Local Notes
This item is not available for online access until July 25, 2024. Digital access copies must be provided for use. Contact the Beinecke Rare Book and Manuscript Library at beinecke.library@yale.edu to request access.
Notes
Advisors: Geanakoplos, John Committee members: Mian, Atif; Davila, Eduardo.
Description based on Dissertations Abstracts International, Volume: 84-02, Section: B.
Summary
This dissertation consists of three essays on the topics of macroeconomics, finance, and climate change.In the first essay I examine how two government policies implicitly encourage homeownership in areas increasingly threatened by climate change. First, the government spends billions of dollars helping rebuild homes and infrastructure after major disasters. Second, government-sponsored enterprises, Fannie Mae and Freddie Mac, do not charge mortgage borrowers for location-specific climate risk. In this paper, I introduce climate risk into a canonical lifecycle model of consumption and housing choice to estimate how removing both distortions would affect the number of homeowners living in areas exposed to climate risk, specifically flood risk. I use Hurricane Sandy's landfall in New Jersey as a natural experiment to quantify the two distortions and discipline the model. My model predicts that jointly removing both distortions would have reduced the number of homes affected by Hurricane Sandy by 20%. I further show that a vast majority of the reduction in the number of homeowners exposed to climate risk could have been achieved by two measures, taxing at-risk homeowners and charging them a flood risk premium as part of the mortgage. This latter policy regime would have reduced residential losses from the storm by 30%, i.e., $2.3 billion, despite continuance of post-disaster public assistance.In the second essay, co-authored with Chase Ross and Sharon Ross, we show that household mortgage leverage has an overlooked benefit: it helps diversify climate risk. We present a simple model to show that, under reasonable levels of higher climate risk, mortgage leverage declines in equilibrium but makes households better off than in a world with no leverage. We use property-level data to confirm the predictions of the model. We show that households with more mortgage leverage reduced their consumption less than their less-levered counterparts after the 2019 Midwest Floods.In the third essay, co-authored with John Geanakoplos, we develop a quantitative, policy-relevant tool---the credit surface---for tracking credit conditions in the housing sector. Given market expectations of growth in home prices, coupled with the default and prepayment rates for mortgages, the credit surface outputs the fair mortgage rate for a loan given the borrower's credit score and desired loan-to-value ratio at origination. Depending on the market environment, the credit surface flattens or tightens, indicating the phase of the leverage cycle the market currently is in. We generate the credit surface for each month between 2002 and 2010, demonstrating its evolution over the course of the boom and bust of the housing market in the U.S.