This paper examines how Los Angeles citywide discontinuous increases in real estate transfer taxes, aimed at preventing homelessness and funding affordable housing, actually affected housing supply. Using the city boundary as a natural experiment, I find that the tax reduced both home sales and new construction. I assemble a novel dataset of building permits, using AI-powered textual analysis of permit descriptions to extract granular information on construction activity. Sellers clustered transactions just below the tax threshold, while developers broadly reduced construction activity rather than pivoting to affordable or multi-family housing. Luxury homeowners increasingly chose remodeling over selling, further constraining supply. These findings suggest that taxing high-end properties may undermine rather than improve housing affordability.

I explore the temporal dynamics of localized policy intervention by examining adjacent counties subject to differing levels of pandemic-related restrictions. Specifically, I find that the imposition of an additional workplace closure measure in a given month is initially associated with a transitory suppression of housing price growth relative to a neighboring counterpart. However, this negative effect does not persist; over time, the trajectory reverses, and a positive adjustment in house price change emerges in the subsequent months.