, agreement to norms, being empowered in the household and living in a specific country for men and women? ........................................................ 58 Regression 3: How is asset ownership associated with individual characteristics? ....................................... 61 Regression 4: How is care need in the household associated with employment for men and women?.... 67 Regression 5: How is use of institutional child care associated with women’s employment? ..................... 68 Regression 6: How is agreement with norms associated with observable characteristics of women such as education, age, owning an asset and employment? ............................................................................................ 69 Regression 7: How is having a say in household decisions associated with observable characteristics of women and men that the women are living together with? .............................................................................. 70 References .......................................................................................................................................................................... 73 3 Table of Figures Figure 1 The third round of Life in Transition Survey covers 34 countries and it is representative at the country level ....................................................................................................................................................................................... 6 Figure 2: The educational attainment levels are similar while employment ratios vary for men and women ..... 9 Figure 3 For households where the primary and secondary respondents are a couple in the majority of countries it is most likely that they are both employed ................................................................................................................ 10 Figure 4 Women are more likely to not work because of household duties and they are more likely to take temporary leave ................................................................................................................................................................. 11 Figure 5 Young people not being in education or employment is common in most of the countries ............... 12 Figure 6 More men attempt to be entrepreneurs compared to women but after they attempt to set up a business, men and women have similar success rates. ................................................................................................................. 13 Figure 7 Sector, location, ownership type and assets of business by gender of the owner .................................. 14 Figure 8 13.7% of the population live in a household with children aged 0-6 while living with an elderly is more common with 26% of the population living with an elderly aged 65 or more ....................................................... 16 Figure 9 The fraction of the population living in three generation households -children (aged 0-17), adults (aged 18-64) and elderly (65+) - is negatively correlated with the percentage of the population living in urban areas .............................................................................................................................................................................................. 17 Figure 10 While most of the population (75.3%) live in a household with no care need, variation between countries is high. ............................................................................................................................................................... 18 Figure 11 In the majority of the countries household members are the primary source of care providers for child care, elderly care and disabled care ....................................................................................................................... 19 Figure 12 Utilization of institutional child care is low in many countries ............................................................... 20 Figure 13 Access to institutional child care is positively correlated with the level of welfare in the country. ... 20 Figure 14 Female employment (of women aged 25-40) is negatively correlated with care need ......................... 21 Figure 16 Women’s ownership of a dwelling or land shows wide variation between countries.......................... 23 Figure 17 In most countries, men are more likely to own the dwelling or land than women. Additionally, owning a dwelling solely or jointly is more common among men than women. The same trend also holds for owning land ...................................................................................................................................................................................... 24 Figure 18 In most countries, it is more common for men to own a dwelling (jointly or solely) than women. The same trend holds for land ownership, but land ownership is less common compared than dwelling ownership .............................................................................................................................................................................................. 25 Figure 19 The older the women and men get the more likely are they to own an asset ....................................... 26 Figure 20 On average women who do not own any assets are less likely to work compared to women who own assets ................................................................................................................................................................................... 26 Figure 21 Owning a dwelling or land by living arrangements of women (% of female population), by country .............................................................................................................................................................................................. 27 Figure 22 Being able to afford adequate heating of the dwelling (% of population) , by country ...................... 28 Figure 23 Norms are effective in women’s asset ownership up to a degree ........................................................... 29 Figure 25 Unlike asset ownership, bank account ownership is more egalitarian across genders - except for a few countries. Wide variation in the levels of access to financial services between countries persists. ..................... 30 Figure 26 While in all the countries people agree that equal rights for women as citizens is important, they believe less that it really exists in their country ............................................................................................................ 32 Figure 27 Men tend to agree more with the norms on the distribution of roles in the household but differences between men and women are small. In many countries, a large share of women also agree with the gender norms regarding the distribution of roles in the household ...................................................................................... 33 4 Figure 28 In all of the countries but one, women agree more with the statement “Women are as competent as men to be business executives”. In almost all of the countries women disagree more with the statement “Men make better political leaders than women do”. ............................................................................................................ 34 Figure 29 Women and men generally have similar opinions for their daughters and sons’ university education .............................................................................................................................................................................................. 35 Figure 30 Women’s participation in economic life is associated with norms at the country level ...................... 36 Figure 31 Regardless of their gender, most individuals think that their opinions are considered in decisions made by the household .................................................................................................................................................... 38 Figure 32 Percent of population having a say in the decision (Mostly me/Shared equally between me and my partner/Shared equally between me and someone else in the household), by country ........................................ 39 Figure 33 For a number of countries LITS III data represents the population either as overeducated (Turkey) or undereducated (Azerbaijan) compared to UNESCO statistics ............................................................................ 42 Figure 34 For a number of countries the LITS III dataset overestimates or underestimates the female employment to population ratio or male employment to population ratio by more than 10 percentage points compared to WDI statistics............................................................................................................................................. 43 Annex 5. Table 1 Men’s characteristics and women’s employment – Regression results ..................................... 58 Annex 5. Table 2 Personal characteristics, country effects and entrepreneurship ................................................. 59 Annex 5. Table 3 Women’s asset or bank account ownership, individual characteristics and country effects . 61 Annex 5. Table 4 Men’s asset or bank account ownership, individual characteristics and country effects ....... 64 Annex 5. Table 5 Care needs in the household and employment outcomes .......................................................... 67 Annex 5. Table 6 Using institutional child care and women’s employment ........................................................... 68 Annex 5. Table 7 Women’s observable characteristics and agreement with norms .............................................. 69 Annex 5. Table 8 Women’s having a say in household decisions ............................................................................. 71 5 Introduction1 The Life in Transition Survey (LITS) is conducted jointly by the World Bank and EBRD and collects information about the socio-economic status of households and individuals along with individuals’ perceptions on social, economic and political issues. It was collected for the first time in 2006 in 29 countries to explore the views and attitudes of people living in the transition countries. The second round of the survey was collected in 2010 in 35 countries. And the last round of the survey which is also the main data source used in this report, was collected between the end of 2015 and the beginning of 2016 in 34 countries2. The 34 countries that LITS III covers are composed of 29 transition countries from the Europe and Central Asia region (including Mongolia) along with Cyprus, Turkey and Greece and two western European countries (Germany and Italy- selected to enable comparisons with developed European countries) (See Figure 1). Approximately 1,500 households were interviewed in each country. When taken together this gives a total sample size of 51,206 households. The sample is representative of the population aged 18 or over at the country level. LITS III is a unique Figure 1 The third round of Life in Transition Survey covers 34 countries and it is dataset ideally suited to representative at the country level conducting gender analysis in the Europe and Central Asia region. LITS was already a rich survey in its previous rounds allowing researchers to analyze the relationships between attitudes and individual and household background. However, unlike previous rounds of the survey, LITS III includes new questions and modules permitting in- depth gender analysis. LITS III includes additional questions on asset ownership, care need in the household and gender norms. In addition, responses for two of the modules (assets and employment) were collected from more than one respondent in the household. The additional sample of secondary respondents corresponds, to an adult from the opposite sex than the main respondent -mainly a spouse or partner, but also older children, or other relatives, allowing an analysis of gender intra-household dynamics3. 1 This report is a product of a grant by the Umbrella Facility for Gender Equality by the World Bank. 2 For details please see EBRD’s Life in Transition III, Methodology Annex. 3 Must be noted that the only condition was the interview of an adult of the opposite sext, when possible, the spouse or partner of the main respondent, but not exclusively. Overall the survey consists of the following modules: (1) household roster collecting basic information on the members of the household such as their age, level of education and gender, (2 and 3) assets modules, (4) attitudes and values, (5) employment and unemployment, (6) entrepreneurship, (7) governance. Annex 1 provides some descriptive information educational attainment and employment in the LITS sample when compared to national statistics 6 Findings suggests that while gender relations, views, and women’s access and use of some opportunities are in a good place, women still face challenges to fully participate in economic activity. Household duties and childcare demands negatively affect women during their reproductive years, which coincide with their peak productive years (25-45 years of age) for whom a need for child care is negatively associated with their employment. On the positive side, these women are 27.3 percentage points more likely to be employed if the household utilizes institutional care for at least one child under the age of 6. Across all countries women are at a disadvantage in terms of holding an asset -e.g. land and/or dwelling. We find that asset ownership is not only important in terms of economic security, but also for women’s voice and agency within the household. Results indicate that for those households where the man owns an asset, the woman is less likely to have a say in day-to-day household spending. Using LITS III this report presents findings on five main topics. The first section focuses on employment of men and women in LITS III. Section two looks at entrepreneurial activity. Section three focuses on household composition, care needs and care provision at the household level. Section four focuses on asset ownership across the genders looking mainly at land and dwelling ownership. Section five looks at perceived gender roles, social norms and role/voice of women in decision making processes in the households across countries4. 4 A note should be made on the use of only LITS III for this analysis. While several reports have documented finding from the previous rounds of the survey, changes and improvements in the sample selection suggest that cross-years comparisons are not robust for some countries. 7 Women’s economic participation Opportunities to earn their own income through participating in economic activities can improve the welfare, bargaining power and economic empowerment of women. Wealth and asset ownership in addition to income are some of the core economic sources that improve women’s bargaining power in the household (World Bank, 2012) Considering that, for example, labor earnings make up the largest share of household income, women’s contribution is key. (World Bank, 2013). Yet, globally, labor force participation rates for women still lag behind those of men at 55.3 percent (compared with 81.7 percent of men), and more women than men work as unpaid family workers, lacking the wages or salaries that allows for them not only to contribute to their household’s economy, but also to save or invest in assets and other goods.5 An increase in women’s economic activity can translate to positive outcomes for women themselves, their children and the society as a whole. At the individual level women’s higher access to resources is associated with a higher involvement for women in decision making in the household. Women’s participation in entrepreneurship programs can lead them to have greater bargaining power in the household and more say in household decisions (Pitt et al., 2003). For women, access to a job could also be instrumental in expanding their social networks and providing new opportunities to build skills (World Bank, 2014). This could then lead to a virtuous cycle of personal improvement and access to economic opportunities. Women’s greater access to resources could also lead to better outcomes for their children. A systematic review of 15 studies on economic resource transfers points out that targeting women in transfers could improve children’s wellbeing in terms of better education and health outcomes (Yoong et al.,2012). At the country level, tapping the potential of women is a key to securing economic growth. It is estimated that the total income loss due to gender gaps in labor force participation and gaps in type of work for Kosovo, Turkey and Albania were 28.2, 22 and 19.8 percent respectively. Globally women also tend to participate less in entrepreneurial activities than men (Global Entrepreneurship Monitor, 2017). This results in productivity losses across the sampled countries. The estimated loss in GDP per capita due to the gender gap in entrepreneurship is of 4 to 7 percent in different regions in the world (Cuberes and Teignier, 2012). Same country-level analysis mentioned before that for the same countries, estimates range from 8.3 percent in Turkey, to 6.9 percent in Serbia (Cuberes et al., 2019). A. Employment According to LITS III data, women and men’s employment outcomes are quite unequal across the sampled countries. Overall, 58.9 percent of women of working age (18 to 64 years old) reported working any time in the last 12 months as opposed to 72.3 percent of men.6 Employment rates are higher for individuals at their prime working age (ages 25-54) but the gap between men and women remains. 67.8 percent of women and 82.3 of men in that age group reported being employed at some time during the past year. In each of the 34 countries it is more common for men to be employed (measured as having worked at any time in the past 12 months (See Figure 2). However, the gap between men and women differs a lot in between countries. The gap is highest in Azerbaijan reaching 51 percentage points and lowest in Germany with only 4.8 percentage points for individuals at their prime working age. But this gap marks important differences in levels of participation for both sexes in different countries, ranging from gaps with low levels of employment for men and women -like Bosnia and Herzegovina, at 33 and 54 percent respectively, to gaps with higher levels, such as in Estonia with participation levels of 60 percent for women, and 72 percent for men. 5 Source: World Bank, World Development Indicators database 6 An individual is assumed to be “employed” if he/she responded “yes” to the question “Did you work during the past 12 months?”. When employment is taken as employment in the past week, 55.5 percent of women and 67.9 percent of men reported being employed. Figure 2: The educational attainment levels are similar while employment ratios vary for men and women A. Employment – worked in the past 12 months B. Male population vs. female population with upper (% of population aged 25-54), by gender and by secondary level education or more (% of country population aged 25-54) Belarus Czech Rep. Tajikistan Ukraine Belarus Russia Poland 100 Estonia Russia Poland Bulgaria 100 Latvia Lithuania Kazakhstan Kyrgyz Rep. Uzbekistan Bulgaria Italy Latvia Germany Lithuania Georgia Estonia Slovak Greece Ukraine Slovenia Rep. Romania Montenegro Greece Hungary Azerbaijan Cyprus % of male population (aged 25-54) 90 Serbia Albania 90 Serbia Kazakhstan Romania Croatia ItalyArmenia % of male population (age 25-54) Bosnia and Cyprus Germany Moldova FYR Macedonia Armenia Herz. Kosovo Bosnia and 80 Uzbekistan Montenegro Mongolia Moldova 80 Turkey FYR MacedoniaAlbania Mongolia Kosovo Herz. Croatia Azerbaijan Slovenia Slovak Rep. Hungary 70 70 Tajikistan Turkey Georgia Czech Rep. 60 Kyrgyz Rep. 60 50 50 40 40 30 30 20 20 10 10 0 0 0 20 40 60 80 100 0 20 40 60 80 100 % of female population (aged 25-54) % of female population (aged 25-54) % employed 45 degree line % with upper secondary education or more 45 degree line Source data: LITS III dataset, sample of primary respondents, weighted The sectors that men and women in LITS III are employed in are substantially different. The top 3 sectors that women work in are services, 26.9 percent of women aged 18-64 who worked in the last year work in the services sector. Public administration and retail trade come second for women, which together account for an additional 27.6 percent of female employment (14.1 percent and 13.5 percent respectively). While a substantial proportion of men are also in the services sector with 19.1 percent of them working there, 28.2 percent of men work in construction or manufacturing sectors which is almost three times the share of women in those sectors. Across sectors, the majority of women and men work under contractual agreements, and while few of them work part-time, on average men work longer hours than women. Among women who work as wage employees or paid interns 75.3 percent work permanently with a written contract. This share is similar for men with 75.1 percent of men working under a permanent written contract. Including temporary and seasonal contracts, the share rises to 85.4 percent for women and 82.5 percent for men. Yet, there is some variation between countries. The share of women working under a contract (permanent, temporary or seasonal worker) drops down to 54.4 percent in Turkey (52 percent for men) while it is highest in Lithuania with 99.7 percent of women working as wage employees or paid interns working under contracts (97.9 percent for men). In households with an adult man and woman (aged 25-54) who are a couple, both being employed is the most common household type (See Figure 3). For such households, if only one member is employed it is more likely to be the man than the woman and this trend holds across all countries. Overall, only 8 percent of the population live in a household where only the woman works in the couple while 29.6 percent of the population live in a household where only the man works. In the majority of the sampled countries (24 out of 34) it is more likely for both the man and the woman to work. In Azerbaijan, Tajikistan, Kosovo, Turkey, Uzbekistan, Armenia, Macedonia and Bosnia and Herzegovina, among households where the primary and secondary respondents are a couple, it is more common for the man to be the sole employed person than it is for both the man and woman to be employed. 9 Figure 3 For households where the primary and secondary respondents are a couple in the majority of countries it is most likely that they are both employed 100 8 % of population in households where primary and 90 secondary respondents are a couple 80 29.6 70 60 50 Neither man nor woman works 40 Only man works 54.3 30 Both man and woman work 20 Only woman works 10 8 0 Croatia Azerbaijan Hungary Italy Kosovo Turkey Armenia Kyrgyz Rep. FYR Macedonia Greece Moldova Ukraine Slovak Rep. Poland Latvia Estonia Kazakhstan Slovenia Germany Serbia Russia Bosnia and Herz. Albania Georgia Mongolia Montenegro Romania Cyprus Belarus Tajikistan Lithuania Bulgaria Czech Rep. Total Uzbekistan Source data: LITS III dataset, sample of primary respondents living in households of all sizes and where primary and secondary respondents are a couple. Both respondents are aged 24-54, weighted. Women are more likely to not work because of household duties and they are more likely to take temporary leave compared to men (See Figure 4). Most women not working and not looking for work state that “household duties” are the main reason for this situation. Overall 59.6 percent of women of prime working age (aged 25-54 years old) report not looking for work due to looking after the family/house while the share of men stating this reason is 26.2 percent. Women are also more likely to not have worked in the previous week due to temporary leave which might be linked to care and household duties. 47.2 percent of prime working age women (aged 25-54) report not going to work in the previous week due to temporary leave while for men this share is 20.5 percent. The employment outcomes of women are significantly associated with their own characteristics, such as their education, but also with some characteristics of the men they live with. Some male characteristics do have a significant association with women’s employment outcomes. The man’s age is negatively associated with the woman’s employment status and the man’s employment status is positively associated with the woman’s employment status. Women’s level of education and their asset ownership are other factors that increase their likelihood of being employed (See Annex 5 for regression results). But men’s level of education or asset ownership are not significantly associated with women’s employment. 10 Figure 4 Women are more likely to not work because of household duties and they are more likely to take temporary leave A. Reasons for not looking for a job B. Reasons for not working in the last 7 days Percent of population not working and not looking for a job Percent of population who worked in the previous (Aged 25-54) year but who did not work in the last 7 days (Aged 25-54) Temporary absence (e.g.… Looking after the family/house Could not find a job Other reasons Other Retired Home or care duties Long term sick/disabled I got fired No need to work Other personal or family reasons No suitable jobs available Student Did not want to work Doesn't want to work My new job (/contract) will… Temporarily sick/injured Chronic illness Have already found a job waiting to start Full time student Waiting for an answer Retired Military service 0 20 40 60 80 100 0 20 40 60 80 100 % of population (aged 25-54) % of population (aged 25-54) Male Female Male Female Source data: LITS III dataset, sample of primary respondents aged 25-54 years old, weighted Young people not being in education or employment is a problem in most of the countries and women are more likely to be in this position than men. Overall, 27 percent of women aged 18-24 and 22.3 percent of men aged 18-24 are neither in employment nor in education. In 24 countries (out of 34) the proportion of young people not in education or employment is higher among women than men (See Figure 5). In Azerbaijan and Lithuania, the gap is more than 20 percentage points. Of all the young people in the sample not in education or employment, 54.6 percent are women. 11 Figure 5 Young people not being in education or employment is common in most of the countries Prevalence of youth not in education or employment, by gender (Percent of population aged 18-24) female male Russia Czech Rep. Poland Estonia Germany Slovak Rep. Italy Slovenia Ukraine Greece Latvia Lithuania Belarus Montenegro Croatia Albania Cyprus Bulgaria Romania Serbia Hungary Mongolia Kosovo Bosnia and Herz. Kazakhstan Moldova FYR Macedonia Turkey Georgia Armenia Tajikistan Kyrgyz Rep. Uzbekistan Azerbaijan 0 20 40 60 80 100 % of population aged 18-24 12 B. Entrepreneurial activity Entrepreneurship is not very common among the population in the LITS III sample, especially among women. Overall, in the 34 countries, 10.9 percent of the male population (aged 18 or more) and 5.8 percent of the female population report having “ever set up a business” -e.g. they have set up their current business or they have set up a business before but they are not involved with it any more. Women are less likely to attempt becoming an entrepreneur but, if they do, their success rates are similar to men. Out of every 1000 men only 146 attempt becoming an entrepreneur, while this number is only 80 for women. Out of the women attempting to set up a business 72.5 percent of them succeed compared to 74.7 percent for men. For both men and women not having enough capital is the primary reason for failing in setting up the business (See Figure 6). Figure 6 More men attempt to be entrepreneurs compared to women but after they attempt to set up a business, men and women have similar success rates. Men Women 1000 Men 1000 Women Not Not Attempted: Attempted: Attempted: 146 Attempted: 80 854 920 Not Succeeded: Not succeeded: 37 109 succeeded: 21 Succeeded: 58 Not enough Not enough capital: 15 capital: 10 Too much Too much bureaucracy: 7 bureaucracy: 2 Couldn't Couldn't afford the afford the bribes: 1 bribes: 0 Competitors Competitors threatened: 3 threatened: 1 Change in Change in personal personal situation: 5 situation: 5 Other: 6 Other: 2 Source data: LITS III dataset, sample of primary respondents aged 18 or more, weighted Women entrepreneurs are concentrated mostly in the services sector and in shops or stands (See Figure 7). As it is the case for employment, compared to male entrepreneurs, a higher share of women entrepreneurs are in the services sector. 29.5 percent of female and 16.8 percent of male entrepreneurs work in this sector. And akin employment, women entrepreneurs are less present in the manufacturing, wholesale trade and construction sectors. In line with this, women entrepreneurs are also less likely to be in industrial estates compared to men. Businesses run by women are mostly concentrated in shops or spaces in other locations, indicating smaller businesses and lower capital. 13 Women entrepreneurs are slightly more likely to be the sole owners of their businesses while their ownership of assets is lower compared to male business owners (See Figure 7 Panel C and D). Overall 67 percent of women entrepreneurs own their businesses solely as opposed to 60.7 percent of men. Ownership of equipment or machinery is the most common type of asset among both male and female entrepreneurs. 54.3 percent of women entrepreneurs own equipment or machinery as opposed to 63.9 percent of men. Compared to male entrepreneurs female entrepreneurs also have a lower ownership of land or building. Figure 7 Sector, location, ownership type and assets of business by gender of the owner Sector of business Location of business Services 16.8 22 29.5 Commercial district/ shop 24 Retail Trade 19.7 21.1 Stand/space in other 20.3 Public Administration 0.1 22.8 16.6 location Nonclassifiable Establishments 4.1 13.3 16.9 Other 19.2 Finance, Insurance, and Real… 3.1 6.4 Agriculture, Forestry, and… 8.2 Traditional market 1.7 4.9 13.1 Manufacturing 17.1 3.3 12 Home 10.8 Wholesale Trade 13.8 2.2 Construction 15 20.2 1.9 Industrial estate 7.9 Transportation and Public… 1.9 0.8 No fixed location 6.8 Mining 0.1 2.2 0 0 10 20 30 40 0 5 10 15 20 25 30 % of employers % of employers Male Female Male Female Ownership of the business Assets of the business 80 70 63.9 67 70 60.4 60 54.3 60 % of employers 50 50 % of employers 40 40 30 29.9 22.5 19.7 30 17.1 21.6 21.9 20 13.3 10 20 11.3 0 10 Sole ownership Joint ownership Joint ownership with household with non- 0 member household Land Building Equipment and member machinery Male Female Male Female Source data: LITS III dataset, sample of primary respondents aged 18 or more who are employers (self-employment is not included), weighted 14 Having a higher education degree is positively correlated with being an entrepreneur for both men and women. According to the regression results where the dependent variable is “ever been an entrepreneur” and controlling for country fixed effects, both women and men are more likely to be an entrepreneur if they have a higher education degree (See Annex Table 2 for regression results). In fact, more educated women are more likely to be entrepreneurs and are also more likely to be employed. Overall more than half of female entrepreneurs have a higher education degree. 61.7 percent of women entrepreneurs (aged 18-64 years old), and only 12.3 percent have less than an upper secondary education degree. This is similar to the education levels of women who reported that they had worked in the past year. 57.6 percent of working women (aged 18-64 years old) have a higher education degree while 12.2 percent have a degree lower than upper secondary education. Hence women entrepreneurs are neither more nor less educated than the overall female work force. At the country level, empowerment of women in household decisions regarding household finance is positively correlated with women’s entrepreneurship7. While this is the case at the country level, at the individual level, according to the regression results financial empowerment in the household is not found to be significantly associated with being an entrepreneur for women. This finding may suggest that rather than financial empowerment of the individual, being in a country where women are more financially empowered overall may make a greater difference to whether or not a woman chooses to become an entrepreneur. This finding points to social and cultural differences at the country level being more important than individual differences. Care: Needs and Provision Among the many factors affecting women’s economic participation, the gender division in terms of care roles and responsibilities is a core one. The burden of eldercare and childcare responsibilities can have serious negative repercussions on women’s economic outcomes. As women spend more time engaging in unpaid, informal care work, they have less time to work in the market. Studies looking at the relationship between caregiving and labor market outcomes show negative impacts on women’s economic participation8. To the traditional childcare needs, in the case of Europe and Central Asia, eldercare needs have been added, given greater longevity by many European societies, as the combined forces of lower fertility and higher life expectancy drive population aging in European societies (Bussolo et al. 2015). As with childcare, these needs may imply an additional task for women in the households. Preferences, policies, institutional factors, such as the availability, accessibility, affordability, and quality of care services, and the alternative opportunities for women, affect decisions around whether to provide care at home or not. Household composition determines care needs in the household and has an impact especially on women. In households with children or elderly members, care need arises naturally. While in larger households care responsibilities may be distributed across adults, the care burden may be higher on women in countries where larger households are less common. There is substantial variation in household composition and demography across the LITS countries, as they are in different moments of their demographic processes. The percentage of the population living in households with children is greatest in younger countries like Tajikistan where 82.6 percent of the population live in such households. By contrast, in post demographic transition countries like Germany only 12.1 percent of the population live in households with children. Tajikistan is the country with the youngest population of those sampled in LITS III, with a median age of 21.5. Germany has the oldest population with a median age of 7 See Annex 2 for the methodology regarding the construction of empowerment indices 8 See Levin et al. (2015) and World Bank (2012) for a summary. 15 45 years old.9. The percentage of the population living in households with young children (ages 0-6) and elderly is an important proxy when looking at care needs, and there is a lot of variation among countries when it comes to this (See Figure 8 Panel A). Overall, 13.7 percent of the population live in households with children aged 0- 6. The share of the population living in a household with young children is particularly high in Mongolia, Tajikistan and Uzbekistan where more than every 1 in 3 people live in such households. Living with elderly people (aged 65+) is more common in the LITS III countries than living with younger children (aged 0-6). Overall, 26 percent of the population lives with an elderly person aged 65 or more as opposed to 13.7 percent of the population living with children aged 0-6. People living in all-elderly households make up 12.2 percent of the total population in the sample. The prevalence of all-elderly households is as high as 20.6 percent in Germany while it drops down to 0.5 percent in Tajikistan10. Note that there is no general pattern in terms of the difference between rural and urban populations for the percentage of people living in all-elderly households. While in Russia, Belarus and Bulgaria, all-elderly households are more common in rural areas, in Turkey, Germany and FYR Macedonia this trend is reversed. While living with an elderly person may mean more care work for the other adults in the household, it may also result in less childcare work if the elderly person is providing care for grandchildren themselves. With an increasing urban population worldwide, it is important to see whether the prevalence of this kind of three-generation household structure is declining. Indeed, the analysis of LITS III data suggests that three-generation households are less common in countries with larger urban populations (See Figure 9). The decline of three-generation households increases the childcare burden on adults and could also increase the need for institutional childcare options. Figure 8 13.7% of the population live in a household with children aged 0-6 while living with an elderly is more common with 26% of the population living with an elderly aged 65 or more A. Population living with children aged 0-6, and mean number of children 100 1 0.8 0.8 % of population living in households with average number of children 80 0.8 0.6 60 0.5 0.4 0.5 0.6 0.4 0.4 children aged 0-6 0.4 40 0.4 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.3 0.2 0.2 0.2 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.2 0.1 0.2 0.2 0.1 20 0.1 0.1 0.2 0.0 0 0 Bulgaria Slovenia Mongolia Tajikistan Germany Italy Serbia Albania Russia Georgia Kosovo Total Turkey Montenegro Greece Romania Croatia Azerbaijan Poland Lithuania Latvia Estonia Kyrgyz Rep. Cyprus Slovak Rep. Belarus Moldova Armenia Uzbekistan Czech Rep. Hungary Ukraine FYR Macedonia Kazakhstan Bosnia and Herz. % of population living with children aged 0-6 average number of children aged 0-6 9 Data is obtained from UN Data for year 2012. 10Note that there is no general pattern in terms of the difference between rural and urban populations for the percentage of people living in all-elderly households. While in Russia, Belarus and Bulgaria, all-elderly households are more common in rural areas, in Turkey, Germany and Macedonia this trend is reversed. 16 B. Population living in households with at least one elderly individual aged 65+, and mean number of elderly 100 1 % of population living in households with average number of elderly 80 0.8 0.6 0.5 0.5 0.5 0.5 0.5 0.5 60 0.4 0.4 0.4 0.5 0.4 0.4 0.4 0.4 0.4 0.6 0.4 0.4 0.4 0.4 0.4 0.3 0.4 0.3 0.3 0.4 0.4 0.3 0.2 0.3 0.3 0.3 0.3 0.3 elderly aged 65+ 40 0.4 0.1 20 0.2 0 Tajikistan 0 Albania Serbia Mongolia Russia Germany Slovenia Italy Bulgaria Kosovo Georgia Turkey Azerbaijan Moldova Latvia Kyrgyz Rep. Montenegro Greece Belarus Total Uzbekistan Cyprus Poland Croatia Lithuania Estonia Romania Armenia Czech Rep. Kazakhstan Ukraine Slovak Rep. FYR Macedonia Hungary Bosnia and Herz. % of population living with elderly aged 65+ average number of elderly aged 65+ Source data: LITS III dataset, sample of primary respondents, weighted Figure 9 The fraction of the population living in three generation households -children (aged 0-17), adults (aged 18-64) and elderly (65+) - is negatively correlated with the percentage of the population living in urban areas 20 Georgia Uzbekistan % of population living in three generation Tajikistan Armenia 18 16 14 y = -0.241x + 20.291 12 households Kyrgyz Rep. Azerbaijan 10 Albania Kazakhstan Macedonia FYRSerbia 8 Romania Bosnia and Herz. Belarus Mongolia 6 Montenegro SlovakCroatia Moldova Rep. Slovenia Russia Bulgaria Latvia Estonia Lithuania Ukraine 4 Greece Poland Hungary Italy CyprusGermany Czech Rep. Turkey 2 0 0 20 40 60 80 100 %of population living in urban Source data: % of people living in 3 generational households is calculated using LITS III dataset, sample of primary respondents, weighted. % of population living in urban is obtained from World Bank World Development Indicators In the LITS III data set, the need for care in households is widespread with one-in-four people in the sample living in a household with a child, elderly person or disabled person in need of care.11 12.3 percent of the population live in a household with a child care need, 10.4 percent in a household with an elderly care need and 4.7 percent in a household with a disabled care need. There is substantial variation between countries both in terms of the extent of their care needs and the types of care that are most needed (See Figure 10), with 13 out of the 34 countries having more than 30 percent of the population live in a household with someone in need of care. In Uzbekistan where the child care need is highest, 48.9 percent of the population 11 This is one of the innovations of LITS III, as questions allowing for the analysis of care need were added. The questionnaire includes a question asking whether each household member needs care and by what source this care is provided to that individual (e.g. household member, public care facility, private care facility etc.). The question is asked for all children (0-6 years old), elderly people older than 75 years old, and household members with a disability. Since disability of the household members is not asked with a separate question, individuals needing care, who are not children (aged 0-6 years old) and not elderly (aged more than 75 years old) were categorized as in need of care for a disability in the analysis. 17 live in a household with a child (aged 0-6) in need of care. In Russia, where care due to disability is highest, 21.3 percent of the population live in a household with a disabled individual in need of care. Population living in a household with an elderly person who has a care need reaches its highest value in Armenia with 11.7 percent of the population living in such households. Figure 10 While most of the population (75.3%) live in a household with no care need, variation between countries is high. Care need in households (% of population living in such households), by country 100 90 80 All types of care % of population 70 60 Elderly care and disabled care 50 40 Child care and disabled care 30 20 Child care and elderly care 10 Only disabled care 0 Only elderly care Only child care Source data: LITS III dataset, sample of primary respondents, weighted Household members are the number one source providing care in the LITS III sample. The high levels of care need observed across countries are most often met by members of the household itself rather than institutions or other sources of care (such as nannies, relatives or friends), and while institutional care is used more for children, for the elderly or the disabled needing care, institutional care use is considerably less common (See Figure 11). Overall, 63.7 percent of the population who live in a household with a child gets care for that child only from a household member. This share is considerably higher for elderly care with 81.4 percent of the population who live in a household with such care need uses a household member to satisfy it. There are only a handful of countries in which household members are not the primary care providers. In Slovenia, Cyprus, Belarus, Latvia and Russia more than 50 percent of the population living in a household with a child receives care from a public or private institution or combination of providers as well as from household members. Germany is the only country in which more than 50 percent of the population living in a household with an elderly requiring care that receives that care from someone other than a family member. Institutional care use for child care varies considerably between countries (See Figure 12). Despite being more commonly used overall when compared to elderly or disabled institutional care, child care is still utilized less in some countries which may be due to either supply or demand side constraints. LITS data shows that institutional child care is close to non-existent in Azerbaijan and Turkey, for the later, we know that access and capacity are limited 12. In contrast in Russia and Belarus, more than 50 percent of the population who live in a household with a child care need receive center-based care. In general, institutional child care use is more common for children aged 4-6 compared to children aged 0-3, as it tends to overlap with pre-primary education, which is compulsory in some countries (share of use of formal care go from 18.3 percent for households with children aged 0-3 to 33.5 percent for those living with children aged 4-6). 12 For Turkey see World Bank (2015). 18 Figure 11 In the majority of the countries household members are the primary source of care providers for child care, elderly care and disabled care A. Child care providers (% of B. Elderly care providers (% of C. Disabled care providers (% of population living in households with population living in households with population living in households with child care need), by country elderly care need), by country disabled care need), by country Czech Rep. 98.3 Azerbaijan 100 100 95.7 100 FYR… 100 Azerbaijan 95.2 Kosovo 98.4 100 89.2 98.3 Uzbekistan 99.6 Moldova 88.3 Romania 98.1 97.6 87.7 97.1 Turkey 97.3 Tajikistan 87.3 FYR Macedonia 96.8 97.3 86.2 93.7 Greece 96.2 FYR Macedonia 83.6 Serbia 93.2 96.1 83.1 92.9 Estonia 96.1 Uzbekistan 81.3 Bosnia and… 92.6 95.6 78.6 91.1 Moldova 95.5 Bosnia and Herz. 76.9 Czech Rep. 87.3 95.2 75.7 86.5 Slovenia 94.7 Armenia 72.5 Slovenia 86.4 93.7 69.7 85.2 Lithuania 92.7 Montenegro 68.2 Croatia 84.9 92.4 67.1 83.4 Latvia 92.3 Lithuania 67.1 Bulgaria 83 92.1 66.2 82.9 Kosovo 92 Kyrgyz Rep. 64.9 Moldova 82.8 82.8 90.5 63.2 Georgia 90.2 Germany 62.1 Kyrgyz Rep. 82.7 82.3 89.9 61.3 Albania 85.4 Italy 60 Hungary 80.4 79.8 85.4 57.1 Croatia Greece 78.7 82.5 Slovak Rep. 54.7 76.8 78.8 52.9 Kazakhstan 74.5 Armenia 76.9 Mongolia 52.5 74.1 73.2 47 Cyprus Belarus 69.5 Germany 65.6 44.7 44.2 63.4 65.1 Latvia 39.2 Slovak Rep. 56 Cyprus 55.2 33.4 14.4 13.8 Total 63.7 Total 81.4 Total 93.1 0 50 100 0 50 100 0 50 100 % of population living in households % of population living in households % of population living in households with elderly care need with disabled care need with child care need Source data: LITS III dataset, sample of primary respondents, weighted Country income is positively correlated with the utilization of institutional care (See Figure 13). Yet some countries do better (or worse) than predicted by their GDP per capita when it comes to institutional care use. Belarus and Turkey, for example, have very similar GDP per capita levels (with 16.6 thousand and 19.0 thousand USD respectively). Yet in Belarus 51.7 percent of the population living in a household with a child care need report using institutional care as opposed to 4.6 percent for the same group in Turkey. 19 Figure 12 Utilization of institutional child care is low in many countries Utilization of institutional care in households (% of population living in households with children in ages 0-3 or 4-6), by country For children aged 0-3 years old For children aged 4-6 years old Source data: LITS III dataset, sample of primary respondents, weighted Figure 13 Access to institutional child care is positively correlated with the level of welfare in the country. Institutional care use for child care vs GDP per capita For children aged 0-3 years old For children aged 4-6 years old Russian 45 80 Federation BelarusLatvia institutional care when there is a child institutional care when there is a child Russian % of population in households using % of population in households using 40 Latvia 70 Poland Federation needing care aged 0-3 years old needing care aged 4-6 years old Slovak 35 Republic Cyprus Slovak Cyprus 60 30 Kyrgyz Belarus Republic Slovenia Kazakhstan 50 Mongolia Kazakhstan Mongolia Albania Serbia 25 Republic Serbia Germany Ukraine Slovenia Germany Montenegro Lithuania Georgia 40 Georgia Croatia Poland Kyrgyz Albania Lithuania 20 Romania Italy Montenegro Italy Ukraine UzbekistanFYR BosniaFYR Uzbekistan and 30 Republic Bulgaria Romania 15 Bulgaria Bosnia Armenia and Macedonia Greece Moldova Macedonia Herzegovina Estonia CroatiaEstonia Tajikistan Hungary Greece 20 Herzegovina 10 Armenia Czech Moldova Tajikistan Hungary Czech Kosovo Turkey Azerbaijan Azerbaijan Turkey 10 Kosovo 5 Republic Republic 0 0 0 10000 20000 30000 40000 50000 0 10000 20000 30000 40000 50000 GDP per capita, PPP (constant 2011 international $) 2015 GDP per capita, PPP (constant 2011 international $) 2015 Child care needs in the household affects women adversely in terms of employment outcomes (at the country level, and the individual level). While we do not know the gender of the household member who is providing care for the children in the LITS III dataset, employment outcomes of men and women suggest that the care burden is only negatively associated with women’s employment -this according to the regression results controlling for age, level of education and country fixed effects for men and women aged between 25 and 40 years old (or aged between 25 and 54 years old). In contrast, other individual characteristics like age and higher education are positively correlated with being employed for both women and men. (See Annex Table 5). 20 A negative relationship between care need (child, elderly or disabled care need) and female employment can also be seen at the country level. In countries where care needs are higher, female employment (of women aged 25-40) is generally lower (See Figure 14). Yet a higher utilization of institutional care seems to favor women’s employment outcomes. At the country level, a positive correlation exists between institutional care use for child care and female employment, where we can assume the causal relationship goes both ways (See Figure 14). At the individual level, institutional child care use is again positively associated with women’s employment. Women living in a household with young children (aged 0-6 years old) are 27.3 percentage points more likely to be employed if the household utilizes institutional care for at least one child13. Presence of other adults in the household is not found to be significantly associated with women’s employment. (See Annex 5 Table 6 for regression results) Figure 14 Female employment (of women aged 25-40) is negatively correlated with care need A. % of female population employed (for women aged 25- B. Institutional care use for child care vs female 40) vs % of population living in a household with care need employment 100 100 % of employed among the female population % of employed among the female population 90 Russia 90 Poland Russia Germany Poland Latvia Germany Latvia Estonia Estonia Belarus Belarus Bulgaria Czech Cyprus Rep. Lithuania Czech Rep. Bulgaria Lithuania Cyprus 80 Italy Croatia Romania 80 Croatia Italy Romania Serbia Ukraine Serbia Ukraine Albania Slovak Rep. Hungary Albania Slovak Rep. Hungary Montenegro Slovenia Montenegro Slovenia 70 Greece 70 Greece Kazakhstan Kazakhstan y = 0.6721x + 42.95 60 Moldova Bosnia and Mongolia 60 Moldova Bosnia and FYR Mongolia aged 25-40 aged 25-40 Herz. Herz. FYR Macedonia Macedonia 50 Kyrgyz Rep. 50 Kyrgyz Rep. Turkey y = -0.8936x + 84.435 Turkey Georgia Georgia 40 Armenia 40 Armenia Uzbekistan Uzbekistan Tajikistan Tajikistan Kosovo Kosovo 30 30 20 Azerbaijan 20 Azerbaijan 10 10 0 0 0 20 40 60 80 100 0 20 40 60 80 100 % of population living in a household with care % of population in households using need instiutional care when there is a child needing care Source data: For institutional care use LITS III dataset is used, sample of primary respondents, weighted. GDP per capita values are obtained from World Bank WDI dataset. According to the regression results controlling for level of education and age of the women, presence of another adult 13 woman in the household and presence of an adult man as well as the country effects, for women aged 25-54. 21 Spotlight: Female Headed Households In the 34 countries included in the study, close to a third of the population live in a household with a female head14. 68.6 percent of the population lives in a household with a male head as opposed to 31.4 percent living with a female household head. Yet in some countries like Latvia and Russia, female headed households are almost as common as male-headed households. In Latvia, 57 percent of the population live in a female headed household. But female headed households come in all forms and in this region, they are older. Female heads are more likely to be older (65+) and to live alone when compared to male household heads. 25.6 percent of the population living in female headed households live in a household where the head is older than 64 as opposed to 19 percent in male headed households. 29.6 percent of the population living in female headed households are women living alone while this is the case for only 9.4 percent of the population living in male headed households. Living alone with children (in single adult households) or living alone in old age is more frequent among women. The proportion of women living alone with children or elderly women living alone was found to be 10.5 percent of the overall female population in the sample countries. 3.3 percent of women live alone with children (aged 0-17) and 7.2 percent of women are aged 65 or more and living alone. Russia, Poland, Latvia, Lithuania, Czech Republic and Estonia are countries where more than 15 percent of women live in these types of households. By contrast, the same categories generally represent a small part of the male population. Only 0.4 percent of men live alone with children and elderly men living alone make up 2.2 percent of the male population. The households where adults live alone with children (single adult households) or elderly living alone are overwhelmingly composed of women. 90.6 percent of adults living alone with children and 78.6 percent of elderly people living alone are female. Looking at the composition of individuals living alone with and without children by age shows that women living alone with children are significantly higher in number. Women living alone without children are more likely to be older while men living alone without children are more likely to be younger. Asset Ownership Women’s asset ownership has consequences as it affects the power balance in the household. Ownership of assets gives women greater economic freedom and hence improves their outside options, enabling them to have a more affordable “exit” strategy (World Bank, 2012). Women are more likely to have a stronger presence in household decision-making processes when they have greater control over assets (Klugman et al., 2014). Evidence from other countries outside of ECA indicates that asset ownership is positive for women. In Nepal, owning land is associated with women being more likely to have the final word in household decisions regarding their own health care and making large household purchases (Allendorf, 2007). In Ecuador, in households where both men and women own real estate the couple is more likely to make joint decisions and to agree on the decision to work and spend their own income (Deere and Twyman, 2012). 14The household head is the person that is reported as “head” by the respondent to the questionnaire who can be the head themselves or by another knowledgable member of the household. 22 Women’s asset ownership is also found to be correlated with more child-friendly spending in the household. Similarly, assets that women bring in to the marriage are found to be associated with an increase in the budget share of education in Bangladesh, India and South Africa (Quisumbing and Maluccio, 2000). A. Land and Dwelling Ownership A wide discrepancy exists between countries in terms of asset ownership for women (See Figure 15). The share of women owning a dwelling or land (solely or jointly) ranges from 15.7 percent in Azerbaijan to 76.7 percent in Ukraine. In Azerbaijan, Turkey, Kosovo and Uzbekistan less than a quarter of women own a dwelling or land. In contrast, in Romania, Russia, Hungary and Ukraine more than two thirds of women own a dwelling or land. Figure 15 Women’s ownership of a dwelling or land shows wide variation between countrie s Ownership of dwelling or land (% of female population), by country Source data: LITS III dataset, sample of female primary respondents, weighted The LITS III dataset allows us to take a detailed look at women’s asset ownership, what drives it and the possible outcomes associated with it. For the 34 countries included in the study, it is possible to examine the asset ownership status of both men and women and link them with a number of other factors at the country or individual level. In the 34 countries examined, women are at a disadvantage overall in terms of holding an asset (See 23 Figure 16 In most countries, men are more likely to own the dwelling or land than women. Additionally, owning a dwelling solely or jointly is more common among men than women. The same trend also holds for owning land Panel A). Dwelling or land ownership (solely or jointly) is more common among men in most countries (27 out of 34). Looking at dwelling and land ownership separately, dwelling ownership solely or jointly is more common among men (See 24 Figure 16 In most countries, men are more likely to own the dwelling or land than women. Additionally, owning a dwelling solely or jointly is more common among men than women. The same trend also holds for owning land Panel B). Overall 51.6 percent of women own a dwelling as opposed to 61.4 percent of men. Land ownership incidence is very low in the sample – mainly driven by the fact that the population is now more urban (72.2 percent of the population lives in urban areas in LITS III countries). Overall 7.3 percent of women own land solely or jointly as opposed to 8.7 percent of men in the sample. 25 Figure 16 In most countries, men are more likely to own the dwelling or land than women. Additionally, owning a dwelling solely or jointly is more common among men than women. The same trend also holds for owning land A. Dwelling or land ownership (% of population), by gender B. Dwelling and land ownership by and by country ownership type (% of population), by gender, overall 100 100 Ukraine 90 Georgia Russia 80 Azerbaijan Bosnia and Herz. Moldova Bulgaria Romania 80 38.5 Belarus 48.4 % of male population Tajikistan Mongolia Croatia Slovak Hungary Rep. % of population Serbia FYR Macedonia Albania MontenegroKyrgyz Rep. Slovenia GreeceCzechLithuania Rep. 70 Kosovo Uzbekistan Italy Kazakhstan Estonia Armenia Germany 60 60 Turkey Poland 92.7 91.3 Latvia Cyprus 50 25.1 40 23.7 40 30 20 36.3 20 10 27.9 3.2 2.7 0 4.1 6 0 Female Male Female Male 0 20 40 60 80 100 Dwelling Land % of female population does not own any Owning dwelling or land 45 degree line owns all jointly owns at least one solely Source data: LITS III dataset, sample of primary respondents, weighted Land ownership is higher among men and women living in rural areas than men and women living in urban areas. However, a higher rural population in the country does not necessarily lead to a higher rate of land ownership. Uzbekistan is the country with the highest rural population share (63.7 percent15) while only 2.7 percent of women and 3.7 percent of men own land in the country, which are among the lowest levels in the sample. By contrast, Bosnia and Herzegovina, which is the country with the second largest rural share of population in the sample, has higher land ownership rates with 14.6 percent of women and 27.4 percent of men owning land. The country with the highest population share owning land is Georgia, where 38.7 percent of men and 27.3 percent of women own land.16 15According to the World Bank World Development Indicators for year 2015. 1646.4 percent of population live in rural areas in Georgia according to World Bank World Development Indicators for year 2015. 26 Figure 17 In most countries, it is more common for men to own a dwelling (jointly or solely) than women. The same trend holds for land ownership, but land ownership is less common compared than dwelling ownership Ownership of at least one dwelling solely, by gender and by Ownership of all dwellings jointly, by gender and by country country female male female male Tajikistan Azerbaijan Turkey Latvia Albania Montenegro Kosovo Uzbekistan FYR Macedonia Kosovo Azerbaijan Kyrgyz Rep. Germany Bosnia and Herz. Mongolia Serbia Uzbekistan Georgia Armenia FYR Macedonia Italy Moldova Bosnia and Herz. Kazakhstan Montenegro Greece Kyrgyz Rep. Armenia Serbia Estonia Slovak Rep. Turkey Lithuania Bulgaria Croatia Romania Cyprus Croatia Belarus Slovenia Albania Hungary Russia Kazakhstan Tajikistan Czech Rep. Mongolia Georgia Slovenia Greece Poland Poland Italy Moldova Czech Rep. Cyprus Germany Belarus Ukraine Ukraine Bulgaria Lithuania Slovak Rep. Latvia Romania Estonia Hungary Russia 0 20 40 60 80 100 0 20 40 60 80 100 Dwelling ownership (%) Dwelling ownership (%) Source data: LITS III dataset, sample of primary respondents, weighted Differences between men and women’s asset ownership is largely due to differences in the rate of sole ownership of dwellings or land. Men are more likely to solely own an asset compared to women. In Azerbaijan where the gap between men and women is largest, only 12.6 percent of women own at least one dwelling solely as opposed to 70.4 percent of men. In 13 of the countries (out of 34) the gap is larger than 20 percentage points. In contrast, ownership of all dwellings jointly is more egalitarian between genders across countries. The largest gap between men and women’s ownership of all dwellings jointly is only 6.3 percentage points, observed in Ukraine. Even when women report that they own an asset, they may not have the right to make decisions on the asset on their own. Indeed, in some countries women are at a particular disadvantage in terms of having the right to sell their assets on their own. In 9 out of 34 countries, more than half the women who own land do not have the right to sell it by themselves. Significant disparities between men and women’s rights to sell land are observable in several countries. In Albania, Kosovo, Azerbaijan and Turkey the proportion of men owning land with the right to sell is more than 20 percentage points higher than the proportion of women. Women’s asset ownership is positively associated with age, higher educational attainment and being employed. Older women are more likely to own assets (See Figure 18). Only 17.6 percent of women aged 18- 24 own a dwelling or land as opposed to 73.2 percent of women aged 65 or more. The same relationship holds for men as well. 24.7 percent of men aged 18-24 own a dwelling or land as opposed to 86.3 percent of men aged 65 or more (See Annex 4 Table 1 and 2 for the cross tabulations). 27 Women with higher education degrees are more likely to own assets.17 60.1 percent of women with higher education own a dwelling or land, which is 11 percentage points more than women with upper secondary education or less. For men, education levels don’t make a difference when it comes to asset ownership. For both, men and women, higher education is found to be positively associated with dwelling ownership in general and ownership of all dwellings jointly18. Sole ownership of a dwelling is instead significantly associated with being a widow, divorcee and with increasing numbers of children in the household. The relationship between education and asset ownership is similar for men as well. Owning a dwelling is positively associated with higher education for men but owning at least one dwelling solely is not significantly related to education as was the case for women19. For sole dwelling ownership among men, the only significant determinant seems to be age and number of children in the household. Unlike women, men’s sole ownership of a dwelling is not affected by marital status. Figure 18 The older the women and men get the more Figure 19 On average women who do not own any assets are likely are they to own an asset less likely to work compared to women who own assets Dwelling or land ownership by age groups and by Percent employed in the last year by asset ownership and gender (%) gender 100 100 Dwelling or land ownership (% of population) 90 employed in the last year (%) 80 80 70 60 60 50 40 40 30 20 20 10 0 0 18-24 25-34 35-44 45-54 55-64 65+ 18-24 25-34 35-44 45-54 55-64 65+ Age groups Female - does not own assets Age groups Female - owns assets Male - does not own assets Female Male Male - owns assets Source data: LITS III dataset, sample of primary respondents, weighted Being employed is positively correlated with men’s and women’s asset ownership. 56.0 percent of women of working age and employed own a dwelling or land. For women who are not employed, this share is lower at 38.7 percent. A similar positive relationship also exists for men (62.1 percent of working men versus 49.4 percent of non-working men). Considering the relationship the other way around, ownership of a dwelling or land increases the likelihood of being employed for both men and women of all ages except for individuals who are older than 65 years old (See Figure 19). Ownership of assets decreases the likelihood of working for men when they are older than 65 (20.6 percent vs 7.5 percent) and it decreases the likelihood of working for 17 Overall in the sample 43.2 percent of women have a higher education degree and 33.5 percent of women have an upper secondary education degree. 18 According to the regression results, controlling for age, marital status, household composition and country fixed effects. In comparison, there is no significant association between sole ownership and educational attainment. 19 Controlling for other individual factors, household composition and country fixed effects. 28 women slightly as well (9.2 percent vs 8.7 percent). Overall, women who do not own assets are less likely to work in all age groups other than the over 65s. Spotlight: Asset ownership of elderly women living alone and single mothers Elderly women living alone interviewed in LITS III are, in general, not asset poor. Vulnerable women are more likely to be widowers or divorcees and this brings a higher likelihood of asset ownership due to inheritance from their husbands who have passed away or shared ownership of assets after a divorce. In almost all 34 countries, elderly women living alone are more likely to own a dwelling or land compared to the overall female population (See Figure 20). It might also be the case that the elderly women are living alone because they can afford it, explaining the observed correlation between asset ownership and elderly women living alone. In contrast, single mothers (adult women living alone with children) are more likely to be asset poor. Adult women living alone with children are more likely to own a dwelling or land as opposed to the overall female population in only 8 of the 34 countries. Germany stands out as the country where asset ownership is the lowest for these women with only 16.1 percent of adult women living alone with children owning a dwelling or land in the country. Figure 20 Owning a dwelling or land by living arrangements of women (% of female population), by country 100 % owning a dwelling or land 80 60 40 20 0 Russia Uzbekistan Czech Rep. Hungary Azerbaijan Germany Slovak Rep. Kosovo Montenegro Albania Armenia Bosnia and Herz. Kyrgyz Rep. Kazakhstan Italy Latvia Poland Bulgaria FYR Macedonia Serbia Tajikistan Cyprus Croatia Slovenia Moldova Total Romania Mongolia Greece Belarus Lithuania Ukraine Georgia Estonia Single female (aged 18-64) with children Elderly female (aged 65+) living alone All women (aged 18+) Source data: LITS III dataset, sample of primary respondents, weighted. Note that for Turkey there is no observation for “Single female with children” and there is only one observation for “Elderly female living alone” While elderly women living alone may not be asset poor, they may be “cash” poor as evidenced by the fact that a large proportion of them reporting not being able to afford heating in the household in some countries (See Figure 21). In Moldova, Azerbaijan and Albania the difference between the population who can afford heating and the share of the elderly women living alone who are able to afford heating is more than 20 percentage points. In all these three countries, elderly women living alone are found to own an asset. However, this does not ensure that they are able to afford heating. On the other hand, in Cyprus, Greece and Estonia the share of women living alone with children who can afford heating is more than 20 percentage points lower than the share of the general population who are able to afford heating. 29 Figure 21 Being able to afford adequate heating of the dwelling (% of population) , by country 100 90 80 70 % of population 60 50 40 30 20 10 0 Russia Hungary Czech Rep. Armenia Azerbaijan Uzbekistan Bulgaria Poland Bosnia and Herz. Slovak Rep. Cyprus Kyrgyz Rep. Kosovo Albania Kazakhstan Latvia Italy Montenegro Germany Tajikistan FYR Macedonia Serbia Total Moldova Croatia Greece Ukraine Mongolia Georgia Lithuania Romania Belarus Slovenia Estonia Single female (aged 18-64) with children Elderly female (aged 65+) living alone Overall (of the country) Source data: LITS III dataset, sample of primary respondents, weighted. Note that for Turkey there is no observation for «Single female with children» and there is only one observation for «Elderly female living alone». No significant differences exist across LITS III countries regarding the laws on property that can account for the difference observed in ownership between men and women. This suggests that differences in women’s asset ownership between countries do not stem from differences in legislation that protect women’s asset ownership20. Informal norms, not only those in the legal books can also affect the gap between men and women’s asset ownership. Yet a country level analysis of the relationship between women’s asset ownership and agreement levels with some questions around norms gives mixed results (See Figure 22). In countries where a higher share of women agrees with women and men’s equality as business executives or political leaders, asset ownership of women seems to be higher. However, women’s agreement with norms about women’s role in the household does not seem to be correlated with women’s asset ownership. Women’s asset ownership is in fact correlated more with women’s education and employment at the country level. It seems that in countries where a higher share of women have an upper secondary education degree or more and in countries where it is more likely for women to be employed, a higher share of women own assets. It should be noted that this relationship goes both ways. Asset ownership, education and employment together make a woman more empowered and are naturally correlated with each other. 20 See Annex 3 for a summary of property ownership laws in the countries covered in LITS III 30 Figure 22 Norms are effective in women’s asset ownership up to a degree A. “Women are as competent as men to be B. “Men make better political leaders than women business executives” do” 100 100 % of female population owning a % of female population owning a 90 Ukraine 90 Ukraine 80 Hungary Russia Romania 80 Russia Hungary Romania Lithuania Slovak Estonia BulgariaRep. Slovak Rep. Lithuania Estonia Bulgaria 70 Poland Czech Belarus Rep. Cyprus 70 Belarus Czech Rep. Cyprus MoldovaPolandSlovenia dwelling or land dwelling or land Moldova Slovenia Georgia Latvia Greece Georgia LatviaGreece 60 Tajikistan Croatia Italy 60 Tajikistan Croatia Italy Germany Kazakhstan Kazakhstan Germany Rep.Mongolia Kyrgyz Bosnia and Mongolia Kyrgyz Rep. Bosnia and 50 50 Serbia Herz. FYR Armenia Serbia Armenia Albania FYR Herz. Albania 40 Montenegro Macedonia 40 Montenegro Macedonia Turkey Uzbekistan Uzbekistan Turkey 30 Kosovo 30 Kosovo Azerbaijan Azerbaijan 20 20 10 10 0 0 0 20 40 60 80 100 0 20 40 60 80 100 % of female population agreeing % of female population disagreeing C. “A woman should do most of the household D. “It is better for everyone involved if the man earns chores even if the husband is not working” the money and the woman takes care of the home and children” 100 100 90 90 % of female population owning a % of female population owning a Ukraine Ukraine 80 Romania Hungary Russia 80 Romania HungaryRussia Lithuania Slovak EstoniaRep. Bulgaria Lithuania Estonia Slovak Rep. Bulgaria Czech 70 Poland Rep. Cyprus Belarus 70 Poland Czech Rep. Cyprus Belarus Moldova Moldova dwelling or land dwelling or land Slovenia Georgia Slovenia Georgia Tajikistan 60 CroatiaGreece Latvia Tajikistan 60 Germany Croatia Greece Latvia Germany Italy Italy Mongolia Kazakhstan Mongolia Kazakhstan 50 Bosnia and Kyrgyz Rep. 50 Bosnia and Kyrgyz Rep. Montenegr Herz. FYRSerbia Armenia Herz. Serbia FYR Armenia 40 Albania 40 Albania o Macedonia Montenegro Macedonia Turkey Uzbekistan Turkey Uzbekistan 30 Kosovo 30 Kosovo Azerbaijan Azerbaijan 20 20 10 10 0 0 0 20 40 60 80 100 0 20 40 60 80 100 % of female population agreeing % of female population agreeing Source data: LITS III dataset, sample of female primary respondents, weighted B. Bank Account Ownership Bank account ownership is the simplest indicator for financial inclusion in the population. Access to financial services is instrumental in expanding opportunities in life and having better life outcomes. Starting a business, investing in education and managing risks all become easier when individuals use financial services. The financial inclusion of women is also a key tool in empowering women. Bank account ownership is measured in LITS III. According to the results, in the 34 countries examined, 65.4 percent of women and 71.1 percent of men own a bank account (solely or jointly). Unlike asset ownership, bank account ownership is more egalitarian across genders - except for a few countries (See Figure 23). Kosovo has the largest gap between the share of men and women owning a bank account. 66.9 percent of men in the country own a joint or sole bank account as opposed to 40.8 percent of women. This country also has one of the widest gaps in labor force participation by sex. While in some countries a gap between men and women exist, inter-country differences account for a larger share of the variation in 31 bank account ownership. In some countries, financial penetration is particularly low. In Azerbaijan, Armenia, Moldova, Tajikistan, the Kyrgyz Republic and Uzbekistan more than two thirds of both men and women do not own a bank account. As with assets, bank account ownership is positively associated with higher education for women. Women with higher education degree are 27.1 percentage points more likely to own a bank account on their own compared to women who have an education lower than upper secondary education. A similar positive relationship also holds for men21. Figure 23 Unlike asset ownership, bank account ownership is more egalitarian across genders - except for a few countries. Wide variation in the levels of access to financial services between countries persists. Female and male population’s bank account (sole) Female and male population’s bank account (joint or sole) ownership, by country ownership, by country 100 Croatia Lithuania Latvia Slovenia Estonia 100 Croatia Lithuania Greece Latvia Germany Slovenia Estonia Cyprus Czech Italy Slovak Rep. Rep. Bosnia and Mongolia Mongolia 90 FYR Serbia Bosnia andSerbia Herz. Herz. 90 FYR Macedonia Macedonia Germany Czech Cyprus Rep. Montenegro Slovak Rep. Poland Belarus 80 Bulgaria Montenegro 80 Bulgaria Russia Belarus Kosovo Ukraine Kosovo % of male population % of male population Russia Turkey 70 Turkey Ukraine 70 Hungary Italy Poland Albania Greece 60 Hungary Albania 60 Georgia Kazakhstan Kazakhstan 50 Georgia 50 Romania Romania 40 40 Uzbekistan Armenia Uzbekistan Armenia 30 30 Kyrgyz Rep. 20 Tajikistan Moldova 20 Kyrgyz Rep. Tajikistan Moldova Azerbaijan Azerbaijan 10 10 0 0 0 20 40 60 80 100 0 20 40 60 80 100 % of female population % of female population owning a bank account (sole) 45 degree line % owning a bank account (joint or sole) 45 degree line Source data: LITS III dataset, sample of primary respondents, weighted 21 See Annex 5 Table 3 for regression results 32 Norms and Voice in Household Decisions Social norms can create gender roles and stereotypes that adversely affect women’s outcomes by constraining their bargaining power within the household. Using data from World Value Surveys (1990, 1995, 1999), Fortin (2005) shows that anti-egalitarian views such as agreement with the statement “When jobs are scarce, men should have more right to a job then women” are negatively associated with female employment rates and positively associated with a gender pay gap. More recently, Kenny and Patel (2017), also using the same data, find a strong correlation between norms, laws and female labor force participation, and between norms and the share of female legislators in a country. Along the same lines, a study looking at the World Values Surveys and European Values Surveys show that gender egalitarian attitudes towards female employment are associated with fertility in a U-shaped curve where an initial increase from a traditional perspective to gender egalitarian views is negatively associated with fertility while beyond a certain threshold more egalitarian views are positively associated with fertility (Arpino, Esping-Andersen, & Pessin, 2015). Since LITS III asks similar questions, this section investigates how transition countries see gender norms. In the 34 countries included in LITS III, most of the population (men and women) think that equal rights for women as citizens are important for their country. However, they do not necessarily think that equal rights for women exist in their countries (See Figure 24). Overall, while about half (55.3 percent) of the population stated that they “agree” or “strongly agree” with the statement “Equal rights for women as citizens exist in (my) country”, the majority of respondents (86.4 percent) “agree” or “strongly agree” with the statement that “Equal rights for women as citizens is important for (my) country”. In most of the countries, men and women think similarly on these issues highlighting a positive fact about awareness of women’s rights, albeit not their full realization. The countries where less than half of the population think that equal rights for women as citizens exist in their country correspond with countries were labor force participation of women is lower than the regional average, as well as other challenges are present. These include countries such as Kosovo, Moldova, Montenegro, Bosnia and Herzegovina, Turkey, Armenia and Ukraine, where s less than 40 percent of the population agrees with equality being a reality. There is an expectation that views regarding gender equality differ depending on the age group of the respondents, and while this is the case, there is no clear pattern that can be observed across the countries in the survey. Looking at the countries with the lowest agreement levels, in Turkey for instance young people (aged 18-24) agree less with the existence of equal rights for women as citizens (24.7 percent) while people older than 65 agree more (38.5 percent), although agreement remains low. In contrast, in Armenia young people agree more with the existence of equality (44.4 percent) while the elderly agree less (35.8 percent). 33 Figure 24 While in all the countries people agree that equal rights for women as citizens is important, they believe less that it really exists in their country “Equal rights for women as citizens are important for the “Equal rights for women as citizens exist in the country” country” Source data: LITS III dataset, sample of primary respondents, weighted While, in general, respondents value gender equality, the belief that men and women’s roles within the household should be equally distributed is not common in many countries. Overall, a bit less than half of the respondents (47.4 percent) agree or strongly agree with the statement “A woman should do most of the household chores even if the husband is not working”, with no clear differences between the views of men and women. Similarly, about half of men and women (50.6 percent of the total population) agree or strongly agree with the statement “It is better for everyone involved if the man earns the money and the woman takes care of the home and children” pointing to the fact that gender norms regarding household roles are still persistent. Looking at country averages, men seem to agree slightly more with the norms on the distribution of roles in the household (Figure 27). In Germany only 4.4 percent of women and 10 percent of men agree with men being the breadwinner and women being the caretaker as opposed to Azerbaijan where this share reaches 90.4 for women and 92.7 percent for men. The widest gap in terms of agreement in shared responsibilities is in Bulgaria, where 30 percent of women and 47 percent of men agree with the idea that housework should be a woman’s responsibility. There is a mixed picture when it comes to women’s roles in public life. Women and men are perceived to be equally competent as business executives by most of the population in most countries, but there is more divergence when it comes to political leadership. Overall 83.8 percent of women and 73.9 percent of men agree with the statement “Women are as competent as men to be business executives”. While women tend to agree more with the statement, the differences between men and women are not wide in the sampled countries. In Belarus, which has the largest difference, the proportion of women that agree is 23 percentage points higher than men. When faced with the statement ““Men make better political leaders than women do”, 53.6 percent of women and 42 percent of men disagree with it. And while women tend to disagree more with the statement as with the case of business executives, the gap is particularly wide (wider than 20 percentage point) in some countries such as Russia, Montenegro, Bulgaria and Poland (Figure 28). In other areas, such as education for boys and girls, there is substantial agreement between women and men. Achieving higher education is seen as important in all the countries both for boys and girls (See Figure 29). Most respondents, regardless of their gender, agree that it is important if their son/daughter achieves 34 a university education, with three quarters of the population thinking that is important that their daughters and sons get a university education. While in the majority of the countries a university education is perceived as important, Poland and the Czech Republic are the only countries where less than half of the population agree with the importance of a university education for their children, with no differences for girls and boys. Figure 25 Men tend to agree more with the norms on the distribution of roles in the household but differences between men and women are small. In many countries, a large share of women also agree with the gender norms regarding the distribution of roles in the household “A woman should do most of the household chores even “It is better for everyone involved if the man earns the if the husband is not working” money and the woman takes care of the home and children” female male female male Germany Germany Poland Poland Czech Rep. Turkey Lithuania Slovenia Italy Italy Croatia Croatia Turkey Estonia Estonia Mongolia Cyprus Bosnia and Herz. Slovak Rep. Lithuania Romania Cyprus Bosnia and Herz. Albania Bulgaria Czech Rep. Greece Montenegro Latvia Greece Montenegro Kosovo FYR Macedonia Latvia Mongolia Romania Albania Serbia Serbia Georgia Moldova FYR Macedonia Georgia Bulgaria Kosovo Slovak Rep. Slovenia Moldova Tajikistan Kazakhstan Ukraine Hungary Hungary Belarus Armenia Ukraine Azerbaijan Uzbekistan Kyrgyz Rep. Russia Kazakhstan Tajikistan Belarus Armenia Russia Kyrgyz Rep. Uzbekistan Azerbaijan 0 50 100 0 50 100 Agreement (%) Agreement (%) Source data: LITS III dataset, sample of primary respondents, weighted 35 Figure 26 In all of the countries but one, women agree more with the statement “Women are as competent as men to be business executives”. In almost all of the countries women disagree more with the statement “Men make better political leaders than women do”. Agreement with “Women are as competent as men to be Disagreement with “Men make better political leaders business executives” than women do” female male female male Poland Uzbekistan Turkey Germany Kyrgyz Rep. Azerbaijan Georgia Mongolia Belarus Kyrgyz Rep. Belarus Moldova Uzbekistan Armenia Albania Kazakhstan Russia Ukraine Kosovo Estonia Estonia Tajikistan Turkey Armenia Serbia Czech Rep. FYR Macedonia Bosnia and Herz. Hungary Montenegro Czech Rep. Georgia Romania Croatia FYR Macedonia Latvia Russia Bulgaria Bulgaria Italy Bosnia and Herz. Slovak Rep. Hungary Poland Mongolia Cyprus Cyprus Lithuania Greece Germany Slovenia 0 50 100 0 50 100 Agreement (%) Disagreement (%) Source data: LITS III dataset, sample of primary respondents, weighted 36 Figure 27 Women and men generally have similar opinions for their daughters and sons’ university education Agreement (percentage of population responding “agree” or “strongly agree”) to the terms “It is important that my daughter/son achieves university education”, by country female for daughters female for sons male for daughters male for sons Poland Turkey Hungary Slovak Rep. Belarus Italy Russia Lithuania Romania Bosnia and Herz. Bulgaria Uzbekistan Tajikistan Armenia Azerbaijan Mongolia Kosovo 0 50 100 Agreement(%) Source data: LITS III dataset, sample of primary respondents, weighted Norms are correlated with women’s economic empowerment at the country level (See Figure 30). In countries where a larger share of women is employed, women agree less with gender norms regarding unequal distribution of roles in the household. Additionally, a positive correlation exists between prevalence of women taking roles in economic life and women’s perception of men and women to be equal as business executives or political leaders. The relationship is stronger for the latter. 37 Figure 28 Women’s participation in economic life is associated with norms at the country level Agreement to “A woman should do most of the household Agreement to “It is better for everyone involved if the chores even if the husband is not working” man earns the money and the woman takes care of the home and children” Uzbekistan 100 Russia 100 Azerbaijan Belarus Kazakhstan Kyrgyz Rep. 90 Kyrgyz Rep. 90 Armenia Tajikistan Russia % of female population agreeing % of female population agreeing Azerbaijan Uzbekistan Armenia Ukraine 80 80 Hungary Belarus FYR Hungary Kazakhstan Moldova 70 Tajikistan Ukraine 70 Slovak Rep. Macedonia Bulgaria Georgia Serbia Slovenia Romania KosovoBosnia Montenegro Greece Latvia 60 60 and Albania Czech Rep. Kosovo Georgia Herz. Cyprus Lithuania Moldova Mongolia Estonia 50 Serbia FYR Albania 50 Croatia Italy Mongolia Bosnia and Macedonia 40 Montenegro Greece Latvia Bulgaria 40 Herz. Romania Slovak CyprusRep. Turkey Slovenia Turkey Estonia 30 Croatia Italy Lithuania 30 Czech Rep. 20 20 Poland Germany 10 Poland Germany 10 0 0 0 20 40 60 80 100 0 20 40 60 80 100 employment among women (%) employment among women (%) Agreement to “Women are as competent as men to be Disagreement to “Men make better political leaders than business executives” women do” Bosnia Mongolia and FYRGreece Slovenia Hungary Cyprus Lithuania Slovak Rep. Italy Bulgaria Russia Latvia 100 Croatia Romania Montenegro Czech Rep. 100 Kosovo Herz. Serbia Armenia Macedonia Tajikistan Estonia Ukraine Kazakhstan Albania Uzbekistan Moldova 90 90 Slovenia % of female population disagreeing Kyrgyz Rep. Belarus Germany Georgia Croatia % of female population agreeing Azerbaijan Germany Cyprus Italy 80 80 Bosnia and Greece Poland 70 Kosovo Herz. FYRMontenegr Bulgaria 70 Turkey Macedonia o Lithuania Georgia Serbia Hungary Latvia Turkey Moldova Czech Rep. 60 Poland 60 Albania Romania Estonia Russia 50 50 Slovak Rep. Armenia Ukraine 40 Belarus Kazakhstan 40 Mongolia Azerbaijan Kyrgyz Rep. 30 30 20 20 Tajikistan Uzbekistan 10 10 0 0 0 20 40 60 80 100 0 20 40 60 80 100 employment among women (%) employment among women (%) Source data: LITS III dataset, sample of primary respondents, weighted While it is expected that younger generations display lower adherence to traditional norms, on average, no significant differences are observed between the young and older women and men. Half (53.7 and 51.1 percent) of older -65 or above women and men agree with the statement “It is better for everyone involved if the man earns the money and the woman takes care of the home and children”. Younger people in 38 the survey (18-24) display a slightly lower level of agreement with the statement (48.9 and 47.5), but overall had a similar opinion22. Women’s agreement with unequal norms is most strongly associated with their level of education but also with their participation in economic life (as measured by employment) and owning an asset23. Higher educational attainment is strongly associated with taking a stance against norms. The negative relationship is especially high for the statement on the distribution of roles in the household. The likelihood of women agreeing with the statement “It is better for everyone involved if the man earns the money and the woman takes care of the home and children” drops by 17.5 percentage points if the woman has a higher education degree. Being employed is also negatively associated with agreeing with this norm and it decreases the likelihood of agreeing by 7.3 percentage points (See Annex 5 Annex 5. Table 7 for regression results). A. Voice in household decisions Women’s decision-making indicators have been regularly seen as signals women’s empowerment or bargaining power in the household. Ibrahim and Alkire (2007) draw on these indicators to suggest that signals of choice or control, particularly over personal decisions and in household decision-making are good proxies for women’s empowerment. LITS III captures some signals of empowerment using similar indicators that, albeit imperfect, can shed some light on how power is distributed or perceived to be among men and women. People generally believe that they have a say in household decisions, and both men and women agree that their opinions are considered in decisions made by the household. Overall, 80.5 percent of women and 81.5 percent of men living in a household with at least two adults (aged 18 or more) from the opposite sex agree that their opinions are considered in household decisions. Only in Poland, Turkey and Germany does this share drop down dramatically to around 50 percent, but still, with no visible gender differences (See Figure 31). But a further look into who thinks has a say in decision-making shows some subtle differences. While indeed similar shares of men and women think that they have a say in financial household decisions (See Figure 30), this does not hold for all countries, and we start seeing gender differences in countries like Azerbaijan, Uzbekistan, and Kosovo, among others. Further breaking down decision-realms, we can see that decisions related to children, women have a stronger say compared to men, which while positive, might also just be a factor of women’s role being tilted towards child care and child rearing. On average employed women report having a stronger say in household decisions across domains, ranging from household finances to child care. 67.8 percent of the female population who are not employed respond “mostly me”, “shared between me and my partner” or “shared between me and someone else in the household” when asked about who makes household decision on savings, investment and borrowing. An additional 20 percentage points (moving agreement levels to 86.3 percent) marks the responses from women who are employed24. This positive relationship can also be seen for decisions about child care. 22 See Annex 4 Tables 3 and 4 for cross tabulations 23 To observe what characteristics of women affect their perception of norms, a regression was conducted with the dependent variables being agreement with each norm (agree or strongly agree), controlling for age, marital status, household composition and country fixed effects. 24 See Annex 4 Table 5 and 6 for cross tabulations 39 Figure 29 Regardless of their gender, most individuals think that their opinions are considered in decisions made by the household Agreement to “My opinions are taken into account in decisions made by the household” (% of population living in households with at least two adults from the opposite gender, responding “agree” or “strongly agree”), by country female male Poland Turkey Germany Kazakhstan Ukraine Moldova Serbia Georgia Kosovo Belarus Czech Rep. Kyrgyz Rep. Croatia Uzbekistan Azerbaijan Romania Slovak Rep. FYR Macedonia Bosnia and Herz. Latvia Albania Mongolia Lithuania Slovenia Montenegro Hungary Greece Italy Russia Tajikistan Estonia Bulgaria Armenia Cyprus 0 50 100 % of population living in households with at least one adult (18+) female and at least one adult (18+) male Source data: LITS III dataset, sample of primary respondents living in households with at least two adults from the opposite gender, weighted. Sample size is 36,459 individuals (Whole sample is 51,206). 40 Figure 30 Percent of population having a say in the decision (Mostly me/Shared equally between me and my partner/Shared equally between me and someone else in the household), by country Managing day-to-day spending and Making large household purchases Savings, investment and borrowing paying bills (e.g. cars, major appliances) female male female male female male Azerbaijan Azerbaijan Azerbaijan Uzbekistan Uzbekistan Kosovo Armenia Kosovo Uzbekistan Kosovo Italy Armenia Albania Turkey Turkey Albania Turkey Albania Italy Kyrgyz Rep. Kyrgyz Rep. Cyprus Mongolia Serbia Serbia Serbia Armenia Kyrgyz Rep. FYR Macedonia Poland FYR Macedonia Italy FYR Macedonia Georgia Cyprus Belarus Mongolia Kazakhstan Kazakhstan Latvia Georgia Georgia Croatia Poland Cyprus Poland Montenegro Tajikistan Bosnia and Herz. Latvia Greece Czech Rep. Bulgaria Ukraine Kazakhstan Croatia Bulgaria Estonia Slovenia Bosnia and Herz. Greece Czech Rep. Mongolia Germany Bosnia and Herz. Croatia Ukraine Tajikistan Montenegro Bulgaria Ukraine Czech Rep. Montenegro Greece Moldova Tajikistan Belarus Germany Slovak Rep. Estonia Slovak Rep. Belarus Russia Slovenia Lithuania Slovak Rep. Latvia Slovenia Germany Estonia Hungary Hungary Hungary Russia Lithuania Russia Moldova Moldova Lithuania Romania Romania Romania 0 50 100 0 20 40 60 80 100 0 20 40 60 80 100 % of population living in households with at least % of population living in households with at least % of population living in households with at least one one adult (18+) female and at least one adult (18+) one adult (18+) female and at least one adult (18+) adult (18+) female and at least one adult (18+) male male male The way the children are raised Looking after the children Social life and leisure activities female male female male female male Albania Tajikistan Uzbekistan Turkey Turkey Azerbaijan Tajikistan Albania Kyrgyz Rep. Kyrgyz Rep. Kyrgyz Rep. Albania Armenia Serbia Turkey Serbia Armenia Kosovo Kosovo Italy Armenia Italy Kosovo Serbia Uzbekistan Uzbekistan FYR Macedonia Mongolia Ukraine Tajikistan Kazakhstan Georgia Kazakhstan FYR Macedonia Slovak Rep. Mongolia Bosnia and Herz. Azerbaijan Italy Georgia FYR Macedonia Georgia Ukraine Bosnia and Herz. Ukraine Slovak Rep. Montenegro Croatia Russia Mongolia Bulgaria Montenegro Russia Bosnia and Herz. Croatia Croatia Moldova Azerbaijan Kazakhstan Montenegro Cyprus Cyprus Moldova Slovak Rep. Czech Rep. Belarus Latvia Belarus Greece Greece Moldova Czech Rep. Germany Greece Hungary Belarus Estonia Lithuania Czech Rep. Lithuania Slovenia Estonia Hungary Bulgaria Bulgaria Lithuania Hungary Slovenia Cyprus Poland Germany Russia Latvia Poland Poland Germany Romania Slovenia Romania Latvia Estonia Romania 0 20 40 60 80 100 0 20 40 60 80 100 % of population living in households with at least % of population living in households with at least 0 20 40 60 80 100 one adult (18+) female and at least one adult (18+) one adult (18+) female and at least one adult (18+) % of population living in households with at least one male male adult (18+) female and at least one adult (18+) male Source data: LITS III dataset, sample of primary respondents living in households with at least two adults from the opposite gender, weighted. Sample size is 36,459 individuals (Whole sample includes 51,206 observations). 41 Women who have a higher education degree, older women, women who own assets and who are employed are more likely to have a say in household decisions25. According to the regression results having a higher education degree is positively associated with women’s likelihood of having a say in household decisions. A higher education degree is most positively associated with having a say in one of the financial household decisions. A woman with a higher education degree is 9 percent more likely to report “mostly me”, “shared between me and my partner” or “shared between me and someone else in the household” for the household decision “Making large household purchases (e.g. cars, major appliances)”. Age, asset ownership and being employed are the other factors that are positively associated with having a say in household decisions. Overall, financial empowerment and overall empowerment is more likely for these types of women. Women’s own characteristics determine their voice in the household more than the characteristics of their partner or other adult man interviewed in the same household. According to the results, the characteristics of the male secondary respondent are generally not associated with women’s say in household decisions. When it comes to the man’s age, his asset ownership and his employment, these characteristics are negatively associated with women’s voice only of the household financial decisions. For instance, if the man owns an asset, the woman is less likely to have a say in day to day household spending (his asset ownership is not associated with other financial decisions in a statistically significant way) while his employment decreases the likelihood of her having a say in large household purchases (his employment is not associated with other financial decisions in a statistically significant way). Also, the older the man is, the weaker her voice is in household decisions on making large household purchases and decisions on savings while the male respondent's age is not statistically significantly associated with rest of the decisions. Conclusions The Life in Transition Survey III offers a unique possibility to look at some factors affecting women’s and men’s life differently across the countries included in the survey. Ranging from women’s employment, entrepreneurship, care needs and provision in the household, asset ownership and attitudes, to norms and values, the results provide a mixed picture of progress and stagnation towards greater gender equality. Women and men alike have (and had for a while) close gaps in educational attainment, but their employment rates are quite unequal across countries, with women being less likely to work for pay because of household duties. Women persist to have a harder time to join the labor market, particularly young ones, where a sizable number appears as not being in education, employment or training. Contrasting with other world regions, the Europe and Central Asia sees fewer attempts to be an entrepreneur among its population. Only 5% of the sampled adult women have ever attempted setting up their own business, and while higher among men, the numbers are still low. Women attempt less to be an entrepreneur but when they do, they succeed at a similar rate in continuing the business compared to men. Women entrepreneurs are slightly more likely to be the sole owner of their businesses while their ownership of assets is lower compared to male business owner. This might have to do with sectoral concentration in services and commerce (shops or stands). 25 See Annex 5 Annex 5. Table 8 Women’s having a say in household decisions for regression result including both primary and secondary respondents’ characteristics and country fixed effects, 42 One of the most frequently found factors affecting women’s economic participation, childcare, features prominently across the 34 countries surveyed in LITS III. One-in-four people on the sample lives in a household with child, elderly or disabled care need. Across countries, household members are the primary providers of care, and although there is evidence of institutional child care, this is not used as much as in average OECD countries, which may be due to either supply or demand side constraint, which if solved will prove beneficial. Women are 27.3 percentage points more likely to be employed if the household utilizes institutional care for at least one child. In the 34 countries examined, women are at a disadvantage overall in terms of holding an asset, a difference that not disappears with age, although women and men alike grow their assets pool with time. When it comes to asset ownership, men are more likely to solely own an asset compared to women. There is a mixed picture when it comes to social norms, who seems to be ‘in transition’. While across the 34 countries included in LITS III, most of the population (men or women) think that having equal rights for women as citizens is important, the actual practice of such equality seems more contested, and the belief in equality in the distribution of men’s and women’s roles in the household is less common than the ideal situation in many countries. Norms acceptance and change are correlated with women’s economic empowerment at the country level. At the intra-household level, women tend to have a more pronounced role in decisions regarding children compared to men, while men seem to have more of a role on financial decision-making in a household. However, given the stronger role women’s own characteristics play in determining their voice in the household, it is expected that in the near future, as women gain space in public and economic life, they’ll also gain space inside the household. 43 Annexes Annex 1 Data Cross Checks Educational Attainment, Employment Rates in LITS III Data LITS III is representative of the countries’ populations. However, cross-checks with other datasets suggest that LITS III, in some cases, tends to over or underrepresent some groups in the population. For a number of countries, the LITS III data seems to represent the population either as overeducated (Turkey) or undereducated (Azerbaijan) when compared to UNESCO statistics (See Figure 31). A comparison using the latest available statistics in UNESCO’s database for the educational attainment of women shows that in 6 countries (out of 19) the difference between UNESCO statistics and LITS III is more than 10 percentage points, the most extreme cases being Azerbaijan and Turkey. In Azerbaijan, according to UNESCO’s statistics 85.5 percent of women aged 25 or more have an upper secondary education or more as opposed to 55.4 percent calculated using LITS III. In contrast, for Turkey, LITS III represents women as overeducated. 28.5 percent of women aged 25 or more have an upper secondary education or more in the country according to UNESSCO’s statistics while using the LITS III dataset, this statistic is calculated to be 55.2 percent. Similar discrepancies can be seen for the educational attainment of men as well. In 10 countries (out of 19), the difference between the two educational attainment values are more than 10 percentage points. Figure 31 For a number of countries LITS III data represents the population either as overeducated (Turkey) or undereducated (Azerbaijan) compared to UNESCO statistics Female population (aged 25+) with upper secondary Male population (aged 25+) with upper secondary level level education or more education or more 100 100 90 80 80 60 70 60 40 50 20 40 30 0 20 Cyprus Czech Rep. Hungary Azerbaijan Bosnia and Herz. Bulgaria Germany Poland Slovak Rep. Latvia Serbia Greece Moldova Slovenia Turkey Estonia Georgia Lithuania Romania 10 0 Czech Rep. Bosnia and Herz. Bulgaria Hungary Slovak Rep. Azerbaijan Cyprus Latvia Poland Germany Moldova Serbia Lithuania Romania Slovenia Turkey Estonia Georgia Greece UNESCO LITSIII UNESCO LITSIII Source data: LITS III dataset, sample of primary respondents aged 25+, weighted. UNESCO statistics are obtained from UNESCO’s database for the countries that have data for 2014 (latest year available), the indicator used is “Educational attai nment: at least completed upper secondary (ISCED 3 or higher), population 25+ years” In terms of employment to population ratio, LITS III over represents or underrepresents the employed population in a number of countries compared to the World Bank’s WDI statistics (See Figure 32). The statistics used for comparison are the modelled ILO estimates for the population aged 15 or more while LITS collects data from individuals who are 18 years old or older. However, a large discrepancy between the two statistics is not to be expected. For 12 countries (out of 33) the LITS III dataset over or underestimates the female employment to population ratio by more than 10 percentage points. This gap is particularly large for Azerbaijan (a difference of 48.6 percentage points). Similarly, for men, for 13 countries (out of 33) the LITS III dataset 44 over or underestimates the male employment to population ratio by more than 10 percentage points. In particular, for the Kyrgyz Republic the two ratios are considerably different (a difference of 28.4 percentage points). i Figure 32 For a number of countries the LITS III dataset overestimates or underestimates the female employment to population ratio or male employment to population ratio by more than 10 percentage points compared to WDI statistics Percent of female population employed 100 90 80 70 60 50 40 30 20 10 0 Russia Czech Rep. Armenia Azerbaijan Bosnia and Herz. Hungary Bulgaria Italy Latvia Montenegro Poland Slovak Rep. Uzbekistan Cyprus FYR Macedonia Germany Kyrgyz Rep. Tajikistan Ukraine Albania Croatia Kazakhstan Moldova Serbia Belarus Estonia Georgia Greece Lithuania Mongolia Romania Slovenia Turkey WDI LITS III Percent of male population employed 100 90 80 70 60 50 40 30 20 10 0 Germany Hungary Poland Russia Czech Rep. Albania Armenia Azerbaijan Bosnia and Herz. Bulgaria Slovak Rep. Uzbekistan Cyprus FYR Macedonia Italy Kazakhstan Montenegro Kyrgyz Rep. Latvia Croatia Moldova Serbia Tajikistan Belarus Estonia Georgia Lithuania Mongolia Romania Slovenia Turkey Ukraine Greece WDI LITS III Source data: LITS III dataset, sample of primary respondents aged 18+, weighted. Individuals are counted as employed if they answered «yes» to the question «Did you work during the past 12 months?» WDI statistics are obtained from World Bank’s World Development Indicators database for the countries that have data for 2014 (latest year available), the indicator used is «Employment to population ratio, 15+, male (%) (modeled ILO estimate) 45 Annex 2. Construction of the Empowerment Indices Using the answers to the question “Who makes the decisions about the following issues in your household?” empowerment indices are calculated for individuals. The issues listed are as follows: • Managing day-to-day spending and paying bills • Making large household purchases (e.g. cars, major appliances) • The way the children are raised • Social life and leisure activities • Savings, investment and borrowing • Looking after the children For each issue individuals select one of these answers: (i) Mostly me, (ii) Shared equally between me and my partner, (iii) Mostly my partner, (iv) Shared equally between me and someone else in the household, (v) Mostly someone else in the household , (vi) Mostly someone else not currently living in the household Financial empowerment index: For each household issue related with finances (issues numbered 1,2 and 5) each individual receives one point if their answer is “mostly me” or “shared between me and my partner equally” or “shared between me and other household member equally” to the question “Who makes the decisions about the following issues in your household? (Question 4.27.A – 4.27.F)”. Hence an individual can take at most 3 points, since there are 3 financial issues being questioned in the survey. It is also necessary to take into account the number of questions answered by the individual. Hence the scores are further adjusted by dividing the score by total number of questions answered, hence the final index ranges between 0 and 1. Overall empowerment index: There is also an overall empowerment index where individual’s answers to all the statements above are calculated. For this index, top score can be 6. Since it is necessary to take into account the number of the questions answered by the individual, the scores are further adjusted by dividing the score by total number of questions answered, hence the final index ranges between 0 and 1. Example: If a woman answers all of the 6 questions and gives the answer «mostly me» to all, then she gets: Financial empowerment index=3/3=1 Overall empowerment index=6/6=1. If a woman answers as «mostly my partner» to issue 1 and 2 and «shared between me and my partner» to the rest, then she gets: Financial empowerment index=1/3=0.33 Overall empowerment index=4/6=0.67 46 Annex 3. Laws on Ownership of property Country What is the default Who legally If the husband Are there Does the law Do Do married Do sons and Do marital property regime? administers marital administers special provide for the unmarried men and daughters female property? property, is provisions valuation of men and married have equal and male spousal consent for major nonmonetary unmarried women have rights to surviving required for major transactions contributions? women have equal inherit assets spouses transactions? concerning equal ownership from their have the marital ownership rights to parents? equal home? rights to property? rights to property? inherit assets? Partial community of Albania property Both must agree N/A Yes Yes Yes Yes Yes Yes Partial community of Armenia property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Azerbaijan property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Belarus property Both must agree N/A No Yes Yes Yes Yes Yes Bosnia and Partial community of Herzegovina property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Bulgaria property Both must agree N/A Yes Yes Yes Yes Yes Yes Partial community of Croatia property Both must agree N/A Yes Yes Yes Yes Yes Yes Cyprus Separation of property Original owner N/A No No Yes Yes Yes Yes Partial community of Czech Republic property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Estonia property Both must agree N/A No Yes Yes Yes Yes Yes FYR Partial community of Macedonia property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Georgia property Both must agree N/A No Yes Yes Yes Yes Yes Deferred community of Separate with spousal Germany property consent N/A No Yes Yes Yes Yes Yes Greece Separation of property Original owner N/A No No Yes Yes Yes Yes Partial community of Hungary property Both must agree N/A Yes Yes Yes Yes Yes Yes Partial community of Italy property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Kazakhstan property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Kosovo property Both must agree N/A No Yes Yes Yes Yes Yes Kyrgyz Partial community of Republic property Both must agree N/A No Yes Yes Yes Yes Yes Latvia Other Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Lithuania property Both must agree N/A Yes Yes Yes Yes Yes Yes Partial community of Moldova property Both must agree N/A No Yes Yes Yes Yes Yes Mongolia Other Other N/A No Yes Yes Yes Yes Yes Partial community of Montenegro property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Poland property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Romania property Both must agree N/A Yes Yes Yes Yes Yes Yes Russian Partial community of Federation property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Serbia property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Slovak Republic property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Slovenia property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Tajikistan property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Turkey property Both must agree N/A Yes Yes Yes Yes Yes Yes Partial community of Ukraine property Both must agree N/A No Yes Yes Yes Yes Yes Partial community of Uzbekistan property Both must agree N/A No Yes Yes Yes Yes Yes Source: World Bank Women, Business and the Law dataset for year 2016 47 Annex 4. Cross tabulations Annex 4. Table 1 Asset and bank account ownership by individual and household characteristics Dwelling Land Dwelling or Bank account Bank account ownership ownership land ownership ownership ownership (Sole) (Sole or joint) F M F M F M F M F M Individual characteristics Age of the primary respondent 18-24 14.6 22.8 3.7 2.8 17.6 24.7 51.7 55.5 54.6 60.4 25-34 35.5 43.3 3.4 3.8 37.2 44.7 56.0 64.1 64.3 70.5 35-44 51.0 64.1 6.8 8.0 53.5 65.6 59.3 59.5 70.0 70.1 45-54 56.8 70.7 7.8 12.8 58.9 72.1 58.3 66.4 67.7 75.7 55-64 64.2 76.9 9.4 10.7 66.6 77.7 59.1 61.9 68.7 74.6 65+ 70.4 85.3 10.9 13.4 73.2 86.3 48.5 51.7 63.4 73.0 Highest education completed Less than (upper) secondary education 49.1 63.1 11.1 10.4 52.5 64.6 44.9 51.1 61.5 69.2 (Upper) secondary education 44.6 53.9 6.4 9.3 47.1 55.7 51.9 59.6 61.6 68.9 Higher education 58.4 66.3 6.0 7.4 60.1 67.3 63.9 64.9 70.4 73.6 Marital status Single (never married) 31.1 30.4 4.0 4.0 32.8 31.9 65.1 67.3 69.2 71.6 Married 48.1 72.8 8.0 10.3 50.9 74.2 48.4 57.2 63.6 71.5 Widowed 77.5 73.9 11.3 15.2 80.4 74.8 59.0 60.7 61.2 65.0 Divorced 69.0 50.2 4.5 5.2 69.8 51.0 70.8 64.4 73.3 66.2 Separated 50.6 50.6 2.4 5.4 51.2 51.0 67.5 71.2 73.6 79.5 Employment status Not employed in the past 12 months 36.2 47.8 6.0 7.0 38.7 49.4 42.7 49.1 50.5 55.1 Employed in the past 12 months 53.8 60.7 6.6 8.1 55.9 62.1 67.5 66.8 76.6 76.8 Household characteristics Gender of the household head Female 66.6 43.6 7.3 7.6 68.4 45.3 64.0 54.7 67.9 62.9 Male 39.4 64.9 7.4 8.9 42.2 66.1 48.6 61.4 63.5 72.7 Household size (effective) 1 67.0 51.0 6.9 5.5 68.5 51.6 70.9 74.7 72.7 76.9 2 57.8 68.0 8.1 9.3 59.9 68.7 55.7 60.8 68.6 74.5 3 44.3 60.5 6.2 8.0 46.5 61.7 57.0 60.1 68.8 70.7 4 47.3 62.6 7.3 9.2 49.6 64.2 49.5 54.1 62.1 70.1 5 or more 29.1 54.5 7.8 11.1 33.6 58.4 35.9 50.4 42.7 56.3 Households with children aged 0-17 Without 54.9 60.4 7.5 8.5 57.0 61.4 58.4 62.1 68.1 72.7 With 43.8 63.9 6.8 9.1 46.9 65.8 48.5 56.3 59.0 67.5 Households with children aged 0-6 Without 53.4 61.8 7.5 8.6 55.7 63.0 56.9 61.2 66.8 72.0 With 39.5 59.4 6.2 8.9 42.3 61.7 45.9 55.1 55.8 65.8 Households with children aged 7-17 Without 53.3 59.8 7.4 8.4 55.4 61.1 57.5 61.5 67.3 72.3 With 45.3 67.4 7.2 9.7 48.5 69.1 47.7 55.8 58.2 66.8 Households with elderly aged 65+ Without 47.3 58.8 5.8 7.3 49.2 59.9 57.9 63.1 66.6 72.0 With 62.2 70.5 11.1 13.1 65.7 72.7 49.4 50.9 62.4 68.1 Households with elderly aged 75+ Without 50.0 60.8 7.0 8.1 52.2 62.1 57.3 61.4 66.8 71.5 With 66.4 68.1 10.7 15.1 69.8 70.2 38.6 49.1 52.8 66.9 Urbanity status Urban 51.6 60.6 4.1 4.7 53.0 61.3 59.8 63.9 69.7 75.3 Rural 51.5 63.7 15.9 18.8 56.6 66.6 43.9 51.3 54.0 60.5 Country Albania 26.9 59.2 9.4 19.6 32.9 62.6 37.7 48.2 46.7 55.5 Armenia 33.4 48.4 5.6 16.7 34.9 53.8 19.6 27.1 21.1 28.5 Azerbaijan 13.1 70.4 3.1 3.1 15.7 72.1 2.6 4.3 5.5 5.0 Belarus 58.0 68.0 4.6 3.8 59.9 68.7 64.4 68.4 71.1 73.8 Bosnia and Herz. 28.1 70.3 14.6 27.4 35.5 74.1 63.7 80.0 65.8 82.2 Bulgaria 60.1 69.4 18.7 19.8 63.1 70.9 68.6 71.0 69.0 71.8 Croatia 45.3 63.8 16.3 20.6 50.9 65.6 89.5 93.4 90.7 94.7 Cyprus 57.5 37.7 21.1 17.3 61.2 44.0 74.9 74.5 89.8 90.1 Czech Rep. 60.7 58.6 5.8 7.8 61.4 59.7 78.3 76.1 89.9 90.0 Estonia 61.6 57.1 9.0 10.4 63.4 58.6 98.0 96.5 99.0 97.3 FYR Macedonia 21.8 59.8 8.9 20.4 28.0 63.5 67.2 76.7 68.3 78.2 48 Georgia 39.7 61.5 27.3 38.7 53.7 74.9 47.8 42.0 53.4 50.4 Germany 44.1 51.6 13.9 6.4 47.9 52.1 79.4 76.5 97.3 95.1 Greece 45.6 55.7 20.5 32.4 53.0 60.5 45.3 54.1 91.4 94.0 Hungary 71.0 64.7 1.8 1.8 71.6 64.8 48.2 48.6 60.0 60.7 Italy 48.4 58.2 3.3 6.9 49.5 59.5 50.4 57.2 86.1 89.8 Kazakhstan 43.9 56.9 7.8 10.8 46.1 58.0 44.6 44.7 51.6 49.6 Kosovo 12.2 51.7 10.5 28.2 19.2 58.1 37.4 65.3 40.8 66.9 Kyrgyz Rep. 28.1 56.4 20.5 22.6 42.2 61.8 12.2 11.2 18.8 17.8 Latvia 49.1 42.7 10.1 11.4 51.4 45.7 91.5 90.8 91.9 91.2 Lithuania 63.1 55.7 18.6 19.2 65.5 60.5 92.8 92.0 92.9 93.2 Moldova 48.0 65.8 22.4 24.8 57.9 72.8 9.9 9.9 10.9 10.7 Mongolia 40.9 64.5 5.6 9.4 44.1 67.6 82.6 84.0 86.8 86.7 Montenegro 25.6 58.7 7.5 16.3 28.8 60.5 69.5 70.9 74.3 76.5 Poland 59.3 47.3 3.5 7.2 59.7 48.6 60.0 56.9 70.4 74.6 Romania 66.4 66.4 16.8 19.6 69.0 69.7 41.3 36.7 43.0 40.1 Russia 70.1 74.0 4.0 6.5 71.1 74.2 57.9 64.5 61.1 69.4 Serbia 30.7 60.6 11.2 23.4 35.9 64.4 71.2 81.2 74.0 83.5 Slovak Rep. 64.6 64.1 6.0 7.7 64.7 65.4 69.2 71.8 85.8 88.1 Slovenia 54.3 59.2 11.3 14.3 56.1 61.3 97.6 98.0 98.1 98.7 Tajikistan 30.3 55.8 27.2 28.6 51.0 67.8 10.2 10.8 11.9 12.5 Turkey 21.2 51.1 1.7 1.6 22.2 51.4 46.7 61.3 48.5 63.5 Ukraine 73.1 78.2 10.4 11.2 76.7 80.4 52.5 60.9 57.7 66.5 Uzbekistan 22.4 56.2 2.7 3.7 24.1 57.2 22.4 28.1 24.2 29.2 Total 51.6 61.5 7.3 8.7 54.0 62.8 55.5 60.3 65.4 71.1 Note: LITS III dataset, sample of primary respondents, weighted. F stands for female respondents and M stands for male respondents. Annex 4. Table 2 Different types of asset ownership by individual and household characteristics Other Land Dwelling Dwelling Land Land dwelling ownership ownership ownership ownership ownership ownership with the (at least one (all jointly) (at least one (all jointly) with the right to sell solely) solely) right to sell F M F M F M F M F M F M Individual characteristics Age of the primary respondent 18-24 66.4 59.4 37.7 63.5 7.2 8.8 7.4 14.0 1.4 1.6 2.3 1.1 25-34 63.5 57.1 48.0 61.5 17.9 29.2 17.6 14.1 2.0 2.6 1.4 1.2 35-44 66.6 63.3 48.2 48.9 22.8 38.0 28.3 26.1 3.2 4.9 3.6 3.1 45-54 68.7 67.8 46.1 63.7 29.5 42.2 27.3 28.6 3.5 9.6 4.4 3.2 55-64 70.5 64.9 59.0 68.2 36.5 47.3 27.7 29.6 5.9 8.1 3.6 2.6 65+ 75.1 60.0 63.3 61.0 43.4 47.2 27.0 38.0 7.3 8.9 3.5 4.5 Highest education completed Less than (upper) secondary education 71.6 65.6 51.7 60.0 25.2 36.3 23.9 26.9 5.2 6.8 5.8 3.6 (Upper) secondary education 64.2 62.1 51.9 66.3 22.0 33.2 22.6 20.7 4.0 6.9 2.5 2.5 Higher education 71.6 62.6 58.5 56.7 34.0 38.7 24.4 27.6 3.7 5.0 2.3 2.4 Marital status Single (never married) 70.5 58.8 47.4 61.3 22.1 19.4 9.0 11.0 2.0 2.7 1.9 1.3 Married 65.3 62.1 43.8 58.8 13.4 40.3 34.8 32.5 3.5 6.9 4.6 3.4 Widowed 80.6 80.5 82.5 74.4 68.6 58.2 8.9 15.7 9.5 13.0 1.8 2.2 Divorced 79.9 64.8 60.7 77.4 56.8 41.0 12.2 9.2 4.2 4.3 0.3 0.8 Separated 50.6 72.6 95.0 92.1 45.7 47.6 4.9 3.0 2.3 5.1 0.1 0.3 Employment status Not employed in the past 12 months 64.3 67.0 53.3 61.6 17.7 30.4 18.5 17.4 3.2 4.9 2.8 2.1 Employed in the past 12 months 68.9 62.5 47.9 60.9 28.0 35.9 25.8 24.8 3.3 5.7 3.3 2.4 Household characteristics Gender of the household head Female 77.8 57.6 68.7 42.1 51.3 12.3 15.3 31.3 5.7 3.5 1.6 4.1 Male 60.6 64.0 42.5 64.0 8.7 40.9 30.7 24.0 2.9 6.5 4.5 2.4 Household size (effective) 1 86.0 84.0 82.1 81.8 66.7 50.9 0.3 0.1 6.7 5.4 0.2 0.0 2 69.6 48.8 53.6 63.1 24.1 35.1 33.6 32.8 4.2 6.4 3.9 2.9 3 57.2 67.3 49.6 60.5 18.8 33.3 25.5 27.2 3.2 5.3 3.1 2.7 4 65.3 63.2 44.9 53.9 17.3 29.6 29.9 33.0 3.2 5.8 4.1 3.4 5 or more 60.1 72.8 36.2 52.3 11.0 37.8 18.1 16.7 3.2 7.2 4.6 4.0 Households with children aged 0-17 Without 71.4 63.3 58.5 66.4 31.1 35.7 23.8 24.7 4.5 6.1 3.1 2.4 With 64.2 61.7 43.1 49.8 20.5 37.8 23.3 26.1 3.3 5.8 3.5 3.3 49 Households with children aged 0-6 Without 70.1 62.2 56.2 62.5 29.5 36.0 24.0 25.8 4.3 6.0 3.2 2.6 With 66.4 66.2 37.7 52.6 17.8 38.5 21.7 20.9 3.1 5.9 3.0 2.9 Households with children aged 7-17 Without 71.1 62.7 56.3 65.1 29.6 35.8 23.7 24.0 4.3 6.0 3.0 2.4 With 64.0 63.3 46.2 48.1 21.6 38.3 23.7 29.1 3.5 6.1 3.8 3.6 Households with elderly aged 65+ Without 67.4 62.5 50.9 61.9 24.7 35.7 22.5 23.0 3.0 5.4 2.8 2.0 With 74.5 63.7 58.4 59.4 35.8 38.4 26.5 32.1 7.0 8.2 4.1 4.9 Households with elderly aged 75+ Without 70.1 62.1 52.6 60.1 26.1 36.1 23.8 24.7 3.9 5.6 3.1 2.4 With 67.2 69.7 63.6 66.6 44.0 38.5 22.4 29.6 6.6 10.0 4.1 5.1 Urbanity status Urban 70.9 63.8 64.7 68.3 28.5 35.0 23.2 25.5 2.8 3.5 1.2 1.2 Rural 65.1 59.9 47.0 56.4 26.5 39.7 25.0 24.0 7.6 12.3 8.3 6.5 Country Albania 44.1 64.4 26.5 52.5 6.4 38.9 20.5 20.4 4.4 8.9 5.1 10.7 Armenia 42.8 52.2 57.8 73.1 18.5 31.3 14.9 17.1 3.1 12.7 2.5 4.0 Azerbaijan 40.1 84.4 47.8 77.1 12.6 70.4 0.5 0.1 0.5 1.2 2.6 1.8 Belarus 59.7 81.3 62.9 65.4 37.8 42.4 20.2 25.6 2.1 2.5 2.6 1.3 Bosnia and Herz. 86.1 83.2 88.1 85.9 20.9 64.8 7.2 5.5 6.3 23.4 8.3 4.0 Bulgaria 42.0 38.7 56.0 60.4 25.6 32.4 34.5 37.0 13.0 14.9 5.7 4.8 Croatia 59.7 57.6 58.2 74.7 26.3 47.7 19.0 16.1 8.7 17.0 7.6 3.6 Cyprus 57.2 58.9 72.6 72.5 37.6 17.0 19.9 20.7 19.4 16.1 1.7 1.2 Czech Rep. 50.6 55.0 50.2 66.8 30.4 28.5 30.3 30.1 3.0 6.0 2.8 1.8 Estonia 57.2 53.1 56.4 61.0 46.4 36.8 15.2 20.2 6.4 8.2 2.5 2.2 FYR Macedonia 58.9 81.9 82.2 88.3 12.1 52.3 9.8 7.5 4.3 18.2 4.7 2.2 Georgia 60.6 74.0 45.2 58.0 31.1 50.2 8.6 11.4 20.2 31.3 7.1 7.5 Germany 80.7 38.7 36.8 42.7 12.7 16.8 31.5 34.8 4.5 2.4 9.4 4.0 Greece 67.1 76.6 77.6 81.6 31.1 39.5 14.5 16.3 18.3 28.6 2.2 3.8 Hungary 54.7 58.9 76.7 61.6 28.0 18.9 43.0 45.8 1.0 0.7 0.8 1.1 Italy 67.0 62.0 51.7 62.0 19.2 32.2 29.2 26.0 2.3 4.9 1.0 2.0 Kazakhstan 70.7 66.9 84.4 90.9 29.3 42.1 14.5 14.8 7.0 9.8 0.9 1.0 Kosovo 67.2 75.9 44.5 67.4 6.9 43.3 5.3 8.4 2.2 16.5 8.3 11.8 Kyrgyz Rep. 64.1 60.3 64.5 66.3 21.8 50.2 6.3 6.2 18.0 21.6 2.5 1.1 Latvia 71.1 65.8 79.8 84.3 45.1 38.5 4.0 4.2 8.9 10.9 1.2 0.5 Lithuania 60.8 58.0 71.1 56.5 44.6 31.9 18.5 23.8 14.0 13.1 4.6 6.0 Moldova 58.1 53.7 79.6 67.2 34.0 45.8 14.0 20.1 20.4 22.1 2.0 2.6 Mongolia 35.0 33.5 50.7 58.6 15.4 41.4 25.4 23.1 4.3 7.3 1.3 2.1 Montenegro 68.3 88.0 74.8 84.1 21.4 56.3 4.2 2.5 3.8 13.6 3.7 2.7 Poland 45.0 46.4 58.0 46.7 31.5 23.7 27.8 23.5 2.1 4.9 1.4 2.2 Romania 31.0 34.6 44.9 57.1 25.9 28.7 40.4 37.8 8.4 12.5 8.4 7.1 Russia 76.9 65.3 85.3 68.5 48.6 48.3 21.5 25.7 2.9 4.8 1.0 1.6 Serbia 72.7 68.3 73.5 83.1 23.0 54.0 7.7 6.6 6.3 19.5 4.9 3.9 Slovak Rep. 46.2 43.9 66.2 59.4 25.2 23.7 39.4 40.4 4.3 4.8 1.7 2.9 Slovenia 58.3 63.7 54.5 62.7 27.5 30.1 26.8 29.1 6.7 8.8 4.6 5.5 Tajikistan 48.7 48.9 12.1 21.4 5.0 28.5 25.3 27.3 3.3 10.9 23.9 17.7 Turkey 57.5 81.3 70.3 97.7 5.6 36.5 15.6 14.6 0.9 1.6 0.9 0.0 Ukraine 52.0 53.3 38.8 36.1 39.4 38.3 33.7 40.0 6.9 6.3 3.6 4.9 Uzbekistan 62.2 80.0 38.7 48.9 18.0 53.7 4.4 2.4 1.0 2.6 1.7 1.1 Total 69.8 62.8 54.2 61.0 27.9 36.3 23.7 25.1 4.1 6.0 3.2 2.7 Note: LITS III dataset, sample of primary respondents, weighted. F stands for female respondents and M stands for male respondents. 50 Annex 4. Table 3 Agreement with norms by individual and household characteristics Equal rights for Equal rights for Women are as Men make better DISAGREE: Men women as citizens women as citizens competent as men political leaders make better are important for my exist in my country to be business than women do political leaders country executives than women do F M F M F M F M F M Individual characteristics Age of the primary respondent 18-24 85.7 82.9 51.8 59.2 81.8 72.5 38.9 47.2 54.1 45.6 25-34 87.2 80.2 55.9 55.4 83.7 73.0 40.6 49.2 53.6 40.9 35-44 86.8 83.7 54.0 53.9 83.6 76.4 38.3 52.6 55.6 38.2 45-54 88.8 84.6 54.9 55.4 85.0 75.1 37.1 46.2 55.4 44.3 55-64 88.4 89.0 56.1 55.5 83.6 76.0 40.8 48.1 53.4 40.9 65+ 90.5 87.5 61.2 47.8 84.4 69.6 39.4 42.1 50.3 43.4 Highest education completed Less than (upper) secondary education 87.7 84.1 55.7 50.1 77.8 67.6 34.5 39.5 52.7 47.3 (Upper) secondary education 88.4 85.2 55.7 54.0 84.1 73.3 37.1 45.0 55.0 42.6 Higher education 88.1 84.1 56.7 56.6 86.8 77.2 43.3 53.5 53.0 39.2 Marital status Single (nevermarried) 88.9 85.5 52.2 57.8 86.8 74.1 30.7 39.6 63.6 51.4 Married 87.0 83.9 56.5 52.2 81.1 74.4 40.2 50.4 51.0 39.2 Widowed 91.3 85.7 59.9 61.7 87.4 73.4 42.6 53.0 51.7 32.4 Divorced 88.9 84.3 55.0 64.0 86.9 68.7 43.0 49.7 53.0 39.6 Separated 85.3 88.5 58.7 44.6 92.7 72.0 41.3 40.3 55.7 54.3 Employment status Not employed in the past 12 months 84.6 81.5 51.8 56.0 80.4 69.0 41.0 45.3 50.6 41.0 Employed in the past 12 months 89.5 84.8 56.8 55.4 85.9 76.9 37.8 50.2 57.2 42.0 Household characteristics Gender of the household head Female 89.2 83.3 58.1 56.8 86.7 78.2 39.9 47.5 54.5 46.1 Male 87.3 84.7 54.4 54.0 81.4 73.2 38.4 47.8 53.0 41.2 Household size (effective) 1 90.9 86.9 61.9 61.4 87.8 71.6 32.2 37.1 62.0 51.1 2 87.2 84.9 54.2 53.8 83.7 73.7 35.5 44.4 55.3 42.7 3 86.9 81.5 53.4 52.6 80.5 74.9 41.0 49.0 51.6 41.7 4 87.6 85.9 53.3 48.6 85.9 74.3 40.4 48.6 55.0 43.3 5 or more 89.6 84.7 61.7 59.2 82.0 75.1 55.4 66.7 38.2 28.2 Households with children aged 0-17 Without 88.0 84.8 55.3 53.6 83.7 73.8 36.1 44.4 55.9 44.1 With 88.4 83.8 58.0 56.1 84.1 74.3 46.3 55.4 48.3 37.1 Households with children aged 0-6 Without 88.0 84.4 55.6 53.4 84.0 74.0 37.0 46.4 55.4 42.9 With 88.7 85.0 59.8 60.1 82.8 73.4 53.3 56.3 41.5 36.4 Households with children aged 7-17 Without 88.1 84.6 55.7 54.0 83.3 74.0 37.7 45.4 54.5 43.6 With 88.3 84.0 57.6 55.8 85.5 73.6 44.5 56.6 50.1 36.1 Households with elderly aged 65+ Without 87.4 83.6 54.8 55.7 83.4 74.8 38.2 47.9 55.1 42.5 With 89.9 87.3 59.5 50.2 84.7 71.2 41.6 47.6 49.7 40.0 Households with elderly aged 75+ Without 87.6 84.3 56.0 55.4 83.7 74.6 38.2 47.4 54.4 42.5 With 92.5 86.3 57.1 43.8 84.6 67.1 47.8 52.1 46.3 36.5 Urbanity status Urban 87.3 83.3 55.5 53.1 83.9 73.7 38.6 46.8 54.2 42.5 Rural 90.2 87.4 57.6 57.7 83.4 74.6 40.6 50.5 51.8 40.7 Country Albania 82.3 83.3 39.1 41.6 86.1 76.4 46.9 66.4 53.1 33.6 Armenia 95.1 88.4 38.6 36.7 90.1 72.4 62.6 71.9 37.4 28.1 Azerbaijan 77.5 68.1 45.4 50.5 76.1 79.0 57.7 86.2 24.8 8.3 Belarus 84.4 76.1 60.8 57.6 80.2 57.2 57.5 72.8 33.8 21.5 51 Bosnia and Herz. 88.6 85.8 35.6 35.7 91.5 88.0 33.0 49.7 66.6 49.7 Bulgaria 91.3 89.1 62.0 65.7 94.4 86.4 36.8 63.8 63.2 36.2 Croatia 92.0 91.3 40.7 43.3 93.1 86.1 23.8 37.3 76.2 62.7 Cyprus 95.8 95.2 44.3 49.9 96.0 95.2 24.9 31.7 75.1 68.3 Czech Rep. 86.4 82.8 43.5 51.3 92.8 89.0 44.2 59.1 55.8 40.9 Estonia 95.4 92.5 55.2 65.3 89.4 88.0 43.3 52.2 52.7 45.6 FYR Macedonia 87.8 86.9 39.4 46.2 90.8 87.3 40.0 51.0 60.0 49.0 Georgia 90.3 82.9 36.4 45.9 76.7 64.1 37.3 49.0 58.3 46.1 Germany 99.1 98.1 68.9 70.5 75.1 62.2 3.6 4.9 78.6 71.6 Greece 93.9 89.3 60.6 69.4 97.1 90.0 32.0 49.2 68.0 50.8 Hungary 90.8 89.2 48.0 51.0 95.6 87.2 42.7 49.2 57.3 50.8 Italy 91.8 87.1 41.6 43.1 94.7 87.1 25.2 34.3 74.8 65.7 Kazakhstan 90.6 86.9 68.8 70.9 86.1 75.4 62.7 67.6 33.4 29.3 Kosovo 89.3 89.5 21.6 28.4 88.8 80.9 35.5 44.1 64.5 55.9 Kyrgyz Rep. 78.9 79.2 57.4 59.8 80.4 76.5 75.8 73.9 22.9 24.3 Latvia 95.2 94.2 61.4 67.5 93.7 90.0 42.3 45.6 57.7 54.4 Lithuania 94.9 93.3 56.0 65.1 96.8 90.5 37.5 44.9 62.5 55.1 Moldova 90.8 92.0 29.7 26.3 85.5 83.3 41.7 57.0 55.4 40.7 Mongolia 92.0 93.1 45.2 48.9 96.0 95.3 72.3 77.3 27.7 22.7 Montenegro 90.5 89.9 32.1 35.4 92.6 83.5 40.1 63.7 59.9 36.3 Poland 94.1 87.6 77.6 70.7 54.2 34.8 4.5 9.1 68.7 47.3 Romania 91.7 92.6 47.7 54.1 93.0 90.4 49.5 55.6 50.5 44.4 Russia 86.4 80.3 60.4 52.3 94.0 80.5 55.1 76.4 44.9 23.6 Serbia 88.6 88.1 45.5 49.5 90.2 80.0 43.3 55.6 56.7 44.4 Slovak Rep. 89.4 90.3 42.8 42.8 95.2 89.3 58.1 71.3 41.9 28.7 Slovenia 93.8 91.5 50.7 53.5 97.7 92.7 19.1 27.2 80.9 72.8 Tajikistan 86.4 86.8 84.3 88.3 89.6 83.0 88.0 92.7 12.0 7.3 Turkey 72.9 69.7 38.6 35.3 62.9 62.2 24.4 31.0 54.8 40.1 Ukraine 79.9 76.0 38.9 38.2 86.2 81.4 61.2 65.8 36.6 32.3 Uzbekistan 91.5 90.2 89.3 91.2 85.8 77.9 82.7 80.2 10.3 12.8 Total 88.1 84.5 56.1 54.4 83.8 73.9 39.2 47.8 53.6 42.0 Note: Sample of primary respondents, weighted. Agreement with the norm takes a value of 1 if the respondent answered the question as “Agree” or “Strongly agree” and it takes a value of 0 if the respondent answered as "Strongly disagree", "Disagree", "Neither disagree nor agree" or “Don’t know”. F stands for female respondents and M stands for male respondents. Annex 4. Table 4 Agreement with norms by individual and household characteristics A woman It is important It is Co-habiting It is better for should do most that my important partners should everyone of the daughter that my son be married involved if the household achieves achieves man earns the chores even if university university money and the the husband is education education woman takes not working care of the home and children. F M F M F M F M F M Individual characteristics Age of the primary respondent 18-24 47.0 45.2 77.1 74.5 78.2 76.0 50.0 47.5 48.9 47.5 25-34 48.4 50.2 77.7 73.8 78.8 76.5 53.7 51.1 49.3 53.6 35-44 46.0 48.9 78.4 75.4 78.0 77.3 50.9 55.0 47.8 52.0 45-54 45.9 46.0 79.5 78.5 78.5 78.4 54.0 50.7 49.7 50.6 55-64 51.1 48.2 71.5 73.4 72.5 73.0 57.2 56.5 51.2 49.6 65+ 48.0 42.6 73.3 64.6 74.4 70.5 61.3 61.3 53.7 51.1 Highest education completed Less than (upper) secondary education 36.9 37.2 66.9 62.5 65.7 65.9 52.2 52.8 47.9 46.6 (Upper) secondary education 43.0 44.0 76.7 73.9 78.4 75.7 52.8 51.8 46.7 47.7 Higher education 57.2 53.7 80.8 78.1 81.2 79.4 58.3 55.7 54.3 55.3 Marital status Single (nevermarried) 38.6 38.5 76.3 72.2 77.1 74.3 38.8 37.1 41.4 39.2 Married 46.0 50.2 75.5 73.5 76.6 75.4 58.7 60.9 50.2 54.9 Widowed 56.7 48.8 77.2 75.0 76.5 76.9 64.0 56.5 57.2 64.6 Divorced 59.2 47.7 76.2 77.6 74.5 80.5 50.8 42.8 56.2 48.0 Separated 52.6 36.4 90.3 78.6 87.6 77.7 48.0 24.7 49.2 31.3 Employment status Not employed in the past 12 months 48.7 46.3 74.6 69.8 75.7 71.7 57.1 51.7 53.0 46.4 Employed in the past 12 months 46.9 48.5 78.6 77.3 78.3 78.2 50.6 52.7 46.8 52.7 Household characteristics 52 Gender of the household head Female 54.4 56.6 75.4 78.7 75.5 79.4 54.7 55.7 53.9 55.5 Male 42.0 45.3 76.7 72.7 77.5 74.8 55.0 53.4 47.1 50.1 Household size (effective) 1 46.5 36.2 76.3 70.8 75.9 72.1 46.6 31.8 43.2 39.0 2 44.5 43.5 71.3 70.2 71.4 72.7 52.9 53.6 48.9 50.0 3 43.0 48.3 77.1 75.5 76.8 74.4 49.7 55.6 48.8 50.3 4 49.8 46.9 80.8 75.9 83.6 81.2 63.8 55.1 52.2 50.8 5 or more 65.5 68.4 83.0 79.8 84.1 82.5 72.8 74.4 64.7 69.8 Households with children aged 0-17 Without 44.1 43.0 74.1 72.1 74.0 74.5 52.2 50.8 47.4 47.0 With 56.1 56.2 81.0 76.9 82.6 77.7 61.5 60.5 56.8 59.9 Households with children aged 0-6 Without 45.7 45.2 75.3 72.3 75.5 74.4 53.3 51.7 48.3 48.8 With 60.8 58.3 82.0 81.0 84.3 82.1 66.0 65.9 62.7 63.9 Households with children aged 7-17 Without 45.8 44.6 74.7 72.9 75.0 75.2 53.3 52.3 48.9 48.7 With 54.9 56.3 81.5 75.9 82.5 76.5 61.1 59.3 55.1 59.4 Households with elderly aged 65+ Without 46.6 46.8 76.9 75.0 77.3 76.2 51.6 51.3 48.5 50.4 With 50.4 48.2 74.5 68.8 75.0 73.1 63.4 62.2 54.4 52.9 Households with elderly aged 75+ Without 46.8 46.6 75.9 74.2 76.3 75.8 52.9 52.7 49.2 50.8 With 56.3 52.8 78.8 66.5 79.8 72.5 74.3 65.2 59.7 52.8 Urbanity status Urban 45.9 44.7 76.7 73.4 77.2 75.3 52.5 51.5 49.0 49.0 Rural 52.5 53.3 74.7 74.0 75.1 76.1 61.7 59.7 53.5 56.2 Country Albania 39.9 43.2 94.6 95.4 95.5 95.6 74.5 72.6 51.1 55.6 Armenia 74.9 76.2 93.8 91.9 95.3 93.2 88.0 87.1 80.6 91.7 Azerbaijan 77.4 88.6 94.9 93.5 94.4 95.2 90.8 89.2 90.4 92.7 Belarus 88.4 80.7 71.1 72.9 70.3 72.8 71.4 66.3 68.7 68.7 Bosnia and Herz. 30.4 41.9 89.6 86.1 89.2 86.0 80.3 81.9 45.8 56.0 Bulgaria 32.4 46.6 88.7 88.6 88.2 87.7 59.6 61.0 61.4 71.2 Croatia 22.1 31.9 81.7 78.3 80.8 77.9 62.0 59.1 38.9 46.8 Cyprus 25.2 25.0 95.3 91.6 95.4 91.4 48.6 50.6 47.5 50.7 Czech Rep. 18.1 27.4 48.9 48.5 50.4 49.1 54.3 53.7 51.6 61.1 Estonia 24.8 26.6 73.7 70.9 74.3 70.2 56.2 47.9 44.8 49.6 FYR Macedonia 34.4 40.1 92.6 91.4 93.7 90.5 83.9 87.9 60.0 60.3 Georgia 44.0 45.6 88.0 85.7 89.0 87.2 76.2 74.5 59.0 68.3 Germany 2.3 1.9 63.8 62.3 63.5 68.5 9.8 11.0 4.4 10.0 Greece 33.1 43.0 92.1 89.0 91.6 89.0 55.2 59.2 53.7 66.8 Hungary 67.8 65.6 66.3 62.8 62.7 62.7 53.1 50.0 65.9 71.5 Italy 19.9 26.2 82.2 79.7 82.8 80.3 43.7 44.4 36.8 38.9 Kazakhstan 85.0 86.5 87.6 87.1 89.1 88.8 73.6 73.3 64.3 71.3 Kosovo 46.0 48.7 96.8 95.7 97.5 96.9 81.2 80.4 53.9 61.4 Kyrgyz Rep. 82.3 86.4 91.3 94.1 90.5 93.3 73.7 76.3 86.0 89.0 Latvia 33.1 32.8 94.6 93.0 94.1 92.2 62.9 57.1 54.1 59.4 Lithuania 19.8 23.2 85.3 82.5 85.5 83.8 75.9 66.9 47.2 54.6 Moldova 42.9 44.5 87.6 84.4 87.7 83.6 73.9 69.2 64.2 66.2 Mongolia 36.5 35.0 94.0 95.0 93.4 94.3 89.2 85.8 44.9 46.4 Montenegro 33.4 47.3 95.1 95.8 96.1 95.6 75.1 80.3 51.7 68.0 Poland 4.0 6.6 35.2 33.3 37.9 33.8 11.9 14.9 6.4 8.8 Romania 26.8 31.9 84.8 85.5 85.9 86.2 74.6 75.0 54.8 51.6 Russia 92.3 87.6 86.5 82.3 86.3 83.2 75.5 70.9 79.8 78.0 Serbia 41.7 48.8 82.6 82.3 81.1 82.7 73.3 76.4 57.0 64.4 Slovak Rep. 25.6 36.9 68.0 68.2 69.3 71.0 73.2 68.0 63.7 66.8 Slovenia 55.7 50.5 61.0 63.7 57.1 62.5 34.7 35.8 27.2 34.8 Tajikistan 62.5 67.6 91.2 91.5 97.4 98.8 93.3 93.6 79.8 84.9 Turkey 23.2 29.2 64.3 61.1 64.7 64.5 38.9 45.3 26.6 28.7 Ukraine 63.2 67.6 81.1 82.5 80.9 81.7 72.4 75.0 73.0 76.0 Uzbekistan 95.1 90.9 93.1 90.2 97.3 95.9 95.7 96.4 77.0 84.7 Total 47.7 47.1 76.2 73.6 76.7 75.5 55.0 53.8 50.3 51.0 Note: Sample of primary respondents, weighted. Agreement with the norm takes a value of 1 if the respondent answered the question as “Agree” or “Strongly agree” and it takes a value of 0 if the respondent answered as "Strongly disagree", "Disagree", "Neither disagree nor agree" or “Don’t know”. F stands for female respondents and M stands for male respondents. 53 Annex 4. Table 5 Having a say in household decisions My opinions are taken Managing day-to-day Making large The way the children into account in spending and paying household purchases are raised decisions made by the bills (e.g. cars, major household appliances) F M F M F M F M Individual characteristics Age of the primary respondent 18-24 75.6 76.0 53.7 56.4 52.1 61.1 56.2 57.4 25-34 80.3 81.3 74.9 80.8 78.1 85.8 85.2 79.4 35-44 80.0 84.0 84.7 80.9 83.3 89.7 90.0 75.9 45-54 81.7 80.7 85.8 82.7 86.4 90.7 91.1 83.1 55-64 84.4 83.6 84.5 79.5 86.1 89.2 91.6 81.7 65+ 79.7 81.2 77.4 79.7 78.0 87.2 80.5 79.5 Highest education completed Less than (upper) secondary education 72.5 75.3 73.9 74.4 75.6 81.8 80.0 71.8 (Upper) secondary education 80.2 80.3 76.1 79.1 75.9 83.3 82.8 76.9 Higher education 85.8 85.0 83.2 78.7 83.9 88.5 89.6 79.8 Marital status Single (neve rmarried) 76.4 76.4 60.2 63.6 62.9 68.2 59.0 61.1 Married 80.7 82.3 81.1 81.2 81.8 89.5 89.9 80.2 Widowed 85.6 87.9 81.4 77.6 76.6 77.8 65.3 72.3 Divorced 83.1 86.0 86.8 81.2 85.9 83.1 88.6 65.3 Separated 94.9 99.0 94.7 76.8 93.4 96.4 96.6 93.4 Employment status Not employed in the past 12 months 76.9 73.2 69.7 72.6 70.3 72.7 79.8 70.7 Employed in the past 12 months 83.7 84.8 85.9 79.6 86.4 89.8 90.4 79.4 Household characteristics Gender of the household head Female 84.5 83.0 85.0 72.7 85.7 83.1 84.5 76.0 Male 79.2 81.2 76.3 79.2 76.9 86.0 85.0 77.7 Household size (effective) 1 2 78.6 79.3 85.5 82.1 85.0 90.3 90.5 81.9 3 79.5 81.7 79.0 76.2 80.4 84.3 89.0 78.1 4 84.4 81.0 71.9 70.6 74.2 79.2 76.9 72.9 5 or more 82.4 87.7 67.6 79.3 67.5 81.3 79.2 73.1 Households with children aged 0-17 Without 78.3 79.1 79.1 76.4 79.6 84.2 84.4 78.2 With 84.7 85.7 77.2 80.7 77.9 87.4 85.6 76.3 Households with children aged 0-6 Without 79.3 80.7 79.0 76.8 79.5 84.8 84.1 77.4 With 87.3 85.5 75.2 83.6 76.7 88.3 88.1 77.1 Households with children aged 7-17 Without 79.3 79.6 78.7 77.5 79.5 84.8 85.2 78.0 With 84.3 87.0 77.5 79.3 77.4 87.0 84.1 75.8 Households with elderly aged 65+ Without 80.0 81.2 78.8 78.1 79.1 86.0 85.8 77.4 54 With 82.1 82.5 77.5 77.7 79.0 83.5 81.9 76.8 Households with elderly aged 75+ Without 80.1 81.6 78.5 78.8 79.2 86.3 85.8 77.2 With 85.0 80.0 77.5 68.9 76.8 75.9 74.0 78.6 Urbanity status Urban 80.6 81.2 79.2 77.3 80.2 85.6 84.0 77.3 Rural 80.4 82.2 76.7 79.5 76.2 84.8 87.1 77.3 Country Albania 94.3 93.7 70.6 69.2 65.7 75.1 72.3 52.7 Armenia 96.7 97.4 67.2 63.1 64.3 73.9 81.8 61.5 Azerbaijan 92.2 94.8 50.0 82.3 49.8 78.8 88.5 70.2 Belarus 89.6 90.1 85.8 79.1 84.6 85.3 89.9 69.3 Bosnia and Herz. 93.8 91.0 80.7 82.1 83.3 87.6 84.9 83.6 Bulgaria 96.7 96.9 83.1 80.7 81.8 87.4 91.4 66.2 Croatia 91.2 90.6 80.3 80.6 82.0 88.1 86.9 82.3 Cyprus 98.4 95.1 74.0 73.3 76.8 82.6 88.5 74.8 Czech Rep. 90.2 91.0 80.9 77.7 82.5 83.1 89.8 79.3 Estonia 96.5 94.5 81.0 68.9 85.2 82.2 90.5 67.9 FYR Macedonia 93.6 93.3 77.7 79.9 76.0 83.8 84.8 77.5 Georgia 89.0 89.1 77.8 87.7 80.6 89.4 85.0 84.0 Germany 56.0 58.5 82.9 79.1 86.8 88.4 95.4 79.6 Greece 95.3 97.0 81.7 75.8 84.4 84.3 90.1 65.5 Hungary 95.0 91.6 87.4 64.7 87.3 84.1 92.7 74.2 Italy 95.7 93.8 70.6 72.1 76.7 86.7 82.1 73.0 Kazakhstan 87.5 89.5 81.0 76.6 80.0 84.6 84.6 82.2 Kosovo 89.1 89.6 58.9 79.9 64.4 80.9 81.9 71.5 Kyrgyz Rep. 90.4 93.8 74.7 79.4 73.4 77.6 79.4 80.8 Latvia 94.1 96.6 79.8 70.0 81.2 87.7 93.5 75.1 Lithuania 94.9 93.9 86.0 63.7 87.4 84.8 90.6 79.8 Moldova 88.4 87.5 89.1 87.5 90.3 88.3 90.0 84.4 Mongolia 94.5 94.8 77.9 55.5 74.9 83.9 83.8 75.0 Montenegro 95.0 95.2 83.2 79.2 80.9 80.0 86.8 76.9 Poland 43.1 44.3 80.7 78.9 80.9 87.2 93.4 80.8 Romania 92.9 93.2 93.8 88.2 91.9 92.5 95.5 88.9 Russia 95.9 95.9 87.5 76.4 85.6 88.1 86.0 83.8 Serbia 88.6 88.8 74.6 75.5 75.4 80.0 81.8 72.6 Slovak Rep. 93.2 94.2 85.0 69.9 85.8 76.9 85.4 75.8 Slovenia 95.0 89.8 87.2 74.7 82.2 81.9 90.8 80.6 Tajikistan 96.2 98.3 84.8 88.1 83.3 87.4 79.1 72.5 Turkey 51.9 54.5 69.9 83.2 69.8 80.7 72.7 72.0 Ukraine 87.9 87.5 83.1 77.7 84.3 82.7 85.3 77.7 Uzbekistan 91.6 97.5 59.5 81.7 55.0 82.3 83.0 71.2 Total 80.5 81.5 78.4 78.0 79.0 85.4 84.9 77.3 Note: The sample is restricted to sample of primary respondents living in households with at least two adults from the opposite gender, weighted. Sample size is 36,459 individuals (Whole sample includes 51,206 observations).For the statement “My opinions are taken account in decisions made by the household” agreement with the statement takes a value of 1 if the respondent answered the question as “Agree” or “Strongly agree” and it takes a value of 0 if the respondent answered as "Strongly disagree", "Disagree", "Neither disagree nor agree" or “Don’t know”. Having a say in household decisions takes a value of 1 if the respondent answered the question as “mostly me”, “shared equally between me and my partner” or “shared equally between me and someone else in the household”. F stands for female respondents and M stands for male respondents. 55 Annex 4. Table 6 Having a say in household decisions Social life and leisure activities Savings, investment and Looking after the children borrowing F M F M F M Individual characteristics Age of the primary respondent 18-24 73.6 79.4 50.3 56.7 57.7 55.6 25-34 86.5 88.4 78.6 86.6 83.2 70.9 35-44 89.1 88.6 81.6 87.3 88.4 63.9 45-54 90.5 89.9 84.9 89.8 90.9 77.2 55-64 91.8 90.3 83.4 88.6 90.3 76.1 65+ 85.5 85.6 78.2 88.6 84.4 74.3 Highest education completed Less than (upper) secondary education 84.0 84.1 74.5 81.9 80.4 64.5 (Upper) secondary education 86.2 87.0 75.8 84.4 83.2 71.4 Higher education 89.6 89.6 82.2 85.8 87.8 70.8 Marital status Single (never married) 79.9 83.5 61.5 65.7 61.5 56.0 Married 88.4 88.5 80.7 88.9 88.5 72.4 Widowed 82.8 79.8 77.8 78.6 73.1 73.5 Separated 94.4 93.6 95.5 94.4 92.8 61.0 Employment status Not employed in the past 12 months 81.5 83.0 67.8 72.0 79.8 67.4 Employed in the past 12 months 92.0 89.8 86.3 88.3 88.6 70.2 Household characteristics Gender of the household head Female 89.2 85.9 85.2 79.2 86.5 71.4 Male 86.4 88.0 75.8 85.8 83.8 69.7 Household size (effective) 1 2 91.1 92.2 85.1 90.7 89.7 78.2 3 89.5 86.5 78.3 82.6 85.7 69.3 4 84.2 81.1 73.1 79.3 80.1 64.1 5 or more 76.2 84.9 65.6 77.6 79.8 65.0 Households with children aged 0-17 Without 88.0 88.2 79.3 83.9 85.8 74.3 With 85.1 86.5 75.6 85.7 83.0 65.4 Households with children aged 0-6 Without 87.6 87.5 78.4 83.8 84.0 71.4 With 84.0 87.9 76.2 88.2 86.6 65.2 Households with children aged 7-17 Without 87.5 88.0 79.3 84.4 86.0 72.2 With 85.6 86.5 74.1 85.0 81.3 65.6 Households with elderly aged 65+ Without 87.4 87.9 77.9 84.2 84.5 69.6 With 86.1 86.7 78.4 85.7 84.7 71.4 Households with elderly aged 75+ Without 87.3 88.0 78.3 84.9 84.4 69.8 With 84.0 83.1 75.0 80.6 86.0 72.7 Urbanity status Urban 87.3 87.3 77.6 83.9 83.8 69.6 56 Rural 86.5 88.4 79.2 86.2 86.3 70.8 Country Albania 77.9 73.1 70.5 74.0 74.8 52.7 Armenia 82.1 81.9 78.2 71.3 80.0 48.1 Azerbaijan 66.9 83.0 44.9 84.3 85.4 66.6 Belarus 92.2 86.1 80.1 87.0 89.0 57.3 Bosnia and Herz. 89.9 91.2 82.9 87.7 85.9 82.4 Bulgaria 89.5 87.1 82.7 86.4 91.8 63.3 Croatia 89.0 91.7 83.8 87.1 87.6 74.3 Cyprus 94.3 88.4 81.5 84.7 88.4 69.1 Czech Rep. 92.5 90.1 84.9 86.2 89.3 75.0 Estonia 95.5 86.7 88.0 83.9 91.4 69.8 FYR Macedonia 84.7 86.7 79.6 83.9 85.5 76.1 Georgia 87.5 90.1 80.6 89.5 84.3 81.8 Germany 91.6 89.8 85.6 88.7 93.5 67.7 Greece 91.6 88.4 82.2 84.2 89.0 49.6 Hungary 93.4 88.8 88.5 87.8 90.2 72.8 Italy 86.6 88.8 68.0 86.2 80.5 60.4 Kazakhstan 85.3 87.4 80.3 84.4 88.3 75.5 Kosovo 81.7 87.1 66.2 80.5 81.7 72.8 Kyrgyz Rep. 75.3 76.5 75.3 79.6 78.9 78.7 Latvia 91.5 87.8 87.1 86.2 95.7 73.0 Lithuania 94.2 89.6 90.5 88.3 91.0 77.5 Moldova 89.9 90.9 85.1 90.5 88.6 80.8 Mongolia 85.5 86.4 83.6 83.3 86.6 69.8 Montenegro 90.9 89.1 84.1 81.6 86.1 74.2 Poland 95.1 91.8 79.4 87.0 94.2 70.6 Romania 95.5 92.2 93.0 89.8 95.4 88.5 Russia 94.7 86.7 88.6 84.7 86.8 72.1 Serbia 83.5 86.0 77.3 76.9 79.7 64.1 Slovak Rep. 91.2 91.2 85.7 82.4 84.7 70.5 Slovenia 95.3 90.8 86.9 86.8 92.1 78.6 Tajikistan 85.0 86.3 81.9 87.8 67.6 55.4 Turkey 77.9 84.5 68.5 78.3 73.7 69.7 Ukraine 87.8 87.8 82.7 83.0 84.0 77.3 Uzbekistan 64.5 85.3 54.1 83.6 83.3 64.4 Total 87.0 87.6 78.0 84.6 84.5 69.9 Note: The sample is restricted to sample of primary respondents living in households with at least two adults from the opposite gender, weighted. Sample size is 36,459 individuals (Whole sample includes 51,206 observations). Having a say in household decisions takes a value of 1 if the respondent answered the question as “mostly me”, “shared equally between me and my partner” or “shared equally between me and someone else in the household”. F stands for female respondents and M stands for male respondents. 57 Annex 4. Table 7 Empowerment indices Financial empowerment index Overall empowerment index Female Male Female Male Individual characteristics Age of the primary respondent 18-24 0.526 0.576 0.580 0.609 25-34 0.771 0.843 0.812 0.827 35-44 0.832 0.857 0.861 0.814 45-54 0.857 0.874 0.884 0.857 55-64 0.847 0.856 0.873 0.851 65+ 0.781 0.849 0.801 0.832 Highest education completed Less than (upper) secondary education 0.748 0.790 0.779 0.771 (Upper) secondary education 0.761 0.822 0.797 0.807 Higher education 0.831 0.840 0.859 0.827 Marital status Single (never married) 0.616 0.654 0.654 0.677 Married 0.812 0.863 0.848 0.839 Widowed 0.790 0.779 0.763 0.771 Divorced 0.862 0.829 0.882 0.820 Separated 0.947 0.892 0.945 0.886 Employment status Not employed in the past 12 months 0.694 0.722 0.747 0.732 Employed in the past 12 months 0.862 0.857 0.881 0.833 Household characteristics Gender of the household head Female 0.854 0.776 0.861 0.774 Male 0.764 0.836 0.803 0.818 Household size (effective) 1 2 0.853 0.875 0.877 0.871 3 0.793 0.807 0.824 0.790 4 0.731 0.761 0.761 0.735 5 or more 0.671 0.790 0.725 0.764 Households with children aged 0-17 Without 0.794 0.813 0.822 0.814 With 0.770 0.843 0.808 0.800 Households with children aged 0-6 Without 0.791 0.816 0.819 0.808 With 0.759 0.864 0.811 0.814 Households with children aged 7-17 Without 0.792 0.820 0.823 0.814 With 0.766 0.836 0.801 0.794 Households with elderly aged 65+ Without 0.786 0.825 0.820 0.808 With 0.784 0.820 0.809 0.812 Households with elderly aged 75+ Without 0.787 0.831 0.821 0.813 With 0.769 0.751 0.778 0.766 Urbanity status Urban 0.791 0.820 0.818 0.805 Rural 0.775 0.833 0.815 0.817 Country Albania 0.690 0.727 0.719 0.664 Armenia 0.698 0.692 0.749 0.680 Azerbaijan 0.483 0.816 0.636 0.773 Belarus 0.835 0.836 0.871 0.788 Bosnia and Herz. 0.823 0.856 0.843 0.858 Bulgaria 0.827 0.849 0.855 0.803 Croatia 0.819 0.850 0.846 0.844 Cyprus 0.773 0.799 0.830 0.796 Czech Rep. 0.828 0.817 0.865 0.819 Estonia 0.848 0.775 0.881 0.770 58 FYR Macedonia 0.777 0.824 0.813 0.816 Georgia 0.795 0.889 0.827 0.876 Germany 0.851 0.854 0.888 0.835 Greece 0.829 0.814 0.860 0.779 Hungary 0.878 0.783 0.892 0.794 Italy 0.718 0.816 0.764 0.794 Kazakhstan 0.806 0.818 0.832 0.816 Kosovo 0.632 0.804 0.721 0.788 Kyrgyz Rep. 0.745 0.789 0.761 0.787 Latvia 0.824 0.813 0.861 0.805 Lithuania 0.879 0.783 0.898 0.801 Moldova 0.885 0.888 0.891 0.870 Mongolia 0.788 0.741 0.820 0.756 Montenegro 0.828 0.803 0.850 0.804 Poland 0.803 0.842 0.835 0.832 Romania 0.929 0.900 0.942 0.898 Russia 0.875 0.824 0.884 0.818 Serbia 0.756 0.773 0.783 0.764 Slovak Rep. 0.852 0.762 0.862 0.778 Slovenia 0.855 0.808 0.881 0.822 Tajikistan 0.833 0.877 0.802 0.797 Turkey 0.693 0.803 0.720 0.783 Ukraine 0.834 0.811 0.848 0.814 Uzbekistan 0.564 0.824 0.665 0.779 Total 0.786 0.824 0.817 0.809 Note: The sample is restricted to sample of primary respondents living in households with at least two adults from the opposite gender, weighted. Sample size is 36,459 individuals (Whole sample includes 51,206 observations). See Annex 2 for the methodology for the construction of empowerment indices. 59 Annex 5 Regression Results Regression 1: How are men’s characteristics like education or employment associated with women’s employment? To answer this question we ran the probit regression below for the sample of working age (between 18 and 64 years old) women (primary respondents) living together with at least one adult (aged 18 or more) man (secondary respondents). P(Being employed ==1 | x) = Φ(µ1Number of children aged 0-6 in the household + µ2Number of children aged 7-17 in the household + µ3Number of elderly aged 65+ in the household + µ4Household size+ α1Age + α2Education + α3Owning a dwelling or land + β1Age of the man + β2Education level of the man + β3Man’s ownership of a dwelling or land + β4Man is employed + Ω1country dummy1 +…+ Ω34country dummy34 + ε) Annex 5. Table 1 Men’s characteristics and women’s employment – Regression results Dependent variable: Being Dependent variable: Being employed employed (in the past 12 months) (in the past week) VARIABLES 1 2 1 2 Household composition Number of children aged 0-6 in the household -0.055*** -0.060*** -0.052** -0.058*** (0.020) (0.020) (0.021) (0.021) Number of children aged 7-17 in the household 0.043** 0.040** 0.040** 0.036* (0.019) (0.019) (0.020) (0.020) Number of elderly aged 65+ in the household -0.010 -0.002 -0.005 0.004 (0.041) (0.042) (0.043) (0.045) Effective household size -0.033** -0.030** -0.027** -0.024* (0.013) (0.013) (0.013) (0.013) Primary respondent Highest education level: Upper secondary education 0.127*** 0.127*** 0.138*** 0.138*** (0.034) (0.034) (0.034) (0.035) Highest education level: Higher education 0.304*** 0.298*** 0.292*** 0.286*** (0.035) (0.035) (0.039) (0.038) Age 0.002 0.001 0.002 0.001 (0.001) (0.001) (0.001) (0.001) Owns dwelling or land 0.087*** 0.083*** (0.027) (0.028) Secondary respondent Highest education level: Upper secondary education 0.023 0.016 0.018 0.010 (0.035) (0.035) (0.036) (0.035) Highest education level: Higher education -0.045 -0.052 -0.043 -0.052 (0.039) (0.039) (0.040) (0.039) Age -0.004*** -0.004*** -0.003*** -0.003** (0.001) (0.001) (0.001) (0.001) Owns dwelling or land -0.001 0.014 (0.028) (0.028) Employed 0.082*** 0.085*** 0.093*** 0.095*** (0.026) (0.026) (0.027) (0.027) Country dummies Controlled for Controlled for Controlled for Controlled for Observations 14,398 14,398 14,398 14,398 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of primary respondent women aged 18 to 64 years old living together with at least one adult man – e.g. secondary respondents (aged 18 years old or more). Regressions also include the country dummies which are not reported here. Regression 2: How is entrepreneurship associated with education, agreement to norms, being empowered in the household and living in a specific country for men and women? To answer this question we ran the probit regression below for the sample of primary respondent women and men aged 18 years old or more separately. 60 P(Ever being an entrepreneur ==1 | x) = Φ(β0 + α1Education + α2Agreement to norms + α3Financial empowerment index + Ω1country dummy1 +…+ Ω34country dummy34 + ε) Annex 5. Table 2 Personal characteristics, country effects and entrepreneurship Dependent variable: Have ever been an entrepreneur (1) (2) (3) VARIABLES Women Men Overall Primary respondent is male 0.0422*** (0.00651) Highest education level: Higher education 0.0425*** 0.0395*** 0.0413*** (0.00830) (0.0119) (0.00719) Agrees with the term "Men make better political leaders than women do" 0.00599 0.0225* 0.0115 (0.00883) (0.0128) (0.00757) Agrees with the term "A woman should do most of the household chores even if the husband is not working" -0.0159 -0.00692 -0.00996 (0.0116) (0.0130) (0.00878) Agrees with the term "It is better for everyone involved if the man earns the money and the woman takes care of the home and children" -0.00398 0.0195* 0.00693 (0.00914) (0.0118) (0.00759) Financial empowerment index 0.00318 0.0405** 0.0219** (0.0122) (0.0168) (0.0104) country – Albania 0.0477* 0.0686** 0.0603*** (0.0252) (0.0333) (0.0210) country - Armenia -0.0162 0.0566 0.0190 (0.0140) (0.0380) (0.0197) country - Azerbaijan -0.0477*** -0.0462** -0.0419*** (0.00636) (0.0204) (0.0109) country - Belarus -0.0136 -0.0210 -0.0172 (0.0148) (0.0247) (0.0139) country - Bosnia and Herz. 0.0172 0.0165 0.0193 (0.0193) (0.0264) (0.0165) country - Bulgaria -0.00283 0.0400 0.0207 (0.0159) (0.0308) (0.0172) country - Croatia 0.0173 0.0377 0.0287* (0.0188) (0.0283) (0.0169) country - Cyprus 0.0312 0.0992*** 0.0651*** (0.0216) (0.0346) (0.0205) country - Czech Rep. 0.0713*** 0.0608* 0.0690*** (0.0271) (0.0325) (0.0214) country - Estonia 0.0316 0.132*** 0.0785*** (0.0224) (0.0410) (0.0232) country - FYR Macedonia 0.0164 0.0157 0.0176 (0.0200) (0.0267) (0.0167) country - Georgia 0.0383 0.0358 0.0401** (0.0235) (0.0310) (0.0195) country - Greece 0.0613** 0.159*** 0.111*** (0.0252) (0.0398) (0.0240) country - Hungary 0.0321 -0.0293 0.00338 (0.0228) (0.0210) (0.0154) country - Italy 0.0613** 0.0670** 0.0672*** (0.0261) (0.0319) (0.0209) country - Kazakhstan 0.0153 0.00662 0.0116 (0.0211) (0.0284) (0.0175) country - Kosovo -0.0272** 0.0113 -0.00506 (0.0111) (0.0256) (0.0139) country - Kyrgyz Rep. 0.0515* 0.0269 0.0397* (0.0286) (0.0311) (0.0210) 61 country - Latvia 0.0336 0.117*** 0.0714*** (0.0221) (0.0388) (0.0221) country - Lithuania 0.00497 0.0201 0.0123 (0.0172) (0.0279) (0.0161) country - Moldova 0.00559 -0.0288 -0.00986 (0.0201) (0.0203) (0.0144) country - Mongolia 0.161*** 0.114*** 0.142*** (0.0387) (0.0388) (0.0278) country - Montenegro 0.0237 -0.00325 0.0133 (0.0209) (0.0248) (0.0164) country - Poland -0.00463 0.0635* 0.0269 (0.0161) (0.0380) (0.0200) country - Romania -0.00967 -0.0316 -0.0194 (0.0149) (0.0204) (0.0125) country - Russia 0.0190 -0.0297 -0.00303 (0.0215) (0.0243) (0.0162) country - Serbia 0.0230 0.0152 0.0207 (0.0212) (0.0267) (0.0171) country - Slovak Rep. 0.0303 0.0710* 0.0516** (0.0227) (0.0364) (0.0215) country - Slovenia 0.0704** 0.0725** 0.0709*** (0.0294) (0.0331) (0.0221) country - Tajikistan 0.0492* -0.0208 0.0125 (0.0273) (0.0228) (0.0173) country - Turkey -0.0123 0.00614 -0.00223 (0.0154) (0.0270) (0.0153) country - Ukraine -0.0236* -0.0132 -0.0190 (0.0132) (0.0274) (0.0145) country - Uzbekistan 0.00977 0.0635* 0.0339* (0.0212) (0.0350) (0.0202) Observations 26,309 20,537 46,846 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of primary respondents aged 18 years old and more. Reference country is Germany in the regressions. 62 Regression 3: How is asset ownership associated with individual characteristics? To answer this question we ran the probit regression below for the sample of female primary respondent and male primary respondents separately. P(Asset ownership ==1 | x) = Φ(α1Education + α2Age + α3Marital status + µ1Number of children aged 0-6 in the household + µ2Number of children aged 7-17 in the household + µ3Number of elderly aged 65+ in the household + µ4Household size + Ω1country dummy1 +…+ Ω34country dummy34 + ε) Annex 5. Table 3 Women’s asset or bank account ownership, individual characteristics and country effects Dependent variables Bank account Dwelling Land Dwelling or Bank account ownership (sole or VARIABLES ownership ownership land ownership ownership (sole) joint) Highest education level: Upper secondary education 0.009 -0.018* -0.000 0.124*** 0.107*** (0.033) (0.011) (0.032) (0.028) (0.024) Highest education level: Higher education 0.138*** -0.018* 0.118*** 0.271*** 0.259*** (0.033) (0.010) (0.032) (0.030) (0.026) Age of the primary respondent 0.011*** 0.001*** 0.011*** -0.001 0.001 (0.001) (0.000) (0.001) (0.001) (0.001) Marital status: Married 0.114*** 0.020** 0.116*** -0.074** 0.035 (0.029) (0.010) (0.029) (0.031) (0.030) Marital status: Widowed 0.161*** 0.025 0.178*** 0.084* 0.016 (0.047) (0.021) (0.044) (0.044) (0.042) Marital status: Divorced 0.157*** -0.005 0.152*** 0.098** 0.082** (0.039) (0.013) (0.038) (0.039) (0.034) Marital status: Separated -0.037 -0.028** -0.043 0.062 0.075 (0.087) (0.014) (0.088) (0.062) (0.054) Number of children aged 0-6 in the household 0.039* -0.005 0.030 -0.010 0.006 (0.022) (0.006) (0.021) (0.021) (0.018) Number of children aged 7-17 in the household 0.052*** 0.000 0.048*** -0.012 0.000 (0.019) (0.005) (0.018) (0.018) (0.016) Number of elderly aged 65+ in the household -0.077*** -0.001 -0.062** -0.040* -0.021 (0.025) (0.007) (0.024) (0.023) (0.020) Effective household size -0.045*** 0.005 -0.038*** -0.010 -0.002 (0.012) (0.003) (0.012) (0.012) (0.010) country - Albania -0.099** -0.023** -0.070 -0.401*** -0.640*** (0.049) (0.010) (0.049) (0.033) (0.027) country - Armenia -0.047 -0.037*** -0.077 -0.520*** -0.700*** (0.053) (0.007) (0.052) (0.018) (0.013) country - Azerbaijan -0.260*** -0.046*** -0.264*** -0.569*** -0.716*** (0.048) (0.005) (0.048) (0.011) (0.011) country - Belarus 0.125*** -0.039*** 0.106** -0.285*** -0.582*** (0.048) (0.007) (0.046) (0.043) (0.041) country - Bosnia and Herz. -0.119** -0.000 -0.074 -0.171*** -0.531*** (0.048) (0.014) (0.048) (0.047) (0.044) country - Bulgaria 0.187*** 0.024 0.176*** -0.150*** -0.533*** (0.044) (0.019) (0.043) (0.048) (0.046) country - Croatia 0.050 0.006 0.066 0.194*** -0.202*** (0.048) (0.015) (0.046) (0.043) (0.066) country - Cyprus 0.202*** 0.042** 0.195*** -0.072 -0.273*** (0.042) (0.021) (0.040) (0.052) (0.070) country - Czech Rep. 0.203*** -0.035*** 0.168*** -0.012 -0.225*** (0.043) (0.008) (0.042) (0.048) (0.064) country - Estonia 0.142*** -0.020** 0.125*** 0.328*** 0.072 (0.046) (0.010) (0.045) (0.034) (0.065) country - FYR Macedonia -0.180*** -0.027*** -0.146*** -0.099** -0.505*** (0.047) (0.009) (0.049) (0.050) (0.047) country - Georgia -0.019 0.070** 0.093* -0.370*** -0.645*** (0.053) (0.028) (0.050) (0.038) (0.027) 63 country - Greece 0.036 0.032* 0.072 -0.352*** -0.206*** (0.049) (0.019) (0.046) (0.037) (0.068) country - Hungary 0.305*** -0.051*** 0.268*** -0.338*** -0.582*** (0.035) (0.005) (0.036) (0.039) (0.038) country - Italy 0.081 -0.053*** 0.046 -0.291*** -0.298*** (0.050) (0.007) (0.049) (0.045) (0.066) country - Kazakhstan 0.050 -0.024** 0.031 -0.419*** -0.668*** (0.049) (0.010) (0.048) (0.032) (0.025) country - Kosovo -0.236*** -0.023** -0.166*** -0.350*** -0.640*** (0.047) (0.011) (0.053) (0.042) (0.027) country - Kyrgyz Rep. -0.066 0.045* 0.056 -0.545*** -0.704*** (0.052) (0.024) (0.050) (0.014) (0.012) country - Latvia 0.004 -0.015 -0.007 0.136*** -0.270*** (0.050) (0.012) (0.049) (0.046) (0.069) country - Lithuania 0.164*** 0.030 0.153*** 0.170*** -0.235*** (0.045) (0.019) (0.044) (0.045) (0.070) country - Moldova 0.076 0.051** 0.138*** -0.551*** -0.707*** (0.049) (0.023) (0.045) (0.012) (0.011) country - Mongolia 0.063 -0.033*** 0.053 0.018 -0.347*** (0.048) (0.008) (0.047) (0.051) (0.065) country - Montenegro -0.157*** -0.028*** -0.159*** -0.141*** -0.482*** (0.047) (0.009) (0.047) (0.048) (0.051) country - Poland 0.166*** -0.046*** 0.131** -0.292*** -0.553*** (0.057) (0.007) (0.057) (0.049) (0.053) country - Romania 0.220*** 0.012 0.206*** -0.425*** -0.683*** (0.042) (0.016) (0.041) (0.031) (0.023) country - Russia 0.263*** -0.053*** 0.236*** -0.370*** -0.671*** (0.047) (0.011) (0.046) (0.045) (0.046) country - Serbia -0.118** -0.018 -0.103** -0.083* -0.466*** (0.049) (0.011) (0.049) (0.050) (0.052) country - Slovak Rep. 0.273*** -0.036*** 0.230*** -0.116** -0.304*** (0.038) (0.007) (0.038) (0.049) (0.064) country - Slovenia 0.148*** -0.016 0.122*** 0.368*** 0.085 (0.046) (0.011) (0.045) (0.025) (0.058) country - Tajikistan 0.064 0.078** 0.226*** -0.539*** -0.708*** (0.053) (0.031) (0.044) (0.016) (0.012) country - Turkey -0.145*** -0.061*** -0.180*** -0.329*** -0.662*** (0.053) (0.006) (0.052) (0.045) (0.035) country - Ukraine 0.267*** -0.012 0.269*** -0.400*** -0.673*** (0.042) (0.013) (0.039) (0.037) (0.032) country - Uzbekistan -0.129** -0.051*** -0.157*** -0.515*** -0.733*** (0.053) (0.006) (0.053) (0.026) (0.016) Observations 28,706 28,705 28,706 28,706 28,706 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of female primary respondents aged 18 years old and more. Reference country is Germany in the regressions. 64 Table 3 (continued) Dependent variables Having the Land Dwelling right to sell ownership ownership (at Dwelling Land Land other with a least one ownership (all ownership (at ownership VARIABLES dwelling document solely) jointly) least one solely) (all jointly) Highest education level: Upper secondary education -0.026 -0.013 -0.039 0.041* -0.001 -0.011** (0.082) (0.018) (0.028) (0.021) (0.009) (0.004) Highest education level: Higher education 0.016 -0.019 0.041 0.081*** -0.003 -0.009* (0.081) (0.022) (0.031) (0.023) (0.007) (0.005) Age of the primary respondent 0.000 0.001 0.007*** 0.003*** 0.001*** 0.000 (0.002) (0.001) (0.001) (0.001) (0.000) (0.000) Marital status: Married -0.031 0.065** -0.135*** 0.221*** 0.005 0.012** (0.073) (0.027) (0.023) (0.020) (0.006) (0.005) Marital status: Widowed 0.074 0.022 0.194*** -0.087*** 0.025 -0.004 (0.093) (0.022) (0.044) (0.029) (0.017) (0.008) Marital status: Divorced 0.053 0.030* 0.128*** 0.002 0.010 -0.015*** (0.090) (0.016) (0.036) (0.034) (0.010) (0.003) Marital status: Separated -0.276* 0.037*** 0.025 -0.104** -0.007 -0.016*** (0.161) (0.010) (0.052) (0.045) (0.010) (0.002) Number of children aged 0-6 in the household 0.103** -0.007 0.091*** -0.033** 0.006 -0.007** (0.050) (0.010) (0.020) (0.016) (0.004) (0.003) Number of children aged 7-17 in the household 0.052 -0.009 0.099*** -0.037*** 0.005 -0.004 (0.052) (0.008) (0.015) (0.013) (0.003) (0.002) Number of elderly aged 65+ in the household 0.037 -0.004 -0.101*** 0.014 -0.004 0.002 (0.051) (0.011) (0.019) (0.016) (0.004) (0.004) Effective household size -0.066** 0.011** -0.100*** 0.046*** -0.004* 0.006*** (0.031) (0.005) (0.011) (0.008) (0.002) (0.001) country - Albania -0.386** -0.384*** 0.066 -0.127*** 0.012 -0.012*** (0.153) (0.120) (0.059) (0.016) (0.017) (0.002) country - Armenia -0.366** -0.055 0.300*** -0.152*** -0.003 -0.014*** (0.147) (0.065) (0.069) (0.013) (0.013) (0.002) country - Azerbaijan -0.398 -0.407*** 0.280*** -0.194*** -0.024*** -0.015*** (0.246) (0.151) (0.072) (0.007) (0.005) (0.002) country - Belarus -0.221 -0.037 0.412*** -0.113*** -0.015* -0.012*** (0.164) (0.078) (0.061) (0.018) (0.008) (0.003) country - Bosnia and Herz. 0.058 -0.172** 0.269*** -0.168*** 0.019 -0.006* (0.124) (0.085) (0.066) (0.010) (0.019) (0.003) country - Bulgaria -0.415*** -0.231*** 0.289*** -0.007 0.065** -0.008** (0.128) (0.088) (0.065) (0.030) (0.032) (0.003) country - Croatia -0.235 -0.063 0.337*** -0.118*** 0.033 -0.007* (0.143) (0.064) (0.064) (0.018) (0.023) (0.003) country - Cyprus -0.237* -0.148* 0.540*** -0.120*** 0.147*** -0.014*** (0.138) (0.081) (0.054) (0.017) (0.045) (0.002) country - Czech Rep. -0.367*** 0.280*** 0.008 -0.010 -0.012*** (0.133) (0.062) (0.031) (0.010) (0.002) country - Estonia -0.315** -0.162 0.444*** -0.125*** 0.013 -0.012*** (0.132) (0.103) (0.058) (0.017) (0.017) (0.002) country - FYR Macedonia -0.206 -0.220* 0.186*** -0.167*** 0.006 -0.013*** (0.168) (0.128) (0.067) (0.010) (0.016) (0.002) country - Georgia -0.196 -0.340*** 0.494*** -0.175*** 0.147*** -0.008** (0.154) (0.096) (0.060) (0.009) (0.051) (0.003) country - Greece -0.154 -0.108 0.391*** -0.143*** 0.113*** -0.014*** (0.138) (0.070) (0.063) (0.014) (0.040) (0.002) country - Hungary -0.315** -0.043 0.274*** 0.101*** -0.023*** -0.016*** (0.146) (0.092) (0.062) (0.038) (0.005) (0.002) country - Italy -0.166 -0.167 0.248*** -0.066** -0.013 -0.020*** (0.147) (0.126) (0.064) (0.027) (0.010) (0.003) country - Kazakhstan -0.149 -0.045 0.421*** -0.154*** 0.032 -0.016*** (0.151) (0.065) (0.061) (0.013) (0.023) (0.002) country - Kosovo -0.082 -0.740*** 0.196*** -0.181*** -0.008 -0.011*** (0.162) (0.090) (0.074) (0.008) (0.012) (0.003) country - Kyrgyz Rep. -0.101 -0.040 0.441*** -0.179*** 0.154*** -0.014*** (0.167) (0.058) (0.067) (0.008) (0.051) (0.002) 65 country - Latvia -0.140 -0.025 0.433*** -0.178*** 0.033 -0.014*** (0.138) (0.049) (0.059) (0.008) (0.023) (0.002) country - Lithuania -0.262* -0.002 0.428*** -0.104*** 0.071** -0.008** (0.139) (0.035) (0.060) (0.019) (0.033) (0.003) country - Moldova -0.261* -0.043 0.430*** -0.141*** 0.143*** -0.014*** (0.148) (0.056) (0.063) (0.015) (0.048) (0.002) country - Mongolia -0.461*** -0.227** 0.223*** -0.076*** 0.010 -0.015*** (0.125) (0.116) (0.065) (0.023) (0.017) (0.002) country - Montenegro -0.154 -0.136 0.255*** -0.177*** 0.000 -0.012*** (0.159) (0.108) (0.063) (0.008) (0.013) (0.002) country - Poland -0.419*** -0.056 0.284*** -0.028 -0.016* -0.016*** (0.157) (0.089) (0.070) (0.033) (0.009) (0.002) country - Romania -0.530*** -0.184** 0.229*** 0.029 0.028 -0.003 (0.102) (0.080) (0.061) (0.033) (0.021) (0.004) country - Russia -0.074 -0.124 0.432*** -0.076*** -0.011 -0.024*** (0.118) (0.079) (0.057) (0.027) (0.012) (0.004) country - Serbia -0.107 -0.113 0.288*** -0.173*** 0.015 -0.011*** (0.149) (0.082) (0.065) (0.010) (0.018) (0.003) country - Slovak Rep. -0.352** -0.061 0.302*** 0.054 0.001 -0.015*** (0.147) (0.075) (0.064) (0.034) (0.013) (0.002) country - Slovenia -0.263* -0.039 0.328*** -0.057** 0.017 -0.011*** (0.141) (0.065) (0.065) (0.025) (0.018) (0.002) country - Tajikistan -0.232 -0.436*** 0.139** -0.124*** 0.008 0.019 (0.187) (0.104) (0.069) (0.019) (0.017) (0.013) country - Turkey -0.220 -0.100 0.101 -0.160*** -0.022*** -0.021*** (0.197) (0.123) (0.066) (0.017) (0.008) (0.003) country - Ukraine -0.335** -0.221** 0.401*** -0.016 0.020 -0.011*** (0.151) (0.089) (0.063) (0.030) (0.021) (0.003) country - Uzbekistan -0.074 0.473*** -0.207*** -0.019** -0.017*** (0.185) (0.065) (0.009) (0.008) (0.002) Observations 2,782 3,300 28,706 28,706 28,705 28,705 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of female primary respondents aged 18 years old and more. Reference country is Germany in the regressions. Annex 5. Table 4 Men’s asset or bank account ownership, individual characteri stics and country effects Dependent variables Dwelling or Bank account Bank account Dwelling Land land ownership ownership VARIABLES ownership ownership ownership (sole) (sole or joint) Highest education level: Upper secondary education -0.036 0.001 -0.033 0.131*** 0.098*** (0.026) (0.009) (0.025) (0.026) (0.021) Highest education level: Higher education 0.076*** -0.017 0.066** 0.198*** 0.189*** (0.026) (0.010) (0.026) (0.031) (0.026) Age of the primary respondent 0.013*** 0.002*** 0.013*** 0.002** 0.002*** (0.001) (0.000) (0.001) (0.001) (0.001) Marital status: Married 0.210*** 0.032*** 0.209*** -0.085*** 0.055** (0.028) (0.010) (0.028) (0.029) (0.028) Marital status: Widowed 0.028 0.047 0.026 -0.053 -0.063 (0.060) (0.051) (0.059) (0.056) (0.051) Marital status: Divorced -0.098* -0.006 -0.095* -0.081 -0.089* (0.053) (0.014) (0.053) (0.055) (0.054) Marital status: Separated -0.101 -0.001 -0.099 -0.016 0.029 (0.066) (0.026) (0.066) (0.067) (0.054) Number of children aged 0-6 in the household 0.039* -0.001 0.033 0.015 0.013 (0.022) (0.007) (0.022) (0.023) (0.020) Number of children aged 7-17 in the household 0.102*** 0.006 0.096*** -0.007 -0.001 (0.017) (0.006) (0.017) (0.020) (0.017) Number of elderly aged 65+ in the household -0.075*** -0.004 -0.061*** -0.101*** -0.062*** (0.021) (0.008) (0.021) (0.024) (0.023) Effective household size -0.054*** 0.001 -0.047*** -0.005 0.002 (0.010) (0.004) (0.010) (0.013) (0.011) country - Albania 0.174*** 0.137*** 0.187*** -0.279*** -0.559*** (0.032) (0.030) (0.029) (0.052) (0.063) country - Armenia 0.103** 0.114*** 0.134*** -0.465*** -0.708*** 66 (0.044) (0.032) (0.040) (0.040) (0.032) country - Azerbaijan 0.273*** -0.028* 0.265*** -0.608*** -0.764*** (0.024) (0.016) (0.022) (0.014) (0.014) country - Belarus 0.197*** -0.023** 0.189*** -0.143*** -0.458*** (0.033) (0.012) (0.032) (0.055) (0.081) country - Bosnia and Herz. 0.240*** 0.202*** 0.251*** 0.084* -0.253*** (0.025) (0.034) (0.022) (0.049) (0.082) country - Bulgaria 0.226*** 0.127*** 0.222*** -0.061 -0.411*** (0.027) (0.027) (0.026) (0.053) (0.079) country - Croatia 0.179*** 0.132*** 0.180*** 0.272*** 0.013 (0.030) (0.028) (0.028) (0.033) (0.069) country - Cyprus -0.096** 0.112*** -0.031 -0.009 -0.126 (0.046) (0.028) (0.044) (0.055) (0.086) country - Czech Rep. 0.149*** 0.014 0.148*** 0.035 -0.063 (0.032) (0.017) (0.031) (0.048) (0.069) country - Estonia 0.096** 0.053** 0.102*** 0.299*** 0.073 (0.038) (0.022) (0.036) (0.030) (0.063) country - FYR Macedonia 0.179*** 0.134*** 0.193*** 0.039 -0.327*** (0.032) (0.031) (0.029) (0.055) (0.086) country - Georgia 0.180*** 0.331*** 0.260*** -0.354*** -0.617*** (0.036) (0.045) (0.025) (0.051) (0.057) country - Greece 0.084** 0.255*** 0.121*** -0.232*** -0.018 (0.037) (0.035) (0.033) (0.051) (0.074) country - Hungary 0.209*** -0.048*** 0.194*** -0.267*** -0.483*** (0.029) (0.008) (0.028) (0.048) (0.069) country - Italy 0.103*** -0.003 0.101*** -0.177*** -0.101 (0.038) (0.016) (0.037) (0.053) (0.078) country - Kazakhstan 0.140*** 0.059** 0.136*** -0.353*** -0.638*** (0.037) (0.023) (0.035) (0.047) (0.053) country - Kosovo 0.197*** 0.230*** 0.223*** -0.078 -0.450*** (0.034) (0.046) (0.029) (0.067) (0.085) country - Kyrgyz Rep. 0.182*** 0.187*** 0.205*** -0.571*** -0.739*** (0.034) (0.038) (0.030) (0.020) (0.021) country - Latvia -0.058 0.063*** -0.033 0.185*** -0.134 (0.045) (0.022) (0.044) (0.043) (0.086) country - Lithuania 0.069* 0.145*** 0.110*** 0.212*** -0.087 (0.039) (0.029) (0.035) (0.040) (0.083) country - Moldova 0.198*** 0.198*** 0.236*** -0.576*** -0.749*** (0.031) (0.034) (0.025) (0.017) (0.016) country - Mongolia 0.220*** 0.045** 0.226*** 0.110** -0.213** (0.028) (0.022) (0.025) (0.048) (0.085) country - Montenegro 0.166*** 0.103*** 0.166*** -0.068 -0.354*** (0.031) (0.026) (0.030) (0.054) (0.082) country - Poland 0.003 0.011 0.006 -0.241*** -0.398*** (0.057) (0.019) (0.056) (0.058) (0.086) country - Romania 0.179*** 0.129*** 0.196*** -0.409*** -0.676*** (0.032) (0.027) (0.029) (0.042) (0.045) country - Russia 0.276*** 0.008 0.265*** -0.190*** -0.483*** (0.034) (0.021) (0.033) (0.057) (0.080) country - Serbia 0.149*** 0.157*** 0.169*** 0.089* -0.235*** (0.033) (0.030) (0.030) (0.050) (0.085) country - Slovak Rep. 0.222*** 0.011 0.214*** -0.016 -0.128 (0.028) (0.016) (0.027) (0.053) (0.078) country - Slovenia 0.137*** 0.076*** 0.143*** 0.353*** 0.165*** (0.034) (0.024) (0.032) (0.020) (0.040) country - Tajikistan 0.216*** 0.252*** 0.272*** -0.573*** -0.751*** (0.032) (0.047) (0.023) (0.021) (0.017) country - Turkey 0.078* -0.054*** 0.066 -0.145*** -0.494*** (0.041) (0.008) (0.040) (0.053) (0.072) country - Ukraine 0.262*** 0.057*** 0.266*** -0.226*** -0.543*** (0.030) (0.022) (0.028) (0.055) (0.075) country - Uzbekistan 0.157*** -0.031*** 0.141*** -0.472*** -0.733*** (0.039) (0.012) (0.039) (0.043) (0.033) Observations 22,500 22,497 22,500 22,499 22,499 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of male primary respondents aged 18 years old and more. Reference country is Germany in the regressions. 67 Table 4 (continued) Dependent variables Having the Dwelling Land right to sell Having the ownership (at Dwelling ownership (at Land other right to sell least one ownership (all least one ownership (all VARIABLES dwelling land solely) jointly) solely) jointly) Highest education level: Upper secondary education -0.057 0.075 -0.029 -0.008 0.007 -0.005 (0.077) (0.053) (0.024) (0.023) (0.007) (0.005) Highest education level: Higher education -0.042 -0.005 0.023 0.049** -0.007 -0.008 (0.073) (0.058) (0.027) (0.024) (0.008) (0.005) Age of the primary respondent 0.001 0.002 0.011*** 0.001* 0.001*** 0.000 (0.003) (0.002) (0.001) (0.001) (0.000) (0.000) Marital status: Married 0.004 -0.050 0.036 0.187*** 0.016** 0.016*** (0.083) (0.065) (0.027) (0.022) (0.008) (0.005) Marital status: Widowed 0.193* 0.108 0.090 0.015 0.039 0.003 (0.104) (0.129) (0.062) (0.072) (0.047) (0.016) Marital status: Divorced 0.050 0.153 0.028 -0.061 -0.001 -0.008 (0.132) (0.096) (0.046) (0.041) (0.012) (0.007) Marital status: Separated 0.100 0.258*** 0.056 -0.151*** 0.005 -0.018*** (0.129) (0.076) (0.063) (0.026) (0.021) (0.004) Number of children aged 0-6 in the household 0.026 0.032 0.138*** -0.074*** 0.005 -0.005 (0.067) (0.039) (0.021) (0.020) (0.006) (0.003) Number of children aged 7-17 in the household 0.006 -0.044 0.129*** -0.018 0.007 0.000 (0.052) (0.033) (0.017) (0.015) (0.005) (0.003) Number of elderly aged 65+ in the household -0.022 -0.036 -0.124*** 0.027 -0.013** 0.009*** (0.054) (0.041) (0.022) (0.018) (0.006) (0.003) Effective household size 0.003 -0.030 -0.106*** 0.044*** -0.005 0.005*** (0.031) (0.019) (0.011) (0.010) (0.003) (0.002) country - Albania 0.211** 0.133** 0.454*** -0.147*** 0.116*** 0.025* (0.087) (0.066) (0.038) (0.018) (0.027) (0.013) country - Armenia 0.132 0.287*** 0.426*** -0.166*** 0.191*** -0.008 (0.118) (0.045) (0.047) (0.017) (0.038) (0.006) country - Azerbaijan 0.317*** 0.303*** 0.640*** -0.217*** -0.010 -0.014*** (0.064) (0.062) (0.016) (0.008) (0.013) (0.005) country - Belarus 0.311*** 0.204** 0.414*** -0.112*** 0.007 -0.016*** (0.051) (0.089) (0.041) (0.024) (0.015) (0.003) country - Bosnia and Herz. 0.314*** 0.328*** 0.591*** -0.197*** 0.290*** -0.006 (0.045) (0.032) (0.022) (0.009) (0.038) (0.006) country - Bulgaria 0.000 0.121* 0.295*** -0.011 0.171*** 0.001 (0.121) (0.071) (0.047) (0.034) (0.031) (0.007) country - Croatia 0.173* 0.259*** 0.463*** -0.155*** 0.204*** -0.008 (0.092) (0.046) (0.035) (0.016) (0.034) (0.005) country - Cyprus 0.174** 0.246*** 0.116** -0.131*** 0.209*** -0.016*** (0.088) (0.049) (0.052) (0.020) (0.036) (0.003) country - Czech Rep. 0.136 0.199*** 0.214*** -0.019 0.055*** -0.013*** (0.099) (0.069) (0.046) (0.031) (0.021) (0.004) country - Estonia 0.140 0.152** 0.311*** -0.112*** 0.100*** -0.010* (0.097) (0.076) (0.046) (0.023) (0.028) (0.005) country - FYR Macedonia 0.307*** 0.341*** 0.557*** -0.193*** 0.241*** -0.014*** (0.051) (0.028) (0.027) (0.010) (0.038) (0.004) country - Georgia 0.271*** 0.175*** 0.541*** -0.183*** 0.417*** 0.007 (0.072) (0.063) (0.032) (0.013) (0.049) (0.010) country - Greece 0.290*** 0.313*** 0.377*** -0.155*** 0.345*** -0.006 (0.056) (0.039) (0.042) (0.017) (0.040) (0.005) country - Hungary 0.179* 0.144 0.108** 0.094** -0.029*** -0.017*** (0.100) (0.130) (0.047) (0.041) (0.007) (0.003) country - Italy 0.202** 0.184** 0.291*** -0.102*** 0.035* -0.016*** (0.098) (0.087) (0.048) (0.027) (0.019) (0.005) country - Kazakhstan 0.240*** 0.363*** 0.465*** -0.173*** 0.140*** -0.018*** (0.082) (0.030) (0.039) (0.015) (0.032) (0.003) country - Kosovo 0.272*** 0.272*** 0.580*** -0.196*** 0.254*** 0.016 (0.067) (0.046) (0.026) (0.010) (0.044) (0.014) country - Kyrgyz Rep. 0.179* 0.246*** 0.556*** -0.197*** 0.319*** -0.018*** 68 (0.109) (0.051) (0.029) (0.010) (0.047) (0.003) country - Latvia 0.225*** 0.302*** 0.355*** -0.197*** 0.141*** -0.018*** (0.075) (0.041) (0.044) (0.009) (0.031) (0.003) country - Lithuania 0.168* 0.137** 0.259*** -0.110*** 0.170*** 0.012 (0.089) (0.069) (0.047) (0.022) (0.033) (0.010) country - Moldova 0.130 0.213*** 0.451*** -0.135*** 0.291*** -0.009 (0.104) (0.056) (0.038) (0.019) (0.041) (0.006) country - Mongolia -0.046 0.179*** 0.439*** -0.109*** 0.102*** -0.012*** (0.129) (0.069) (0.038) (0.023) (0.029) (0.004) country - Montenegro 0.327*** 0.300*** 0.549*** -0.201*** 0.170*** -0.009* (0.039) (0.038) (0.027) (0.009) (0.031) (0.005) country - Poland 0.081 0.039 0.170*** -0.103*** 0.041* -0.011** (0.126) (0.112) (0.058) (0.028) (0.023) (0.005) country - Romania -0.032 0.134* 0.242*** -0.039 0.148*** 0.011 (0.134) (0.070) (0.048) (0.032) (0.030) (0.009) country - Russia 0.272** 0.202* 0.451*** -0.111*** 0.041* -0.016*** (0.110) (0.104) (0.046) (0.031) (0.024) (0.006) country - Serbia 0.233*** 0.310*** 0.522*** -0.194*** 0.229*** -0.006 (0.074) (0.037) (0.030) (0.010) (0.034) (0.006) country - Slovak Rep. 0.064 0.123 0.233*** 0.020 0.042** -0.010** (0.123) (0.079) (0.052) (0.037) (0.019) (0.005) country - Slovenia 0.207** 0.176*** 0.266*** -0.062** 0.096*** 0.004 (0.082) (0.061) (0.046) (0.028) (0.025) (0.008) country - Tajikistan 0.102 -0.105 0.428*** -0.125*** 0.178*** 0.054** (0.140) (0.095) (0.047) (0.025) (0.040) (0.022) country - Turkey 0.323*** 0.377*** 0.402*** -0.180*** -0.008 (0.056) (0.026) (0.044) (0.019) (0.012) country - Ukraine 0.135 -0.052 0.353*** -0.026 0.068*** 0.004 (0.132) (0.096) (0.049) (0.038) (0.023) (0.008) country - Uzbekistan 0.296*** 0.150 0.611*** -0.226*** 0.016 -0.020*** (0.072) (0.098) (0.026) (0.009) (0.017) (0.003) Observations 2,579 3,839 22,500 22,500 22,497 21,728 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of male primary respondents aged 18 years old and more. Reference country is Germany in the regressions. Regression 4: How is care need in the household associated with employment for men and women? To answer this question we ran the probit regression below for the samples of primary respondent women and men separately. Both samples are restricted to the population aged 25 to 40 years old. P(Being employed ==1 | x) = Φ(α1Age + α2Education + β1Child care need in the household + β2Elderly care need in the household + β3Disabled care need in the household + Ω1country dummy1 +…+ Ω34country dummy34 + ε) Annex 5. Table 5 Care needs in the household and employment outcomes Dependent variable: Being employed (in the past 12 months) 25-40 year olds 25-54 year olds VARIABLES Women Men Overall Women Men Overall Age 0.012*** 0.005*** 0.010*** 0.003* 0.002* 0.002*** (0.003) (0.002) (0.002) (0.001) (0.001) (0.001) Highest education level: Upper secondary education 0.077** 0.081*** 0.078*** 0.119*** 0.057*** 0.090*** (0.033) (0.020) (0.019) (0.032) (0.015) (0.018) Highest education level: Higher education 0.251*** 0.146*** 0.198*** 0.273*** 0.130*** 0.202*** (0.037) (0.027) (0.023) (0.038) (0.019) (0.022) At least one child aged 0-6 receives child care -0.174*** 0.035* -0.065*** -0.157*** 0.020 -0.058*** (0.030) (0.020) (0.018) (0.027) (0.021) (0.017) 69 At least one elderly aged 75+ receives care 0.030 -0.286 -0.168 0.007 -0.197* -0.112 (0.068) (0.192) (0.158) (0.056) (0.113) (0.077) At least one disabled receives care -0.069 -0.034 -0.047 -0.051 -0.074** -0.057** (0.046) (0.036) (0.030) (0.036) (0.035) (0.026) Country dummies Controlled for Controlled for Controlled for Controlled for Controlled for Controlled for Observations 8,373 6,976 15,349 14,843 12,693 27,536 Robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of primary respondent women and men aged 25 to 40 years old for the first three regressions and sample of primary respondent women and men aged 25 to 54 years old for the last three regressions. Regressions also include the country dummies which are not reported here. Regression 5: How is use of institutional child care associated with women’s employment? To answer this question we ran the probit regression below for the sample of primary respondent women aged 25 to 40 years old and living together with at least one child aged 0 to 6 years old. P(Being employed ==1 | x) = Φ(α1Age + α2Education + α3Use of institutional child care + α4 Presence of at least one other adult female (aged 18 or more) + α5 Presence of at least one adult male (aged 18 or more) + Ω1country dummy1 +…+ Ω34country dummy34 + ε) Annex 5. Table 6 Using institutional child care and women’s employment Dependent variable: Being employed (in the past 12 months) VARIABLES Ages 25-40 Ages 25-54 Age of the primary respondent 0.012*** 0.005 (0.005) (0.003) Highest education level: Upper secondary education 0.186*** 0.153*** (0.057) (0.056) Highest education level: Higher education 0.326*** 0.284*** (0.052) (0.053) Use of institutional child care 0.273*** 0.221*** (0.046) (0.046) Presence of at least one other adult female (aged 18 or more) 0.004 0.073 (0.052) (0.051) Presence of at least one adult male (aged 18 or more) 0.103 0.032 (0.066) (0.062) Country dummies Controlled for Controlled for Observations 2,930 3,447 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Probit regressions include the sample of primary respondent women (aged 25 to 40 years old for the first regression and aged 25-54 for the second regression) and living in a household with at least on child aged between 0 and 6 years old. Regressions also include the country dummies which are not reported here. 70 Regression 6: How is agreement with norms associated with observable characteristics of women such as education, age, owning an asset and employment? To answer this question we ran the probit regression below for the sample of primary respondent women for each norm. Employment variable is defined for the individuals aged between 18 and 64 years old, hence the sample drops down to women aged 18 to 64 years old as well. P(Agreement to the norm Z ==1 | x) = Φ(µ1Number of children aged 0-6 in the household + µ2Number of children aged 7-17 in the household + µ3Number of elderly aged 65+ in the household + µ4Household size α1Education + α2Age + α3Marital status + α4Owning a dwelling or land + α5 Being employed + Ω1country dummy1 +…+ Ω34country dummy34 + ε) Annex 5. Table 7 Women’s observable characteristics and agreement with norms Dependent variables: Agreeing to the norms Equal Equal Women are Men make A woman It is It is important Co-habiting It is better rights forrights for as competent better should do important that my son partners for everyone women aswomen as men to be political most of the that my achieves should be involved if citizens areas business leaders than household daughter university married (-) the man important citizens executives women do chores even if achieves education (+) earns the for myexist in (+) (-) the husband university money and country (+) my is not education the woman country working (-) (+) takes care of (+) the home and children VARIABLES (-) Number of children aged 0-6 in the household 0.004 0.018 0.004 0.030 -0.001 0.016 0.024 -0.016 -0.001 (0.010) (0.021) (0.011) (0.021) (0.019) (0.017) (0.018) (0.022) (0.023) Number of children aged 7- 17 in the household 0.001 0.018 0.018* -0.017 -0.016 0.009 0.010 -0.026 -0.014 (0.008) (0.018) (0.010) (0.017) (0.019) (0.015) (0.017) (0.019) (0.018) Number of elderly aged 65+ in the household -0.004 -0.000 -0.005 -0.001 0.010 -0.018 -0.038 0.057* -0.063* (0.016) (0.035) (0.017) (0.031) (0.029) (0.026) (0.032) (0.033) (0.038) Effective household size 0.009 -0.008 -0.010 -0.002 0.006 -0.001 0.001 0.017 0.014 (0.005) (0.012) (0.007) (0.012) (0.011) (0.009) (0.010) (0.013) (0.012) Highest education level: Upper secondary education 0.020 0.029 0.052*** -0.058** -0.089*** 0.062** 0.081*** -0.010 -0.103*** (0.014) (0.029) (0.017) (0.027) (0.026) (0.024) (0.023) (0.030) (0.027) Highest education level: Higher education 0.038*** 0.000 0.056*** -0.105*** -0.142*** 0.083*** 0.092*** -0.045 -0.175*** (0.015) (0.031) (0.020) (0.030) (0.029) (0.028) (0.027) (0.030) (0.029) Age 0.000 0.000 0.001 0.000 0.001 -0.000 0.000 0.001 0.001 (0.000) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) Marital status: Married -0.005 0.028 -0.055*** 0.039 0.061** -0.006 -0.002 0.193*** 0.037 (0.016) (0.030) (0.019) (0.030) (0.029) (0.025) (0.025) (0.033) (0.030) 71 Marital status: Widowed 0.033* -0.000 -0.040 0.019 0.082** -0.028 -0.025 0.132*** 0.040 (0.019) (0.049) (0.036) (0.048) (0.042) (0.048) (0.047) (0.047) (0.049) Marital status: Divorced -0.002 -0.020 -0.040 0.038 0.079* -0.002 -0.043 0.052 0.035 (0.024) (0.046) (0.036) (0.040) (0.043) (0.039) (0.048) (0.044) (0.039) Marital status: Separated -0.037 -0.022 -0.034 0.090 0.019 0.132*** 0.087** 0.094 0.029 (0.049) (0.070) (0.057) (0.074) (0.073) (0.024) (0.034) (0.062) (0.062) Owns dwelling or land 0.012 0.052** 0.022 0.001 -0.041* 0.008 -0.007 -0.007 -0.048** (0.012) (0.024) (0.016) (0.023) (0.022) (0.019) (0.019) (0.024) (0.022) Employed 0.016 0.033 0.015 0.017 -0.012 0.050*** 0.041** 0.000 -0.073*** (0.011) (0.022) (0.014) (0.022) (0.020) (0.019) (0.019) (0.022) (0.021) Controlled Controll Controlled Controlled Controlled Controlled Controlled Controlled Country dummies for ed for for for for for Controlled for for for Observations 21,991 21,977 21,505 21,370 21,670 20,194 20,161 21,274 21,438 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 Note: Reported here are the marginal effects. Regressions regarding agreement in norms are probit regressions. They include the sample of primary respondent women aged 18-64. Agreement with the norm takes a value of 1 if the respondent answered the question as “Agree” or “Strongly agree” and it takes a value of 0 if the respondent answered as "Strongly disagree", "Disagree", "Neither disagree nor agree" or “Don’t know”. Country dummies are also controlled for but not reported here. Regression 7: How is having a say in household decisions associated with observable characteristics of women and men that the women are living together with? To answer this question we ran the probit regression below for the sample of primary respondent women aged 18 to 64 years old living together with at least one adult man (e.g. a secondary respondent) for each decision (For empowerment indices regressions are linear regressions). P(Agreement to the norm Z ==1 | x) = Φ(µ1Number of children aged 0-6 in the household + µ2Number of children aged 7-17 in the household + µ3Number of elderly aged 65+ in the household + µ4Household size α1Education + α1Education + α2Age + α3Owning a dwelling or land + α4 Being employed + β1Age of the man + β2Education level of the man + β3Man’s ownership of a dwelling or land + β4Man is employed + Ω1country dummy1 +…+ Ω34country dummy34 + ε) 72 Annex 5. Table 8 Women’s having a say in household decisions Dependent variables: Having a say in household decisions, financial empowerment index, overall empowerment index Managing Making Savings, The way Social life Looking Financial Overall day-to-day large investment the and leisure after the empowerm empowerm spending household and children activities children ent ent and paying purchases borrowing are raised bills (e.g. cars, major VARIABLES appliances) Household composition Number of children aged 0- 6 in the household 0.069*** 0.079*** 0.083*** 0.067*** 0.044*** 0.058*** 0.091*** 0.094*** (0.015) (0.015) (0.014) (0.011) (0.009) (0.014) (0.013) (0.011) Number of children aged 7- 17 in the household 0.053*** 0.049*** 0.039*** 0.030*** 0.020** 0.009 0.059*** 0.052*** (0.013) (0.013) (0.011) (0.009) (0.009) (0.010) (0.011) (0.009) Number of elderly aged 65+ in the household 0.016 0.052* 0.055* 0.012 0.018 0.037 0.053** 0.044** (0.030) (0.029) (0.029) (0.026) (0.016) (0.031) (0.024) (0.020) Effective household size -0.047*** -0.040*** -0.052*** -0.034*** -0.026*** -0.025*** -0.055*** -0.051*** (0.009) (0.009) (0.009) (0.007) (0.005) (0.007) (0.008) (0.006) Primary respondent Highest education level: Upper secondary education 0.011 0.028 0.024 0.006 0.009 0.021 0.015 0.011 (0.023) (0.021) (0.024) (0.017) (0.016) (0.018) (0.019) (0.016) Highest education level: Higher education 0.058** 0.090*** 0.038 0.046** 0.031* 0.055** 0.052** 0.045*** (0.029) (0.024) (0.030) (0.019) (0.017) (0.023) (0.021) (0.017) Age 0.007*** 0.007*** 0.006*** 0.005*** 0.003*** 0.005*** 0.007*** 0.007*** (0.001) (0.001) (0.001) (0.001) (0.000) (0.001) (0.001) (0.001) Owns dwelling or land 0.053*** 0.044** 0.030* 0.028** -0.003 0.037** 0.039*** 0.030*** (0.019) (0.017) (0.018) (0.014) (0.011) (0.017) (0.014) (0.011) Employed 0.104*** 0.085*** 0.131*** 0.039*** 0.036*** 0.034* 0.110*** 0.082*** (0.020) (0.018) (0.019) (0.015) (0.013) (0.019) (0.017) (0.014) Secondary respondent Highest education level: Upper secondary education -0.024 -0.028 -0.041 -0.024 -0.003 -0.006 -0.033 -0.020 (0.028) (0.023) (0.027) (0.018) (0.016) (0.020) (0.021) (0.017) Highest education level: Higher education -0.019 -0.048* -0.050* -0.007 -0.020 -0.027 -0.033 -0.023 (0.033) (0.025) (0.030) (0.018) (0.017) (0.022) (0.022) (0.017) Age -0.001 -0.003*** -0.003*** -0.000 -0.001 -0.000 -0.003*** -0.002*** (0.001) (0.001) (0.001) (0.001) (0.000) (0.001) (0.001) (0.001) Owns dwelling or land -0.058*** -0.008 -0.006 -0.022* -0.010 -0.008 -0.025* -0.021* (0.019) (0.017) (0.018) (0.012) (0.011) (0.016) (0.014) (0.011) Employed -0.008 -0.052*** -0.008 -0.007 -0.007 0.017 -0.017 -0.006 (0.018) (0.017) (0.017) (0.013) (0.011) (0.019) (0.014) (0.012) Controlled Controlled Controlled Controlled Controlled Controlled Controlled Controlled Country dummies for for for for for for for for Observations 14,273 14,214 13,982 12,157 14,035 11,943 14,297 14,305 R-squared 0.219 0.227 Robust standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1 73 Note: Regressions regarding the decisions are probit regressions and reported here are the marginal effects.. They include the sample of primary respondent women aged 18-64 with a secondary respondent. (male) Having a say in household decisions takes a value of 1 if the respondent answered the question as «mostly me», «shared equally between me and my partner» or «shared equally between me and someone else in the household». Financial empowerment and overall empowerment indices takes a value between 0 and 1. For further details on the construction of these indices see Annex 2. Country dummies are also controlled for but not reported here. 74 References Allendorf, K. (2007). Do Women's Land Rights Promote Empowerment and Child Health in Nepal? World Development, 35(11), 1975-1988. Arpino, B., Esping-Andersen, G., & Pessin, L. (2015). How Do Changes in Gender Role Attitudes Towards Female Employment Influence Fertility? A Macro-Level Analysis. European Sociological Review, 31(3), 370-382. doi: 10.1093/esr/jcv002 Berlinski, S., & Galiani., S. (2007). The effect of a large expansion of pre-primary school facilities on preschool attendance and maternal employment. Labor Economics, 14(3), 665-680. Crespo, L., & Mira, P. (2013). Caregiving to Elderly Parents and Employment Status of European Mature Women. Review of Economics and Statistics, 96(4), 693-709. doi: 10.1162/REST_a_00426 Cuberes, D., & Teignier, M. (2012). Gender gaps in the labor market and aggregate productivity. Sheffield Economic Research Paper Series. Sheffield, UK: University of Sheffield. Cuberes, D., Munoz-Boudet, A.M. and Teignier, M. (2019). How Costly Are Labor Gender Gaps? Estimates by Age Group for the Balkans and Turkey. Eastern European Economics, 57(1), pp.86-101. Deere, C. D., & Twyman, J. (2012). Asset ownership and egalitarian decision making in dual-headed households in Ecuador. Review of Radical Political Economics 44(3), 313-320. Del Boca, D. (2002). The effect of child care and part time opportunities on participation and fertility decisions in Italy. Journal of Population Economics, 15(3), 549-573. Doss, C. (2006). The Effects of Intrahousehold Property Ownership on Expenditure Patterns in Ghana. Journal of African Economies, 15(1), 149-180. Fortin, N. M. (2005). Gender Role Attitudes and the Labor-market Outcomes of Women across OECD Countries. Oxford Review of Economic Policy, 21(3), 416-438. doi: 10.1093/oxrep/gri024 Global Entrepreneurship Monitor. (2017). GEM Global Report 2016/17. London: Global Entrepreneurship Research Association (GERA). Ibrahim, S., & Alkire, S. (2007). Agency and empowerment: A proposal for internationally comparable indicators. Oxford development studies, 35(4), 379-403. Kenny, C. and Patel, D., (2017). Gender laws, values, and outcomes: Evidence from the World Values Survey. Center for Global Development. Working Paper 452. Washington D.C. Klugman, J., Hanmer, L., Twigg, S., Hasan, T., McCleary-Sills, J., & Santamaria, J. (2014). Voice and Agency: Empowering Women and Girls for Shared Prosperity. Washington D.C.: International Bank for Reconstruction and Development / The World Bank. McKinsey Global Institute (MGI). (2015). The Power of Parity: How Advancing Women's Equality Can Add $12 Trillion to Global Growth. London, San Francisco and Shanghai: McKinsey and Company. Muñoz Boudet, A. M., Petesch, P., Turk, C., & Thumala, A. (2013). On Norms and Agency : Conversations about Gender Equality with Women and Men in 20 Countries. Washington D.C.: World Bank. Pitt, M. M., Khandker, S. R., & Cartwright, J. (2003). Does Micro-Credit Empower Women?: Evidence from Bangladesh Policy Research Working Paper;No. 2998. Washington D.C.: World Bank. Quisumbing, A. R., & Maluccio, J. A. (2000). Intrahousehold allocation and gender relations: New empirical evidence from four developing countries. Washington D.C.: International Food Policy Research Institute. World Bank. (2012). World Development Report 2012: Gender Equality and Development. Washington D.C.: World Bank. World Bank. (2013). World Development Report 2013: Jobs. Washington D.C.: World Bank. World Bank. (2014). Gender at Work : A Companion to the World Development Report on Jobs. Washington D.C.: World Bank. Yoong, J., Rabinovich, L., & Diepeveen, S. (2012). The impact of economic resource transfers to women versus men: a systematic review. London: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London. 75 76