“The Distributional Consequences of Climate Change: The Role of Housing Wealth, Expectations, and Uncertainty”

Job Market Paper

In the United States, the majority of the median household's wealth is in the form of a single climate-exposed asset: their home. The value of that home nonlinearly reflects current expectations and uncertainty about the future of its local economy and climate. To quantitatively understand the impact of different climate scenarios on household wealth, I build a dynamic spatial model of the U.S. with 1713 locations. Forward-looking heterogeneous households choose location, housing, and housing investment in spatially-segmented housing markets, endogenizing local rent and real estate prices. I find the global solution under aggregate climate uncertainty using a novel but simple and highly general deep learning method. To quantify the model, I harmonize and combine recent estimates on the impact of global climate conditions on local productivity, amenities, maintenance costs, and disaster risk. I calibrate a stochastic global climate process using projections from the climate literature and discipline uncertainty using historical homeowners insurance premia and payouts. In the calibrated model, climate change, in expectation, exacerbates existing wealth inequality. While wealthier households are more exposed to climate risk through housing assets, in equilibrium they benefit from the resulting endogenous risk premium on those assets. This risk premium and its regressive implication arise only in the global solution.