Why Is Housing So Hard to Build? Four Papers on the Collective Action Problem of Spatial Proximity

Abstract

Housing has spatially concentrated costs and spatially diffuse benefits. This dissertation studies how this imbalance of costs and benefits creates scale-dependent preferences regarding the supply of new housing. Furthermore, this dissertation measures how institutions harnessing those preferences have led to an undersupply of new housing, deepening the urban affordability crisis. The first paper of this dissertation measures the spatial sensitivity of residents to the location of new housing. Using both national experimental data and city-specific behavioral data, I find that renters living in expensive cities express NIMBY (`Not In My Back Yard') attitudes towards market rate housing at a level similar to homeowners. However, these renters still largely support an overall increase in their city's housing supply. This conflict between supporting housing citywide, but not in one's neighborhood reflects the collective action problem of spatial proximity. Both homeowners and renters may support new housing and its positive effects citywide, yet they are prone to defect and block the development of new housing in their own neighborhood. The second paper shifts from renter attitudes to homeowner attitudes, seeking to identify when homeowners who typically oppose new supply will vote in favor of an increase in the housing supply. Again, using both national and city-specific data, I measure the tradeoffs homeowners make between their personal interest in seeing their home values appreciate and their simultaneous belief that citywide housing prices are too high. By using respondents' expectations of the effect of new supply on personal and city prices, I find that these `conflicted' homeowners are willing to forego appreciation of their home value to achieve lower prices, but not accept losses in their home value. This finding not only supports the use of home value insurance as a policy solution, but also is the first test of the role of prospect theory and loss aversion in sociotropic voting.

The third paper returns to the spatial sensitivity of renters and homeowners by framing the overall housing supply as a public good. I use a survey experiment to test whether NIMBYism is linked to concerns of neighborhoods free-riding and avoiding the costs of this public good. Using a survey experiment, I alter the spatial allocation of a proposed 10\% increase in the housing supply, then measure respondent's support for the policy. I find that when homeowners and renters must accept new housing in their neighborhood, they prefer allocations that eliminate free-riding by distributing the housing equally across all neighborhoods.

Finally, the fourth paper examines how NIMBY preferences can work through neighborhood-level institutions to cause breakdowns in collective action. Absent the formal power to veto new development, these neighborhood planning institutions may increase the de facto political power of local opposition. Using time-series permitting data for 52 cities from 1960 to 2015, I find that the implementation a neighborhood planning institution decreases the annual permitting of housing units within the same city by roughly 35 percentage points, an effect driven almost exclusively by a decrease in multifamily housing. By increasing each neighborhood's influence in the permitting process, these local institutions appear to threaten the neighborhoods' collective interest in permitting enough new housing to maintain affordability citywide.