
Gang-related explosions have become an increasingly visible form of violent crime in Sweden, but little is known about their effects on local housing markets. This study examines how explosions affect residential property prices in Stockholm and whether the effects vary across locations and with repeated exposure. We combine geocoded data on 146 explosions between 2021 and 2024 with housing transactions and estimate hedonic difference-in-differences models using propensity score matching and an event-study approach. The results show that explosions are associated with lower housing prices within 500 metres. The average price reduction is approximately 2%, while areas exposed to multiple explosions experience substantially larger declines of around 9%. The event-study results show no systematic differences in price trends before explosions but significant price changes afterwards. The effects also vary geographically, with the estimated decline concentrated in Stockholm Municipality. Explosions occur disproportionately in neighbourhoods with lower incomes, lower educational attainment, and higher unemployment. These findings show that the economic effects of explosive violence extend beyond direct physical damage and are reflected in local housing markets. As explosions are concentrated in relatively disadvantaged neighbourhoods, their effects on property values may contribute to existing spatial inequalities.
Black poor people are much more likely to be homeless than white or Hispanic poor people. Why? The main correlates of disparities involving Black people are high rents, low rental vacancy rates, and segregation, with rents being by far the most important. I try to explain why with a model of selectivity on the part of landlords in a monopolistically competitive housing market, and with a model of informal housing markets. I also discuss why racial disparities are much greater in California and New York than in the rest of the country and why US homelessness is concentrated so heavily in those two states. All three questions are closely related.
In a field experiment on Airbnb in Paris during and after the 2024 Olympic Games, we examine how discrimination responds to temporary market disruptions. During the Olympics, booking requests from male and North African-sounding guests were significantly less likely to be accepted than those from female or French-sounding guests. After the Games, discrimination against male guests disappeared, and bias against North African-sounding guests, though still present, declined substantially. Our results suggest that the higher prejudice during the Olympic Games was driven primarily by increased host selectivity in neighborhoods experiencing the strongest Airbnb market tensions. Exploiting variation in Airbnb's Instant Book feature, we find that ethnic discrimination is primarily statistical, while gender discrimination is more consistent with taste-based preferences for women. Beyond acceptance rates, North African-sounding guests also experienced slower and less enthusiastic positive responses, pointing to both strategic and implicit forms of unequal treatment.
We develop a framework to holistically test for and monitor the impact of different types of shocks affecting a country’s housing market, yet originating from housing-external sources. We classify events along three dimensions leading to testable predictions: supply versus demand, prices versus quantities, and immediate versus staggered. These dimensions translate into guidance about which data type, statistical measure, and testing strategy should be used. To perform such tests, suitable statistical models are needed which we implement as a hierarchical hedonic model for prices and a count model for volumes. We apply this framework to the Austrian real estate market together with two disruptive events: the COVID-19 pandemic and affordability barriers due to stricter lending standards, increased living costs, and higher mortgage interest rates. The results are largely consistent with expectations when following our proposed guidelines and selecting the suggested combination of data, models, and testing strategy. By that, some housing market puzzles are resolved. Deviating from the a priori classification exercise means that some developments would have been undetected or wrongly interpreted. Further, adopting our framework consistently in the future would lead to better comparable research results and enable meta-conclusions from the bulk of studies available across time and space.