This paper investigates the two types of housing wealth effects: the “pure” wealth effect, and the collateral effect. We incorporate mortgage equity withdrawals (MEW) and the influence of mortgage liberalization into the Campbell and Mankiw (1989) model. Based on U.S. data during the 1978Q1-2012Q3 period, our empirical results suggest that the collateral effect adds to housing marginal propensity to consume (MPC), not “in addition to” but “instead of” the “pure” housing wealth effect. Furthermore, mortgage liberalization significantly amplifies the collateral effect. Conditional on the use of MEW and the share of the non-GSEs market-based financial intermediaries’ mortgage holdings to total home mortgage, housing wealth has an average MPC of 1.78 cents, a maximum of 6.07 cents. With the relaxed access to mortgage credit, by 2007, the MEW shock explained close to 40% of the forecasting variance of consumption growth.