Abstract
We analyze the effects of borrower-based macroprudential policy at the household-level. For identification, we exploit administrative Dutch tax-return and property ownership data linked to the universe of housing transactions, and the introduction of a mortgage loan-to-value limit. The regulation reduces mortgage leverage, with bunching in its limit. Ex-ante more-affected households substantially reduce overall leverage and debt servicing costs but consume greater liquidity to satisfy the regulation. Improvements in household solvency result in less financial distress and, given negative idiosyncratic shocks, better liquidity management. However, fewer households transition from renting into ownership. All of these effects are stronger for liquidity-constrained households.
