A Nonparametric Estimation of the Local Zipf Exponent for all US Cities

In this short paper we apply the methodology proposed by Ioannides and Overman (2003) to estimate a local Zipf exponent using data for the entire twentieth century of the complete distribution of cities (incorporated places) without any size restrictions in the US. The results reject Zipf’s Law from a long term perspective, as the estimated values are close to zero. However, decade by decade we find evidence in favour of Zipf’s Law. We also see how periods in which the Zipf exponent grows with city size are interspersed with others in which the relationship between the exponent and city shares is negative.