The Costs of Housing Regulation: Evidence From Generative Regulatory Measurement
103 Pages Posted: 29 Nov 2023
Date Written: September 14, 2024
Abstract
We present a novel method called "generative regulatory measurement'' that uses Large Language Models (LLMs) to interpret statutes and administrative documents. We demonstrate its effectiveness in analyzing municipal zoning codes, achieving 96% accuracy in binary classification tasks and a 0.92 correlation in predicting minimum lot sizes. Applying this method to U.S. zoning regulations, we establish five facts about American zoning: (1) Housing production disproportionately happens in unincorporated areas without municipal zoning codes. (2) Density in the form of multifamily apartments and small lot single family homes is broadly limited. (3) Zoning follows a monocentric pattern with regional variations, with suburban regulations particularly strict in the Northeast. (4) Housing regulations can be clustered into two main principal components, the first of which corresponds to housing complexity and can be interpreted as extracting value in high demand environments. (5) The second principal component associates with exclusionary zoning.
Keywords: housing regulation, zoning codes, large language models, natural language processing, artificial intelligence, municipal ordinances, retrieval augmented generation
JEL Classification: R52, R58, K11, O38, R31, C81
Suggested Citation: Suggested Citation