Report No: ACS7935 Republic of India India: Women, Work and Employment 26 February 2014 SASDS SOUTH ASIA CONTENTS ACRONYMS AND ABBREVIATIONS iv SUMMARY vi 1. GENDER AND LABOR FORCE PARTICIPATION IN INDIA 1 1.1 Introduction 1 1.2 The ‘Economic Empowerment’ Framework 1 1.3 Women’s Labor Force Participation: India and Other Countries 2 1.4 The Indian Growth Story 3 1.5 India’s Labor Force in the ‘High Growth’ Period 4 1.6 Barriers to Women’s Work Participation 6 1.7 How ‘Gender and Economic Empowerment’ Unfolds 7 2. KEY CORRELATES OF LABOR OUTCOMES 8 2.1 Focus on Three Significant Correlates 8 2.2 Household Economic Status and Work 8 2.3 Household Wages and Women’s Labor Force Participation 9 2.4 Labor Force Participation and Social Disadvantage 10 2.5 Education and Work 12 2.6 Interactions between Education and Economic/Social Status 15 2.7 Returns to Education 17 2.8 The ‘Two India’s’ are Three for Women and Work 19 3. GENDER AND EMPLOYMENT BY SECTOR 20 3.1 The Labor Market and Gender 20 3.2 Changes in the Time of Economic Growth 22 3.3 High Risks, Low Returns for Women 24 3.4 Formal Employment 25 3.5 Some Areas of Growth in Women’s Employment 26 3.6 Lessons from these Experiences of Job Growth for Women 32 4. GENDER AND EMPLOYMENT BY STATE 34 4.1 Labor Outcomes in the States 34 4.2 Wage Rates and Ranges 40 i 5. INTRA-HOUSEHOLD AND OTHER FACTORS AFFECTING WOMEN’S LABOR OUTCOMES 43 5.1 When Women Do Not Work 43 5.2 The Social Underpinnings of ‘Not Working’ 44 5.3 Strengthening ‘Agency’ to Enhance Women’s Work 47 6. LAND, INFRASTRUCTURE AND WOMEN’S WORK 52 6.1 Access to Land and Productive Assets 52 6.2 Participation in Agricultural Work 53 6.3 Operationalizing Women’s Claims to Land 56 6.4 Infrastructure and Economic Empowerment 57 6.5 Inequality in Access to Infrastructure and Work Participation 60 7. POLICIES RELATED TO WOMEN’S EMPLOYMENT 63 7.1 Empowering Policies and Laws 63 7.2 Economic Policies and Related Programs 65 8. CONCLUSIONS AND NEXT STEPS 67 8.1 Future research 68 ANNEXURES (INCLUDING UPDATED TABLES & FIGURES) 70 LIST OF BACKGROUND PAPERS 83 REFERENCES 84 List of Tables 1.1 Number and Distribution of Work Force Participants (Usual Principal and 4 Subsidiary Status workers aged 15 and above), 1993-94 to 2009-10 1.2 Labor Force Participation Rates of Men and Women (UPSS workers aged 15 and 4 above), Rural and Urban areas, 1993-94 to 2009-10 2.1 Mean Male Household Wages and Female LFPRs, Rural areas, 2004-05 and 2009-10 11 2.2 LFP Rates for Women and the Ratios of Male to Female LFPRs by Social Group, 2009-10 11 2.3 Labor Force Participation among Rural and Urban Women by Education Levels 15 2.4 Summary Results of a Regression Analysis of Wage Rates 19 3.1 Gender Differences in Activity Status of Workers in Key Sectors 25 4.1R Classification of States by Gender Gaps (M-F) in Shares of Workers in Different 41 Occupational Groups in Rural Areas, 2009-10 4.1U Classification of States by Gender Gaps (M-F) in Shares of Workers in Different 42 Occupational Groups in Urban Areas, 2009-10 ii List of Figures 1.1 Female Labor Force Participation Rates in Selected Countries (15-65 years), 2010 2 2.1 Labor Force Participation Rates of Women and Men by MPCE Decile Groups, All-India 9 Rural and Urban, 1993-94 and 2009-10 2.2 Labor Force Participation Rates of Different Social Groups, 2009-10 12 2.3 Labor Force Participation among 15-64 Year-olds, 2009-10 13 2.4 Labor Force Participation of Men and Women by Education, Rural and Urban Areas, 13 2009-10 2.5 LFPR among Different Social Groups by Education, 2009-10 17 2.6 Wage Regression Coefficients and Returns to Education for Men and Women in 2009- 18 10 3.1 Share of Women in Each Industry Group, NSS 2009-10 21 3.2 Distribution of Female and Male Work Force, NSS 2009-10 21 4.1 Gender Gap in LFPRs by state, 2011-12 34 4.2 LFPRs for Rural, Urban and All Males, Females and Persons by MPCE Deciles 36 5.1 Male and Female Labor Force Participation Rates by Age Group, 2009-10 46 List of Annexures (Including Updated Tables & Figures) Figures 2.1 Labor Force Participation Rates of Women and Men by MPCE Decile Groups, All-India 70 Rural and Urban, 1993-94 to 2011-12 2.2 Labor Force Participation Rates of Different Social Groups, 2011-12 70 2.3 Labor Force Participation among 15-64 Year-olds, 2011-12 71 2.4 Labor Force Participation of Men and Women by Education, Rural and Urban Areas, 71 2011-12 2.5 LFPR among Different Social Groups by Education, 2011-12 72 3.1 Share of Women in Each Industry Group, NSS 2011-12 73 3.2 Distribution of Female and Male Work Force, NSS 2011-12 73 4.1 Gender Gap in LFPRs by State, 2011-12 74 4.2 LFPRs for Rural, Urban and All Males, Females and Persons by MPCE Deciles 74 Tables 1.1 Number and Distribution of Work Force Participants (Usual Principal and Subsidiary 77 Status Workers) Aged 15 and above, 1993-94 to 2011-12 (millions; percentages in brackets) 1.2 Labour Force Participation Rates of Men and Women (UPSS workers aged 15 and 77 above), Rural and Urban Areas, 1993-94 to 2011-12 (percent) 2.1 Table 2.1 Mean Male Household Wages and Female LFPRs, Rural Areas, 2004-05 and 77 2011-12 2.2 Table 2.2 LFP Rates for Women and the Ratios of Male to Female LFPRs by Social 78 Group, 2011-12 2.3 Table 2.3 Labor Force Participation among Rural and Urban Women by Education 78 Levels 2.4 Average Wage Rates of Women and Men by Level of Education, Rural and Urban areas 79 3.1 Gender Differences in Activity Status of Workers in Key Sectors 80 4.1R Distribution of Male and Female Workers in Different Occupational Groups in Rural 81 Areas of States, 2011-12 (percent) 4.1U Distribution of Male and Female Workers in Different Occupational Groups in Urban 82 Areas of States, 20011-12 (percent) iii ACRONYMS AND ABBREVIATIONS BCG Boston Consulting Group BPO Business Process Outsourcing CII Confederation of Indian Industrie CWW Casual Wage Workers FAO Food and Agricultural Organization GDP Gross Domestic Product GEWDR Gender Equality World Development Report GOI Government of India HBW Home-based Workers ICT Information and Communication Technology IFAD International Fund for Agricultural Development INR Indian Rupees IT Information Technology ITES Information Technology-Enabled Services LFG Labor Force Participation LFPR Labor Force Participation Rate MGNREGS Mahatma Gandhi National Rural Employment Guarantee Scheme MKSP Mahila Kisan Sashaktikaran Pariyojana MPCE Monthly Per capita Consumption Expenditure MSME Ministry of Small and Medium Enterprises/ Micro, Small and Medium Enterprises MVA Multivariate Analysis NABARD National Bank for Agriculture and Rural Development NCO National Classification of Occupations NGO Non-Government Organization NIC National Industrial Classification NMEW National Mission for Empowerment of Women NRLM National Rural Livelihoods Mission NSS National Sample Survey NSSO National Sample Survey Organization NPSD National Policy on Skills Development OAW Own Account Workers OBC Other Backward Classes pp Percentage Points RWW Regular Wage Workers SASDS South Asia Social Development Sector SASDO South Asia Sustainable Development Office SASSD South Asia Sustainable Development Sector SC Scheduled Castes SDI Software Development Industry SEW Self Employed Workers iv SHG Self Help Group ST Scheduled Tribes UFW Unpaid Family Workers UPSS Usual Principal and Subsidiary Status v SUMMARY Overview Since economic liberalization in the early 1990s, India has experienced high economic growth and made considerable progress in gender equality in areas such as primary education. However, it fared poorly on gender-parity in labor force participation (LFP). During the period between 1993-94 and 2011-12, female labor force participation rate (LFPR) remained consistently low as compared to male participation. More alarming is the fact that female participation rate declined steadily during the same period, particularly in rural areas. The low level along with declining trend in rural female LFP poses a serious threat of ‘missing gender’ in the labor force. Although economic growth added jobs for both men and women in India till 2005, Indian women lost jobs in the next seven years, while men continued to gain, thereby widening the gender gap. The actual figures in 2012 suggest that approximately 35 to 40 million women are ‘missing’ from the labor force, had female LFP grown at the same rate as it had between 1999 and 2005.1 This represents a troubling trend considering the potential of these women to contribute to the country’s productivity. Figure 1: Female labor force An international comparison suggests that gender gap in LFPRs participation rate (%) across South Asian Countries is large in India, followed by only some countries from the Middle East, Asia and North Africa where socio-religious 83.1 68.2 restrictions on women created larger gaps. In the regional 59.8 57.2 38 context, female LFPR in India also fares poorly as compared to 30.3 23 15.9 its South Asian neighbors viz. Nepal, Bangladesh, Bhutan, Sri Lanka and Maldives. According to World Development India Sri Lanka Nepal Bangladesh Afghanisthan Bhutan Pakistan Maldives Indicators, India ranks 6th among 8 countries from the South Asia region in terms of female LFP— just above Pakistan and Source: WDI, World Bank Afghanistan (Figure 1). To better understand the existing situation, this report investigates gender and female labor force dynamics by drawing mostly on data from five rounds of the National Sample Survey, India, between 1993-94 and 2011-12. Key findings from the study are grouped below in three sections. First section describes the dynamics of female LFP looking at its evolution in previous two decades. The next section presents the drivers of low level of female participation and its declining trend. The last section proposes possible areas of action. 1 Based on World Bank staff calculation on number and distribution of work Force participants aged 15 and above, 1993-94 to 2011-12 (millions). The growth rate between 1999-00 and 2004-05 has been used to project female LFP in 2011-12. The missing numbers are calculated by taking the difference between projected and actual number. vi 1 Dynamics of female LFP Female LFP has been both low and declining. The LFPR of women has consistently been less than half that of men in both rural and urban areas. Consecutive rounds of NSS data show that during the period between Fig. 2: Female LFP: Both low and declining 1999-00 and 2011-12, the male participation rate in India 80 was reasonably high and hovered around slightly more 60 than 70 percent. The female LFPR, on the other hand, 40 remained consistently low (at around 30 percent) as 20 compared to the male rate. The same source of data also 0 underscores that the female participation rate declined 1999-00 2004-05 2009-10 2011-12 steadily in India over time, with the decline being more All Males All Females Female LFPR pronounced in rural areas. The low level of participation and declining trend in participation rate among women poses a serious threat of the ‘missing gender’ in the labor force. The 68th Round of the NSS (2011-12) provides evidence that the gender gap has continued to increase. Over the last decade, the number of female workers along with female LFPR declined in rural areas. The number of rural women workers dropped by 18.7 million between 2004-05 and 2009-10 in absolute terms. By 2011-12, 1.3 million more females dropped out of the work force— accounting for a total decline of 20 million in female LFP between 2004-05 and 2011-12 in rural areas. Based on trends up to 2005, 160 million women should have been in the work force by 2012 but the actual number was only 126 millions in 2012, which suggests that around 35 million women are ‘missing’ from the labor force. Female LFPR in rural and urban areas exhibited diverse trends in the last decade. According to NSS data, the female LFPR exhibited a fluctuating trend in rural areas. Between 1999-00 and 2004-05, it increased by 4 percentage points after recovering from a slide between 1993-94 and 1999-00. It then declined again by 13.4 percentage points by 2011-12. The urban female LFPR, on the other hand, was roughly persistent around 20 percent throughout this period. Fig 3: Rural -Urban distribution of Female Labour Force Participation, 1993-94 to 2011-12 One of the reasons for stagnancy in urban LFPR is that job (%) opportunities for urban female workers are confined in 60 few sectors. Case studies of garment manufacturing, the software industry and private household services identify 40 that workspaces have been emerging slowly for women, in manufacturing, construction, trade & repair and 20 software development industries. Manufacturing, across 0 a number of low-income states, including U.P, M.P. and 1993-94 1999-00 2004-05 2009-10 2011-12 Orissa, have had relatively high percentages of women Rural Females Urban Females workers in the urban areas, with the exceptions of Bihar and Jharkhand, where women’s work is depressed. vii Construction has been a growing sector for women’s employment particularly in urban areas as well. Among Low Income States, the tribal states of Chhattisgarh, Jharkhand and Orissa have higher shares. Trade and Repair Services have a very low overall share of rural women workers (2.3 percent), but a higher percentage of urban women workers (10.0 percent). Overall, female workforce participation is ‘skewed’ in the primary sectors. In 2011-12, almost two- thirds (62 percent) of all women workers were in agriculture, whereas, only about 13 percent were in manufacturing, and Fig. 4: Sectoral composition of female labor force participation,2012 15 percent (or less) were in service sectors such as Agriculture & fishing construction, education and trade & repair services. 2% 6% Manufacturing 2% 5% Construction 4% These general trends suggest that women face barriers to 6% Trade and repair entry in male-dominated technology-related work. Industrial services Education 13% 62% modernization involving technology has meant that men may Other COP services displace women (e.g., in mechanized garment factories), Private households sending women to ‘lower-end’ jobs, particularly in the urban Others areas. On an average, women got only one out of eight jobs that were created in the urban areas— the place considered to be the locus of India’s recent GDP growth. As a result, women work mainly and at the bottom of the pyramid, often in informal jobs or home-based work. Formal opportunities are negligible, and can only be accessed by educated and middle–income women. Changes over time show an increasingly educated female work force in urban areas. The share of women in the urban labor force with tertiary, technical and professional education has increased over the past decades. Between 1999-00 and 2011-12, the share of women with graduate and higher degrees increased from 16.9 to 24.0 percent in the urban labor force. In Fig. 5: LFP among Rural and Urban rural areas, on the other hand, uneducated women had the Women by Education Level highest LFPR all throughout. However, it declined from 74 Illiterate Upto Primarymiddle percent in 1999-00 to 56 percent in 2011-12. During the same 80.0 Upto Higher Sec Graduate & above period, participation of women with primary & middle, and 60.0 secondary & higher secondary education increased by 7 and 10 40.0 percentage points respectively. 20.0 0.0 1999-00 2011-12 1999-00 2011-12 Gender gaps in participation are very large among socio- Rural Urban religious disadvantaged groups. The gender gap in LFPRs is almost 50 percentage points among the socio-religious disadvantaged groups (Figure 6). Among SC, ST, OBC, Muslims and others, the gap is lowest among the tribal groups, albeit having increased from 22.7 percentage points in 1993-94 to 33.9 percentage points in 2011-12.2 Muslim women, on the other hand, had the lowest LFPRs of all women (20.2 percent). While Muslim men had participation rates comparable to others, the lowest female participation rate 2 It is important to note that these socio-religious categories are not homogeneous as geographic and other cultural contexts also shape norms. Thus, LFPRs among ST women, for example, in the Northeast states, reflect the particular culture and history of women’s development in that region which are quite different from those of the Central Indian region. viii led to the highest gender gap amongst Muslims (61.1 percentage points in 2011-12). Women in the ‘Others’ category had very low LFPRs (23.6 percent) as well, and there were large gender gaps (52.1 percentage points) among these higher castes and other Fig. 6: Labor force participation rate across religions. Socio Religious Group, 2011-12 Male Female At the state level, three distinct patterns were observed 100.0 between households’ economic status and female LFPR: (i) 75.0 states where work participation was greatly depressed and 50.0 25.0 relatively flat irrespective of economic status (‘Bihar pattern’); 0.0 (ii) states where there was a sharp decline in female LFP from Muslim Others OBC SC ST Total the poorest to the wealthiest (‘U.P. pattern’); and (iii) states where there were high participation among the poorest and wealthiest and a strong missing middle, especially for women (‘Himachal pattern’). There were also some unique situations, e.g., Gujarat which showed a sharp decline in the lower end and ‘flatness’ at the upper end of the ladder; and Andhra Pradesh with an unusual ‘soaring middle.’ At the state level, economic growth per se did not enhance women’s LFP. Women’s work participation increases with State Domestic Product (SDP) – up to a point. Female LFP rose linearly with per capita state domestic product (PCSDP) up to a level of about INR 13,000 (in 2009-10 at 1993-94 prices).3 However, the percentage changes were not related, demonstrating that growth per se did not enhance women’s LFP. This was confirmed by analysis of data for the period when LFP increased among women (1999-2000 to 2004-05), and held in both rural and urban areas in both periods. Combined Participation Rates (CPRs) in education and labor force show strong gender differentials at the state level. Rural men’s combined participation rates (CPRs) in labor market and education did not differ across states with low, medium or high per capita State Domestic Product (PCSDP). However, both LFPRs and enrolment rates among rural women are directly correlated with per capita state domestic product. The poorest state group clearly had very low opportunities for women’s work and also low educational enrolment of girls. In terms of age group, education and work participation together provide some insights on state level variations in education or jobs for their youth.4 Among 15-24 year-old men, CPRs in both rural and urban 3 Nine of 20 states included in this analysis were below a PCSDP of INR 13,000. 4 Enrolment and LFP rates are computed using the same population denominator and can therefore be added to get a ‘Combined Participation Rate’ (CPR). In India overall, the CPR of 15 -19 year-old males was around 96 percent and that of women was 66.3 percent in rural areas and 77.1 percent in urban areas, i.e., about a quarter of urban 15-19 year-old girls and one-third of rural girls of this age were neither studying nor working in 2009-10. The CPRs for rural and urban males in the 20- 24 year age group were around 98 percent. However, females in this age group had considerably lower CPRs, especially in urban areas. Thus, significant proportions of 20-24 year-old women were neither in education nor in the labor force. The reasons for this include the low value associated with female education; social barriers to education and employment; the primacy of marriage, child-bearing and involvement in family care; unavailability of appropriate employment opportunities; unacceptable employment conditions, and so on. In Chapter 4 we will examine the impact of some of these issues on young women’s activities. ix areas were higher than the work participation rates of 25-64 year-olds. For young women, on the other hand, CPRs varied much more widely. In seven states (including four low income states) over 50 percent of 15-24 year-old women were out of school and work. Over the 16-year period of economic growth, the CPRs of rural-urban and men-women increased in some states and declined in others, with no discernible relationship with the level of development, location, or other characteristics of states. However, one welcoming change is that younger women are slowly filling the gaps. 2 Drivers of low female labor force participation A segmented labor market, low mobility and low paying unsecured jobs appear to play important roles in holding women back. Low mobility restricted women to being grouped in certain industries and occupations, viz. basic agriculture, human resource management or client interactions, rather than to the technical, managerial or strategy-related jobs. Labor market segmentation is also influenced by gender biases, socio-cultural practices and geographies. Even in its most dynamic and ‘new’ aspects, the Indian labor market upholds the prevailing socio-cultural beliefs and behaviors that ‘homemaking’ is women’s primary role. This is manifested widely, e.g., in the employment of girls from an early age in the domestic service sector, the preference of both workers and employers for home-based work in the garment manufacturing sector, and the constraints to progression of women in the software development industry. Sectors such as domestic service and garment-making are considered ‘suitable for women.’ Although the latter is an important area in manufacturing and Indian exports, it is associated with women’s traditional work of sewing and tailoring, leading the sub-sector to function in some ways that are less than modern. If women’s work continues to be driven largely by households’ economic Fig.7: Female LFP declines with higher male necessity and not by any belief in the intrinsic value of work earnimgs in the household Female LFPR 2004-05 Female LFPR 2011-12 to women and of women’s work to families and society, these 75 Female LFPR(%) deep-seated practices will continue to hold sway. 50 25 A direct relationship was observed between stable earnings by men and low women labor force participation. Men with 0 Poorest Second Third Fourth Richest Quintile Quintile Quintile Quintile Quintile regular wage or salary had the highest proportion of spouses Household wage quilntiles who were not in the labor force. Based on 2004-05 and 2011- 12 NSS data, the pattern showed that higher the increases in male wages, greater the decline in the female LFPR. The study also finds that Labor force participation among both women and men is highest among poor households with the lowest consumption- expenditure levels and declines with higher expenditure. Spatial variation in male-female wage gap across states appears to be significant. Average wage rates for rural women are considerably lower than those for men in all work categories. For example, the ratio of female to male wage was 34 percent in Tamil Nadu, 54 percent in Maharashtra, 55 percent in Andhra, 58 percent in Kerala, and 83 percent in Uttar Pradesh. Moreover, the market wage for female workers was less than stipulated minimum wage in 11 states (Kanchi, 2010). Such wide gaps can be considered potential disincentives for female workers to join the labor force. x Strict social norms dictate gender roles and low female LFP. The main explanations for India’s very low female LFP are the strict norms that dictate gender roles and behaviors, including women’s seclusion and low investment in girls’ education and skills. Women did not benefit from the jobs generated by economic growth because of the lack of fit between the jobs and women’s endowments, due to low education of girls in the past. When women have to make a choice, the following social priorities often come in the way:  Life-cycle matters. Marriage and child-rearing are associated with lower LFP among women and higher LFP among men. Women in India still marry young and are expected to bear children in quick succession. Young brides and mothers are often secluded; women who have worked before marriage or motherhood may withdraw from work during the childrearing years and seek to re-enter it thereafter.  Care work is first, paid work secondary for women. The responsibility that women bear for care work is another prime reason for them not working. In the vast majority of Indian households, care work is invariably ‘women’s work’ while earning is ‘men’s business,’ giving rise to strong gender asymmetries. Social norms in India still by and large militate against men sharing in care work, so decisions for women not to work are related to life-cycle and household needs. The inadequacy of labor-saving utilities and support services such as child care ties women to homes, or results in the ‘double burden’ on women.  Families’ social status is considered enhanced by women not working in the Indian social context where men are expected to provide for their families. This is particularly true of those who have moved upward to acquire middle- or high-income status – hence, the ‘missing middle.’ Women with regular wage/salaried spouses have the highest likelihood of not being in the labor force.  Women often enter the labor force on adverse terms including low activity status, poor earnings, low-skilled and traditionally-inherited occupations. Out-sourcing in many areas of manufacturing has loosened the chains of inherited work, as has drawing women into factories and work-sheds; and these have enabled a wider acquisition of skills (and even dropping of caste identities).  In sum ‘women working’ depends to a great extent on the acceptance by communities and households. Women work if income is desperately needed by their household (as amongst the poorest), or if the net value added (financially and socially) by their work exceeds the costs of ‘care time foregone.’ 3 Direction: Areas for action There is need for deliberate interventions to enhance women’s employment. The overall experience of the 1993-2012 period suggests that we cannot assume that growth will improve women’s labor outcomes. India must take deliberate steps for women to participate in further growth. From policy to practical intervention, it requires action to reduce gender disparities in labor outcomes, increase women’s contribution to GDP and growth, and the economic empowerment of women overall. The actions range from improving education and skills among girls to increasing organization of women in xi work, access to services, and provide infrastructure to facilitate access to markets for regular paid work outside the home. Overall, loosening the social norms that constrain women’s work participation. Women are not a homogeneous group—so, no single solution will fit all. Future policy-making for women’s economic empowerment must distinguish between the three broad groups of women: (i) low-income women who need better paid/higher productivity work; (ii) the ‘missing middle’ who need work that is in flexible and in keeping with their skills, education and status; and, (iii) women in paid work, including young, mobile, relatively more-educated women who could benefit from emerging economic opportunities. They need to be supported through measures such as improved transport, enhanced safety in public spaces, and arrangements for child and elderly care. Within workplaces, issues such as sexual harassment, ‘invisible’ biases in recruitment and promotion, and flexible work and benefits need to be addressed. Connect jobs with women. The above differentiation entails two broad approaches for creating women’s work opportunities: ‘bringing work to women’ and ‘bringing women to work.’ One way to bringing work to women is by developing local economies that will increase women’s work within their reach. Similarly, ‘bringing women to work’ will place women in jobs, particularly, in the growing areas of the economy. Align skills with the market. India has a young population that will continue to enlarge the work force at least until 2026. It consists of over 200 million men and women between 15 and 24 years of age. The fact that almost three-quarters of women in this age-group and half of men are not yet working provides an opportunity to prepare them better for work whether they are in or out of secondary or tertiary education. Enhance skills for micro, small and medium enterprises (MSMEs). MSMEs are a rapidly growing sector in India, employing about 70 million people and contributing about 45 percent of industrial production and 40 percent of India’s exports. However, only about 7 percent of 30 million enterprises are women- led, pointing to the scope to increase women’s participation. To conclude, there are three important takeaways overall. (1) Efforts to increase women's economic participation and productivity need to take into account many different economic and social circumstances in which women are currently positioned. (2) 'Economic empowerment' requires more than assets, work or income. Education that brings with it new aspirations, and public policies that support changing gender roles play essential parts in helping men and women to break out of traditional roles and expectations. (3) There is a two-way relationship between the development of new ‘social & gender identities’ and ‘sustainable economic growth with shared prosperity’— each making the other a more realistic possibility. xii 1. GENDER AND LABOR FORCE PARTICIPATION IN INDIA 1.1 Introduction Despite two decades of high economic growth and considerable progress towards gender equality in some areas (such as primary education), India fares poorly on gender parity in economic participation. The female labor force participation rate (LFPR) in 2009-10 (31.6 percent) was not only substantially lower than men’s (80.4 percent) but compared unfavorably also with women in most countries. In 2011- 12, the LFPR for females had come down to 30.7 percent. Although economic growth added jobs for both men and women in India until 2004-05, in the five years that followed Indian women lost about 18 million jobs, leading to the widest gender gap that India has seen in the LFPR over the past 30 years.5 This report presents an overview of women’s labor force participation in the context of economic growth in India. Based on an analysis of available data and a review of the literature, it documents how labor outcomes and behaviors have changed by gender since India’s economic reforms began in 1992- 93, and examines whether economic growth has been a catalyst for addressing gender differences in labor outcomes. 1.2 The ‘Economic Empowerment’ Framework In this report labor outcomes are situated within the economic empowerment framework of the 2012 World Development Report on ‘Gender Equality and Development’ (World Bank, 2012). 'Economic empowerment’ is defined broadly in the GEWDR as the accumulation of endowments, the use of endowments to utilize economic opportunities and generate incomes, and the application of endowments to take action affecting individual and household well-being (or agency). Endowments encompass education, health, land and other assets such as financial resources that women and men accumulate during their lifetimes. Endowments enable individuals to utilize economic opportunities, make choices and take actions, i.e., to exercise agency. The WDR framework has been modified to show economic empowerment as the outcome of a balance between human capital endowments (e.g., education and health), social capital (capital built as a result of being members of community structures, informal institutions and/or social networks) and economic capital (land, assets, infrastructure, and so on). 6 In addition to examining opportunities in the labor market (via labor force participation, both a 5 . However, over the next two years only 1.9 million more females were added to the work force. Data from the 2011-12/68th Round of the NSS, that were released after the data analyses and drafting of this report were completed, show that the gender gap has continued to increase. The number of male workers and the male LFPR increased after 2009-10, but both these figures declined among women. Rural female labor force participation rate which was 49 % in 2004-05 and came down to 37.3% in 2009-10 has come further down to 35.6% in 2011-12. In terms of absolute numbers, there was a drop in the number of rural women workers by 18.7 million between 2004-04 and 2009-10. Over the next two years, there was a further drop of 1.3 million more females from the work force. So something clearly is going on in rural India. Patterns observed in 2009-10 continue in 2011-12 for the various age categories and socio- economic groups. 6 This framework is located within societal norms regarding gender roles and other basic values that inform household and community behavior. 1 cause and effect of economic empowerment), the report presents and discusses gender differentials in labor outcomes by age, education, social groups, location, economic levels and family characteristics. The labor market is discussed in terms of occupational sectors (industry groups) and activity categories (formal and informal work arrangements), and trends over the past two decades. The enabling environment that generates or supports the ‘three capitals’ – India’s policies and laws governing gender and work and some relevant programs – is also reviewed. 1.3 Women’s Labor Force Participation: India and Other Countries India’s female labor force participation rate is lower than that of women in most other countries (Figure 1.1). These include Nepal, Bangladesh and Sri Lanka in South Asia and most countries in Africa, Asia and Latin America – in fact all countries excluding some significant Islamic ones such as Egypt and Iran, and Pakistan and Afghanistan in the region. The gender gap in LFP rates is also large in India, with only a few countries in the Middle East, Asia and North Africa, where socio-religious restrictions on women prevail, having larger gaps. Inter-country comparisons also show that the LFPR of urban women in India is especially low. Figure 1.1 Female Labor Force Participation Rates in Selected Countries (15-65 years), 2010 90 80 70 60 50 Female LFPR 40 30 20 10 0 Argentina Norway Vietnam Mexico Lao PDR Korea, Dem. Rep. Indonesia Libya Portugal Ghana Bhutan Hong Kong SAR, China Malaysia Kuwait Sri Lanka China Maldives Nicaragua Bahrain Ethiopia Afghanistan Peru Australia Kenya Bangladesh Japan Colombia Iran, Islamic Rep. Guatemala Mauritius South Africa India Morocco Nepal Myanmar Thailand Brazil Kyrgyz Republic Chile Cambodia Mali Turkey Pakistan Saudi Arabia Zimbabwe Bulgaria French Polynesia Nigeria Egypt, Arab Rep. Lebanon Azerbaijan Czech Republic United Arab Emirates United States Countries Note: The blue bars include a range of high-income, middle-income and low-income countries around the world; the red bars are countries in South Asia, and India is shown in dark red. Source: World Development Indicators, World Bank, Washington D.C., 2010. 2 1.4 The Indian Growth Story Low female labor force participation has persisted in India despite two decades of high economic growth. This section describes how growth unfolded in India to set the stage for further analysis of male and female LFP during this period. In response to a macroeconomic financial crisis in 1991, India liberalized its economy substantially in five important areas: fiscal policy (particularly taxation), the capital account and foreign direct investment, international trade and tariffs, the financial sector (banks, loans, deposits, insurance, the stock market, and so on), and the role of the state (particularly in licensing, revenue generation and expenditure). GDP growth increased from 0.8 percent in 1991-92 to 5.3 percent in 1992-93, 6.2 percent in 1993-94, and over 7 percent in each of the next three years (Acharya, 2002). This was accompanied by real growth in manufacturing, exports, investment (especially industrial) and savings. The second half of the decade saw slower GDP growth (down to 4.8 percent), resulting in the 1990’s averaging out to slightly below 6 percent. Both industrial growth and agricultural growth rose and fell – the former fell from 7 to about 4.5 percent; the latter had risen to almost 5 percent in the first half of the decade, but declined to 2 percent in the second half. It is the services sector that helped maintain economic growth in the 1990’s, increasing from 6.7 to 7.8 percent, and contributing 60 to 70 percent of growth in that decade. This ‘boom and bust’ during the 1990’s was followed by a spurt of growth between 2003 and 2008 (the most rapid since Independence), averaging over 8 percent. Manufacturing grew by 8.8 percent per year, and services by 9 percent. Industrial growth was wide-ranging – from basic commodities and beverages to textiles and transport (Panagariya, 2008). The leading sectors were capital goods, communications and business services. During this period, exports of goods and services doubled in current monetary terms and as a percentage of GDP; gross investment, foreign investment, savings and remittances all rose substantially. Although the GDP growth rate fell to about 4 percent in 2007-08 it rallied in the three subsequent years, declined to a low 5.0 percent in 2012-13, and is expected to fall below this in 2013-14. Were the periods of growth accompanied by increases in labor force participation, a component of ‘total factor productivity’? There is a strong relationship between GDP growth and the growth of working-age populations across countries (Bloom and Williamson, 1998); and growth in India’s working-age population up to 2025 was expected to contribute (at least) 1.3 percent to the GDP growth rate (Rodrik and Subramanian, 2004). An implicit assumption, however, was that jobs would be created for (or by) the increasing work force. Unfortunately, the evidence of the past 20 years in India does not show that increases in GDP and labor demand go together. Despite GDP growth in the 1990’s, employment growth slowed down relative to the 1980’s in both rural and urban areas. This drop in (or lack of) employment elasticity of growth – commonly known as ‘jobless growth’ in India – held in the aggregate and across sectors, with the exception of a few tertiary sectors.7 7 . These were transport, storage and communications, finance, insurance, real estate, and business services. See Chapter 3. 3 1.5 India’s Labor Force in the ‘High Growth’ Period Table 1.1 shows the size and composition of the work force over the 16 years covered by the surveys. While 61 million persons were added to the work force between 1999-2000 and 2004-05, only 7.9 million were added between 2004-05 and 2009-10. And while the male work force increased by 25.9 million during this last crucial five-year period, distributed across rural and urban areas, the number of women in the work force actually declined by 18.2 million in this time, specifically in rural areas. This decline is significant cause for concern. Table 1.1 Number and Distribution of Work Force Participants (Usual Principal and Subsidiary Status workers aged 15 and above)8, 1993-94 to 2009-10 (millions; percentages in brackets) 1993-94 1999-00 2004-05 2009-10 Rural Males 181.2 (64.9) 192.4 (65.9) 214.8 (64.2) 227.9 (69.30) Rural Females 98.2 (35.1) 99.5 (34.1) 119.7 (35.8) 101.0 (30.7) Rural Persons 279.4 (77.7) 291.8 (75.5) 334.5 (74.8) 328.9 (72.2) Urban Males 63.6 (79.5) 76.3 (80.6) 89.5 (79.3) 102.6 (81.1) Urban Females 16.4 (20.5) 18.4 (19.4) 23.4 (20.7) 23.9 (18.9) Urban Persons 80.0 (22.3) 94.7 (24.5) 113.0 (25.2) 126.5 (27.8) All Males 244.8 (68.1) 268.6 (69.5) 304.4 (68.0) 330.5 (72.6) All Females 114.6 (31.9) 117.8 (30.5) 143.1 (32.0) 124.9 (27.4) All Persons 359.4 (100.0) 386.5 (100.0) 447.5 (100.0) 455.4 (100.00) Source: Raveendran, BG 2012:6 Table 1.2 Labor Force Participation Rates of Men and Women (UPSS workers aged 15 and above), Rural and Urban areas, 1993-94 to 2009-10 (percent) 1993-94 1999-00 2004-05 2009-10 Rural Males 87.6 85.4 85.9 82.5 Rural Females 48.6 45.4 49.0 37.3 Rural Persons 68.4 65.8 67.7 60.2 Urban Males 80.0 78.8 79.2 76.1 Urban Females 23.6 20.9 24.4 19.5 Urban Persons 53.3 51.0 53.5 48.6 All Males 85.4 83.4 83.8 80.4 All Females 41.9 38.1 41.8 31.6 All Persons 64.2 61.3 63.4 56.4 Source: Raveendran, 2012 BG:6 8 . ‘Usual Principal and Subsidiary Status’ (UPSS) combines ‘Usual Principal Status’ (UPS) and ‘Subsidiary Status’ (SS). UPS refers to a person’s major activity status during the 365 days preceding the survey (i.e., what s/he spent most of her/his time doing during the reference period. SS refers to the major activity status of a person for 30 days or more during the reference period. 4 Gender differences in labor force participation rates. Clearly evident in Table 1.2, the LFP rates of women have been consistently low and less than half those of men in both rural and urban areas. Further, all LFPRs were lowest in 2009-10, and the decline was entirely contributed by women. Over the 16-year period there was a reduction of 10.3 percentage points (pp) in the LFPR of women, and about 5.0 pp in the LFPR of men. The reduction in the rural female LFPR was 11.3 pp, and in urban areas it was 4.1 pp. In both areas the 16-year declines, as well as those between 2004-05 and 2009-10, were lower among men. The GEWDR pointed out that income growth by itself does not lead to greater gender equality on all fronts (World Bank, 2012). How growth has ‘played out’ is important, as well as how formal and informal institutions and markets have functioned and interacted with the realm of household decision- making. For example, the availability of income due to growth may have led to more schools and, with larger markets creating opportunities for women’s work participation, more girls may have been sent to school, a result of household decision-making. In the Indian context the lack of responsiveness of the gender gap in LFP during India’s high growth period may in part be due to the very low levels of secondary and tertiary education among women, and the poor health of women, both of which make for poor human capital. Clearly, economic endowments and ‘institutions’ in this sphere were also not able to contribute adequately to improve labor outcomes for women, particularly because of regressive societal norms. The larger context within which these analyses were conducted was that of ‘jobless growth’ (Ghosh, 2009a).9 Liberalization in India has not been associated with feminization of the labor force, the exception being agriculture and a few export-oriented sectors.10 Further, the share of formal employment fell from 9 percent in 1999-2000 to 7 percent in 2009-10. Hence, the challenge of having growth associated with the growth of ‘good jobs’ is even greater. Conversely, high levels of labor force participation are viewed as desirable for growth, and a small gender gap in LFP is not only good from an equality perspective but is also ‘smart economics’ (World Bank, 2006). The economic literature on gender inequality holds that reductions in gender gaps are associated with faster economic growth. This has been shown in several theoretical papers that emphasize that lower gender inequality is associated with faster human capital accumulation (e.g., Galor and Weil, 1996; Doepke and Tertilt, 2009; Lagerlof, 2003; Blackden et al., 2006) as well as empirically (Hill and King, 1995; Dollar and Gatti, 1999; Tzannatos, 1999; Klasen, 2002; Klasen and Lamanna, 2009). 9 While economic growth has been unprecedented, levels of inequality and social exclusion have also been marked. It is th increasingly accepted that growth without equity cannot lead to sustainable development. The 11 Five Year Plan recognized that growth had bypassed many groups such as a majority of the large workforce in the informal sector, women, Scheduled Castes, Scheduled Tribes, minorities and prioritized the massive effort at employment creation, of the right quality, and decent conditions of work for all sections of population. See GOI, 2007 for more details. There are inherent inequalities within society which are leading to increasing social exclusion. Feminization of poverty and the feminization of agriculture is on the increase. See also, Mehta et al., 2011: 134, Basu, 2008 ; Chaudhury, S. and M. Ravallion 2006 and Rao et al., 2007. 10. While women are increasingly the backbone of the agriculture sector, there has been an overall decline in the share of agriculture in employment from 70 percent in 1993-94 to 53 percent in 2009-10. 5 1.6 Barriers to Women’s Work Participation What have been the barriers to women's work participation, and what are the related policy implications? Some of the key barriers are identified in Figure 1.3 which organizes them into Individual, Household, Community, Market and Geographic groups and identifies some broad related phenomena (e.g., Ability, Need, Discrimination, and so on). There are supply-side barriers such as low competencies (in the Individual arena), the social norm of marriage and motherhood being ‘preferred’ for women over work (shown in the Community and Household arenas), and women’s care responsibilities being considered their primary role (a situation that plays out in the Household, Community and Job Market arenas). Male and female workforce decisions are inter-dependent, and women’s participation, in particular, is subject to being able to balance work and care. Another set of barriers is related to the inadequate demand for women workers. This includes the poor alignment of jobs with women’s skills, working conditions that discourage female participation, and low rates of formal employment generation (all in the Job Market arena), and constraints in the financial and land markets (Other Markets). This report discusses many (but not all) of these barriers with a view to identifying what could be done about them to progress toward more gender-equal work participation. Figure 1.2 Barriers to Entry into the Labor Market for Indian Women HOUSEHOLD •Care responsibilities - (children,elderly) •Social norms- ’permission’, COMMUNITY ‘safety’,’trust’ (Mobility*) •Socio-cultural norms •Violence - stigmatization GEOGRAPHIC •Formal or informal arrangements for •Food availability care work •Transport (affordability) INDIVIDUAL availability and costs •Infrastructure •Health •Availability of •Competencies training (Literacy, Skills) JOB MARKETS •Working conditions,hours OTHER MARKETS •Wages (low) •Land •Unaligned jobs and •Financial skills •Informality AVAILABILITY 6 1.7 How ‘Gender and Economic Empowerment’ Unfolds In subsequent chapters of this summary report we examine gender differences in India’s labor outcomes as determined by ‘consumption expenditure,’ social groups and education (Chapter 2), and also look at labor market variations by sectoral/occupational categories and work arrangements within the formal and informal sectors,11 (Chapter 3) and by geographies, including rural and urban, and different states (Chapter 4). Disparities in labor force participation and wages are strong across geographical regions, between occupational/economic sectors and activity categories, and among social, educational and economic groups, and gender differentials persist throughout. In addition to the individual correlates of labor force participation and aspects of the job market that have been analyzed quantitatively, we discuss some of the Household and Community barriers in Figure 1.3, notably, family/care responsibilities and women ’s agency/organization (in Chapter 5), and access to ‘Other Markets’ land and infrastructure (Chapter 6). It may be clarified that some factors are both correlates of LFP and outcomes. For example, consumption expenditure at the household level would be a correlate of LFP but LFP would also affect current consumption expenditure. Chapter 7 presents an overview of policies in India that are related to women’s employment and Chapter 8 provides the report’s conclusions. 11 . In the NSS, economic activities are classified into ‘industry groups’ and the surveys up to 2009-10 used the National Industrial Classification – 2004 (NIC-2004). Unless otherwise specified, this report uses NIC-2004, but it also refers to the industry groups as ‘economic activities,’ ‘occupations’/occupational groups or sectoral activities (sometimes grouped into the Primary, Secondary and Tertiary sectors, and at other times disaggregated into more specific economic activities). An alternative classification, the National Classification of Occupations – 2004 (NCO-2004) is used in some cases, with appropriate identification (and explanations, where necessary). 7 2. KEY CORRELATES OF LABOR OUTCOMES 2.1 Focus on Three Significant Correlates This chapter first presents the relationship between labor outcomes and household consumption expenditure levels, an indicator of ‘economic status’ or ‘standard of living’ and of relative poverty or wealth in a given geography. This is followed by an examination of LFP among different social groups. The chapter then discusses two complementary aspects of another significant correlation of labor outcomes, education: the role of education in determining LFP among women and men, including the nature of the work they are involved in, and India’s recent educational progress with a gender perspective. Present trends in employment and earning are the result of past trends in education, and present trends in education portend labor outcomes in the years to come. Thus, gender inequalities in education provide insights to the gender dynamics of economic empowerment; these can in turn help us understand what needs to be done in education to improve gender equality in the labor market. Of the three correlates discussed in this chapter, education is the most amenable to intervention to enhance labor force participation. 2.2 Household Economic Status and Work Labor force participation among both women and men is highest in the households with the lowest consumption-expenditure levels and declines with higher expenditure.12 While LFP rates for rural men in 2009-10 hovered around 85 percent in the first three monthly per capita consumption expenditure (MPCE) deciles, they were between 40 and 42 percent among rural women in these groups (Figure 2.1). Thereafter there was a gradual decline in LFPRs with MPCE deciles. Rural women from the poorest deciles had a 25 percent higher participation rate than those in the richest decile. Urban women had extremely low participation rates, far lower than those of urban men. The LFPR declined from about 25 percent for the poorest decile to about 15 percent for the eighth and ninth deciles, but increased slightly in the ‘richest’ decile to 19.5 percent. This participation pattern is referred to as a ‘U-shaped curve.’ Over time the poorest women fared the worst. The decline in LFPR among women between 1993-94 and 2009-10 was about 15 percentage points (pp) among rural women in the poorest decile (which had the highest LFPRs) (i.e., from 57.3 to 42.5 percent), but only 5 pp among rural men (from 89.5 to 84.8 percent). The same decline occurred in the top decile of men, but it was less pronounced among this decile of women compared with other women – only about 8 pp. The poorest women were the most significant ‘losers’ of work over the 16-year period of economic growth. 12 . The National Sample Survey includes data on household consumer expenditure, i.e., expenditure on food, tobacco and intoxicants, fuel and light, clothing and footwear, miscellaneous goods and services, and durable articles that are consumed by each household. These items may be acquired from the market, through exchanges of goods or services, gifts, charity or borrowing, home-grown stock or free collection; their monetary value is used to compute consumption expenditure. For a household, the monthly per capita consumption expenditure (MPCE) is its total consumer expenditure divided by its size and expressed on a monthly (30-day) basis. Households can be ordered on the basis of their MPCE and divided into ten numerically-equal groups, giving ‘MPCE deciles.’ It is important to note that the monetary range of a decile is variable between rural and urban areas and among states. Hence, the deciles are a measure of relative poverty or affluence in a given location and are not absolute levels. (The decile groups were computed for each of the major states and grouped smaller states and union territories to take into account state-wise variations in the levels and patterns of consumption.) 8 The fall in the five years between 2004-05 and 2009-10 was also most significant for the poorest groups. For example, there was a reduction of over 10 pp in the poorest decile of rural women. For the next four deciles the LFPRs were slightly lower than the first decile. From the sixth decile onwards, there was a further reduction. Rural women’s LFPRs in 2009-10 were 2 to 4 pp lower than in 2004-05 among all deciles, and the LFPRs of urban women were also significantly lower. Figure 2.1 Labor Force Participation Rates of Women and Men by MPCE Decile Groups, All-India Rural and Urban, 1993-94 and 2009-10 Rural Urban 100 100 90 90 80 80 70 70 LFPRs LFPRs 60 60 50 50 40 40 30 30 20 20 10 10 0 0 Decile Group Decile Group Rural Male 1993-94 Rural Male 2009-10 Urban Male 1993-94 Urban male 2009 -10 Rural Female 1993-94 Rural Female 2009-10 Urban Female 1993-94 Urban Female 2009-10 Source: World Bank staff The MPCE curve shows that women’s work participation is strongly embedded in poverty. Whether (and, if so, how) their work helps their households out of poverty or just to survive is an open question. This calls for greatly improving the returns to poor women’s labor, be it in wage work, farming or the myriad ‘own account’ production or trade activities in which poor Indian women are engaged. As the poorest women had the greatest decline in LFPRs during the period of high economic growth there is a great need to create alternative employment opportunities for them. 2.3 Household Wages and Women’s Labor Force Participation There is a direct relationship between increasing household earnings by men and decreasing women’s participation (Table 2.1) – the greater the increase in male wages, the greater the decline in the female LFPR (Neff et al., 2012). This effect has been ascribed to the withdrawal of social ‘permission’ for women to work (Klasen and Pieters, 2012; Srivastava and Srivastava, 2010); women exercising the option of not working in activities that they do not like or consider appropriate to their status or qualifications (Das and Desai, 2003); and women taking on more household responsibilities (Neff et al., 2012). It is also supported by research that indicates that women’s labor force participation increases in times of economic crisis (e.g., Klasen and Pieters, 2012; Rangarajan et al., 2011; Himanshu, 2011). Significant distress-related ‘feminization of work’ is also found in regions experiencing drought or 9 famine. Under these circumstances women’s work increases in the farm sector (rather than the non- farm sector) as a result of the household ‘coping strategy’ of women managing agriculture while men migrate in search of other work (Abraham, 2009). Table 2.1 Mean Male Household Wages and Female LFPRs, Rural areas, 2004-05 and 2009-10 Household 2004-05 2009-10 Change in Change in Wage Quintiles Mean Female Mean Male Female Mean Female Male LFPR Household LFPR Male LFPR Household (Percent) Wages (INR) (Percent) Household (Percentage Wages Wages Points) (INR) (Percent) Poorest 158.9 64.6 319.2 50.3 100.9 ‐14.3 Quintile Second 270.7 58.5 544.8 41.8 101.3 ‐16.7 Quintile Third Quintile 388.0 54.3 735.7 39.1 89.6 ‐15.2 Fourth Quintile 607.7 43.2 1076.0 34.0 77.1 ‐9.2 Richest 1326.5 29.6 1902.2 25.2 43.4 ‐4.4 Quintile All 305.3 53.4 614.0 40.4 101.1 ‐13.0 Source: Neff et al. (2012) based on NSS. 2.4 Labor Force Participation and Social Disadvantage13 Socio-religious background is significantly related to women’s labor force participation. Several socio- religious groups in India are both economically and socially disadvantaged, notably the Scheduled Castes (SCs or Dalits), Scheduled Tribes (STs, some of whom are known as Adivasis) and Muslims. An examination of their labor outcomes, and of gender differentials in these, is important to understand the extent to which the enhancement of work opportunities must be made socially (in addition to gender) inclusive. When these groups were compared, hardly any differences were found among them in men’s work participation, but women’s LFPRs were markedly different (Figure 2.2). The highest rates were among STs and SCs, and the lowest among Muslim women (Table 2.2). High participation among Scheduled Tribe women is usually ascribed to the central role they play in tribal culture and economy.14 Tribal groups are amongst the poorest in India but experienced a consistent decline in women’s LFPRs through the period of the four surveys (data not shown) as a result of reduced access to the natural resources that are the bases of their livelihoods, including forest-based, farm and non-farm production. 13 . In this report (as in most Indian documents), those belonging to the ‘Scheduled Tribes’ or ‘Scheduled Castes’ are regarded as belonging to these specific social groups, irrespective of their religion. The ‘Other Backward Classes’ (OBCs) are a separate group, excluding Muslims. And all Muslims except those classified as Scheduled Tribes are another group. The remaining socio- religious groups are combined into an ‘Others’ category. 14 . Scheduled Tribes constitute about 8.25 percent of India’s population and most live in areas that are difficult to reach. SCs account for 17 percent of the Indian population and are spread throughout rural and urban areas of the country. 10 Gender gaps in Participation are also very large in these disadvantaged groups. The gender gap in LFPRs was also lowest in the tribal group, but it increased from 22.7 percentage points in 1993-94 to 35.0 pp in 2009-10. This is sometimes ascribed to ‘cultural mainstreaming,’ a process that, inter alia, erodes the more gender-egalitarian aspects of tribal society and imposes more patriarchal values and practices.15 Dalit women had the second highest LFPRs among social groups most likely, again, due to their poverty, and the gender gap in the Dalit LFPRs was about 38-39 pp until 2004-05 but increased further to 46.5 pp in 2009-10. Expanding work opportunities for Dalits is needed also to help them break out of stigmatized situations and occupations and obtain ‘decent’ work to reduce poverty. Although Muslim men had participation rates comparable to Others, Muslim women had the lowest LFPRs of all women, and the gender gap was highest amongst Muslims (62.8 pp in 2009-10). While Muslim populations in India are also amongst the poorest, women’s LFPRs are low among them as social norms, apparently stronger than the pressures of poverty, result in their seclusion, a situation that also calls for special approaches to enhance women’s work participation.16 Women in the Others category had very low LFPRs and there were large gender gaps among these higher castes and other religions. This situation arises from the higher economic and education levels of these women. As discussed above, in higher status households women tend to withdraw from paid market work. Indeed, a process of ‘Sanskritization’ is occurring in women’s work participation among all social groups, leading to the decrease of differences in the freedoms of ‘mobility’ and ‘autonomy.’17 Table 2.2 LFP Rates for Women and the Ratios of Male to Female LFPRs by Social Group, 2009-10 Female LFPR M:F LFPR Ratio Muslims 18.3 4.4 Others 23.2 3.3 All Women 31.6 2.5 Other Backward Classes 35.1 2.3 Scheduled Castes 36.3 2.3 Scheduled Tribes 49.9 1.7 Source: World Bank staff 15 . It is important to note that these socio-religious categories are not homogeneous as geographic and other cultural contexts also shape norms. Thus, LFPRs among ST women, for example, in the Northeast states, reflect the particular culture and history of women’s development in that region which are quite different from those of the Central Indian region. 16 . LFPRs fall off more sharply after the age of 45 years for STs, Dalits and women from Other Backward Classes compared with their better-off counterparts, possibly because of worse health among the poorer groups. Details about other disadvantaged groups including the disabled and female-headed households were not available. However,o’Reilly has written extensively on the discrimination faced by disabled women in employment. This is a global trend. See, O’Reilley, 2003, 2007 for more details. 17 . ‘Sanskritization’ is a term used by the Indian anthropologist M.N. Srinivas to describe the process by which lower caste groups adopt the values and practices of higher caste groups. 11 Figure 2.2 Labor Force Participation Rates of Different Social Groups, 2009-10 100 90 80 70 LFP Rate 60 50 40 30 20 10 0 Scheduled Scheduled Muslim OBC Others All Tribe Caste Social Groups Rural Males Rural Females Urban Males Urban Females Source: Raveendran, 2012 BG 2.5 Education and Work The relationship between education and LFP is clearly ‘U-shaped’. Figure 2.3 shows that the highest work participation rates are amongst illiterate men and women (except for post-graduates and above, among the latter). LFPRs decline thereafter to higher secondary school completers but increase again among those with tertiary education. The male/female ratios in LFPRs are roughly the same among Illiterates and college graduates, but are dramatically higher (unfavorable to women) between these levels. While households with better economic levels support girls’ school education they have yet to make the social transition to women working for the intrinsic value of this. They choose marriage and motherhood for their young women over further education or work. But there is also a lack of jobs suited to women with ‘middling’ education. An alternative explanation is that these women are ‘unemployable’ but this will be discussed later. The increase in women’s LFP with tertiary education (surmounting some of these barriers) is due in part to the availability of good opportunities for the (relatively small) numbers of women educated to this level. Although college-educated women still have lower employment rates than their male counterparts, tertiary education has had a greater effect on women’s LFPRs over the past two decades than on men’s. Technical/professional education in particular has a strong effect on women’s work – there is little gender difference among the professionally qualified in urban areas. Rural-Urban Differences are Significant in Women’s LFPRs. Throughout the period of the surveys, rural-urban differences in men’s LFPRs were marginal, but there were significant differences in women’s LFPRs as noted earlier, with urban rates being much lower. This was particularly so among the less educated (Figure 2.4). The very low overall urban female LFPR of 19.5 percent is special cause for concern as (not only were urban areas the locus of India’s economic growth during this period) but the urban population also includes women graduates who have the highest urban LFPRs. 18 However, they have the same LFPR as women with secondary-school education in rural areas, and much lower rates 18 See Bhalla and Kaur (nd) for a discussion on why the lowest LFPR was for urban women. 12 than illiterate or primary-schooled rural women, a pattern that is not seen among men. Although this is at least partly an income effect, it points to the constraints faced even by educated women in urban areas. Male and female LFPRs converge among rural Professional Graduates but other gender gaps remain large. Figure 2.3 Labor Force Participation among 15-64 Year-olds, 2009-10 100 94.9 95.9 92.9 84.4 93.2 86.8 82.0 80.7 90 73.3 80 64.9 70 80.5 60 73.1 70.8 67.4 56.5 LFPR 50 65.8 61.6 59.1 59.7 46.0 40 52.1 42.2 30 40.1 45.5 Male 35.6 24.8 20 32.6 Female 10 25.4 14.5 16.8 Total 0 Education Level Source: Sankar, 2013:5 Figure 2.4 Labor Force Participation of Men and Women by Education, Rural and Urban Areas, 2009-10 100 90 80 70 60 LFPR 50 40 Rural Male 30 Rural Female 20 Urban Male 10 Urban Female 0 Education Levels Source: Sankar,2013:6 13 Changes over time show an increasingly educated work force. Using a broader grouping of education levels, Table 2.3 presents LFPRs for rural and urban women over a decade, and the shares of each group in the female labor force. While uneducated women had the highest LFPR in rural areas, in urban areas this applied to those with tertiary education. Urban women with tertiary education, however, showed the steepest decline in LFPR between 2004-05 and 2009-10. The steepest decline in rural areas was amongst uneducated women, while the LFPRs of their urban counterparts did not decline. In sum, while all educational groups experienced a drop in LFPRs between 2004-05 and 2009-10, uneducated and tertiary educated women, the ‘highest performing’ groups in the U-shaped distribution, were worst affected. In general, the trend in female LFPR during this period (as we have seen earlier) was a rise between 1999-00 and 2004-05 and fall between 2004-05 and 2009-10 for each group (giving an overall decline), except among primary-schooled urban women who saw a net increase. There was also a slight shift in the educational level of women in the labor force – a smaller proportion of uneducated women in 2009-10 compared with 1999-00 in both rural and urban areas, and higher proportions with primary and secondary education in rural areas, and with tertiary education in urban areas. These shifts in the labor force mirror the changes in education levels during this period. Table 2.3 Labor Force Participation among Rural and Urban Women by Education Levels 1999-2000 2004-05 2009-10 Rural Education Levels Percent LFP Percent of LFP Percent LFP of LF Rate LF Rate of LF Rate All Education Levels 100 41.7 100 45.0 100 34.9 No Education 74.0 47.0 67.0 50.2 56.6 39.7 Primary School 14.4 36.1 16.5 41.2 22.0 35.5 Secondary School 10.6 26.4 14.4 32.5 18.6 24.9 Post-Secondary School 1.0 42.1 2.1 50.1 2.8 38.7 Urban Education Levels Percent LFP Percent of LFP Percent LFP of LF Rate LF Rate of LF Rate All Education Levels 100 20.6 100 24.3 100 20.0 No Education 41.5 26.7 35.7 30.6 27.9 24.6 Primary School 15.2 17.4 17.2 23.5 17.8 21.5 Secondary School 26.4 13.9 25.5 15.5 28.1 12.8 Post-Secondary School 16.9 31.7 21.6 37.2 26.2 30.7 Source: World Bank staff Technical Education is especially beneficial for women’s work participation. Among people with tertiary education, there are differences in LFP with the type of education. Women with technical professional education have higher LFPRs than those with other tertiary education in both rural and urban areas. In contrast with most other participation rates that declined between 2004-05 and 2009- 10, the LFPR for women technical professional graduates increased. The ratio of male/female LFPRs shows that this group also had the greatest gender parity. 14 2.6 Interactions between Education and Economic/Social Status Women’s LFPRs decline with increasing MPCE at all levels of education (except Graduates and above among whom the third and fourth deciles have higher LFPRs than others). This suggests that the ‘missing middle’ in the rural women’s work-MPCE curve could be addressed through higher education – i.e., that women in the middle MPCE deciles would join the labor force if they had at least high-school education, a finding with considerable policy import. In urban areas only among women with Graduate education and above did participation rates remain relatively high with increasing economic status. To break work barriers urban women needed to have completed college. While education is intrinsically empowering for women, to increase women’s work participation India must achieve at least secondary and preferably tertiary education for girls. Education has profound effects on the LFP of women in the disadvantaged social groups. As Figure 2.5 shows, men’s LFPRs remain fairly stable across educational categories, while women’s participation rates are U-shaped in all the different socio-religious groups. However, the pattern is somewhat different for each group, and the LFPRs attained also vary. While almost similar percentages of women in the four disadvantaged groups are illiterate (about 60 percent of SC and ST women, and over 50 percent of OBC and Muslim women), about 55 percent of illiterate ST women were in the labor force, but less than 40 percent of SC women. While there are large gender gaps in the LFPRs of Dalits at all levels of education, the gap closes completely among professionally-educated Dalits, but not among the other disadvantaged groups, nor even among ‘Others.’ This is most likely the impact of India’s reservations policies which have provided both education and employment opportunities within reach of SC women. Regression analysis confirms many of the relationships discussed above.19 Women’s labor force participation clearly shows a U-shaped relationship with education. Among men, only those with Higher Secondary education ‘dip.’ Both urban men and women participate less in the labor force than their rural counterparts (urban coefficients were -0.03819 and -0.110401, respectively, which also show that urban residence affects women much more strongly). Those belonging to the SC, ST and OBC communities had higher LFP rates while Muslim women had significantly lower participation rates than those in the Others category (regression coefficient -0.127). As will be discussed in Chapter 4, while marital status did not deter men from working (married men had a higher LFPR than unmarried men, regression coefficient 0.1526), married women participated less in the labor market than unmarried women (regression coefficient -0.1210).20 Indeed, other econometric analysis has provided various explanations for the levels of female work participation observed in India. The explanations range from women’s entry into the labor force being driven by poverty to the absence of ‘preferred’ job opportunities which discourage women from seeking work, to gendered practices in the labor market excluding women (Sundaram and Tendulkar, 2004; Das, 19 . The methodology is presented in Sankar (2012BG). 20 . Field surveys provide a different perspective. Sudarshan and Bhattacharya’s study of urban Delhi showed that women with no schooling reported higher labor force participation than women who have completed schooling. Close to half the women who were educated above graduation joined the workforce (Sudarshan and Bhattacharya, 2008:12-13). 15 2006; Soni-Sinha, 2006). The most likely explanation depends on which segment of the work force is considered. For the work force as a whole, Mukhopadhyay and Tendulkar (2006) analyzed the combined impact of a range of variables available in the NSS: literacy, age, marital status, household size, region, per capita household expenditure and social group. Their estimated equations had a low overall explanatory power, pointing to missing variables, among which data on earnings and other sociological factors were possibly important. Additional analysis confirmed a high congruence between the activity status categories of wives and husbands. For example, men who were regular wage or salaried workers had the highest proportion of spouses not in the labor force. Controlling for income levels the more secure a husband’s earnings (regular salaried wage status), the lower was the probability of a woman entering the labor force. Adding a constructed variable for ‘couple disparity in education,’ the results showed that, ceteris paribus, this variable reduced the probability of the wife being in the labor force. Using data from a subset of the NSS in which earnings information is also collected, and estimating earnings functions separately for women and men, the difference between male and female wages was partly explained by a ‘characteristic effect’ while the residual, ‘discrimination effect,’ ranged from 38 to 67 percent for rural and urban groups (Mukhopadhyay and Tendulkar, 2006). Overall, the data confirmed the relevance of supply-side explanations although, for educated and skilled women, gendered discrimination within the labor market might lead to withdrawal. 16 Figure 2.5 LFPR among Different Social Groups by Education, 2009-10 100 Dalits Adivasis 100 80 80 60 60 LFPR 40 LFPR 40 20 Female Male 20 0 0 Education Levels Education Levels 100 OBC 100 Muslims 80 80 60 60 LFPR LFPR 40 40 Female 20 20 Male 0 0 Education Levels Education Levels 100 Others 80 60 LFPR 40 Female 20 Male 0 Education Levels Source: Sankar, 2013:10-11 2.7 Returns to Education Annex Table 2.1 gives the average wage rates by levels of education for women and men in rural and urban areas over the 16-year period of the surveys. The higher returns for each group with increasing education are evident.21 So too are the gender gaps in returns at all levels of education, which occur in both rural and urban areas and all four survey years. In general, the male/female ratios are higher at 21 See also, Rao et al.,2007 17 lower levels of education – that is, increasing education, particularly secondary and tertiary education, improves gender equality in wage rates. Among the less educated, gender equality is lower in urban areas while among the more educated it is lower in rural areas. Significantly, there appears to have been no improvement in equality over time. Figure 2.6 shows returns to education and wage regression coefficients in graphical form. Figure 2.6 Wage Regression Coefficients and Returns to Education for Men and Women in 2009-10 3 100 85 Female Female Wage Regression Coefficients 2.5 Returns to Education 80 71 68 60 60 2 60 52 43 1.5 36 40 24 28 2724 1 15 0.5 20 43 9 -11 0 0 -6 -0.5 -20 Education Levels Education Levels Simple Mincerian Augmented Mincerian Simple Mincerian Augmented Mincerian 2.5 100 Male Male Wage Regression Coefficients 82 2 80 76 Returns to Education 60 1.5 60 46 45 4642 1 40 28 27 37 23 0.5 20 11 14 11 15 5 0 0 -5 -6 -0.5 -20 Education Levels Education Levels Source: Sankar, 2013BG:21 18 Table 2.4 Summary Results of a Regression Analysis of Wage Rates All Rural Urban Female Positive effects Positive and larger effects than in rural areas  Technical education  Formal work  Age  Completed years of education  Principal status Negative effect  Formal work * Positive but smaller effect than in  Principal status rural areas Positive effects for women Technical education  Urban residence Male Positive effects Positive and larger effects than in  Technical education  Formal work rural areas  Technical education  Principal status  Age * Negative effect  Formal work  Education  Female sex, i.e., female  Technical education * wage rates are lower than  Principal status * male Positive but smaller effects than in rural areas  Completed years of education  Formal work Note: The items with an asterisk (*) showed a larger effect on wages of that sex than on the wages of the opposite sex. 2.8 The ‘Two India’s’ are Three for Women and Work While Indians often refer to ‘two India’s’ or ‘Bharat and India,’ i.e., an India that lives in poverty and tradition and one that is modern and affluent, this analysis of the economic and social correlates of women’s labor outcomes suggests that there are three India’s in the arena of women’s work. The first is where women are poor, uneducated and socially disadvantaged and have high rates of participation, primarily because they need to work and earn for their households to survive. In the second, they are relatively better-off but not ‘emancipated’ – their work participation is low either because there is no economic need for them to work, because the household values their presence at home more than their potential work, or because they are strongly constrained by social norms in the household as well as the labor market. In the third India, women are well-off and participate in work, most likely because with education their horizons and ‘identity’ have changed and their earnings make work worthwhile. 19 3. GENDER AND EMPLOYMENT BY SECTOR 3.1 The Labor Market and Gender In order to identify how gender imbalances in labor outcomes arise and thus could be addressed we turn to a discussion of women’s and men’s occupations (i.e., the ‘Industry Groups’ or sectors in which they work) and how they are engaged (their ‘Activity Status’).2223 Including an examination of the labor markets in rural and urban areas, this chapter explores the trends in industry groups in relation to economic growth in India over the period 1993-94 to 2009-10, and particularly the crucial five years (2004-05 to 2009-10) when women’s labor force participation fell considerably. The discussion of Activity Status encompasses issues of informality, self-employment and wage work. The subsequent section consists of case studies of three sub-sectors in which women’s employment in India has been growing – Garment Manufacturing, Private Household Services, and the ‘white collar’ Software Industry and identifies lessons from these sectors that are useful for the future. Over 85 percent of India’s workforce is in five sectors of the economy: Agriculture, Manufacturing, Trade and Repair Services, Construction, and Transport, Storage and Communications. Two additional sectors (Public Administration and Defense and Education) gain importance in urban areas. Constraints to growth in these sectors would affect labor outcomes adversely. A notable case in point is Agriculture, where growth has been severely constrained although it continues to be the dominant area of work for both women and men. The overlap of LFP in this sector with social and economic disadvantage is striking: for example, 81 percent of female agricultural workers are SCs, STs and OBCs, and 55 percent are casual laborers. The proportions of women workers in these sectors illustrate the low equilibrium between female labor supply and demand. Only 27 percent of the total workforce in India is female (31 percent in rural and 20 percent in urban areas, amounting to 128 million women workers). Women account for more than one-third of workers only in Agriculture, Education, and Health and Social Work, and for more than one-fourth in Manufacturing, and Other Community, Social and Personal Services as well (Figure 3.1). In most other sectors they are only between five and ten percent of workers. 22 . The ‘National Industrial Classification’ of workers (NIC – 2004) used by the NSS has 17 ‘Industry groups’ at the single digit level (see Figure 3.1 and Table 3.1 for the list). 23 Kabeer, 2012:12 draws attention to the fact that many of the disadvantages faced by women from low income and socially marginalized groups are also faced by men from these groups but gender exacerbates the situation. 20 Figure 3.1 Share of Women in Each Industry Group, NSS 2009-10 60 56.37 50 Percent of Women among 40.51 42.18 40 33.54 Workers 30 25.49 25.46 20 13.51 15.73 13.44 10.15 11.25 12.04 7.57 8.59 10.20 10 2.55 0 Industry Groups Source: Sankar, 2013 BG:13 The distribution of the female workforce is skewed toward the primary sectors (Figure 3.2). Two-thirds of all women workers are in Agriculture, about one-tenth in Manufacturing, and five percent or less in several other sectors including Construction, Education and Trade and Repair Services. While the agricultural workforce was being ‘feminized’ in earlier periods, there was a steep decline of women in agriculture between 2004-05 and 2009-10. Nevertheless, 80 percent of the rural female workforce continues to depend on this sector. The distribution of women workers across sectors (e.g., almost 71 percent in the primary sectors, about 15 percent in the manufacturing and construction, and under 13 percent in services) can be compared with the percentage shares of GDP accounted for by agriculture and allied sectors (14.64), Industry (28.27) and Services (57.09) in 2009-10. Figure 3.2 Distribution of Female and Male Work Force, NSS 2009-10 Agri, hunting, forestry (70.2) Agri, hunting, forestry (48.5) Female Manufacturing (10.5) Male Construction (12.7) Construction(4.6) Trade (10.8) Trade (3.5) Manufacturing (10.7) Education (3.4) Transport & commerce (5.6) Other social services (1.8) Public admin & defence (2.1) Others (1.6) Other social services (1.8) Health & social work (1.1) Education (1.7) Public admin & defence (0.9) Hotel & restaurant (1.4) Hotel & restaurant (0.8) Real estate (1.3) Transport & commerce (0.4) Financial intermediaries (0.9) Real estate (0.4) Mining (0.7) Financial intermediaries (0.4) Health & social work (0.5) Mining (0.3) Fishing (0.5) Fishing (0.1) Others (0.4) Electricity, gas, water (0) Electricity, gas, water (0.3) Source: Sankar 2013 BG:13 21 3.2 Changes in the Time of Economic Growth During the fall in women’s employment between 2004-05 and 2009-10, women exited from both long- term and short-term work, i.e., they were not pushed from being Principal Status workers to Subsidiary Status as the latter also declined in absolute numbers and in its proportion of all women workers. This may have been due in part to the fall in seasonal agricultural tasks (in which women are engaged) during the drought experienced throughout the country in 2009-10. However, it signals the lack of alternative sources even of temporary work (which is greatly needed by the poor) as well as the inadequate diversification of livelihoods among long-term workers. Some shifts between unpaid family work and casual wage work suggest the importance of the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) in shoring up women’s work during this time.24 Wage employment. MGNREGS 25which began in 2005 was designed with a gender perspective (Aiyar and Samji,2009), explicitly seeking to address some of the challenges that poor working women face, such as gender inequality in wages and a lack of child care (Holmes,et al., 2010:viii). It has provided work and incomes to women in the poorest households: for example, women accounted for almost half of all person-days of work generated by the program up to March 2012, exceeding the ‘quota’ of 33 percent26. Research shows that women were paid equal wages and maintained control over them (Dreze and Khera, 2011:54-55; Vanitha and Murthy, 2011:417). Among the major reasons for these successes are: the operational guidelines of MGNREGS take women’s needs and constraints explicitly into account; training programs for managers include modules related to gender issues; and the impact of the program on women is a criterion in monitoring and evaluation27. A lacuna that has been reported, however, is that the poorest and most vulnerable women are often excluded because they cannot produce the documents required for enrolment. To reach its full potential to enhance women’s work MGNREGS needs (among other things) to increase its coverage in states where it is low (e.g., Uttar 24 One field survey in Andhra Pradesh showed that that the overall rate of female labour-force participation in the age group of 16 to 60 years has increased substantially, by 13 percentage points, while for males it fell marginally between 2007 and 2009 – 10. This rise in female labour-force participation is largely driven by casual labour (Afridi et al., 2013: 5).According to Lahoti and Swaminathan, NREGA had a positive impact on FLFP rates and there was a smaller decline in the NREGA districts between 2004- 05 and 2007-08 than in non NREGA districts (Lahoti and Swaminathan, 2013: 10). 25 Formerly known as NREGS (National Rural Employment Guarantee Scheme). 26 By the end of March 2012, almost 50 percent of workers on MGNREGS sites were women. Dutta et al., (2012) use the 2009- 10 NSS data to point out that the female share in works under MGNREGS is greater than their share of work in the casual wage labor market across all States (Dutta et al, 2012: 62). This, as the authors point out, is a very high rate for a country where women’s labor force participation is normally very low. Ghosh points out that women are participating in the scheme much more actively than they participated in all forms of recorded work (Ghosh, 2009 b). See also, Khera and Nayak, 2009:3. 27 Another reason cited for success is the impact of social mobilization. Shariff’s survey of 3200 households across 16 of the most deprived districts of North India revealed that a woman’s participation depended to a great extent on whether she was participating in the panchayats, school committees, mahila mandals, etc. Participation in panchayati raj linked institutions was crucial for women’s participation in MGNREGS (Shariff, 2009: 8). The Kudumbashree neighbourhood self -help group program is the major reason for such a high participation of women in MGNREGS in Kerala. The researchers found that 75 percent of the MGNREGS workers were Kudumbashree members. See also Dasgupta and Sudarshan, 2011:9. 22 Pradesh and Bihar)28; ensure that social norms do not interfere with women’s access to the program; and provide flexible working hours and the mandated child care facilities (Holmes et al., 2010:24). Over the 16-year period, there was a good increase in women workers in Construction and Education.29 Construction achieved a very significant increase, rising above a 5 percent share. However, the shares of women workers in important sectors such as Manufacturing, Trade and Repair Services, Hotels and Restaurants and Private Household Services declined. Just between 2004-05 and 2009-10 there were significant reductions in Agriculture, Manufacturing and Private Household Services. Overall, there was a drop of 20.5 million female workers over this period. Women have very low shares of jobs in a number of sectors which have expanded through processes of growth, such as Trade and Repair Services, Hotels and Restaurants, Construction, Public Administration and Defense, Financial Intermediation, and sub-sectors such as Information Technology Services. In the last three of these, over 50 percent of workers are formal. Education, and Health and Social Work, where women have greater shares of jobs (although still only 6.3 percent of the total), are also more formal. In the first two sectors above, over 90 percent of workers are informal. High levels of informality also pervade occupations in which women have larger shares such as Agriculture, Private Household Services, and Textiles and Apparel Manufacturing. The nature of interventions required to increase women’s participation in formal and informal sectors of course varies considerably. There is a greater divergence of male and female activities in urban areas, suggesting stronger gender stereotyping of urban work. Despite their overall low work participation rate in urban areas, women had a greater share than men in Private Household Services and were over 40 percent of workers in Education, and Health and Social Work. They were disproportionately represented in Trade and Repair Services (especially in the informal sector) and Other Community, Social and Personal Services and men in Manufacturing and Construction. Urban employment increased over the 16-year period but the male workforce grew at an average annual rate seven times that of women (especially in Manufacturing) – i.e., on average, women got only one out of eight jobs that opened up in urban areas where most of India’s GDP growth took place. In rural areas there were few additional jobs, and the male Annual Growth Rate was six times that of women. Rural/urban disparities in sectoral growth were greatest in Agriculture and Manufacturing. The 28 While the national figures for the percentage of female workers engaged in MGNREGS works are impressive, state-wise figures present quite a different story. Thus, for instance, in Uttar Pradesh only 17 percent of workers were women; 18 percent of workers were women and in Assam, 20 percent of workers were women in the financial year 2011-2012. These figures are much lower as compared to Kerala, where 93 percent of workers were women, and Tamil Nadu, where 74 percent of workers were women (Merin and Mohandas, 2012:152; Sudarshan, 2009:18).Rajasthan reported 69 percent women workers, Himachal Pradesh reported 60 percent women workers and in Andhra Pradesh, 58 percent were women workers. Out of 20 states where work was going on, 5 states did not fulfill the stipulated minimum of 33 percent participation of women. Dasgupta and Sudarshan explain these variations as a result of a variety of reasons which include socio-cultural norms, mobility and intra- household allocations of roles and responsibilities; levels of care responsibilities, number of adult women in the house; other market opportunities and market wages for men and women and mobilization work done by NGOs and activists (Dasgupta and Sudarshan, 2011: 9). 29 . Some sectors with small numbers of women and low (below 2 percent) shares of all women workers (including Health and Social Work, Financial Intermediation, Public Administration and Defense, Electricity Gas and Water Supply and Transport, Storage and Communication) also increased. 23 latter declined in rural areas but grew in urban areas, while the primary sectors declined in urban areas. Accelerated growth occurred in both rural and urban areas in Construction, Financial Intermediation, and Public Administration and Defense between 2004-05 and 2009-10. 3.3 High Risks, Low Returns for Women Informality defies boom. Despite the decades of high economic growth, the Indian workforce remains highly informal – almost 95 percent of women and 91 percent of men workers are so, and the trend has been one of increasing informality. Home is where the work is. The dominance of informality and slow growth of formal work are both causes and effects of a pattern of work organization that emanates from adjustment to labor regulation or evasion of it. For example, the growth of exports after liberalization led to out-sourcing of work by growing or new industries, and home-based work in turn enabled the expansion of a range of market goods. Women’s low mobility – their implicit and, often, explicit seclusion – supports the predominance of home-based work which includes unpaid family work, self-employed own account work, and wage work such as piece-rate work in garment manufacturing. Thus, women continue to be concentrated in home-based work and increasingly so. While in 1999-2000, 11 percent of men and 35.5 percent of women in informal non-agricultural work worked from their own homes, by 2004- 05 this had changed only marginally for men (to 12.3 percent) but had increased to 51.3 percent for women (Jhabvala et al., 2007). Women workers are concentrated at the bottom of the pyramid. The lowest status workers in a hierarchy of work arrangements that drops from Regular Salaried/Wage Workers (RWW) at the top to Own Account Workers (OAW), Home-based Workers (HBW), Casual Wage Workers (CWW) and Unpaid Family Workers (UFW) at the bottom are largely rural and predominantly female. They account for about half of women workers and are one-third unpaid family workers and two-thirds casual wage workers (the poorest women who work outside the home). The ‘better’ half of women workers is about two-thirds self-employed workers (SEW), mostly own account workers and a few Employers and one- third regular wage workers. Over the 16-year period, women self-employed workers declined and wage workers (both regular and casual) increased. Overall, the need is not over to stop the decline in the number of women workers but also the downward ‘status’ shift of women workers, enhancing women’s abilities to rise to better paid self-employed or regular wage work. Gender affects activity status in different sectors. The above distribution of activity status applies strongly to rural workers in Agriculture and Manufacturing (and is, in fact, derived from these sectors because of their predominance in overall employment). Table 3.1 summarizes some gender differences and similarities in activity status in different sectors in rural and urban areas. Education has an impact on activity status. More-educated women are usually regular wage or self- employed workers while the less-educated are casual workers or self-employed in Agriculture. Women make up sizeable proportions of skilled workers in some areas – for example, 30 percent of skilled agricultural workers, 29 percent of technical professionals, 19 percent each of professionals and of 24 people in crafts and trades, 16 percent of clerks, and 14 percent each of services-shops-sales people, and legislators-officers-managers. Table 3.1 Gender Differences in Activity Status of Workers in Key Sectors Men Women Rural Largely complementarity Complementarity between Agriculture between UFW and CWW (seen UFW and CWW in many sectors) Both RWW and CWW increased OAW and RWW increased but Manufacturing during the 18-year period UFW and CWW decreased Construction SEW decreased Amost all CWW Trade and Repair 90 percent SEW, distributed 15 percent RWW and CWW Services evenly among OAW and UFW Private Household 56% CWW Over two-thirds RWW Services Only small differences in distribution by sex and activity status in Education both rural and urban areas Urban Manufacturing (large Two-thirds SEW of which two- Two-thirds RWW and CWW employer) thirds are OAW Private Household Much less CWW as in rural areas Services Education Five-sixths RWW RWW doubled among rural females and increased by 45% among Overall urban females over 18 years Gender gaps in wage rates have increased over time due to lower growth in female wages even during the period of economic growth. This is most likely a significant cause of the decline in women’s LFP. The higher increase in male wage rates may have contributed also to lower female participation, but it is important to remember that most of the decline in women’s LFP took place in the poorest groups who receive the lowest wages. Wage rates increase with middle levels of education and increases in education improve gender equality in wage rates, but gender wage gaps remain high at all levels of education, in both rural and urban areas, and over time. Increases in women’s wages do result in their increased participation, providing important policy levers. 3.4 Formal Employment Formal employment is available mostly to educated and middle-income women who have the freedom to undertake (usually) full-time work. Increasing numbers of women with higher levels of education, greater desire among young women to delay marriage and child-bearing and greater value attached to 25 women’s earnings and work itself are the supply-side drivers of formal employment. Formal work provides a regular income, some social security, social and cultural capital, and savings potential, with benefits to both individual and aggregate productivity. Nevertheless, formal work is also considered to have ‘dis-benefits’ including the neglect of child and elderly care (or a double burden on the working woman) and exposure to public spaces and work places with the attendant dangers of sexual harassment and gender-based violence. 3.5 Some Areas of Growth in Women’s Employment While women’s work participation was declining overall, employment opportunities for them grew in some sub-sectors. Understanding how these successfully provided work to women could help to identify the practical measures and policies that would promote more gender-equal employment in India’s future. Three buoyant areas were: garment manufacturing, private household services and information technology. Case studies of these sub-sectors prepared for this report and discussed below bring out a number of important issues surrounding the growth of women’s work (Chakravarty,2012BG). 3.5.1 Garment Manufacturing As discussed earlier, Manufacturing is the second largest employer of women in rural areas after Agriculture, and the primary employer of urban women. In some sub-sectors such as garments, electronics and leather products, the percentages of women workers are higher than in others.30 The export component of these sub-sectors is generally higher than that of other manufacturing activities. In the early exporting countries of East Asia, export production depended largely on the expansion of female labor; this makes these sectors attractive for increasing women’s employment in India as well. Labor-intensive products exported by developing countries are, by nature, low technology ones, and this is especially true in the clothing industry.31 The increasing cost competition in export markets leads entrepreneurs to search for newer sources of cheap labor. Young women are ready to work not only for very low wages but also for longer hours under exceedingly inhospitable working conditions. Indeed, the gender gap in wages accounts in part for the differential distribution of male and female workers in different branches of industry. An additional reason behind women’s increasing absorption in export- oriented manufacturing is their high turn-over which makes for a flexible work force. Women workers are less likely to protest layoffs, a corollary of the ‘oriental docility’ that inhibits them from joining unions or agitating against managements. Women’s work in garment manufacturing in developing countries in general and South Asia in particular is of two main types. Women are found in large numbers in home-based production, often without any direct relationship to the market – i.e., with an intermediary or contractor operating between them and the market (Kabeer, 2001; Ghosh, 2002); and they also work in substantial numbers in garment 30 . This does not mean that there are larger numbers of women than men working in these sectors. At the micro level, however, there may be factories with predominantly women workers. 31 . However, in recent years some technological modernization has taken place in this industry. We will return to this issue later. In contrast, the production processes of electronic goods are highly technology-intensive and being modernized continuously. 26 factories, especially in export-processing units and zones. Their numbers in home-based work are increasing in line with the increasing informality of women’s employment,(Mahadevia, 2001; Kantor, 2005) and as a result of labor regulations that cause firms to opt for decentralized production, sub- contracting or labor-saving technological modernization, the last leading to fewer workers in factories. Enhancing home-based work. Improving market access for home-based garment producers can increase the economic success of these workers. Access to markets may call in turn for better skills and investment in equipment to improve product quality. The lack of institutional credit is a constraint to such investment as women usually do not own property to use as collateral. The formation of self-help groups or co-operatives (discussed in Chapter 4) offers a solution to some. SEWA’s experience of organizing self-employed women in a variety of trades – from home-based garment producers to forest- based gum collectors – into producer groups in order to build and own assets has greatly enhanced their capacity (Bhatt, 2006). In sum, the adverse effects of women’s dependence on middlemen in the growing home-based subsectors could be partially mitigated by membership of credit and producer groups with better direct access to markets. Equalizing the factory experience. Women workers also constitute a significant percentage of the workforce in garment manufacturing firms, for example, in export parks. Studies of garment factories in Hyderabad have indicated that export firms often prefer women workers (Chakravarty, 2004; 2006; 2007). Managers found them sincere in their work and less likely than men to participate in unions and strikes which are costly to export firms with highly time-sensitive delivery schedules. However, there has been a decline in the share of women workers in factories in the face of increasing technological sophistication. Although much of the production machinery, methods and tasks in this industry in India are behind state-of-the-art, a substantial amount of technological modernization is taking place in the larger apparel-manufacturing firms, and changes in the production process have led to some significant changes in certain job categories. Men are increasingly taking over as female candidates for the ‘new’ highly-skilled jobs are less likely to be exposed to technical education and work experience. In Tiruppur (Tamil Nadu) and Delhi, Singh and Sapra, (2007) found that women workers were concentrated in the lower categories of work in garment production and were often discriminated against in wages. However, in Hyderabad while women did dominate lower categories of workers, they were also in higher categories such as Designers and Supervisors, even in technologically more advanced firms (Chakravarty, 2004; 2006). There was career progression among women, with no evidence of wage discrimination. Some managers who preferred women workers said that they found it difficult to find women with at least primary school education and the expertise needed, for example, to be a Supervisor in a garment unit. Hence, they opted to train their most efficient women workers in order to replace at least some male supervisors with women. Obtaining workers was also difficult – managers wanted younger people with basic schooling, averring that schooling attainments play a very important role in job-learning abilities. Some firms opted to train people from nearby villages, employing more women than men, perhaps because of a tradition of engaging women in certain production activities. Thus, technological modernization may be less of a constraint to the growth of women’s participation in garment 27 manufacturing than local shortages of educated women. Women may lose opportunities in the labor market, even when they are preferred, because of gender inequalities in other spheres of development. Increasing the demand for industrial labor in the economy may be a proximate way to increase wages in the unorganized sector in general and garment manufacturing in particular, as Singapore demonstrated. This process would be facilitated by managements’ preferences for women workers because of their lower propensity to be involved in trade union activity. Based on detailed discussions with women workers, Chakravarty ,2007 pointed out important reasons for their apparent union-averse behavior. Some women were the main breadwinners for their families (because of widowhood or separation) and a strike would be extremely costly as it would mean a loss of pay. Closure of a factory or lay-off would be disastrous. It is more difficult for a woman than a man to find a new job, particularly as she is unable to venture far from home, and also because she is less endowed in terms of education and information. 3.5.2 Private Household Services Paid urban domestic work is highly feminized with an increasing presence of girl children in recent years. ‘Private household service’ is emerging as an important area of work for women in India on account of growing demand in urban areas as well as from other countries. There has been an increasing concentration of women in this essentially low-paid and irregular work in the informal urban economy. About 65 percent of workers in the sub-sector are women; and around 7 percent of urban working women are concentrated in this service alone. The share of domestic work in all service work done by women increased from 11.8 to 27.1 percent between 1999-2000 and 2004-05 (Neetha, 2009). Domestic service is also ‘girl children’s work’ – among children in the age-group of 5 to 15 years who were working as domestics, around 60 percent were girls (Chakravarty, 2007). The increase in domestic service is associated with the increase in the size of the Indian middle class since the 1990’s and the participation of upper class women in work. Demand for domestic workers is likely to grow with urbanization. The case of West Bengal. Recent changes in West Bengal’s economy, led by a boom in the services sector, especially in and around metropolitan Kolkata, have opened up opportunities in both higher and lower status work, and there has been a significant increase in the participation of women between the ages of 25 and 34 years as professionals as well as domestic workers. Although still a poor performer in adult women’s work participation, West Bengal has shown increases in the LFPR of the richest class of urban women in recent years – and they are in need of domestic help for child care in the absence of institutional support mechanisms such as crèches, full-day schools or day-care centers. Consequently, while domestic service occurs throughout India, it constitutes 77 percent of all personal services provided by urban women in West Bengal compared with 48 percent nationally (Chakravarty, 2007). Domestic workers who live in the cities or commute to them daily are mostly adult women, while live-in workers are usually girl children, migrants from rural areas of the state (Kundu, 2008; Chakravarty and Bose, 2011; Chakravarty and Chakravarty, 2012).32, 33 32 . The high rate of participation of girl children in paid work in West Bengal (which has a historically low female work participation rate) makes the state somewhat unique (Chakravarty et al., forthcoming). However, similar trends are emerging 28 Improving the Quality of Employment. Many adult women domestic workers are the chief breadwinners of their families and command significant decision-making power within the family (Neetha, 2004; Chakravarty and Bose., 2011; Chakravarty and Chakravarty, 2012). However, the quality of their employment is often at issue, with its terms and conditions varying significantly. The individual nature of domestic service makes workers highly vulnerable to exploitation. While those living in the city may have some bargaining power as a result of their exposure to urban culture, and may be able to obtain higher wages, limited working hours, and so on, new migrants, ‘live-in domestics’ and girl children are often subject to poor treatment by employers (as well as by recruitment agencies). Domestic workers do not have key labor protection (exemplifying the informal economy in India in general) so that making this work a ‘decent’ livelihood option for women calls for several issues to be addressed including a minimum wage, weekly holiday, adequate notice before dismissal, access to medical and maternity benefits, and pension schemes. Some of these aspects are covered by the Domestic Workers (Registration, Social Security and Welfare) Acts of 2008 and 2010. Formation of domestic workers’ associations could help workers ensure implementation of these Acts. 3.5.3 ‘White Collar’ Work in the Software Industry While educated urban women are concentrated in the Education, and Health and Social Work sectors, women’s participation in Information Technology (IT) work is growing visibly. IT-related services are primarily worked by middle- and upper-middle class women in urban India in contrast with garment manufacturing and domestic service which employ mostly poor women. The Indian IT industry is divided into software development and IT-enabled services (ITES).34 Typically women are concentrated in some jobs in the software industry and left out in others. Although there is a significant percentage of women in the Software Development Industry (SDI), roles appear to be gender-stereotyped, and women’s benefits are reportedly not comparable to those of men (Kelkar et al., 2002). While there are some highly-qualified women software engineers, women are concentrated largely in jobs such as Human Resource Management, Marketing, and Finance, with employers explaining this as due to women’s better interpersonal skills. Women are usually not found in more technical jobs despite the high educational qualifications of those being recruited in this sector. Less equal among more equals. Detailed discussions with women software engineers in Bangalore threw some light on the behavior of employers (Chakravarty, 2012BG). Women engineers were typically in U.P. (see NSS 2009-10) but the routes and reasons are likely to be different from those in West Bengal. In this state, the downward trend in industrial job opportunities after Independence, accompanied by large-scale in-migration of women, men and children from bordering East Pakistan (now Bangladesh), led to an unprecedented increase in the labor force under conditions of sluggish investment. In such a situation, in their frantic search for means of survival, the poor refugee women gradually drove out men in the host population who were engaged in domestic service in urban areas by offering to work for low - often no - wages. The replacement of male domestic workers by females was further facilitated by the gradual decline in inter-state migration due to the lack of employment opportunities in West Bengal. The next stage in the changing profile of domestic service was set by the increase in rural girls migrating in search of urban employment. 33 . These findings are supported by surveys conducted by Save the Children (2004, 2006 and 2009). 34 . Published data are not disaggregated by gender and so it is difficult to determine the exact number or percentage of women employees working at various levels of the IT industry. Some estimates reveal that women constituted about 21 percent of the total IT work force in India in 2001. While this percentage is believed to have increased, more recent data are not available. 29 left out of strategic work such as team-building, as it involved taking teams or clients out for a drink to build closer relationships. Women – especially those with small children – preferred to go home early or could not participate in late evening social gatherings. Women often forego – or are not offered – promotion opportunities as they cannot move to a different city or country, either taking or leaving their families behind (Kelkar et al., 2002; Chakravarty, 2012BG; Arun et al., 2002; 35 Heeks, 2001). Thus, the primacy of women’s care responsibilities persists even in this labor market segment and seriously thwarts their prospects. In software development firms women are found mostly at the lower levels and there is a gradual decrease in their presence up the ladder, especially in decision-making positions (Chakravarty, 2006; Ghosh, 2002; Kelkar et al., 2002). At higher levels, 14-16 hour days are the norm, and it is often necessary to continue work at home. While men manage these work demands by not undertaking housework, women find it taxing to manage such work as well as their homes. While flexi- time, telecommuting and so on are intended to create easier work situations for women, these are of less help as women go up the ladder, if they do. As software companies are in the formal sector it is difficult to find differences in salaries for the same work, but there is evidence of more subtle discrimination. Managements mention that women are more honest, ‘less demanding,’ ‘cheaper’ and hard-working; with the same qualifications men demand – and presumably get – a lot more than do women. Why do highly qualified women behave in the same way as the barely-educated garment workers discussed earlier? The reasons too are similar: women tend to accept less remuneration in exchange for considerations such as location and timings. Money is not the only criterion women use when they decide on a job. Employers simply exploit this socio- culturally determined timidity of women across all classes. To redress some of these entrenched beliefs and practices, the attitudes of both men and women need to be addressed (beginning in school!) to create gender-sensitivity all round and question the prevailing role models. Without such change, efforts such as establishing crèches or skilling women for higher jobs will not support career women adequately. More equal among less equals? IT-enabled Services (ITES) include call centers, customer interaction cells, back-office operations, insurance claims processing, medical transcription, and so on. These services can be provided by women or men with limited skills. The proportion of women working in ITES is much greater than that in the software development industry in India, and their situation appears to be better (Mirchandani, 2005). Based on a study of call centers in Mumbai, Gothoskar (2000) reported that the basic qualifications required were a Bachelor’s degree with good communication skills and a good American accent (Mirchandani, 2005)! Nevertheless, workers found their jobs extremely boring, lacking in creative and challenging features. Another major difficulties faced by employees in this sector is the precariousness of employment. For example, the typically small medical transcription centers, that are increasing rapidly even in the garages and basements of Indian cities, employ people ‘on training’ for short periods of six months or so. They close overnight if they do not get contracts, many because of the dearth of 35 . Arun et al. (2002) reported that some mothers in a survey in Kerala found software development jobs so stressful that they gave them up and settled for teaching in computer training centers. 30 investment in their business. A number of studies report that there has been no significant change in the salary structure in the industry over the past ten years (Chakravarty, 2012BG; Ramesh, 2004). Apart from salaries, employees may not receive other formal benefits although they may have the standard insurance and social security premiums deducted from their pay. However, there is a low incidence of discrimination against women in ITES. In the largest Indian medical transcription center, Healthscribe, 70 percent of employees and over 40 percent of team leaders and managers were female (Kelkar et al., 2002). Most employees were unaware of gender discrimination in the workplace. As the skills and education required for jobs in this field centers also are not very high, women from the middle or lower-middle class obtain the jobs easily but the jobs are not only monotonous, they have few career development opportunities. Young men aiming for rapid advancement do not remain in these jobs for long, and employers prefer married women with children who will not move away based on studies of some Indian and Malaysian firms, Mitter (2001 and 2005) pointed out that women prefer less challenging jobs when trying to balance home and work, and are often ready to accept lower pay, as discussed earlier. Again, these are strategies to survive the gender discrimination in society which affects women’s labor market behavior. Patterns of interaction between workers tend to promote gender-stereotyping of jobs in ITES firms. Call centers in Indian cities provide ample opportunity for task segregation on the basis of gender as many centers offer several distinct processes (Mirchandani, 2005). However, managers deny a relationship between gender and skills: “It’s more the personality; it doesn’t really matter about men or women,” said one manager (Mirchandani, 2005). Reaping the IT harvest. IT-enabled services have been the fastest growing export in India since the 1990’s, and expectations of growth continue.36 The sector could provide employment for educated women, offering relatively advantageous terms. While there is no doubt that software development and ITES have opened new employment opportunities with lower gender barriers for the educated middle class, the availability of and access to this employment requires access to relevant hardware – but the penetration of technology is highly restricted even within urban areas of India: 400 million Indians still have no access to electricity. Moreover, in order to use the technology, a certain level of education is essential. While both literacy and school enrolment have improved in recent years, about 60 percent of the population eligible for secondary and higher secondary education were out of school even in 2004-05; gross enrolment ratios for secondary and higher secondary education for girls were about 45 and 25 percent, respectively (Planning Commission, 2009). Therefore, one of the most important public interventions to improve the possibility of women’s employment in ITES is to ensure an increase in educational attainments of girls. Finally, ITES jobs are concentrated among better-off youth who have been educated in English; it is difficult for students educated in vernacular schools to enter this sector. This is likely to remain the case even if educational performance improves. 36 . Given the sector’s dependence on Western markets uncertainty is high. Hence, the Indian market for ITES needs to be developed. 31 3.6 Lessons from these Experiences of Job Growth for Women The three case studies above bring out a number of points that are worth highlighting. Labor market segmentation by gender is influenced by socio-cultural practices and, related, by geographies. Even in its most dynamic and ‘new’ aspects the Indian labor market upholds the prevailing socio-cultural beliefs and behaviors that ‘homemaking’ is women’s primary role. This is manifest widely e.g., in the employment of girls from an early age in the domestic service sector, the preference of both workers and employers for home-based work in the garment manufacturing sector, and the constraints to progression of women in the software development industry. Sectors such as domestic service and garment-making are considered ‘suitable for women.’ Although the latter is an important area in manufacturing and Indian exports, it is associated with women’s traditional work of sewing and tailoring, leading the sub-sector to function in some ways that are less than modern. Employers prefer home-based work because it enables them to skirt labor laws and to not deal with workers (as sub- contractors are available); and women comply because they can earn while carrying out their domestic responsibilities. If women’s work continues to be driven largely by households’ economic necessity and not by any belief in the intrinsic value of work to women and of women’s work to families and society, these entrenched practices will continue to hold sway. Whether home-based or factory-based, working conditions require improvement. To ease the dichotomy between home-based and factory-based work, women home-based workers could be encouraged and assisted to organize into producer groups, co-operatives or small/medium enterprises for better access to markets, credit, information, and improved returns to their labor. Factory-workers in garment manufacturing and similar sub-sectors also have significant needs to address including safe buildings, proper facilities, health and social protection, and some job security. While expanding this sector could serve to increase women’s employment, economic growth and exports, it is imperative to improve working conditions and enforce relevant regulations. Women face barriers to entry in male-dominated sectors, particularly technology-related work which is still seen mainly as men’s work. Industrial modernization involving technology has meant that men may displace women (e.g., in mechanized garment factories), sending women to ‘lower-end’ jobs. However, some modern work spaces have emerged for women, such as the software development industry and ITES. Women have joined in large numbers and remained in ITES which requires less education (and assertiveness); but it also pays less, has few benefits, is more precarious and has less career progression. To expand the numbers of women in such fields, increases in secondary education, greater exposure to technology, and better English language and communication skills are needed. Gender-stereotyping has led to women being considered more suited to the ‘softer’ work, for example, human resource management or client interactions in SDI, rather than to the technical, managerial or strategy-related jobs which involve long and late working hours. Hence, the numbers of women decline up the ladder. Gender-equal practices and better role models are needed to empower female employees. Women are considered dependable employees because, among other reasons, they are not upwardly or outwardly mobile. The corollary of these desired traits, however, is that women accept 32 lower pay and poor working conditions either because they are greatly dependent on the job they have, or because they want the flexibility provided by the less rewarding job. There is a disjunction between the education levels that employers seek and the work preferences of women with such education. Managers find it difficult to get women with even primary education to work, for example, in the garment sector or with secondary education for the ITES sector. While this emanates from the inadequate schooling of girls, strong related reasons are the lower wages and low status associated with these jobs relative to the expectations of educated woman candidates. Educated women (or their family members) ‘choose’ to stay at home rather than taking on low-status low-return work. Rather than provide better pay, perks or working conditions, employers have addressed this problem by training and promoting women internally into jobs that require higher competencies. Informal work could be formalized through appropriate mechanisms. Although demand for domestic workers emanates from households with formal women workers, the former have virtually no labor protection. They need minimum wages to be established, better working conditions, benefits such as maternity leave, health care, protection against sexual harassment, and some job security. Domestic workers’ associations could lobby for these, and enactment of long-pending legislation (the Domestic Workers’ Welfare and Social Security Bill) and enforcement of existing laws wouldbe useful toward these goals. Ultimately, several improvements are needed both inside and outside the world of formal work. Foremost among those ‘outside’ is the education of young girls – also to enable them to obtain appropriate information for work, to prepare and reach for the opportunities in the job market, and to be assertive in work situations. Within work, women must be encouraged and assisted to organize to gain better access to equal pay, benefits, support systems and rights, and to be seen by themselves, their families and employers as productive workers. 33 4. GENDER AND EMPLOYMENT BY STATE This chapter is devoted to a discussion of state-level variations first, in labor force participation and the correlates discussed in Chapter 2 and then in gender gaps in sectoral participation and wage rates. The situations of low-income states are highlighted. 4.1 Labor Outcomes in the States A detailed analysis of gender and labor outcomes at the state level, focusing on India’s low-income (LIS) and Special Category states (SCS), reveals a host of enigmas and opportunities also. 4.1.1 Social Aspects Women’s labor force participation is strongly related to states’ socio-cultural characteristics. Women’s LFPRs were high in hill states (e.g., Himachal Pradesh, also a Special Category state, which saw early increases in girls’ education and has high male outmigration), Andhra Pradesh, where poor women have been mobilized widely to enhance livelihoods, and in a low-income tribal state (Chhattisgarh) (Figure 4.1).37 These states maintained their lead in women’s LFP throughout the 16-year period of economic growth despite very different economic levels – high, middle and low median MPCE levels, respectively. Andhra and Himachal are the only states where both men’s and women’s work participation are high, suggesting that state-led efforts can foster more gender-equal work participation. Rajasthan, a low-income state, is among the top five states with high participation among both women and men. Figure 4.1 Gender Gap in LFPRs by state, 20011-12 100.0 90.0 80.0 L 70.0 F 60.0 50.0 P 40.0 30.0 R 20.0 LFPR Male 10.0 0.0 LFPR Female Uttarakhand Orissa Delhi Kerala Punjab Karnataka West Bengal Chhattisgarh Tamilnadu Other States Jharkhand Bihar Gujarat Uttar Pradesh Total Maharashtra Haryana Other NE States Himachal Pradesh Rajasthan Madhya Pradesh Assam Andhra Pradesh Gender Gap Statea/UTs 37 . Andhra also has a numerically large tribal population, but this is a small percentage of its total population (about 6.8 percent compared with the Indian average of 8.25 percent). 34 The states with low gender gaps in LFP have better gender parity indicators on the whole or have predominant cultures that specifically favor women’s participation in work (e.g., the hill and tribal states). Higher gender gaps are found in states with low sex ratios that are also richer, such as Punjab and Haryana, and in some of the poorest states (e.g., Bihar and Assam) where there is low labor demand in general. In some urbanized states such as Delhi, both male and female LFPRs are low because of higher incomes and educational attainments, on the one hand, and shortages of employment and stressful living conditions among the poor, on the other. Gender plays out differently in rural and urban areas. An examination of men’s and women’s work participation separately (in groups of states) suggests a negative relationship between the levels of participation of the two sexes in rural areas. This implies that attempts to enhance work in rural areas need to pay special attention to gender so that women do not lose ‘good’ jobs while men gain them, or vice versa. The negative relationship was weaker in urban areas, suggesting (again) that urban labor demand may be more gender-stereotyped (i.e., that the sexes do not readily substitute each other). While work participation among rural women is highly variable across states that of urban women is less so, being quite low everywhere. Education does not explain declines in LFP. Work force declines for young rural men were not accounted for by increased educational enrolments in several states, including four of the LIS/SCS and two important high-income states. In some better-off states, however, educational enrolment exceeded the decline in LFPRs. Among rural women, education did not compensate for work force declines in most states, and many had very large gaps. Notable exceptions, showing increase in their CPRs is Bihar, among the LIS and SCS, respectively. In essence, both supply and demand factors appear to operate for education and labor force participation, and do so independently of each other. Combined Participation Rates (CPRs) show strong gender differentials and state variations. Examining education and work participation together tells us which states have not provided education or jobs for their youth (15-24 year-olds).38 Among 15-24 year-old men, CPRs in both rural and urban areas reached higher levels than the work participation rates of 25-64 year-olds, due in part due to the recent expansion of educational enrolment and in part to attrition of the older workforce. Between 2 and 15 percent of young men were neither in education nor in work in individual states; in six states (including four LIS) the percentages exceeded 10. For young women the CPRs varied much more widely – in seven states (including four LIS/SCS) over 50 percent of 15-24 year-old women were out of school and work. Two of the worst-affected states were high-income states. State incomes clearly do not help to fill the 38 . Enrolment and LFP rates are computed using the same population denominator and can therefore be added to get a ‘Combined Participation Rate’ (CPR). In India overall, the CPR of 15-19 year-old males was around 96 percent and that of women was 66.3 percent in rural areas and 77.1 percent in urban areas, i.e., about a quarter of urban 15-19 year-old girls and one-third of rural girls of this age were neither studying nor working in 2009-10. The CPRs for rural and urban males in the 20- 24 year age-group were around 98 percent. However, females in this age-group had considerably lower CPRs, especially in urban areas. Thus, significant proportions of 20-24 year-old women were neither in education nor in the labor force. The reasons for this include the low value associated with female education; social barriers to education and employment; the primacy of marriage, child-bearing and involvement in family care; unavailability of appropriate employment opportunities; unacceptable employment conditions, and so on. In Chapter 4 we will examine the impact of some of these issues on young women’s activities. 35 gender gaps in schooling and/or work. Over the 16-year period of economic growth, the CPRs of rural/urban men/women increased in some states and declined in others, with no discernible relationship with the level of development, location, or other characteristics of states. School or work for youth: Younger women are filling the gaps. When this age-group is considered on its own, the variation in men’s LFP across states decreases, while that in women’s LFP increases. In effect, the participation of young women in work is reducing the overall gender differential in work participation, a welcome development. 4.1.2 Economic Levels State-level LFPRs decline with increasing median MPCE. This study also examined the links between household economic status and LFP at the state level using MPCE deciles and groups of these, as well as groups of states. Whether low or high in their median MPCE levels, the states had varied patterns of work force participation by MPCE. Generally, LFPRs declined with increasing ‘wealth’ – the consistent exception was the wealthiest decile of urban women among whom work participation increased (even though urban women had the lowest LFPRs) as we also saw at the national level (Figure 4.2 A-F). Different LFPR patterns by MPCE deciles were identified among the states: a ‘Bihar pattern’ in which work participation is greatly depressed and relatively flat across all deciles, especially for women; a ‘U.P. pattern’ where sharp declines are seen from the poorest to the wealthiest decile for all groups (rural/ urban men/women); and a ‘Himachal pattern’ which has a more pronounced ‘U-shape’ (high participation among the lowest and highest deciles and a strong missing middle), especially for women. There were also some unique situations, e.g., Gujarat which showed a sharp decline in the lower deciles and ‘flatness’ at the upper end; and Andhra Pradesh with an unusual ‘soaring middle.’ These two states appear to be addressing the ‘missing middle’ problem better than others. Figure 4.2 LFPRs for Rural, Urban and All Males, Females and Persons by MPCE Deciles A. All-India, 2009-10 100 Rural Male 90 80 Rural Female 70 Rural Person 60 50 Urban Male LFPR 40 Urban Female 30 20 Urban Person 10 0 All Male All Female All Person MPCE Decile 36 B. Bihar, 2009-10 100 90 Rural Male 80 Rural Female 70 60 Rural Person 50 Urban Male LFPR 40 30 Urban Female 20 Urban Person 10 0 All Male All Female All Person MPCE Deciles C. Uttar Pradesh, 2009-10 100 90 Rural Male 80 Rural Female 70 60 Rural Person LFPR 50 Urban Male 40 30 Urban Female 20 Urban Person 10 All Male 0 All Female All Person MPCE Deciles D. Himachal Pradesh, 2009-10 100 90 Rural Male 80 Rural Female 70 60 Rural Person 50 LFPR Urban Male 40 30 Urban Female 20 Urban Person 10 0 All Male All Female All Person MPCE Deciles 37 E. Gujarat, 2009-10 100 Rural Male 90 80 Rural Female 70 60 Rural Person LFPR 50 Urban Male 40 30 Urban Female 20 Urban Person 10 0 All Male All Female All Person MPCE Deciles F. Andhra Pradesh, 2009-10 100 Rural Male 90 80 Rural Female 70 Rural Person 60 50 Urban Male LFPR 40 Urban Female 30 20 Urban Person 10 0 All Male All Female All Person MPCE Deciles Source: Raveendran 2012 BG:55-58 While poor households have high women’s LFP, the poorest states have low work opportunities for women. An analysis by per capita State Domestic Product (SDP) in which states were grouped into low, middle and high SDP groups (and similarly MPCE deciles) showed that while rural men’s CPRs did not differ across SDP or decile groups, the decline in LFPR from low to middle and high MPCE groups was steeper with increasing SDP (i.e., from low to middle to high SDP groups). The enrolment ratios of rural males in the nine groups were clearly the inverse of the LFPRs, and the percentage of inactive rural males did not vary. Rural women presented quite a different picture. The decline by MPCE groups occurred in all three types of states and got steeper as states’ wealth increased. In essence, the poorest states had the lowest educational enrolment and work participation among rural women, while the middle and best-off states had higher levels of these, particularly of work participation. As we saw at the national level, the highest LFPRs occurred amongst the poorest households in each state group. The poorest states clearly have very low opportunities for women’s work, and also low educational enrolment of girls. This disheartening combination is manifest in the high percent of inactive rural women in the group of poor states. 38 Women’s work participation does increase with SDP – up to a point. While the total labor force increased by about 95 million in the 16-year period spanned by the four Employment-Unemployment surveys, LFP rates declined over this period (i.e., the increase in jobs over the period did not absorb the increase in the working-age population). Further, as mentioned earlier, female LFP fell both proportionately and absolutely between 2004-05 and 2009-10. What, if any, was the relationship between this and the states’ economic levels? There was some association between female LFP and SDP, both current and with a lag (from SDP in 2004-05 to LFP in 2009-10) (Joseph and Andres, 2012BG). Female LFP rose linearly with SDP up to a level of about INR 13,000 (in 2009-10 at 1993-94 prices).39 However, the percentage changes were not related, demonstrating that growth per se did not enhance women’s LFP. This was confirmed by analysis of the data for the period when LFP increased (1999-2000 to 2004-05), and it held across rural and urban areas in both periods. 4.1.3 Sector Differences States vary greatly in male and female participation levels in the four largest sectors of the Indian economy (Agriculture, Manufacturing, Construction, and Trade and Repair Services) and a fifth group, all ‘Other Industries’ combined. Table 4.1R and 4.1 U shows the variation in gender gaps in the five industry groups in rural and urban areas at state level. India as a whole has more female than male workers in Agriculture, and this is true in most states. In essence the supply of female labor is high in the primary sectors and there is evidence of increasing female shares in several states – a real and present opportunity to enhance women’s wages and productivity in this sector. On the other hand in states where rural women are highly (i.e., over 90 percent) confined to Agriculture (e.g., Chhattisgarh among LIS), it is advisable to diversify employment opportunities. Manufacturing in a few states – as diverse as Kerala and West Bengal – has relatively high percentages of rural women workers, suggesting that in other states (the majority) the share of women in the sector (and the overall low 7.4 percent in 2009-10) could be increased. The situation is somewhat better among urban women workers overall (27.5 percent) and across a number of states including six of the eight LIS (e.g., over 39 percent in U.P, M.P. and Orissa; the exceptions are Bihar and Jharkhand, where women’s work is depressed as a whole). There is considerable scope for improving women’s participation in this sector. Construction has been a growing sector for women’s employment and still has vast opportunity, particularly in urban areas where women’s shares have been depressed. Among LIS, the tribal states of Chhattisgarh, Jharkhand and Odisha have higher shares. Trade and Repair Services have a very low overall share of rural women workers (2.3 percent), but a higher percentage of urban women workers (10.0). However, the latter share has been declining and there are wide state variations. For both reasons, significant attempts are needed to enhance women’s participation in this sector. It is an important sector for urban women in Assam, Bihar and Jharkhand among LIS, and in the other Northeast states (which are Special Category States). Women’s participation has been growing in the Other Industries group but the share of rural women workers in these is only 5.5 percent. In urban areas, the 39 . Nine of 20 states included in this analysis were below an SDP of INR 13,000. 39 share of women workers in Other Industries varied from 27 to 75 percent (average 44 percent), the more urbanized states having higher shares. 4.2 Wage Rates and Ranges Gender wage gaps and women’s risks are high. In general, compared with men, women do work that earns less (4 to 40 percent of the average per capita income) and is riskier – more insecure, seasonal, and without benefits or progression, although entry barriers may be lower and flexibility higher. Even in agriculture where women predominate, their market wages are a fraction of men’s in all states – for example, 34 percent in Tamil Nadu, 54 percent in Maharashtra, 55 percent in Andhra, 58 percent in Kerala, and 83 percent in Uttar Pradesh. While this appears contrary to the higher status of women in the South, it reflects the availability of women for agricultural work in these areas (and/or their relegation to low-paid agricultural tasks). While the market wage for male agricultural laborers is less than the statutory minimum wage in two states, for women this is so in 11 states (Kanchi, 2010). Average wage rates for rural women are considerably lower than those for men in all work categories, and state-level variations are wide. For example, the male/female ratio in the average wage rates for casual rural workers was 1.77 in 2009-10. In Kerala, the Indian state most renowned for gender equality, the average wage rates of casual male workers was 1.9 times that of women. For urban workers, states with strong anti-female biases (e.g., Haryana) as well as those with greater gender equality (Kerala, Tamil Nadu) had male/female ratios over 2 for casual workers, and several states approached a ratio of 3 among regular workers. It appears that states with low wage rates had low increases between 2004-05 and 2009-10, and only a few states with higher wages had high growth rates (most had low increases). The LIS Bihar stands out for the low level and negative growth of wages for urban women casual workers in this period. While gender equality in wage rates appears to be greater among formal workers, the numbers and shares of formal workers among women are low. The male/female ratio in average wage rates for formal rural workers was 1.37 and for formal urban workers it was 1.17. Most states also had lower average wage rates for women compared with men in all other urban work categories. The gender gap in wages is a significant reason for the differential distribution of women and men workers in industries. Factors influencing wage rates. Formal work and technical qualifications are the most significant correlates of wage rates for rural women (Raveendran, 2012BG). Technical qualifications are a less important determinant of wage rates for urban than for rural women, but age, principal worker status, formal work, and education are more important correlates. Formal work is also a more important determinant of wage rates among urban women than urban men. The correlates of wage rates did not change markedly when state and district variations were considered. Other research has found that the difference between male and female wages was partly explained by their characteristics, but there was a sizeable residual ‘discrimination effect’ especially in urban areas. 40 Table 4.1R Classification of States by Gender Gaps (M-F) in Shares of Workers in Different Occupational Groups in Rural Areas, 2009-10 Gender Gaps AFF MFG CONS TRS OTH Rural Less than - 20 Jha, HP, Mah, WB Del Pun, UK, Ass, Guj, Har, UP -20 to -10 TN, Odi, Raj, OS, Chh, AP, Bih, Ker -10 to 0 Kar, ONE, MP MP, ONE, Odi, TN, Raj, Chh, ONE WB Jha, AP, Bih, Kar, Ker 0 to 10 Del Ass, Guj, Har, HP, WB, OS, Odi, TN, AP, Ass, Bih, AP, Ass, Bih, Mah, Pun, Raj, AP, Ass, Guj, Guj, Har, HP, Guj, Har, Kar, UP, Chh, UK, OS Mah, Kar, MP Kar, Ker, MP, Ker, MP, Mah, Mah, Odi, Pun, Odi, Pun, TN, Raj, TN, UP, UP, Jha, Chh, WB, Jha, Chh, UK, OS UK, ONE, OS 10 to 20 WB Del Jha, UK, Pun, Del HP, Raj, ONE UP, Bih, Har, Ker Above 20 HP Del AFF = Agriculture and Fishing; MFG = Manufacturing; CONS = Construction; TRS = Trade and Repair Services, and; OTH = Other occupations AP – Andhra Pradesh Ass – Assam Bih – Bihar Guj – Gujarat Har – Haryana HP – Himachal Pradesh Kar – Karnataka Ker – Kerala MP – Madhya Pradesh Mah – Mahrashtra Odi - Odisha Pun – Punjab Raj – Rajasthan TN – Tamil Nadu UP – Uttar Pradesh WB – West Bengal Jha – Jharkhand Chh – Chhattisgarh UK – Uttarakhand Del – Delhi ONE – Other North East States OS – Other States Source: World Bank staff based on Raveendran 2012 BG 41 Table 4.1U Classification of States by Gender Gaps (M-F) in Shares of Workers in Different Occupational Groups in Urban Areas, 2009-10 Gender Gaps AFF MFG CONS TRS OTH Urban Less than - 20 Bih, Raj Del -20 to -10 UK, HP, Odi, Pun, AP, Ass, TN, MP, Bih, Guj, Har, TN Odi, UP, WB, Ker, Mah, Pun, WB, Jha, UK, OS -10 to 0 AP, Ass, Guj, Jha, HP, Kar, Ker, Raj, Odi, Chh ONE Ass, Kar, MP, Har, Kar, ONE, Chh, UK, ONE UP, Chh MP, Mah, Chh, UP, WB 0 to 10 Ker, Del, OS Bih, Guj, Har, AP, Ass, Guj, OS, AP, OS, Ker, AP, HP, Raj, Mah, Pun, Jha, OS HP, Kar, MP, Mah, Jha, TN, ONE Mah, TN, UP, WB, Jha, ONE 10 to 20 Del Bih, Har, Ker, Ass, Bih, UK, Odi, Pun, Raj, UK Guj, Har, HP, Kar, MP, TN, WB, Raj, Chh Above 20 Odi, Pun, UP, Del AFF = Agriculture and Fishing; MFG = Manufacturing; CONS = Construction; TRS = Trade and Repair Services, and; OTH = Other occupations AP – Andhra Pradesh Ass – Assam Bih – Bihar Guj – Gujarat Har – Haryana HP – Himachal Pradesh Kar – Karnataka Ker – Kerala MP – Madhya Pradesh Mah – Mahrashtra Odi - Odisha Pun – Punjab Raj – Rajasthan TN – Tamil Nadu UP – Uttar Pradesh WB – West Bengal Jha – Jharkhand Chh – Chhattisgarh UK – Uttarakhand Del – Delhi ONE – Other North East States OS – Other States Source: World Bank staff based on Raveendran 2012 BG 42 5. INTRA-HOUSEHOLD AND OTHER FACTORS AFFECTING WOMEN’S LABOR OUTCOMES 5.1 When Women Do Not Work40 Having explored the key correlates of gendered labor outcomes and the labor market across space, time and sectors, we now examine the reasons for why there are such large gender differentials in the labor market in India and why women, in particular, have such poor labor outcomes. This chapter first summarizes the social issues underlying women’s work participation and then discusses a key strategy – that of organizing women into Self-help Groups – that has been used in India to improve women’s labor outcomes directly as well as through enhancing their agency in the household and society. A caveat: women workers are undercounted. One explanation for India’s low female labor force participation arises from the definitions and methods used by the labor force-enumerating efforts, the NSS and the Census. NSS data (which are the more frequently collected and used, including in this report) may not capture informal work adequately, particularly women’s work, as a result of either investigator or respondent bias (or both). Informal workers are difficult to access and their work difficult to assess as it is carried out in a variety of locations, (Sudarshan and Bhattacharya, 2008:4; NCEUS, 2007:77)41 may involve multiple activities, be seasonal, part-time and poorly paid,42 with payments being delayed and/or made to someone other than the worker. Women’s own perceptions of their work and economic roles are low, especially if their work is carried out at home alongside other domestic duties Sudarshan and Bhattacharya, 2008: 9,15; NCEUS 2007: 81. Smaller surveys that probe more deeply can capture missing work and missing women, and time-use studies provide insights into the mix of activities that women do and how they manage them, but such efforts are few and far between (NCEUS, p.77. Even better-off educated women are affected by under-counting. For example, women’s work in providing and nurturing social capital for family-run businesses is rarely identified as work and difficult to measure (Sudarshan, 2012BG). Family businesses are numerous and widespread in India, found across a wide spectrum of industries and services and in economically-important sectors such as gems and jewelry, yet women’s roles in them generally elude surveys. Women’s contributions to the Indian economy are also under-counted insofar as non-SNA activities (which include most of women’s care work) are excluded. Counting these activities enhances estimates of women’s contribution to India’s GDP by at least a third, from 19 to 25 percent (Raveendran, 2010). 40 . This chapter incorporates information from Sudarshan (2011BG and 2012BG) and Urzua and Palloni (2012BG). 41 The NCEUS Report points out that “in 2004-05 only half of the total non-agricultural workforce and as little as one-third of the women workers worked in conventionally designated workplaces. This implied that about 89 million men and only about 10 million women workers had a conventional place of work,either of their own or belonging to the employer. The enterprises in the unorganised sector are mainly unregistered units. Only 40 per cent of the total and 18 per cent of the women workers in the unorganised sector had a designated place of work” (NCEUS, 2007: 77). 42 See Government of India’s report on employment (GOI, 2010b). See also Chen et al., 1999: 603-601. 43 5.2 The Social Underpinnings of ‘Not Working’ The household is ultimately the unit of decision-making about women’s work. Decisions are strongly related to household earnings (and its resultant ‘status’) and usually beyond women’s control. Gender asymmetry is clearly evident in the Indian context where, in the vast majority of households, domestic work is invariably ‘women’s work’ while earning is ‘men’s business.’ Women often enter the labor force on adverse terms including low activity status, poor earnings, gender stereotyping of jobs and being confined to inherited occupations. In India many occupations are associated with caste and passed on from one generation to the next with both entry and exit barriers. While preserving knowledge and skills, this can lock women and men into low-skill, low-return occupations. The higher than average levels of women’s work participation among SCs, STs and the poor are related to these phenomena as also the fact that they are amongst the poorest of the poor..43 The strong association of work and family also entrenches gender biases – girls frequently can only enter occupations held by their mothers or mothers-in-law. In this context there is a clear need to improve the efficiency and returns of such ‘traditional’ occupations, foremost among which are Agriculture, and Handloom and Handicrafts. Out-sourcing to home-based workers in many areas of Manufacturing has, to some extent, broken the chains of inherited work, as has drawing women into factories and work- sheds, all of which have enabled a wider acquisition of skills and even the dropping of caste identities. Women’s work depends to a great extent on the acceptance by communities of the specific work that women do and, indeed, of them working at all. Women can aspire to work within a limited physical and ‘activity’ space. They work if income is desperately needed by their households (as amongst the poorest), or if the net value added (financially and socially) by their work exceeds the costs of ‘care time’ foregone. Greater support for women to work does come, when their economic contributions are necessary or valued by the household, or its social status enhanced. Care work is first, paid work secondary. Indeed the responsibility that women bear for care work is a prime reason for them not being in the labor force, as in many other contexts.44 Indian society strongly ordains the roles of both spouses – women as wives, mothers and homemakers, and men as breadwinners. These gendered roles result in women’s decisions not to work being related to reproduction, child care, household needs and the opportunity costs of working outside the home. 45 Social norms in India still by and large militate against men sharing in care work, and household 43 . See NCEUS 2007:8 44 . For the unorganized sector, theIntegrated Child Development Services ( ICDS) program is available throughout the country in both rural and urban areas, and the Rajiv Gandhi National Crèche Scheme in limited areas. The ICDS program provides basic nutrition, health and pre-school inputs to children under six, but its timings limit the extent of child care provided to women who work a full day. The timings may also be unsuited, for example, to agricultural work. There have been several policy pronouncements for child care., notably the National Policy for Children, 1974, National Policy on Education, 1986, Shramshakti Report of 1988 which drew attention to the impact of the lack of child care facilities on women’s work participation, health and also the well-being of small children. According to the Report of the Working Group on Child Rights for the 12th Five Year Plan (2012-2017), inspite of the recognition of the need for early child care, investment in this area has been low (GOI, 2011d:11). According to the Working Group report, there have been policy gaps and poor implementation and insufficient priority has been given to this area (GOI, 2011d:11). 45 . There may well be cases of women deciding to stay at home to look after younger siblings so that older female children can go to school. 44 arrangements need to be modified when women enter the market economy. Unfortunately, the arrangements are not always adequate and can result in a heavy ‘double burden’ on women. The inadequacy of labor-saving utilities and support facilities also ties women to the home or exacerbates the double burden. Women’s and families’ status are considered enhanced by women not working. This is particularly true of those who have moved upward to acquire middle- or high-income status – hence, the ‘missing middle.’ The higher the disparity in education between husband and wife, the lower the probability of the latter being in the labor force; and women with regular wage/salaried spouses have the highest likelihood of not being in the labor force, and even more so when male earnings are secure. Women’s work decisions are secondary to men’s and may be related particularly to the latter’s work activities and patterns. Women often move between wage labor and unpaid work on account of situations faced by their spouses. When men obtain work in the market economy, women may carry out home-based work linked to the local economy. In rare instances, men manage local economic tasks and women seek work in the market economy. Both men and women working in local economies is often limited to gender-specific roles, while both may work in the market economy at higher levels of education and income. Life-cycle matters. Marriage and child-rearing result in lower LFP among women and higher LFP among men. As Figure 5.1 shows, the labor force participation rate among men stabilizes in the 25-29 year age- group and, among women, in the 35-39 year age-group. The first two five-year age-groups  15-24 year- olds  have low LFPRs because many of them are still in education and/or have not yet entered the labor market. In India, women still marry young and are expected to bear children in quick succession – hence, many in the 15-24 year age-group have transitioned from education to marriage and childbearing and are not in the labor force for these reasons. Young brides and mothers are often secluded so that going out to work is virtually taboo. Even women who have worked before marriage or motherhood may withdraw from work during the child-rearing years and seek to re-enter the labor force thereafter. To absorb these women, flexible work opportunities and supportive policies, especially easy access to information, capital, support services and perhaps new skills, are needed by informal workers, and flexible work arrangements during the child-rearing years and options to resume careers by formal workers. 45 Figure 5.1 Male and Female Labor Force Participation Rates by Age Group, 2009-10 96.9 99.0 99.3 99.2 98.5 96.5 92.3 100 90 78.6 76.8 80 70 60 69.8 70.9 71.2 70.6 Male LFPR 50 37.0 64.6 66.0 64.1 40 53.1 52.0 Female 30 40.2 40.1 40.1 40.9 33.3 33.9 total 20 32.2 27.5 26.0 27.3 10 0 15.3 Age Groups Source: World Bank staff Gender norms. The paramount explanations for India’s abysmally low female LFP, and for the U-shaped relationships between female LFP and either education or standard of living are the strict norms that dictate gender roles and behaviors, including seclusion and low investment in girls’ education and skills. The explanations that emerge from regression analysis for the loss of jobs among women between 2004-05 and 2009-10 are only partial. First, only a small proportion of the reduction was explained by 15-24 year-old women not entering the labor market because they continued in secondary or tertiary education. Another small section did not do so because increased household incomes caused decreases (in need increased status or both). By and large, women’s work participation declined mostly in rural areas and among the poor, suggesting that the lack of labor demand may be a main cause of the loss of jobs among women. Unsuitable jobs. Indeed, there are supply-side reasons for women not working including the lack of opportunities suited to women’s gendered roles, such as part-time work. Even when women consciously opt out of work this may be due to the unsuitability of work contents or conditions, particularly for educated/better-off women who have the choice not to work. Women’s work participation is also constrained by workplace practices such as gender-stereotyping of jobs, recruitment biases, poor remuneration, lack of benefits, distant locations, inadequate safety, and so on. Data limitations. A few of the explanations for poor and declining female LFP that have been brought out by qualitative studies are difficult to ascertain with available macro data, including the lack of jobs suited to women’s endowments or ‘preferred’ by them; increased demands of care work on women’s time (in part because of increases in daughters’ schooling); and decreased mobility among women due to regressive social norms, increasingly inadequate support facilities, and declining safety in public places. The lack of fit between jobs that may be generated by economic growth (creating demand for skills) and women’s endowments is due to low educational investment in girls and results in gender- imbalanced employment growth, leaving the majority of women workers in unskilled casual labor or self-employed in low income-generating activities. Labor market explanations include gender- 46 stereotyping of jobs which work partially for women’s LFP but mostly against it, as do other forms of discrimination against women. The need is for deliberate interventions to enhance women’s employment. The overall experience of the 1993-2010 period shows that we cannot assume that economic growth will improve women’s labor force participation in either quantity or quality. This means that India must make deliberate interventions for women to participate in further growth. Reducing gender disparities in labor outcomes, increasing women’s contribution to GDP and growth, and the economic empowerment of women require actions – from policy to practical interventions – to increase their presence in non- traditional and high-value activities. The actions range from improving education and skills among girls to increasing access to credit and financial services, expanding jobs and support facilities to enable women to take on regular paid work outside the home, and loosening the social norms that constrain women’s work participation. An important policy prescription that arises is to create more part-time jobs for women of different levels of education. These could provide work during the 15 to 20 years that women spend rearing children, and thereby enable careers. More decentralized work and more diverse options need to be available, particularly for lower income groups. For home-based producers, skills and organization that could help enhance productivity and enterprise development need to be expanded widely. These economic efforts must be accompanied by better social protection measures, enforcement of labor regulation, drudgery reduction and information provision. 5.3 Strengthening ‘Agency’ to Enhance Women’s Work46 Organizing women is a key strategy. One approach to increasing women’s work participation that has been spreading through India over the past two decades has been that of organizing women (in the informal sector) into local-level groups (Carr, Chen and Jhabvala, 1999). Women’s groups have been seen as an instrument to enhance women’s agency or empower them by providing them information, credit, access to social and economic services, and increasing their social capital to meet other strategic needs (Harris, 2007). Since the 1980’s, thrift (or savings) and credit groups set up by NGOs and/or government programs (in some instances with support from external agencies such as IFAD and the World Bank) have facilitated access to microcredit to improve income-generating activities, food security, access to land for cultivation, skills, and so on. Self-help groups (SHGs) as most are now called have increased women’s access to finance, arguably more for consumption smoothing and retiring high- cost debt than for enhancing productivity (Shah, Rao and Vijay Shankar, 2007). Aggregating into Village Organizations and Federations or Producer Groups across villages, SHGs have worked together to strengthen their members’ economic and social activities. An ‘SHG-Bank Linkage Programme’ supported by NABARD, the National Bank for Agriculture and Rural Development, has expanded over the past 20- odd years, enabling links between SHGs or their Federations and commercial banks to help the women’s organizations obtain larger loans for productive purposes. In some cases, public or private intermediaries have provided training and support services to the organizations. The National Rural 46 . This section incorporates material from Dand and Nandi (2012BG). 47 Livelihoods Mission, initiated in 2011, is building on a series of earlier programs using SHGs to strengthen human, social and economic capital among poor women. An unfinished agenda. There are now over six million women’s SHGs across the country, each with 10- 20 members. If women were not members of more than one SHG, if groups were distributed evenly across the country’s states and villages, and if all SHGs were functioning productively, this number would most likely cover the estimated 70 million poor rural households in India.47 Unfortunately, none of these three conditions obtain, pointing to some key aspects of the large agenda before SHG-based programs. The concentration of SHGs in the southern states emphasizes the need for the northern and eastern low-income states to improve their SHG-related efforts. 48 In addition to credit, a great deal of non-financial support is needed to make SHGs (or the relevant clusters) into fully productive organizations; and SHGs need to be seen as primary producers’ organizations rather than as business for the intermediary organizations that promote them (which partly underlies the duplication of membership). In the economic arena, SHGs need to help poor women to diversify their sources of income, particularly by obtaining skills and adopting new income-generating activities, and to increase their productivity. Stronger federations and producer organizations are needed to up the game for women left in low-productivity activities. Productive borrowing. One fairly ubiquitous success of SHGs has been to reduce the dependence of the poor on money-lenders, enabling borrowing at lower interest rates. There is evidence also that SHG members shift over time from using group credit for consumption needs (such as food, education, rituals, dowry, etc. – purposes for which banks do not lend) to productive purposes; but there is little detailed information on how this happens or could be made to happen.49 There is an overall need to provide SHGs (and their promoters) with clear guidance on organizational management, skill and capacity development, and the wide range of tasks inherent in productive activities. 47 . SHGs have not ‘caught on’ as well in urban areas except in some states such as Kerala. This is in part because the promot ing programs either did not have urban counterparts (e.g., the program for Development of Women and Children in Rural Areas) or were late starting in urban areas (e.g., the Swarnajayanti Shahari Rozgar Yojana, and the National Urban Livelihoods Mission). At a policy level it has been assumed that the presence of banks in urban areas makes access to finance easier for the urban poor, but this is not always the case. 48 The expansion has not been uniform across States with the Southern States having a much larger presence of the SHG movement in terms of absolute numbers of SHGs, the bank credit extended to them as well as the coverage of poor persons (Dand and Nandi, 2012BG:19). According to Dand and Nandi, 2012BG: 27, the pan India geographic spread of SHGs has been lopsided with the 9 states of the Northern region covering less than 20% households as compared to the Southern region comprising of 7 states and 2 UT’s where the coverage has been almost 75%. The intermediary regions remain the states o f West Bengal, Mizoram, Tripura and Maharashtra where the coverage has been between 50-75% and in the rest of the states the coverage has been between less than 50% but more than 75%. 49 Dand and Nandi, 2012BG :43 report that the Kudumbashree experience of Kerala has in fact several examples of large transfer of assets in women’s neighbourhood groups which have enhanced livelihoods. A case in example is that of banana cultivation for 30,000 women farmers of Trivandrum district, for which investments of Rs. 40 crores were leveraged from National Horticulture Mission and Panchayats while Kudumbashree put in Rs. 1.5 crores. However, the convergence did not come easily. Similar investments for mushroom cultivation in Allepey and pineapple in Kochi districts have been made to enhance livelihoods. Such examples illustrate the potential of livelihoods strengthening through women’s groups by transfer of resources to women’s groups. 48 Agency in the community. The instrumentality of SHGs in helping to deliver financial services (and some social services) has increased women’s involvement in public management and contributed to increasing women’s visibility and ‘mobility’ in more ways than one. Indeed, SHGs are credited with having increased women’s political participation, particularly in panchayats to which women SHG members have been elected on account of the confidence and prominence they have gained as group leaders.50 However, SHGs’ internal management is not necessarily democratic – many are led by more educated or higher status women who do not share power and may usurp most of the groups’ benefits. 51 Although informal in nature, SHGs would benefit from more assistance for efficient and democratic management (of the organization, its activities and finances) to reach their goals sooner and broaden their achievements. While changing the existing power structure in villages clearly cannot be a goal of SHGs, they must at the least change the dynamics between poorer and better-off women members, also working to reduce caste divides among them. Despite early successes – such as the Anti-arrack Movement which was spearheaded by women’s groups and addressed a ‘felt need’ of women in Andhra Pradesh, SHGs have not been very successful in addressing many of women’s practical and strategic needs. These include information, skills, functional literacy, health care, child care, and reduced domestic violence and obnoxious public-space behaviors. This is perhaps an extension of their limited success in collective efforts even in the economic sphere (such as group enterprises or marketing). There are many other social justice issues that also need to be addressed including investment in girls, elimination of dowry, and treating elderly women and widows with dignity. Reducing poverty. The jury is still out on the ability of SHGs to lift households out of poverty, which is one of the key outcomes sought by SHG-based programs. Not only are annual savings low among groups and access to bank credit limited and often delayed, but (in addition to SHGs being least prevalent in the poorer states) ‘the poorest of the poor’ and low caste women may not be members (or active members) of groups.52 On the other hand, membership in two or more groups may increase the 50 Almost 60% of the savings and credit groups of Swayam Shikshan Prayog, an NGO based in Maharashtra, reported that they were considered a strong lobby and had a voice in selection of candidates and hold discussions with them on women’s issues and priorities. Parthasarthy links the exercise of democratic values and rights and responsibilities by women in panchayats through the experiences gained by being in the groups (Parthasarthy, 2005). See Dand and Nandi, 2012BG: 54 for more details. 51 Dand and Nandi, 2012BG:53 draw attention to a Nirantar study which was conducted in 2007 and covered 2750 micro-credit based SHGs across 16 states, which says that 69% of the women who were getting the opportunity to come into leadership roles were literate, and literacy emerged as the single most important determinant of power and leadership in the groups. Women who came into leadership gained access to many more opportunities and resources as compared to members of the groups. Leaders of the groups had a much greater opportunity to participate in capacity building inputs and thus also had better access to information. For 65% of groups, it was the group leaders who had participated in most training programmes. Group leaders also had more access to the resources. 46% of the large loans were availed of by the SHG leaders alone, even though they were only 13% of the total membership. It is evident from the findings that literacy creates access to leadership which in turn leads to access to other opportunities, such as credit and capacity building. 52 Dand and Nandi, 2012BG quote from a Planning Commission document which says that “The focus on credit provisioning for micro-enterprises has increased but the challenge in the micro credit movement is to reach out to the poorest and most vulnerable and marginalized populations like SC/STs, single women and other socially excluded communities. Many of these SHG groups have low levels of credit absorption, low skill base and low asset base, and find it hard to create economic enterprise. At the heart of the problem lies the need for appropriate institutional mechanisms to address illiteracy, lack of investment, poor credit worthiness, mobilization, and other structural exclusions. These will need to be addressed to realise the vision of financial inclusion.” http://planningcommission.nic.in/aboutus/committee/wrkgrp12/wcd/wgrep_women.pdf 49 risk of indebtedness, and thereby of households slipping deeper into poverty. Under these circumstances, building assets is a moot point; and a most troubling dimension from the point of view of women’s agency is that many activities within SHGs (including thrift, repayment, multiple borrowing, and income generation) can occur at the cost of women meeting their own needs, for example, for food or health care. Achieving ‘economic inclusion’ by engaging women in labor market activities that provide good returns evenly and sustainably without compromising costs is by far the most important agenda for SHG proposals. Self-employment. Since 2011 the National Rural Livelihoods Mission (NRLM) has been working to enhance livelihood opportunities, particularly for poor women, in addition to providing access to financial services through SHGs. As described earlier, SHG programs have been relatively successful in mobilizing and organizing women and, in some states, encouraging their political participation in panchayats. However, important areas for improvement are: ensuring that SHGs are robust in terms of the social capital developed and use of the financial capital they access; improving the productivity of ‘traditional’ activities and greatly expanding non-traditional livelihood activities; including the ‘poorest of the poor’ and developing viable livelihood options for them; monitoring improvements in women’s/households’ economic status, particularly their rise out of poverty; and addressing women’s strategic needs through collective action. Power at home.53 The extent to which participation in SHGs has empowered women within their own households also remains a question.54 Anecdotal evidence suggests, on the one hand, that women are Notwithstanding the problems faced by SHGs and the fact that in many areas the poorest of the poor have been excluded, there is evidence to show that in some areas, there has been positive impact on the nutritional levels of women as also an improvement in school attendance of daughters. Dand and Nandi, 2012BG:44 report that there is overwhelming evidence that SHGs have contributed to consumption smoothening which in case of poor households has led to improved food security. See also, Srinivasan,N. (2012): 103. The micro finance sector report quotes several studies which have measured impacts of SHGs on Food security. Naila Kabeer and Helzi Noponen’s study (2004), found that participants in PRADAN were less likely to experience food shortages, had better access to clear drinking water, and had higher levels of awareness of health and family planning issues, such as causes of malaria and diarrhoea. Sh and Liu (2009) found strong positive correlations between micro- credit and nutritional intake. New and existing clients experienced improvements in nutrition through an increase in protein consumption, while new clients also benefited from greater caloric intake. For new SHG participants, there is a 9 per cent increase in caloric consumption and a 17 per cent increase in protein consumption. For old participants, there is an 8 per cent increase in protein consumption but no impact on caloric intake. This evidence suggests that microfinance can help diversify consumption and enable families to access higher quality foods. (http://info.worldbank.org/etools/docs/reducingpoverty/case/82/fullcase/India%20SHGS%20Full%20Case.pdf 53 This section has drawn from Burra (2012 BG) 54 There is some evidence that organized and mobilized women’s collectives have gained, not only economically, but also with respect to issues such as more autonomy within the household, more control over decision-making and less domestic violence. The study by Murthy et al of the UNDP-supported South Asia Poverty Alleviation Programme in Andhra Pradesh found that only 45 percent of members reported domestic violence as against 74 percent of non-members. The incidence of wife-beating was significantly lower for members at 9 percent as compared to non-members, where it was 29 percent (Murthy et al., 2005:90-91; Rajivan, 2005: 157). The incidence of violence at the workplace was also significantly lower for members at 29 percent as compared to 62 percent for non-members (Murthy et al., 2005: 91). Although, not planned for not envisaged, both practical and strategic needs of women were addressed. But there is contrarian evidence as well: Kabeer reports from Bangladesh that violence had increased against women in some cases because they were earning and were able to stand up to their husbands but it had also decreased in other cases because they could now contribute to household budgets (Kabeer, 2010). 50 more appreciated when they have access to money and the means for better livelihoods, and that they have broken barriers such as the gendered division of labor. On the other hand, resentment of their better ‘status’ appears to result in their continuing to be marginalized in important household decision- making and in being victims of domestic violence. Clearly, social empowerment does not necessarily follow financial improvement and there is greater scope for the enhancement of women’s voice and agency through SHG programs. Patriarchy lives and reigns within most Indian households and must be challenged both from within and ‘without’ in order for its grip to be loosened. One of the greatest hurdles that remain is to change the ‘low value’ ascribed to women – even those who work and earn – fairly widely in Indian society. In the end, there is little doubt that, of existing approaches to women’s economic empowerment, the self-help group is the most widespread and promising. But there are clearly many needs for SHGs to achieve even the core goals they are ascribed – financial inclusion, livelihood improvement and agency for poor women – leave alone the myriad others they seek to attain. Moreover, while organization and agency are key needs for women’s social and economic empowerment, ‘voice alone’ is not enough. In the next chapter we shall discuss two other important needs for more gender-equal productive work – ownership of or access to land, a need faced by the bulk of women workers who are agriculturists; and access to infrastructure to facilitate work participation in terms of enabling work directly (e.g. through better transport, electricity, etc.) as well as indirectly by reducing the burden of household care work. Whether the decrease in domestic violence happens because men realize the power of the collective behind the women or because of gender sensitization of men as a result of project activities or because of the income women bring in, it is not axiomatic that domestic violence can be reduced through programs and policies. 51 6. LAND, INFRASTRUCTURE AND WOMEN’S WORK 6.1 Access to Land and Productive Assets55 Given that the majority of India’s workers, both men and women, are in agriculture, the conditions of work in this sector are of paramount importance in terms of continued participation as well as returns to their labor. Although several ‘conditions’ are involved, including access to water, energy and equipment for irrigation, inputs such as fertilizer and seeds, credit to purchase all these, markets, and so on, land is perhaps the most fundamental of these – a sine qua non of agricultural work and productivity. Also fundamental is access to infrastructure such as roads, electricity, water and sanitation, and so on. This chapter examines the relationships between these aspects and labor outcomes for women and men. There is grave inequality in land-holding in India, with the poorest – small and marginal farmers – owning only a small proportion of the total land area. While over 60 percent of the agricultural work force is female and 84 percent of rural female workers are engaged in agriculture, women own only about 11 percent of cultivable land (FAO, 2011; Kelkar,2013a).56 In a study conducted in Karnataka in 2009-10 by Swaminathan et al.,2012 of 4110 households, women’s shares of land decreased as land- holding increased – e.g., women owned 18 percent of the land in marginal farm households but only 11 percent in medium and large farm households. (This is reminiscent of the lower work participation of women in higher income groups.) Gender equality was greater in the ownership of other assets such as livestock, consumer durables and jewelry, but livestock ownership is low among small land-holders! While the numbers of female-headed households and operational holdings in their names have increased over time, the vast majority is landless, and the average size of women’s holdings is lower than men’s (UN Women and RDI, 2011). Pressures on land add to the complexities of women’s access to it, which include problems arising from location, mode of acquisition, the type and extent of rights, and the larger social and political economy that shapes state policies and priorities within which claims are negotiated (Rao, 2011a). Even when women do own land they may not control agricultural decisions pertaining to it.57 They face many barriers to accessing inputs (including labor) and finances, which constrain their working the 55 . Sections 6.1 to 6.3 are substantially based on Rao (2012BG). 56 This issue was discussed by the Planning Commission Working Group on Disadvantaged Farmers including Women where it was pointed out that “Gender is a major source of social disadvantage. Few women own lands in their own right due to male bias in transfer of land by families, the state and in the functioning of markets…. Also few women have the financial resources for leasing land on their own. Hence, on the one hand, women are major contributors to agricultural production, and increasingly so as more men than women have moved out of agriculture….” (GOI, 2011b:5). Kelkar, in a recent study, pointed out that unless women get ownership and control of land and other productive assets, management skills, technological empowerment as well as employment opportunities, the process of economic development will only reinforce gender inequalities (Kelkar, 2011:59-68; Kelkar, 2013a: 5) Swaminathan et.al. , household survey in Karnataka found that even when women have access to land, this does not lead to equality within the household. One reason for this is that “asset acquisition of married women is largely mediated through the spouse. Women are largely reported as co-owners of agricultural land and houses that their husbands have either inherited or purchased. Although this might render them empowered enough to participate in the decision-making process, husbands do not necessarily agree with their wife’s perceptions” Swaminathan et.al. 2012:24. 57 According to Velayudhan, in the 23 villages in Gujarat where WGWLO research was done, only 11.81 percent women owned land. Of the 403 cases of women owning land, 47.89 percent owned land after they became widows (Velayudhan, 2009: 77) 52 land; hence, they have lower productivity (Rao,2008). Even state-provided inputs disadvantage women as these are provided on the basis of land-holding size and social norms are also at work in the public service delivery systems. Access to information is gender unequal, including that provided through modern technologies because, for example, fewer women own mobile phones. The lack of land titles excludes women from membership of relevant organizations and from the benefits of many development schemes (Ramakrishnan and Nagar, 2011). It severely constrains access to agricultural credit and even to the Farmers’ Credit Card scheme. In this context, access to microcredit from the savings and loan groups discussed in the previous chapter is important for women. Many women borrow from SHGs, for example, to pay for agricultural inputs even when they do not own the land, or to acquire livestock or poultry that could fetch them a cash income. Persistent rural poverty and disadvantage and an increasingly feminized agricultural sector call for attention to the constraints faced by women farmers – among which the lack of control over land and productive assets is key. Recent policies to strengthen the entitlements of and support to women farmers include joint titles to land, but their implementation has been slow because land represents more than a physical asset or material wealth – it underscores power and authority in Indian society. There are many reasons for women not owning land, but a significant constraint is lack of recognition by families and communities of their legal rights. Equitable laws are necessary but not sufficient to ensure that women are able to realize their rights. Realizing claims involves confronting social, religious, political and economic obstacles at family, community, market and state levels. 6.2 Participation in Agricultural Work The feminization of agriculture has many dimensions.58 Female participation in the sector remained fairly stable and higher than male participation (which declined) during the period of high economic growth. As discussed earlier, about two-thirds of men and women in agriculture are self-employed, and one-third is in casual wage work. In the recent past, unpaid family workers (who are mostly women) replaced hired wage workers and casual labor. While this reflects women’s increased responsibility for agricultural operations, it also represents a downturn in returns to women’s labor. The number of days of work available to women casual agricultural workers is declining due to new infrastructure, technology, inputs and cropping patterns. Male contributions to joint work in households have also declined, with women performing the larger proportion of the work. The stagnation of agriculture in the recent past and diversification of male employment has meant that agriculture’s contribution to the incomes of agricultural households has declined. Agriculture provides much lower wages to casual workers than non-agricultural work (not the least because half of female casual workers are SCs and STs), and even among them there is a gender wage gap – women earned 71 percent of male wages on average in 2007-08 (NSSO, 2010). In several states the wages of women workers were below the minimum wage. This situation for women – uncertain and low paid work – is in part due to their lack of land, productive assets and capital. Even female-owned enterprises 58 Feminization of agriculture, as Rao points out, is not about increasing the number of women workers per se nor about their ownership and control of land. Feminization refers to the rise in the proportion of female to male workers (Rao 2011b:210). 53 in rural areas are small and have low returns – the investment in them is one-third that in male enterprises, and twice the proportion give a daily return that is lower than the minimum wage. Gender, caste and access to land are important in obtaining rural non-agricultural employment with higher returns, and this is in turn important for investments in land and agricultural assets. If a woman owns productive assets, the probability of her participating in household decisions and receiving better wages increases. Indeed, owning productive assets may facilitate women’s participation in household decision- making in a way that work participation does not. In the important area of agriculture, policies and programs cover a range of needs related to agricultural growth, efficient use of natural resources, land and other reforms, rejuvenating agricultural practices and productivity, and technical training and inputs (GOI, 2000). While some initiatives provide necessary support to women farmers, the key strategic need – to enhance women’s control over land and productive assets – is not addressed. Programs are limited in their coverage of the vast numbers of farmers and women who need to be assisted (GOI,1997,2002,2012). A great deal of effort is also required for them to tackle the socio-cultural norms that blur recognition of women farmers and their rights. To focus technical and extension services on women farmers, agricultural program personnel need to be trained well to identify women’s needs and provide relevant and gender-sensitive support (GOI, nda). Program activities must be designed and contextualized to reach women across the social, economic and physical barriers they face. Because of the large number and wide spread of women agricultural workers, gender concerns and approaches need to be mainstreamed across the sector (GOI,2012:80; GOI,2011b:24). Asset Ownership 6.2.1 Constraints to Women’s Asset Acquisition When women’s ‘control’ over rural resources is precipitated by male out-migration it may or may not have positive effects, as women are often left with heavy workloads and are economically and socially vulnerable (Rao 2012:9). In the prevailing context of low returns for women’s work, increasing costs of living, and male migration under economic pressure, poor women’s resources are almost entirely directed to household subsistence rather than asset accumulation. They may invest in children’s education, gifts and dowry (as a way of building social capital), jewelry and other mobile assets rather than land or livestock (Swaminathan et al.,2011). With their limited earnings, women appear to look for assets that are moveable and liquid, and can provide security, especially in times of crisis. Jewelry seems to be the most widely made investment from women’s earnings. Land is usually inherited or purchased – mostly by men, and may be acquired jointly under some government schemes.59 In the end, labor is women’s biggest asset – but requires considerable support to increase both productivity and use. 59 . The Central Ministry of Rural Development is proposing to provide land in women’s names in future, rather than jointly as this has had its drawbacks. 54 As discussed in Chapter 5, women’s acquisition of assets through membership of SHGs and access to credit has been severely limited. In a significant study, women’s loans from SHGs were found to be used to meet household requirements (consumption and working capital for household farms or enterprises), and only 10 percent to improve or buy assets that were owned by the men of the house, including land and family businesses (Garikapati, 2011). Significantly, women repaid the loans from wage work they did in addition to household production and care work. Gender outcomes of land disposal. Over the past two decades, growth in India has driven the acquisition of rural agricultural and common lands by the state and private sector for development of economic zones and infrastructure. Whether through voluntary sales or involuntary ‘handover,’ loss of land has often meant that women lost their resource base, their labor was marginalized, and (even officially) they were relegated to ‘housewife’ status. Money received was often spent on consumer goods or business activities that reflected male preferences, or on risky activities such as land speculation, money-lending or alcohol consumption.60 There were few if any gains to women’s work or welfare, such as investments in health or technology to reduce their work burdens. ‘Unintended’ consequences. Links between women’s work, property ownership and gender-based violence have been drawn by a number of studies. For example, women with cash incomes experienced more domestic violence than those without (Kishor and Johnson, 2004), while women without property were more likely to experience violence than those owning property (ICRW, 2006; Panda and Agarwal, 2005). 61 Other studies, however, have found that an increase in a wife’s assets may induce marital violence. There are differences by culture and class, in addition to variations among families. The extent to which women accept or contest patriarchal norms is critical. Although women are legally permitted to inherit property, including land, social norms still hold daughters’ claims to property in their natal families as illegitimate and threatening to social relations within the family. Joint titling between spouses of housing acquired from public schemes appears to protect women from domestic violence and other ‘ills’ such as abandonment, to make them more assertive and gain them respect in addition to security of tenure.62 In effect, inheritance or ownership of property improves 60 Rao et al., 2007:54 gives the example of sale of agricultural land in f Kazakpur village in Gorakhpur to land developers. While men took up a range of activities including petty trade, brokering, rickshaw pulling or private jobs in the city, no such opportunity was available to the women, who were increasingly confined to their homes. There are other examples mentioned in this report from the states on Uttar Pradesh, Maharashtra and West Bengal. 61 Panda and Agarwal’s study of 500 Kerala households, both in rural and urban areas showed that only 7 percent of women with house and land faced physical violence as against 84 percent of assetless women who faced physical violence (Panda and Agarwal, 2005: 836). Chowdhry, 2012:54 in her study covering 6 districts of Haryana also indicates “… that property certainly scored high with women in their ability to deal with violence. Possessing means of production not only entails possessing a source of income but also source of authority/power/status and mobility—leading to access to education and health facilities— all these are negatively related to violence. … But as they themselves pointed out, how many women are in a position to acquire or inherit property or even other productive assets? This leaves out a vast number of women, especially those belonging to lower caste and economic categories. Therefore, factors other than property also assume importance. These identified by women were: education and employment.” 62 But as Collin points out, the number of states in India that have implemented the central government directions are few. Only the four states of Karnataka, West Bengal, Assam and Madya Pradesh took on joint-titling and it is not clear why some states did and others have not Collin (2013:4). Secondly, 62 joint titling often requires pressure from outside as is the case in 55 women’s status and empowers them – but there is often a need to address the male insecurity and violence that arises as a result. 6.2.2 Strategies to Overcome Constraints Current pressures on land and the imperatives of growth increase the importance of ensuring women’s access to land, property and assets to enhance their economic inclusion. Women must be viewed as legitimate claimants not only by the law but by society, community and family. For this, awareness (beginning with women themselves) is crucial. Social mobilization is playing an important role in helping people assert their land rights, within which attention to gender equity needs to be strengthened. 63 The example of women’s collective action through the Kudumbashree program in Kerala shows the potential of coordination across institutions including banks, government departments, panchayats and women’s groups to secure women’s access to land, develop and cultivate it, and improve household food security. 64 Efforts to improve women’s land rights need to include enhancement of women’s capacities, sensitization of local institutions and the bureaucracy, and changing public attitudes. While other forms of ‘groups’ such as cooperatives (e.g., in the dairy sector) have also been successful in supporting women economically, and could be expanded to other sectors with due attention also to strategic needs to achieve gender equality, a ubiquitous need is for intermediary organizations to promote women’s rights to land and assets, assisting grassroots groups as well as women in panchayats. 6.3 Operationalizing Women’s Claims to Land Women’s ownership and control of only a fraction of household resources, even while they contribute a large share of household productive and care work, disadvantages them across a range of areas from decision-making to expenditure within the household, access to other resources and services, legal claims and employment in the public sphere. Their remaining in low-paid or unpaid home-based informal work is often due to this lack of access. To reduce this constraint they have to be seen by family and community as legitimate claimants to property. Asset ownership must cease to be the double-edged sword it is in terms of gender-based violence and women’s status. Maharashtra where the Jameen Adhikar Andolan (JAA) (Land Rights Movement) ensured that women’s names were included in the land records in many districts of the state (Rao, nd :15) 63 Velayudhan reports on the work of WGWLO, the Working Group for Women and Land Ownership in Gujarat which has had some success in ensuring land rights for women. In Andhra Pradesh, Velayudhan describes the case of the Andhra Pradesh Vyvasaya Vrithodaala Union (APVVU) which was formed as a federation of trade unions in 2001. As a result of APVVU struggle, approximately 1,82,000 acres of land have been taken from landlords who illegally occupied public land. About 90% of this land was distributed among women – 60% to dalits and 30% to adivasis and 10% to backward castes in Anantpur and Kurnool. (Velayudhan, 2009: 78) 64 Other efforts include that of the Andhra Pradesh Mahila Samatha Society where women’s grou ps have been able to jointly lease land and borrow from banks to pay for inputs. The Deccan Development Society, an NGO based in Andhra Pradesh, has organized thousands of women around the issue of land and they now have de facto control over fallow land with financial support from the state government (GOI,2011b). Ekta Parishad, a people’s organization led a strong movement for land reform and this forced the government to set up a Land Reforms Committee in the Ministry of Rural Development (Rao,2011a:10) Similarly, the Working Group for Women and Land Ownership (WGWLO) in Gujarat has organized NGOs and community-based organizations around land rights for women and are working closely with the state government. As a result of these efforts, in some areas, women have been able to access kisan (farmer) credit cards, widow benefits, ration cards, bank accounts and inclusion of local grains within the PDS (Rao, 2011a:10). 56 Given women’s lack of resources, markets too do not work for women, leaving a few state mechanisms to address the situation. Effective implementation mechanisms are required for laws, policies and programs to ensure that women are reached and assisted to act, and that higher-level institutions in turn are responsive to their claims. Government can legislate to make women equal partners in land inheritance and ownership, and recognize women as farmers and ensure their access to resources such as credit, inputs, tools, membership of service and marketing agencies, irrespective of land title. It can sensitize land officials and amend procedures to ensure effective monitoring and implementation of laws, and provide women assistance to enhance their legal awareness and support to deal with administration. Multilateral agencies can help by strengthening mechanisms to monitor India’s legal commitments and their implementation, similar to CEDAW reporting mechanisms. They can support context-specific research to deepen understanding of the institutional constraints to securing rights, recognition and representation for women; and assist the establishment of a sex-disaggregated database on land- holdings. Finally, civil society can help to improve understanding of contextual realities, document positive elements of customary laws and rights that provide the framework for informal institutions to influence social recognition, and increase information-sharing and capacity-building among all stakeholders. 6.4 Infrastructure and Economic Empowerment Does infrastructure development increase labor participation and empower women through better access to resources and markets, or improvements in intra-household efficiency? Several important links have been hypothesized between physical infrastructure and women’s economic activity. As women and girls are largely responsible for household tasks related to water supply, cooking, cleaning, and so on, improvements in water supply or energy (electricity or cooking fuel) are expected to reduce the time they spend on housework, freeing up time for productive activities (or leisure). Improved water supply systems may also enable girls (or boys, in some cases) to go to school.65 Safe sanitation reduces the time spent and hazards of ‘going to the fields,’ and improves hygiene, potentially reducing infection and time spent on illness care. Toilets in schools increase the enrolment and attendance of girls, especially at the secondary level. Electricity improves lighting also, facilitating productive as well as care work and reducing indoor air pollution. It can improve access to information through radio, television, phone networks, and so on – these information and communication technologies (ICTs) have the potential to connect producers to markets and job-seekers to opportunities as well as enable a host of social and economic services including banking, bill payments, etc. Rural women can be spared time on firewood collection by the creation of local infrastructure providing cleaner fuel such as LPG. Improvements in health brought about by cleaner water and fuel also benefit women in terms of more time for productive activities and less on care. Better health also means less expenditure on illness care, increasing the share of 65 The World Bank report on impact of infrastructure cites several field studies to bring out the positive impact of roads, electricity and cooking fuels on health, education and productivity. See World Bank (nd). 57 household income available for investment in other family or work-related activities. 66 There is evidence that these impacts are greater for women than men. Better transport could facilitate job-related and market transactions, increasing the efficiency and productivity of both home-based and market work. Roads can reduce back-breaking labor carrying water or fuel home or produce to market which is particularly important in the context of the out- migration of men from their homes. Extension of road networks and more efficient public transport systems enable people to travel to work, school, health facilities and so on. In the Indian context, where many villages are not yet connected by all-weather roads or regular bus systems, the opening up of the hinterland through better transport can have transformative effects on lives and livelihoods. 6.4.1 What the Data Show67 ‘Income effects’ are strong in the availability of intra-household infrastructure and work participation of men and women. Both NSS and Census data shed light on the relationship between infrastructure and labor participation. Census data in 2001 showed that the availability of electricity as a source of lighting was associated with higher work participation among urban men but not rural men nor rural or urban women; and later NSS data (2009-10) showed a significant relationship between electricity for lighting and work participation of rural salaried men. Salaried work and electricity are both associated with higher household incomes. In contrast, the use of firewood as fuel (which signals poorer households), was associated with higher work participation among women. Where improved fuels (e.g., LPG) were used, there was a negative relationship. Both these relationships are ‘income effects’ showing that more poor women work while fewer women from better-off households do so. On the other hand, men’s work participation increased with kerosene use by households, consistent with the finding that men’s work participation increases from the lowest to the middle MPCE deciles. Similarly, in those households with indoor water taps (i.e., the better-off), women’s work participation was lower (in both rural and urban areas) than those whose women walked a distance to collect water – these are poorer households where more women work. 68The opposite was true for men – work decreased when water was far away and increased when it was nearby (in the house). ‘Better-off’ households most likely invest in making water more easily available within the house but, while this reduces their domestic drudgery, women in such households do not work. The relationships between infrastructure and work participation can be further nuanced by the type of work arrangements. Regular/salaried work, self-employment and casual work have been compared. NSS data showed positive relationships between access to water within the household, availability of toilets and baths, electricity and pucca housing and regular salaried work among rural and urban males 66 Köhlin, et al.,2012 have analyzed more than 25 field level studies to bring out the impact of infrastructure on health, education, improvement in productivity and drudgery. They cite Koolwal and van de Valle (2010) study which argues that infrastructure development could play an important role in increasing women’s off-farm labor participation and thus improve women’s situation by enhancing their control over resources, financial independence, bargaining power within the household, etc. 67 This section is based on the paper by Dewan and Raju (2013BG). 68 See IFC, 2012 58 and females. Tube-wells and hand-pumps (as sources of water), water away from the house, and rural road connectivity were negatively related to rural women’s participation in regular/salaried work. While the type of housing had no impact on men’s work participation in rural and urban areas, the impacts on women’s participation were significant: women with pucca housing were not in the workforce (most likely because they were well-off). Access to a toilet and bath within the house also indicates higher levels of income, and the impacts on work were negative for rural areas, while there was no association in urban areas. Radio, television and telephones were significantly negatively related to rural women’s participation, but positively related to men’s participation. Although significant relationships were observed above between several indicators and regular salaried male and female workers in both rural and urban areas, this was not the case for self-employed workers. No indicator was significant for female self-employed workers in rural areas, but distance to water (away from the house) and kachcha housing were significant for self-employed work among urban women. Six indicators had a significant relationship with self-employment among rural men: distance to water, well (as the source of water) and kachcha and semi-pucca houses were positively related, while electricity (for lighting) and pucca houses were negatively related, indicating that men in households with these facilities were not self-employed. The relationships between infrastructure and employment were weakest for casual workers, suggesting that the availability of physical infrastructure does not affect work participation of this group – they are mostly very poor and hence compelled to work regardless of their access to infrastructure. The only significant relationships were negative ones between urban male casual workers and access to a toilet within the household, and between female casual workers and pucca housing and electricity – all demonstrating the income effect. The 2008-09 NSS data showed more relationships with the participation of both male and female casual workers – access to water near the household, hand- pumps and rural roads had positive relationships with rural work participation; and access to sanitation (toilets and baths) within the house showed significant negative relationships, suggesting that rural households with these facilities are not engaged in casual work, most likely because they are better-off. In urban areas, access to water near the house and hand-pumps had a positive relationship with the participation of male and female casual workers, and access to electricity and sanitation facilities within the house had negative relationships. Again, a better living standard was associated with lower casual work. Thus, overall the data show that the relationship between work participation and infrastructure is highly related to income. Women from poorer households have higher work participation rates and lower access to improved infrastructure, and those from better-off households have higher access to infrastructure and tend to be out of the labor force. The availability of water from outside the house (only) is significantly related to the participation of rural women as casual labor, i.e., in households that are very poor, women perform the lowest type of work. The link between rural women’s self- employment and the availability of LPG as a fuel suggests that when modern amenities are available, women can respond by increasing their involvement in income-generating activities. 59 Multivariate analysis of the significantly-related indicators of infrastructure and work participation identified the amenities that had the greatest impact on work participation rates.69 Across different segments of the female workforce the physical infrastructure that was most significant was the availability of water and access to sanitation, particularly toilets, within the premises. In sum, the availability of infrastructure impacts men’s and women’s work participation differently and sometimes in opposite ways. The impacts varied for casual workers, the self-employed and regular/salaried workers, and the relationships varied across rural/urban areas and states. Thus, location and work arrangements are important correlates of the impact of infrastructure on work participation. Access to several types of infrastructure is determined by income. This is especially true for regular/salaried workers, some of whom are in the ‘middle’ sections of the labor force. 6.5 Inequality in Access to Infrastructure and Work Participation70 Many personal, household and locational characteristics render access to infrastructure uneven, creating ‘inequality of opportunity.’ In addition, in order to produce positive outcomes, the use of these services is important as they may be within reach (accessible) but not utilized due to a lack of information, cultural habits, social norms, and so on. Further, the quality of services is important in determining use and, therefore, outcomes. Our infrastructure analysis was deepened by examining inequalities in access along with use and quality indicators and found great variation in access to services as well as use, particularly in sanitation. Location and education of the head of the household (HOH) were the main contributors to the ‘Human Opportunity Index’ (HOI), the measure of access. Rural households were typically less served by infrastructure, and the impacts of education of the HOH suggest an income effect. Caste was also a significant contributor to inequality – low caste status contributed over 30 percent to the inequality of access to clean water sources, over 15 percent to inequality of access to telephone lines, and 10 percent to differences in access to sanitation and energy sources. In contrast, the ‘sex of the HOH’ had a negligible effect in explaining the HOI at the national level (but was significant in state-level analysis as discussed below).71 69 Katcha house and distance to water (away) explained 11 percent of the WPR of urban female self-employed workers. Availability of water within the house and access to sanitation (toilets) within the premises had a significant positive impact on rural female salaried workers. All the significant indicators together explained about 40 percent of the WPR. Tube-well or hand-pump as the water source, and access to a toilet within the house had significant negative impacts. Distance to water had a major influence on women’s WPR. Rural roads had a significant negative impact. Pucca and semi-pucca housing together explained one-third of the WPR of urban female salaried workers. Access and availability of water within the premises had a significant positive impact. Tube-well or hand-pump near the house, and access to sanitation facilities within the premises were most significantly related to work in the most vulnerable segment of the work force (female casual workers). Access to toilets within the premises had a significant negative coefficient. Rural roads had a significant positive impact but explained only 6 percent of the WPR, suggesting that employment opportunities for this group were largely within the village. Electricity, tube-well/hand-pump near the house, and access to sanitation within the premises were significantly related among urban female casual laborers. When estimating regressions separately, tube-wells/hand-pumps were more important than the distance to water, but accounted for less than 10 percent of the WPR. Access to a toilet facility within the house was more significant than a bath, explaining nearly 30 percent of urban women’s participation in economic activity. 70 This section is based on the background paper by Picon and Echnique (2013BG). 71 As the data on coverage/access/use of services cannot be disaggregated by sex, the sex of the head of the household was the only gender-related indicator that could be used in this analysis. Typically, in India only a small percent of households have female heads. 60 Comparing the distribution of states using an ‘infrastructure index’ with their distribution by female work participation rates revealed a complex situation. An inverse relationship in some small urban states/UTs (e.g., Delhi, Goa) suggests that high incomes and high levels of infrastructure dampen the female WPR, consistent with the income effect. This effect is also at work when low infrastructure is accompanied by high female participation (e.g., in Chhattisgarh). Some larger southern states (e.g., Tamil Nadu, Karnataka) are in the second quartile of infrastructure availability, but the highest quartile of female work participation because their greater cultural gender equality loosens this relationship. However, some of the poorest states (e.g., Bihar, Jharkhand) are in the lowest quartile for both infrastructure and female work – most likely because of their poor economies. Uttar Pradesh is similar, being in the third quartile for infrastructure and having amongst the lowest female WPRs. These findings are consistent with those in Chapter 3. Mapping the relationship of infrastructure with female work participation at the district level showed that there is a limited direct relationship between the two variables. A few districts, mostly in Himachal Pradesh, Kerala and the Northeast, have both high infrastructure and female work participation; many more are in the bottom quartiles for both variables and a similar number are in the middle quartiles. Allowing a variation by one quartile, many more districts show a relationship between medium-high FLFP and medium-high infrastructure. These are mostly in Uttarakhand, Haryana, Rajasthan and Andhra Pradesh. Indeed, these data support the inverse relationship between infrastructure and women’s work due to the income/poverty effect found in the aggregate analysis. The coverage of different states with infrastructure is highly variable. When a ‘dissimilarity index’ (D- index) was calculated for three infrastructure services in each of the 34 states/UTs, using the ‘penalty’ imposed by variations in household characteristics, it showed greater inequality in access to improved sanitation across states than in electricity and water supply. Jharkhand, a low-income tribal state, had the largest D-indices for access to improved sanitation and electricity, and one of the highest for access to improved water sources. Gender differentials arise at the state level. While location and education of the HOH were the most important household characteristics explaining the HOI across states, access to services was not evenly distributed among female- and male-headed households in a number of states (in contrast with the negligible contribution of the sex of the HOH at the country level). In a number of states, access appeared biased towards male-headed households. The greatest contribution of this characteristic to inequality was in the states with high coverage with water and sanitation. Its contribution to inequality of access to water was 10 percent or more in Puducherry, Chandigarh, Meghalaya and Delhi. In contrast with gender, however, caste contributed 72 percent of the inequality of access to improved sanitation services in Manipur and large proportions also in some other states (e.g., 40 percent in Kerala and 33 percent in Punjab). It also contributed high proportions of the inequality of access to improved water sources in West Bengal (47 percent), Tripura (45 percent) and Manipur (38 percent). The HOI and D-index allow identification of states where inequality of access must be specifically targeted. In effect, the challenge of increasing access to infrastructure across India is compounded by the inequality caused by certain characteristics of households, including rural location, those whose 61 heads have low (or no) education, low caste, large size, and female-headedness, most of which characteristics are closely related to poverty. Further analysis. Unfortunately, analysis of the impacts of infrastructure on labor force participation is constrained by the lack of a database directly linking the two with access to and use of infrastructure disaggregated by gender. Clearly, even to strengthen the evidence of a relationship between infrastructure and work participation better information of this nature is needed. There are also many potential specific topics for further research through field surveys and smaller in-depth/qualitative studies: (i) gender and social exclusion from infrastructure availability, access, affordability and use; (ii) gender differences in the impacts of infrastructure on health and educational attainments; (iii) the relationships between infrastructure, gender-specific economic participation, and levels of intensive and extensive economic activities (in different states in order to capture regional differences); this could include the uptake of emerging employment opportunities; and (iv) the gender-disaggregated impacts on work participation of ‘new’ infrastructure services such as markets, banks and training centers. Having examined the individual, household and societal correlates of and some key needs (skills, organization, land ownership and infrastructure) for women’s economic participation in chapters 2 to 6, we turn in chapter 7 to the policy environment that could enable such participation. The chapter also provides some final conclusions. 62 7. POLICIES RELATED TO WOMEN’S EMPLOYMENT 7.1 Empowering Policies and Laws72 This chapter provides an overview of key policies, laws and programs in India that have had a bearing on gender differences in labor outcomes. Policies and programs to empower women in India have expanded greatly over the past four decades. A number of Committee reports, Five-Year Plans, the National Policy for Empowerment of Women, government machinery and civil society efforts have strived to improve gender equality in India. The links between women’s employment and economic growth were specifically addressed in the 11th (2007-2012) and 12th (2012-2017) Five-Year Plans (GOI, 2007; 2012). Among many other proposals the 12th Plan includes improving the employability of women through training in skills that are in demand in the market, particularly to enable women to move out of low-skilled and low-paid work; improving access to land, productive assets and other agricultural resources for better productivity; strengthening women’s collectives; and increasing women’s participation in the organized economy. India also has numerous laws mandating appropriate employment conditions, equality in wages, social security and other benefits, and regulation of industrial relations that cover both men and women. Individual laws apply to the organized (public and private) and/or unorganized sector and a specific law provides broad-based social security measures to the unorganized sector including home-based workers, self-employed workers and daily-wage workers. In addition, women-specific laws provide maternity benefits for all women workers and protect them from sexual harassment in the workplace. In addition to legislation, the organized sector has employee associations and trade unions to protect workers’ interests. These are largely absent for the unorganized sector (which encompasses most of India’s workers). For them the enforcement of existing regulations is poor, job security low, working conditions often hazardous, and wages variable, often ‘minimum’ or lower. Several legal provisions for women are inadequate or not followed or may even inhibit women from working. For example, the definition of ‘workplace’ in the important Factories Act excludes many women workers and their economic activities, such as home-based and self-employed work even though this may provide inputs to factories. Consequently, such workers (who include the majority of women workers) are ineligible for the welfare provisions of the Act. Although aimed at ensuring women’s physical security provisions that prohibit women from working between 7 pm and 6 am discourage factories and other establishments from hiring women as extra (male) workers would be needed to cover night shifts. Other provisions prevent women from doing certain tasks, such as underground work in mines, resulting in women being relegated to other, often lower-paid, work. The Maternity Benefits Act (1961) also has some protective provisions, e.g., prohibiting women from being made to work for six weeks after giving birth. However, the related entitlements under the Act 72 . This chapter incorporates material from Burra (2012BG) and Chauhan, Pal and Abraham (2013BG). 63 such as those to normal pay or against termination of employment are often not provided. As elsewhere, employers shy away from employing women so that they do not have to provide the mandated maternity benefits. Better enforcement is needed to ensure that these mandates are met and do not undermine women’s employment prospects. Similarly, the Factories Act mandates that establishments hiring more than 30 women provide crèches and facilities for mothers to breastfeed. Employers may sidestep this requirement by under-reporting women employees or not hiring women. The Act could be amended to require these facilities in establishments with more than 30 workers of any sex. The Equal Remuneration Act (1976), designed to prevent discrimination in employment, training and wages on the basis of sex, may be thwarted in practice by men and women being assigned ‘different’ work, enabling different pay. While the Act provides for Advisory Committees, Inspectors and registers to be maintained by employers, and for penalties on employers who fail to implement these provisions, there is limited monitoring of these. Informal workers, in particular, need much better legal protection and enforcement. For example, the Unorganised Workers’ Social Security Act (2008) (UWSSA) is silent on important aspects such as penalties for non-payment of wages or delays in payment. Studies show that workers are not provided with basic amenities, crèches, safety measures or social protection, pointing to poor implementation of this Act. There is no regulation yet of work in Private Household Services as homes remain outside the purview of labor legislation. For the most part these workers do not have wage standards, social security, maternity benefits, regulated working conditions, or any other labor rights that could be enforced. Courts have held that those providing ‘personal services’ do not constitute ‘workmen’ for the purpose of the Trade Unions’ Act. While a few states have enacted laws governing some aspects of such work (e.g., minimum wages), and domestic workers are included under the UWSSA and Sexual Harassment Act, there are large gaps in legal support to this widespread and growing form of women’s work. Payment of compensation and dispute resolution are other areas where women appear at a disadvantage. Some legislative provisions assume, for example, that women are inept at managing their finances, and assign related decisions to others. Key roles are provided in dispute resolution to organizations such as trade unions to which few women subscribe, thus undermining women’s rights to bring disputes forward. Women working in hospitals are in a grey area in dispute resolution as courts are allowed to judge on a case-by-case basis whether a hospital is an ‘industry’ that falls in the ambit of the Industrial Disputes Act. The Sexual Harassment Act stipulates that prior to conducting an enquiry the relevant committee must work to settle the matter through conciliation, which belies the power relationship inherent in sexual harassment. It also includes penalties for ‘false charges,’ which could inhibit women from raising genuine sexual harassment complaints. On the whole, although India’s laws are supportive of women’s work, some amendments are needed to accommodate current contexts, and additional legislation is necessary for important groups of informal workers. Monitoring, enforcement and grievance redressal also need to be strengthened for the benefits of legislation to be ensured. 64 7.2 Economic Policies and Related Programs Some of the key policies and programs related to women’s economic empowerment are discussed below. Agriculture. In the important area of agriculture, policies and programs cover a range of needs related to agricultural growth, efficient use of natural resources, land and other reforms, rejuvenating agricultural practices and productivity, and technical training and inputs. While some initiatives provide necessary support to women farmers, the key strategic need – to enhance women’s control over land and productive assets – is not addressed. Programs are limited in their coverage of the vast numbers of farmers and women who need to be assisted. A great deal of effort is also required for them to tackle the socio-cultural norms that blur recognition of women farmers and their rights. To focus technical and extension services on women farmers, agricultural program personnel need to be trained well to identify women’s needs and provide relevant and gender-sensitive support. Program activities must be designed and contextualized to reach women across the social, economic and physical barriers they face. Because of the large number and wide spread of women agricultural workers, gender concerns and approaches need to be mainstreamed across the sector. Wage employment. The Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) which began in 2005 was designed with a gender perspective, explicitly seeking to address some of the challenges that poor working women face, such as gender inequality in wages and a lack of child care. It has provided work and incomes to women in the poorest households: for example, women accounted for almost half of all person-days of work generated by the program up to March 2012, exceeding the ‘quota’ of 33 percent. Research shows that women were paid equal wages and maintained control over them. Among the major reasons for these successes are: the operational guidelines of MGNREGS take women’s needs and constraints explicitly into account; training programs for managers include modules related to gender issues; and the impact of the program on women is a criterion in monitoring and evaluation. A lacuna that has been reported, however, is that the poorest and most vulnerable women are often excluded because they cannot produce the documents required for enrolment. To reach its full potential to enhance women’s work MGNREGS needs (among other things) to increase its coverage in states where it is low (e.g., Uttar Pradesh and Bihar); ensure that social norms do not interfere with women’s access to the program; and provide flexible working hours and the mandated child care facilities. Self-employment. Since 2011 the National Rural Livelihoods Mission (NRLM) has been working to enhance livelihood opportunities, particularly for poor women, in addition to providing access to financial services through SHGs. As described in Chapter 4, SHG programs have been relatively successful in mobilizing and organizing women and, in some states, encouraging their political participation in panchayats. However, important areas for improvement are: ensuring that SHGs are robust in terms of the social capital developed and use of the financial capital they access; improving the productivity of ‘traditional’ activities and greatly expanding non-traditional livelihood activities; including the ‘poorest of the poor’ and developing viable livelihood options for them; monitoring 65 improvements in women’s/households’ economic status, particularly their rise out of poverty; and addressing women’s strategic needs through collective action. Child care. Several laws mandate the provision of child care facilities for organized sector workers as discussed earlier, but implementation even by formal employers needs to be monitored and strengthened. For the unorganized sector, the ICDS program is available throughout the country in both rural and urban areas, and the Rajiv Gandhi National Crèche Scheme in limited areas. The ICDS program provides basic nutrition, health and pre-school inputs to children under six, but its timings limit the extent of child care provided to women who work a full day. The timings may also be unsuited, for example, to agricultural work. This aspect of the program could be strengthened, but there are competing claims for program attention including, importantly, the prevention and care of malnutrition among mothers and young children, and timely health inputs to them. Health insurance. The importance of health for productivity has not been covered in our report, and has received less attention generally. Health insurance is not readily available for many types of workers even in the formal sector. The Rashtriya Swasthya Bima Yojana (RSBY), a cashless health insurance scheme, was begun in 2008 and is being phased in across the country and across different types of unorganized workers, including street vendors and domestic workers. This contributory scheme covers a wide variety of health problems as well as maternity care and hospitalization, and early feedback suggests it holds promise. 66 8. CONCLUSIONS AND AREAS OF FURTHER RESEARCH Despite two decades of robust economic growth in India, the LFPR is much lower than men’s and also fairs poorly compared with women in most countries. Three major correlations of labor outcomes were sought. LFPRs are highest in households with lowest consumption expenditure levels and decline with higher expenditure. As men’s wages increase, women’s participation decreases. Groups suffering from social disadvantage also revealed gender gaps in participation. Illiterate men and women (with some exceptions) had the highest work participation rates. Women in better-off households with a middle school education had lower LFPRs because social norms promote marriage and motherhood. The best-off women with high technical and professional qualifications had higher LFPs than before and there was hardly any gender difference in urban areas. Even though urban areas were the locus of India’s GDP growth, on the average women got only one out of eight jobs that were created there. Women are to be found mainly in informal jobs, often in home- based work and at the bottom of the pyramid, with gender gaps in wage rates increasing over time. Formal employment, negligible as it is, can be accessed only by educated and middle –income women. Case studies of garment manufacturing, the soft-ware industry and private household services bring out the influence of socio-cultural practices on labor market segmentation by gender; the need to improve working conditions whether in the factory or in the home and gender stereotyping that acts as a barrier to the entry of females into male-dominated sectors. A study of different states brings out the links between women’s LFPR and the socio-cultural characteristics. There is wide variation as to women’s LFPRs in different sectors in different states, with the supply of female labor high in the primary sector. Statistical and econometric techniques to measure women’s work have to be supplemented by smaller surveys and time use studies to give a better picture of social reality. Ultimately, it is social norms that determine how LFPRs behave. Acceptance by communities and households for women to go out and work is a key factor. The gendered roles of wife, mother and caregiver offer little space for women to work. The social status of families goes up when women stay at home and this is a core value in Indian society. SHGs are one means to organize women to improve their agency and show signs of promise and potential to reduce poverty and improve gender relations in households. As agriculture is the main source of employment for both women and men, conditions of work are central. In an agricultural economy, where small and marginal farmers own only a small proportion of land, women are doubly disadvantaged because, by and large, they do not own land, are not recognized as farmers by the state machinery and find it difficult to assert their rights in a male-dominated society. Many studies have shown how the provision of physical infrastructure like electricity, water and sanitation and roads reduce drudgery and improve health and educational outcomes. 67 8.1 Future research During the course of preparing this report, certain issues have arisen that call for further study and research. These are discussed briefly below.  There are wide variations in LFPRs both across sectors and States. In order to bring out the complexity and contextuality of these variations, more studies are needed at the sector and State levels.  One hypothesis presented in the report is that women with a middling education have tended to stay at home because of social norms. There are rival explanations such as the one that argues that mothers are substituting their time so that their older female children can go to school. It has also been pointed out that there is a lack of fit between the qualifications of this segment of women and the jobs available in the market. Further research and study are needed to assess the validity of rival hypotheses.  The focus of this report has been upon women, work and employment. The notion of the ‘empowerment of women’ deserves to be unpacked conceptually and in practical terms. Such a study would inter alia necessarily consider the ‘economic’ aspects of empowerment as well and should be based on primary research in different locations.  This report has emphasized the role of SHGs in promoting women’s economic activities as also in the arena of intra-household decision-making. A more detailed balance sheet of SHGs and their impact on poverty has to be drawn up in order to help orient policy and program initiatives of government departments and other relevant institutions.  It is also necessary to look at a broader canvas where social movements, poor people’s organizations, producer groups and other kinds of women’s organizations to understand the interplay between the larger contexts and women’s vulnerability and agency.  NSS data may not capture women’s informal work adequately and also their contribution to the Indian economy. Smaller surveys that probe more deeply can capture ‘missing work’ and ‘missing women’. More such surveys need to be commissioned.  The last time use survey of men, women and children was done was in 1996. It would be useful to have a fresh survey done by the same organisation so that data are comparable and there is better understanding of the time that men and women spend on various productive and care activities.  Some studies have brought out the ‘income effect’ where women in better-off households – with infrastructure – tend to stay at home on account of social norms. More primary research in this area is called for. 68  One of the themes of the report has been the role and importance of social norms of the wider community and society in the work and lives of women. To understand their dynamics better, studies need to be commissioned across the different axes of caste, class and religion as also regions and the urban-rural divide.  Given the multiplicity of governmental departments and initiatives at both State and Central levels, there are myriad programs with broadly similar goals being implemented. The need for convergence amongst these programs is often expressed but detailed studies –that bring out the complexities organizations working in the field face – are needed.  It is important to examine why childcare facilities have not been provided inspite of several recommendations of various government committees and the implications of lack of child care facilities for FLFP.  It will be useful to commission research that analyzes gender dynamics from the perspectives of men’s roles and the ways in which they support or hinder a more balance female/male participation in the labor force. 69 ANNEXURES (INCLUDING UPDATED TABLES & FIGURES) Figure 2.1 Labor Force Participation Rates of Women and Men by MPCE Decile Groups, All-India Rural and Urban, 1993-94 to 2011-12 Source: World Bank staff Figure 2.2 Labor Force Participation Rates of Different Social Groups, 2011-12 90.0 80.0 70.0 60.0 L F 50.0 P 40.0 R 30.0 20.0 10.0 0.0 ST SC OBC Muslim Others Total Socio-Religious Groups Rural Male Rural Female Urban Male Urban Female Source: Raveendran, 2012 BG 70 Figure 2.3 Labor Force Participation among 15-64 Year-olds, 2011-12 100.0 90.0 80.0 L 70.0 60.0 F 50.0 P 40.0 R 30.0 Male 20.0 10.0 Female 0.0 Person Educational Level Source: Sankar, 2013:5 Figure 2.4 Labor Force Participation of Men and Women by Education, Rural and Urban Areas, 2011-12 100.0 90.0 80.0 L 70.0 60.0 F 50.0 P 40.0 Rural Male R 30.0 20.0 Rural Female 10.0 Urban Male 0.0 Urban Female Educational Level Source: Sankar,2013:6 71 Figure 2.5 LFPR among Different Social Groups by Education, 2011-12 Source: Sankar, 2013:10-11 72 Figure 3.1 Share of Women in Each Industry Group, NSS 2011-12 80.0 67.0 70.0 Percentage of women 60.0 50.0 43.7 41.4 40.0 36.1 28.5 29.1 30.0 20.0 14.4 11.6 15.1 15.5 16.1 12.3 12.2 7.8 10.9 10.0 2.0 Share (%) 0.0 Industry Group Source: Sankar, 2013 BG:13 Figure 3.2 Distribution of Female and Male Work Force, NSS 2011-12 Source: Sankar 2013 BG:13 73 Figure 4.2 LFPRs for Rural, Urban and All Males, Females and Persons by MPCE Deciles A. All-India, 2011-12 B. Bihar, 2011-12 74 C. Uttar Pradesh, 2011-12 D. Himachal Pradesh, 2011-12 75 E. Gujarat, 2011-12 F. Andhra Pradesh, 2011-12 Source: Raveendran 2012 BG:55-58 76 Table 1.1 Number and Distribution of Work Force Participants (Usual Principal and Subsidiary Status Workers) Aged 15 and above, 1993-94 to 2011-12 (millions; percentages in brackets) 1999-00 2004-05 2009-10 2011-12 Rural Males 192.4 (65.9) 214.8 (64.2) 227.9 (69.3) 232.6 (70.0) Rural Females 99.5 (34.1) 119.7 (35.8) 101.0 (30.7) 99.7 (30.0) Rural Persons 291.8 (75.5) 334.5 (74.8) 328.9 (72.2) 332.3 (71.1) Urban Males 76.3 (80.6) 89.5 (79.3) 102.6 (81.1) 108.4 (80.0) Urban Females 18.4 (19.4) 23.4 (20.7) 23.9 (18.9) 27.0 (20.0) Urban Persons 94.7 (24.5) 113.0 (25.2) 126.5 (27.8) 135.4 (28.9) All Males 268.6 (69.5) 304.4 (68.0) 330.5 (72.6) 341.0 (72.9) All Females 117.8 (30.5) 143.1 (32.0) 124.9 (27.4) 126.8 (27.1) All Persons 386.5 (100.0) 447.5 (100.0) 455.4 (100.00) 467.7 (100.0) Source: Raveendran, BG 2012:6 Table 1.2 Labour Force Participation Rates of Men and Women (UPSS workers aged 15 and above), Rural and Urban Areas, 1993-94 to 2011-12 (percent) 1999-00 2004-05 2009-10 2011-12 Rural Males 85.4 85.9 82.5 81.3 Rural Females 45.4 49 37.3 35.6 Rural Persons 65.8 67.7 60.2 58.7 Urban Males 78.8 79.2 76.1 76.3 Urban Females 20.9 24.4 19.5 20.5 Urban Persons 51 53.5 48.6 49.1 All Males 83.4 83.8 80.4 79.7 All Females 38.1 41.8 31.6 30.7 All Persons 61.3 63.4 56.4 55.5 Source: Raveendran, 2012 BG:6 Table 2.1 Mean Male Household Wages and Female LFPRs, Rural Areas, 2004-05 and 2011-12 Household 2004-05 2011-12 Change in Change in Male Wage Mean Female Mean Male Female mean male female LFPR Quintiles male LFPR household LFPR household (Percentage wage points) household (Percent) wages (Percent) (Percent) wages Poorest 301 -36.7 Quintile 94.4 68.1 396 57.8 Second 468 -171.5 Quintile 210.3 57.4 679 38.8 Third Quintile 280.4 55.2 838 38.5 558 -241.9 Fourth 671 -323.6 Quintile 357.2 51.9 1028 33.5 77 Richest 1149 -906.0 Quintile 936.4 36.5 2085 30.5 All 451.3 53.9 1190 40.2 739 -411.0 Table 2.2 LFP Rates for Women and the Ratios of Male to Female LFPRs by Social Group, 2011-12 M:F LFPR Female LFPR Ratio Muslims 20.2 4.0 Others 23.6 3.2 All Women 30.7 2.6 Other Backward Classes 32.4 2.5 Scheduled Castes 34.7 2.3 Scheduled Tribes 50.2 1.7 Table 2.3 Labor Force Participation among Rural and Urban Women by Education Levels 1999-2000 2004-05 2009-10 2011-12 Rural Percent Percent Percent Percent Educational Levels of LF LFPR of LF LFPR of LF LFPR of LF LFPR All Educational Levels 100.0 41.7 100 45 100 34.9 100 35.6 No Education 74.0 47 67 50.2 56.6 39.7 56.2 41.6 Upto primary 14.4 36.1 16.5 41.2 22 35.5 21.4 36.0 Upto Higher Secondary 10.6 26.4 14.4 32.5 18.6 24.9 20.0 25.2 Post Higher Secondary 1.0 42.1 2.1 50.1 2.8 38.7 2.3 35.8 Urban Percent Percent Percent Percent Educational Levels of LF LFPR of LF LFPR of LF LFPR of LF LFPR All Educational Levels 100 20.6 100 24.3 100 20 100.0 20.5 No Education 41.5 26.7 35.7 30.6 27.9 24.6 26.0 23.9 Upto primary 15.2 17.4 17.2 23.5 17.8 21.5 18.4 22.3 Upto Higher Secondary 26.4 13.9 25.5 15.5 28.1 12.8 31.6 14.3 Post Higher Secondary 16.9 31.7 21.6 37.2 26.2 30.7 24.0 32.4 Source: World Bank staff 78 Table 2.4 Average Wage Rates of Women and Men by Level of Education, Rural and Urban areas 2011-12 Rural Urban All Level of Education M:F M:F M:F Male Female Male Female Male Female Ratio Ratio Ratio Illiterate 141.6 98.8 1.4 183.6 117.1 1.6 149.8 102.5 1.5 Primary and 104.2 1.5 1.7 1.5 below 151.6 203.6 122.8 166.8 109.9 Middle 172.1 109.7 1.6 233.5 134.6 1.7 196.9 118.7 1.7 Secondary and 179.2 1.4 1.2 1.3 Higher Secondary 252.1 362.5 314.0 310.1 247.9 Graduate and 365.8 1.4 1.3 1.3 above 514.7 800.1 614.7 728.6 564.3 2009-10 Illiterate 97.4 68.0 1.4 132.3 82.9 1.6 103.5 70.6 1.5 Primary and 105.3 70.2 1.5 146.1 94.0 1.6 117 75.9 1.5 below Middle 123.5 72.6 1.7 180.3 106.0 1.7 145.3 83.8 1.7 Secondary and 187.2 137.6 1.4 296.8 257.3 1.2 248.2 199.6 1.2 Higher Secondary Graduate and 374.0 281.7 1.3 629 503.5 1.2 568 459.7 1.2 above 2004-05 Illiterate 52.2 34.2 1.5 82 45.8 1.8 57.4 36.1 1.6 Primary and 63.7 1.6 94.2 56.5 1.7 74.3 44.8 1.7 below Middle 78.4 1.9 108.6 62.3 1.7 92.1 48.7 1.9 Secondary and 136.8 1.3 189.9 169.1 1.1 168.1 140.3 1.2 Higher Secondary Graduate and 261.6 1.5 366.5 267.9 1.4 339.2 249.4 1.4 above 1999-2000 Illiterate 43 1.5 68.3 41.2 1.7 47.4 30.4 1.6 Primary and 54.4 1.2 83.9 52.9 1.6 64.5 46.9 1.4 below Middle 68.6 1.8 96.9 64.5 1.5 81.4 47 1.7 Secondary and 115 1.1 161.7 139.4 1.2 142.3 124 1.1 Higher Secondary Graduate and 216.5 1.4 278.6 235.4 1.2 262.6 222.4 1.2 above 1993-94 Illiterate 22.5 1.5 37.5 21.3 1.8 25 16.1 1.6 Primary and 28.8 1.6 44 22.3 2 34.1 18.9 1.8 below Middle 37.5 1.7 52.1 31.3 1.7 44.5 25.6 1.7 Secondary and 60.3 1.2 77.4 68.9 1.1 70.4 62.7 1.1 Higher Secondary 79 Graduate and 94.5 1.3 127.5 99.1 1.3 118.8 95.4 1.2 above Table 3.1 Gender Differences in Activity Status of Workers in Key Sectors Men Women Rural Largely complementarity Complementarity between Agriculture between UFW and CWW (seen UFW and CWW in many sectors) Both RWW and CWW increased OAW and RWW increased but Manufacturing during the 18-year period UFW and CWW decreased Construction SEW decreased Amost all CWW Trade and Repair 90 percent SEW, distributed 15 percent RWW and CWW Services evenly among OAW and UFW Private Household 56% CWW Over two-thirds RWW Services Only small differences in distribution by sex and activity status in Education both rural and urban areas Urban Manufacturing (large Two-thirds SEW of which two- Two-thirds RWW and CWW employer) thirds are OAW Private Household Much less CWW as in rural areas Services Education Five-sixths RWW RWW doubled among rural females and increased by 45% among Overall urban females over 18 years 80 Table 4.1R Distribution of Male and Female Workers in Different Occupational Groups in Rural Areas of States, 2011-12 (percent) Male Female State/UT AFF MFG CONS TRS OTH AFF MFG CONS TRS OTH Andhra Pradesh 52.5 50.7 67.3 74.6 76.6 47.5 49.3 32.7 25.4 23.4 Assam 77.2 88.7 91.4 95.9 84.5 22.8 11.3 8.6 4.1 15.5 Bihar 89.1 85.0 97.4 95.2 91.9 10.9 15.0 2.6 4.8 8.1 Gujarat 64.8 81.9 81.1 87.7 82.2 35.2 18.1 18.9 12.3 17.8 Haryana 66.7 91.0 93.2 99.5 89.2 33.3 9.0 6.8 0.5 10.8 Himachal Pradesh 30.4 87.8 84.0 87.5 79.6 69.6 12.2 16.0 12.5 20.4 Karnataka 64.1 67.3 91.7 81.8 82.5 35.9 32.7 8.3 18.2 17.5 Kerala 61.6 56.2 79.6 87.2 70.6 38.4 43.8 20.4 12.8 29.4 Madhya Pradesh 67.2 69.6 76.8 85.6 88.6 32.8 30.4 23.2 14.4 11.4 Maharashtra 54.0 79.6 82.5 79.8 83.0 46.0 20.4 17.5 20.2 17.0 Odisha 67.0 55.7 81.5 89.7 80.7 33.0 44.3 18.5 10.3 19.3 Punjab 59.1 74.2 98.4 92.9 77.4 40.9 25.8 1.6 7.1 22.6 Rajasthan 48.8 82.2 71.2 87.5 83.3 51.2 17.8 28.8 12.5 16.7 Tamil Nadu 61.1 55.1 45.5 71.9 80.0 38.9 44.9 54.5 28.1 20.0 Uttar Pradesh 67.0 76.5 95.9 93.8 84.6 33.0 23.5 4.1 6.2 15.4 West Bengal 81.2 48.8 91.8 92.0 79.2 18.8 51.2 8.2 8.0 20.8 Jharkhand 62.8 73.5 96.2 91.5 87.7 37.2 26.5 3.8 8.5 12.3 Chhattisgarh 53.9 68.5 65.3 77.5 76.1 46.1 31.5 34.7 22.5 23.9 Uttarakhand 39.6 72.7 96.6 92.4 91.3 60.4 27.3 3.4 7.6 8.7 Delhi 100.0 77.7 100.0 40.7 87.5 0.0 22.3 0.0 59.3 12.5 Other NE States 61.8 49.2 59.1 73.8 77.5 38.2 50.8 40.9 26.2 22.5 Other States 63.8 75.5 81.2 76.6 75.5 36.2 24.5 18.8 23.4 24.5 INDIA 63.9 66.6 82.6 87.9 82.2 36.1 33.4 17.4 12.1 17.8 AFF = Agriculture and Fishing; MFG = Manufacturing; CONS = Construction; TRS = Trade and Repair Services, and; OTH =Other occupations 81 Table 4.2U Distribution of Male and Female Workers in Different Occupational Groups in Urban Areas of States, 2011-12 (percent) Male Female State/UT AFF MFG CONS TRS OTH AFF MFG CONS TRS OTH Andhra Pradesh 58.2 61.9 89.6 80.6 78.9 41.8 38.1 10.4 19.4 21.1 Assam 74.4 84.0 98.5 92.9 80.9 25.6 16.0 1.5 7.1 19.1 Bihar 78.6 88.5 95.5 95.0 90.9 21.4 11.5 4.5 5.0 9.1 Gujarat 65.9 85.1 92.8 93.0 76.2 34.1 14.9 7.2 7.0 23.8 Haryana 70.6 93.7 97.2 96.7 75.8 29.4 6.3 2.8 3.3 24.2 Himachal Pradesh 41.3 75.5 91.0 89.3 78.7 58.7 24.5 9.0 10.7 21.3 Karnataka 78.2 66.2 93.9 92.1 74.6 21.8 33.8 6.1 7.9 25.4 Kerala 70.1 61.0 93.1 80.5 66.4 29.9 39.0 6.9 19.5 33.6 Madhya Pradesh 73.0 77.6 92.2 90.3 80.9 27.0 22.4 7.8 9.7 19.1 Maharashtra 56.9 76.4 89.3 91.2 74.3 43.1 23.6 10.7 8.8 25.7 Odisha 66.6 70.3 79.6 84.7 84.1 33.4 29.7 20.4 15.3 15.9 Punjab 63.1 85.5 98.0 96.6 72.7 36.9 14.5 2.0 3.4 27.3 Rajasthan 41.6 72.8 87.3 92.6 80.1 58.4 27.2 12.7 7.4 19.9 Tamil Nadu 64.5 66.5 91.5 84.8 74.6 35.5 33.5 8.5 15.2 25.4 Uttar Pradesh 77.8 75.7 96.1 96.0 84.1 22.2 24.3 3.9 4.0 15.9 West Bengal 84.1 75.7 97.3 89.5 70.3 15.9 24.3 2.7 10.5 29.7 Jharkhand 73.6 85.0 95.5 95.8 84.9 26.4 15.0 4.5 4.2 15.1 Chhattisgarh 52.4 75.2 65.4 86.3 65.2 47.6 24.8 34.6 13.7 34.8 Uttarakhand 78.8 82.5 99.4 95.2 79.5 21.2 17.5 0.6 4.8 20.5 Delhi 100.0 92.4 98.0 88.7 78.9 0.0 7.6 2.0 11.3 21.1 Other NE States 64.5 49.7 94.1 70.5 78.1 35.5 50.3 5.9 29.5 21.9 Other states 87.9 82.8 87.9 87.5 76.1 12.1 17.2 12.1 12.5 23.9 INDIA 67.2 75.8 91.6 90.0 76.5 32.8 24.2 8.4 10.0 23.5 AFF = Agriculture and Fishing; MFG = Manufacturing; CONS = Construction; TRS = Trade and Repair Services, and; OTH = Other Occupations 82 LIST OF BACKGROUND PAPERS Burra, Neera (2012) Gender Equality in Labour-related Public Policies and Programs. Chakravarty, Deepita (2012) Indian Women at Paid Work: Does their Concentration in Specific Sectors Say Something Positive? Chauhan, Devika, Poulomi Pal and Anita Abraham (2013) Impact of existing Labour Laws and Practices on Participation of Women in the Workplace. Cuberes, David and Marc Teignierz (2012) Gender Gaps in the Labor Market and Aggregate Productivity: the Case of India. Dand, Sejal and Subhalakshmi Nandi (2012) Examining SHG’s Through a Gender Lens: Implications for Women’s Empowerment. Dewan, Ritu and Swati Raju (2013) Impacts of Physical Infrastructure on Women’s Workforce Participation Rates in Rural and Urban India. Joseph, George and Luis Alberto Andres (2012) Gender and Growth in India: An Outline. Picon, Mario and Juan Agustin Echenique (2013) Inequality of Opportunities in the Access to Infrastructure services: An Application to India. Rao, Nitya (2012) Gender Differentials in Access to Land and Productive Assets. Raveendran, G. (2012) Analysis of Gender and Labour Force Participation with Four Rounds of NSS Data. Ray, Srabashi, Preet Rustagi, Sumit Mazumdar, Abhishek Kumar and Ann George (2013) Inequality of Opportunity in Education and Labour Market Outcomes in Rural Bihar – A Gender Perspective. Sankar, Deepa (2012) Examining Education and Labor Force Participation Progress through a Gender Lens in India. Shenoy, Meera (2012) Investing in Endowments: Employment-related Skills. Sinha, Anushree (2012) Review of Gender-Aware Macro Models Integrating Intra-Household Bargaining Behaviors. Sudarshan, Ratna M. (2011) Women's Economic Empowerment and Macro Economic Growth in India. Sudarshan, Ratna M. (2012) Understanding the Formal-informal and ‘Missing Workers’ in Women's Labour Force Participation. Urzua, Sergio and Giordano Palloni (2012) Female Labor Force Trends in India. 83 REFERENCES Abraham, Vinod (2009) ‘Employment Growth in Rural Distress-Driven?’ Economic and Political Weekly, 44(16), 97-104. Acharya, Shankar (2002) ‘India: Crisis, Reforms and Growth in the Nineties,’ Center for Research on Economic Development and Policy Reform Working Paper 139, Stanford University, California, USA. Afridi , Farzana Abhiroop Mukhopadhyay, and Soham Sahoo, (2013) “Female Labour-Force Participation and Child Education in India: The Effect of the National Rural Employment Guarantee Scheme’ Young Lives Working Paper 95, March Aiyar, Y. and S. Samji (2009) “The National Rural Employment Guarantee Act: Design, Process and Impact”, Ministry of Rural Development, New DelhiArun, Shoba, Thankom Arun and U. Devi (2002) ‘ICT, Gender and Development: Women in Software Production in Kerala,’ Journal of International Development, 14(1): 39-50 Basu, K. (2008) “India’s Dilemmas: The Political Economy of Policymaking in a Globalised World”, Economic and Political Weekly, Vol. 43, No.5, 2-8 February Bhalla, Surjit, S. & Ravinder Kaur(nd) Labour Force Participation of Women in India: Some facts, some queries; Asia Research Centre Working Paper 40, London School of Economics & Political Science Bhatt, Ela R. (2006) We are Poor but So Many, Oxford University Press, New Delhi. Bhattacharya, Manasi et al. (2009) ‘Marital Violence and Women’s Employment and Property Status: Evidence from North Indian Villages’, IZA DP No. 4361, August Burra (2011) ‘Developing Women’s Skills for Improving Economic Prospects’, Report prepared for the World Bank, draft, November, New Delhi Blackden, Mark, Sudharshan Canagarajah, Stephen Klasen and David Lawson (2006) Gender and growth in Sub-Saharan Africa. UNU-WIDER Research Paper No. 2006/37. Bloom, David E. and Jeffrey G. Williamson (1998) ‘Demographic Transitions and Economic Miracles in Emerging Asia’, World Bank Economic Review 12, 419-455. Carr, M., Chen, M., & Jhabvala, R. (1999). ‘Introduction’. In M. Carr, M. Chen, & R. Jhabvala, Speaking Out: Women's Economic Empowerment in South Asia (pp. 1-17). New Delhi: Vistaar Publications. Chaudhury, S. and M. Ravallion (2006) “Partially Awakened Giants: Uneven Growth in China and India,” Policy Research Working Paper 4069, World Bank, Washington, D.C. Chakravarty, Deepita (2004) ‘Expansion of Markets and Women Workers: A case of Indian Garment manufacturing,’ Economic and Political Weekly, Vol. 39, No. 46: 4910-4916. Chakravarty, Deepita (2006) ‘Quality of Employment and Organization of Work in the Indian IT industry,’ in collaboration with Dr. Loraine Kennedy, member of the UR23 of the IRD, IFREDE, France under the framework of the study ‘Decentralised industrialization and Political Industrialization in India’. Chakravarty, Deepita (2007) ‘Overview and journey of home based workers in South Asia,’ paper presented at SEWA-UNIFEM Policy Conference on Home Based Workers in South Asia, 2007, New Delhi. Chakravarty, Deepita and Indranil Bose (2011) ‘Industry, Labour and the State: Emerging Relations in the Indian State of West Bengal,’ Journal of South Asian Development, Sage Publications. 84 Chakravarty, Deepita and Ishita Chakravarty (2012) ‘For Bed and Board Only: Women and Girl Children Domestics in Post Partition Calcutta (1951-1981),’ Modern Asian Studies, Cambridge University Press. Chen, Martha, Jennefer Sebstad and Lesley O’Connell (1999) ‘Counting the Invisible Workforce: The Case of Homebased Workers’, World Development, Vol. 27, No.3, pp.603-601 Chowdhry, Prem (2012) ‘Reduction of Violence Against Women: Property Ownership & Economic Independence in Rural Haryana’, UN Women Collin, Matthew (2013) ‘Joint-titling of land and housing: Examples, causes and consequences’, Oxford Policy Management, Economic and Private Sector Professional Evidence and Applied Knowledge Services, Helpdesk Request, July CII and BCG (2010) Confederation of Indian Industries and Boston Consulting Group (2010) Indian Manufacturing: The Next Growth Orbit, BCG, NY. Das, Maitreyi (2006) ‘Do Traditional Axes of Exclusion Affect Labour Market Outcomes in India,’ Washington D.C. World Bank: Social Development Paper no. 97. Das, Maitreyi and S. Desai (2003) ‘Why are Educated Women Less Likely to be Employed in India? Testing Competing Hypotheses,’ Social Protection Discussion Paper Series No. 0313, Social Protection Unit, Human Development Network, The World Bank, May. Dasgupta, Sukti and Ratna M. Sudarshan (2011) “Issues in labour market inequality and women’s participation in India’s National Rural Employment Guarantee Programme, ILO Working Paper No. 98, Policy Integration Department, International Labour Office, Geneva, February Doepke, Matthias, Michele Tertilt (2009) ‘Women's liberation: What's in it for men?’ Quarterly Journal of Economics 124(4), 1541-1591. Dollar, David and Roberta Gatti (1999) ‘Gender Inequality, Income and Growth: Are Good Times Good for Women? ‘Mimeograph, World Bank, Washington, DC. Drèze, Jean and Reetika Khera (2011) “The Battle for Employment Guarantee” in Reetika Khera (edited) The Battle for Employment Guarantee, Oxford University Press, New Delhi Dutta, P, R. Murgai, M. Ravallion, and W. V. Dominique (2012) ‘Does India’s Employment Guarantee Scheme Guarantee Employment’, Economic and Political Weekly, Vol. XLVII, No 16, April 12 Galor, Oded and David N. Weil (1996) ‘The Gender Gap, Fertility, and Growth,’ American Economic Review, 85(3), pp. 374-387. Garikipati, S. (2011). Microcredit and women's empowerment: Through the lens of time use data from rural India. Brussels, CEB WP No. 11/034, Solvay Brussels School of Economics and Management, Centre Emile Bernheim. Ghosh, Jayati (2002) ‘Social Policy in Indian Development,’ prepared for the UNRISD project on Social Policy in a Development Context in the UNRISD, November, Draft, Geneva. Ghosh, Jayati (2009a) Never Done and Poorly Paid: Women’s Work in Globalising India, New Delhi: Women Unlimited. Ghosh, Jayati (2009b) ‘Equity and Inclusion through Public Expenditure: The Potential of the Nregs’, New Delhi: Paper for International Conference on NREGA, 21–22 January 85 GOI (1997) “Ninth Five Year Plan” Vol.2, Planning Commission, Government of India, New Delhi GOI (2000) ‘National Agricultural Policy’, Ministry of Agriculture and Cooperation, Government of India GOI (2002) “Tenth Five Year Plan, Sectoral Policies and Programmes, 2002-2007”, Vol. II, Planning Commission, Government of India, New Delhi GoI (2007) Eleventh Five Year Plan document,Volume III. Agriculture, Rural Development, Industry, Services and Physical Infrastructure, Planning Commission, Government of India, New Delhi GOI (2009)“Vision of the National Skill Development Initiative in India” Ministry of Labour and Employment, Government of India, March GOI (2010) ‘Annul Report To The People on Employment’, Government of India, Ministry of Labour and Employment, New Delhi, July. GOI (2010b) “Engendering Public Policy. A Report on the Work of the Working Group of Feminist Economists during the Preparation of the Eleventh Five-Year Plan: 2007-2012”, Planning Commission, Government of India, New Delhi, April GOI (2011a) ‘Working Group Report on Elementary Education and Literacy’, GOI (2011b) ‘Twelfth Plan Working Group on Disadvantaged Farmers including Women’, Final Report, Submitted to the Planning Commission, Government of India, New Delhi GOI (2011c)“Report of The Working Group on Decentralised Planning in Agriculture for XII Plan” submitted to Planning Commission, Government of India, New Delhi, November. Chair: R S Deshpande GOI (2011d) “Report of the Working Group on Child Rights for the 12th Five Year Plan (2012-2017)”, submitted to the Ministry of Women and Child Development, Government of India, New Delhi GOI (2012a) “Annual Report 2011-12”, Ministry of Agriculture, Government of India, New Delhi GOI (2012b) “MGNREGA Sameeksha”, An Anthology of Research Studies on the Mahatma Gandhi National Rural Employment Guarantee Act, 2005 2006–2012, Ministry of Rural Development Government of India, Orient Blackswan Private Limited, New Delhi GOI (nd (a)) “Policy Framework for Agricultural Extension”, Extension Division, Department of Agriculture & Cooperation, Ministry of Agriculture, Government of India, New Delhi http://agricoop.nic.in/welcome.html Gothoskar Sujata (2000) ‘Teleworking and Gender’, Economic and Political Weekly, Vol. XXXV, No.26, Grogan and A. Sadanand (2009) ‘Electrification and the Household,’ University of Guelph, Working Paper, www.isid.ac.in Harris, John (2007) ‘Antinomies of Empowerment: Observations on Civil society, Politics and Urban Governance in India’,Economic and Political Weekly, Vol.42, No.26, pp.2716-2724 Heeks, Richard (2001) ‘Indian Software Export Figures,’ IDPM, University of Manchester, UK. Hill, Anne M. and Elizabeth M. King (1995) ‘Women's education and economic well-being.’ Feminist Economics 1(2),126. Himanshu (2011) ‘Employment Trends in India: A Re‐examination,’ in: Economic and Political Weekly, 46, 37, 43‒59. 86 Holmes, Rebecca, Nidhi Sadana and Saswatee Rath (2010) “Gendered risks, poverty and vulnerability in India : Case study of the Indian Mahatma Gandhi National Rural Employment Guarantee Act (Madhya Pradesh)”, Overseas Development Institute and Indian Institute of Dalit Studies, October Hong Tan, Sangeeta Goyal, Yevgeniya Savchenko (2007) “2006 Baseline Study of India’s Industrial Training Institutes: Institutional Performance and Employment of ITI Graduates”, Prepared for the Directorate General for Employment and Training Ministry of Labor and Employment, Government of India South Asia Region, Human Development Department (SASHD), March, The World Bank .” http://planningcommission.nic.in/aboutus/committee/wrkgrp12/wcd/wgrep_women.pdf ICRW (2006) ‘Report on violence against women. International Center for Research on Women’. Available online www.icrw.org/html/issues/violence.html. IFC (2012) ‘Gender Impact of Public Private Partnerships Literature Review Synthesis Report , International Finance Corporation, London, UK, November Jha, Praveen (2009) “The Well-Being of Labour in Contemporary Indian Economy: What’s Active Labour Market Policy Got to Do With It?”Paper prepared for a background report for the Economic and Labour Market Analysis Department, International Labour Office, ILO, Geneva Jhabvala, Renana, Jeemol Unni and Shalini Sinha (2007: 9) ‘Overview and journey of home based workers in South Asia,’ paper presented at SEWA-UNIFEM Policy Conference on Home Based Workers in South Asia, January 2007, New Delhi. Joseph, George and Luis Alberto Andres (2012) ‘Gender and Growth in India: An Outline’. Kabeer, Naila (2001) ‘Resources, Agency, Achievements. Reflections on the Measurement of Women’s Empowerment’, in A. Sisask, (ed.), Discussing Women’s Empowerment: Theory and Practice, SIDA studies Stockholm, SIDA, pp. 17–54 Kabeer, Naila (2012)’Women’s economic empowerment and inclusive growth: labour markets and enterprise development,’ Centre for Development Policy & Research, School of Oriental & African Studies, Discussion Paper 29/12, October Kabeer, Naila. and Helzi Noponen (2004) ‘Social and economic impacts of PRADAN’s Self Help Group Microfinance and Livelihoods Promotion Program: analysis from Jharkhand, India’ Imp-Act Working Paper No. 11, IDS, Brighton www.ids.ac.uk/impact/publications/working¬_papers.html Kabeer, Naila, Simeen Mahmud and Sakiba Tasneem ( 2011) ‘Does paid work provide a pathway to women’s empowerment?’ Empirical findings from Bangladesh. IDS Working Paper, Vol 2011, No 375. Kanchi, Aruna (2010) ‘Women Workers in Agriculture: Expanding Responsibilities and Shrinking Opportunities.’ ILO Asia- Pacific Working Paper Series. New Delhi: International Labour Organisation. Kantor, Paula (2005) ‘Determinants of Women’s Micro Enterprise Success in Hyderabad, India: Empowerment and Economics,’ Feminist Economics, 11:3, 63-83. Kelkar, Govind (2011) ‘Gender and productive assets: Implications for women’s economic security and productivity’, Economic and Political Weekly, June 4-10, XLVI, no. 23, pp. 59-68. Kelkar, Govind (2013a) ‘The Fog of Entitlement: Women and Land in India’, paper prepared for presentation at the Annual World Bank Conference on Land and Poverty, The World Bank, Washington, DC, April 8-11 87 Kelkar, Govind (2013b) ‘At the Threshold of Economic Empowerment: Women,Work and Gender Regimes in Asia, ILO Asia-Pacific Working Paper Series Kelkar, Govind, Girija Shrestha and N. Veena (2002) ‘IT Industry and Women’s Agency: Explorations in Bangalore and Delhi’, Gender and Development Studies, Sage Publications. Khera, Reetika and Nandini Nayak (2009)”Women workers and perceptions of the National Rural Employment Guarantee Act in India”, Paper presented at the FAO-IFAD-ILO Workshop on Gaps, trends and current research in gender dimensions of agricultural and rural employment: differentiated pathways out of poverty, Rome, 31 March - 2 April Klasen, Stephen (2002) ‘Low schooling for girls, slower growth for all?’ Cross-country evidence on the elect of gender inequality in education on economic development. World Bank Economic Review 16(3), 345-373. Klasen, Stephen and Francesca Lamanna (2009) ‘The Impact of Gender Inequality in Education and Employment on Economic Growth: New Evidence for a Panel of Countries,’ Feminist Economics, 15: 3, pp. 91-132. Klasen, Stephen and J. Pieters (2012) ‘Push or Pull? Drivers of Female Labor Force Participation during India’s Economic Boom,’ IZA DP No. 6395, Institute for the Study of Labor (IZA), Bonn, February. Köhlin Gunnar, Erin O. Sills , Subhrendu K. Pattanayak, Christopher Wilfong, Rasmus Heltberg, Robin Mearns (2011) ‘Energy, Gender and Development What are the Linkages? Where is the Evidence?’ A background paper for the World Development Report 2012 on Gender Equality and Development, The World Bank Washington, DC Kundu, Amit (2008) ‘Conditions of Work and Rights of the Female Domestic Workers of Kolkata,’ Indian Journal of Labour Economics. Lagerlof, Nils (2003) ‘Gender equality and long run growth.’ Journal of Economic Growth 8, 403-426. Lahoti , Rahul and Hema Swaminathan (2013) ‘Economic Growth and Female Labour Force Participation in India’ Working Paper No. 414, Indian Institute of Management, Bangalore Mahadevia, Darshini (2001) 'Informalization of Employment and Poverty in Ahmedabad,’ in Amitabh Kundu and A. N. Sharma (eds.) Informal Sector in India: Problems and Policies, New Delhi: Institute for Human Development. Mehta, Aasha, Kapur et al., (2011) India Chronic poverty Report. Towards Solutions and New Compacts in a Dynamic Context, Indian Institute of Public Administration and Chronic Poverty Research Centre Merin S., Thadathil and Vineeth Mohandas (2012) “Impact of MGNREGS on Labour Supply to Agricultural Sector of Wayanad District in Kerala”, Agricultural Economics Research Review, Vol. 25(No.1) January- June Mirchandani, Kiran (2005) ‘Gender Eclipsed? Racial Hierarchies in Transnational Call Center,’ work Author(s): Reviewed work(s): Source: Social Justice, Vol. 32, No. 4 (102), Race, Racism, and Empire: Reflections on Canada, pp. 105-119 Published by: Social Justice/Global Options. Mitra, Amit (2008) Convergence of agricultural interventions in Maharashtra’s distressed districts. Working Paper 1 on social and gender aspects, Rome, IFAD. 88 Mitter, Swasti and Cecilia Ng (2001) ‘Teleworking and Teletrade in India: Combining diverse perspectives and visions,’ Economic and Political Weekly, XXXV, 26. Mukhopadhyay, Swapna and SureshTendulkar (2006) ‘Gender differences in Labour Force Participation in India,’ an analysis of NSS data, ISST, Working paper GN (III) WP2. Murthy, Ranjani,K., K.Raju and Amitha Kamath with research team (2005) “Towards Women’s Empowerment and Poverty Reduction: Lessons from the Andhra Pradesh South Asia Poverty Alleviation Programme” in Burra, Neera, Joy Deshmukh-Ranadive and Ranjani K.Murthy (eds) Micro-credit, Poverty and Empowerment. Linking the Triad, Sage Publications, New Delhi Neff, Daniel, Kunal Sen and Veronica Kling (2012) ‘The Puzzling Decline in Rural Women’s Labor Force Participation in India: A Reexamination,’ GIGA Working Papers No 196, German Institute of Global and Area Studies, Hamburg, May. Neetha N. 2004. Making of female breadwinners:migration and social networking of women domestics in Delhi, Economic and Political Weekly, April 24, 2004, pp 1681 – 1688. NCEUS (2007) Report of Conditions of Work and Promotions of Livelihoods in the Unorganised Sectors, National Commission for Enterprises in the Unorganised Sector, August, NIRD(2010) “Research Study on Changing Gender Relations through MGNREGS”, Sikkim State Report, July 2010, Centre for Wage Employment and Poverty Alleviation, National Institute of Rural Development, Rajendranagar, Hyderabad NSSO (2010) ‘Education in India: 2007-08 Participation and Expenditure NSS 64th Round (July 2007 – June 2008)’, National Sample Survey Office, National Statistical Organisation, Ministry of Statistics and Programme Implementation, Government of India, May O.Reilly, Arthur (2003) ‘The Right to Decent work of Persons with disability, Working Paper , International Labour Organization, Geneva O.Reilly, Arthur (2007) ‘The right to decent work of persons with disabilities,’ International Labour Organization, Geneva Palit, Amitendu (2009) “Skills Development in India: Challenges and Strategies,” ISAS Working Paper No. 89, September Panagariya, Arvind (2008) India: The Emerging Giant, Oxford University Press, New Delhi. Panda, Pradeep and Bina Agarwal (2005) Marital violence, human development and women’s property status in India’ World Development 33(5): 823-850. Parthasarathy, S. K. (2005) ‘Awareness, Access, Agency: Experiences of Sawayam Shikshan Prayog’ in Mirco-finance and Women's Empowerment. In Neera Burra, J. D. Ranadive, & R. K. Murthy, Micro- Credit, Poverty and Empowerment: Linking the Triad (pp. 200-244). New Delhi: Sage Publications. Planning Commission (2007) ‘Micro-finance and poverty alleviation: The Eleventh Five-Year Plan (2007/08–2011/12).’ Development Policy Division, Planning Commission, New Delhi. http://planningcommission.nic.in/aboutus/committee/strgrp11/str11_fpa.pdf Ramachandran, V (2003) “Gender Inequality in Education in India”, Background paper prepared for the Education for all Global Monitoring Report 2003/4, “Gender and Education for All: The Leap to Equality”, UNESCO 89 Ramachandran, V (2009) “Towards Gender Equality in Education”, Education for All – Mid Decade Assessment; National University for Education Planning and Administration. Rajivan, Anuradha (2005)”Micro-credit and Women’s Empowerment: A Case Study of SHARE Micro- Finance Limited” in Burra, Neera, Joy Deshmukh-Ranadive and Ranjani K.Murthy (eds) Micro-credit, Poverty and Empowerment. Linking the Triad, Sage Publications, New Delhi Ramakrishnan, Rajesh and Shailesh Nagar (2011) “ Small-holder Women Farmers in India’, A Discussion Paper for Oxfam India, New Delhi, Draft, January Ramesh, Golait (2004) ‘Cyber Coolies in BPO: Insecurities and Vulnerabilities of Non-Standard Work,’ Economic and Political Weekly, January 31. Rangarajan, C., Padma Iyer Kaul and Seema (2011) ‘Where Is the Missing Labour Force?’ in: Economic and Political Weekly, 46, 39, 68‒72. Rao, Nitya, Arjan Verschoor, Ashwini Deshpande, Amaresh Dubey (2010) “Gender Caste and Growth Assessment – India,” Report to Department for International Development, University of East Anglia. Rao, N (2008) "Good women do not inherit Land": Politics of Land and Gender in India, Social Science Press and Orient Blackswan, New Delhi Rao, Nitya (2011a) Women’s access to land: An Asian perspective. Paper presented at the Expert Group Meeting on ‘Enabling rural women’s economic empowerment: institutions, opportunities and participation,’ Accra, Ghana, 20-23 September. Rao, Nitya (2011b) ‘Gender, land and resource rights in India’. In Du grain à moudre. Genre, développement rural et alimentation. (Dir.) C. Verschuur. 209-245. Actes des colloques genre et développement. Berne : DDC-Commission nationale suisse pour l’UNESCO; Genève : IHEID. Rao, N (2012) Male ‘providers’ and female ‘housewives’: A gendered co-performance in rural North India. Development and Change. 43(5) Rao, Nitya (nd) ‘The Politics of Land and Gender: Critical Issues and Debates’,draft for discussion Rao, Nitya, Arjan Verschoor, Ashwini Deshpande, Amaresh Dubey (2010) “Gender Caste and Growth Assessment – India,” Report to Department for International Development, University of East Anglia. Raveendran, G. (2010) ‘Contribution of women to the national economy,’ ILO Asia Pacific Working paper- http://www.ilo.org/wcmsp5/groups/public/---asia/---ro-bangkok/---sro-new_delhi/documents /publication/wcms_146136.pdf Raveendran, G. (2012) ‘Analysis of Gender and Labour Force Participation with Four Rounds of NSS Data’ Rodrik, Dani, and Arvind Subramanian (2004) ‘Why India Can Grow at 7 Percent a Year or More: Projections and Reflections,’ IMF Working Paper 04/118, Washington, D.C. Sankar, Deepa (2012) ‘Examining Education and Labor Force Participation Progress through a Gender Lens in India’ Save the Children (2004) Evaluation of Project ‘Comprehensive Intervention on Child Domestic Work,’ Kolkata. Save the Children (2006) ‘Abuse Among Child Domestic Workers,’ Kolkata Save the Children (2009) Small Hands, Big Work: No More at your Home. 90 Shah, Mihir, Rangu Rao and P.S. Vijay Shankar (2007) ‘Rural Credit in Twentieth Century India: An Overview of History and Perspectives’ Economic and Political Weekly, Special Articles, 1351-1364 Shariff, Abusaleh (2009) “Putting People to Work: What can we learn from India’s Mass Employment Scheme? International Food Policy Research Institute New Delhi Office, New Delhi, IndiaSingh, Navsharan and Mrinalini Kaur Sapra (2007: 45) ‘Liberalization in Trade and Finance: India’s Garment Sector’ in Barbara Harris-White and A. Sinha (ed) Trade Liberalisation and India’s Informal Economy. New Delhi: Oxford University Press, pp 42-127. Soni-Sinha (2006) ‘Where are the women?’ Gender, labour and discourse in the NOIDA export processing zone and Delhi, Feminist Economics, 12 (3)- pp 335-365. Srinivasan, N.(2012) Microfinance India. State of the Sector Report 2011, Sage Publications and ACCESS Development Services, New Delhi Srivastava, Nisha and Ravi Srivastava (2010) ‘Women, Work, and Employment Outcomes in Rural India,’ in: Economic and Political Weekly, 45, 28, 49‒63. Sudarshan, Ratna, M.(2009) “Examining India’s National Regional Employment Guarantee Act: its impact and women’s participation”, Social Protection in Asia Working Paper Issue 05, May Sudarshan, Ratna M. (2012) ‘Understanding the formal-informal and ‘missing workers’ in women's labour force participation’. Sudarshan, Ratna, M. and Shrayana Bhattacharya (2008) ‘Through the Magnifying Glass: Women’s Work and Labour force participation in urban Delhi’, ILO-Asia Pacific Working Papers Series, International Labour Organization, Sub-regional Office for Asia and the Pacific, New Delhi, May Sundaram, Krishnamurthy and Suresh Tendulkar (2004) ‘The poor in the Indian labour force in the 1990s,’ Working Paper No. 128 (Delhi School of Economics, Delhi University). Swaminathan, Hema, Suchitra J Y, Rahul Lahoti (2012), ‘Individual property rights and bargaining outcomes: evidence from intra-household asset data’, Indian Institute of Management, Bangalore, Working Paper No.364 Tzannatos, Zafiris (1999) ‘Women and Labor Market Changes in the Global Economy: Growth Helps, Inequalities Hurt and Public Policy Matters,’ World Development Vol. 27, No. 3, pp. 551-569. UN Women and RDI (2011) Gender and land tenure security: Challenges and barriers to women’s entitlement to land in India. UN Women. New Delhi Vanitha , S. M. and P. S. Srikantha Murthy (2011)”An Economic Analysis of MGNREG Programme in Mysore District of Karnataka” in Agricultural Economics Research Review, Vol. 24 (Conference Number) Velayudhan, Meera (2009 )’Women’s Land Rights in South Asia: Struggles and Diverse Contexts’, Economic and Political Weekly, Vol xliv No 44, october 31, pp.74-79 World Bank (2006) World Development Report 2006. Equity and development Washington: World Bank World Bank (2010) India’s Employment Challenge. Creating Jobs, Helping Workers, Oxford University Press, New Delhi 91 World Bank (2012) World Development Report 2012: Gender Equality and Development, (http://siteresources.worldbank.org/INTWDR2012/Resources/7778105-1299699968583/7786210- 1315936222006/Complete-Report.pdf World Bank (nd) Impact Evaluation for Infrastructure, General Guidance and Existing Evidence, Washington, DC. 92