How do gender norms shape the relationship between women's labor force participation (LFP) and political engagement? While existing research suggests that economic resources enhance political participation, we have yet to engage with the idea that women's LFP itself deviates from gender norms. I argue that when LFP is norm-deviant, women compensate for their employment by decreasing political engagement. Specifically, in restrictive normative environments, women who work face potential sanctions from their communities. This leads them to offset one form of norm deviance (employment) by conforming to norms in another domain (politics). I test the observable implications of my argument using survey and administrative election data from three settings in India as well as cross-national survey data. Results from a variety of research designs show that These findings complicate assumptions that economic empowerment directly translates into political empowerment. They suggest that policies aimed at increasing women’s employment must also address entrenched gender norms to create meaningful pathways for political inclusion.

Studies about the determinants of electoral manipulation demonstrate that perpetrators tend to target specific voters, including rural voters, lower-income voters, and otherwise marginalized groups. However, little is known about how citizens' gender can affect the nature of electoral manipulation. I use the widespread phenomena of gender-segregated voting to conceptualize how citizens' gender can affect the allocation of electoral manipulation at the ballot box. I then test how gender can affect electoral fraud and violence at the ballot box in the case of Pakistan's 2013 national election. I draw on unique election-observer-level data sets that overcome various difficulties in measuring electoral fraud to find that women's polling stations and booths experience higher levels of electoral manipulation. The nature of observation, voter demographics, and electoral competition do not drive this effect. My findings have implications for improving electoral integrity and ensuring citizens' equality in democratic processes.

Formerly titled, “Roadblocks remain: constraints to women’s political participation in Pakistan” (co-authored with Sarah Thompson)

How can governments encourage political participation by all? In this study, we ask why certain groups are less likely to vote solely based on where they are assigned to vote, and argue that mobility plays an important role. We focus on Pakistan, the world's sixth-most populous country. Pakistan is among the lowest-ranked countries for women's political participation despite instituting reforms like single-gender polling stations. In this paper, we leverage 2018 polling station data to show that mixed-gender polling stations increase women's turnout. We also present descriptive findings to show that chance assignment to certain types of polling locations make women, but not men, more likely to turn out. We then use a survey experiment to test one possible explanation—mobility. Constraints to mobility have been shown to negatively impact women's educational and labor force choices, but their impact on women's political participation has not been directly tested. We find that when women's mobility is constrained by a lack of male accompaniment, or they expect to travel along a predominantly male route, their likelihood of turnout decreases. We also find that women are more likely to vote in areas familiar to them (i.e., primary schools where they drop their children and girls' schools they attended). Our study implies that strategies to increase women's political participation in developing democracies should take seriously the role of women's mobility.

GLD Working Paper

Co-authored with with Nivedita Narain and Soledad Artiz Prillaman

Gendered imbalances in political spaces remain widespread and calls for “women's empowerment” are now commonplace. Despite the need for and benefits from women’s political empowerment, few frameworks exist to conceptualize and assess the extent of prevailing political power imbalances, particularly at the individual level. Further, unlike economic empowerment, political empowerment — the ability to choose when and how one interacts with political institutions—has received less scholarly attention. Measurement of gendered political behavior has generally focused on actual political participation. These measures, and many of the models they test, fail to capture the core aspect of political empowerment—the presence of agency. Our paper proposes a theoretical framework to conceptualize political empowerment at an individual level, highlighting the importance of conceptualizing agency. We use this framework to develop a set of survey questions to construct a measure of women’s political empowerment at the individual-level. We test the reliability and validity of this measure using data from an original survey with roughly 1,000 women in rural India. We then highlight the importance of conceptualizing and measuring political empowerment by demonstrating the empirical distinctness of empowerment from participation, evaluating variation in the predictors of empowerment and participation, augmenting the learnings from a series of replications focused on shifting participation, and providing new empirical evidence of the link between empowerment and within household political preferences.

How do capital investments predict political participation? Existing research links investments to wealth, which increases political participation by expanding monetary resources. This paper introduces an overlooked mechanism: risk. I argue that households relying on risky, singular income sources, like agriculture, view government benefits as a form of insurance and depend on the state for additional inputs like infrastructure. This reliance on the state drives investors into the political arena. I use agricultural investments in India as a case to show that these concentrate household income and increase reliance on government support, spurring political participation. I use a matched difference-in-differences design on a nationally representative panel survey in India to test this idea. This paper shows that people who invest in agricultural machinery are more likely to attend public meetings than a comparable control group. I then use public meeting transcripts and the specific case of the tractor to substantiate that investors attend these meetings to make claims on the government.