The empirical literature has demonstrated that housing assets exhibit larger wealth effects than stocks (or, more broadly, financial assets), which is often interpreted as a larger MPC (Marginal Propensity of Consumption) out of housing wealth. Still, the question remains as to whether this stylized fact has anything to do with the collaterality of housing assets. We build a household consumption and portfolio choice model with two risky assets, housing and stocks, whereby housing can be used

more »

... collateral to borrow against. The optimizing agent's preference and investment opportunity set generate implications of different MPCs for groups characterized by their respective asset/debt portfolios. Under calibrated parameters from macro data, the model exhibits the highest MPC for households who simultaneously borrow against housing asset and invest in stocks. We examine the Panel Study of Income Dynamics (PSID) micro data of homeowners and find no evidence of this implied collateral effect on non-durable consumption.