2026 Theses Doctoral

The Politics of Gender Norms: Evidence from Rural Tanzania

Montano, Beatrice

Why do harmful practices against women persist across the Global South? Gender-based violence (GBV) is one of the most pervasive human-rights violations in the world: one in three women worldwide experiences physical and/or sexual violence in her lifetime, and one in five is married before the age of 18. In the 60 villages of rural Tanzania where this dissertation is based, the situation is even more dire. These practices are the most visible surface of an underlying system of gender norms that governs nearly every consequential decision in the lives of women in much of the Global South. The three papers in this dissertation study how gender norms are politically contested, privately enforced within households, and reproduced across generations. While each chapter stands on its own, the main contribution of the dissertation is to establish that beliefs about the social world---how individuals read the attitudes, risks, and expectations that surround them---shape the political and household decisions that keep harmful practices in place.

The first paper is about contestation: it shows that gender norms can be politically disrupted when citizens learn how much agreement for change already exists in their community. Cultural attitudes are constantly evolving, yet behavioral changes lag behind---often taking the form of sticky social norms. One example is early marriage, which affects 40% of women in Tanzania despite being widely opposed. I demonstrate that misperceptions about community attitudes perpetuate practices harmful to women by misallocating political action, and that correcting misperceptions works to redirect it. While various forms of "nudging" have been used to address gender norms, no previous study has considered their effect on political mobilization. In a face-to-face experiment with 3000 villagers, I provide accurate information about community attitudes toward early marriage and measure subsequent political behavior. After learning they have community support, women who oppose early marriage prefer to direct their advocacy efforts toward political elites instead of focusing on peer persuasion: they strategically target their advocacy towards perceived political opportunities rather than work to build a consensus that already exists. This pattern is also observed among untreated women who already believed others to oppose early marriage, which indicates that this strategic approach to politics is a natural part of women's political choices. These findings suggest that inaccurate beliefs about social consensus can trap communities in harmful gender norms by leading citizens to misallocate political effort, even when broad support for change already exists.

The second paper is about enforcement: it shows how gender norms govern adult intra-household resource decisions, and how information about GBV-risk tightens, rather than loosens, that grip. In patriarchal settings, men govern whether their wives travel, making male resource allocation the proximate determinant of female mobility. I ask whether increased awareness of GBV-risk makes husbands more likely to invest in safe transport or to restrict travel altogether. Boda-boda motorcycle taxis are the primary mode of transport in the study area and GBV-risk is salient. I formalize the theoretical ambiguity in a structural model that separates a safety preference margin---how much a husband values safe transport conditional on travel---from a mobility support margin---whether he funds the trip at all---and connects each to a distinct empirical estimate. Using a conjoint experiment and an incentivized willingness-to-pay elicitation embedded in a randomized field experiment with about 1200 people across 34 villages, I find that GBV awareness leaves men's safety preferences unconditionally high while reducing their willingness to fund travel by 20%. Gatekeeping and genuine protective concern need not be mutually exclusive.

The third paper is about acquisition: it shows how gender norms are acquired in childhood, and through what perceptual machinery. The dominant answer to where children's gender attitudes come from---that parents transmit their own attitudes through household exposure---fails on contact with data from 680 households. Within villages, knowing a parent's gender attitudes is essentially uninformative about their children's. What predicts children's attitudes is not what parents actually think, but what children believe their parents think.
And those beliefs are not random errors: children systematically misread their parents as more aligned with the local community norm than they truly are, anchoring their perceptions toward the village mean in a pattern that is directional, proportional, and village-specific. Parents themselves misperceive their community---but their misperceptions transmit to neither how children read them nor how children form their own attitudes. Children construct their community reading independently. Patriarchal norms persist across generations not because families reliably transmit them, but because communities reliably model them.