Abstract
The median US household holds a majority of its wealth in a single climate-exposed asset: the home. I study the implications of this fact for the distributional consequences of climate change, using a dynamic, stochastic, heterogeneous-agent model with 1713 locations. Forward-looking equilibrium real estate prices and rents endogenously respond to climate news shocks in spatially-segmented markets. Households participate on both sides of each market. To quantify climate impacts on economic fundamentals, I harmonize recent estimates of local productivity, amenities, energy costs, and disaster damage sensitivities. I find the model’s global solution under aggregate climate uncertainty with a simple but general deep learning method introduced in a companion paper. In the calibrated model, a switch from widespread climate denial to widespread climate acceptance causes an effective transfer of housing wealth across regions of $41bn immediately and $507bn over the following century. Migration exacerbates this by amplifying housing price responses. The spatially-equalizing effects of migration only dominate 50 years post-shock. Climate uncertainty causes ongoing regressive wealth transfers through higher equilibrium rents, borne mostly by poorer households.
Details
Presentation Type
Paper Presentation in a Themed Session
Theme
KEYWORDS
Environmental, Urban, Spatial, Housing, Machine Learning, Deep Learning, Heterogeneity, Climate
