By Pedro Gete, Professor of Finance at IE University, Athena Tsouderou, Ph.D. candidate at IE University, and Susan M. Wachter, Professor of Real Estate and Finance at The Wharton School of the University of Pennsylvania
Hurricane Harvey made landfall on the coast of Houston, Texas, in August of 2017, and Hurricane Irma followed in September hitting southern Florida. These two hurricanes rank in the top five of the costliest storms on record in the United States, destroying thousands of homes due to flooding and/or strong winds. While major hurricanes with the potential to cause severe economic damage and a wave of mortgage defaults is not a new phenomenon, climate change is expected to bring larger and more frequent catastrophes in the near future. The U.S. has a new financial market that trades specifically on the risk of mortgage default in the U.S. (called Credit Risk Transfers). This new market for Credit Risk Transfers allows further investigation into an important question: the effect of climate risk on mortgage pricing.
In our paper, we study the Credit Risk Transfer market during a period of stress caused by Hurricanes Harvey and Irma. We can be the first to measure how the investors who buy these securities price credit risk in U.S. mortgages absent any government intervention. That is, how much compensation they demand for taking on the risk of mortgage defaults while being exposed to climate change risk. Through simulations, we also yield predictions for other crises and the design of the U.S. housing system, which is unique in the world because the U.S. government currently absorbs nearly all the risk from mortgage default.
We find that the current system encourages households to be exposed to climate risk, because it prevents them from internalizing the hurricane risk of the homes they purchase. On the positive side, the U.S. housing system achieves a powerful macroeconomic policy that lowers costs of mortgage credit in a crisis. For example, in the first quarter of the COVID crisis, the cost of mortgage credit would have been 21% higher if the private market –and not the U.S. government– was taking on the credit risk. Thus, our paper contributes to two active debates: financial consequences of climate risk, and how to reform the U.S. housing finance system.
Our paper provides policymakers with precise estimates on how mortgage rates will react if they are totally market-determined and exposed to risks of default. Our paper has several important insights for policy:
First, under the current system, with the U.S. government absorbing all credit risk, geographical differences in mortgage cost are muted. Thus, the current system does not generate any incentive to avoid settling down in hurricane-exposed locations. Households pay the same costs for a mortgage to buy a home exposed to a hurricane than to buy one that is safe. In other words, the current system encourages climate risk taking. Mortgage prices do not help households to internalize hurricane risk.
Second, the government absorbing credit risk is a powerful macroeconomic policy to help with recessions, like fiscal or monetary policy. Using our model calibrated to match the empirical estimates from the hurricanes, we can measure
the stimulus provided by the government guarantees that absorb credit risk. For example, for the COVID crisis that so far increased mortgage defaults by 114%, without the government guarantees the cost of mortgage credit would have been 21% higher. Higher cost of credit would mean lower prices, less construction, less wealth and higher unemployment.
Thus, our two previous results show that in the optimal system likely we do not want to completely remove the government guarantees to mortgage credit risk. Moreover, inequalities in access to credit will increase without government intervention because the guarantees help especially the riskier households who are usually low-income (Gete and Zecchetto 2018). The optimal level of intervention in mortgage markets is an issue that warrants further investigation.
