Working Paper: CEPR ID: DP16806
Authors: Francisco Amaral; Sebastian Kohl; Martin Dohmen; Moritz Schularick
Abstract: This paper makes the first comprehensive attempt to study within-country heterogeneity of housing returns. We introduce a new city-level data set covering 15 OECD countries over 150 years and show that national housing markets are characterized by systematicspatial variation in housing returns. Total returns in large agglomerations are close to 100 basis points lower per year than in other parts of the same country. The excess returns outside the large cities can be rationalized as a compensation for higher risk, especiallyhigher co-variance with income growth and lower liquidity. Real estate in diversified large agglomerations is comparatively safe.
Keywords: Asset Returns; Housing Risk; Superstar Cities; Regional Housing Markets
JEL Codes: G10; G12; N90; R21; R31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Location (large agglomerations) (R12) | Lower total returns (G19) |
| Location (large agglomerations) (R12) | Returns (Y70) |
| Higher risk (covariance with income growth and lower liquidity) (G19) | Lower total returns (G19) |
| Lower covariance between housing returns and income growth in large cities (R21) | Higher idiosyncratic risk outside large cities (R12) |
| Lower rental returns in large cities (R21) | Lower total returns (G19) |