Policy Research Working Paper 9350 Political Reservation and Female Labor Force Participation in Rural India Klaus Deininger Songqing Jin Hari K. Nagarajan Sudhir K. Singh Development Economics Development Research Group August 2020 Policy Research Working Paper 9350 Abstract Despite income growth, fertility decline, and educational for participation in the labor force, and income as well expansion, women’s labor force participation in rural India as intrahousehold bargaining in the short and medium dropped precipitously over the last decade. This paper uses term. Political empowerment through reservation affected nationwide, individual-level data allow to explore whether women’s but not men’s participation in public works, but random reservation of village leadership for women also women’s participation in labor markets, income, and affected their access to suitable job opportunities, demand participation in key household decisions, with a lag. This paper is a product of the Development Research Group, Development Economics. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://www.worldbank.org/prwp. The authors may be contacted at kdeininger@worldbank.org, jins@msu.edu, ssingh12@worldbank.org, and hknagarajan@gmail.com. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Political Reservation and Female Labor Force Participation in Rural India Klaus Deininger1, Songqing Jin2, Hari K. Nagarajan3, Sudhir K. Singh1, 2 1 World Bank, Washington DC 2Michigan State University, East Lansing MI 3 Indian Institute of Management Ahmadabad, India Key Words: Political reservation, Female labor force participation, India, Public Work, NREGA. JEL Codes: H3; I38; O1 1 Acknowledgement: 1818 H St. NW, Washington DC, 20433; Tel 202 4730430, fax 202 522 1151. The authors can be contacted at kdeininger@worldbank.org; Jins@msu.edu, hknagarajan@gmail.com, and ssingh12@worldbank.org, respectively. We gratefully acknowledge funding support from World Bank, DFID, Brown University, ICSSR, and NIRD. The views presented in this paper are those of the authors and do not necessarily reflect those of the World Bank Group, its Executive Directors, or the countries they represent. We would like to thank the IRMA team led by J. P Singh, Anupam Chaterjee, Vipul Jain, and the Survey team led by T.K. Krishan for excellent support in data collection. 1. Introduction Since the 1990s, India has experienced robust economic growth, declines in fertility, expansion of education, and improved access to infrastructure, all factors that are generally associated with sustained increases in female labor force participation (Klasen 2019). Yet, female labor force participation remained low by global standards and in rural areas declined from 48 percent in 1984 to 33 percent in 2012 (Andres et al. 2017). Many rural women reduced their labor force participation and dropped out at high rates that may be difficult to reverse (Sarkar et al. 2019). As the associated loss of income may affect outcomes including women’s income and autonomy as well as household decisions on children’s education and health (Afridi et al. 2016), this is of relevance for human and physical capital accumulation and India’s ability to take advantage of its ‘demographic dividend’ in the longer term. Identifying ways to reverse or at least arrest this decline is thus a priority for policy (Fletcher et al. 2017) and especially so in light of current crises affecting the country. The literature suggests that, on the supply side, higher real wages in rural areas triggered a negative income effect, the size of which outweighed associated potential increases in labor supply (Mehrotra and Parida 2017), a tendency possibly reinforced by changes in educated women’s returns to home production vs. market participation (Afridi et al. 2018). At the same time, agricultural mechanization and manufacturing’s rising capital intensity reduced female labor demand as many women lack the education and skills that would allow them to move to higher-paying sectors. This interpretation and the importance of demand side rationing is supported by women’s strong response to workfare programs (Desai 2018; Sarkar et al. 2019). Beyond economic factors, social norms likely contributed to declining female labor force participation. In rural areas, having married women work outside the home reflects badly on their family and is deemed an indication of low status (Eswaran et al. 2013). Men’s opposition to work by their spouses indeed reduced women’s take-up of employment (Bernhardt et al. 2018). Such norms change only slowly (Kandpal and Baylis 2019) and show high persistence across generations (Dhar et al. 2019), implying that changes in gender stereotypes may be needed to trigger sustained change in female labor force participation with attendant benefits. Reservation of village leadership positions for women is an intervention with the potential to affect female labor force participation directly, by providing public goods desired by women (Chattopadhyay and Duflo 2004) and by expanding their ability to access workfare job opportunities (Deininger et al. 2019). It may also influence labor market outcomes indirectly, by giving women voice (Iyer et al. 2012) and affecting stereotypes and attitudes regarding women’s ability to perform leadership functions (Beaman et al. 2012), the status of girls vs. boys (Kalsi 2017), and the value of adolescent girls’ school enrollment (O'Connell 2018). Yet, the literature on effects of reservation on female labor force participation is largely limited to 2 looking at labor supply for public works and remains scant and inconclusive: For the manufacturing sector, Ghani et al. (2014) find that female reservation triggered an expansion of the number of informal woman- owned establishments without leading to higher female employment. In Uttar Pradesh, Bose and Das (2018) find that having a female leader increased female interest in public works as measured by the number of job cards issued and demand for work under the National Rural Employment Guarantee Scheme (NREGS) but failed to affect actual employment. In Andhra Pradesh, Afridi et al. (2017) find that in village councils where leadership was reserved for women, NREGS implementation was less efficient and leakage higher than in those where this was not the case, a finding attributed to women leaders’ limited formal education and experience. In this paper, we use individual data to assess the short- and medium-term impact of female reservation on female labor supply, identify mechanisms that might underpin such changes and, for a subsample for which such data were collected, explore impacts on female empowerment. Identification relies on the fact that, in each period, villages to be reserved were randomly chosen. We analyze individuals’ outcomes by matching data on some 66,362 individuals in 23,350 households over India’s 12 main states from the Rural Economics and Demographic Survey (REDS) to villages’ reservation history. As our data were collected when NREGS was active, we assess if exogenous exposure to female leadership improved women’s ability to take advantage of this program, potentially catalyzing broader changes. By providing estimates of the impact of female reservation in current and the previous election periods, we can assess longer-term effects on labor force participation, agency, demand for work, and involvement in household decision-making. Three findings stand out: First, contemporaneous reservation affected local governance as measured by the quality of NREGS implementation but had no measurable impact on female labor supply. Second, beyond the reserved period, female leadership reservation had sustained effects on female labor supply to public workfare and to private sector labor markets. Effects were quantitatively large (half a standard deviation) and most pronounced for married women. Third, past reservation also increased women’s income, their demand for work, and their participation in household decisions relating to spending on food items, health, and education, pointing towards potential to affect norms in the longer term. Our paper contributes to the literature in at least two respects: First, we add to the evidence regarding the impact of gender quotas by showing that, even if women leaders may lack experience or have to contend with male backlash (Gangadharan et al. 2016) so that pre-exiting gaps cannot be fully closed (Iyer and Mani 2019), politically empowering women can have positive effects in the medium term, consistent with the notion that agency problems may hinder female political participation (Casas-Arce and Saiz 2015). Second, we show that one of the avenues for political reservation to affect behavioral norms is by improving women’s economic participation, control over resources, and bargaining power. Although workfare can 3 catalyze such participation in labor markets (Deininger et al. 2019), we also find significant reservation- induced impacts on participation in regular labor markets. This supports the notion that female labor force participation, bargaining power, and control over income interact with changes in gender norms (Field et al. 2019), which can also be brought about by specific measures to change attitudes (Dhar et al. 2018) and (Jensen 2012) via long-term training and provision of information. The rest of the paper is organized as follows. Section 2 describes the institutional background and context by documenting the paradox of India’s secular decline of female labor force participation and discusses the origin, nature, and evidence of impact of the country’s reservation policy as well as its Employment Guarantee Scheme. Section 3 describes the data and estimation strategy, including balance tests to ascertain that random assignment of reservation status as mandated by legislation that was indeed implemented. Section 4 presents results regarding impacts of reservation on (i) female labor force participation (separately for NREGS-related and other employment) and the heterogeneity of these impacts by marital status and age; (ii) individual income, desire to work, and participation in household decision-making; and (iii) voice in terms of affecting the way NREGS is implemented and tests for their robustness. Section 5 concludes by discussing policy implications and suggestions for further research. 2. Background and institutional context We show that, despite favorable external conditions such as increased levels of income and education and declining fertility over the last decades, India’ level of female labor force participation declined from an already low level. This is likely due to a combination of supply- and demand-side factors including strong social norms. We discuss how reservation of local political leadership for women could possibly reverse this trend by altering social norms and, in interaction with other government policies such as NREGS, generate mutually reinforcing feedback loops between economic and political empowerment. 2.1 India’s declining female labor force participation: Evidence and policy implications Determinants and effects of female labor market participation within and across countries have been studied by a large literature. Early studies often assumed that, due to changes in countries’ economic structure, education, and fertility that are associated with growth, labor force participation would display a U-shaped relationship with income. While evidence in support of this hypothesis is weak (Gaddis and Klasen 2014), there is a strong link between female empowerment and labor force participation. Gender-friendly legal reforms have, since the 1970s, consistently triggered higher levels of female labor force participation (Hyland et al. 2019). Similarly, greater voice, in terms of women’s participation in legislative bodies, is associated with higher female labor force participation (Lv and Yang 2018). 4 The fact that India is characterized by some of the most glaring stark levels of gender inequality globally implies women’s involvement in wage work was traditionally low. 1 Yet, although sustained growth in GDP, education, and access to key infrastructure (electricity, cooking gas and piped water) vastly improved Indian women’s lives since the early 1990s, women’s labor force participation stagnated in urban areas (Klasen and Pieters 2015) and declined in rural ones, especially after 2005, for the married, and those in the 15 - 24 year age bracket (Andres et al. 2017). Changes in returns to home vs. market production may have adversely affected educated women’s labor force participation, especially before 1999 (Afridi et al. 2018). Yet, higher wealth and income by other household members also reduce women’s probability of entry to the labor force and increases the likelihood of their exit (Sarkar et al. 2019). Could policies help reverse or at least arrest this trend? Access to job opportunities has been identified as an important factor (Das et al. 2019); in fact agricultural mechanization and increased capital intensity in manufacturing limit opportunities for low-skilled females who mainly worked as casual agricultural labor while increased real wages resulted in a negative income effect that outweighed potential increases in labor supply (Mehrotra and Parida 2017). The fact that provision of low-skilled employment opportunities for women via workfare is associated with reduced female labor force exit (Sarkar et al. 2019) and significantly increased women’s participation in the work force (Desai and Joshi 2019) is often taken as support for the notion that job creation holds the key to increased female labor force participation (Chatterjee et al. 2015). Access to roads or transport is also associated with increased access to nonagricultural employment that affects women more than men, especially in communities with more egalitarian gender norms (Lei et al. 2019). Other measures to empower women can reinforce this (Fletcher et al. 2017). Social norms also have an important role in mitigating female autonomy (Debnath 2015). Evidence on spouses' preferences and community attitudes towards work by married women in central India suggests that women's labor force participation may negatively affect their spouses’ social standing, leading to many husbands being opposed to their wives' taking up of employment (Bernhardt et al. 2018). Interventions to change social norms may thus hold promise to increase female labor force participation in the medium term. Indeed, while short-term interventions involving testimonies by working women or discussions within the household had no effect (Dean and Jayachandran 2019), young rural women who, over a 3-year period, were offered training to acquire skills needed to join the business process outsourcing industry were significantly less likely to get married or have children during this period, choosing instead to enter the labor market or obtain more schooling or postschool training (Jensen 2012). Similarly, financial literacy 1 India ranks 149 of 153 in the Economic Participation and Opportunity sub-index of the 2020 World Economic Forum’s Global Gender Gap Index, before only Pakistan, the Republic of Yemen and Iraq. Although levels of gender inequality across Indian regions vary with agricultural endowments that affect demand for and value of female labor (Carranza 2014), such intra-country variation cannot explain low overall levels of female participation. 5 training and transfer of NREGS wage payments to women’s own accounts increased women’s labor supply and reduced social stigma associated with female work. Effects were concentrated in households with stronger norms against female work and consistent with increased bargaining power (Field et al. 2019). 2.2 Can political reservation affect females’ labor market outcomes? Reservation of village council leadership positions for women and scheduled castes (SCs) or tribes (STs) was introduced in India in 1993 to among others overcome long-standing inequalities and discrimination. The share of seats reserved for women is fixed at the state level and, unlike reservation for SCs, 2 seats to be reserved for women are selected randomly in every election. Female leadership has been shown to change the nature and quality of public goods supplied locally, e.g. by women leaders providing goods such as water and roads preferred by women (Chattopadhyay and Duflo 2004) and establishing role models (Beaman et al. 2009). Female leadership reservation is associated with higher rates of breastfeeding and immunization as well as higher child survival (Bhalotra and Clots-Figueras 2014). Children’s exposure to reservation in utero or early in life is associated with improved learning outcomes in primary school (Pathak and Macours 2017). 3 While reservation may trigger male backlash in the short term (Gangadharan et al. 2016), it can alter social norms and attitudes in the longer term. Female leadership increases women’s level and quality of political participation, their ability to contribute to public goods, and leaders’ accountability (Deininger et al. 2015). 4 Exposure to female leaders acting as role models triggered higher school enrollment by adolescent girls, especially those from poorer and less educated households (O'Connell 2018). It narrowed gender gaps (Beaman et al. 2012), improved female labor force participation (Duflo 2005; Iyer et al. 2012), and raised educational attainment and aspirations by girls. Changes in beliefs regarding gender roles and greater voice by women are argued to be central reasons for increased survival of higher-birth-order girls where local seats were reserved for women (Kalsi 2017). Enhanced female participation in program oversight, civic engagement, and electoral participation in ‘reserved’ villages all point towards potential complementarities between political and economic empowerment (Deininger et al. 2019). The National Rural Employment Guarantee Scheme (NREGS) has been designed to expand demand for unskilled work, especially by women. Building on the country’s long tradition of food-for-work schemes (Dutta et al. 2012; Subbarao 1997), this program guarantees up to 100 days of employment per year to 2 Beyond gender, pradhan (village council’s headship) seats can also be reserved for scheduled castes and tribes. As seats are not allocated randomly and evidence suggests that politicians’ incentives to allocate benefits along party lines may blunt such quotas’ effects (Dunning and Nilekani 2013), we focus on female reservation only. For discussion of caste reservation, see (Kaletski and Prakash 2016) and (Chin and Prakash 2011). 3 In Spain, quotas resulted in slightly better electoral results for parties most affected, suggesting that without the quota, party leaders were not maximizing electoral results due to agency problems hindering female representation in political institutions (Casas-Arce and Saiz 2015). 4 Similar outcomes are observed in West Bengal (Beaman et al. 2010), South India (Besley et al. 2005) and urban Mumbai (Bhavnani 2009). Length of exposure to women politicians is also linked to more formal sector entrepreneurship (Ghani et al. 2014). 6 households that have registered locally and established eligibility by obtaining a job card.5 Unskilled labor supplied by locals is expected to build productive assets (access roads, water harvesting structures, etc.) to increase agricultural productivity. NREGS explicitly encourage female participation by paying equal wages to men and women and requiring that a minimum share of work be performed by women. While there is considerable heterogeneity in program implementation and use of IT, e.g. electronic payment of wages directly into beneficiaries’ accounts (Muralidharan et al. 2016), major program-induced effects have been confirmed in three areas. First, NREGS increased wages, especially for women (Azam 2012), in the dry season (Imbert and Papp 2015a), and for the unskilled (Berg et al. 2014). Second, by providing a predictable source of income, it helped reduce seasonal short-term migration (Imbert and Papp 2015b), encouraged diversification of cropping patterns (Gehrke 2017), and improved agricultural productivity (Deininger et al. 2016). Finally, as the program is self-targeting, distributional effects have been largely positive: NREGS enhanced consumption (Bose 2017) and asset accumulation by the poor (Deininger and Liu 2013), affecting health (Ravi and Engler 2015), primary school participation (Islam and Sivasankaran 2015), learning outcomes in primary (Mani et al. 2014), though not secondary schools (Shah and Steinberg 2015), gender-based violence (Amaral et al. 2015), and female empowerment (Afridi et al. 2016). Yet, despite the far-reaching positive impacts on social outcomes and economic empowerment (Duflo 2005; Iyer et al. 2012), the literature finds links between political reservation and labor force participation to be ambiguous. Using state-level data, Ghani et al. (2014) find that female reservation did not increase female employment in the manufacturing sector although it triggered an expansion of the number of woman-owned establishments in the unorganized sector. In Uttar Pradesh, Bose and Das (2018) show that having a female leader increased the number of job cards issued and demand for work under NREGS but not actual program- induced employment. In Andhra Pradesh, NREGS implementation was less efficient and leakage higher in ‘reserved’ compared to unreserved village councils, a finding attributed to women leaders’ more limited education and experience (Afridi et al. 2017). 3. Data and econometric approach We use descriptive data to check for balance in pre-program characteristics between ever and never reserved villages and differences in program-affected variables that, if allocation was random, can be interpreted as causal interpretation. Data are consistent with random allocation of reservation, suggest it brought to power leaders with less formal education, and point towards gender differences in the impact of reservation on labor market participation at the extensive and intensive margins. 5 Applicants are eligible to receive a job card containing photos of all adult household members free within 15 days of application. The indicative work demands by job-card holders lead to elaboration of an annual plan that, once ratified by the village assembly, is transmitted for consolidation at the district level, although in practice a more top-down process is often followed, based on central budget allocations. 7 3.1. Data and descriptive statistics To explore possible links between political and economic empowerment, we use individual data from a complete enumeration of all adult residents in 190 villages in 13 states implemented in 2014/15 as part of the long-running ARIS-REDS panel. 6 Information was collected on 275,677 individuals in 91,984 households. Of these 23,350, generally the most disadvantaged ones, had a job card allowing household members to apply for work under NREGS. To obtain a conservative estimate of reservation-induced effects, we limit our analysis to these households. In addition to standard demographic and socio- economic characteristics at individual and household level, the survey obtained detailed information on actual and desired labor market participation at individual level. For individuals who participated in NREGS, data were gathered on key features of program implementation including whether dated work receipts were issued, payment was deposited directly in beneficiaries’ own account, if they were paid less than the statutory wage and, if yes, whether a complaint was lodged. For a subsample of states with traditionally high levels of discrimination against women, an extra module was administered asking about individuals’ involvement in key household-level decisions. In addition, a village questionnaire was also administered to, among others, elicit characteristics of all village leaders elected from 2005 together with election details, including if the election was ‘reserved’. Table 1 illustrates the timing of panchayat elections in sample states. Most states held elections in 2005/06 so that the local government had been recently elected when NREGS was launched in 2006-2008. Another round of elections was held in 2010 or 2011 and the village council leaders elected then had just completed their terms when our data were collected. Random assignment of female leadership reservation to villages provides an opportunity to assess if exposure to female leadership in the current or immediately preceding election period improved women’s ability to take advantage of labor market opportunities in NREGS or the private sector although we are unable to analyze impacts of reservation and NREGS separately. Household, individual, and village characteristics are reported in tables 2-4 separately for the entire sample (col. 1) and for villages that had or had not been reserved in the two previous election periods (cols. 2 and 3) with p-values from testing for equality of means between ever and never reserved groups reported in col. 4. 7 If, as stipulated by law, villages to be reserved were chosen randomly, covariates unaffected by the 6 The original survey, in 1971, was based on a representative sample of about 4,500 households in 252 villages in 16 states. Subsequent rounds took place in 1982, 1999, and 2006. While resource limitations precluded expansion of this exercise to all states, villages in the states of Andhra Pradesh, Bihar, Chhattisgarh, Haryana, Jharkhand, Madhya Pradesh, Rajasthan, Tamil Nadu, Uttar Pradesh, Maharashtra, Orissa and West Bengal were revisited in 2014/15 by IRMA with funding support from Brown University, German Development Institute, and the World Bank. 7 Tables providing a more detailed distinction between villages that have been reserved now and in the previous period are included in appendix tables A1 to A3. For those villages that are reserved in current period, previous period and reserved in either current or previous period, respectively, while column 4 reports the means of these characteristics of village councils that are never reserved. Relevant p-values in cols. 6-9 do not allow us to reject the hypothesis that relevant variables were balanced between the different types of villages. 8 program should be balanced between reserved and non-reserved villages while differences in any program- related outcomes can be interpreted as causal effects. Table 2 panel A presents data on the 23,350 households with job-cards and their 66,362 working-age members in sample villages. The average household includes 4.5 individuals, has a head who is aged 49 years, spent 3.8 years in school, is married in 85%, widowed in 13.6%, and female in 11.6% of cases. The data further show that 88% of sample households are Hindus, 42% belong to scheduled castes or tribes, 58% own agricultural land and 48% had a proper (pucca) house. Panel B presents means at individual level, highlighting that 29% had education at primary, 21% between primary and high school, and 11% above high school level. Neither individual nor household characteristics differ significantly between ever and never reserved villages, allaying fears about random assignment of villages to female leadership not having been adhered to. Data at village level in table 3 (panel A) suggest that sample villages are typical of rural India with population of 450 to 520 households (2,500 to 2,800 individuals), mostly Hindu (≈90%) and about one- third belonging to scheduled castes (≈21%) or tribes (≈11%). Agriculture remains the main income source for 56% of households. Some 50% of villages can access a good road or primary health care within one kilometer and 92 percent have access to a primary or secondary school within 3 km. Pradhan characteristics in panel B suggest that, in ever reserved villages, the share of pradhans who either held or contested the position of village leader before is slightly but not significantly lower in villages that had been reserved compared to those that had not. At the same time we find significant differences in leaders’ attributes between the two types of villages, consistent with the notion that female reservation opened the way for less educated non-Hindu leaders: while only 26% and 14% of leaders in ever reserved villages had secondary or higher education and 48% were Hindus, corresponding figures for never reserved villages are 42%, 19%, and 64%, respectively. 8 Table 4 presents information on individuals’ actual and desired labor market participation, involvement in household decision-making and, if they participated in NREGS, program implementation and governance with data for males in cols. 1-4 and for females in cols. 5-8. In line with the literature, data show that labor force participation rates and number of days worked by men (86% participation with 185 days worked annually) exceed those for women (62% and 64 days). Significant gender differences are visible in the way labor days are allocated across sectors. Men spend close to 50% of working time in non-agricultural casual employment followed by agricultural self-employment in (39%), casual labor in agriculture (33%), and 8 Beyond gender, pradhan seats can also be reserved for scheduled castes and tribes. As seats are not allocated randomly and evidence suggests that politicians’ incentives to allocate benefits along party lines may blunt such quotas’ distributive effects (Dunning and Nilekani 2013), we do not deal with this in detail and instead refer readers to (Kaletski and Prakash 2016) and (Chin and Prakash 2011) for further discussion. 9 salaried work (7%) and rather limited use of NREGS which accounts for less than 5% of their time. Women, by contrast, rely much more on employment in agriculture and workfare as they spend more than 60% of their time in agriculture (32% self-employed and 29% in casual labor), followed by NREGS (27%) and non-agricultural casual labor (10%). Such disproportional reliance on unskilled agricultural work makes women more susceptible to being displaced by agricultural mechanization (Mehrotra and Parida 2017) with access to workfare possibly providing a safety net uptake of which could be affected by women’s voice. As these variables may be affected by female leadership reservation, testes for significance of differences in cols. 4 and 8 are of interest. We find time use, reservation-induced effects are more pronounced for females than for males: while there is no difference in labor force participation for males between ever (87%) and never (86%) reserved villages and males even work and earn significantly more in never (188 days and Rs. 66,000) vs. ever (182 days and Rs. 63,724) reserved villages, the opposite is true for women for whom labor force participation (67% vs. 58%), number of days worked per year (67 vs. 61), and total earnings (Rs. 22,490 vs. Rs. 19,804) are all significantly higher in ever vs. never reserved villages. At the same time, willingness to work more is significantly higher for males and females in ever vs. never reserved villages. The difference is larger for women than men (9.1 vs. 4.5 percentage points), possibly pointing towards greater rationing for female labor market participation (Desai 2018). Reservation also appears to affect adherence to program rules and, for indicators in which women were particularly disadvantaged, allowed them to achieve gender parity. In ever reserved villages, the share of women who got a dated work receipt and were paid directly into their bank account increased from 62% to 68% and from 80% to 91%, respectively. Reservation does not seem to have affected the share of females who were under-paid (about 45% for ever and never reserved villages) and increased it for males (35% in never vs. 41% in ever reserved villages), though close to two-thirds of those who did not get paid the set amount did launch a complaint, much higher than those who did so in never reserved villages (39% of men, and 46% of women). For the smaller sample where such data were collected, evidence on involvement in decisions on food, non-food, health, and education suggests reservation led to significant, though quantitatively modest, increases in involvement in all these decisions by males as well as females; with 76% in ever vs 70% in never reserved villages, potential reservation-induced effects are largest for females’ participation in education decisions. 3.2. Econometric approach To assess impacts of political preference on women’s economic empowerment, we use the fact that, in each period, a predetermined share of villages is randomly chosen to have the leadership position reserved for a 10 woman. 9 Data on current and previous reservation status allows us to test for persistence of such effects, i.e., if -in line with the notion that gender attitudes change slowly with individuals altering their attitude only after having been exposed to female leadership for some time (Beaman et al. 2012)- past reservation of a village for female leadership affects current outcomes. Synergies between political and economic empowerment (Deininger et al. 2019) would yield the same result. Letting v denote villages, i individuals, and t time, we assess the impacts of female reservation on outcome variables relating to individual i’s labor force participation as well as other outcome variables by estimating the following equation. 1 2 𝑌𝑌𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 𝑅𝑅𝑣𝑣 + 𝛽𝛽2 𝑅𝑅𝑣𝑣 + 𝛽𝛽3 𝑿𝑿𝑖𝑖𝑖𝑖 + 𝛽𝛽4 𝑽𝑽𝒗𝒗 + 𝑢𝑢𝑑𝑑 + 𝜀𝜀𝑖𝑖𝑖𝑖 (1) 1 where Yiv is the outcome variable of interest for individual i in village v, 𝑅𝑅𝑣𝑣 is an indicator variable that equals one if council leadership in village v was reserved for women in the most recent election (i.e., the pradhan at the time of the survey was a woman who assumed her position as a result of reservation) and 2 zero otherwise; 𝑅𝑅𝑣𝑣 is an indicator variable that equals one if council leadership in village v had been reserved for a woman in the previous election and zero otherwise; 10 X is a vector of household and individual controls; V is a vector of village and pradhan characteristics; 𝑢𝑢𝑑𝑑 a district fixed effect; and εiv an error term. Our main interest is in β1 and β2, the parameter estimates of current or past reservation on individual outcomes relative to the base category of a village never having been reserved. To explore the gender dimension of reservation, we let fiv be an indicator variable taking a value of one if the respondent is female and zero otherwise. With interactions between respondent’s gender and current or past reservation, our estimating equation becomes: 1 1 2 2 𝑌𝑌𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 𝑅𝑅𝑣𝑣 + 𝛽𝛽3 (𝑅𝑅𝑣𝑣 × 𝑓𝑓𝑖𝑖𝑖𝑖𝑖𝑖 ) + 𝛽𝛽2 𝑅𝑅𝑣𝑣 + 𝛽𝛽4 (𝑅𝑅𝑣𝑣 𝑣𝑣 + 𝑢𝑢𝑑𝑑 + 𝜀𝜀𝑖𝑖𝑖𝑖 (2) × 𝑓𝑓𝑖𝑖𝑖𝑖 ) + 𝛽𝛽5 𝑿𝑿𝑖𝑖𝑖𝑖 + 𝛽𝛽6 𝑉𝑉 where parameters are as above and the main difference from other studies is that the parameters estimated are gender-specific. In other words, β1 and β2 are the estimated impact of current or past reservation on men and β1 + β3, as well as β2 + β4 are estimated impacts of current and past reservation on women so that the household-level impact of current reservation is given by β2 + β4. The significance of linear combinations of estimated parameters can be tested via F-tests which are reported in the results tables throughout. 9 In 2009/10 all states in our sample except Bihar and Madhya Pradesh (where the share was 50 percent) required a third of villages to reserve the pradhan position for a woman. By 2015 all except Haryana and Uttar Pradesh had increased the share of panchayats required to reserve seats for women to 50 percent. Whatever the overall share, because a village’s reservation status is exogenously given it does not affect our analysis. For a detailed discussion of how randomization is implemented, see Dunning and Nilekani (2013) and Chattopadhyay and Duflo (2004). 10 To illustrate: R1v for villages in Andhra Pradesh equals one if, in this village, the 2011 election was reserved for a woman and R2v equals one if in this village the 2006 election had been reserved. Similarly, for villages in Orissa R1v and R2v equal 1 if the 2012 or 2007 elections were reserved. 11 4. Results and discussion Regressions at household- and individual-level suggest that reservation had no concurrent impact on female labor force participation but affected modalities of NREGS implementation, e.g. if work receipts were issued and those receiving less than the stipulated wage complained. Past reservation is estimated to have led to gains in female labor force participation at the extensive and the intensive margin. Significant part of these impacts materialized via higher NREGS participation and married individuals, especially women, benefited most. Greater labor force participation in turn seems to have triggered improvements in women’s income, demand for work, and intra-household bargaining power. 4.1 Impacts of reservation on female labor market participation Table 5 reports results from regressions of labor force participation without and with gender-differentiated effects that correspond to equations (1) and (2) in panels A and B, respectively. Beyond results for overall participation along the intensive (col. 1) and extensive (col. 4) margin, estimated coefficients are reported separately for NREGS-related activities (cols. 3 and 6) and all activities except NREGS (cols. 2 and 5). Concurrent reservation is estimated to have had no impact on participation at the extensive margin. At the intensive margin, there is some evidence that introduction of NREGS crowded out non-NREGS activities with a marginally significant increase in NREGS days (coefficient of 0.128 in col. 6 of panel A) substituting for a reduction in non-NREGS related labor supply (coefficient of -0.092 in col. 5). Differentiating by gender in panel B suggests that this is driven by male labor supply. We thus cannot reject the hypothesis that, during the reserved period, there is no impact of reservation on either the extent or the intensity of overall female labor market participation. By contrast, we find highly significant gender effects of reservation in the previous period: the likelihood of labor market participation overall is estimated to have increased by 2.7 percentage points (col. 1), an effect comprised of estimated increases by 6.3 and 2.2 points for NREGS and non-NREGS work (cols. 2 and 3), respectively. A similarly highly significant overall effect -with an elasticity of 0.257 for NREGS- and 0.124 for non-NREGS-related work, respectively (cols. 6 and 5) emerges at the intensive margin. Disaggregating these effects by gender in panel B highlights that virtually all long-term impacts can be attributed to changes in women’s rather than men’s labor market participation. F-tests in the bottom rows of table 5 indicate that estimated impacts of reservation on women’s labor supply (α2 + β2) are significant at the 1% level throughout. Past reservation is estimated to have led to an 8.2 percentage point increase in the likelihood of female labor force participation, comprised of estimated increases of 15 and 6.5 percentage points in women’s likelihood of participating in NREGS and non-NREGS work, respectively. With 44% 12 overall (col. 4) -33% for non-NREGS (col. 5) and 49% for NREGS work (col. 6)- estimated elasticities at the intensive margin are even larger. Thus, even though reservation seems not to have affected women’s labor market participation in the short term, it brought a significant number of women to the labor force and increased existing participants’ labor supply in the medium term. Given the requirement for NREGS to offer conditions favorable for females, it is not too surprising to find estimated coefficients for work performed under this program to be consistently larger than for non NREGS-related work. At the same time, coefficients for non NREGS work are significant throughout and suggest that, beyond potentially affecting the modalities under which workfare was provided, reservation increased women’s demand for paid work. This is consistent with the notion of the program having performed a catalytic role, affecting social norms by having female leaders act as role models and changing level and quality of women's political participation (Deininger et al. 2015). Regressions at household level suggest that access to job cards was not affected by a village’s current or past reservation status. 4.2 Heterogeneity of effects If, as the literature suggests, the scope for labor market participation is particularly limited for married women (Eswaran et al. 2013), reservation-induced effects may be more pronounced for this group, either by providing them with economic resources and social connections that they would not otherwise have access to or by helping to change their husbands’ attitude to general gender roles and particularly female labor force participation (Bernhardt et al. 2018). To test this, we run the above regressions separately for the sub-samples of married and unmarried individuals. Results from doing so in Table 6 indeed support this notion, suggesting estimated effects are consistently more significant and larger for married than for unmarried individuals: First, in contrast to insignificant aggregate effects of concurrent reservation on labor supply in the total sample, current reservation is estimated to increase married women’s likelihood of labor force participation by 1.7 percentage points with marginal significance. The main channel for concurrent effects to materialize is via NREGS-related work, participation in which is estimated to increase by 4.1 percentage points as a result of reservation irrespectively of gender (panel A, col. 6), largely by substituting for self-employment by males and, with a slightly smaller point estimate, females. Aggregate effects of current reservation on unmarried individuals’ participation are insignificant (panel B col. 1): while the negative effect on self-employment (col. 2 and 3) is consistent with findings for married individuals, reservation has no significant effect on NREGS participation by unmarried ones, consistent 13 with the notion that they have access to different opportunities in the labor market or different returns to work at home (Afridi et al. 2018). 11 Second, past reservation is estimated to have had a gender-differentiated impact whereby a reduction in the likelihood of married males’ participation -by 3.7 percentage points- is more than compensated for by an increase in married females’ propensity to participate to yield a net increase of 9.1 percentage points overall due to female reservation. Disaggregating by type of labor suggests that most of this effect can be attributed to increased participation in NREGS activities, estimated to increase by 17 percentage points, versus 6.2 percentage point gain in non-NREGS activities. The comparison of estimated elasticities at the intensive margin between NREGS and non-NREGS demonstrates even large difference between the two types of job activities (0.62 for NREGS vs. 0.30 for non-NREGS activities). By comparison, for unmarried individuals, we find evidence of smaller effects of past reservation that do not differ by gender and are less dominated by NREGS. For example, past reservation would increase female’s probability to participate in non-NREGS and NRGES by 4.7 percentage points and 6.7%, respectively. The gain in intensive margins between non-NREGS and NRGES are even more similar as the estimated elasticities for NREGS and non-NREGS activities are 0.27 and 0.24, respectively. 4.3 Impact pathways To explore if reservation affected supply- or demand-side factors, we report effects on modalities of NREGS implementation that are likely to have affected the supply of jobs and women’s bargaining power within the household separately. Results from regressions (1) and (2) with the key indicators of program implementation in table 7 suggest that current as well as past reservation helped improve quality of program implementation in several dimensions: The share of those who received a dated receipt for work performed under NREGS (col. 1) increased significantly during the reserved period and beyond (with elasticities of 27% and 50%, respectively). The likelihood of lodging complaints in case of under-payment also increased in the reserved period (with an elasticity of about 27%), though no longer thereafter (col. 4). Significant lagged effects are observed for an increased likelihood of wages being paid directly into beneficiaries’ account (col. 2) with an estimated elasticity of 17%; the likelihood of complaints for underpayment being addressed (col. 6 with an elasticity of 24%) and possibly as a result, a reduction in the likelihood of under- payment (col. 4). While reservation has undeniably improved program governance and thus enhanced females’ ability to access jobs under NREGS, none of these effects are gender-specific; to the contrary, for some, mainly lodging and response to complaints, women are estimated to lag men. Regressions distinguishing non-NREGS related work for married and unmarried individuals along the extensive (table A5) and intensive (table 11 A6) margins are included in the appendix. 14 Results from regressions in table 8 allow us to explore if reservation increased women’s demand for work as well as their income and bargaining power. As one would expect if reservation relaxes constraints to female labor supply, it triggers significant lagged increases of women’s individual income, estimated to have increased by some 78% and females’ (but not males’) demand for work by some 16 percentage points. Although not available for the entire sample, data on intra-household bargaining power support the notion of a role-model effect of reservation having, with a lag, led to higher levels of female autonomy: The share of women who participate in decision-making on food, health, and education is estimated to have increased by 16, 14, and 7 percentage points, respectively. We conclude that, beyond improving supply of jobs that are suitable and attractive for females, reservation enhanced female decision-making autonomy and their potential and actual participation in the labor force. A possible interpretation of the above evidence is that the role model effect provided by past female leaders enhanced women’s ability to take advantage of changes in the availability of jobs, including those made available via NREGs, available to everybody. 5. Conclusion and policy implications Motivated by the recent decline of female labor force participation in India, this paper explores if random reservation of political leadership positions for women affects women’s labor force participation as well as supply- and demand-related factors. While there is no contemporaneous effect, past leadership reservation for women significantly increased females labor supply by allowing individuals to join the labor force and increasing the amount of time spent working by those already in work. While large part of the observed effects is attributable to females’ improved ability to take advantage of public workfare under NREGS, female participation in non-NREGS labor markets (especially non- agricultural casual and self-employment) expands as well. Estimated effects are stronger for married than for unmarried women. Labor force participation allows women to obtain higher levels of individual income, increases their demand for work, and affects bargaining power by enhancing their participation in intra- household decision making on spending for consumption, health, and education. Avenues to enhance these effects by combining them with targeted provision of information and training to change not only norms regarding women’s labor force participation but also equip them with the skills to adapt to changing labor market conditions are a priority area for further research. 15 Table 1: Timing for panchayat elections, NREGS roll-out and data collection 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Andhra Pradesh √ √ Bihar √ √ Chhattisgarh √ √ Haryana √ √ √ Maharashtra √ √ √ Madhya Pradesh √ √ √ Orissa √ √ √ Rajasthan √ √ Tamil Nadu √ √ √ Uttar Pradesh √ √ West Bengal √ √ √ Note: Lightly shaded areas indicate the period of roll-out of NREGS. Phase 1 of the program was rolled out in Feb 2006 in 200 districts, phase 2 of the program was rolled out in April 2007 in next 130 districts and Phase 3 of the program was rolled out in April 2008 in remaining districts of India. The darker shading in 2014 indicates the time of data collection for the survey used in the analysis. √ indicates the timing of panchayat election in states of India. 16 Table 2: Household and individual level summary statistics Reservation Status Total Difference Ever Never Panel A: Household characteristics Female head 0.116 0.115 0.117 -0.002 Head's age 49.2 49.3 48.9 0.400 Head's education 3.800 3.830 3.750 0.080 Head married 0.848 0.842 0.855 -0.013 Head widowed /separated 0.136 0.143 0.128 0.015 Household size 4.480 4.480 4.470 0.010 Males 15-65 years 1.650 1.660 1.630 0.030 Males 15-65 years 1.560 1.570 1.550 0.020 Children <15 years 1.080 1.070 1.110 -0.040 Female children <15 years 0.530 0.520 0.540 -0.020 Max. educ. in hh (years) 14.260 14.420 14.070 0.350 Hindu 0.888 0.882 0.894 -0.012 SC/ST 0.419 0.404 0.438 -0.034 Owns agricultural land 0.579 0.599 0.555 0.044* Has pucca house 0.476 0.483 0.467 0.016 # observations 23,350 12,678 10,672 Panel B: Individual characteristics Female 0.490 0.491 0.489 0.002 Age 39.8 39.9 39.7 0.200 Educ. primary. 0.213 0.208 0.218 -0.010 Educ. up to high school 0.291 0.291 0.292 -0.001 …up to graduate 0.110 0.116 0.102 0.014 Others 0.015 0.015 0.014 0.001 Married 0.748 0.742 0.754 -0.012 Unmarried 0.174 0.176 0.172 0.004 No. of obs. 66,362 34,707 31,655 Note: Author’s own calculation from 2014/15 REDS follow-up survey. To test difference in means, p values from regressions with district fixed effects and standard errors clustered by village panchayat are reported in the last column. 17 Table 3: Village level summary statistics Reservation Status T-test of Total Ever Never difference Panel A: Village characteristics Population 2,619 2,472 2,766 -294 Households 483 448 519 -710 SCs 0.208 0.204 0.212 -0.008 STs 0.115 0.120 0.110 0.010 Share of Hindu 0.898 0.901 0.894 0.007 Share in agric. 0.563 0.582 0.543 0.039 Has prim. school 0.921 0.905 0.937 -0.032 Has sec. school 0.926 0.926 0.926 0.000 Has prim health center 0.537 0.558 0.516 0.042 Has pucca road 0.505 0.516 0.495 0.021 Dist. to district HQ (km) 51.170 50.440 51.900 -1.460 Dist. to town (km) 14.940 13.970 15.910 -1.940 Dist. to bus station (km) 4.240 4.980 3.490 1.490 Dist. railway station (km) 25.130 25.590 24.660 0.930 Dist. to post office (km) 1.940 1.990 1.880 0.110 Panel B: Pradhan's Characteristics Earlier contested 0.158 0.137 0.179 -0.042 Held position before 0.474 0.442 0.505 -0.063 Up to high school 0.263 0.263 0.263 0.000 High sec. & above 0.342 0.263 0.421 -0.158 Higher education 0.163 0.137 0.189 -0.052 SC 0.537 0.579 0.495 0.084 ST 0.116 0.116 0.116 0.000 OBC 0.126 0.105 0.147 -0.042 OC 0.216 0.200 0.232 -0.032 Hindu 0.563 0.484 0.642 -0.158 Muslim 0.089 0.084 0.095 -0.011 No. of obs. 190 95 95 Note: Author’s own calculation from 2014/15 REDS follow-up survey. To test difference in means, p values from regressions with district fixed effects and standard errors clustered by village panchayat are reported in the last column. 18 Table 4: Summary statistics of labor days and labor force participation rate by reservation status Total Reservation status Test Total Reservation status Test Ever Never Ever Never Males Females Panel A: Labor supply Participated in labor market 0.86 0.87 0.86 0.01 0.62 0.67 0.58 0.09*** … self-empl. in agric. 0.39 0.41 0.37 0.04*** 0.32 0.35 0.29 0.06*** … self-empl. in non-agric. 0.06 0.05 0.06 -0.01*** 0.01 0.01 0.01 0.00* … casual labor in agric. 0.33 0.35 0.32 0.03** 0.29 0.31 0.28 0.03*** … casual labor in non-agri. 0.49 0.49 0.49 0.00 0.1 0.1 0.09 0.01*** … in NREGA 0.23 0.25 0.21 0.04*** 0.27 0.31 0.22 0.09*** … regular salaried work 0.07 0.07 0.07 0.00 0.01 0.01 0.01 0.00 No of days worked 184.8 182.3 187.6 -5.30*** 64.1 67.2 60.8 6.40*** … self-empl. in agric. 19.9 21.8 17.8 4.00*** 12.2 14.2 10.0 4.20*** … self-empl. in non-agric. 13.8 12.5 15.2 -2.70*** 2.3 2.3 2.4 -0.10 … casual labor in agric. 33.2 32.1 34.3 -2.20*** 23.3 21.9 24.7 -2.80*** … casual labor in non-agri. 90.6 88.5 92.8 -4.30*** 12.1 12.1 12.2 -0.10 … in NREGA 6.6 7.5 5.7 1.80*** 10.7 13.2 8.0 5.20*** … regular salaried work 20.8 19.9 21.8 -1.90** 3.5 3.6 3.4 0.20 Individual income (Rs.) 65,000 63,724 66,000 -2276* 20,986 22,490 19,804 2686** Would like to work more 0.274 0.296 0.251 0.05 0.316 0.359 0.268 0.09 If participated in NREGS work Got dated receipt 0.734 0.744 0.721 0.02* 0.660 0.684 0.615 0.07*** Paid directly to bank account 0.889 0.905 0.865 0.04*** 0.866 0.905 0.795 0.11*** Was paid less than was due 0.388 0.412 0.350 0.06*** 0.454 0.453 0.463 -0.01 If less, did complain 0.557 0.649 0.390 0.26*** 0.590 0.658 0.464 0.19 No. of obs. 34,427 17,990 16,437 31,935 16,717 15,218 Panel B: Intra-household decision making Participates in decisions on …. … food 0.655 0.669 0.638 0.03*** 0.839 0.851 0.824 0.03*** … nonfood 0.828 0.835 0.819 0.02** 0.761 0.769 0.751 0.02** … health 0.798 0.807 0.786 0.02*** 0.866 0.875 0.855 0.02*** … education 0.854 0.862 0.844 0.02*** 0.737 0.763 0.704 0.06*** No. of obs. 12,284 5,390 6,894 11,395 4,978 6,417 Note: Author’s own calculation from 2014/15 REDS follow-up survey. As discussed in the text, funding constraints required to limit collection of information on intra-household decision to 5 states (Gujarat, Uttar Pradesh, Maharashtra, Orissa, and West Bengal). To test difference in means, p values from regressions with district fixed effects and standard errors clustered by village panchayat are reported in the last column. 19 Table 5: Effects of political reservation on labor supply at extensive and intensive margin Participation No. of days worked Total No NREGS NREGS only Total No NREGS NREGS only Panel A Res. now (α1) 0.000 -0.016 0.034 -0.038 -0.092** 0.128* (0.008) (0.010) (0.022) (0.038) (0.045) (0.067) Res. before (α2) 0.027*** 0.022** 0.063*** 0.166*** 0.124*** 0.257*** (0.010) (0.010) (0.019) (0.049) (0.045) (0.064) Observations 66,362 66,362 66,362 66,362 66,362 66,362 R-squared 0.28 0.317 0.242 0.377 0.405 0.230 Test: F test (α1+α2=0; p val) 0.004 0.590 3.57e-05 0.013 0.546 5.54e-06 Panel B Res. now (α1) -0.022 -0.033 0.027 -0.139 -0.167* 0.039 (0.027) (0.022) (0.041) (0.117) (0.094) (0.145) Res. before (α2) -0.035 -0.011 -0.018 -0.131 -0.039 -0.158 (0.024) (0.021) (0.041) (0.121) (0.097) (0.158) Res now × fem (β1) 0.035 0.033 0.015 0.167 0.169 0.072 (0.050) (0.045) (0.062) (0.229) (0.191) (0.224) Res. before × fem (β2) 0.117** 0.076* 0.168** 0.571** 0.372* 0.645** (0.046) (0.042) (0.072) (0.230) (0.191) (0.275) Observations 66,362 66,362 66,362 66,362 66,362 66,362 R-squared 0.285 0.318 0.249 0.381 0.406 0.236 Dep. Var Mean 0.76 0.693 0.252 3.65 3.324 1.466 … males 0.87 0.84 0.23 4.51 4.34 0.71 … females 0.63 0.51 0.27 2.69 2.11 0.92 Test: F test (α1+α2=0; p val) 0.062 0.090 0.707 0.088 0.083 0.548 F test (β1+β2=0; p val) 0.007 0.030 0.0434 0.013 0.016 0.0424 F test (α1+ β1 =0; p val) 0.609 0.998 0.393 0.822 0.988 0.310 F test (α2+ β2=0; p val) 0.002 0.008 0.00198 0.001 0.003 0.00250 F test (α1+ β1+α2+ β2=0; p val) 0.002 0.021 0.00171 0.003 0.007 0.00238 Note: ‘Reserved now’ and ‘reserved before’ are indicator variables of whether village panchayats are reserved in the current or the previous panchayat periods and the sample is limited to those who worked under NREGS. Control variables included throughout but coefficients on which are not reported include household size, composition, land ownership, and the head’s marital status, gender, age, and education; village-level access to road, distance to town and district HQ, population, share of SCs, STs, and key religions; years since the last village election; pradhan characteristics (education, caste, religion, previous tenure and candidacy for office) and for individual-level regressions individuals’ gender, marital status, age, education and their squared terms. Standard errors are clustered at panchayat level. Robust standard errors reported in parentheses and multiple hypotheses tests are adjusted using the Bonferroni method. *** p<0.01, ** p<0.05, * p<0.10. 20 Table 6: Effect of reservation status on labor force participation by marital status and types of employment Participation No. of days worked Total No NREGS NREGS only Total No NREGS NREGS only Panel A: Married Res. now (α1) -0.012 -0.022 0.035 -0.089 -0.110 0.144 (0.022) (0.023) (0.035) (0.103) (0.100) (0.126) Res. before (α2) -0.037* -0.012 -0.033 -0.143 -0.035 -0.166 (0.022) (0.023) (0.041) (0.112) (0.107) (0.152) Res now × fem (β1) 0.038 0.036 0.008 0.152 0.154 0.050 (0.042) (0.042) (0.049) (0.200) (0.186) (0.183) Res. before × fem (β2) 0.128*** 0.074* 0.204*** 0.621*** 0.340* 0.779*** (0.042) (0.041) (0.067) (0.219) (0.174) (0.265) Observations 50,872 37,258 50,872 50,872 37,258 50,872 R-squared 0.284 0.335 0.247 0.417 0.461 0.249 Dep. Var Mean 0.804 0.738 0.285 3.866 3.544 0.913 Test: F test (α1+α2=0; p val) 0.0539 0.217 0.968 0.0979 0.255 0.911 F test (β1+β2=0; p val) 0.000632 0.0302 0.0195 0.00339 0.0337 0.0204 F test (α1+ β1 =0; p val) 0.272 0.536 0.156 0.576 0.665 0.0714 F test (α2+ β2=0; p val) 0.000195 0.0223 1.10e-05 0.000151 0.0165 2.89e-05 F test (α1+ β1+α2+ β2=0; p val) 1.70e-05 0.00742 3.16e-05 0.000126 0.00606 3.55e-05 Panel B: Unmarried Res. now (α1) -0.028* -0.031* 0.005 -0.184** -0.191** 0.018 (0.016) (0.017) (0.019) (0.093) (0.095) (0.066) Res. before (α2) -0.003 0.007 0.015 0.013 0.047 0.032 (0.019) (0.017) (0.020) (0.109) (0.099) (0.072) Res now × fem (β1) 0.024 0.017 0.030 0.195 0.159 0.130 (0.026) (0.023) (0.026) (0.133) (0.109) (0.102) Res. before × fem (β2) 0.070** 0.040 0.052 0.317* 0.221* 0.210 (0.033) (0.026) (0.035) (0.168) (0.132) (0.140) Observations 15,490 14,197 15,490 15,490 14,197 15,490 R-squared 0.272 0.282 0.269 0.300 0.305 0.295 Dep. Var Mean 0.603 0.545 0.146 2.864 2.593 0.494 Test: F test (α1+α2=0; p val) 0.194 0.260 0.403 0.193 0.196 0.588 F test (β1+β2=0; p val) 0.0322 0.0657 0.0676 0.0237 0.0158 0.0648 F test (α1+ β1 =0; p val) 0.839 0.475 0.0955 0.921 0.743 0.0503 F test (α2+ β2=0; p val) 0.00349 0.0179 0.00921 0.00829 0.0129 0.0119 F test (α1+ β1+α2+ β2=0; p val) 0.0296 0.181 0.00113 0.0227 0.0629 0.00143 Note: ‘Reserved now’ and ‘reserved before’ are indicator variables of whether village panchayats are reserved in the current or the previous panchayat periods and the sample is limited to those who worked under NREGS. Control variables included throughout but coefficients on which are not reported include household size, composition, land ownership, and the head’s marital status, gender, age, and education; village-level access to road, distance to town and district HQ, population, share of SCs, STs, and key religions; years since the last village election; pradhan characteristics (education, caste, religion, previous tenure and candidacy for office) and for individual-level regressions individuals’ gender, marital status, age, education and their squared terms. Standard errors are clustered at panchayat level. Robust standard errors reported in parentheses and multiple hypotheses tests are adjusted using the Bonferroni method *** p<0.01, ** p<0.05, * p<0.10. 21 Table 7: Effects of reservation on NREGS governance Get dated Payment to Payment less If less, did Complaint receipt account than assessed complain addressed Panel A Res. now (α1) 0.268*** 0.024 -0.014 0.271*** -0.014 (0.039) (0.030) (0.043) (0.015) (0.008) Res. before (α2) 0.496*** 0.174*** -0.444*** -0.048 0.241*** (0.037) (0.032) (0.034) (0.093) (0.057) R-squared 0.604 0.698 0.390 0.558 0.687 Test: F test (α1+α2=0; p val) 5.80e-11 0.00355 3.79e-07 0.0123 0.00124 Panel B Res. now (α1) 0.264*** 0.029 -0.004 0.294*** -0.007 (0.038) (0.030) (0.039) (0.023) (0.014) Res. before (α2) 0.488*** 0.175*** -0.456*** -0.036 0.254*** (0.041) (0.034) (0.035) (0.100) (0.058) Res now × fem (β1) 0.013 -0.017 -0.046 -0.052 -0.009 (0.018) (0.021) (0.035) (0.040) (0.038) Res. before × fem (β2) 0.040* -0.008 0.034 -0.067** -0.058*** (0.022) (0.010) (0.040) (0.026) (0.019) Obs. 6,736 6,736 6,736 2,747 2,747 R-squared 0.605 0.698 0.390 0.560 0.688 Dep. var mean 0.712 0.883 0.408 0.568 0.420 Test: F test (α1+α2=0; p val) 1.57e-10 0.00323 1.27e-07 0.00559 0.000463 F test (β1+β2=0; p val) 0.0931 0.346 0.848 0.000621 0.0684 F test (α1+ β1 =0; p val) 9.74e-07 0.745 0.440 3.04e-08 0.554 F test (α2+ β2=0; p val) 0 6.27e-06 3.14e-09 0.253 0.00126 F test (α1+ β1+α2+ β2=0; p val) 0 0.0108 6.78e-05 0.127 0.0126 Note: ‘Reserved now’ and ‘reserved before’ are indicator variables of whether village panchayats are reserved in the current or the previous panchayat periods and the sample is limited to those who worked under NREGS. Control variables included throughout but coefficients on which are not reported include household size, composition, land ownership, and the head’s marital status, gender, age, and education; village-level access to road, distance to town and district HQ, population, share of SCs, STs, and key religions; years since the last village election; pradhan characteristics (education, caste, religion, previous tenure and candidacy for office) and for individual-level regressions individuals’ gender, marital status, age, education and their squared terms. Standard errors are clustered at panchayat level. Robust standard errors reported in parentheses and multiple hypotheses tests are adjusted using the Bonferroni method *** p<0.01, ** p<0.05, * p<0.10. 22 Table 8: Impact of reservation and women’s participation in households’ day to day decision making Individual Wants to Participation in household decisions on…. Income work more Food Nonfood Health Education Res. now (α1) -0.071 0.013 0.022 -0.002 0.009 0.050* (0.217) (0.044) (0.040) (0.039) (0.035) (0.026) Res. before (α2) -0.195 0.006 -0.044 0.145** 0.081 0.006 (0.221) (0.047) (0.047) (0.056) (0.050) (0.035) Res now × fem (β1) 0.076 0.025 -0.104 -0.012 -0.028 -0.078*** (0.427) (0.060) (0.067) (0.042) (0.024) (0.027) Res. before × fem (β2) 0.982** 0.163** 0.213*** -0.001 0.064*** 0.066** (0.390) (0.071) (0.049) (0.039) (0.020) (0.032) Observations 66,362 66,362 22,571 22,571 22,571 22,571 R-squared 0.286 0.252 0.260 0.206 0.216 0.188 Dep. Var Mean 9.118 0.296 0.754 0.801 0.838 0.802 Test: F test (α1+α2=0; p val) 0.299 0.741 0.590 0.000212 0.00177 0.0414 F test (β1+β2=0; p val) 0.00859 0.0460 0.105 0.781 0.175 0.648 F test (α1+ β1 =0; p val) 0.984 0.318 0.0746 0.749 0.614 0.308 F test (α2+ β2=0; p val) 0.00264 0.000463 0.000425 0.0133 0.00464 0.0923 F test (α1+ β1+α2+ β2=0; p val) 0.00404 0.000176 0.0458 0.00165 0.000766 0.283 Note: ‘Reserved now’ and ‘reserved before’ are indicator variables of whether village panchayats are reserved in the current or the previous panchayat periods. Regressions for desire to work and individual income include the entire sample whereas those for intra-household bargaining is limited to the states of Gujarat, Uttar Pradesh, Maharashtra, Orissa, and West Bengal where a supplemental questionnaire on intra-household bargaining was administered. Control variables the coefficients of which are not reported include household size, composition, land ownership, and the head’s marital status, gender, age, and education; individuals’ gender, marital status, age, education and their squared terms; village-level access to road, distance to town and district HQ, population, share of SCs, STs, and key religions; years since the last village election; pradhan characteristics (education, caste, religion, previous tenure and candidacy for office). Standard errors are clustered at village panchayat and robust standard errors reported in parentheses and multiple hypotheses tests are adjusted using the Bonferroni method. *** p<0.01, ** p<0.05, * p<0.10. 23 Appendix Tables Table A1: Village Level Summary Statistics Reservation status Difference test Total Now Before Ever Never 2 vs 5 3 vs 5 4 vs 5 2 vs 3 Panel A: Village characteristics Population 2619 2565 2554 2472 2766 0.641 0.662 0.446 0.977 Households 483 471 441 448 519 0.574 0.417 0.350 0.679 SCs 0.208 0.209 0.214 0.204 0.212 0.914 0.951 0.737 0.864 STs 0.115 0.117 0.077 0.120 0.110 0.851 0.376 0.784 0.337 Share of Hindu 0.898 0.897 0.883 0.901 0.894 0.911 0.761 0.811 0.699 Share in agric. 0.563 0.550 0.565 0.582 0.543 0.874 0.654 0.320 0.777 Has prim. school 0.921 0.878 0.904 0.905 0.937 0.187 0.471 0.422 0.657 Has sec. school 0.926 0.905 0.981 0.926 0.926 0.627 0.166 1.000 0.089 Has prim health center 0.537 0.541 0.577 0.558 0.516 0.751 0.481 0.563 0.689 Has pucca road 0.505 0.514 0.462 0.516 0.495 0.810 0.703 0.773 0.569 Dist. to district HQ (km) 51.17 52.08 45.66 50.44 51.90 0.975 0.355 0.788 0.282 Dist. to town (km) 14.94 14.54 12.04 13.97 15.91 0.512 0.083 0.325 0.290 Dist. to bus station (km) 4.24 4.59 4.06 4.98 3.49 0.272 0.548 0.290 0.206 Dist. railway station (km) 25.13 23.75 24.84 25.59 24.66 0.828 0.970 0.821 0.830 Dist. to post office (km) 1.94 2.06 1.81 1.99 1.88 0.668 0.858 0.768 0.618 Panel B: Pradhan's Characteristics Earlier contested 0.158 0.176 0.212 0.137 0.179 0.956 0.633 0.429 0.617 Held position before 0.474 0.419 0.500 0.442 0.505 0.267 0.952 0.386 0.372 Up to high school 0.263 0.270 0.231 0.263 0.263 0.918 0.668 1.000 0.619 High sec. & above 0.342 0.216 0.308 0.263 0.421 0.005 0.178 0.022 0.249 Higher education 0.163 0.122 0.212 0.137 0.189 0.235 0.750 0.329 0.177 SC 0.537 0.608 0.481 0.579 0.495 0.144 0.872 0.247 0.159 ST 0.116 0.108 0.115 0.116 0.116 0.876 0.994 1.000 0.899 OBC 0.126 0.122 0.115 0.105 0.147 0.631 0.592 0.385 0.916 OC 0.216 0.162 0.288 0.200 0.232 0.267 0.451 0.599 0.090 Hindu 0.563 0.459 0.596 0.484 0.642 0.017 0.585 0.028 0.133 Muslim 0.089 0.081 0.058 0.084 0.095 0.759 0.436 0.801 0.619 # observations 190 74 52 95 95 Note: Author’s own calculation form survey. For test of difference in mean, p values reported in the last column are based on regressions with district fixed effects and standard errors are clustered by village panchayat. 24 Table A2: Household and Individual Level Summary Statistics Reservation status Difference test Total Now Before Ever Never 2 vs 5 3 vs 5 4 vs 5 2 vs 3 Panel A: Household characteristics Have a job card 12 25.36 33.26 31.63 30.78 21.30 0.072 0.567 0.142 0.393 Female head 0.116 0.119 0.126 0.115 0.117 0.571 0.076 0.617 0.201 Head's age 49.15 49.25 49.72 49.34 48.92 0.683 0.721 0.722 0.785 Head's education 3.80 3.83 4.14 3.83 3.75 0.252 0.181 0.271 0.164 Head married 0.848 0.838 0.828 0.842 0.855 0.88 0.224 0.895 0.136 Head widowed /sep. 0.136 0.146 0.153 0.143 0.128 0.57 0.189 0.598 0.832 Household size 4.48 4.37 4.37 4.48 4.47 0.617 0.022 0.809 0.562 Males 15-65 yrs 1.65 1.62 1.61 1.66 1.63 0.752 0.002 0.718 0.503 Males 15-65 yrs 1.56 1.54 1.54 1.57 1.55 0.574 0.016 0.713 0.114 Children <15 yrs 1.08 1.02 1.03 1.07 1.11 0.472 0.027 0.64 0.391 Female children <15 yrs 0.53 0.49 0.50 0.52 0.54 0.038 0.942 0.505 0.134 Max. educ. in hh (yrs) 14.26 14.31 13.94 14.42 14.07 0.095 0.601 0.119 0.292 Hindu 0.888 0.869 0.914 0.882 0.894 0.969 0.524 0.98 0.686 SC/ST 0.419 0.402 0.362 0.404 0.438 0.478 0.605 0.589 0.021 Owns agricultural land 0.579 0.563 0.618 0.599 0.555 0.04 0.144 0.062 0.231 Has pucca house 0.476 0.469 0.543 0.483 0.467 0.274 0.124 0.253 0.447 # observations 23,350 10,926 6,889 12,678 10,672 Panel B: Individual characteristics Female 0.49 0.492 0.493 0.491 0.489 0.935 0.386 0.924 0.106 Age 39.84 40.01 40.23 39.94 39.72 0.805 0.431 0.785 0.405 Educ. primary. 0.213 0.215 0.205 0.208 0.218 0.611 0.541 0.696 0.069 Educ. up to HS 0.291 0.294 0.3 0.291 0.292 0.853 0.301 0.939 0.080 up to graduate 0.110 0.113 0.131 0.116 0.102 0.8 0.224 0.848 0.153 Others 0.015 0.014 0.017 0.015 0.014 0.967 0.323 0.894 0.341 Married 0.748 0.742 0.735 0.742 0.754 0.542 0.695 0.513 0.481 Unmarried 0.174 0.174 0.18 0.176 0.172 0.894 0.597 0.997 0.505 # observations 66,362 29,042 17,705 34,707 31,655 Note: Author’s own calculation form survey. For test of difference in mean, p values reported in the last column are based on regressions with district fixed effects and standard errors are clustered by village panchayat. 12 Total households in the survey was 91,984. 25 Table A3: Summary statistics or labor supply, decision making and NREGS assessment Reserved Reserved Total Total Now Before Ever Never Now Before Ever Never Males Females Panel A: Labor Supply Total Labor force participation rate 0.86 0.86 0.85 0.87 0.86 0.62 0.66 0.71 0.67 0.58 … self-employed in agriculture 0.39 0.37 0.39 0.41 0.37 0.32 0.31 0.33 0.35 0.29 … self-employed in non-ag. 0.06 0.05 0.05 0.05 0.06 0.01 0.01 0.01 0.01 0.01 … casual labor in agriculture 0.33 0.36 0.33 0.35 0.32 0.29 0.30 0.29 0.31 0.28 … casual labor in non-agriculture 0.49 0.48 0.42 0.49 0.49 0.10 0.09 0.08 0.10 0.09 … in NREGA 0.23 0.25 0.22 0.25 0.21 0.27 0.32 0.38 0.31 0.22 … regular salaried work 0.07 0.07 0.07 0.07 0.07 0.01 0.01 0.01 0.01 0.01 No of days worked 184.84 180.28 177.61 182.29 187.62 64.13 66.79 73.88 67.19 60.76 …self-employed in agriculture 19.86 19.69 22.28 21.75 17.80 12.17 12.49 14.69 14.17 9.97 … self-employed in non-ag. 13.78 11.67 13.47 12.49 15.19 2.34 2.40 1.78 2.26 2.43 … casual labor in agriculture 33.16 34.99 37.11 32.13 34.28 23.27 22.21 25.15 21.93 24.74 … casual labor in non-agriculture 90.58 86.89 77.75 88.52 92.83 12.12 11.46 9.78 12.07 12.19 … in NREGA 6.64 7.48 7.21 7.52 5.68 10.72 14.43 19.10 13.20 7.99 … regular salaried work 20.82 19.55 19.79 19.89 21.84 3.50 3.82 3.38 3.57 3.43 Individual income (Rs.) 65,000 58,000 62,500 63,724 66,000 20,986 17,900 20,350 22,490 19,804 Panel B: Decision making and NREGS assessment Participates in household decisions on …. … food 0.839 0.808 0.874 0.824 0.851 0.655 0.591 0.826 0.638 0.669 … nonfood 0.828 0.797 0.902 0.819 0.835 0.761 0.719 0.830 0.751 0.769 … health 0.866 0.836 0.916 0.855 0.875 0.798 0.758 0.891 0.786 0.807 … education 0.854 0.827 0.886 0.844 0.862 0.737 0.667 0.792 0.704 0.763 Would like to work more 0.274 0.297 0.269 0.296 0.251 0.316 0.375 0.437 0.359 0.268 If participated in any NREGS work Got dated receipt 0.734 0.798 0.400 0.744 0.721 0.660 0.738 0.642 0.684 0.615 Was paid directly to bank account 0.889 0.903 0.888 0.905 0.865 0.866 0.903 0.836 0.905 0.795 Experienced delay in payment 0.993 0.998 0.982 0.992 0.995 0.991 0.998 0.986 0.993 0.988 delayed Was paid less than was due 0.388 0.446 0.451 0.412 0.350 0.454 0.486 0.619 0.453 0.463 If less, did complain 0.557 0.650 0.428 0.649 0.390 0.590 0.659 0.489 0.658 0.464 Complaint addressed 0.435 0.467 0.336 0.466 0.379 0.396 0.378 0.357 0.378 0.429 No. of obs. 34,427 15,018 9,135 17,990 16,437 31,935 14,024 8,570 16,717 15,218 Note: Author’s own calculation form survey. 26 Table A4: Impact of reservation on wages received Agricultural wage Non-agricultural wage NREGS wage Res. now (α1) -0.038 -0.033 0.028 0.022 0.011 -0.001 (0.031) (0.026) (0.047) (0.050) (0.025) (0.017) Res. before (α2) -0.019 -0.009 -0.023 -0.037 -0.016 -0.015 (0.041) (0.040) (0.064) (0.064) (0.021) (0.020) Res now × fem (β1) 0.001 0.001 0.008 (0.006) (0.011) (0.009) Res. before × fem (β2) -0.009 -0.001 -0.003 (0.007) (0.011) (0.010) Observations 37,250 37,250 24,803 24,803 15,721 15,721 R-squared 0.338 0.336 0.088 0.084 0.495 0.495 Dep. Var Mean 5.080 5.080 5.542 5.542 4.839 4.839 Test: F test (α1+α2=0; p val) 0.126 0.278 0.926 0.775 0.843 0.492 F test (β1+β2=0; p val) 0.350 0.981 0.767 F test (α1+ β1 =0; p val) 0.220 0.660 0.706 F test (α2+ β2=0; p val) 0.670 0.540 0.361 F test (α1+ β1+α2+ β2=0; p val) 0.189 0.766 0.619 Note: ‘Reserved now’ and ‘reserved before’ are indicator variables of whether village panchayats are reserved in the current or the previous panchayat periods and the sample is limited to those who worked under NREGS. Control variables included throughout but coefficients on which are not reported include household size, composition, land ownership, and the head’s marital status, gender, age, and education; village-level access to road, distance to town and district HQ, population, share of SCs, STs, and key religions; years since the last village election; pradhan characteristics (education, caste, religion, previous tenure and candidacy for office) and for individual-level regressions individuals’ gender, marital status, age, education and their squared terms. Robust standard errors reported in parentheses and multiple hypotheses tests are adjusted using the Bonferroni method. *** p<0.01, ** p<0.05, * p<0.10. 27 Table A5: Effect of reservation status on labor force participation by marital status and types of employment Total Self-employed Casual labor in NREGS Salaried Ag Non-Ag Ag Non-Ag Work Panel A: Married Res. now (α1) -0.012 -0.057*** -0.027*** -0.007 0.016 0.035 -0.002 (0.022) (0.016) (0.008) (0.030) (0.022) (0.035) (0.009) Res. before (α2) -0.037* -0.002 0.005 0.020 -0.049** -0.033 0.003 (0.022) (0.017) (0.007) (0.030) (0.024) (0.041) (0.008) Res now × fem (β1) 0.038 0.022 0.018** -0.039 -0.002 0.008 0.005 (0.042) (0.014) (0.008) (0.044) (0.032) (0.049) (0.009) Res. before × fem (β2) 0.128*** 0.020 -0.007 0.010 0.076* 0.204*** 0.000 (0.042) (0.013) (0.009) (0.043) (0.040) (0.067) (0.010) Observations 50,872 50,872 50,872 50,872 50,872 50,872 50,872 R-squared 0.284 0.479 0.051 0.194 0.346 0.247 0.075 Dep. Var Mean 0.804 0.401 0.0398 0.356 0.318 0.285 0.0365 Test: F test (α1+α2=0; p val) 0.0539 0.00548 0.0373 0.666 0.303 0.968 0.897 F test (β1+β2=0; p val) 0.000632 0.0135 0.370 0.573 0.172 0.0195 0.686 F test (α1+ β1 =0; p val) 0.272 0.0326 0.176 0.0910 0.489 0.156 0.615 F test (α2+ β2=0; p val) 0.000195 0.359 0.724 0.291 0.232 1.10e-05 0.610 F test (α1+ β1+α2+ β2=0; p val) 1.70e-05 0.398 0.137 0.621 0.175 3.16e-05 0.476 Panel B: Unmarried Res. now (α1) -0.028* -0.068*** -0.010* -0.040** -0.021 0.005 -0.000 (0.016) (0.020) (0.005) (0.018) (0.024) (0.019) (0.010) Res. before (α2) -0.003 0.019 -0.001 -0.021 -0.021 0.015 0.011 (0.019) (0.020) (0.007) (0.019) (0.023) (0.020) (0.009) Res now × fem (β1) 0.024 -0.007 0.020** -0.009 0.015 0.030 0.012 (0.026) (0.020) (0.008) (0.024) (0.024) (0.026) (0.010) Res. before × fem (β2) 0.070** 0.014 -0.009 0.010 0.036 0.052 0.001 (0.033) (0.023) (0.008) (0.023) (0.023) (0.035) (0.011) Observations 15,490 15,490 15,490 15,490 15,490 15,490 15,490 R-squared 0.272 0.328 0.032 0.190 0.227 0.269 0.082 Dep. Var Mean 0.603 0.247 0.0227 0.206 0.249 0.146 0.0611 Test: 0.489 0.431 0.149 0.405 0.432 0.353 0.239 F test (α1+α2=0; p val) 0.194 0.0740 0.141 0.0175 0.124 0.403 0.340 F test (β1+β2=0; p val) 0.0322 0.761 0.182 0.953 0.0697 0.0676 0.341 F test (α1+ β1 =0; p val) 0.839 0.000416 0.161 0.0200 0.792 0.0955 0.163 F test (α2+ β2=0; p val) 0.00349 0.129 0.107 0.643 0.535 0.00921 0.174 F test (α1+ β1+α2+ β2=0; p val) 0.0296 0.147 0.972 0.0120 0.725 0.00113 0.0403 Note: ‘Reserved now’ and ‘reserved before’ are indicator variables of whether village panchayats are reserved in the current or the previous panchayat periods and the sample is limited to those who worked under NREGS. Control variables included throughout but coefficients on which are not reported include household size, composition, land ownership, and the head’s marital status, gender, age, and education; village-level access to road, distance to town and district HQ, population, share of SCs, STs, and key religions; years since the last village election; pradhan characteristics (education, caste, religion, previous tenure and candidacy for office) and for individual-level regressions individuals’ gender, marital status, age, education and their squared terms. Robust standard errors reported in parentheses and multiple hypotheses tests are adjusted using the Bonferroni method. *** p<0.01, ** p<0.05, * p<0.10. 28 Table A6: Effect of reservation status on labor force participation days by marital status and types of employment Total Self-employed Casual labor in NREGS Salaried Ag Non-Ag Ag Non-Ag Work Panel A: Married Res. now (α1) -0.089 -0.219*** -0.138*** -0.070 0.087 0.144 -0.006 (0.103) (0.062) (0.042) (0.135) (0.115) (0.126) (0.051) Res. before (α2) -0.143 0.089 0.028 0.114 -0.275** -0.166 0.013 (0.112) (0.076) (0.041) (0.137) (0.127) (0.152) (0.044) Res now × fem (β1) 0.152 0.100** 0.096** -0.191 0.008 0.050 0.028 (0.200) (0.049) (0.045) (0.201) (0.173) (0.183) (0.053) Res. before × fem (β2) 0.621*** 0.049 -0.041 0.018 0.397* 0.779*** 0.001 (0.219) (0.049) (0.048) (0.203) (0.212) (0.265) (0.060) Observations 50,872 50,872 50,872 50,872 50,872 50,872 50,872 R-squared 0.417 0.466 0.050 0.205 0.363 0.249 0.077 Dep. Var Mean 3.866 1.436 0.210 1.496 1.598 0.913 0.203 Test: F test (α1+α2=0; p val) 0.0979 0.126 0.0442 0.762 0.266 0.911 0.897 F test (β1+β2=0; p val) 0.00339 0.0315 0.399 0.493 0.159 0.0204 0.671 F test (α1+ β1 =0; p val) 0.576 0.0404 0.195 0.0356 0.387 0.0714 0.516 F test (α2+ β2=0; p val) 0.000151 0.0773 0.692 0.326 0.308 2.89e-05 0.682 F test (α1+ β1+α2+ β2=0; p val) 0.000126 0.819 0.151 0.395 0.176 3.55e-05 0.455 Panel A: Unmarried Res. now (α1) -0.184** -0.221*** -0.048* -0.201** -0.100 0.018 0.003 (0.093) (0.070) (0.027) (0.079) (0.124) (0.066) (0.061) Res. before (α2) 0.013 0.115 -0.006 -0.043 -0.100 0.032 0.057 (0.109) (0.077) (0.035) (0.081) (0.120) (0.072) (0.051) Res now × fem (β1) 0.195 -0.009 0.105** -0.051 0.106 0.130 0.064 (0.133) (0.071) (0.042) (0.102) (0.121) (0.102) (0.057) Res. before × fem (β2) 0.317* 0.051 -0.047 0.027 0.160 0.210 0.003 (0.168) (0.085) (0.043) (0.100) (0.118) (0.140) (0.065) Observations 15,490 15,490 15,490 15,490 15,490 15,490 15,490 R-squared 0.300 0.322 0.031 0.196 0.229 0.295 0.083 Dep. Var Mean 2.864 0.846 0.118 0.867 1.238 0.494 0.340 Test: F test (α1+α2=0; p val) 0.193 0.258 0.161 0.0229 0.153 0.588 0.347 F test (β1+β2=0; p val) 0.0237 0.605 0.191 0.854 0.0695 0.0648 0.371 F test (α1+ β1 =0; p val) 0.921 0.00107 0.132 0.00527 0.960 0.0503 0.151 F test (α2+ β2=0; p val) 0.00829 0.0280 0.112 0.868 0.622 0.0119 0.232 F test (α1+ β1+α2+ β2=0; p val) 0.0227 0.496 0.898 0.0118 0.622 0.00143 0.0511 Note: ‘Reserved now’ and ‘reserved before’ are indicator variables of whether village panchayats are reserved in the current or the previous panchayat periods and the sample is limited to those who worked under NREGS. Control variables included throughout but coefficients on which are not reported include household size, composition, land ownership, and the head’s marital status, gender, age, and education; village-level access to road, distance to town and district HQ, population, share of SCs, STs, and key religions; years since the last village election; pradhan characteristics (education, caste, religion, previous tenure and candidacy for office) and for individual-level regressions individuals’ gender, marital status, age, education and their squared terms. Robust standard errors reported in parentheses and multiple hypotheses tests are adjusted using the Bonferroni method. *** p<0.01, ** p<0.05, * p<0.10. 29 References: Afridi, F., A. Mukhopadhyay and S. Sahoo. 2016. "Female Labor Force Participation and Child Education in India: Evidence from the National Rural Employment Guarantee Scheme." IZA Journal of Labor & Development, 5(1), 1-27. Afridi, F., V. Iversen and M. R. Sharan. 2017. "Women Political Leaders, Corruption and Learning: Evidence from a Large Public Program in India." Economic Development and Cultural Change, 66(1), 1-30. Afridi, F., T. Dinkelman and K. Mahajan. 2018. "Why Are Fewer Married Women Joining the Work Force in Rural India? A Decomposition Analysis over Two Decades." Journal of Population Economics, 31(3), 783-818. Amaral, S., S. Bandyopadhyay and R. Sensarma. 2015. "Employment Programmes for the Poor and Female Empowerment: The Effect of Nregs on Gender-Based Violence in India." Journal of Interdisciplinary Economics, 27(2), 199-218. Andres, L., et al. 2017. "Precarious Drop: Reassessing Patterns of Female Labor Force Participation in India." Azam, M. 2012. "The Impact of Indian Job Guarantee Scheme on Labor Market Outcomes: Evidence from a Natural Experiemnt." IZA Working Paper 6548. Beaman, L., E. Duflo, R. Pande and P. Topalova. 2012. "Female Leadership Raises Aspirations and Educational Attainment for Girls: A Policy Experiment in India." Science, 335(6068), 582. Berg, E., S. Bhattacharyya, R. Rajasekhar and R. Manjula. 2014. "Can Public Employment Schemes Increase Equilibrium Wages? Evidence from a Natural Experiment in India." Department of Economics, University of Bristol, UK, The Centre for Market and Public Organisation. Bernhardt, A., et al. 2018. "Male Social Status and Women's Work." AEA Papers and Proceedings, 108, 363-367. Bhalotra, S. and I. Clots-Figueras. 2014. "Health and the Political Agency of Women." American Economic Journal: Economic Policy, 6(2), 164-197. Bose, N. 2017. "Raising Consumption through India’s National Rural Employment Guarantee Scheme." World Development, 96, 245-263. Bose, N. and S. Das. 2018. "Political Reservation for Women and Delivery of Public Works Program." Review of Development Economics, 22(1), 203-219. Carranza, E. 2014. "Soil Endowments, Female Labor Force Participation, and the Demographic Deficit of Women in India." American Economic Journal: Applied Economics, 6(4), 197-225. Casas-Arce, P. and A. Saiz. 2015. "Women and Power: Unpopular, Unwilling, or Held Back?" Journal of Political Economy, 123(3), 641-669. Chatterjee, U., R. Murgai and M. G. Rama. 2015. "Job Opportunities Along the Rural-Urban Gradation and Female Labor Force Participation in India." The World Bank, Policy Research Working Paper Series: 7412. Chattopadhyay, R. and E. Duflo. 2004. "Women as Policy Makers: Evidence from a Randomized Policy Experiment in India." Econometrica, 72(5), 1409-1443. Chin, A. and N. Prakash. 2011. "The Redistributive Effects of Political Reservation for Minorities: Evidence from India." Journal of Development Economics, 96(2), 265-277. Das, M. B., et al. 2019. "Does Culture Matter or Firm? Demand for Female Labor in Three Indian Cities." The World Bank, Policy Research Working Paper Series: 8736. Dean, J. T. and S. Jayachandran. 2019. "Changing Family Attitudes to Promote Female Employment." AEA Papers and Proceedings, 109, 138-142. Debnath, S. 2015. "The Impact of Household Structure on Female Autonomy in Developing Countries." Journal of Development Studies, 51(5), 485-502. Deininger, K. and Y. Liu. 2013. "Welfare and Poverty Impacts of India's National Rural Employment Guarantee Scheme: Evidence from Andhra Pradesh." The World Bank, Policy Research Working Paper Series: 6543. Deininger, K., S. Jin, H. K. Nagarajan and F. Xia. 2015. "Does Female Reservation Affect Long-Term Political Outcomes? Evidence from Rural India." Journal of Development Studies, 51(1), 32-49. Deininger, K., H. K. Nagarajan and S. K. Singh. 2019. "Women's Political Leadership and Economic Empowerment: Evidence from Public Works in India." Journal of Comparative Economics. Desai, S. 2018. "Do Public Work Programs Increase Women's Economic Empowerment? Evidence from Rural India." IHDS working paper 2018-02, National Council of Applied Economic Research, New Delhi. 30 Desai, S. and O. Joshi. 2019. "The Paradox of Declining Female Work Participation in an Era of Economic Growth." Indian Journal of Labour Economics, 62(1), 55-71. Dhar, D., T. Jain and S. Jayachandran. 2018. "Reshaping Adolescents' Gender Attitudes: Evidence from a School-Based Experiment in India." National Bureau of Economic Research, Inc, NBER Working Papers: 25331. Dhar, D., T. Jain and S. Jayachandran. 2019. "Intergenerational Transmission of Gender Attitudes: Evidence from India." Journal of Development Studies, 55(12), 2572-2592. Duflo, E. 2005. "Why Political Reservations?" Journal of European Economic Association, 3(2-3), 668-678. Dunning, T. and J. Nilekani. 2013. "Ethnic Quotas and Political Mobilization: Caste, Parties, and Distribution in Indian Village Councils." American Political Science Review, 107(1), 35-56. Dutta, P. V., R. Murgai, M. Ravallion and D. van de Walle. 2012. "Does India's Employment Guarantee Scheme Guarantee Employment?" Economic and Political Weekly, 47(16), 55-64. Eswaran, M., B. Ramaswami and W. Wadhwa. 2013. "Status, Caste, and the Time Allocation of Women in Rural India." Economic Development and Cultural Change, 61(2), 311-333. Field, E. M., et al. 2019. "On Her Own Account: How Strengthening Women's Financial Control Affects Labor Supply and Gender Norms." National Bureau of Economic Research, Inc, NBER Working Papers: 26294. Fletcher, E. K., R. Pande and C. Troyer Moore. 2017. "Women and Work in India: Descriptive Evidence and a Review of Potential Policies." Harvard University, John F. Kennedy School of Government, Working Paper Series. Gaddis, I. and S. Klasen. 2014. "Economic Development, Structural Change, and Women's Labor Force Participation: A Reexamination of the Feminization U Hypothesis." Journal of Population Economics, 27(3), 639-681. Gangadharan, L., T. Jain, P. Maitra and J. Vecci. 2016. "Social Identity and Governance: The Behavioral Response to Female Leaders." European Economic Review, 90, 302-325. Gehrke, E. 2017. "An Employment Guarantee as Risk Insurance? : Assessing the Effects of the Nregs on Agricultural Production Decisions." The World Bank, Policy Research Working Paper Series: 8015. Ghani, E., W. R. Kerr and S. D. O'Connell. 2014. "Political Reservations and Women's Entrepreneurship in India." Journal of Development Economics, 108, 138-153. Hyland, M., S. Djankov and P. K. Goldberg. 2019. "Gendered Laws." The World Bank, Policy Research Working Paper Series: 9080. Imbert, C. and J. Papp. 2015a. "Labor Market Effects of Social Programs: Evidence from India's Employment Guarantee." American Economic Journal: Applied Economics, 7(2), 233-263. Imbert, C. and J. Papp. 2015b. "Short-Term Migration, Rural Workfare Programs and Urban Labor Markets: Evidence from India." Working Paper, University of Oxford, Oxford. Islam, M. and A. Sivasankaran. 2015. "How Does Child Labor Respond to Changes in Adult Work Opportunities? Evidence from Nrega." Harvard University, Cambridge, MA. Iyer, L., A. Mani, P. Mishra and P. Topalova. 2012. "The Power of Political Voice: Women's Political Representation and Crime in India." American Economic Journal: Applied Economics, 4(4), 165-193. Iyer, L. and A. Mani. 2019. "The Road Not Taken: Gender Gaps Along Paths to Political Power." World Development, 119, 68- 80. Jensen, R. 2012. "Do Labor Market Opportunities Affect Young Women's Work and Family Decisions? Experimental Evidence from India." Quarterly Journal of Economics, 127(2), 753-792. Kaletski, E. and N. Prakash. 2016. "Does Political Reservation for Minorities Affect Child Labor? Evidence from India." World Development, 87, 50-69. Kalsi, P. 2017. "Seeing Is Believing- Can Increasing the Number of Female Leaders Reduce Sex Selection in Rural India?" Journal of Development Economics, 126, 1-18. Kandpal, E. and K. Baylis. 2019. "The Social Lives of Married Women: Peer Effects in Female Autonomy and Investments in Children." Journal of Development Economics, 140, 26-43. Klasen, S. and J. Pieters. 2015. "What Explains the Stagnation of Female Labor Force Participation in Urban India?", 29(3), 449- 478. 31 Klasen, S. 2019. "What Explains Uneven Female Labor Force Participation Levels and Trends in Developing Countries?" World Bank Research Observer, 34(2), 161-197. Klasen, S. 2020. "From ‘Metoo’ to Boko Haram: A Survey of Levels and Trends of Gender Inequality in the World." World Development, 128(forthcoming). Lei, L., S. Desai and R. Vanneman. 2019. "The Impact of Transportation Infrastructure on Women's Employment in India." Feminist Economics, 25(4), 94-125. Lv, Z. and R. Yang. 2018. "Does Women's Participation in Politics Increase Female Labor Participation? Evidence from Panel Data Analysis." Economics Letters, 170, 35-38. Mani, S., J. R. Behrman, S. Galab and P. Reddy. 2014. "Impact of the Nregs on Schooling and Intellectual Human Capital." GCC Working Paper Serice, GCC 14-01, University of Pennsylvania, Population Studies Center, Philadelphia, PA. Mehrotra, S. and J. K. Parida. 2017. "Why Is the Labour Force Participation of Women Declining in India?" World Development, 98, 360-380. Muralidharan, K., P. Niehaus and S. Sukhtankar. 2016. "Building State Capacity: Evidence from Biometric Smartcards in India." American Economic Review, 106(10), 2895-2929. O'Connell, S. D. 2018. "Political Inclusion and Educational Investment: Estimates from a National Policy Experiment in India." Journal of Development Economics, 135, 478-487. Pathak, Y. and K. Macours. 2017. "Women’s Political Reservation, Early Childhood Development, and Learning in India." Economic Development and Cultural Change, 65(4), 741-766. Ravi, S. and M. Engler. 2015. "Workfare as an Effective Way to Fight Poverty: The Case of India's Nregs." World Development, 67, 57-71. Sarkar, S., S. Sahoo and S. Klasen. 2019. "Employment Transitions of Women in India: A Panel Analysis." World Development, 115, 291-309. Shah, M. and B. M. Steinberg. 2015. "Workfare and Human Capital Investment: Evidence from India." National Bureau of Economic Research, Inc, NBER Working Papers: 21543. Subbarao, K. 1997. "Public Works as an Anti-Poverty Program: An Overview of Cross-Country Experience." American Journal of Agricultural Economics, 79(4), 678-683. 32