The impact of the spatial population distribution on economic growth: Evidence from the United States

We introduce a new class of measures called spatial population concentration (SPC) designed to quantify how many people live on average within a given radius of every person within a geographic area. The distinguishing feature relative to existing measures is that ours are the first to satisfy scale invariance, replication invariance, sensitivity to transfers, aggregativity, sensitivity to neighborhoods, invariance to uninhabited regions, concurrently. We calculate the SPC measure at the US county level for various radii and identify that the strongest relationship with subsequent economic growth can be found for a 25-km radius. Interacting SPC with diverse infrastructure metrics does not alter the optimal radius. This indicates that regional policies such as infrastructure projects which influence density should strategically target the 25-km distance range to maximize the growth impact. We also show that the importance of SPC has declined over time.