In this paper we examine whether house price changes drive mortgage-based equity extraction. To do this we use longitudinal survey data with subjective information about current and expected future house prices to calculate unanticipated house price changes. We link this information at the individual level to high quality administrative records with information about mortgage borrowing as well as savings in various financial instruments. We find a marginal propensity to extract out of unanticipated housing wealth gains to be 2-5 per cent. We find no evidence that the effect is driven by collateral constraints. Instead, the effect is driven by about 11 per cent of the observations where the respondent is recorded having actively taken out a new mortgage. These results point to the existence of a housing wealth effect that is intimately connected to the functioning of the mortgage market, and this suggests that monetary policy could amplify the interaction between unanticipated housing wealth gains and spending by affecting the passthrough on interest rates on mortgage loans.