We study the the relationship between homebuyers’ beliefs about future house price changes and their mortgage leverage choices. From a theoetical perspective, this relationship is ambiguous, and depends on the magnitude of a “collateral adjustment friction” that reduces pessimistic homebuyers’ willingness to live in substantially smaller houses. When households primarily maximize the levered return of their property investment, more pessimistic homebuyers reduce their leverage to purchase smaller houses. On the other hand, when considerations such as family size pin down the desired property size, pessimistic homebuyers reduce their financial exposure to the housing market by making smaller downpayments to buy similarly-sized homes. To determine which scenario better describes the data, we empirically investigate the cross-sectional relationship between beliefs and leverage choices in the U.S. housing market. Our data combine mortgage financing information and a housing market expectations survey with anonymized social network data from Facebook. The survey shows that an individual’s belief distribution about future house price changes is affected by the recent house price experiences of her geographically distant friends, allowing us to exploit these experiences as quasi-exogenous shifters of individuals’ house price beliefs. We find that more pessimistic homebuyers choose higher leverage, in particular in states where default costs are relatively low, as well as during periods when prices are expected to fall on average. Overall, our results show how data from online social network services can help researchers test models with heterogeneous beliefs, allowing us to provide evidence for an important role of individuals’ beliefs in their financial decision-making.