Public housing projects are hotly debated especially for their impact on neighboring properties. On the one hand development of housing units could enhance available amenities and increase nearby house prices. On the other hand concentration of low-income households could create negative spillovers that reduce prices. For large projects, the expansion of the housing stock could reduce prices as well. A large body of literature provides estimates for the impact of public housing projects in developed countries. Yet, to the best of our knowledge, there is no rigorous empirical analysis of the impact of such projects in developing countries. In this paper, we study a large public housing project known as the Mehr housing project in Iran to shed light on this question in a developing country. In 2007 the Iranian government revealed a plan to facilitate the construction of 2 million affordable apartments across almost all cities in the country. We use the exact delivery time of Mehr housing units and their postal regions to set up a difference-in-differences strategy for the estimation of the impact of Mehr units on existing nearby house prices.

Using the universe of house transactions for 19 large cities in Iran between 2010 and mid-2019, we compare house price changes in Mehr postal regions to non-Mehr postal regions around the time of Mehr housing units delivery. Our results show that after the delivery of Mehr units, house prices in Mehr postal regions decline by around 11 percent relative to prices in non-Mehr postal regions. This effect is significant at the 5 percent level and is robust to several specification checks like the inclusion of city-by-time fixed effects, allowing for differential trends for suburban locations, and regions with higher initial property values. We also provide suggestive evidence on the role of disamenity effects by looking at the heterogeneity of results across different house areas, cities, and over time.