Regional Heterogeneity and Monetary Policy

Joseph Vavra, Associate Professor of Economics

Most immediately I am interested in using the CRISM data to continue my paper “Regional Heterogeneity and Monetary Policy” with Martin Beraja, Andreas Fuster and Erik Hurst. Andreas Fuster is currently at the NY Fed, so we used the CRISM data through the NY fed for the current work on this paper, but he is going to be leaving in the near future at which point we will no longer have access to the data through the fed or be able to work on revisions of the paper.

In that paper, we argue that time-series variation in the regional distribution of housing equity has important implications for monetary policy through its effects on refinancing activity. Reductions in interest rates can potentially spur spending as households lock in lower monthly payments and extract housing equity, but the ability to do this depends on the level of equity that households have in their homes. We use the CRISM data to construct the spatial distribution of housing equity and then use it to explore a variety of refinancing and spending responses in the immediate aftermath of QE1. In particular, we show that there is less refinancing, less cash out equity extraction conditional on refinancing and less auto spending in low equity locations after QE1. These low equity locations like Vegas and Phoenix where lots of borrowers are underwater also had the deepest recessions, so QE mostly benefited the locations doing relatively well. We then use a structural model to interpret the aggregate implications of these cross-region refinancing differences and to argue that time-series variation in the distribution of equity will lead to time-series variation in the consequences of monetary policy.

While the baseline results for this paper have already been done using CRISM data, I am interested in extending the analysis beyond just QE1. We currently have not explored the other QE episodes, and have not computed the equity distribution in a long time-series up to the present. So I’m interested in extending our analysis over the last decade and exploring the spatial distribution of refinancing activity over the housing recovery and how that has interacted with interest rate movements over this period. We have also currently done little with the individual data and have focused mainly on the implications at the regional level since this is the level of variation of most house price/equity shocks during the QE episode.

Regional Heterogeneity and the Refinancing Channel of Monetary Policy, with Martin Beraja, Andreas Fuster, Erik Hurst, Joseph Vavra, The Quarterly Journal of Economics, Volume 134, Issue 1, February 2019, Pages 109–183,