Housing Betas

This paper documents new stylized facts about returns and cashflow growth rates on stocks and housing over decade-long holding periods. While cashflow growth rates on the two assets comove positively, their returns comove negatively until the Global Financial Crisis and positively thereafter. These facts present a puzzle for representative-agent models that imply positive return comovement for assets with similar cashflows. I consider a heterogeneous-agent model with segmented stock and housing markets connected through credit. News about the aggregate economy generates negative return comovement. Recent shifts such as wealthier homebuyers and institutional housing purchases reduce the importance of credit and segmentation.