How do housing bubbles affect other economic sectors? We show that in the presence of collateral constraints, a bubble initially raises housing credit demand and crowds out credit to non-housing firms. If the bubble lasts, however, housing credit repayments raise banks’ net worth and expand credit supply, so that crowding-out eventually gives way to crowding-in. This is consistent with evidence from the recentSpanish housing bubble. Initially, credit growth of non-housing firms was lower at banks with higher bubble exposure, and firms relying on these banks exhibited lower credit and output growth. During thebubble’s last years, these effects reversed.