The Long- and Short-Run Relationships Between House Prices and Rents: Evidence from 18 OECD Countries

Asset-pricing theory predicts that in equilibrium the housing and rental markets should be closely interconnected. I analyse both the long-run and short-run relationships between real house prices and real rents using a new quarterly dataset for 18 OECD countries from 1968:Q2 to 2021:Q4. I find that only in few countries Engle-Granger and Johansen tests provide evidence of cointegration between the two variables. Granger-causality tests show that rent growth and house price growth have predictive power on each other in some countries. However, by using a rolling-window framework, I document that these longrun and short-run relationships vary substantially over time. Three main factors might explain this discrepancy between theory and empirical evidence: (1) different quality of the dwellings on sale and to rent, (2) data aggregation, and (3) rental markets regulations.