Missing Links: Social Networks & Local Housing Markets

This paper examines whether homebuyers with better-connected social networks achieve higher capital gains on their housing purchases. The central hypothesis is that buyers whose family members, colleagues, and classmates reside near a target neighbourhood possess superior local information (about underpriced properties, improving streets, and motivated sellers) which translates into better purchasing decisions and higher subsequent appreciation. I develop a theoretical model in the style of Kurlat (2016) of competitive housing market with heterogenous information. Locally 'tied' buyers draw from a more favourable expertise distribution, improving their ability to identify high-quality houses. The model yields a key sign-reversal prediction: tie exposure improves outcomes for high-quality purchases (information channel) but worsens them for low-quality ones, as the preference premium buyers pay to live near their network generates overpayment relative to fundamental value. To test these predictions empirically, I construct a spatially-weighted information centrality measure using Dutch administrative microdata on family, colleague, and classmate networks from Statistics Netherlands (CBS). The empirical estimation is identified using neighbourhood-quarter sale fixed effects and the central mechanism is further tested using empirical predictions derived from the model.