The Urban Wage Premium: Sorting, agglomeration economies or statistical artifact? The problem of sampling from lognormals
Vladislav Vysotsky and
Papers from arXiv.org
Economic explanations for the urban wage premium (UWP) fall in two categories: sorting of more productive individuals into larger cities, or agglomeration externalities. We present a third hitherto neglected mechanism: a statistical artifact arising from sampling lognormally distributed wages. We show how this artificial UWP emerges, systematically and predictably, when the variance of log-wages is larger than twice the log-size of workers sampled in the smallest city. We present an analytic derivation of this connection between lognormals and increasing returns to scale using extreme value theory. We validate our results analyzing simulated data and real data on more than six million real Colombian wages across more than five hundred municipalities. We find that when taking random samples of $1\%$, or less, of all Colombian workers the estimated real and artificial UWP are both $7\%$, and become statistically indistinguishable, yet both significantly larger than zero. This highlights the importance of working with large samples of workers. We propose a method to tell whether an estimate of UWP is real or an artifact.
New Economics Papers: this item is included in nep-eff, nep-geo and nep-ure
Date: 2018-07, Revised 2018-10
References: View references in EconPapers View complete reference list from CitEc
Citations: Track citations by RSS feed
Downloads: (external link)
http://arxiv.org/pdf/1807.09424 Latest version (application/pdf)
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
Persistent link: https://EconPapers.repec.org/RePEc:arx:papers:1807.09424
Access Statistics for this paper
More papers in Papers from arXiv.org
Bibliographic data for series maintained by arXiv administrators ().