Sumit Agarwal, Cristian Badarinza, and Wenlan Qian, National University of Singapore

The Effectiveness of Housing Collateral Tightening Policy

Agarwal, Badarinza, and Qian show that macroprudential policies can lead to adverse selection in the market for residential mortgage loans. They exploit a unique loan-level data set and a policy experiment in Singapore that differentially targets mortgage contracts for second homes. For a horizon of up to one year after the policy roll-out, the researchers document a significant composition change towards a riskier type of borrowers that are twice as likely to become delinquent relative to a comparable non-treated cohort. Ex ante, these borrowers are not different in terms of observable characteristics, but they have lower behavioral credit scores and worse histories of credit card repayment. Consistent with the hypothesis of adverse selection towards a pool of more optimistic investors that fail to have a correct assessment of the policy impact and housing market growth, most of the effects appear to be driven by individual experiences of past price appreciation. Using a separate data set on bankruptcy proceedings, the researchers also document a higher probability of default in the market for investment properties, occurring immediately after the time of the policy change.