This reply refutes the objection raised by Levy (2009) about the fit of the upper tail of the city size distribution in Eeckhout (2004). I show that the method on which his conclusion is based is unsubstantiated. The visual interpretation of the fit on log-log plots is misleading. In addition, the methodology used to estimate a truncated subsample of the distribution while testing its significance against a distribution with prespecified parameters is ill-founded. The main conclusion is that Gibrat's law holds: city sizes follow proportionate growth, thus giving rise to a lognormal size distribution, tail included. (JEL R11, R12, R23)
No takes yet. Share an insight, caveat, or question.
Jan Eeckhout (2009) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: