ABSTRACT
Sports facilities across the globe are increasingly being sited in the urban core. The existing literature in sports and urban affairs fails to consider that the magnitude of negative externalities attributed to sports facilities could vary by the built environment, that the social costs of locating a facility in the urban core may be greater than the social costs of locating it in more suburban areas due to the greater development density in the former. This study test this hypothesis using daily police incident-level data from Sacramento, California in 2016, when a professional basketball team moved from a more suburban arena to a new one in the city’s downtown. Using a doughnut-hole specification in a triple difference-in-difference framework, it is concluded that police response time to incidents in the immediate vicinity of the downtown arena during event periods is on average 7.4 per cent longer, or about 33 seconds. No such delay is observed for incidents near the suburban arena during event periods. Both conclusions are robust to a placebo.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1 New facilities in suburban and ex-urban areas are more common outside of the United States—Allianz Riveria in Nice, France and Sivas 4 Arena in Turkey, for example. In contrast, of the more than 25 new sports facilities for the five professional sports leagues in the US, only on has been built outside of the urban core (SunTrust Park in Cobb County, Georgia).
2 Month year fixed effects were also tested but the joint test statistic was not statistically different from zero, so they were excluded.
3 Since not every incident in the data file has a description, the bottom and upper 10 per cent of calls were excluded, or incidents with response times less than two minutes or more than 20 minutes. These calls are assumed to fall into the other excluded categories.