LAUSR

Most US households invest a substantial fraction of their wealth in an individual asset: their family home. This paper studies the empirical properties of idiosyncratic house price risk using a novel micro-dataset of house sales and remodeling permits from the metropolitan areas of Los Angeles, San Diego, and San Francisco. Unlike what is usually assumed in the literature, I find that idiosyncratic housing risk does not scale with time. Moreover, the fraction of capital gains variance determined by the idiosyncratic component decreases over a house holding period. I analyze the effects of idiosyncratic risk on household welfare, using a quantitative portfolio model. Exposure to idiosyncratic shocks has substantial effects on the trade-off between owning and renting, especially for short holding periods.