Research


Working Papers:

  • Housing Market Congestion and Internal Migration In Major European Cities

    Presentations: 76th European Meeting of the Econometric Society*, 13th European Meeting of the Urban Economics Association*, University of Exeter*, Brunel University London*, Toulouse School of Economics*, RWI Essen

    Housing plays a crucial role in influencing inter-regional relocations. Employing a parsimonious spatial search and matching model, this research reveals that congestion faced by renters could lead to out-migration—a phenomenon distinct from the well-documented "lock-in" and cost-related channels. Moreover, the correlation between congestion and inter-regional relocation is likely state-dependent, wherein migration flow directions may vary based on the tightness of the market. The central mechanism is the trade-off between the thick-market effect and congestion externality. If the latter dominates, an increase in congestion leads to reduced housing consumption, incentivising out-migration. Conversely, if the former dominates, such an increase could result in in-migration. Leveraging a proprietary micro-dataset obtained from a prominent property consulting company, I construct a congestion measure aligned with the theoretical framework for 34 major European cities spanning 2009-2021. Subsequently, I demonstrate a positive overall correlation between congestion and out-migration, while emphasising the state-dependent nature of this relationship, thereby validating the model’s insights. These novel empirical patterns are the first to unveil the association between congestion and inter-regional relocation. Policy analysis utilising the theoretical framework underscores potential adverse implications of imposing rent subsidies in cities with tight housing markets. In contrast, expanding the housing stock could be welfare-enhancing. The essence of these policy implications lies in the search and matching frictions; outcomes would be different if the housing market were Walrasian.

Work in Progress: