Published online by Cambridge University Press: 10 September 2021
This paper studies the role of monetary policy for the dynamics of US mortgage debt, which accounts for the largest part of household debt. A time-varying parameter vector autoregressive (VAR) model allows us to study the variation in the sensitivity of mortgage debt to monetary policy. We find that an identically sized policy shock became less effective over time. We use a dynamic stochastic general equilibrium model to show that a fall in the share of adjustable rate mortgages (ARMs) can replicate this finding. Calibrating the model to the drop in the ARM share since the 1980s yields a decline in the sensitivity of housing debt to monetary policy which is quantitatively similar to the VAR results. A sacrifice ratio for mortgage debt reveals that a policy tightening directed toward reducing household debt became more expensive in terms of a loss in employment. Counterfactuals show that this result cannot be attributed to changes in monetary policy itself.