Job Market Paper

Legal Segregation, Ethno-Religious Boundaries, and Urban Inequality in India’s Disturbed Areas Act

I study Gujarat’s Disturbed Areas Act (DAA), a rare setting where urban segregation is created and enforced through formal law restricting inter-religious property transactions within designated police-station jurisdictions. I assemble and geocode a new panel of DAA notifications (1991--2024) and link these boundaries to neighborhood religious composition measured from electoral rolls (via name-based classification), data on housing markets and density, data on school inputs and outcomes, and data on violence. Using staggered difference-in-differences event studies (Sun and Abraham; Callaway and Sant’Anna), I estimate dynamic effects and show that heterogeneity by local DAA intensity is central: housing outcomes diverge sharply by exposure intensity, schooling inputs and failure rates deteriorate over the medium run alongside compositional shifts in enrollment, and violence exhibits no differential pre-trends that would support a purely preventive rationale for designation.

Other Works in Progress

The Intergenerational Impacts of Residential Segregation and Discrimination on Muslims in Urban India

I develop and estimate a parsimonious neighborhood-choice model linking residential segregation to intergenerational mobility for Muslims in urban India, where neighborhood quality is proxied by local school quality and parental location choices shape children’s skill formation. Discrimination operates as a “minority tax” that raises the cost for Muslims of accessing higher-quality (typically Hindu-majority) neighborhoods, generating endogenous segregation even conditional on skill. I discipline the model using parent--child linkages in the India Human Development Survey panel (2005, 2012) and estimate the discrimination wedge via simulated method of moments. The model implies that neighborhood access and discrimination can account for a sizable share of the Hindu--Muslim skill gap (larger in Gujarat than nationally), and counterfactual reductions in discrimination meaningfully narrow long-run inequality and raise Muslim welfare.

Segregation and Unequal Access to School Resources: Evidence from Caste-based School Composition Thresholds in India (with Moumita Das, Gagandeep Sachdeva, and Kartik Srivastava)

We study how local caste-based composition thresholds generate discrete changes in schooling markets in India. Using national school census data (UDISE, 2005--2017), we recover district-specific “tipping points” in baseline disadvantaged-caste share using a fixed-point strategy and then map these into predicted village-level cutoffs to reflect local market conditions. We estimate jump-and-kink specifications around the predicted cutoff using Regression Discontinuity methods and show that crossing the threshold coincides with discontinuous shifts in enrollment composition and sharp breaks in multiple dimensions of observable school resources (infrastructure, staffing, and fiscal inputs), consistent with a regime change driven by sorting, administrative responses, and capacity constraints rather than smooth linear adjustment. Heterogeneity by management type (government vs.\ private) further supports an interpretation of market segmentation and reallocation across school types at the threshold.

The Impact of the US-China Trade War on Indian Firms (with Pulak Ghosh,  Aakash Kalyani, and Manpreet Singh)

This project studies how Indian firms responded to the US-China trade war by exploiting firm-level panel data on exports, imports, and investments from the CMIE Prowess database. We construct industry-year level measures of exposure to US tariffs on Chinese products and estimate the differential investment response among Indian firms across trade categories. Using firm, year, and industry fixed effects in panel regressions, we find preliminary evidence that Indian exporters to the US, particularly those not heavily reliant on Chinese inputs, increased investment significantly post-tariffs, consistent with a substitution effect in global value chains. This investment response was heterogeneous by firm size and age, with younger and more productive firms driving the gains.