Efficient Urbanization for Mexican Development


  •  David Mayer-Foulkes    

Abstract

By applying Data Science techniques we find strong evidence that urbanization plays a key role in the process of development in Mexico, This process necessarily involves government action and therefore must be the subject of policy. We suggest that there are ways of streamlining the government’s role in providing the public goods of urbanization that can combine with and stimulate the competitive economic context. We apply Data Science techniques including visualization of the full universe of the object of study, and application of the Random Forest Classifier and Regressor machine learning algorithms, to municipal firm number growth obtained from Mexico’s full Directory of Economic Units for 2012 and 2016. These are aggregated at the municipal level by employment scales and one-digit production sectors, and combined with municipal demographic census data. Our visualization exercises also show that the dynamics of firm and population numbers is complex, such as in a changing fractal.


This work is licensed under a Creative Commons Attribution 4.0 License.