Selected Work in Progress
Selected Work in Progress
Dissertation chapter in progress
In the United States, property taxation is the main revenue instrument for local governments, and strengthening the welfare state at the local level is heavily reliant on property tax financing. In recent years, progressive local governments are increasingly trying to fund more public investments in education and early childhood services with property tax levies. However, there has been little research so far on how property tax hikes themselves influence fertility - the earliest and foundational private investment in children. Assuming children are normal consumer goods (Becker 1981), and given that property taxes can significantly shift disposable income and cost of housing (a key input in raising a child) for owners and renters, higher property taxes would - all else equal - reduce the demand for children in the current period through both income and substitution effects. If low-SES groups respond by reducing or delaying childbearing at a higher rate, the beneficiaries of property-tax-funded programs may increasingly skew toward wealthier families. Exploiting close school tax referenda from Ohio using a Regression Discontinuity design, this paper produces novel causal evidence on the effects of property tax levy on fertility and heterogeneities between groups.
Dissertation chapter in progress
Are women and men fundamentally different? And can grappling with this question help one resist the force of gender stereotypes? I hypothesize that prompting girls to interrogate the categorical boundaries embedded in gender schemas can help them succeed in counter-stereotypical domains. I test this hypothesis using a randomized survey experiment conducted in 14 secondary schools in Northern Nigeria in Spring 2025. Girls in each school were randomly assigned to one of three equal-sized groups: Treatment 1 (T1), Treatment 2 (T2), or a pure control (N = 474). Girls in T1 and T2 were exposed to a vignette describing a successful, counter-stereotypical female peer excelling in locally male-dominated domains. In T1, the vignette was followed by writing prompts designed to highlight shared characteristics and capabilities between boys and girls, while T2 received the vignette only. Preliminary results suggest heterogeneous effects by pre-intervention self-efficacy on both achievement expectation immediately after treatment and Mathematics performance in a national exam within a few weeks of treatment.
Dissertation chapter in progress
A large focus of family planning policies in developing countries is on the supply of contraceptive commodities. However, evidence on the effectiveness of supply-side interventions in boosting modern contraceptive use remains mixed. This study aims to estimate the causal effects of increased availability of modern contraceptives on use and fertility outcomes in Senegal, where unintended pregnancy rates are among world's highest. I exploit the staggered implementation of the Informed Push Model program which strengthened contraceptives supply chains nationwide since 2012 and analyze impact on fertility behavior and outcomes using a Difference-in-Differences design and data from the Demographic and Health Surveys (DHS) and Service Provision Assessment (SPA) surveys.
Working paper. Blog on gendered language.
with Didier Alia, Prasun Gangopadhyay and C. Leigh Anderson
Explanations for technology adoption by small-scale producers (SSPs) predominantly focus on observed socioeconomic factors and technology attributes. We posit that where gender norms persist, a farmer’s attitude toward the gender of a technology’s name can influence choice between functionally undifferentiated products. Using a discrete choice experiment in India, we test this by asking SSPs to choose which one to buy between two otherwise identical improved seed varieties - one with a male-gendered name and one female-gendered. We test two pairs of names and find statistically significant differences in choices between male and female farmers in one pair, where female farmers are 11.6 percentage points less likely than male farmers to choose the male-gendered alternative; but not in the other pair. Our findings suggest gendered language used to name a technology can play a role in adoption decision-making by SSPs and induce systematically different choices between male and female farmers.
Behavioral economics research in stereotype threat: New approaches to an established field and lessons for policy
Working paper