On a hot summer day in 2015, I sat down alongside Sushila, a middle-aged woman who is the leader of her village’s women’s movement.Footnote 1 We sat on a woven cotton blanket on her front porch sipping chai while she recounted her transformation from being a woman constrained to the domestic sphere and forbidden to leave her house without permission into being a woman with power and voice in village politics today. Sushila’s personal transformation mirrored a transformation in her village, where historically women were not even political outsiders, but seen as apolitical, and now are a political force to be reckoned with.
Reflecting on years past, Sushila shared that she was married at the age of twenty to a man she had met only once. As is customary in much of India, she moved in with her husband and his parents just after they got married. This left Sushila more than 100 kilometers from her family and the friendships she had developed throughout childhood and adolescence. In her new village, her only ties were through her husband. She grew nostalgic as she reflected on her life before marriage, stating simply that during that time, “she was free.” Her prior freedom stood in stark contrast to her life after marriage, where Sushila fulfilled the set of obligations expected of her as caretaker of the house despite her own desire to work. Instead, she was expected to cook and clean so that her husband could tend to the small plot of land that was their primary source of income. Sushila also helped harvest the crops and tend to the soil, but her family (like most official definitions) did not consider this work. The customary assumption is that he who plows the fields reaps the financial rewards from the crops. Yet plowing is the exclusive purview of men. And so, despite putting in much of the work to harvest and produce the crops sold at the market, Sushila received no credit for her family’s earnings from the land.
Like most of the women I spoke to over six years of fieldwork in rural India,Footnote 2 Sushila’s life revolved around her domain: the household. She fetched the water, prepared the meals, and cared for her children, spending the majority of her day inside the house. The presiding norms – the invisible rules arbitrating which behaviors were deemed possible and permissible and which merited social sanction – defined the division of labor in Sushila’s house as well as where she was allowed to travel, to whom she could speak, and how she was expected to spend her time. Sushila’s husband, Nandkishor, spent the majority of his days outside the house. He would leave in the morning to work in their fields or take their crops to market. When he completed his business, he would often spend time in the village center, chatting with other men before returning home. Nandkishor, like most men, did not need to ask permission to leave the village, go to the market, or socialize with others in the community.
Even as I spoke with Sushila, a cluster of men sat together huddled around a deck of cards a few hundred yards away. Sushila said it was common for men in the village to sit together to drink and gamble. To Sushila, these activities were not merely reflective of the differences in the social lives of men and women, but tied to more insidious challenges faced by women in her village. She shared that alcoholism was a challenge in their village and that many women lived in fear of male aggression. She told me about a woman in a nearby village who had approached the local elected official to help her file a legal claim against her husband for regularly assaulting her while intoxicated. The official suggested adjudicating the matter informally rather than taking it to the courts. He convened a meeting of the family, at which he chastised her husband for the violence he perpetrated. He told the woman that she need not approach the police, as the matter was now resolved.Footnote 3 This story had convinced Sushila that her local institutions would not protect her from men’s violence and power.
Sushila, like women in much of the world, is subject to a political order rooted in patriarchy, the de facto allocation of power to elder males. This political order is marked by the absence of women and the centrality of men in most domains of politics and the devaluation of women’s voices in political decision-making. This book exposes the patriarchal political order and documents its tangled relationship with many structures of political power.
Many have called attention to the persistent and prevalent gender inequalities in politics, noting women’s underrepresentation in electoral office, at the ballot box, and in other interactions with the state. Such inequalities are most prevalent in the Global South: World Values Survey data (2005–2022) reveal in Figure 1.1 that men participate in politics in these countries at substantially higher rates than women (reflected in the positive gender gaps). These inequalities have largely been attributed to characteristics that men have but women lack: money, time, skills, social status, inclinations, and opportunities. According to these arguments, women could enter politics if only they had what men have; their level of engagement in politics is their (often rational) response under these constraints.
Yet there is an important and unexplained puzzle in women’s political participation: while women are markedly less politically present than men between elections, women vote at high rates, almost equal to those of men (see Figure 1.1).Footnote 4 This pattern of substantially larger gender gaps in non-electoral political participation is present across democracies in the Global South, but nowhere is it more acute than in India, where the gender gap is roughly 2 percentage points for voting, but nearly 20 percentage points for non-electoral political participation.Footnote 5 More than half of the women who voted on election day were absent from politics afterward. In fact, India has the lowest level of women’s non-electoral political participation of the ninety-one democratic countries surveyed.
Why do women vote but not participate in politics between elections? Prior explanations of gendered political behavior fail to explain this puzzling pattern of participation, rarely distinguishing between the drivers of electoral and non-electoral political behavior. Yet they provide two foundational explanations of gendered political behavior: women’s relatively lower levels of political participation are due to resource inequalities (lack of money, time, and skills) and social inequalities (lack of social status and inclination). Access to resources lowers the costs of political participation by facilitating the accumulation of relevant information and easing the financial and procedural barriers to participation.Footnote 6 Social inequalities, largely seen as the product of norms that socialize women into domestic and docile roles and sanction those who deviate from these prescribed roles, also condition the costs and perceived benefits of political participation.Footnote 7
Both explanations yield an expectation that development, and its consequent economic growth and norm renegotiation, will improve women’s political participation. Economic growth is assumed to generate political inclusion, and accordingly, the gendered lag in access to economic prosperity perpetuates women’s exclusion.Footnote 8 Economic growth increases incomes, expands job opportunities, and improves state capacity and, as a result, service provision. Such economic gains enable households to invest more evenly in both genders; therefore, economic growth ensures more gender-equal access to the constituents of development: health, education, and earning opportunities.Footnote 9 Women’s greater access to economic resources increases their bargaining power within the household and provides easier access to information, broader networks outside the household, and incentives to invest in young girls’ education.Footnote 10 Even cultural- or norm-based explanations of women’s exclusion suggest that the norms constraining women’s behavior are most likely to erode when they gain economic power.Footnote 11 As the value of women’s production rises (as opposed to their value in reproduction), norms related to which behaviors are considered acceptable for women tend to shift.Footnote 12 Girls are then socialized alongside boys into roles associated with economic productivity.
But these economic and social inequalities are only half of the story. Many of these models treat women as atomized individuals endowed with resources and constraints that either facilitate or hinder their political action.Footnote 13 But they fail to identify how physical and psychological coercion constrain women’s political agency and action. Feminist theorists and scholars of empowerment (who largely emanate from the Global South and South Asia in particular) have long drawn attention to the coercive structures that control women’s behavior.Footnote 14 They focus on the concepts of freedom and agency, or the ability to act in line with one’s strategic life goals, as pivotal to our understanding of gender inequities. In the domain of politics, feminist theorists have honed in on the household as a locus of disempowerment for women.Footnote 15 More recent research, also largely rooted in South Asia, has empirically shown how household members manipulate and constrain women’s political behavior.Footnote 16
I combine these two paradigms – rational and coercive explanations of women’s political behavior - and proffer a strategic answer to the puzzle of women’s political participation: many women participate in politics only when it serves the interests of men. Women’s voting is of benefit to men in systems of clientelist mobilization (where electoral support is exchanged for private benefits), while their more general political participation threatens male authority without reaping rewards for men.
While it is true that many women lack the resources that incline and support men’s political action, these resources alone do not explain the variance in women’s political participation. In addition, we must ask: who benefits from women’s political exclusion? On close inspection, it is the men in Sushila’s community, including those in her household and the elites who run village politics, who gain from her submission to the political order and who have the power to (often violently) enforce this submission. Gendered patterns of social and economic inequalities and the capacity for coercion rooted in patriarchal norms and permissive legal structures delineate de facto authority and power and generate incentives to build and maintain a political order that maximizes men’s welfare – the patriarchal political order.Footnote 17
The patriarchal political order is thus defined not only by women’s limited political participation but also by their political disempowerment – their inability to exercise free choice. For many women, men are the strategic actors deciding their political behavior. Or, more accurately, women are the pawns in men’s political games, enabling men to extract greater spoils.
At the center of these political games is the household. Patriarchy is fundamentally based on the allocation of power within the household to elder men. The household has long been considered a critical unit of analysis. It was once treated as a cohesive and aligned unit,Footnote 18 but later revealed to be a space for bargaining and negotiation as household members navigate distinct preferences but joint decisions.Footnote 19 In addition to being spaces of collective decision-making, households are domains of coercion. One in three women around the world reports violence at the hands of a male household member.Footnote 20
In the patriarchal political order, the household is the fundamental unit of political organization. Thus, when men dominate the household, they dominate politics. In close-knit political communities like the thousands of villages in India, in which electoral patronage and clientelistic exchange are commonplace,Footnote 21 political entrepreneurs benefit from organizing politics around households.Footnote 22 By treating the household, the fundamental organizing social structure, as a political unit, the costs of political mobilization, particularly with patronage, are lower. But households are also the principal domain of patriarchy, where patriarchal hierarchies are most explicitly defined. Those with power, both legitimate and coercive, within the household, namely elder men, have incentives to maximize their personal gains from politics by subordinating other household members. Such systems of political mobilization and exchange under patriarchy thus benefit from the alignment of households (and the subordination of the women within them).Footnote 23
In this book, I will show that households in rural India behave as a unit in political decision-making but that women’s agency is often subordinated by strategic and powerful men.Footnote 24 The coercive unitary household enforces women’s voting and restricts women’s non-electoral participation because these responses align with the incentives of those with bargaining and coercive power in the household. The institutionalization of the coercive unitary household yields a structure of politics organized around identities shared within households (namely caste) and where men inhabit the center of village politics and women exist on the periphery. These facts align with a political order strategically built on women’s political exclusion and explain the puzzle of women’s political participation.
Women’s political exclusion therefore persists because it benefits those with economic, normative, and coercive power. No amount of money, education, social status, inclination, or opportunity will enable women’s political empowerment unless it also allows them to contest male coercion. Yet women can challenge this political order even where patriarchal norms remain strong, and without changes in their stocks of the resources thought to facilitate political action.
How is the patriarchal political order unraveled? Returning to my day with Sushila and her experiences at present, Sushila discussed a recent village assembly meeting, where she sat among a mass of women at the front. She described speaking up at the meeting as a representative of her women’s group and articulating their concerns – a lack of water, an absent teacher, and the prevalence of domestic violence at the hands of inebriated husbands – demanding responsiveness from local politicians. She also documented how this political action was met with challenges: her husband’s disapproval of her newfound political voice and the experience of being forcibly removed alongside other women from an earlier village meeting, as it “was not their place.” Yet she remained strident in her desire for women to politically mobilize.
Sushila’s public presence and informal community leadership mark a drastic shift from her life right after marriage. She attributes this change to her joining a women-only credit group, known in India as a self-help group (SHG), seven years prior. She joined this group so that she and her family could access cheaper credit and learn about new farming techniques. But when asked what has changed most in her life since joining the SHG, she replied decisively that it was the depth of her connection with other women in the community. She stated, “We overcame our fear when we met together. Alone, we were very frightened. We took each other’s support when we were together. With the support of our sisters, our fear disappeared gradually.”
I will show that women can gain autonomy from the household and challenge the patriarchal political order through collective action. This collective action is made possible by strong political ties, a common gender consciousness, and social solidarity among women. Sushila’s experiences reflect the power generated by women’s collective action. A credit group may seem an unlikely place for political empowerment. Yet, as I will causally demonstrate, such institutions enable women’s autonomy from the household and can in some cases build women’s social solidarity around a shared gender identity.Footnote 25 In turn, this social solidarity can foster collective action to demand political agency and representation. Dense and solidaristic ties among women, built on norms of reciprocity and trust, channeled toward demands for political representation, are effective at increasing women’s political participation and countering subsequent male backlash.
This book documents the patriarchal political order and then unravels it by demonstrating the power of public policy and women’s action to reshape Indian women’s political lives. At its core, this book is about the nature of governance in Indian villages, and how existing governance structures, including those of clientelism, are built on the sustained political exclusion of women. It accounts for an entire gender system that subordinates women – the patriarchal political order – highlighting the complexity of their political inclusion and the ways in which identity can shape power in democratic systems. It sheds light on the political worlds and networks in which women reside and illustrates how women’s most intimate network, the household, shapes their political behavior. It highlights the central role of violence and coercion in suppressing women’s political voices, but also demonstrates how (and when) policies can give women the tools to overcome this subordination. This book is also about development and the unconventional and unanticipated ways it is tied to women’s political representation in modern democracies. It examines cases of women, like Sushila, who have found their voice in politics, and unearths the process and instruments of their political empowerment.
In this endeavor, I analyze a variety of novel data sources, including surveys and interviews, and multiple methodologies, such as natural experiments and network analysis, to explore the experiences of women across villages in one state of India, Madhya Pradesh, that are in most ways indistinguishable but where women’s political lives differ in important ways. I analyze data on women’s and men’s political networks from an original census of all adult residents of six Indian villages and reveal the substantial gender inequalities in political ties and influence as well as the particular importance of extra-household political relationships for women’s political behavior. It is abundantly apparent that village political networks center on men. These data, alongside data from several hundred qualitative interviews in the same communities, also demonstrate how households and communities jointly constitute women’s political behavior, often through the deployment of fear and violence. I then wield the tools of causal inference to evaluate the unintended consequences of a series of public policies designed to increase women’s financial inclusion through microcredit groups (SHGs) for women’s political empowerment. I find that bringing women together outside of their homes can foster a collective identity, generate collective action, and ultimately increase political agency and participation.
The Puzzle of Women’s Political Participation in India
As India is the world’s largest democracy and the home of the world’s most expansive political gender equality policy,Footnote 26 we would expect the political lives of Indian women to be flourishing. Multiple studies have documented the immense gains in women’s representation generated by an electoral quota policy that ensures that more than 1.3 million women hold elected office in local governments.Footnote 27 This policy has enabled women to gain access to previously exclusionary political institutions and to accumulate power to achieve their unique demands.Footnote 28 As expected by those who tout institutional solutions to political gender gaps, this quota policy has also been shown to raise women’s political participationFootnote 29 and shift attitudes about the acceptability and competence of women’s political leadership.Footnote 30 India’s most recent national election had the highest turnout of women in history, where more women turned out than men in many parts of the country.
However, the cascade of gender equality envisioned with the widespread imposition of electoral quotas appears to have left the majority of India’s women behind. World Values Survey data (2005–2022) show that only one in five women in India reported having engaged in any of the measured political actions other than voting (i.e., protesting, petitioning, striking, political occupation, and other political action). Yet nearly half of Indian men reported engaging in politics outside elections, ranking alongside countries lauded for political representation, such as Finland, the Netherlands, Japan, and Germany. Women in India ranked alongside those in Russia, South Africa, and Mexico, countries known for political inequality. Representative data from the Indian Human Development Survey, shown in Figure 1.2, corroborates women’s low levels of non-electoral political participation: fewer than 10 percent of women reported attending village assembly meetings, the cornerstone of village government, in a majority of Indian states, which is all the more puzzling given their high (and rising) voter turnout.
Local gender quotas have also failed to transform women’s political representation outside of reserved seats in local offices. Data from four state election commissions reveal that women win only around 10 percent of seats not reserved for them in local elections.Footnote 31 Furthermore, as of 2019, only 14 percent of members of parliament and 7 percent of members of state legislative assemblies were women (see Figure 1.3). Panel A of Figure 1.3 demonstrates that women’s representation in the national parliament has improved only marginally over the past six decades and remains steadily below the global average for democracies. In 2019, India ranked among the ten democracies with the lowest national parliamentary representation of women.Footnote 32 Panel A of Figure 1.3 also shows that women’s electoral representation in India is markedly below that of Scheduled Castes (SCs) and similar to that of Scheduled Tribes (STs), even though women make up roughly 48 percent of the population as compared to 17 percent and 9 percent for SCs and STs, respectively.Footnote 33 Women’s representation in state assemblies is even worse than in the national parliament: in no state do women account for more than 15 percent of state legislative assembly members (see Panel B of Figure 1.3). Clearly, widespread descriptive representation driven by a quota policy, even one as expansive as India’s, neither explains nor closes the bulk of the gender gap in political participation and representation.
Scholars of norms would suggest that these persistent and stark gender gaps are rooted in deeply entrenched patriarchal norms.Footnote 34 India is indeed an extreme example with respect to the strength of its patriarchal attitudes. As Figure 1.4 demonstrates, more than 50 percent of women and 65 percent of men in India state that they believe men make better political leaders than women, and this is representative of general patterns in South Asia. While such patriarchal attitudes are common across the globe, they are particularly acute in Africa, the Middle East, and South Asia. The extent of these beliefs among both women and men makes India a hard case in which to observe instances of women’s political empowerment, for if a majority of people do not believe women can be competent political actors, why would they support their political inclusion?
Yet there are important cases of women challenging these norms and actively engaging in politics. For example, the Gulabi Gang is an informal group of women in north India known for their pink saris who have fought to reduce domestic violence and improve women’s political representation. Sushila’s story is another example of empowerment where women in the same community and without a shift in patriarchal norms (as evidenced by the resistance they faced) became politically active. Furthermore, norms alone do not explain the patterns of political participation displayed in Figure 1.1. A 2022 report produced by the Pew Research Center using novel survey data found substantial variation across Indian states with respect to the belief that men make better political leaders than women.Footnote 35 Strikingly, the states with the highest rates of women’s political participation (see Figure 1.2) are those with the strongest patriarchal attitudes. Himachal Pradesh, Kerala, Andhra Pradesh, and Assam (the four states with the highest reported village assembly participation by women) ranked in the top half with respect to beliefs that men make better political leaders; and Odisha, Delhi, and Punjab (three of the four states with the lowest women’s village assembly participation) ranked in the bottom half, with the vast majority reporting beliefs of gender equality.
The last common alternative explanation for women’s low political participation centers on resource inequalities across the genders. And, here too, stark inequality persists in India. Most notably, the sex ratio in India remains one of the most skewed in the world, with 877 women for every 1,000 men.Footnote 36 As Sen first noted, millions of women are missing in India.Footnote 37 And while girls have made strides in education, now achieving near-identical educational attainment as boys, women’s labor force participation remains one-third that of men’s and is the lowest in the G-20 with the exception of Saudi Arabia. And, as in much of the world, gender wage gaps persist: women in India earn 28 percent less than men on average.Footnote 38
However, these economic and resource inequalities imperfectly explain women’s non-electoral political participation in India. All three panels of Figure 1.5 – which uses World Values Survey data to illustrate the relationship between education, income, caste category, and non-electoral political participation – show that resources have only a limited ability to predict non-electoral participation. Panel A, for example, shows that the gender gap in non-electoral political participation is greatest when comparing women and men with no formal, middle, or secondary education. The gender gap disappears only for those who have completed a university degree. Similarly, Panel B documents that the gender gap in non-electoral participation increases with household income. While surprising, this pattern mirrors that found with respect to female labor force participation, where labor force participation is largely uncorrelated with educational attainment in India.Footnote 39 Furthermore, my analysis of survey data from men and women in Madhya Pradesh reveals that 78 percent of the variation in men’s and women’s political participation is unexplained by differences in educational attainment, labor market participation, income, land ownership, and free time.Footnote 40 This corroborates studies from the Global South that find little or no evidence of a link between economic development and women’s political participation.Footnote 41
These patterns reveal that women’s political participation is puzzling both because of the way participation is distributed across voting and non-electoral participation and because of the limited predictive power of institutions, norms, and resources alone. This book provides an alternative explanation of broader patterns of politics in rural India that explains women’s continued absence from politics.
The Political Lives of Indian Women: Argument in Brief
The Patriarchal Political Order
Women’s lives are shaped by the institutions in which they operate, starting at the most micro level, the household.Footnote 42 Households are also the ascribed domain of women in societies that adhere to patriarchal norms.Footnote 43 While men are expected to provide economically for their families, women are expected to handle the domestic responsibilities. Men, particularly elder men, exercise de facto authority in household decisions under patriarchy. Patriarchal authority in the household is upheld by a broader social system that elevates and institutionalizes men’s authority and power. Patriarchal norms are deployed to constrain women’s autonomy from the household by, for example, limiting their labor force participation, mobility, and social interactions.Footnote 44
I argue that in rural India, women’s political lives revolve around their household, and that many women lack complete agency over their political decision-making. Instead, households make political decisions jointly, and because of intra-household inequalities and the potential for socially sanctioned and legally tolerated coercion by men, women often concede political authority and representation to the men in their families. Men, as a result, generally act as their households’ political representatives, which enables them to drive household behavior, advocate for their specific interests, and build political capital.
This pattern – what I call the institution of household political cooperation – yields high rates of electoral participation from both men and women in the household, because all-household voting is strategically valuable for those with power in the household. The more household representatives who show up to vote, the greater the likelihood of achieving the household’s preferred electoral outcome, and the greater the likelihood of patronage and clientelistic benefits for its political representatives. Women’s voting therefore serves the interests of the men in their household.
However, women (and younger men) are not expected to participate in non-electoral politics, because it has little strategic value for those with power in the household. Since the marginal return of an additional household representative participating in these spaces is low, women’s non-electoral participation does not improve the outcomes for household members who have bargaining and coercive power over women’s political behavior. In fact, women’s non-electoral participation is likely to incur social costs for both women and the men in their household, as such norm-deviant behavior will be sanctioned to ensure the continuation of patriarchal dominance.
Household political cooperation therefore implies the appearance of equality in voting but limited autonomy over vote choice in practice and the persistence of gender-based political inequalities in non-electoral political institutions. Further, men’s preferences dominate demand-making, and women’s distinct interests remain underrepresented.
Household cooperation can be both a rational response to a service delivery system that privileges households as an institution and the irrational consequence of within-household patterns of coercion and inequality. Given a political system in which many goods are either privately or locally distributed, and where patronage is a common means of service delivery, households – and, particularly, household members who have power – benefit from mutual cooperation. Households have a shared set of interests rooted in common household identities, though women also have a set of unique demands separate from those of the men in their household.Footnote 45 Yet, given intra-household power hierarchies and the potential for coercion, households can also be wielded as a tool for the benefit of the powerful. The promise of an entire household’s electoral and political support increases the expected likelihood of political returns for the political head of the household who brokers these relationships. And with the potential for coercion, household cooperation is not efficient, especially given the likelihood that women will internalize the preferences of men as a self-protection mechanism. As Sen powerfully suggests, “the family identity may exert such a strong influence on our perceptions that we may not find it easy to formulate any clear notion of our own individual welfare.”Footnote 46
Household political cooperation incentivizes strategic politicians to court the household vote and lessens their costs to electoral mobilization. Patronage and responsiveness are easier to deliver to households than to individuals. And if household behavior is aligned, there is no electoral loss associated with mobilizing only those with power.Footnote 47 Those in dominant positions within the household therefore have incentives to subordinate others in the household and ensure joint household political cooperation. Political elites too, have incentives to maintain a political system centered on the household and to challenge contestation to it.
Men’s participation in politics becomes self-perpetuating as they accumulate political skills, capital, and networks. Ties between politicians and male household heads deepen, cementing the link between political institutions and household institutions. The institution of household political cooperation becomes socially upheld as it facilitates and sustains patriarchal dominance and the prevailing (clientelist) political system. Those with power in the system – dominant men and political elites – have a stake in maintaining this institution and are expected to resist attempts to dismantle it. Backlash against attempts to renegotiate these norms or threats to men’s political authority is enabled by male coercive power;Footnote 48 broader patterns of inequality give men the resources and networks to organize and maintain a system in which they are dominant. This political system – what I term the patriarchal political order – yields a politics centered on the household and around men.
The Power of Women’s Collective Action
When patriarchal norms and gender-based inequalities enable men to strategically uphold a political order that elevates their political authority in both the household and the community, the key to women’s political inclusion is autonomy from the household, but only when it enables women to credibly challenge the likely resistance from those benefitting from the patriarchal political order. Autonomy from the household promises women both material benefits from the greater representation of their distinct interests and psychological benefits from the ability to voice these interests freely. But autonomy challenges the patriarchal political order. Those with power in this order are likely to resist giving up that power and therefore women’s autonomy. Backlash is to be expected from those with the greatest stake in the patriarchal order and to those challenges that most threaten their power.Footnote 49
How do women gain autonomy from the household and the freedom to make their own political decisions? Shifts in women’s individual endowments are unlikely to overcome the coercive structures of the patriarchal political order.Footnote 50 Instead, I argue that women’s collective action is a successful response to male power and coercion, even without a priori structural change.Footnote 51 Autonomy from the household and collective action alongside other women enables women to credibly contest these household and societal patterns of male political dominance, including coercive backlash.
Organizing collectively, however, bears high transaction costs: women must have sufficient information about their interests and who would share these interests (informational costs), women must be able to negotiate as a group about their priorities and strategies (bargaining costs), and women must be able to enforce sustained collective action (enforcement costs).Footnote 52 Given these transaction costs, I argue that collective action is most likely when women are connected to each other and their relationships involve political discussion, when they have a common framework of interests rooted in a shared identity, a recognition of shared injustice, and the belief that their action could address those injustices,Footnote 53 and when they have high levels of social solidarity built on trust and norms of reciprocity.Footnote 54
Given these high transaction costs and a dominant political order centering women’s political lives on the household, solidaristic collective action of this nature is unlikely to emerge endogenously. External intervention is often needed to disrupt self-perpetuating power structures and facilitate social change. As Batliwala, a leading theorist of and activist for women’s empowerment in India, argued, “the process of demanding justice does not necessarily begin spontaneously, or arise automatically from the very conditions of subjugation. The process of empowerment must therefore be induced or stimulated by external forces.”Footnote 55
I, therefore, consider the role of public policies, specifically the creation of economically oriented women’s groups (SHGs), in creating autonomy from the household and expanding and deepening women’s political relationships. SHGs are the largest policy intervention aimed at women’s empowerment both globally and within India. SHGs are principally economic institutions that offer microcredit to poor women, but they also bring together women with shared interests, provide an institutionalized space for discussion and to explore gendered interests, and foster the development of civic skills through deliberation and information sharing.Footnote 56 SHGs thus create the conditions for collective action.
While many have highlighted the tenuous benefits of SHGs for women’s economic empowerment,Footnote 57 I argue that if (and when) these groups yield solidarity among women, their political participation is likely to rise. By engaging in group-based collective action, women can jointly challenge political power structures and demand political representation. As a result, membership in SHGs can increase women’s political participation even when social norms and household dynamics continue to reinforce their exclusion from politics.
I study a snapshot of the lives of women in the thick of attempting to renegotiate gender-biased norms and generate social change. When successful, I suggest that this will not only yield a renegotiation of power through women’s presence but also a likely restructuring of politics. Collective action among women is expected to center on their distinct and underrepresented interests, including gender equality and public goods provision.Footnote 58 However, the strategies that women can deploy to navigate the complex structures of coercion are numerous. Yet it is this coercion that creates the conditions that unite women in solidarity against their continued oppression.
Collective action by women, therefore, has the potential to corrode the patriarchal political order. The process of change will not be linear, as many will contest this transformation, but if successful, it can cascade into more inclusive and representative politics and even broader improvements in governance and development.
The Stakes
Many historical legal institutions and even major academic models have not treated women’s limited political participation as a problem. The argument is generally that men and women share the same preferences, so it does not matter who participates; the outcome will be the same. This is particularly true under household-based political decision-making. Legal institutions such as coverture, which gave men legal authority over their wives, were designed with this assumption in mind. Unitary models of the household also assume that since the household has common preferences, it may even be efficient (household welfare maximizing) for women not to participate in politics.Footnote 59
Like these models, I also suggest that the household often behaves as a unit. But unlike these models, I show that women are first and foremost subjects – not citizens – of their households. In this book, I demonstrate that intra-household preferences are diverse, and that women and men have distinct desires related to the functioning of government. Importantly, as I will show with evidence from women’s collective action across India, when women do politically act, they do so in accordance with these interests. My analysis builds on a large literature that has demonstrated the substantial differences in political preferences between women and men across the globeFootnote 60 and work that documents the specific demands raised by women’s movements.Footnote 61 Recent empirical work from Khan in Pakistan reveals that not only do such preference gaps exist but also women are more likely than men to subvert their preferences in favor of those of their spouses.Footnote 62 Strikingly, this is most likely when the preference differentials are largest. So, while households may act as a unit, there is little evidence that doing so maximizes the welfare of the household; instead, it is most likely to maximize the welfare of those with power in the household.
This book, therefore, has implications for democracy: What is democracy with limited agency and unequal representation? Sen questioned whether growth without freedom can really be considered development.Footnote 63 The same can be asked of democracy; even if men perfectly represent women’s demands (although abundant evidence suggests this is unlikely), what is lost for democracy from women’s lack of freedom and representation?Footnote 64 Dahl explicitly argued the need for inclusivity, in addition to electoral contestation, for democratic functioning.Footnote 65 Mansbridge reflected on the many impacts of inclusion and representation in democracies, suggesting that inclusion matters not only because of the differences in outcomes that may be expected from representing distinct interests, but also because political inclusion creates social meaning for underrepresented communities and bestows legitimacy on democratic institutions.Footnote 66 Understanding the underrepresentation of women, and particularly the limits of their political agency, will enable a deeper appreciation of the limits and possibilities of democracy and its institutions.
The state of Indian democracy has received substantial academic attention in recent years, and new evidence has challenged conventional wisdoms.Footnote 67 As in many of the political debates in India since independence, this work has focused on the representation of the marginalized. Since it is one of the most ethnically diverse countries in the world, understanding how (and when) democratic institutions reflect and represent such diversity is critically important. A long history of work, largely centered on caste, has explored the nature of political mobilization in India and the conditions under which identities become politically salient.Footnote 68 Past research has also considered how the design of political institutions, particularly the implementation of electoral quotas, creates a politics of identity and enables more inclusive elite political institutions.Footnote 69 Others have focused on the substantial heterogeneity in the functioning of Indian democracy and considered the role of identity and diversity in explaining the outcomes of governance.Footnote 70 Exploring inclusion in institutions of patronage and clientelism, which are most prevalent in the Global South, has been particularly important. Chandra revealed the strong link between ethnicity and patronage in Indian politics.Footnote 71 Current research explores how patronage networks operate on the ground and how they contribute to larger patterns of political organization.Footnote 72 Like these studies, this book centers on the link between political institutions and identity, bringing the attention to gender and, more specifically, women citizens. It introduces new insights into the study of clientelism and identity by exposing how the structure of gender relations and the primacy of the household as a political unit perpetuate and enable clientelist machines. Accordingly, this book also has implications for governance.
The way in which women are incorporated into politics has important implications for policy and development. I show in this book that women’s political empowerment may itself drive both women’s economic empowerment and broader development. I document how women deploy their newfound political voices to demand gender equality and equal and programmatic access to public services for all. Having spent decades as political outsiders who have limited access to the spoils of politics, women recognize the importance of and demand high-quality governance.Footnote 73 Ample evidence from around the world suggests that policy outcomes shift in line with women’s preferences when they gain the right to vote or are elected to office. In their seminal study, Chattopadhyay and Duflo show that village chairpersons in rural India invest “more in infrastructure that is directly relevant to the needs of their own genders.”Footnote 74 Others have shown that women’s political representation has led to an overall rise in expendituresFootnote 75 as well as increased spending on public healthFootnote 76 and education.Footnote 77 Women’s political inclusion can also lead to the enactment of policies that institutionalize gender equality and outlaw gendered coercion.Footnote 78 And autonomous women’s movements have been shown to precede and precipitate government action, particularly on issues pertinent to women.Footnote 79
Given the high stakes of women’s political inclusion for democratic functioning, governance, and development, a thorough understanding of the mechanisms of women’s exclusion and empowerment can improve the design of public policies aimed at empowerment. Policy attention to date has largely focused on the economic roots of women’s disempowerment; policies have mostly sought to increase women’s economic agency through financial inclusion, access to education and jobs, and protection of inheritance rights.Footnote 80 India’s largest policy aimed at empowering women, which has touched the lives of more than 75 million women, is the promotion of SHGs – small, women-only microcredit groups designed to improve livelihoods through financial inclusion. These policies assume there is a link between women’s economic empowerment and broader development and inclusion. Yet, as I will show, mobilizing women into SHGs has a much larger impact on their political behavior than on their economic well-being. Most policies that seek to address gender inequality in politics have concentrated on women’s electoral representation; the cornerstone of policymaking to increase women’s political inclusion is the use of electoral quotas. A total of 130 countries, including all countries in South Asia, have introduced some form of quota to ensure the representation of women in political office.Footnote 81 And while these policies have made important headway in women’s political inclusion, I have already demonstrated how they have only a limited capacity to fundamentally transform outcomes for ordinary women. A key observable implication of this book is that even the most expansive gender equality policies will imperfectly achieve their goals unless they support women’s counter-resistance to coercion. A surprising finding of this book is that some of the most successful public policies for women’s political empowerment never intended such an outcome, which highlights the need to reevaluate our approaches to political empowerment.
Empirical Strategy
Case Selection
The evidence I bring to bear in this book combines first-person narratives and extensive data analysis. Gathering a compelling body of evidence to prove the existence of a patriarchal political order required directly communicating with more than 9,000 rural Indian citizens. The paucity of data on women’s political behavior, particularly outside of voting, necessitated large-scale original data collection to draw more general inferences. This process generated a depth of understanding of the political lives of these citizens and allowed me to draw comparisons from a wide range of lived experiences. I concentrated on a limited geographic area to enable a causal research design, but I broaden the scope of analysis to the entire country where available data permits.
The majority of the book is based on original research conducted over six years in five districts in the Indian state of Madhya Pradesh (outlined in black in Figure 1.2): Balaghat, Betul, Dindori, Hoshangabad, and Mandla. Madhya Pradesh is the second-largest state in India and has a population of more than eighty-five million, approximately the same as that of Germany or Turkey. It is located in the poorer central belt of the country, and has historically had worse outcomes for women and stronger patriarchal norms. One in three residents of the state lives in poverty, and more than three-quarters reside in rural areas. As in much of India, women are particularly economically disadvantaged. In 2018, women in the state had an average rural labor force participation rate of 17 percent (55 percent for men) and average wages of INR 726, or $10.60, per week (INR 1,831, or $26.73, for men).Footnote 82 These statistics reflect slightly higher female labor force participation than the national average but substantially lower female wages than the national average of INR 1,296, or $18.92.
Importantly, Madhya Pradesh is a richly diverse state. While the official language is Hindi and more than 90 percent of its population is Hindu, it was once a center of Muslim rule. Bhopal, the present-day capital of Madhya Pradesh, was India’s second-largest Muslim state during British colonial rule; it was ruled by four Muslim women from 1819 to 1926 (the Begums of Bhopal). Furthermore, Madhya Pradesh has the largest tribal population in the country; more than forty tribal communities comprise roughly 27 percent of its rural population (see Table A1.1). Tribal populations, also known as Adivasis, are economically worse off than most other ethnic groups in India. On average, they have lower incomes, lower literacy rates, higher rates of maternal and child mortality, and own less land.Footnote 83 Due to their historical exclusion and continued inequality, tribal populations as a group have received formal protections under the law, including through reservations in various domains and limited rights to self-governance. Yet, despite ST numerical dominance in many of the villages studied in this book, Chapter 6 shows that political influence and power is often concentrated in OBC and General Category (GC or upper caste) communities.
The distinct and rich ethnic diversity prevalent in Madhya Pradesh makes this state a valuable place for inquiry and provides an opportunity to explore how gender operates across these intersecting identities. Norms and practices vary in important ways across the many cultures within India and even within villages. The villages I study exhibit substantial ethnic diversity, largely divided among Hindu caste groups and indigenous tribes. This diversity is important to our understanding of how the gender system operates. Tribal populations are often assumed to have much more gender-equal norms, as evidenced through higher female labor force participation and higher sex ratios. Yet evidence abounds that patriarchal norms dominate these communities.Footnote 84 According to their codified norms and rules, women face harsher sanctions for deviating from marriage and birth norms. I show in Chapter 2 that attitudes toward gender equality in these villages vary little across caste groups. In most tribal cultures, men retain rights over children and property and are the de facto household heads, while women have little authority within the household. In this regard, women find common ground across their intersecting identities. Throughout this book, I evaluate the experiences of women from a variety of backgrounds, all of whom are subject to patriarchal norms in their households and communities.Footnote 85 I pay close attention to how caste and gender dynamics intersect but also reveal the depth of commonality in women’s experiences across the country.
Similarly, while socioeconomics and ethnic diversity may vary across states and districts in India, women’s underrepresentation in politics does not. Figure 1.2 demonstrates that, in the domain of non-electoral political participation, the highest rate of reported participation in village assembly meetings by women was 41% (Kerala), and women’s participation in these meetings topped 15% in only four states (Assam, Himachal Pradesh, Andhra Pradesh, and Kerala). Compared to the rest of India, Madhya Pradesh reports relatively low rates of female non-electoral political participation (3%), although these rates are similar to those in neighboring states. Only 27 of its 230 state representatives are women, and female voter turnout was 68 percent in the 2019 parliamentary elections, which was just above the national average of 67 percent.
I situated this study in India (and, more specifically, in Madhya Pradesh) for three reasons. First, it represents a conservative and challenging case in which to identify positive effects on women’s political empowerment given its particularly low levels of women’s political participation and representation. Yet women vote at high rates. The strength of this puzzle of participation opens up opportunities for inquiry, which allow us to observe more clearly what dynamics explain these patterns and to hold constant many of the factors historically attributed to women’s political subordination, such as their relative economic position and gender norms. Second, India and Madhya Pradesh more specifically are domains of policy experimentation. In the regions I study, I observe substantial variation in women’s political participation. Much of this variation, as I will show in the second half of the book, is driven by the large body of governmental and nongovernmental actors working to identify policy levers for women’s empowerment. My decision to focus on the six aforementioned districts was rooted in a series of policy experiments run in these districts by one such nongovernmental organization (NGO). And given the depth of patriarchal norms in this region, the identified solutions will provide strong evidence of what is likely to work even under challenging conditions. Finally, I chose to center this book on India as it is a diverse and consequential subcontinent with 665 million women, representing 17 percent of the global female population. The large majority of empirical research on gendered political behavior has been concentrated in the Global North, primarily in the United States. There remains much to be learned from taking an expanded geographic scope, and India is a clearly important case. Beyond its importance as the world’s largest democracy, conclusions from India can help identify more general patterns of gendered exclusion that are often masked in contexts where norms are harder to observe.Footnote 86
Mixed Methods and Causal Identification
I test these arguments in rural India using a diverse set of data sources and empirical methods. Six years of research, including more than a year spent in India, inform the research design and data collection. A key challenge associated with examining relational theories and identifying the causal effects of social relations for behavior is differentiating the effect of social ties from the rest of the social system. Furthermore, a principal reason why we have such a poor understanding of women’s political behavior in India is a lack of data from women. Rural public opinion polls rarely stratify their samples based on gender due to the difficulty of speaking with women away from their husbands; those who do focus mostly on voting, which, due to household coercion, may not be an informative measure of women’s political agency.
This book overcomes these methodological challenges by analyzing observational qualitative and quantitative data alongside a series of natural experiments that manipulated theoretically important variables. Combined, these data and methodologies generate a preponderance of evidence that supports the existence of household political cooperation, the patriarchal political order, and the value of household autonomy and women’s social solidarity for women’s political behavior. I draw on three principal data sources, which I describe in greater detail in the subsequent chapters.
First, I utilize data from an original survey conducted in 2016 in rural Madhya Pradesh of more than 5,000 women and 2,500 of their husbands across 376 villages. In each village, fifteen women and eight of their husbands were asked a series of questions about their political behaviors and attitudes. Second, I draw on data from a second original survey in 2019 that randomly sampled 6 villages from the 2016 sample of villages and censused all 3,565 adult residents of those villages. This census survey sought to measure political network characteristics and employed a series of name-generator questions to map respondents’ networks. In both surveys, data were collected in person by trained surveyors. Given the sensitive nature of the questions and concerns about social desirability bias, respondents were gender matched to surveyors, and all surveys were conducted in complete privacy. Third, I analyze data from semi-structured, qualitative interviews with 120 randomly sampled women and 80 of their husbands across a second random subsample of twenty villages. These interviews sought to understand how women make political decisions and to capture how women understand their own political agency and participation. I additionally pair these three original data sets with nationally representative survey data from the Indian Human Development Survey (IHDS)Footnote 87 and data on all reported instances of women’s collective action from the Armed Conflict Location & Event Data Project (ACLED)Footnote 88 to demonstrate the generalizability of my argument across India.
I analyze these data using a combination of analytic description and causal inference methodologies. To document the existence of household political cooperation, the nature of women’s political participation conditional on a norm of such cooperation, a broader political order built around women’s exclusion, and the role of political ties in each, I use observational methods to describe the wealth of data provided in my original surveys. I also deploy network analysis tools to map and document the gendered nature of political connections in rural Indian villages. I use both qualitative reporting and quantitative coding of interview transcripts to add richness to this description and better examine the harder-to-observe mechanisms of women’s political subordination.
To more precisely estimate whether particular factors drive women’s political behavior, I examine a series of natural experiments that shocked women’s access to two NGO empowerment interventions. The two interventions I study sought to increase women’s autonomy from the household and solidarity with other women. The first mobilized women into SHGs, the women-only microcredit groups described earlier. The second intervention applied a gender consciousness-raising program within SHGs to build solidarity and trust among women. Since only some villages were eligible for these interventions – eligibility was arbitrarily determined – I can causally estimate their impact on women’s political behavior. The samples for both surveys and the qualitative interviews included villages that did and did not receive these interventions, so the survey and interview data are used to estimate the causal effects of these programs.
The 376 villages in Madhya Pradesh in which data were collected were sampled with these two sets of analyses in mind. As a result, the villages fall into three categories: (1) those that have not received any SHG-related NGO program, (2) those that have received only the SHG program, and (3) those that have received both the SHG and gender consciousness-raising programs. Respondents were sampled in two ways for the 2016 Madhya Pradesh sample survey. In the first set of villages with no SHG programs, women were randomly sampled from census lists subsetting to adult, ever-married women. In the second and third sets of villages, where the SHG program had been implemented, women were randomly sampled from lists of SHG members. These sampling strategies and their implications for the research design are described in the relevant chapters. As I describe in Chapter 7, a supplemental survey in 2016 revisited some SHG villages and surveyed a sample of women who had been randomly chosen from census rosters to ensure comparability with non-SHG villages. Table 1.1 describes the mapping of the survey samples with the samples used in each set of analyses.
No SHG Program | SHG Program | SHG Program + Gender Consciousness-Raising Program | |
---|---|---|---|
MP Sample Survey | 76 villages | 226 villages | 74 villages |
MP Census Survey | 2 villages | 4 villages | |
MP Interviews | 10 villages | 10 villages | |
Sample analyzed by chapter | |||
Chapters 4, 5, 6 | X | ||
Chapter 7 | Control (76 villages) | Treatment (76 villages) | |
Chapter 8 | Control (74 villages) | Treatment (74 villages) |
Overview of the Book
This book presents a theory of coercive political power with four core arguments: (1) The household is the primary unit of political decision-making in rural India, (2) women’s political participation aligns with the strategic interests of the men in their household given inequalities in bargaining and coercive power, (3) this yields a larger political system sustained by women’s political subordination, and (4) autonomy – or freedom to make political decisions distinct from the household – and solidaristic collective action among women enable women’s greater political participation.
The first part of the book (Chapters 2 and 3) describes the conditions surrounding women’s political lives and develops a theory of their political exclusion as rooted in power relations. Chapter 2 sets the stage by defining and describing the manifestations of patriarchy and inequality in India that serve as the backdrop for theorizing. I describe the system of patriarchy in rural India and how widespread and entrenched patriarchal norms generate a patriarchal social order that centers women in the household. I demonstrate how patriarchal norms have enabled the use of violence to control and dominate women, including by internalizing the acceptability of this means of coercion. I further highlight the role of legal and political institutions in perpetuating this social order. I additionally document the state of women’s and men’s political participation in rural India, revealing substantial disparities in political participation between men and women and, even more strikingly, between different forms of participation within women.
In Chapter 3, I build a theory of women’s political behavior under these conditions. I argue for the defining role of the household in women’s political lives. I describe the conditions that would lead to household political cooperation – joint household political decision-making – and hypothesize its implications for individual women’s political behavior and the structure of political organization more broadly. I then develop a series of expectations about how and when women will become active political citizens, arguing for the importance of women’s autonomy from the household, their social solidarity with each other, and the potential for women’s collective action to transform their political lives.
The second part of the book (Chapters 4–6) empirically proves the existence of household political cooperation, demonstrates how it relates to women’s political participation, and documents the patriarchal political order. Chapter 4 brings the household into focus and demonstrates how it denies women political agency and constrains their political participation. Drawing on the census survey and interview data, I document the alignment of the household in political decision-making and the authority of elder men in these decisions. I show that women lack autonomy in their vote choice and are often coerced into compliance with the wishes of the heads of household. I further document the inefficiency of household cooperation for women and demonstrate its perpetuation as rooted in coercion and strategic political mobilization.
In Chapter 5, I show that power within the household and autonomy from the household most strongly predict women’s political participation. Using the census data, I estimate the determinants of women’s (and men’s) political participation. Comparing within households, bargaining power is associated with women’s non-electoral political participation, though not their voting. Comparing within villages, autonomy from the household is a clear predictor of women’s political participation. In fact, the behavior of women who enjoy a high degree of autonomy from the household mirrors that of men. I also provide suggestive evidence that intra-household coercion plays a role in women’s political participation by showing a lack of correlation between political interest and participation in the household for women but not men and a negative correlation between regressive gender attitudes of the dominant male household member and women’s political participation.
Building on this, Chapter 6 illustrates that household political cooperation begets a system of political organization that centers on men. Using network data from the census survey, I describe the structure of the overall village political network, including gender homophily of political ties, centrality in the entire network, and the average degree of connectivity between individuals and political elites. I show that village political networks are structured such that men comprise the center of the network, while women remain on the periphery. As a result, influence is concentrated among men, and village politics is structured around men’s other intersecting identities, namely caste. I then compare the size and composition of women’s and men’s political networks and show that women are connected to village politics largely through the men in their household. For men, women do not register as political actors. Household political cooperation thus implies strong limits on women’s access to power, influence, and information and yields a broader gender and political system that perpetuates male dominance.
The third part of the book (Chapters 7–8) generates causal evidence showing that autonomy from the household and collective action among women increase women’s non-electoral political participation. Chapter 7 provides cause for optimism: women’s participation in apolitical women’s groups enhances their political agency and doubles their political participation. Leveraging a natural experiment to identify the impact of access to SHGs, I show that access to spaces outside of the household with other women generates solidaristic collective action oriented toward women’s political participation that succeeds in changing women’s political behavior: SHG members were significantly and substantially more likely to participate in politics than nonmembers. Further, this impact is evident in the larger village political network; women are more densely politically connected, and gender emerges as a more salient political cleavage. This positive impact of SHGs occurs despite no change in women’s economic resources.
Chapter 8 demonstrates that such effects can be augmented by stimulating solidarity among women with a focus on gender consciousness-raising, but that such actions generate backlash. I test the importance of social solidarity in stimulating women’s collective action by exploiting arbitrary variation in the delivery of a gender consciousness-raising program to SHGs. I show that women are more likely to undertake collective action after identifying shared experiences of deprivation and forming a bond based on their gender identity. This collective action is also more likely to be aimed at women’s strategic interestsFootnote 89 – their interests rooted in their patriarchal suppression – and therefore garners more resistance from men in the community, including through increased experiences of (public) violence and harassment. I show that women navigate this resistance through their collective strength and solidarity.
In the final part of the book (Chapters 9 and 10), I explore my argument’s implications for (and generalizability to) broader patterns of women’s political representation and governance in India. Chapter 9 reveals that the circular patterns of norm renegotiation manifest at the national level in India’s broader women’s movement. I describe the history of this movement and then use ACLED data on all women-led protest events in India from 2016 to 2021 to illustrate the breadth of women’s collective mobilization and the range of demands raised. Women most often come together to protest more explicitly gendered issues, such as gender-based violence. However, many women-led protests focus on other demands, including improved government accountability and service delivery. The nature of women’s demand-making suggests possibilities for both gender equality and improved governance with their political inclusion. Finally, I document broader patterns of resistance to women’s collective action at the national scale, documenting a range of explicit instances of violent backlash and summarizing the rise of the men’s rights movement in India. This provides further evidence of male coercion and suggests conditions under which women’s collective action can succeed.
I conclude in Chapter 10 by taking stock of the evidence and looking ahead at the long-run implications of women’s political inclusion for broader processes of development and social change. I argue that women’s political inclusion hinges on their ability to navigate resistance and co-optation. If they are able to achieve real political representation, I suggest it is likely to yield important changes to governance and development more broadly.