Rent control and the supply of affordable housing

Abstract

We generate the first cross-city panel dataset of rent control reforms and estimate their effect on the supply of rental housing overall and across varying levels of affordability. To identify reforms, we use machine learning algorithms to analyze over 76,000 newspaper articles from 7000 news outlets, spanning 27 metropolitan areas and >4000 census places across the US between 2000 and April of 2021. We then manually validate identified articles to ensure accuracy and combine these data with rental unit counts by affordability level, created using Census microdata. To assess the impact of rent control reforms on rental supply, we employ a two-way fixed effects model with place specific time trends and examine the robustness of our results with a staggered treatment design. Our results provide evidence that more restrictive rent control reforms are associated with a 10-percent reduction in the total number of rental units in a city. When stratified by affordability (based on U.S. Department of Housing and Urban Development definitions of affordability), these reforms lead to an increase in the availability of units affordable to extremely low-income households by about 52 % (with a lower-bound effect equal to 11 %), offset by a decline in units affordable to higher-income households of about 46 % (with a lower-bound estimate equal to 4 %). These findings highlight the complex trade-offs inherent to rent control policies, illustrating differential impacts across income groups and underscoring the nuanced nature of such interventions.